Free Trial

Featured This month

  • OCT biomarkers in diabetic macular edema: detection, quantification, and monitoring

    altris for dme
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min.

    Table of Contents

    The pathogenesis of OCT biomarkers in diabetic macular edema (DME) is complex and multifactorial. The key mechanism is disruption of the inner blood–retinal barrier resulting from chronic hyperglycemia.

    1. Introduction. Brief overview of diabetic macular edema (DME)
    2. Main OCT biomarkers of DME
      2.1 Signs of DME on OCT
      2.2 Quantitative parameters for monitoring DME on OCT

    DME management and patient education: key aspects of the modern approach
    References

    Introduction. Brief Overview of Diabetic Macular Edema (DME)

    The pathogenesis of OCT biomarkers in diabetic macular edema (DME) is complex and multifactorial. The key mechanism is disruption of the inner blood–retinal barrier resulting from chronic hyperglycemia. This leads to increased vascular permeability, plasma extravasation, and fluid accumulation within the retinal layers. Inflammatory processes, cytokine and growth factor activation—particularly vascular endothelial growth factor (VEGF)—also play an important role by further increasing vascular permeability and sustaining chronic edema. For this reason, DME is now regarded not only as a vascular disorder but also as a neurodegenerative and inflammatory pathology.

    Morphologically, DME is characterized by retinal thickening, the formation of intraretinal cystoid spaces, accumulation of subretinal fluid, and progressive photoreceptor damage. Importantly, these structural changes often develop long before the onset of clinical symptoms. A patient may not experience significant visual deterioration, while irreversible changes are already occurring at the microstructural level. This underscores the critical importance of early diagnosis and regular monitoring.

    Current clinical guidelines, including those of the American Academy of Ophthalmology, emphasize that the timely detection of DME and the early initiation of treatment significantly improve functional outcomes. However, effective patient management is impossible without precise instrumental monitoring, particularly with optical coherence tomography (OCT), which enables assessment of both the presence and progression of the pathological process.

    OCT has become the imaging modality that fundamentally transformed the approach to the diagnosis and treatment of DME. It provides noninvasive visualization of the retina with micron-level resolution, allowing detailed analysis of its layered structure.

    Moreover, modern tomographic systems enable a transition from qualitative to quantitative assessment. Measurement of central retinal thickness, lesion area, and other parameters provides objective monitoring of disease progression. This is especially important in the era of personalized medicine, when treatment decisions are based not only on the clinical picture but also on precise numerical indicators.

    Thus, modern management of DME cannot be envisioned without the systematic use of OCT and the analysis of its biomarkers. Ophthalmology has progressed from simply detecting edema to achieving a deeper understanding of microstructural changes and their clinical significance.

    The aim of this article is to summarize current evidence on OCT biomarkers of diabetic macular edema, their roles in detection, quantitative assessment, and disease monitoring, and the practical aspects of their use in clinical practice to optimize patient management.

    2. Main OCT Biomarkers in DME  

    The modern approach to DME is based on a fundamentally new understanding of OCT’s role. OCT biomarkers in DME enable not only assessment of fluid presence but also determination of its type, localization, severity, and impact on key functional structures, particularly the photoreceptors. This is critically important because DME may have different pathogenic mechanisms across patients, ranging from vascular and inflammatory processes to a predominantly tractional component.

    Particular attention should be paid to the role of OCT biomarkers in predicting the course of DME and treatment response. Features such as disorganization of the retinal inner layers (DRIL), the condition of the ellipsoid zone, and the presence of hyperreflective foci are now considered important indicators for predicting disease progression and functional outcomes. These biomarkers enable clinicians to anticipate which patients are more likely to respond favorably to anti-VEGF therapy and in which cases a less favourable visual outcome should be expected.

    Each structural element observed on an OCT scan carries its own clinical significance. These findings allow the physician to answer several key clinical questions:

    • how active the disease process is,
    • whether the condition is acute or chronic,
    • which structures have already undergone irreversible changes,
    • which pathogenic mechanism predominates,
    • and, most importantly, what the likely therapeutic response will be.

    Thus, OCT interpretation extends far beyond simple visual assessment. It becomes an analytical process in which the clinician integrates morphological features with quantitative measurements and the patient’s clinical data.

    According to current international recommendations, no single parameter—such as central retinal thickness—can be sufficient for clinical decision-making. Instead, analysis of a combination of morphological patterns, quantitative indicators, and their temporal changes is recommended.

    The transition from static assessment to dynamic monitoring is particularly important via OCT biomarkers in dme. The rate of change, stability, or progression of individual biomarkers is often more informative than their absolute values. For example, a slight but persistent increase in intraretinal fluid may have greater clinical significance than a single high retinal thickness measurement.

    In addition, modern technologies have considerably expanded the capabilities of OCT analysis. The use of automated retinal layer segmentation, quantitative fluid volume assessment, and artificial intelligence algorithms helps reduce subjective interpretation and improve the reproducibility of results. This is especially important both in clinical practice and in scientific research, where accuracy and standardization are essential.

    In summary, the contemporary concept of OCT biomarkers in DME is based on three interconnected levels of analysis:

    1. Morphological level — identification of structural changes and edema type.
    2. Quantitative level — measurement of retinal thickness, fluid volume, and lesion area.
    3. Prognostic level — assessment of the risk of progression and treatment response.

    It is precisely this multilevel evaluation that enables a transition from standardized treatment protocols to a personalized approach, in which the therapeutic strategy is determined by each patient’s OCT biomarker profile.

    detect dme

    What are the key OCT biomarkers for monitoring diabetic macular edema (DME)?

    The most clinically relevant OCT biomarkers in DME include:

    • Intraretinal Fluid (IRF) — cystic fluid accumulation within retinal layers, strongly associated with active edema and vision impairment.
    • Subretinal Fluid (SRF) — fluid beneath the neurosensory retina, often linked to inflammatory activity and treatment response.
    • Hyperreflective Foci (HRF) — small reflective spots that may indicate inflammation, lipid extravasation, or retinal tissue damage.
    • Disorganization of Retinal Inner Layers (DRIL) — disruption of inner retinal architecture associated with poorer visual outcomes.
    • Ellipsoid Zone (EZ) Disruption — damage to photoreceptor integrity, often correlated with reduced visual acuity.
    • Central Retinal Thickness (CRT) — a widely used quantitative metric for assessing edema severity and treatment response.
    • Neurosensory Retina Atrophy — thinning and structural loss that may reflect chronic retinal damage and disease progression.

    Monitoring these biomarkers over time helps clinicians evaluate disease activity, predict visual prognosis, assess therapeutic response, and optimize individualized treatment strategies. AI-powered OCT analysis further improves reproducibility and enables scalable, quantitative longitudinal monitoring.

    2.1 OCT Features of Diabetic Macular Edema (DME)

    There are several distinct OCT patterns associated with DME, including diffuse retinal thickening, cystoid macular edema (intraretinal cystic spaces), and serous retinal detachment (subretinal fluid).

    1. Diffuse retinal thickening

    This is characterized by a uniform increase in macular thickness, resulting from fluid accumulation in the extracellular space of the neurosensory retina due to disruption of the inner blood–retinal barrier and an imbalance between fluid leakage and the resorptive capacity of the retinal pigment epithelium (RPE).

    When cystic spaces are present, their diameter must not exceed 50 μm; otherwise, the edema is classified as cystoid.

    In chronic edema (lasting more than 6–9 months), irreversible photoreceptor damage may occur, along with the development of retinal atrophy.

    Key features:

    • may be the only finding in early stages
    • requires careful quantitative assessment (central retinal thickness, CRT measurement)
    • Macular thickness dynamics are an important biomarker of treatment efficacy

    2. Cystoid macular edema (intraretinal cystic spaces)

    This is a key structural marker of DME. On OCT, it appears as hyporeflective, round or oval cavities, predominantly located in the inner nuclear layer (INL) and the outer plexiform layer (OPL).

    Clinical significance:

    • reflects vascular hyperpermeability
    • indicates active edema
    • large and confluent cysts may suggest a chronic process and are associated with a worse functional prognosis

    Long-standing cysts can lead to mechanical stretching of retinal tissue and secondary photoreceptor damage.

    3. Serous retinal detachment (subretinal fluid)

    Subretinal fluid refers to the accumulation of fluid between the neurosensory retina and the retinal pigment epithelium (RPE).

    On OCT, it appears as a hyporeflective space above the RPE and is associated with neurosensory retinal detachment.

    Clinical interpretation:

    • may be a marker of active disease
    • in some cases is associated with a better response to anti-VEGF therapy
    • causes less photoreceptor damage than chronic intraretinal cysts

    Although subretinal fluid is generally associated with a relatively better visual prognosis, its presence requires careful monitoring and should be considered when planning anti-VEGF treatment.

    dme progression

    2.2 Quantitative parameters for monitoring DME on OCT

    After morphological assessment, the next step is quantitative analysis. Currently, several key quantitative parameters can be obtained:

    • Central retinal thickness (CRT) – the most widely used parameter
    • Macular volume
    • Fluid quantity and volume

    These measurements enable precise monitoring of treatment response and help guide decisions regarding injection intervals.

    3. Management of DME and patient education: key aspects of the modern approach

    Modern management of diabetic macular edema (DME) is based on a comprehensive, personalized strategy in which OCT plays a central role. Today, therapeutic decisions are influenced by the morphological type of edema, disease activity, integrity of neurosensory retinal structures, individual patient characteristics, comorbidities, and prognostic biomarkers.

    A key principle of the contemporary approach is the integration of structural OCT biomarkers in DME into clinical decision-making. These biomarkers not only help determine whether treatment is necessary, but also assist in selecting the optimal therapeutic modality, assessing response, and timely adjustment of management strategy.

    Additional factors influencing therapy selection include:

    • Presence of disorganization of the retinal inner layers (DRIL) and disruption of the ellipsoid zone (EZ)
    • Response to previous treatments
    • Systemic comorbidities (renal impairment, hypertension, adherence/compliance issues)

    Treatment

    Anti-VEGF therapy

    Anti-VEGF agents (aflibercept, ranibizumab, bevacizumab) remain the first-line treatment for DME, as they directly target the key pathogenic mechanism—vascular hyperpermeability.

    Newer agents with extended durability are emerging, including implantable drug delivery systems.

    However, not all patients respond equally to anti-VEGF therapy. Therefore, OCT biomarker analysis is crucial: for example, a predominance of intraretinal cystic changes is usually associated with a good response to anti-VEGF, whereas a high number of hyperreflective foci or signs of chronic edema may indicate a significant inflammatory component and support consideration of steroid therapy.

    Intravitreal corticosteroid implants

    Steroids are used in cases of chronic and refractory DME, insufficient response to anti-VEGF therapy, and in patients with a pro-inflammatory phenotype.

    Laser therapy

    Although laser treatment has become less central in current practice, it remains useful in selected clinical scenarios. Subthreshold micropulse laser is more commonly used in patients with focal edema without involvement of the foveal center.

    Thus, treatment decisions today are no longer universal; they are based on an individualized OCT-based patient profile.

    Role of OCT in treatment

    OCT accompanies the patient throughout all stages of treatment and performs several key functions:

    • determination of indications for initiating therapy (presence of fluid, macular thickening, involvement of the foveal region)
    • assessment of treatment response (reduction of fluid, normalization of thickness, structural restoration)
    • detection of resistance or partial response
    • optimization of injection intervals (treat-and-extend or pro re nata strategies)

    A particularly important aspect is that OCT can detect subclinical changes. For example, minimal fluid accumulation may appear before any subjective deterioration in vision. This enables timely treatment adjustments and the prevention of functional loss.

    In addition, OCT helps avoid both under- and overtreatment. In patients with a stable anatomical profile and no fluid, injection intervals can be gradually extended, reducing the burden on both the patient and the healthcare system.

    dme monitoring

    Monitoring

    The frequency of follow-up depends on disease stage and activity:

    • active treatment phase – monthly visits with OCT control
    • stabilization phase – every 2–4 months
    • long-term follow-up – individualized, depending on recurrence risk and associated risk factors

    It is important to emphasize that monitoring must remain regular even in the absence of symptoms. DME can progress asymptomatically, and only OCT allows objective assessment of retinal status.

    Dynamic follow-up is critical: comparison of sequential scans provides the most valuable information about disease progression.

    What is important to explain to the patient

    Effective management of DME is not possible without active patient participation; therefore, communication is a key component of treatment.

    The patient must clearly understand that:

    • DME is a chronic condition requiring long-term monitoring
    • treatment aims to stabilize and slow disease progression, not always to fully restore vision
    • interruption of therapy without medical advice may lead to deterioration
    • regular visits and OCT monitoring are critical, even if vision appears stable

    It is especially important to explain the role of OCT to patients. Showing scans and explaining changes significantly improves treatment adherence.

    Lifestyle and systemic control

    Since diabetic macular edema (DME) is a complication of a systemic disease, control of the patient’s overall health is of critical importance.

    Key recommendations include:

    • optimal glycemic control
    • blood pressure management
    • correction of lipid profile
    • healthy diet rich in antioxidants
    • regular physical activity
    • smoking cessation

    DME infographics

    Psychological aspects and treatment adherence

    DME often follows a long and fluctuating course, which may lead to treatment fatigue or reduced motivation in patients. Many patients underestimate the severity of the condition, especially in early stages when visual acuity is still preserved.

    In this context, OCT becomes not only a diagnostic tool but also a communication instrument. Visualization of pathological changes helps patients better understand the disease and the necessity of treatment.

    Establishing a partnership between physician and patient is essential for successful long-term management.

    Conclusion

    OCT biomarkers in DME now allow not only precise diagnosis but also an approach that goes far beyond traditional retinal assessment. Thanks to its high resolution and ability to visualize microstructural changes, OCT enables the detection of subtle abnormalities before clinically significant symptoms appear. This opens the way to a new level of patient management—shifting from descriptive assessment to quantitative evaluation of pathological changes, their dynamics, and treatment response. Furthermore, OCT biomarker analysis allows prediction of disease course, identification of progression risk, and individualization of therapeutic strategies for each patient.

    Modern DME management is not just diagnosis and treatment, but a comprehensive clinical decision-making system based on objective, standardized data. There is a clear shift from subjective interpretation of fundus changes to structured analytics, where every parameter matters: retinal thickness, presence of intra- or subretinal fluid, status of outer retinal layers, and macular architectural disruption. OCT has become the key tool transforming clinical practice, making it more precise, reproducible, and evidence-based. It allows clinicians not only to confirm the presence of pathology but also to better understand its nature, activity, and potential reversibility.

    Ultimately, effective DME management today is the result of synergy between modern imaging technologies, clinical reasoning, and active patient engagement. Proper interpretation of OCT images must be integrated into the overall clinical picture, taking into account systemic factors, diabetes duration, and individual patient characteristics. In this process, OCT acts as a central link—a bridge between diagnosis and treatment—uniting all oct biomarkers in diabetic macular edema components into a coherent clinical system. This approach leads to better functional outcomes, preservation of vision, and significant long-term improvement in patients’ quality of life.

    FAQ

    1. How can AI help detect OCT biomarkers in DME?

    AI can automatically identify key OCT biomarkers such as intraretinal fluid, subretinal fluid, hyperreflective foci, and retinal layer disruptions, helping clinicians detect disease activity faster and more consistently.

    2. Why is quantitative biomarker analysis important in DME?

    Quantification enables objective measurement of biomarker volume, area, and progression over time, supporting treatment decisions, therapy response assessment, and longitudinal patient monitoring.

    3. Which OCT biomarkers are most relevant for monitoring DME progression?

    Commonly monitored biomarkers include intraretinal fluid (IRF), subretinal fluid (SRF), hyperreflective foci (HRF), disorganization of retinal inner layers (DRIL), and ellipsoid zone disruption.

    4. How can automated OCT analysis improve clinical workflow?

    Automated analysis reduces manual interpretation time, improves reproducibility, standardizes reporting, and helps clinicians prioritize patients who may require closer follow-up or treatment adjustments.

    References

    1. https://pubmed.ncbi.nlm.nih.gov/38460657/
    2. https://brief.euretina.org/research/association-of-retinal-oct-biomarkers-with-reading-performance-in-patients-with-diabetic-macular-edema-dme
    3. https://www.mdpi.com/2075-4418/14/1/76
    4. https://www.sciencedirect.com/science/article/pii/S1572100024000814
    5. https://link.springer.com/article/10.1186/s40942-023-00473-w
    6. https://www.medscape.com/viewarticle/1001580
    7. https://www.cureus.com/articles/227801-innovations-in-diabetic-macular-edema-management-a-comprehensive-review-of-automated-quantification-and-anti-vascular-endothelial-growth-factor-intervention#!/

     

  • ‍RWE in Ophthalmology: Challenges of Collection and Processing

    RWE
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    3 min.

    Pharma has no shortage of data in ophthalmology—EHRs, imaging repositories, claims, registries—but the industry still faces a persistent RWE gap when it comes to turning that data into commercially actionable insight. 

    This gap becomes especially critical in the context of modern ophthalmic therapies, where timing, disease stage, and adherence directly impact outcomes—and, by extension, market performance. Without precise, scalable ways to identify eligible patients (e.g., early-stage AMD, DME with specific biomarkers), commercial teams are left relying on proxies or delayed claims data. At the same time, tracking real-world outcomes remains reactive and retrospective, limiting the ability to support value-based narratives, optimize field strategy, or respond dynamically to physician behavior.

    For commercial and market access leaders, closing the RWE gap is no longer a “data strategy” initiative—it’s a growth imperative. The organizations that will lead are those that move beyond passive data aggregation toward active, AI-driven interpretation of multimodal ophthalmic data, enabling near real-time patient identification and outcome tracking. This shift not only strengthens evidence generation but directly translates into sharper targeting, more credible value communication, and ultimately, stronger adoption curves in an increasingly competitive therapeutic landscape.

    chart

    RWE Key Components

    Real-world evidence (RWE) in ophthalmology is fundamentally multimodal—it emerges from the combination of

    • OCT imaging, 
    • clinical metrics, 
    • treatment patterns, 
    • and patient context over time.

    When these components are connected, they move beyond descriptive data and become decision-grade insight for commercial, medical, and market access teams.

    Imaging (OCT): the gold mine of RWE

    OCT is the anchor, but only when integrated with outcomes, treatment data, and longitudinal context does it unlock its full commercial value—turning raw data into actionable RWE that can directly shape strategy and growth.

    Leverage data from past or ongoing clinical trials with the ability to standardize it within a unified ecosystem. OCT segmentation model can enable you to extract robust, clinically meaningful insights from your trial data to:

    • Gain a deeper understanding of patient responses to treatments by characterizing disease progression and outcomes over time.
    • Segment patient populations into subgroups to evaluate biomarker-driven profiles and build predictive analytics—helping streamline future trials by refining endpoints and optimizing inclusion and exclusion criteria.

    All that, as well as to build one of the most extensive and comprehensive real-world evidence (RWE) databases in ophthalmology may be quite possible within one OCT-vendor neutral data analysis platform ecosystem. 

    Clinical outcomes: functional reality

    Clinical measures—most notably visual acuity (VA), intraocular pressure, and physician-reported assessments—represent the functional impact of disease and treatment. These endpoints are still central to regulatory and commercial narratives, but in isolation they are often lagging indicators. By the time vision declines, disease progression may already be advanced. When paired with imaging, however, clinical outcomes provide the critical link between anatomical change and patient benefit, strengthening real-world value stories and payer communication.

    RWE delivers critical insights into how therapies perform in routine clinical practice—such as healthcare outcomes, treatment adherence, and protocol efficiency—often revealing patterns that differ from those observed in controlled clinical trials.

    GA AI

    Treatment data: what actually happens in practice

    Treatment data captures real-world behavior—which therapies are used, dosing frequency, switching patterns, and adherence. This is where the gap between clinical trial protocols and actual practice becomes visible. For commercial teams, this layer reveals drop-off points, under-treatment, and competitive dynamics at a granular level. When combined with OCT and outcomes, this approach makes it possible to understand not just what is happening but why—for example, whether discontinuation is driven by lack of response, disease stabilization, or operational constraints.

    OCT segmentation enables high-throughput processing and deep interrogation of large-scale datasets, enhancing the interpretation of real-world data. It can validate existing assumptions, uncover new patterns, and support hypothesis generation and testing. This analysis provides a clearer view of treatment efficacy and safety, disease progression in real-world settings, and a more precise understanding of target patient populations.

    Demographics and patient context

    Demographics (age, gender, geography) and broader patient context (comorbidities, access to care) provide the segmentation layer for RWE. These factors influence disease prevalence, treatment eligibility, and adherence patterns. While less granular than imaging, they are essential for market sizing, targeting, and equity considerations, helping commercial leaders understand where the highest-value opportunities—and barriers—exist across populations.

    Data analysis includes population-level distributions of retinal layer thickness and fluid volumes; longitudinal tracking of layer and fluid changes over time; and assessment of retinal layer attenuation/loss (depletion maps) and atrophic regions (Figure 2), among other endpoints. 

    Longitudinal progression: the real differentiator

    The true power of RWE lies in its longitudinal nature—tracking how patients evolve over time across imaging, outcomes, and treatment. This enables identification of disease trajectories, early signals of response or non-response, and optimal intervention windows. For pharma decision-makers, longitudinal RWE transforms static snapshots into predictive insight, supporting earlier intervention strategies, more precise patient journeys, and stronger, evidence-backed differentiation in crowded markets.

    The Core Problem

    There are core RWE problems worth mentioning. pay attention to the following bottlenecks: fragmented data, unstructured imaging, lack of standardization, and broken longitudinal tracking.

    • Fragmented data: Patient information is dispersed across multiple systems, limiting the ability to generate a unified, comprehensive view.
    • Unstructured imaging: Large volumes of imaging data, such as OCT scans, remain unstructured and difficult to analyze at scale.
    • Lack of standardization: Variability in data formats and clinical protocols hinders consistent analysis and comparison.
    • Broken longitudinal tracking: Incomplete or disconnected patient timelines prevent accurate assessment of disease progression and treatment outcomes over time.

    The core issue isn’t volume; it’s fragmentation and lack of standardization. OCT scans sit in one system, visual acuity in another, treatment histories elsewhere, often unstructured or inconsistently coded. As a result, even well-resourced teams struggle to answer seemingly simple questions like: Who are the untreated but eligible patients? or Which cohorts are actually benefiting from therapy in real-world settings?

    However, AI is transforming the RWE in ophthalmology research to: 

    • Design smarter, biomarker-driven trials
    • Estimate disease burden and patient volumes
    • Track outcomes, safety, and progression
    • Strengthen regulatory and market access strategies

    Resources to aid ophthalmologists in evaluating the quality of RWE are available, such as the Good Research for Comparative Effectiveness (GRACE) principles, which can support the evaluation of observational comparative effectiveness studies.

    GRACE checklist to support ophthalmologists in the evaluation of RWE:

    GRACE list

    Data Methods
    ✓ Were treatment and/or important details of treatment exposure adequately recorded for the study purpose in the data source(s)? ✓ Was the study (or analysis) population restricted to new initiators of treatment or those starting a new course of treatment?

     

    ✓ Were the primary outcomes adequately recorded for the study purpose? ✓ If one or more comparison groups were used, were they concurrent comparators? If not, did the authors justify the use of historical comparison groups?

     

    ✓ Was the primary clinical outcome(s) measured objectively rather than subject to clinical judgment? ✓ Were important confounding and effect-modifying variables taken into account in the design and/or analysis?

     

    ✓ Were primary outcomes validated, adjudicated, or otherwise known to be valid in a similar population? ✓ Is the classification of exposed and unexposed person-time free of “immortal time bias”?

     

    ✓ Was the primary outcome(s) measured or identified in an equivalent manner between the treatment/intervention group and the comparison group? ✓ Were any meaningful analyses conducted to test key assumptions on which primary results are based?

     

    ✓ Were important covariates that may be known confounders or effect modifiers available and recorded?  

     

    *Table adapted from Dreyer NA et al. J Manag Care Pharm 2014; 20 (3): 301–308 (Table 1). While used with permission of the publisher, the publisher disclaims all endorsement of any organization, product or technique as a matter of policy.

    AI in RWE

    Why Imaging Is Critical?

    OCT is a stepping stone to understanding Geographic Atrophy. Since the approval of the first therapy targeting geographic atrophy in early 2023, interest in the disease has increased dramatically. At the same time, a growing number of clinical trials are underway, evaluating the safety and efficacy of multiple investigational compounds.

    Imaging—particularly OCT (optical coherence tomography)—is the backbone of meaningful real-world evidence in retinal disease, because it captures what clinical codes and claims data simply cannot: anatomical change over time. Without structured OCT data, RWE becomes fragmented and largely inferential, relying on indirect proxies like treatment patterns or visual acuity alone. This creates a major blind spot in understanding disease progression, especially in chronic degenerative conditions where structural deterioration often precedes functional loss.

    In geographic atrophy (GA), this gap is especially critical for therapies such as Syfovre (pegcetacoplan) and Izervay (avacincaptad pegol). These treatments are designed to slow structural progression, not just improve symptoms, meaning their real-world impact can only be properly assessed through consistent, longitudinal imaging markers—lesion growth, retinal layer integrity, and atrophy expansion. When OCT data is unstructured or missing, it becomes impossible to reliably track these anatomical endpoints across time and across care settings.

    As a result, RWE datasets without standardized OCT integration fail to support robust patient journey reconstruction, dilute treatment effect signals, and limit the ability to identify responders vs non-responders. For pharma and clinical stakeholders, this means missed opportunities to demonstrate value, optimize patient selection, and build predictive models that depend on continuous structural imaging rather than episodic clinical snapshots.

    The Way Forward

    AI-driven structuring of imaging data is emerging as the missing link between raw clinical information and truly actionable real-world evidence (RWE). In ophthalmology, vast volumes of OCT scans remain underutilized because they are stored as unstructured images, making large-scale analysis slow, inconsistent, and often impractical. 

    By applying advanced algorithms to automatically segment retinal layers, detect biomarkers, and standardize measurements, platforms like Altris AI transform imaging data into structured, quantifiable, and interoperable datasets. This enables pharma and clinical teams to move beyond anecdotal insights toward statistically robust, evidence-driven decision-making.

    With AI-powered structuring, imaging data becomes scalable and longitudinal by design. Instead of isolated snapshots, clinicians and researchers gain continuous, comparable measurements across time, patients, and sites. This unlocks real-time monitoring of disease progression and treatment response, supports precise patient stratification, and accelerates cohort identification for therapies such as GA treatments. 

    Ultimately, structured imaging powered by AI bridges the gap between clinical practice and research—turning OCT into a high-value, real-time RWE engine that is both clinically meaningful and commercially actionable.

    References:

    https://europe.ophthalmologytimes.com/view/assessing-the-quality-of-real-world-evidence-in-retinal-diseases

    https://onlinelibrary.wiley.com/doi/full/10.1111/aos.14698

    https://www.visionacademy.org/media/3251/download

     

popular Posted

  • OCT biomarkers in diabetic macular edema: detection, quantification, and monitoring

    altris for dme
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min.

    Table of Contents

    The pathogenesis of OCT biomarkers in diabetic macular edema (DME) is complex and multifactorial. The key mechanism is disruption of the inner blood–retinal barrier resulting from chronic hyperglycemia.

    1. Introduction. Brief overview of diabetic macular edema (DME)
    2. Main OCT biomarkers of DME
      2.1 Signs of DME on OCT
      2.2 Quantitative parameters for monitoring DME on OCT

    DME management and patient education: key aspects of the modern approach
    References

    Introduction. Brief Overview of Diabetic Macular Edema (DME)

    The pathogenesis of OCT biomarkers in diabetic macular edema (DME) is complex and multifactorial. The key mechanism is disruption of the inner blood–retinal barrier resulting from chronic hyperglycemia. This leads to increased vascular permeability, plasma extravasation, and fluid accumulation within the retinal layers. Inflammatory processes, cytokine and growth factor activation—particularly vascular endothelial growth factor (VEGF)—also play an important role by further increasing vascular permeability and sustaining chronic edema. For this reason, DME is now regarded not only as a vascular disorder but also as a neurodegenerative and inflammatory pathology.

    Morphologically, DME is characterized by retinal thickening, the formation of intraretinal cystoid spaces, accumulation of subretinal fluid, and progressive photoreceptor damage. Importantly, these structural changes often develop long before the onset of clinical symptoms. A patient may not experience significant visual deterioration, while irreversible changes are already occurring at the microstructural level. This underscores the critical importance of early diagnosis and regular monitoring.

    Current clinical guidelines, including those of the American Academy of Ophthalmology, emphasize that the timely detection of DME and the early initiation of treatment significantly improve functional outcomes. However, effective patient management is impossible without precise instrumental monitoring, particularly with optical coherence tomography (OCT), which enables assessment of both the presence and progression of the pathological process.

    OCT has become the imaging modality that fundamentally transformed the approach to the diagnosis and treatment of DME. It provides noninvasive visualization of the retina with micron-level resolution, allowing detailed analysis of its layered structure.

    Moreover, modern tomographic systems enable a transition from qualitative to quantitative assessment. Measurement of central retinal thickness, lesion area, and other parameters provides objective monitoring of disease progression. This is especially important in the era of personalized medicine, when treatment decisions are based not only on the clinical picture but also on precise numerical indicators.

    Thus, modern management of DME cannot be envisioned without the systematic use of OCT and the analysis of its biomarkers. Ophthalmology has progressed from simply detecting edema to achieving a deeper understanding of microstructural changes and their clinical significance.

    The aim of this article is to summarize current evidence on OCT biomarkers of diabetic macular edema, their roles in detection, quantitative assessment, and disease monitoring, and the practical aspects of their use in clinical practice to optimize patient management.

    2. Main OCT Biomarkers in DME  

    The modern approach to DME is based on a fundamentally new understanding of OCT’s role. OCT biomarkers in DME enable not only assessment of fluid presence but also determination of its type, localization, severity, and impact on key functional structures, particularly the photoreceptors. This is critically important because DME may have different pathogenic mechanisms across patients, ranging from vascular and inflammatory processes to a predominantly tractional component.

    Particular attention should be paid to the role of OCT biomarkers in predicting the course of DME and treatment response. Features such as disorganization of the retinal inner layers (DRIL), the condition of the ellipsoid zone, and the presence of hyperreflective foci are now considered important indicators for predicting disease progression and functional outcomes. These biomarkers enable clinicians to anticipate which patients are more likely to respond favorably to anti-VEGF therapy and in which cases a less favourable visual outcome should be expected.

    Each structural element observed on an OCT scan carries its own clinical significance. These findings allow the physician to answer several key clinical questions:

    • how active the disease process is,
    • whether the condition is acute or chronic,
    • which structures have already undergone irreversible changes,
    • which pathogenic mechanism predominates,
    • and, most importantly, what the likely therapeutic response will be.

    Thus, OCT interpretation extends far beyond simple visual assessment. It becomes an analytical process in which the clinician integrates morphological features with quantitative measurements and the patient’s clinical data.

    According to current international recommendations, no single parameter—such as central retinal thickness—can be sufficient for clinical decision-making. Instead, analysis of a combination of morphological patterns, quantitative indicators, and their temporal changes is recommended.

    The transition from static assessment to dynamic monitoring is particularly important via OCT biomarkers in dme. The rate of change, stability, or progression of individual biomarkers is often more informative than their absolute values. For example, a slight but persistent increase in intraretinal fluid may have greater clinical significance than a single high retinal thickness measurement.

    In addition, modern technologies have considerably expanded the capabilities of OCT analysis. The use of automated retinal layer segmentation, quantitative fluid volume assessment, and artificial intelligence algorithms helps reduce subjective interpretation and improve the reproducibility of results. This is especially important both in clinical practice and in scientific research, where accuracy and standardization are essential.

    In summary, the contemporary concept of OCT biomarkers in DME is based on three interconnected levels of analysis:

    1. Morphological level — identification of structural changes and edema type.
    2. Quantitative level — measurement of retinal thickness, fluid volume, and lesion area.
    3. Prognostic level — assessment of the risk of progression and treatment response.

    It is precisely this multilevel evaluation that enables a transition from standardized treatment protocols to a personalized approach, in which the therapeutic strategy is determined by each patient’s OCT biomarker profile.

    detect dme

    What are the key OCT biomarkers for monitoring diabetic macular edema (DME)?

    The most clinically relevant OCT biomarkers in DME include:

    • Intraretinal Fluid (IRF) — cystic fluid accumulation within retinal layers, strongly associated with active edema and vision impairment.
    • Subretinal Fluid (SRF) — fluid beneath the neurosensory retina, often linked to inflammatory activity and treatment response.
    • Hyperreflective Foci (HRF) — small reflective spots that may indicate inflammation, lipid extravasation, or retinal tissue damage.
    • Disorganization of Retinal Inner Layers (DRIL) — disruption of inner retinal architecture associated with poorer visual outcomes.
    • Ellipsoid Zone (EZ) Disruption — damage to photoreceptor integrity, often correlated with reduced visual acuity.
    • Central Retinal Thickness (CRT) — a widely used quantitative metric for assessing edema severity and treatment response.
    • Neurosensory Retina Atrophy — thinning and structural loss that may reflect chronic retinal damage and disease progression.

    Monitoring these biomarkers over time helps clinicians evaluate disease activity, predict visual prognosis, assess therapeutic response, and optimize individualized treatment strategies. AI-powered OCT analysis further improves reproducibility and enables scalable, quantitative longitudinal monitoring.

    2.1 OCT Features of Diabetic Macular Edema (DME)

    There are several distinct OCT patterns associated with DME, including diffuse retinal thickening, cystoid macular edema (intraretinal cystic spaces), and serous retinal detachment (subretinal fluid).

    1. Diffuse retinal thickening

    This is characterized by a uniform increase in macular thickness, resulting from fluid accumulation in the extracellular space of the neurosensory retina due to disruption of the inner blood–retinal barrier and an imbalance between fluid leakage and the resorptive capacity of the retinal pigment epithelium (RPE).

    When cystic spaces are present, their diameter must not exceed 50 μm; otherwise, the edema is classified as cystoid.

    In chronic edema (lasting more than 6–9 months), irreversible photoreceptor damage may occur, along with the development of retinal atrophy.

    Key features:

    • may be the only finding in early stages
    • requires careful quantitative assessment (central retinal thickness, CRT measurement)
    • Macular thickness dynamics are an important biomarker of treatment efficacy

    2. Cystoid macular edema (intraretinal cystic spaces)

    This is a key structural marker of DME. On OCT, it appears as hyporeflective, round or oval cavities, predominantly located in the inner nuclear layer (INL) and the outer plexiform layer (OPL).

    Clinical significance:

    • reflects vascular hyperpermeability
    • indicates active edema
    • large and confluent cysts may suggest a chronic process and are associated with a worse functional prognosis

    Long-standing cysts can lead to mechanical stretching of retinal tissue and secondary photoreceptor damage.

    3. Serous retinal detachment (subretinal fluid)

    Subretinal fluid refers to the accumulation of fluid between the neurosensory retina and the retinal pigment epithelium (RPE).

    On OCT, it appears as a hyporeflective space above the RPE and is associated with neurosensory retinal detachment.

    Clinical interpretation:

    • may be a marker of active disease
    • in some cases is associated with a better response to anti-VEGF therapy
    • causes less photoreceptor damage than chronic intraretinal cysts

    Although subretinal fluid is generally associated with a relatively better visual prognosis, its presence requires careful monitoring and should be considered when planning anti-VEGF treatment.

    dme progression

    2.2 Quantitative parameters for monitoring DME on OCT

    After morphological assessment, the next step is quantitative analysis. Currently, several key quantitative parameters can be obtained:

    • Central retinal thickness (CRT) – the most widely used parameter
    • Macular volume
    • Fluid quantity and volume

    These measurements enable precise monitoring of treatment response and help guide decisions regarding injection intervals.

    3. Management of DME and patient education: key aspects of the modern approach

    Modern management of diabetic macular edema (DME) is based on a comprehensive, personalized strategy in which OCT plays a central role. Today, therapeutic decisions are influenced by the morphological type of edema, disease activity, integrity of neurosensory retinal structures, individual patient characteristics, comorbidities, and prognostic biomarkers.

    A key principle of the contemporary approach is the integration of structural OCT biomarkers in DME into clinical decision-making. These biomarkers not only help determine whether treatment is necessary, but also assist in selecting the optimal therapeutic modality, assessing response, and timely adjustment of management strategy.

    Additional factors influencing therapy selection include:

    • Presence of disorganization of the retinal inner layers (DRIL) and disruption of the ellipsoid zone (EZ)
    • Response to previous treatments
    • Systemic comorbidities (renal impairment, hypertension, adherence/compliance issues)

    Treatment

    Anti-VEGF therapy

    Anti-VEGF agents (aflibercept, ranibizumab, bevacizumab) remain the first-line treatment for DME, as they directly target the key pathogenic mechanism—vascular hyperpermeability.

    Newer agents with extended durability are emerging, including implantable drug delivery systems.

    However, not all patients respond equally to anti-VEGF therapy. Therefore, OCT biomarker analysis is crucial: for example, a predominance of intraretinal cystic changes is usually associated with a good response to anti-VEGF, whereas a high number of hyperreflective foci or signs of chronic edema may indicate a significant inflammatory component and support consideration of steroid therapy.

    Intravitreal corticosteroid implants

    Steroids are used in cases of chronic and refractory DME, insufficient response to anti-VEGF therapy, and in patients with a pro-inflammatory phenotype.

    Laser therapy

    Although laser treatment has become less central in current practice, it remains useful in selected clinical scenarios. Subthreshold micropulse laser is more commonly used in patients with focal edema without involvement of the foveal center.

    Thus, treatment decisions today are no longer universal; they are based on an individualized OCT-based patient profile.

    Role of OCT in treatment

    OCT accompanies the patient throughout all stages of treatment and performs several key functions:

    • determination of indications for initiating therapy (presence of fluid, macular thickening, involvement of the foveal region)
    • assessment of treatment response (reduction of fluid, normalization of thickness, structural restoration)
    • detection of resistance or partial response
    • optimization of injection intervals (treat-and-extend or pro re nata strategies)

    A particularly important aspect is that OCT can detect subclinical changes. For example, minimal fluid accumulation may appear before any subjective deterioration in vision. This enables timely treatment adjustments and the prevention of functional loss.

    In addition, OCT helps avoid both under- and overtreatment. In patients with a stable anatomical profile and no fluid, injection intervals can be gradually extended, reducing the burden on both the patient and the healthcare system.

    dme monitoring

    Monitoring

    The frequency of follow-up depends on disease stage and activity:

    • active treatment phase – monthly visits with OCT control
    • stabilization phase – every 2–4 months
    • long-term follow-up – individualized, depending on recurrence risk and associated risk factors

    It is important to emphasize that monitoring must remain regular even in the absence of symptoms. DME can progress asymptomatically, and only OCT allows objective assessment of retinal status.

    Dynamic follow-up is critical: comparison of sequential scans provides the most valuable information about disease progression.

    What is important to explain to the patient

    Effective management of DME is not possible without active patient participation; therefore, communication is a key component of treatment.

    The patient must clearly understand that:

    • DME is a chronic condition requiring long-term monitoring
    • treatment aims to stabilize and slow disease progression, not always to fully restore vision
    • interruption of therapy without medical advice may lead to deterioration
    • regular visits and OCT monitoring are critical, even if vision appears stable

    It is especially important to explain the role of OCT to patients. Showing scans and explaining changes significantly improves treatment adherence.

    Lifestyle and systemic control

    Since diabetic macular edema (DME) is a complication of a systemic disease, control of the patient’s overall health is of critical importance.

    Key recommendations include:

    • optimal glycemic control
    • blood pressure management
    • correction of lipid profile
    • healthy diet rich in antioxidants
    • regular physical activity
    • smoking cessation

    DME infographics

    Psychological aspects and treatment adherence

    DME often follows a long and fluctuating course, which may lead to treatment fatigue or reduced motivation in patients. Many patients underestimate the severity of the condition, especially in early stages when visual acuity is still preserved.

    In this context, OCT becomes not only a diagnostic tool but also a communication instrument. Visualization of pathological changes helps patients better understand the disease and the necessity of treatment.

    Establishing a partnership between physician and patient is essential for successful long-term management.

    Conclusion

    OCT biomarkers in DME now allow not only precise diagnosis but also an approach that goes far beyond traditional retinal assessment. Thanks to its high resolution and ability to visualize microstructural changes, OCT enables the detection of subtle abnormalities before clinically significant symptoms appear. This opens the way to a new level of patient management—shifting from descriptive assessment to quantitative evaluation of pathological changes, their dynamics, and treatment response. Furthermore, OCT biomarker analysis allows prediction of disease course, identification of progression risk, and individualization of therapeutic strategies for each patient.

    Modern DME management is not just diagnosis and treatment, but a comprehensive clinical decision-making system based on objective, standardized data. There is a clear shift from subjective interpretation of fundus changes to structured analytics, where every parameter matters: retinal thickness, presence of intra- or subretinal fluid, status of outer retinal layers, and macular architectural disruption. OCT has become the key tool transforming clinical practice, making it more precise, reproducible, and evidence-based. It allows clinicians not only to confirm the presence of pathology but also to better understand its nature, activity, and potential reversibility.

    Ultimately, effective DME management today is the result of synergy between modern imaging technologies, clinical reasoning, and active patient engagement. Proper interpretation of OCT images must be integrated into the overall clinical picture, taking into account systemic factors, diabetes duration, and individual patient characteristics. In this process, OCT acts as a central link—a bridge between diagnosis and treatment—uniting all oct biomarkers in diabetic macular edema components into a coherent clinical system. This approach leads to better functional outcomes, preservation of vision, and significant long-term improvement in patients’ quality of life.

    FAQ

    1. How can AI help detect OCT biomarkers in DME?

    AI can automatically identify key OCT biomarkers such as intraretinal fluid, subretinal fluid, hyperreflective foci, and retinal layer disruptions, helping clinicians detect disease activity faster and more consistently.

    2. Why is quantitative biomarker analysis important in DME?

    Quantification enables objective measurement of biomarker volume, area, and progression over time, supporting treatment decisions, therapy response assessment, and longitudinal patient monitoring.

    3. Which OCT biomarkers are most relevant for monitoring DME progression?

    Commonly monitored biomarkers include intraretinal fluid (IRF), subretinal fluid (SRF), hyperreflective foci (HRF), disorganization of retinal inner layers (DRIL), and ellipsoid zone disruption.

    4. How can automated OCT analysis improve clinical workflow?

    Automated analysis reduces manual interpretation time, improves reproducibility, standardizes reporting, and helps clinicians prioritize patients who may require closer follow-up or treatment adjustments.

    References

    1. https://pubmed.ncbi.nlm.nih.gov/38460657/
    2. https://brief.euretina.org/research/association-of-retinal-oct-biomarkers-with-reading-performance-in-patients-with-diabetic-macular-edema-dme
    3. https://www.mdpi.com/2075-4418/14/1/76
    4. https://www.sciencedirect.com/science/article/pii/S1572100024000814
    5. https://link.springer.com/article/10.1186/s40942-023-00473-w
    6. https://www.medscape.com/viewarticle/1001580
    7. https://www.cureus.com/articles/227801-innovations-in-diabetic-macular-edema-management-a-comprehensive-review-of-automated-quantification-and-anti-vascular-endothelial-growth-factor-intervention#!/

     

  • ‍RWE in Ophthalmology: Challenges of Collection and Processing

    RWE
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    3 min.

    Pharma has no shortage of data in ophthalmology—EHRs, imaging repositories, claims, registries—but the industry still faces a persistent RWE gap when it comes to turning that data into commercially actionable insight. 

    This gap becomes especially critical in the context of modern ophthalmic therapies, where timing, disease stage, and adherence directly impact outcomes—and, by extension, market performance. Without precise, scalable ways to identify eligible patients (e.g., early-stage AMD, DME with specific biomarkers), commercial teams are left relying on proxies or delayed claims data. At the same time, tracking real-world outcomes remains reactive and retrospective, limiting the ability to support value-based narratives, optimize field strategy, or respond dynamically to physician behavior.

    For commercial and market access leaders, closing the RWE gap is no longer a “data strategy” initiative—it’s a growth imperative. The organizations that will lead are those that move beyond passive data aggregation toward active, AI-driven interpretation of multimodal ophthalmic data, enabling near real-time patient identification and outcome tracking. This shift not only strengthens evidence generation but directly translates into sharper targeting, more credible value communication, and ultimately, stronger adoption curves in an increasingly competitive therapeutic landscape.

    chart

    RWE Key Components

    Real-world evidence (RWE) in ophthalmology is fundamentally multimodal—it emerges from the combination of

    • OCT imaging, 
    • clinical metrics, 
    • treatment patterns, 
    • and patient context over time.

    When these components are connected, they move beyond descriptive data and become decision-grade insight for commercial, medical, and market access teams.

    Imaging (OCT): the gold mine of RWE

    OCT is the anchor, but only when integrated with outcomes, treatment data, and longitudinal context does it unlock its full commercial value—turning raw data into actionable RWE that can directly shape strategy and growth.

    Leverage data from past or ongoing clinical trials with the ability to standardize it within a unified ecosystem. OCT segmentation model can enable you to extract robust, clinically meaningful insights from your trial data to:

    • Gain a deeper understanding of patient responses to treatments by characterizing disease progression and outcomes over time.
    • Segment patient populations into subgroups to evaluate biomarker-driven profiles and build predictive analytics—helping streamline future trials by refining endpoints and optimizing inclusion and exclusion criteria.

    All that, as well as to build one of the most extensive and comprehensive real-world evidence (RWE) databases in ophthalmology may be quite possible within one OCT-vendor neutral data analysis platform ecosystem. 

    Clinical outcomes: functional reality

    Clinical measures—most notably visual acuity (VA), intraocular pressure, and physician-reported assessments—represent the functional impact of disease and treatment. These endpoints are still central to regulatory and commercial narratives, but in isolation they are often lagging indicators. By the time vision declines, disease progression may already be advanced. When paired with imaging, however, clinical outcomes provide the critical link between anatomical change and patient benefit, strengthening real-world value stories and payer communication.

    RWE delivers critical insights into how therapies perform in routine clinical practice—such as healthcare outcomes, treatment adherence, and protocol efficiency—often revealing patterns that differ from those observed in controlled clinical trials.

    GA AI

    Treatment data: what actually happens in practice

    Treatment data captures real-world behavior—which therapies are used, dosing frequency, switching patterns, and adherence. This is where the gap between clinical trial protocols and actual practice becomes visible. For commercial teams, this layer reveals drop-off points, under-treatment, and competitive dynamics at a granular level. When combined with OCT and outcomes, this approach makes it possible to understand not just what is happening but why—for example, whether discontinuation is driven by lack of response, disease stabilization, or operational constraints.

    OCT segmentation enables high-throughput processing and deep interrogation of large-scale datasets, enhancing the interpretation of real-world data. It can validate existing assumptions, uncover new patterns, and support hypothesis generation and testing. This analysis provides a clearer view of treatment efficacy and safety, disease progression in real-world settings, and a more precise understanding of target patient populations.

    Demographics and patient context

    Demographics (age, gender, geography) and broader patient context (comorbidities, access to care) provide the segmentation layer for RWE. These factors influence disease prevalence, treatment eligibility, and adherence patterns. While less granular than imaging, they are essential for market sizing, targeting, and equity considerations, helping commercial leaders understand where the highest-value opportunities—and barriers—exist across populations.

    Data analysis includes population-level distributions of retinal layer thickness and fluid volumes; longitudinal tracking of layer and fluid changes over time; and assessment of retinal layer attenuation/loss (depletion maps) and atrophic regions (Figure 2), among other endpoints. 

    Longitudinal progression: the real differentiator

    The true power of RWE lies in its longitudinal nature—tracking how patients evolve over time across imaging, outcomes, and treatment. This enables identification of disease trajectories, early signals of response or non-response, and optimal intervention windows. For pharma decision-makers, longitudinal RWE transforms static snapshots into predictive insight, supporting earlier intervention strategies, more precise patient journeys, and stronger, evidence-backed differentiation in crowded markets.

    The Core Problem

    There are core RWE problems worth mentioning. pay attention to the following bottlenecks: fragmented data, unstructured imaging, lack of standardization, and broken longitudinal tracking.

    • Fragmented data: Patient information is dispersed across multiple systems, limiting the ability to generate a unified, comprehensive view.
    • Unstructured imaging: Large volumes of imaging data, such as OCT scans, remain unstructured and difficult to analyze at scale.
    • Lack of standardization: Variability in data formats and clinical protocols hinders consistent analysis and comparison.
    • Broken longitudinal tracking: Incomplete or disconnected patient timelines prevent accurate assessment of disease progression and treatment outcomes over time.

    The core issue isn’t volume; it’s fragmentation and lack of standardization. OCT scans sit in one system, visual acuity in another, treatment histories elsewhere, often unstructured or inconsistently coded. As a result, even well-resourced teams struggle to answer seemingly simple questions like: Who are the untreated but eligible patients? or Which cohorts are actually benefiting from therapy in real-world settings?

    However, AI is transforming the RWE in ophthalmology research to: 

    • Design smarter, biomarker-driven trials
    • Estimate disease burden and patient volumes
    • Track outcomes, safety, and progression
    • Strengthen regulatory and market access strategies

    Resources to aid ophthalmologists in evaluating the quality of RWE are available, such as the Good Research for Comparative Effectiveness (GRACE) principles, which can support the evaluation of observational comparative effectiveness studies.

    GRACE checklist to support ophthalmologists in the evaluation of RWE:

    GRACE list

    Data Methods
    ✓ Were treatment and/or important details of treatment exposure adequately recorded for the study purpose in the data source(s)? ✓ Was the study (or analysis) population restricted to new initiators of treatment or those starting a new course of treatment?

     

    ✓ Were the primary outcomes adequately recorded for the study purpose? ✓ If one or more comparison groups were used, were they concurrent comparators? If not, did the authors justify the use of historical comparison groups?

     

    ✓ Was the primary clinical outcome(s) measured objectively rather than subject to clinical judgment? ✓ Were important confounding and effect-modifying variables taken into account in the design and/or analysis?

     

    ✓ Were primary outcomes validated, adjudicated, or otherwise known to be valid in a similar population? ✓ Is the classification of exposed and unexposed person-time free of “immortal time bias”?

     

    ✓ Was the primary outcome(s) measured or identified in an equivalent manner between the treatment/intervention group and the comparison group? ✓ Were any meaningful analyses conducted to test key assumptions on which primary results are based?

     

    ✓ Were important covariates that may be known confounders or effect modifiers available and recorded?  

     

    *Table adapted from Dreyer NA et al. J Manag Care Pharm 2014; 20 (3): 301–308 (Table 1). While used with permission of the publisher, the publisher disclaims all endorsement of any organization, product or technique as a matter of policy.

    AI in RWE

    Why Imaging Is Critical?

    OCT is a stepping stone to understanding Geographic Atrophy. Since the approval of the first therapy targeting geographic atrophy in early 2023, interest in the disease has increased dramatically. At the same time, a growing number of clinical trials are underway, evaluating the safety and efficacy of multiple investigational compounds.

    Imaging—particularly OCT (optical coherence tomography)—is the backbone of meaningful real-world evidence in retinal disease, because it captures what clinical codes and claims data simply cannot: anatomical change over time. Without structured OCT data, RWE becomes fragmented and largely inferential, relying on indirect proxies like treatment patterns or visual acuity alone. This creates a major blind spot in understanding disease progression, especially in chronic degenerative conditions where structural deterioration often precedes functional loss.

    In geographic atrophy (GA), this gap is especially critical for therapies such as Syfovre (pegcetacoplan) and Izervay (avacincaptad pegol). These treatments are designed to slow structural progression, not just improve symptoms, meaning their real-world impact can only be properly assessed through consistent, longitudinal imaging markers—lesion growth, retinal layer integrity, and atrophy expansion. When OCT data is unstructured or missing, it becomes impossible to reliably track these anatomical endpoints across time and across care settings.

    As a result, RWE datasets without standardized OCT integration fail to support robust patient journey reconstruction, dilute treatment effect signals, and limit the ability to identify responders vs non-responders. For pharma and clinical stakeholders, this means missed opportunities to demonstrate value, optimize patient selection, and build predictive models that depend on continuous structural imaging rather than episodic clinical snapshots.

    The Way Forward

    AI-driven structuring of imaging data is emerging as the missing link between raw clinical information and truly actionable real-world evidence (RWE). In ophthalmology, vast volumes of OCT scans remain underutilized because they are stored as unstructured images, making large-scale analysis slow, inconsistent, and often impractical. 

    By applying advanced algorithms to automatically segment retinal layers, detect biomarkers, and standardize measurements, platforms like Altris AI transform imaging data into structured, quantifiable, and interoperable datasets. This enables pharma and clinical teams to move beyond anecdotal insights toward statistically robust, evidence-driven decision-making.

    With AI-powered structuring, imaging data becomes scalable and longitudinal by design. Instead of isolated snapshots, clinicians and researchers gain continuous, comparable measurements across time, patients, and sites. This unlocks real-time monitoring of disease progression and treatment response, supports precise patient stratification, and accelerates cohort identification for therapies such as GA treatments. 

    Ultimately, structured imaging powered by AI bridges the gap between clinical practice and research—turning OCT into a high-value, real-time RWE engine that is both clinically meaningful and commercially actionable.

    References:

    https://europe.ophthalmologytimes.com/view/assessing-the-quality-of-real-world-evidence-in-retinal-diseases

    https://onlinelibrary.wiley.com/doi/full/10.1111/aos.14698

    https://www.visionacademy.org/media/3251/download

     

  • PBM Monitoring on OCT: Drusen Progression

    PBM Monitoring on OCT: Drusen Progression
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min.

    Introduction: Role of PBM in Retinal Disease Management

    Photobiomodulation (PBM), also referred to as low-level light or laser therapy, has emerged as a promising non-invasive therapeutic strategy in ophthalmology, particularly for the management of retinal diseases. PBM utilizes low-energy light in the red-to-near-infrared spectrum (typically 600–1000 nm) to modulate cellular function through photochemical rather than thermal mechanisms. 

    According to  Retina Today, PBM is being used as an adjunctive or alternative treatment for several retinal diseases, including age-related macular degeneration (AMD), diabetic retinopathy (DR), and diabetic macular edema (DME).  Its advantages include a favorable safety profile, non-invasive delivery, and relatively low cost compared to conventional therapies.  Furthermore, advances in light-emitting diode (LED) technology have facilitated broader clinical application by enabling safe, uniform, and cost-effective retinal illumination. 

    Key OCT Biomarkers to Track During PBM Therapy

    Optical coherence tomography (OCT) has become an indispensable tool for monitoring retinal structure and treatment response in patients undergoing photobiomodulation (PBM) therapy. Given the mechanism of PBM—targeting mitochondrial function, reducing oxidative stress, and modulating inflammation—several OCT-derived biomarkers are particularly relevant for assessing therapeutic efficacy of dry AMD progression and in other retinal disorders.

    In this article, I will show how to quantify and monitor OCT biomarkers for effective PBM monitoring.

    1. PBM monitoring on OCT: Drusen Progression

    For patients on photobiomodulation (PBM), OCT monitoring of drusen is about one core question: Are we stabilizing or reversing RPE–Bruch’s membrane dysfunction, or is the eye still progressing toward atrophy? So, on B-scans, drusen are seen as:

    • RPE elevations (dome-shaped or irregular)
    • Material between RPE and Bruch’s membrane
    • Variable internal reflectivity

    Here are the key drusen biomarkers to track under PBM:

    1. Drusen volume (MOST important)
    • Measured via OCT segmentation (cube scans)
    • It represents the total disease burden

    As the PBM goal here is the stabilisation or reduction in drusen volume

    Red flag:

    • Continuous increase → disease progression
    1. Drusen height & area
    • Local structural impact on photoreceptors

    PBM signal:

    • Flattening = potential response
    • Increasing height = worsening RPE dysfunction
    1. Internal reflectivity
    • Homogeneous vs heterogeneous content

    An important nuance is that increasing heterogeneity may indicate:

    • calcification
    • regression OR collapse before atrophy

    So, it needs to be correlated with other signs. 

    You may observe Drusen regression patterns. However, not all regression is good.

    “Good” regression:

    • Gradual flattening
    • No photoreceptor loss
    • Stable RPE

     “Bad” regression (collapse):

    • Sudden disappearance
    • Followed by:
      • RPE loss
      • outer retinal thinning

    Leads to geographic atrophy (GA)

    To summarise, for PBM-treated patients, prioritize:

    • Drusen volume trend (longitudinal)
    • Photoreceptor integrity (EZ/ONL)
    • Signs of atrophy risk (HRF, collapse patterns)

    Dry AMD progression matters in PBM monitoring. PBM is currently aimed at:

    • early → intermediate dry AMD

    So when you monitor drusen on OCT, you’re not just tracking morphology — you’re tracking disease trajectory: Is the eye staying in intermediate AMD, or moving toward advanced stages (GA / nAMD)?

    2. Central retinal thickness (CRT)

    In addition to PBM monitoring on OCT: Drusen progression, one of the primary biomarkers is also central retinal thickness (CRT), which reflects changes in retinal edema and overall retinal integrity. Reductions in CRT during PBM therapy may indicate decreased inflammatory activity and improved fluid homeostasis, particularly in conditions such as diabetic macular edema (DME) and neovascular retinal diseases. However, in non-exudative conditions such as dry age-related macular degeneration (AMD), CRT changes may be subtle, necessitating the evaluation of additional structural parameters.

    3. The outer retinal layers

    The outer retinal layers, especially the integrity of the ellipsoid zone (EZ) and external limiting membrane (ELM), represent critical biomarkers of photoreceptor health. PBM has been associated with improved mitochondrial activity within photoreceptors, and preservation or restoration of EZ continuity on OCT may serve as a surrogate marker of functional recovery. Disruptions in these layers are strongly correlated with visual impairment, making them highly relevant endpoints in PBM studies.

    4. Retinal pigment epithelium (RPE) 

    Another key biomarker is retinal pigment epithelium (RPE) morphology, including the presence and evolution of drusen, subretinal drusenoid deposits (SDD), and RPE irregularities. PBM has been hypothesized to enhance RPE function and reduce oxidative burden, potentially leading to stabilization or regression of drusen volume over time. Quantitative drusen analysis using OCT can therefore provide insight into disease modification, particularly in intermediate AMD.

    5. Hyperreflective foci (HRF)

    Hyperreflective foci (HRF) are also important indicators of retinal inflammation and microglial activation. A reduction in HRF number or density during PBM therapy may reflect decreased inflammatory signaling, aligning with the known anti-inflammatory effects of light-based treatment. Similarly, subretinal and intraretinal fluid—when present—should be carefully monitored, as their resolution may indicate improved retinal barrier function and treatment response.

    PBM Treatment Monitoring Protocol 

    1. Patient Selection and Baseline Assessment

    Appropriate patient selection is crucial for optimizing the outcomes of photobiomodulation (PBM) therapy in the progression of dry AMD. Current evidence supports its use primarily in non-exudative retinal diseases, particularly intermediate Age-related Macular Degeneration, as well as emerging applications in Diabetic Retinopathy and Diabetic Macular Edema.

    Inclusion considerations:

    • Intermediate AMD (presence of drusen and/or subretinal drusenoid deposits)
    • Stable retinal conditions without active neovascularization
    • Best-corrected visual acuity (BCVA) is sufficient for functional monitoring

    Exclusion criteria:

    • Active neovascular AMD or significant intraretinal/subretinal fluid
    • Recent anti-VEGF injections (unless PBM is used adjunctively in controlled settings)
    • Significant media opacity affecting light delivery or imaging quality

    Baseline evaluation should include:

    • Visual function testing: BCVA, contrast sensitivity
    • Structural imaging: spectral-domain OCT (mandatory)
    • Optional advanced imaging: OCT angiography (OCTA) for vascular assessment
    • Key OCT biomarkers (baseline reference):
      • Central retinal thickness (CRT)
      • Ellipsoid zone (EZ) integrity
      • Retinal pigment epithelium (RPE) status and drusen volume
      • Presence of hyperreflective foci (HRF)

    Establishing a robust baseline is essential, as PBM-induced changes are often gradual and require longitudinal comparison.

    2. Treatment Session Procedure

    PBM is delivered using low-level light in the red-to-near-infrared spectrum (typically ~600–1000 nm), most commonly via LED-based systems designed for retinal applications.

    Standard session workflow:

    1. Patient preparation
      • No pharmacologic dilation is typically required (device-dependent)
      • Proper alignment and fixation ensured
    2. Device application
      • Light delivered trans-pupillary using controlled, non-thermal energy
      • Multi-wavelength protocols (e.g., combinations of ~590 nm, 660 nm, 850 nm) are commonly used in clinical studies
    3. Treatment duration
      • Typically, a few minutes per eye per session (device-specific)
      • Sequential or simultaneous bilateral treatment, depending on the system
    4. Safety monitoring
      • PBM is non-invasive and well-tolerated
      • No significant adverse retinal effects have been reported in the current literature
      • Monitor for discomfort or visual disturbances

    1. Immediate post-session:
    • No recovery time required
    • Patients resume normal activities immediately

    The mechanism of action—enhancing mitochondrial activity and reducing oxidative stress—does not produce immediate anatomical changes, reinforcing the need for structured follow-up.

    3. Treatment Series and Frequency

    PBM is not a single-session therapy for observing dry AMD progression or any other condition, but is administered as a treatment series, followed by monitoring and potential retreatment cycles.

    Typical treatment regimen (based on clinical studies):

    • Induction phase:
      • 2–3 sessions per week
      • Duration: 3–5 weeks
      • Total: ~9–12 sessions per cycle
    • Follow-up period:
      • Reassessment at 1–3 months post-treatment
      • OCT imaging to evaluate structural response
    • Retreatment strategy:
      • Repeat cycles every 4–6 months, depending on disease progression and response
      • Individualized based on OCT biomarkers and functional outcomes

    Monitoring during and after therapy:

    • Short-term (during treatment):
      • Limited structural change expected
    • Intermediate-term (1–3 months):
      • Possible reduction in drusen volume
      • Stabilization of EZ and RPE integrity
      • Decrease in HRF (inflammatory markers)
    • Long-term:
      • Disease stabilization rather than reversal is the primary goal

    Outcome measures:

    • Functional: BCVA, contrast sensitivity
    • Structural: OCT biomarkers (drusen, EZ, CRT, HRF)
    • Optional: OCTA vascular parameters

    A structured PBM protocol integrates careful patient selection, standardized treatment delivery, and longitudinal OCT-based monitoring. The therapy is best suited for chronic, non-exudative retinal conditions, where its cumulative biological effects—rather than immediate anatomical changes—drive clinical benefit. Consistent imaging and biomarker tracking are essential for guiding retreatment decisions and evaluating long-term efficacy. 

     Clinical Application and Results

    The integration of digital technologies and artificial intelligence (AI) into retinal imaging has significantly enhanced the ability to monitor treatment response in photobiomodulation (PBM) therapy. Given that PBM induces gradual, often subtle structural and functional changes, advanced analytical tools are essential for detecting and quantifying these effects with precision and reproducibility. Here are some real cases of application in Ophthalmology (Dry AMD, DME, etc).

    Dry AMD case

    Photobiomodulation (PBM) has been clinically evaluated primarily in patients with early-to-intermediate Age-related Macular Degeneration, where no widely accepted disease-modifying therapy exists. The most robust evidence comes from the LIGHTSITE clinical trial program, in which PBM is delivered as multiwavelength light therapy (590, 660, and 850 nm) in repeated treatment cycles (typically 9 sessions over 3–5 weeks, repeated every 4 months). 

    Across the LIGHTSITE I–III studies, PBM has consistently demonstrated functional improvements, particularly in best-corrected visual acuity (BCVA) and contrast sensitivity, and has shown favorable safety outcomes, with no evidence of phototoxicity. In LIGHTSITE II, PBM-treated eyes showed a mean ~4-letter gain in BCVA at 9 months, with approximately one-third of patients achieving ≥5-letter improvement, while sham-treated eyes showed minimal change. Earlier studies also reported improvements in contrast sensitivity, microperimetry, and reductions in drusen burden, suggesting both functional and anatomical benefits  .

    More recent data from the pivotal LIGHTSITE III trial further support these findings, demonstrating statistically significant gains in visual acuity compared with sham treatment, with mean improvements exceeding 5 letters and a substantial proportion of patients achieving clinically meaningful gains. At 24 months, PBM-treated eyes showed sustained visual improvement (+6.2 letters) and a reduced progression to geographic atrophy (6.8% vs 24.0% in controls), suggesting potential disease-modifying effects.

    However, despite these encouraging results, meta-analyses indicate that overall effect sizes remain modest and that variability across studies, small sample sizes, and protocol heterogeneity limit definitive conclusions regarding long-term clinical benefit. Thus, while PBM represents a promising and biologically plausible therapy for dry AMD, its role in routine clinical practice continues to evolve, with ongoing studies needed to confirm durability, optimal patient selection, and real-world effectiveness.

    DME 

    Photobiomodulation (PBM) has been explored as a non-invasive adjunctive or alternative therapy for Diabetic Macular Edema, targeting key pathogenic mechanisms, including mitochondrial dysfunction, oxidative stress, and chronic inflammation.

    Unlike anti-VEGF therapy, which primarily addresses vascular permeability, PBM aims to restore retinal metabolic balance through light-induced activation of mitochondrial pathways. Early clinical studies using red-to-near-infrared wavelengths (typically ~630–850 nm) have demonstrated reductions in central retinal thickness (CRT) and improvements in retinal morphology on OCT, alongside stabilization or modest gains in best-corrected visual acuity (BCVA). These effects are particularly notable in mild-to-moderate DME and in patients with non-center-involving edema, where PBM may reduce inflammatory signaling and improve fluid homeostasis.

    Clinical data, although still limited compared to age-related macular degeneration, suggest that PBM may have value as an adjunct to standard of care, potentially reducing treatment burden in patients requiring repeated intravitreal injections. Some studies report decreased intraretinal fluid and improvement in OCT biomarkers following PBM treatment cycles, with a favorable safety profile and no evidence of retinal damage.

     However, results remain heterogeneous, with variability in treatment protocols, patient populations, and outcome measures. Importantly, PBM has not yet demonstrated efficacy comparable to anti-VEGF therapy in center-involving DME, and its role is best considered complementary rather than substitutive at this stage. Larger randomized controlled trials are needed to define optimal dosing strategies, identify responder phenotypes, and clarify long-term functional benefits in DME management.

    Why is OCT Critical for PBM Monitoring?

    Optical Coherence Tomography (OCT) is essential for monitoring photobiomodulation (PBM) therapy because PBM aims to induce subtle, progressive structural and functional changes in the retina—especially in conditions like dry AMD progression. 

    Unlike anti-VEGF treatments, where effects can be more immediate, PBM outcomes are gradual and microstructural, such as changes in drusen volume, RPE integrity, outer retinal layers, and choriocapillaris perfusion. These changes are often invisible on fundus photography or visual acuity alone, making OCT the only practical tool for objective, layer-by-layer tracking over time. 

    Serial OCT scans allow clinicians to detect early signals of response (e.g., drusen regression or stabilization) and differentiate them from natural disease progression, which is critical for validating PBM efficacy in real-world practice.

    AI-Assisted OCT Analysis

    Artificial intelligence–driven analysis of optical coherence tomography (OCT) enables automated, quantitative assessment of retinal biomarkers critical to PBM monitoring. Machine learning and deep learning algorithms can segment retinal layers and identify pathological features such as drusen, hyperreflective foci (HRF), and fluid compartments with high accuracy.

    In the context of PBM, AI tools provide:

    • Automated retinal layer segmentation, including ellipsoid zone (EZ) and retinal pigment epithelium (RPE)
    • Quantification of drusen volume and distribution, particularly relevant in Age-related Macular Degeneration
    • Detection and tracking of subtle structural changes over time that may not be apparent on qualitative review

    These capabilities are especially important because PBM effects are often incremental rather than dramatic, requiring sensitive longitudinal comparison.

    Longitudinal Tracking and Progression Analysis

    Digital platforms enable time-series analysis of OCT data, allowing clinicians to monitor the disease trajectory before, during, and after PBM therapy. Automated registration of sequential scans ensures that the same retinal locations are compared over time.

    Key advantages include:

    • Change detection algorithms for early identification of treatment response
    • Trend analysis of biomarkers such as central retinal thickness, EZ integrity, and HRF density
    • Objective progression metrics, reducing inter-observer variability

    Such tools are critical for distinguishing true therapeutic effects from natural fluctuations in disease, particularly in slowly progressing conditions like dry AMD progression.

    Digital tools and AI are transforming PBM monitoring by enabling precise, quantitative, and longitudinal assessment of retinal biomarkers. From automated OCT analysis to AI-driven decision support and remote monitoring, these technologies address the key challenge of PBM therapy—detecting subtle, progressive changes over time, like in dry AMD progression, etc. Their continued development will be essential for standardizing PBM protocols and optimizing patient outcomes in retinal disease management.

    Altris for PBM monitoring on OCT: Drusen Progression +40 biomarkers for Research Purposes

    Altris has contributed to PBM monitoring on OCT: Drusen progression, as well as 40+ other biomarkers and 30+ pathologies, which may be monitored with the system. Altris enhances this process by turning OCT into a quantitative, standardized monitoring system rather than a subjective review. It automatically segments retinal layers and biomarkers (e.g., drusen, hyperreflective foci, fluid), calculates precise volumetric metrics, and enables longitudinal comparison across visits with high reproducibility.

    How does Altris assist with the monitoring of the main biomarkers of PBM therapy?

    Drusen 

    Detection: ✔️

    Quantification: ✔️ (area, volume, thickness)

    Tracking over time: ✔️

    This is one of the strongest use cases, which is critical for effective PBM monitoring.

    Central Retinal Thickness (CRT)

    Detection: ✔️

    Standard ETDRS maps: ✔️

    Longitudinal tracking: ✔️

    Outer retinal layers (EZ, ONL, etc.)

    Segmentation: ✔️ (layer-based)

    Quantification: ✔️ (thickness, integrity)

    Disruption detection: ✔️

    This is very important for PBM response and detection of early functional damage.

    monitoring

    RPE (Retinal Pigment Epithelium)

    Detection (layer): ✔️  

    Elevation (drusen): ✔️

    Atrophy signs: ✔️ (via hypertransmission, thinning)

    Important for: drusen interpretation and GA risk, though, not always a standalone numeric biomarker.

    Hyperreflective foci (HRF)

    Detection: ✔️

    Localization: ✔️

    Counting / burden tracking: ✔️

    These are high-value biomarkers for progression risk in PBM monitoring.

    Assistance like this allows clinicians to track PBM response objectively, identify responders vs non-responders earlier, and generate consistent reports for clinical decision-making or research. In short, while OCT provides the necessary imaging depth, Altris unlocks its full value for PBM by making subtle retinal changes measurable, comparable, and clinically actionable.

    Conclusion

    PBM represents a novel and biologically plausible therapeutic modality that targets key pathological mechanisms in retinal disease. By enhancing mitochondrial function, reducing oxidative stress, and modulating inflammation, PBM holds significant potential to complement existing treatment strategies and improve outcomes in retinal disease management. However, further research is required to fully define its role in routine clinical practice.

    Despite the promising findings, the clinical integration of PBM remains in an evolving stage. Variability in treatment parameters—including wavelength, dose, and treatment protocols—has limited standardization and comparability across studies.  Moreover, much of the current evidence is derived from small-scale clinical trials and preclinical models, underscoring the need for large, randomized controlled trials to establish optimal treatment regimens for dry AMD progression and to assess long-term efficacy in other eye pathologies. 

    In this context, OCT—especially when enhanced with AI-driven analysis—plays a critical role in advancing PBM adoption. Quantitative OCT biomarkers such as drusen volume, outer retinal integrity, and subtle structural changes provide objective endpoints for assessing therapeutic response. AI-based platforms further enable precise, reproducible, and longitudinal analysis of these changes, helping to standardize evaluation, identify responders earlier, and strengthen the clinical evidence base for PBM.

    FAQ Section

    1. How do I objectively measure response to PBM therapy?

    Clinicians look for quantifiable OCT biomarkers, not just visual acuity:

    • Drusen volume (regression or stabilization)
    • Outer retinal layer integrity (EZ, RPE)
    • Hypertransmission / atrophy areas

    The challenge: changes are subtle → require precise, longitudinal OCT comparison.

    2. Which OCT biomarkers are most relevant for PBM monitoring?

    The most discussed and clinically relevant:

    • Drusen volume/area
    • RPE atrophy  
    • Hypertransmission
    • Ellipsoid Zone (EZ) disruption/loss
    • Hyperreflective foci (secondary)

     For GA specifically:

    Overlap of RPE atrophy + hypertransmission + EZ loss = key composite metric.

    3. How often should I monitor patients on PBM?

    Typical real-world patterns:

    • Baseline OCT before starting PBM
    • Follow up every 3–6 months
    • More frequent (monthly) in studies.

    4. How do I distinguish PBM effect from natural AMD progression?

    Distinguishing the effect of PBM from the natural progression of AMD remains one of the key clinical challenges. AMD typically progresses slowly and can show natural fluctuations, while PBM-related changes tend to be gradual and relatively modest. To differentiate between the two, clinicians rely on consistent OCT metrics tracked over time, comparing trends rather than single visits. Bilateral analysis—evaluating treated versus untreated eyes—can provide additional context, while assessing the rate of change, such as slowing of drusen growth or stabilization of atrophic areas, helps determine whether observed changes are likely treatment-related rather than part of the disease’s natural course.

    5. Do I need AI/software for PBM monitoring, or is manual OCT review enough?

    Whether AI/software is needed for PBM monitoring versus manual OCT review is an increasingly important question in clinical practice. While manual assessment can provide a general, qualitative understanding, it is often variable, time-consuming, and limited in its ability to detect subtle changes. PBM, however, requires identification of micron-level structural differences and high reproducibility across visits to accurately assess treatment response. AI-based OCT analysis addresses these challenges by enabling automated segmentation of key biomarkers, delivering precise volumetric measurements, and supporting reliable longitudinal tracking in standardized units such as mm², mm³, and percentage change. This level of consistency also helps clinicians more confidently distinguish responders from non-responders, making monitoring more objective and clinically actionable.

    References:

    https://pmc.ncbi.nlm.nih.gov/articles/PMC11488463/

    https://link.springer.com/article/10.1007/s40135-025-00340-x

    https://www.ophthalmologytimes.com/view/photobiomodulation-shows-the-power-of-light#:~:text=PBM%20is%20performed%20through%20a,6%2Dmonth%20follow%2Dup

    https://link.springer.com/article/10.1007/s40135-025-00340-x#:~:text=Purpose%20of%20review,wavelengths%20used%20and%20treatment%20protocols

    https://d-nb.info/136218389X/34

    https://www.frontiersin.org/journals/ophthalmology/articles/10.3389/fopht.2024.1388602/full

    https://retinatoday.com/articles/2020-may-june/photobiomodulation-as-a-treatment-in-dry-amd 

    https://lumithera.com/

    https://espansionegroup.it/it/

     

     

  • Geographic Atrophy Retina OCT Biomarkers: Detection, Quantification, and Monitoring

    GA altris ims
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min.

    Introduction. Overview of Geographic Atrophy (GA) as a Late Stage of Dry AMD

    Geographic atrophy (GA) is a chronic progressive retinal degeneration that represents part of the late stage of age-related macular degeneration (AMD). It is characterized by gradual and irreversible atrophy of photoreceptors, the retinal pigment epithelium (RPE), and the choriocapillaris. As a result, a persistent defect of neurosensory tissue develops, which clinically manifests as central vision loss, the appearance of central scotomas, and reduced contrast sensitivity.

    Atrophic lesions typically originate in the outer retinal layers and gradually expand, involving the macula and fovea. Over time, this leads to irreversible visual impairment and a significant decline in quality of life. In the early stages, patients may not experience noticeable changes in visual acuity. However, involvement of the central foveal region may lead to a sudden functional deterioration, with patients reporting difficulties in reading, recognizing faces, and working with fine details.

    GA is considered one of the leading causes of clinically significant central blindness among people over the age of 60 in developed countries. With the aging population, the prevalence of this condition continues to increase, creating a substantial social and economic burden. In addition to reduced visual acuity, GA significantly affects patients’ quality of life.

    Geographic atrophy retina OCT, together with modern digital image analysis algorithms, has become a key tool in the diagnosis, monitoring, and evaluation of OCT biomarkers predicting GA progression. OCT provides cross-sectional imaging of the retina with microscopic resolution, enabling detailed assessment of individual retinal structures—from the inner retinal layers to the RPE–Bruch’s membrane–choriocapillaris complex. This technology has enabled the transition from subjective ophthalmoscopic assessment to objective structural analysis.

    The advantages of OCT in the diagnosis and monitoring of GA include its non-invasive nature, high reproducibility, ability to detect early structural changes, and accurate quantitative measurements. Structural alterations at the level of photoreceptors and the RPE often occur long before they become visible on ophthalmoscopy or fundus photography. Proper recognition of OCT biomarkers of GA is essential not only for disease diagnosis but also for personalizing treatment strategies, predicting the risk of progression, and evaluating therapeutic outcomes.

    The purpose of this article is to summarize current scientific evidence on OCT biomarkers of geographic atrophy, including their morphological definition, quantitative parameters, prognostic significance, and role in monitoring disease progression. Particular attention will be given to the practical aspects of OCT in clinical practice, interpretation of longitudinal changes, and effective communication with patients regarding the expected course of the disease.

    2. Main OCT Biomarkers of Geographic Atrophy

    Modern understanding of GA morphology has been largely shaped by the work of international expert groups, particularly the Classification of Atrophy Meetings (CAM) Group. The CAM group proposed standardized terminology and clear OCT-based criteria for retinal atrophy, enabling harmonization of diagnostic approaches in both clinical practice and multicenter studies.

    The CAM group recommends spectral-domain OCT (SD-OCT) as the preferred imaging modality for detecting GA-related changes, as it allows identification of the earliest signs of developing atrophy.

    2.1 OCT Signs of Geographic Atrophy

    The following three features form the basis for the standardized OCT definition of GA:

    • Loss of the outer retina
    • Loss of the retinal pigment epithelium (RPE) ≥250 µm in diameter
    • Choroidal hypertransmission ≥250 µm in diameter

    1. Loss of the Outer Retinal Layers

    On OCT B-scans this manifests as:

    • disruption or loss of the ellipsoid zone (EZ)
    • absence of the interdigitation zone
    •  thinning or complete loss of the outer nuclear layer (ONL)
    • thinning (atrophic changes) of the neuroepithelium above the lesion

    This feature reflects the loss of photoreceptors, which are the primary functional elements responsible for central vision.

    2. Loss of the Retinal Pigment Epithelium (RPE)

    The CAM group established a threshold of 250 µm in the largest horizontal dimension to define clinically significant atrophy.

    AI detection of RPE atrophy OCT appears as:

    • absence or severe thinning of the hyperreflective RPE band
    • a well-defined border between preserved and atrophic RPE

    3. Choroidal Hypertransmission

    Due to the loss of the RPE, light penetrates more deeply into the underlying layers, resulting in increased visualization of the choroid.

    On OCT this appears as:

    • Increased visibility of the choriocapillaris layer
    • Clear correspondence with the area of RPE defect

    Classification of Outer Retinal Atrophy Associated with AMD

    • Complete RPE and outer retinal atrophy (cRORA)
    •  Incomplete RPE and outer retinal atrophy (iRORA)
    • Complete outer retinal atrophy (cORA)
    • Incomplete outer retinal atrophy (iORA)

    2.2 OCT Parameters for Monitoring Geographic Atrophy

    Once the diagnosis is established, OCT biomarkers predicting GA progression and  quantitative monitoring of disease progression becomes critical. 

    1. Morphological Triad

    RPE atrophy, choroidal hypertransmission, and neuroepithelial atrophy represent the hallmarks of complete retinal atrophy.

    This triad defines retinal atrophy within the lesion area and allows differentiation between complete and incomplete atrophy using structural criteria.

    2. Area of Geographic Atrophy (mm²)

    Quantitative measurement of GA area is a key parameter in both clinical practice and research.

    OCT segmentation enables highly reproducible calculation of the affected area. Modern OCT systems allow:

    • automatic segmentation of atrophy boundaries
    • calculation of the GA area in mm²
    • comparison of measurements between visits

    The annual enlargement rate of the GA area is an objective marker of disease progression and correlates with functional visual outcomes. Importantly, the GA area may increase even when visual acuity remains stable.

    The area of GA served as the primary endpoint in clinical trials evaluating the intravitreal therapies Syfovre and Izervay, which were recently approved by the FDA as treatments to slow GA lesion growth.

    AI-based algorithms further improve the precision and reproducibility of measurements, which is particularly important for long-term monitoring.

    Modern OCT systems provide GA area measurements in mm², and comparisons between visits provide an objective measure of disease dynamics. Even when patients do not perceive changes, increasing lesion area confirms disease progression.

    3. Distance Between GA Lesions and the Fovea

    An important quantitative parameter is the distance between the foveal center and the nearest border of the atrophic lesion.

    This parameter has direct functional significance. Decreasing distance over time correlates with declining visual function: the closer GA approaches the fovea, the higher the risk of sudden vision loss.

    Patients with GA lesions approaching the fovea have a poorer prognosis and often require more intensive monitoring and therapeutic interventions.

    This parameter also allows objective risk prediction and supports:

    • early referral to specialized ophthalmology centers
    • discussion of potential vision loss with patients

    2.3 Predictors of GA Development and Progression

    GA frequently develops as a consequence of drusen involution or structural alterations of the RPE.

    GA lesions in AMD may arise in association with:

    • certain drusen types (large or confluent drusen, reticular pseudodrusen)
    • previous choroidal neovascularization
    • RPE detachment or RPE tear
    • refractile deposits
    • vitelliform lesions

    Geographic Atrophy as a Result of Drusen Involution

    Drusen are localized accumulations of pathological material (photoreceptor metabolic by-products) between the RPE and Bruch’s membrane. They may change in number, size, and morphology. 

    Regression or disappearance of drusen, as well as structural changes observed on OCT, represent predictors of progression to GA. Regular monitoring allows early detection of potentially dangerous changes.

    Types of Drusen

    drusen involution

    1. Hard Drusen

    • round, well-defined yellow-white lesions
    • diameter ≤63 µm
    • usually asymptomatic

    2. Soft Drusen

    • medium: 63–125 µm
    • large: >125 µm
    • poorly defined borders
    • may enlarge and merge
    • associated with diffuse retinal dysfunction

    3. Confluent Drusen

    Formed by the merging of several soft drusen. 

    4. Drusenoid RPE Detachment

    An area of confluent drusen in the macula with a diameter exceeding 350 µm according to AREDS.

    5. Cuticular Drusen

    • located between the RPE and Bruch’s membrane
    • small in diameter but numerous
    • often confluent
    • steep, sloping sides (“saw-tooth” appearance)
    • may disrupt RPE structure
    • represent a risk factor for progression to GA

    6. Reticular Pseudodrusen

    • deposits located in the subretinal space between photoreceptors and the RPE
    • associated with poor visual prognosis
    • strongly linked with GA development

    GA develops particularly rapidly in the presence of reticular pseudodrusen.

    Predictors of GA Development in Eyes with Drusen

    • large numbers of drusen, particularly in the central macula
    • regression of drusen
    • structural changes such as heterogeneous internal reflectivity

    These predictors help identify patients at high risk for GA development and are valuable for optimizing monitoring intervals and potential preventive strategies.

    Another predictor of faster GA lesion formation is hyperreflective foci. These are small intraretinal hyperreflective dots, often located above drusen and typically associated with local disruption of the RPE structure. They likely represent migrating RPE cells and activated microglia. A tiny blue spot is a hyperreflective foci area detected by Altris  automated GA segmentation OCT:

    foci

    Their presence significantly increases the risk of GA development within the next few years (in some studies up to five-fold within two years).

    Clinical Importance of Predictors

    Identifying high-risk patients allows clinicians to:

    • individualize OCT monitoring frequency
    • initiate treatment earlier
    • predict functional vision loss
    • discuss expected disease progression with patients in a timely manner.

    Management of Geographic Atrophy and Patient Education

    Management of patients with GA today extends far beyond simple observation. It involves an active, structured strategy that combines regular OCT monitoring, timely initiation of therapy, risk-factor modification, and comprehensive patient education.

    The main goal is to slow disease progression and reduce the rate of atrophy expansion while preserving the central fovea for as long as possible. GA Progression quantified via Altris:

    ga area quantification

    The Role of OCT

    Effective GA management is impossible without high-quality OCT monitoring.

    OCT enables clinicians to:

    • quantify the area of atrophy
    • determine the rate of lesion expansion
    • measure the distance to the fovea
    • analyze outer retinal layer integrity
    • identify predictors of rapid progression

    Monitoring is recommended every 3–6 months, and when intravitreal therapy is used, OCT should be performed before each injection to assess disease activity and lesion growth rate.

    OCT also serves as a powerful motivational tool: showing patients the dynamics of structural changes helps them better understand the need for treatment and regular follow-up visits.

    Patients should be informed that GA may progress without sudden visual deterioration. Structural OCT changes often precede functional vision loss, making regular examinations essential even when visual acuity appears stable.

    Current Treatment Options

    ga therapies

    Intravitreal Therapy

    • Izervay (avacincaptad pegol)
    • Syfovre (pegcetacoplan)

    For the first time in decades, FDA-approved treatments are available that slow the expansion rate of GA lesions. Although they do not restore lost vision, slowing visual decline is an important clinical goal.

    Patients should clearly understand that treatment slows progression but does not restore vision. Proper expectation management improves treatment adherence and reduces disappointment.

    Nutritional Supplements

    Formulations based on AREDS / AREDS2 have been shown to reduce the risk of progression from intermediate AMD to advanced stages.

    Patients should be informed that these supplements do not treat GA, but may have preventive value at earlier stages.

    What Patients Must Understand

    1. Progressive Nature of GA

    GA is a chronic progressive disease. The area of atrophy almost always increases over time. The rate of progression varies depending on morphological characteristics.

    Patients should understand that treatment aims to slow, not completely stop, disease progression.

    2. Importance of Lifestyle

    Although lifestyle modification has limited influence once GA is established, recommendations remain relevant:

    • smoking cessation
    • blood pressure and lipid control
    • antioxidant-rich diet and omega-3 fatty acids
    • regular physical activity

    These factors improve overall vascular health and may reduce systemic inflammation.

    3. Psychological Adaptation

    Progressive central vision loss often leads to anxiety, fear of blindness, and reduced social activity.

    It is important to discuss:

    • low-vision aids (magnifiers, telescopic glasses, electronic magnifiers)
    • support resources for people with low vision

    Psychological support significantly improves adaptation and quality of life.

    Patient Partnership: The Foundation of Success

    Modern management of dry AMD is no longer hopeless. With approved therapies and evidence-based preventive strategies, clinicians can meaningfully influence the rate of disease progression.

    However, the effectiveness of any strategy depends on collaboration between the physician and the patient.

    Patient education regarding:

    • the nature of the disease
    • the role of regular OCT monitoring
    • treatment possibilities and limitations
    •  the importance of lifestyle modification

    is an essential component of modern GA management.

    FAQs

    Which OCT biomarkers are predictive of GA progression to look for?

    Key biomarkers include hypertransmission defects, RPE atrophy, photoreceptor loss, ellipsoid zone disruption, hyperreflective foci, and reticular pseudodrusen. These structural changes are strongly associated with GA development and progression in AMD. 

    How does AI OCT help prioritize patients at risk of GA progression?

    AI for GA systems identifies high-risk biomarkers and calculates progression rates, enabling clinicians to triage patients for closer monitoring or treatment.

    Can AI detect multiple retinal pathologies in addition to GA?

    Many platforms detect 70+ retinal pathologies and biomarkers simultaneously on OCT scans. Altris detects and quantifies 40+ retina biomarkers and 40+ pathologies. 

    How can AI quantify geographic atrophy on OCT scans?

    AI algorithms automatically segment GA lesions and calculate lesion area, retinal layer loss, and biomarker overlap, providing objective measurements in millimeters or percentages.

    Can AI OCT support treatment decisions for GA therapies?

    AI can measure structural parameters such as EZ loss or RPE integrity, which may help evaluate treatment response or disease activity. Altris applies Flags to filter out the eligible patients then.

    Can AI detect early GA before it becomes clinically visible?

    Yes. AI models can identify subtle structural abnormalities on OCT, such as EZ disruption or early hypertransmission, enabling earlier detection of atrophy.

    Which OCT metrics should be monitored to track GA progression?

    Clinically relevant metrics include: GA lesion area (mm²), rate of lesion growth, distance from lesion margin to the fovea, percentage of macular involvement. AI can automatically calculate and track these parameters over time.

    How to efficiently measure geographic atrophy on OCT?

    To efficiently measure Geographic Atrophy on Optical Coherence Tomography (OCT), clinicians should identify key biomarkers such as RPE loss, outer retinal thinning, and choroidal hypertransmission, then quantify the atrophy area (mm²) using en-face OCT or automated segmentation tools. Tracking lesion size and its distance to the fovea over time allows accurate monitoring of disease progression. AI-assisted OCT platforms can automate detection and measurements, making longitudinal assessment faster and more consistent.

    References

    1. Guymer RH, Rosenfeld PJ, Curcio CA, et al.
      Incomplete retinal pigment epithelium and outer retinal atrophy in age-related macular degeneration: Classification of Atrophy Meeting report.
      Ophthalmology.
      Available at: https://pubmed.ncbi.nlm.nih.gov/38387826/
    2. Natural history and progression of geographic atrophy in AMD.
      ScienceDirect.
      Available at: https://www.sciencedirect.com/science/article/pii/S2468653023006681
    3. OCT Spotlight: Characterizing Geographic Atrophy Development and Progression.
      Retina Today.
      Available at: https://retinatoday.com/articles/2025-apr/oct-spotlight-characterizing-ga-development-and-progression
    4. Automated monitoring of geographic atrophy using OCT imaging.
      Scientific Reports.
      Available at: https://www.nature.com/articles/s41598-023-34139-2
    5. Classification of Atrophy Meeting (CAM) consensus for OCT-based atrophy classification in AMD.
      American Academy of Ophthalmology Journal.
      Available at: https://www.aaojournal.org/article/S0161-6420(17)31703-7/abstract
    6. Identifying Geographic Atrophy Biomarkers.
      Optometric Management.
      Available at: https://www.optometricmanagement.com/issues/2025/october/identifying-geographic-atrophy-biomarkers/
    7. FDA Approval Announcement for Izervay (avacincaptad pegol).
      Astellas Pharma Newsroom.
      Available at:
      https://newsroom.astellas.com/2023-08-05-Iveric-Bio-Receives-U-S-FDA-Approval-for-IZERVAY-TM-avacincaptad-pegol-intravitreal-solution-,-a-New-Treatment-for-Geographic-Atrophy

     

  • Altris AI Receives Health Canada Approval

    AI Ophthalmology and Optometry | Altris AI Altris Inc.

    Altris AI Receives Health Canada Approval, Reinforcing Its Position as a Globally Trusted AI Decision Support Platform for OCT Analysis

    Regulatory clearance marks a pivotal milestone in Altris AI’s international expansion and its mission to bring clinical-grade AI to eye care worldwide.

    10 March 2026 — Altris AI, a leading provider of AI-powered decision support for optical coherence tomography (OCT) analysis, today announced it has received approval from Health Canada | Santé Canada, Canada’s federal health regulatory authority. This clearance represents a significant step forward in the company’s global growth strategy and its commitment to meeting the highest standards of medical device safety and clinical reliability.

    For a company scaling across international markets, regulatory approvals are far more than administrative milestones — they are foundational growth enablers. Health Canada approval strengthens Altris AI’s international positioning, opens new pathways for future regulatory submissions across key markets, and delivers a clear signal to the global healthcare community: Altris AI is built for real-world clinical practice.

    The approval confirms that Altris AI’s clinical and technical validation withstands the rigorous scrutiny of Health Canada’s regulatory review process, which evaluated the platform across five critical dimensions: clinical evidence supporting efficacy and safety claims; risk management protocols; cybersecurity safeguards; quality management systems; and intended use claims. Each of these pillars reflects the standard that modern AI-driven medical devices must meet before being trusted in clinical settings.

    Altris AI’s platform serves optometrists, ophthalmologists, and pharmaceutical organizations by providing intelligent, reliable support for interpreting OCT scans — one of the most widely used diagnostic tools in eye care. By surfacing clinically relevant findings with speed and precision, Altris AI empowers clinicians to make more informed decisions, improve patient outcomes, and increase the efficiency of ophthalmic workflows.

    “This approval is external confirmation that our platform meets the standards required for medical-grade AI,” said  Maria Znamenska, Chief Medical Officer at Altris AI. “Health Canada’s review is thorough, evidence-based, and internationally respected. Receiving this clearance validates not just our technology, but the entire approach we’ve taken — building AI that clinicians can trust, in environments where accuracy is not optional.”

    The Health Canada clearance follows a broader regulatory strategy that positions Altris AI as a compliant, audit-ready platform for healthcare systems worldwide. As global regulators increasingly scrutinize AI in medical devices, early and consistent compliance across multiple jurisdictions will become a decisive competitive differentiator.

    Altris AI’s mission is to accelerate the transition of AI in eye care from innovation to infrastructure — not as a replacement for clinical expertise, but as a trusted decision-support layer that elevates the standard of care across every point of practice.

    About Altris AI Altris AI is an AI Decision Support platform for OCT analysis, designed to support optometry, ophthalmology, and pharma with clinically validated, regulatory-compliant technology. The company is committed to expanding access to intelligent eye care diagnostics globally.

  • AI in Optometry: 5 Real Applications

    ai in optometry
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    9 min.

    Highlights: AI in optometry is revolutionizing clinical decision-making by allowing eye care professionals to analyse B-scans with greater precision, consistency, and confidence right at the point of care.

    • AI for fundus analysis is another way to automatically evaluate fundus scans in an optimized way throughout all devices.
    •  AI-driven deep learning algorithms can detect and quantify retinal and optic nerve pathologies in OCT images, enabling earlier identification of diseases such as glaucoma, age-related macular degeneration (AMD), and other retinal conditions. 
    • AI-assisted analysis enhances clinical efficiency by helping clinicians triage scans, monitor disease progression over time, and focus on clinically significant findings, rather than relying solely on manual scan reviews. 
    • Other AI-powered tools provide objective visual insights for clinicians and patients, improving the accuracy of triage and treatment monitoring and enhancing patients’ understanding of retinal health.
    • AI enables optometrists to manage more complex ocular cases in primary care, facilitating earlier detection, risk stratification, and informed referral decisions based on objective insights.
    • AI chatbots in optometry inform about eye symptoms, guide whether to seek care (urgent vs. routine), suggest possible conditions, support patients and help decide next steps, etc. Or manage eye care specialists’ daily routine.

    Introduction 

    Artificial intelligence is increasingly shaping healthcare by enhancing clinicians’ ability to interpret complex medical data and make earlier, more informed decisions. AI in optometry is especially important in OCT imaging, where it is essential to correctly interpret subtle structural changes to identify eye disorders early.

    AI for optometrists can therefore more reliably and consistently detect and track retinal and optic nerve disorders by integrating deep learning into OCT data. In routine optometric practice, risk management and referral decision-making are enhanced by converting OCT images into understandable, actionable findings.

    Discover how optometry AI tools redefine optometry by improving diagnostic accuracy, clinical efficiency, and the quality of patient care in 5 real cases.

    1. AI Decision Support for OCT Analysis

    One of the most effective AI applications in optometry is AI decision support for OCT analysis. 

    AI decision-support systems are increasingly applied to OCT imaging to assist optometrists in interpreting complex retinal and optic nerve data. Here’s one of the real cases when AI brings ultimate use to practitioners:

    Thus, AI-powered platforms like Altris use deep learning algorithms to automatically detect and quantify structural changes, highlight areas of concern, and track progression over time via an AI OCT pathology detection tool.

    By analysing patterns across large datasets of retinal scans, the system can flag subtle abnormalities that may be difficult to identify manually, segment them automatically, and provide structured, visual insights that help clinicians make more informed, consistent decisions while monitoring patient eye health. It does everything eye care specialists do, but faster, error-free, and unbiased.

    In particular, Altris AI also applies deep learning to OCT scans to automatically detect and highlight complex retinal and optic nerve changes. 

    The system quantifies abnormalities, tracks progression over time, and provides visual insights that help optometrists interpret scans more accurately and consistently, again supporting far more informed clinical decisions.

    2. AI for Fundus Analysis

    AI for Fundus Analysis is another way to automatically evaluate fundus scans in an optimized way throughout all OCT devices. Among the top 3 AI software solutions for fundus imaging, there are:

    Auroraa AI (Optomed) is an advanced artificial intelligence platform integrated with Optomed’s handheld and tabletop fundus cameras, designed to detect multiple retinal abnormalities including diabetic retinopathy, glaucoma, and age‑related macular degeneration; it provides immediate, automated screening results to support clinicians and improve early disease detection. 

    Beammed’s AI‑powered fundus cameras  pair intelligent image analysis with portable retinal imaging to enable early detection of diabetic retinopathy and other retinal conditions, leveraging deep learning algorithms to highlight pathology and help streamline screening programs.

    Cybersight AI (Orbis) offers an AI‑driven diagnostic support tool focused on interpreting fundus images to assist eye care providers in low‑resource settings and telemedicine programs, combining machine learning with expert clinical guidance to improve access to retinal disease screening globally. Here’s how fundus tool may look like:

    Such systems provide severity scores, highlight areas of concern, and track changes over time, giving clinicians objective, reproducible insights to support their decisions. Since AI can detect early signs of conditions like glaucoma, diabetic retinopathy, and age-related macular degeneration, it often spots subtle changes that are hard to see with the naked eye. 

    3. AI for Automated Visit Scheduling

    AI‑powered appointment scheduling systems are digital tools that, alongside any modern AI OCT pathology detection tool or similar, use artificial intelligence — including natural language processing (NLP), predictive analytics, machine learning, and automated communication — to handle clinical scheduling tasks usually done manually by staff. These tools can:

    • let patients self‑schedule online, by chat, voice, or text at any time,
    •  automatically confirm, remind, reschedule, or cancel appointments,
    • optimize schedules based on provider availability and patient needs, and
    • predict and prevent clinic inefficiencies such as no‑shows. 

    In essence, the system acts as a digital receptionist and smart scheduler, integrating with clinic practice management software, EHRs, and CRM systems to manage workflows seamlessly. Best EHR systems include:

    best ehr

    Elation EHR

    Elation EHR is a cloud-based electronic health record designed primarily for independent primary care practices. It focuses on simplifying clinical workflows, documentation, and patient engagement, with strong charting tools and longitudinal patient records. It’s used to help physicians deliver personalized care while reducing administrative burden.

    Epic EHR

    Epic is one of the most widely used enterprise EHR systems globally, typically implemented by large hospitals and health systems. It integrates clinical, administrative, and billing functions into a single platform, supporting everything from patient records to population health management. It’s used to coordinate care at scale and improve interoperability across departments and organizations.

    Tebra EHR

    Tebra combines electronic health records with practice management, billing, and patient communication tools, targeting small to mid-sized medical practices. It streamlines front-office and clinical operations in one system, helping practices manage scheduling, documentation, and revenue cycle efficiently.

    Nextech

    Nextech EHR is a specialty-focused EHR designed for fields like ophthalmology, dermatology, and plastic surgery. It includes tailored templates, imaging integration, and workflow tools specific to these specialties. It’s used to enhance clinical efficiency and documentation accuracy in specialized practices.

    InSync EHR

    InSync EHR is a cloud-based EHR built for behavioral health and therapy practices. It supports telehealth, scheduling, documentation, and billing, with features tailored to mental health workflows. It’s used to improve care coordination and streamline operations for therapists and behavioral health providers.

    Oracle Health EHR

    These tools are designed to improve operational efficiency, reduce administrative burden, and enhance the patient journey in healthcare settings. They offer 24/7 availability, personalized booking flows, and real-time updates, making them a powerful part of your workflow.

    ModMed EMA

    ModMed EMA (Electronic Medical Assistant) is an AI-driven, specialty-specific EHR developed by Modernizing Medicine. It uses structured data and adaptive templates to support clinical decision-making and documentation. It’s used by specialists to increase efficiency, improve outcomes, and reduce time spent on manual data entry.

    “Up to 71% of U.S. hospitals now use predictive AI technologies for scheduling and related automation.”

     

    ai in optometry infographics

    Indeed, AI in optometry for appointments, self-scheduling, and other administrative tasks can optimize routine workflows and offer a range of benefits for both opticians and their visitors. They:

    • Remove Administrative Burden

    AI takes over repetitive scheduling tasks — booking, confirmations, cancellations, preregistration, and follow‑ups — freeing front‑desk staff and nurses to focus on patient care rather than paperwork. 

     Historically, nurse managers and front‑desk staff can spend up to 40% of their day on scheduling tasks, and automating this saves hours of labour. 

    • Provide 24/7 Booking & Patient Flexibility

    Patients no longer have to call during office hours. AI scheduling tools enable self‑service booking, changes, and cancellations at any time via chatbots, voice interfaces, or online portals. 

    This has become especially crucial, as 40% of healthcare appointments are requested outside normal business hours — times when traditional phone booking is impossible. 

    • Reduce No‑Shows & Better Attendance

    Automated reminders — via SMS, email, or voice — consistently reduce no‑show rates, which can otherwise waste clinic time and revenue. Clinics using these tools have reported:

    • No‑shows dropped from 20% to as low as 7% with automated reminders.

    • In some settings, AI virtual assistants reduced missed visits by up to ~73%. 

    AI can also predict patients likely to skip visits and prompt engagement before issues arise. 

    • Enhance Resource Use

    Instead of manually guessing who should be booked and where, such AI in optometry:

    ✔ matches patients with the right clinician,

    ✔ avoids double bookings and schedule conflicts,

    ✔ spreads appointments evenly to reduce bottlenecks, and

    ✔ improves utilization of staff time and rooms. 

    AI scheduling can increase provider utilization by 15–25% and reduce wait times by 15–30%. 

    • Improve Patient Experience

    Patients appreciate convenience. Access to online or chatbot booking correlates with improved satisfaction — one study showed satisfaction scores can rise by over 20% when patients can self‑manage appointments. 

    AI also reduces inbound phone volume by 25–40%, allowing clinics to serve patients more efficiently. 

    For instance, Tele-optometry decision support that offers

    • AI pre-analyses remote exams
    • Flags cases requiring in-person referral
    • Supports non-specialist reviewers

    has the following workflow impact:

    • Scales remote care
    • Consistent quality
    •  Faster review cycles

    Used in:

    • Large optometry chains
    • Retail vision centres
    • Franchise-based practices
    • Rural clinics
    • Community health centres
    • Mobile eye clinics, etc.

    4. AI Workflow & Practice Optimization 

    So, AI-assisted OCT analysis has become helpful in retina & glaucoma screening, in follow-ups, progression tracking, and other workflows. Meaning, what happens with OCT with the help of an AI OCT pathology detection tool is helpful in many ways:

    • OCT scans are automatically analysed
    • Pathologies are flagged (fluid, thinning, progression risk, etc.)
    • Clinicians review AI output before raw scans.

    So, they get 

    • Faster exam review
    • Fewer missed subtle findings
    • Consistent interpretation across doctors, etc.

    But real-world AI applications in optometry for clinic management do not stop there: scheduling, reminders, patient triage, administrative automation, analytics, and beyond may also be supported by specific optometry AI tools. Here are a few examples.

    AI triage tools for urgent eye issues

    • AI pre-screens OCT/fundus images
    • Exams are auto-prioritized by severity
    • High-risk patients are flagged before the visit

    Workflow impact:

    • Smarter scheduling
    • Faster routing to specialists
    • Less cognitive load on staff

    Used in:

    • High-volume optometry chains
    • Tele-optometry services

    An example of AI diagnostic software for optometry and ophthalmology can be IDx-DR AI Diagnostic System for Detecting Diabetic Retinopathy.

    Automated appointment reminders 

    Automated appointment reminders are AI- or rules-based systems, not yet independent chatbots or AI assistants, that automatically notify patients about upcoming eye exams via SMS, email, WhatsApp, or voice calls, without staff involvement.

    They usually trigger:

    • 7 days before (prep + reschedule window)
    • 48–72 hours before (confirmation)
    • 24 hours or same day (final reminder)

    which makes them still a new generation of automation tools for eye care. Well-designed systems, like the majority of AI in optometry tools, support:

    • HIPAA / GDPR-compliant messaging
    • No clinical advice in reminders
    • Audit logs of sent communications
    • Opt-out controls

    This makes them safe for both routine and medical eye care appointments.

    For example: EyeCloudPro 

    But why do no-shows happen in optometry? Like in any other service sphere, there are real reasons why: 

    • Routine exams feel “non-urgent.”
    • Long booking lead times (2–6 weeks)
    • Patients forget dilation/prep requirements
    • Elderly patients miss calls or misremember dates
    • Parents forget pediatric appointments
    • No easy way to confirm or reschedule

    Across outpatient care (including optometry), automated reminders typically achieve:

    • 20–40% reduction in no-shows
    • 5–10% increase in appointment confirmations
    • Up to 30% fewer last-minute cancellations
    • 1–2 hours/day staff time saved (no manual calls)

    In optometry specifically, clinics with long routine exam cycles often see results closer to the upper end of those ranges. What automated reminders do here is directly target forgetfulness + reduce friction. No ordinary staff can do that to such an extent. But AI can.

    Therefore, by keeping patients aware, ready, and involved, automated appointment reminders help optometry clinics reduce no-show rates. Practices may increase patient flow, maximize chair time, and enhance attendance by providing timely, customized notifications through preferred channels—all without adding to the administrative burden.


    Patient communication and optometrists’ education apps

    Patient communication, as well as optometrist education systems and applications, for mobile and desktop usage, support clearer understanding, better engagement, and more consistent care delivery. 

    For example: Chatbots in Healthcare from Capacity

    capacity

    These tools help patients understand their eye health and treatment plans, while enabling optometrists to stay informed through structured education, clinical updates, and decision-support resources—improving outcomes without increasing chair time:

    • AI translates OCT findings into plain language
    • Visual overlays show “what changed” and “why it matters.”
    • Used chairside or via patient portal

    Workflow impact:

    • Better patient understanding
    • Higher treatment acceptance
    • Shorter explanation time per visit

    Used in 

    • Routine eye exams
    • OCT review appointments
    • Retina & glaucoma visits
    • Anti-VEGF treatment discussions
    • Glaucoma therapy initiation
    • Long-term monitoring plans, etc.

    Altris Education application, as an example of a unique tool specifically designed for eye care specialists’ training:

    5. Chatbots for Consultation

    A separate category is AI chatbots and virtual assistants that help with patient follow-up, education, and communication, improving day-to-day patient communication and more. With the AI Help Assistant feature, you can create an AI chatbot trained on your specific content from any platform you like. Chatbots for consultation in optometry offer clear, practical benefits for both clinics and patients:

    • 24/7 Q&A
    • Absolutely personalized follow-up instructions
    • Higher patients satisfaction rate

    Furthermore, they provide instant, 24/7 responses to the most common eye-care questions, helping patients understand symptoms, prepare for visits, and follow post-exam instructions without even waiting for staff availability. 

    Chatbots can assist with pre-consultation triage by gathering symptoms, visual complaints, and medical history, allowing optometrists to focus on higher-value clinical tasks. 

    They also improve patient engagement and adherence by delivering personalized education, reminders, and care instructions in simple, easy-to-understand language. 

    Overall, optometry chatbots dramatically reduce administrative workload, shorten response times, and support more efficient, patient-centered care while maintaining consistent communication quality.

    Types of Chatbots in Eye Care & Optometry

    agent

    1. Symptom & Triage Chatbots

    These ask users about eye symptoms, guide whether to seek care (urgent vs. routine), suggest possible conditions, and help decide next steps. Example: Ada Health

    A study published in  Eye showed that large‑language‑model‑based chatbots (e.g., ChatGPT) could answer common ophthalmology questions with high accuracy and clarity — scoring higher than alternative generative models for diagnostic and triage‑related queries (accuracy and comprehensiveness metrics showed ChatGPT did well on standard patient questions).

    Another clinical evaluation found that AI (GPT‑4) correctly identified the appropriate diagnosis among the top three options in up to 93% of ophthalmology cases and correctly assessed urgency levels in most cases — performance comparable to that of trainees. 

    2. Real‑World Chatbots for Eye Care

     The patient describes eye symptoms and uploads images (e.g., photos or OCT scans).

    An AI assistant organizes symptoms and images for review, then a real ophthalmologist chats with the patient to guide care. Key inputs here:

    • Collect symptoms in natural language
    •  Translate or clarify reports
    •  Help the clinician interpret visuals and decide next steps

    It combines AI symptom intake with real human consultation, making remote triage efficient.

    AI in optometry real chatbot examples:

    3. DocsBot AI for Optometry Services

    Practice‑focused chatbot helping clinics answer FAQs, provide pre‑appointment instructions, and automate patient engagement. Benefits:

    • Instant responses to common patient queries

    • 24/7 availability for basic information

    • Can free up staff time by handling routine communication, such as allowing self-booking, etc. 

    Patients express high satisfaction with AI self‑booking capabilities (up to 85% positive ratings).”

    Here are some more real AI chatbot applications in eye care:

    Tool / Platform Primary Focus
    DocsBot AI Patient FAQs & practice engagement 
    ThriveDesk AI AI customer support for optometry 
    Voiceflow AI Agent Custom appointment/scheduling chatbot 
    MedReception AI Manage eye exams, contact lens orders, and optical retail coordination
    OptoAI AI Assistant Knowledge and clinical support agent 
    Pod AI AI phone/communication agent  

     

    As highlighted, there is a wide array of chatbot types in optometry, ranging from patient-facing virtual assistants to AI-powered communication platforms.  

    With features such as automated appointment scheduling, pre-visit coaching, FAQ handling, 24/7 patient engagement, and basic clinical decision support, these optometry AI systems offer substantial value.

    By streamlining administrative tasks and improving patient education, these AI chatbots free up clinicians’ time to focus on direct care.

    Overall, the integration of AI-driven chatbots is revolutionizing the delivery of eye care. They enhance operational efficiency, reduce missed appointments, support timely patient triage, and improve adherence to care plans. By combining automation with intelligent decision support, AI not only optimizes clinic workflows but also elevates patient outcomes and satisfaction, marking a  transformative shift in modern optometry practice.

    Conclusion

    AI is rapidly becoming a practical and valuable tool in optometry, particularly for analysing OCT imaging. By enabling more consistent interpretation of complex retinal and optic nerve data, AI in optometry supports earlier identification of disease-related changes, more efficient triage, and improved longitudinal monitoring. 

    Beyond clinical efficiency, AI enhances patient communication by translating OCT findings into clear visual insights, supporting better understanding and engagement. As a result, optometrists are better equipped to manage more complex cases in primary care, make informed referral decisions, and deliver higher-quality, data-informed eye care—positioning AI as a meaningful complement to clinical expertise rather than a replacement.

    FAQs

    Is AI for optometry safe?

    AI in optometry is generally safe when it is properly validated, regulated, and used as a clinical support tool rather than a replacement for professional judgment. It can improve screening accuracy, enable earlier detection of eye diseases, and streamline workflows, but it also carries risks such as diagnostic errors, data privacy concerns, and bias if systems are poorly trained or over-relied upon. For safe use, optometrists must remain responsible for final decisions; patients should be informed when AI is involved; and tools should comply with medical regulations (such as FDA or CE approval) and data protection standards, such as GDPR or similar, depending on the region.

    What’s an AI OCT pathology detection tool?

    An AI OCT pathology detection tool is a software system that uses artificial intelligence to analyse optical coherence tomography (OCT) images of the eye and automatically identify signs of disease, such as macular degeneration, glaucoma, or diabetic retinopathy; it assists clinicians by highlighting abnormalities and suggesting potential diagnoses, but it is designed to support—rather than replace—professional interpretation, and its safety and effectiveness depend on proper validation, regulatory approval, and clinician oversight.

    Can AI help reduce no-shows and long wait times in an optometry practice?

    Yes. AI applications in optometry are limitless. AI can help reduce no-shows and long wait times in optometry practices by supporting appointment scheduling, reminders, and workflow optimization—such as identifying scheduling patterns, highlighting bottlenecks, and enabling more efficient patient flow—while assisting staff with better resource planning and communication.

    What can chatbots and AI assistants do for an optometry practice?

    Chatbots and AI assistants can help an optometry practice automate patient communication, streamline scheduling, and improve clinical efficiency by answering FAQs 24/7, booking and confirming appointments, sending reminders, pre-screening patients with symptom checkers, collecting medical history before visits, and triaging urgent cases. They can also support front-desk staff by handling insurance questions, guiding patients to the right services (e.g., OCT, glaucoma screening, contact lens exams), following up after visits, and reactivating inactive patients through personalized messaging. Internally, AI in optometry assistants can summarize patient data, flag high-risk cases, analyse trends in no-shows or referrals, and help with marketing automation — ultimately reducing administrative workload, improving patient satisfaction, and increasing practice revenue.

    What kinds of clinical or diagnostic support can AI provide in eye exams?

    AI can support eye exams by assisting with image review, pattern recognition, and data organization—such as highlighting features in retinal images, supporting consistency in exam review, and providing quantitative reference information—while remaining a complement to clinician judgment rather than a replacement for clinical decision-making.

    How can AI increase optical revenue and overall patient satisfaction?

    AI in optometry can increase optical revenue and patient satisfaction by helping practices streamline workflows, reduce wait times, support personalized patient communication, and enhance the in-practice experience through clearer visualization and education tools—leading to more efficient operations, improved engagement, and higher-quality service delivery.

    References:

    Luhmann, U. F. O. (2015). Innate immunity in age-related retinal degeneration. Acta Ophthalmologica. https://doi.org/10.1111/j.1755-3768.2015.0144

    https://remidio.us/solutions/teleophthalmology-telehealth/

    https://medpick.in/product/idx-dr-ai-diagnostic-system-for-detecting-diabetic-retinopathy/

    https://www.rcophth.ac.uk/academic-and-research/eye-journal/ 

    https://webeyeclinic.com/

    https://webeyeclinic.com/how-it-works/

    https://docsbot.ai/industry/optometry-services

    https://www.voiceflow.com/ai/optometrists

    https://healthus.ai/service/ai-chatbot-appointment-module/

    https://arxiv.org/abs/2511.09394

    https://ajbsr.net/data/uploads/6387.pdf

    https://www.simbo.ai/blog/the-role-of-ai-chatbots-in-revolutionizing-appointment-scheduling-and-automated-rescheduling-to-enhance-patient-convenience-and-reduce-administrative-burden-2858961/

    https://www.simbo.ai/blog/the-role-of-conversational-ai-in-revolutionizing-appointment-scheduling-and-reducing-no-show-rates-in-optometry-practices-1067939/


    https://www.simbo.ai/blog/automating-appointment-scheduling-with-ai-chatbots-reducing-no-shows-and-streamlining-patient-management-processes-2264453/

    https://www.simbo.ai/blog/how-ai-chatbots-are-transforming-appointment-scheduling-and-reducing-no-shows-in-healthcare-facilities-3576997/


    https://www.simbo.ai/blog/the-role-of-conversational-ai-in-automating-patient-appointment-scheduling-and-enhancing-healthcare-access-and-engagement-3558513/


    https://www.oscarchat.ai/blog/ai-chatbots-for-healthcare-clinics-improve-patient-support-and-appointment-scheduling/

     

  • Altris becomes the winner of VSP Vision Challenge at Vision Expo

    AI Ophthalmology and Optometry | Altris AI Altris Inc.
    1 min.

    ORLANDO, FL — Altris, an IMS for OCT analysis with ROU AI models, has been named the Judge’s Winner of the 2026 VSP Vision Innovation Challenge, one of the eyecare industry’s most prestigious startup competitions. The award was presented live on March 13, 2026, on the Innovation Stage at Vision Expo inside the Orange County Convention Center in Orlando, Florida.

    Produced by RX and The Vision Council in collaboration with VSP Vision™, the 2026 VSP Vision Innovation Challenge drew applications from companies around the globe — more than half of which were venture-backed, collectively representing over $300 million in funding. After a rigorous selection process, Altris AI was chosen as one of four finalists and participated in an intensive four-week startup bootcamp before delivering a live pitch to a distinguished panel of industry judges.

    Judges recognized Altris for its clinical impact, technological sophistication, and potential to fundamentally transform the eyecare experience. The judging panel represented a diverse cross-section of industry leaders, including executives, clinicians, and investors.

    “Winning the VSP Vision Innovation Challenge is a powerful validation of what we’re building at Altris AI. Our mission is to put the most advanced retinal intelligence in the hands of every eye care professional — and this recognition from a world-class panel of judges confirms that the industry is ready for this transformation.”

    Altris AI CEO, Andrey Kuropyatnyk

    About Altris

    Altris AI serves as a “second set of eyes” for eye care specialists, identifying more than 70 retinal biomarkers from optical coherence tomography (OCT) scans ( AI models are used for Research Use Only). The platform enables providers to match patients to the most appropriate treatments, devices, and clinical trials with objective, data-driven precision. Altris system is FDA-cleared and HIPAA-compliant, and integrates seamlessly with OCT scanners from nine major manufacturers.

    By combining advanced deep learning algorithms with a clinically intuitive interface, Altris AI reduces diagnostic variability, supports earlier detection of sight-threatening conditions, and frees eye care professionals to focus on what matters most: patient outcomes.

    About the VSP Vision Innovation Challenge

    The VSP Vision Innovation Challenge is a global competition designed to source, support, and accelerate early-stage startups and technologies advancing the eyecare experience. This year’s finalists represented a broad spectrum of innovation — from AI-driven diagnostics and exam automation to digital education and accessibility-focused smart solutions.

    In addition to pitching live, finalists exhibited at Vision Expo’s Innovation Center, a dedicated emerging-technology destination featuring more than 20 next-generation startups spanning AI, augmented and virtual reality, and advanced diagnostics.

    “Innovation in eyecare is accelerating, which is why it’s crucial for the industry to stay actively engaged. We launched the VSP Vision Innovation Challenge to connect emerging technologies with the stakeholders they aim to serve, and this year is no exception. We’re proud to support solutions that advance care for both providers and patients.”

    — Will Flanagan, Head of VSP Global Innovation Center Programs and Partnerships

    Looking Ahead

    With this recognition, Altris AI continues to accelerate its mission of making retinal AI accessible to every eye care professional. The company will leverage the visibility and industry connections gained from the challenge to deepen partnerships with providers, expand clinical integrations, and advance its platform’s capabilities.

    The next VSP Vision Innovation Challenge will take place at Vision Expo 2027, scheduled for March 10–13 in Las Vegas. Applications are expected to open later this fall.

    About Altris Inc.

    Altris AI is a clinical-grade SaaS platform that empowers eye care specialists with AI-driven retinal analysis. By identifying 70+ retinal biomarkers from OCT scans, Altris AI helps providers deliver more precise, evidence-based care. The platform is FDA-approved, HIPAA-compliant, and compatible with OCT devices from nine leading manufacturers.

  • Drusen on OCT: Detection, quantification, and tracking

    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min.

    Introduction

    Drusen remains one of the main biomarkers of age-related macular degeneration (AMD). They play a prognostic role and reflect the stage of the disease. Distinguishing drusen parameters provides a personalized risk profile for the transition to geographic atrophy or neovascular AMD. Everyone working with AMD patients should know how to detect, quantify, and track drusen on OCT.

    What are the types of drusen?

    Drusen are accumulations of pathological material of lipid-protein nature, localized under the PES. They reflect impaired transport and exchange between the retinal pigment epithelium and Bruch’s membrane. Historically, they are divided into hard, soft, reticular pseudodruses (or subretinal drusenoid deposits) and other less common types (confluent, pachidruses) as well as other retinal OCT biomarkers for drusen segmentation.

    Hard drusen

    On ophthalmoscopy, they are small, rounded, clearly delineated foci of yellowish-white color. On OCT, they look like local deposits of hyperreflective material under the PES with a diameter of no more than 63 microns. In small quantities (up to 8), they are not a sign of pathology. They are asymptomatic in most patients.

    Soft drusen

    Soft drusen are larger than hard drusen and appear as extensive foci with blurred edges on the fundus. On OCT, they are dome-shaped and elevated above the PES and are divided into medium (63-125 μm) and large (more than 125 μm) in size. They are more strongly associated with AMD progression, especially when accompanied by pigmentary abnormalities and other OCT biomarkers (hyperreflective foci, destruction of the ellipsoidal zone, etc.). Soft drusen can enlarge and merge. An area of ​​merging drusen with a diameter exceeding 350 μm is called a drusenoid detachment of the PES.

    Soft drusen highlighted

    Soft drusen detected by Altris IMS. AI models are for Research Use Only. Not for use in diagnostic purposes. 

    Confluent drusen

    These are multiple small deposits under the PES, which can occur in relatively young patients; on FAG they often show a “starry sky” appearance. On OCT, there are multiple small symmetrical elevations of the PES, small in diameter (like hard drusen), but more numerous, prone to merging. The course is variable: some patients maintain a stable course for years, some have an increased risk of complications and transition to the late stages of AMD.

    Reticular pseudodrusen (or subretinal drusenoid deposits)

    They differ fundamentally in their localization, being located above the PES (in the subretinal space). They contain some common proteins with soft drusen, but differ in lipid composition. Due to their close location to the important photoreceptor layer, they are more often combined with a decrease in visual function, and also carry a higher risk of progression to late AMD (especially characterized by a rapid transition to geographic atrophy (GA) and the development of macular neovascularization (MNV) type 3).

    What are the levels of drusen?

    The AREDS size classification is still useful in clinical practice: small <63 μm, medium 63–124 μm, large ≥125 μm. Analyses confirm that the 5-year risk of progression to late AMD increases with the number and size of drusen in both eyes and especially with the presence of reticular pseudodrusen. In the NICE guidance for the management of patients with AMD (2018), the risk of progression also depends on the size and type of drusen, as well as the presence of associated pathological changes (pigmentary abnormalities, vitelliform deposits).

    The OCT era has added powerful quantitative metrics with AI for drusen measurement and monitoring:

    • drusen height (μm),
    • area (mm²),
    • volume (mm³),
    • topography (central ring within 1.5 mm; parafovea 3–5 mm),
    • dynamics of changes and associated biomarkers (hyperreflective foci, ellipsoidal zone disruption, presence of hypertransmission zones, etc.).

    A practically significant increase in the volume of drusen in the macular region over a year/two correlates with structural and functional deterioration (destructive changes in the photoreceptor layer, changes in ONL thickness, visual acuity). Data from multicenter projects (such as MACUSTAR) confirm the repeatability of measurements and the possibility of comparison between devices, provided that the correct algorithms are used.

    What do drusen look like on OCT?

    On B-scan OCT, classic hard and soft drusen are localized deposits of hyperreflective material between the PES and Bruch’s membrane (under the PES). Reflectivity can be uniform or heterogeneous depending on the structure and stage of development. Reticular pseudodruses are localized between the photoreceptor layer and the PES (above the PES) – this is the key difference from conventional drusen. On OCT images, they appear as tubercles in the subretinal space that remodel the outer layers of the retina (in particular, the ellipsoidal zone), and on en face, they are visualized as punctate structures, usually connected in a mesh pattern.

    A: Soft drusen. B: Hard drusen (Source) Another classic white and black scan

    In addition to the drusen themselves, clinically significant are hyperreflective foci, destruction of the ellipsoidal zone, thinning of the outer layers/ONL, formation of hyperreflective foci in OCT or geographic atrophy with the effect of hypertransmission – it is the combinations of these features that form prognostic models of the transition of intermediate AMD to late stages. The combination of these biomarkers consistently exceeds single morphometric thresholds.

    En Face Optical Coherence Tomography Illustration

    En Face Optical Coherence Tomography Illustration of the Trizonal Distribution of Drusen and Subretinal Drusenoid Deposits in the Macula (Source)

    As we can see, en face and linear OCT scans help to differentiate different types of drusen and track their progression dynamics. Modern deep learning models for AI drusen examination and en face analysis, like Altris.AI, reliably detect and segment classic drusen from subretinal drusenoid deposits, improving repeatability and reporting speed. You may see the difference from the classic white and black image analysis here:

    Confluent drusen are highlighted in red

    Confluent drusen are highlighted by Altris IMS. AI models are used for Research Use Only. Not for use in Diagnostic Purposes. 

    How to measure drusen size?

    Here we can find how drusen are measured:

    1) Classical size scale (AREDS):

    Orientation on diameter or equivalent on planar reconstructions: <63, 63–124, ≥125 μm. Convenient, but does not take volume/height or topography into account.

    2) Quantitative OCT analysis of PES elevation:

    On ZEISS CIRRUS instruments, the Advanced RPE Analysis module automatically calculates the area and volume of PES elevation in standard 3 and 5 mm rings around the fovea; the minimum height that the system consistently includes in quantitative results is about 19–20 μm. This provides repeatable metrics and a common “language of numbers” for clinical and research purposes.

    3) Morphometric rule for differentiation of drusen and drusenoid detachment of PES:

    By basal width: <350 μm – drusen, ≥350 μm drusenoid detachment of PES.

    4) AI segmentation and 3D morphometry:

    Deep networks segment Bruch’s membrane, PES, and ellipsoidal zone, as well as PES elevation on OCT, calculating drusen height/area/volume and generating dynamics maps. Validation work in 2023–2025 will demonstrate robustness between different OCT devices, which is critical for multicenter networks. Besides, you may track drusen progression on OCT AI tool and stay informed ahead of time to prevent more severe pathology changes in advance.

    Can drusen exist without macular degeneration?

    Yes, and this is possible in the following cases.

    Small (<63 μm) single drusen may occur in the elderly in the absence of other signs of AMD and concomitant risk biomarkers (hyperreflective foci, ellipsoidal zone abnormalities). In this phenotype, the 5-year risk of progression is low; routine monitoring at an interval of 1 time per year is sufficient, if possible, with recording quantitative indicators on OCT (volume/area of ​​PES elevation) for comparison in dynamics. The patient should be informed that the fact of “small drusen” alone does not equal a diagnosis of AMD and does not require treatment, but it is advisable to maintain lifestyle modification (blood pressure control, smoking cessation, a healthy diet).

    Confluent drusen are sometimes found in younger patients; they do not always fit into the classic models of AMD. Tactics – individual observation with an emphasis on high-quality OCT documentation (the same scan and control of concomitant biomarkers). In the absence of “red flags”, a 6-12 month follow-up interval is sufficient.

    Understanding Macular Degeneration

    Understanding Macular Degeneration (Source)

    Hereditary dystrophies (EFEMP1-related; associated phenotypes are Doyne’s cellular degeneration of the retina and Leventis’ malady) form drusen-like deposits without the typical pathogenesis inherent in AMD. They have an autosomal dominant inheritance pattern and are characterized by yellow-white deposits, like drusen, accumulating under the PES, often in the peripapillary zone. The clinical picture may include gradual vision loss, impaired contrast sensitivity, or metamorphopsia. In this case, timely detection of the phenotype (age of onset, family history, symmetry, characteristic fundus appearance) and referral for medical and genetic counseling with a subsequent individual follow-up plan, including monitoring of possible complications (neovascularization, atrophic changes).

    Drusen vs. drusenoid detachment of PES

    Drusen are local elevations of PES above Bruch’s membrane due to deposits of pathological material under PES. Usually multiple, of different diameters, with a tendency to merge with the formation of larger, topographically continuous areas of PES elevation.

    Drusenoid detachment of the pigment epithelium is formed from a larger conglomerate of drusenoid material, which in turn is formed as a result of the fusion of drusen.

    Another differentiating drusen and drusenoid deposits subtypes on multimodal imaging samples

    Another differentiating drusen and drusenoid deposits subtypes on multimodal imaging samples

    On B-scan OCT, it has smooth edges, uneven reflectivity, and often retains communication with neighboring drusen. On en face visualization, a conglomerate of elevation is visible, which corresponds to the zone of changes in the PES-Bruch’s membrane complex. In the absence of fluid inside the lesion, we are talking about drusenoid detachment of PES; if homogeneous hyporeflectivity is visualized under PES, this is serous detachment of PES, and if there are signs of a neovascular membrane according to OCTA or FAG, this is fibrovascular detachment of PES. Therefore, in doubtful cases, it is advisable to add OCTA to exclude hidden MNV.

    The main morphometric rule: basal width ≥350 μm (in the horizontal projection of the OCT slice favors drusenoid detachment of PES. In some situations, we also pay attention to the content (serous/optically empty space, signs of vascularization), PES profile, and associated biomarkers, since PES detachment is more often associated with the risk of transition to HA or the formation of neovascularization.

    What is the best treatment for drusen?

    Drusen are not treated as a separate nosology. They are a structural biomarker of AMD, and also have prognostic value for assessing the further development and rate of progression of the disease.

    Optimal tactics for detecting drusen:

    Optimal tactics for detecting drusen may include the following

    Risk modification: 

    • smoking cessation,
    •  blood pressure control,
    • metabolic profile,
    • diet.

    Dietary supplements based on AREDS 2: 

    • taking antioxidant complexes (lutein, zeaxanthin, vitamins C and E, zinc, copper) reduces the risk of transition to late AMD by approximately 25% within 5 years (according to AREDS 2).

    Quantitative monitoring on OCT: 

    • record the volume/area/height of drusen and their dynamics, distinguish between drusen types, detect other concomitant signs of AMD progression (hyperreflective foci, destructive changes in the ellipsoidal zone, pigmentary anomalies, vitelliform material deposition, signs of formation of foci of geographic atrophy).
    • Individualize observation intervals (depending on the type of drusen, the dynamics of their structural changes and other risk factors).
    • Among the new promising methods of treating dry AMD at the drusen stage is multiwavelength photobiomodulation.

    Multiwavelength photobiomodulation:

    This method is aimed at stopping or regressing the progression of dry AMD by modulating mitochondrial activity and consists of the use of specific light (red and near-infrared spectrum from ~590 to 850 nm), which can reduce oxidative stress in retinal cells, inflammation and apoptosis of PES cells.

    The efficacy as a potential treatment approach has remained controversial until recently: studies have shown only temporary improvement in visual function and reduction in drusen volume (not maintained for 6 months).

    Updated data from the LIGHTSITE III study were presented at the ARVO 2025 conference. They showed that photobiomodulation can significantly slow the decline in visual acuity and reduce the rate of expansion of HA zones

    Recently, the FDA approved photobiomodulation for the treatment of AMD.

    For complications:

    • Neovascular AMD– anti-VEGF.
    • Geographic atrophy – injectable drugs (inhibitors of the C3 and C5 complement system), approved by the FDA

    The role of AI drusen quantification OCT

    The role of AI: automated drusen-volume measurement in OCT is now a reality. IT allows automated segmentation and counting (3D volume, area, height), identification of reticular pseudodruses and other signs of AMD, and compilation of prognostic profiles.

    In practice, applying an OCT drusen-counting algorithm reduces variability in assessments and helps personalize visit frequency. Additionally, home OCT monitoring models with AI analysis are being developed, indicating that broader AI support for AMD is fast approaching.

    Conclusion

    Drusen on OCT are more than just a sign of AMD. They have become one of the most important biomarkers of age-related macular degeneration and a kind of “compass” in the daily practice of an ophthalmologist. Today we understand that:

    Drusen come in different types, and, accordingly, carry different prognostic information: hard, soft, confluent, and reticular pseudodrusen. Each type carries a different risk and requires a different surveillance strategy.

    Drusen levels are no longer limited to diameter, height, volume, dynamics, and structural features as well as accompanying OCT biomarkers have also become important. It is the combination of these parameters that allows us to predict the transition to the late stages of AMD.

    OCT has changed the game: drusen can now be seen in 3D, segmented automatically, build PES elevation maps, and compare data between visits. Thanks to this, the doctor receives a lot of information about the evolution of the disease.

    AI sets a new standard: algorithms can accurately calculate drusen volume, identify their subtypes, generate prognostic profiles, and reduce interobserver variability. This translates data from subjective descriptions into objective, reproducible numbers.

    Drusen classification on OCT using AI allows not only ascertaining the presence of drusen, but also differentiating their type, objectively measuring their number and parameters, and tracking their dynamics via AI drusen quantification on OCT. For the doctor, this means identifying risk factors in the early stages of retinal disease, accurately comparing data between visits, and prescribing the correct therapy promptly.

    Home monitoring is the future that has already begun: the first FDA-approved solutions with “OCT + AI” are currently used to monitor fluid in neovascular AMD, but they pave the way for daily structural monitoring of drusen as well. This means that in the near future, the patient may be able to monitor their own retina at home, and the doctor may be able to see the dynamics in real time.

    In the treatment of drusen wet or dry AMD, the main goal remains not to “remove drusen,” but to minimize risks (smoking, diet, systemic factors), prescribe AREDS2-based complexes, timely detect complications, and apply already available therapies (anti-VEGF in INM, C3 and C5 inhibitors of the complement system in HA). Among the new promising methods for treating dry AMD at the drusen stage is multiwavelength photobiomodulation.

     

    It is important to remember when communicating with the patient: drusen is not a therapeutic target, but a structural “compass”. We do not “treat drusen.” Instead, we systematically reduce risks (smoking, blood pressure, nutrition), use drugs based on the AREDS2 formula, and most importantly, we regularly measure their quantitative parameters in dynamics. When complications appear and the transition to a late stage occurs, we prescribe treatment based on the same objective OCT metrics. Thus, instrumental accuracy and AI analytics turn drusen into a manageable marker that helps to timely detect the risks of AMD progression.

    Thus, drusen on OCT have become a bridge between morphology and prognosis. They provide an opportunity to build a long-term strategy for preserving vision. Today, the doctor is required not only to see drusen, but also to quantitatively measure, assess in dynamics, calculate the risk, and explain to the patient his individual risks. It is thanks to these approaches that we are moving towards a new paradigm – personalized ophthalmology, where decisions are made based on objective digital data, enhanced by artificial intelligence.

    Sources:

      1. https://pubmed.ncbi.nlm.nih.gov/39558093/
      2. https://jamanetwork.com/journals/jamaophthalmology/fullarticle/2765650
      3. https://link.springer.com/article/10.1007/s00417-024-06389-x
      4. https://iovs.arvojournals.org/article.aspx?articleid=2804052
      5. https://www.ophthalmologyscience.org/article/S2666-9145(25)00182-4/fulltext
      6. https://www.nature.com/articles/s41433-024-03460-z
      7. https://www.ophthalmologytimes.com/view/arvo-2025-update-on-the-lightsite-iii-study-in-amd
  • Central Retinal Vein Occlusion CRVO OCT: Detection and Modern Approaches to Monitoring and Treatment

    crvo
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    3 min.

    Introduction

    Central Retinal Vein Occlusion (CRVO) OCT is one of the most common and clinically significant vascular disorders affecting the eye, often resulting in substantial visual impairment. This condition ranks second among causes of vision loss due to vascular disease, after diabetic retinopathy, placing a considerable burden on both healthcare systems and patients’ quality of life. Epidemiological studies show that the prevalence of RVO increases with age, and in populations with concomitant cardiovascular disease, the risk of developing occlusion rises severalfold.

    Despite a long history of study, it is the breakthroughs in instrumental diagnostics over the past decade that have fundamentally changed our approach to recognizing and managing RVO. Previously, assessment of the macula and retinal vasculature relied primarily on ophthalmoscopy. While still an important tool, it has inherent limitations.

    Optical coherence tomography (OCT) has revolutionized diagnostic standards. With its high resolution and ability to capture subtle structural changes within the retinal layers, OCT has become indispensable for determining disease severity, monitoring treatment efficacy, and conducting long-term follow-up. It allows for the detection of minimal early signs of edema, subclinical structural damage, and initial manifestations of ischemia—changes that were practically inaccessible for dynamic assessment 10–15 years ago.

    This level of precision is particularly critical for patients at increased risk of RVO. The most vulnerable groups include individuals with arterial hypertension, diabetes mellitus, glaucoma, coagulation disorders, as well as older adults, in whom the vascular walls may already have undergone degenerative or sclerotic changes.

    Importantly, modern RVO treatments require objective dynamic monitoring. OCT enables precise evaluation of structural changes, tracking of therapeutic response, and individualization of treatment strategies, helping to avoid both overtreatment and undertreatment.

    Thus, the role of OCT today goes far beyond simple visualization: it is a key tool for prognostic assessment, patient stratification, optimization of therapeutic decisions, and timely detection of complications.

    crvo

    1. What RVO Is and Why It Occurs?

    Central Retinal Vein Occlusion (CRVO) OCT is a disruption of venous blood outflow in the retina due to partial or complete vein occlusion. As a result, the following occur:

    • Blood stasis
    • Increased venous pressure
    • Impaired capillary perfusion
    • Retinal edema, especially in the macular area
    • Risk of neovascularization

    Early detection is critical, as prompt treatment—particularly for macular edema—significantly increases the chances of preserving or restoring vision. Delayed diagnosis can lead to progression of ischemia, neovascularization, neovascular glaucoma, and persistent macular dysfunction.

    RVO also has important systemic implications: patients with a history of RVO have a higher risk of acute cardiovascular events (myocardial infarction, stroke, heart failure) compared with the general population. This emphasizes the need for comprehensive management, involving not only ophthalmologists but also other specialists, such as cardiologists.

    Central vs. Branch Retinal Vein Occlusion: Pathogenesis Differences

    • Central Retinal Vein Occlusion (CRVO) occurs when blockage happens at the level of the lamina cribrosa. Compression, arterial wall thickening, or thrombotic processes disrupt blood outflow from the entire retina. Typical signs include:
      • Diffuse hemorrhages
      • Marked macular edema
      • Increased risk of optic disc and iris neovascularization due to severe ischemia
      • Generally worsen prognosis than branch occlusions
    • Branch Retinal Vein Occlusion (BRVO) usually occurs at arteriovenous crossings, where a thickened artery compresses a vein, causing localized occlusion. Characteristic features include:
      • Localized edema and hemorrhages
      • Clear segmental distribution
      • Prognosis is generally better than that of CRVO, though macular edema may persist

    Key Risk Factors for RVO

    Modern studies and guidelines identify the following as the main risk factors:

    • Arterial hypertension
    • Atherosclerosis and age-related vascular changes
    • Diabetes mellitus (even without diabetic retinopathy)
    • Glaucoma and elevated IOP
    • Hypercoagulable states, thrombophilia
    • Obstructive sleep apnea
    • Age >50 years

    Rare cases of RVO associated with thromboembolic complications after COVID‑19 infection or vaccination have also been reported, highlighting the ongoing relevance of thrombotic mechanisms.

    Impact on Microcirculation and Vision

    RVO leads to:

    • Impaired normal venous outflow
    • Sharp elevation of hydrostatic venous pressure
    • Damage to the blood-retinal barrier
    • Leakage of plasma and cellular elements into the retinal interstitium, causing macular edema
    • Development of ischemic zones
    • Over time, thinning of inner retinal layers, neuroepithelial atrophy, and damage to the photoreceptor layer

    These changes are best assessed with OCT, which enables precise patient stratification and treatment planning. Timely diagnosis, proper monitoring, and early therapy are essential.

    fluid progression

    2. OCT Signs of Retinal Vein Occlusion: Detecting Subtle Changes

    With the advent of OCT, detection of structural retinal changes in RVO has significantly improved—even at early stages without obvious clinical signs.

    Acute Stage Changes (first weeks after occlusion)

    • Macular edema:
      • Cystic spaces in inner retinal layers (INL, OPL)
      • Increased central retinal thickness
      • Subretinal fluid (serous neurosensory detachment)
    • Intraretinal hemorrhages: appear on OCT as hyperreflective areas with shadowing of underlying layers
    • Ischemia indicators:
      • Hyperreflectivity of neuroepithelium
      • Cotton-wool spots

    Chronic Stage Changes (months later)

    • Chronic ischemic and atrophic changes (thinning of inner retinal layers)
    • Disruption of photoreceptor layer (ELM and EZ)
    • Disorganization of inner retinal layers (DRIL)
    • Persistent edema (>6 months) indicates chronic RVO requiring therapeutic adjustment

    AI for OCT thus allows both acute diagnosis and long-term monitoring of ischemic progression or tissue remodeling.

    tissues

    rvo

    crvo

    3. Assessment of Macular Changes in RVO Using OCT

    Central retinal vein occlusion crvo OCT is now considered the gold standard for diagnosing, monitoring, and assessing treatment response in macular edema, including that associated with RVO.

    OCT is highly sensitive for:

    • Quantitative and qualitative analysis (central retinal thickness [CRT], macular volume [MV], size and number of cystic spaces, DRIL, photoreceptor layer integrity)
    • Evaluating treatment response
    • Detecting minimal residual cysts
    • Predicting visual acuity outcomes

    Typical OCT Findings in RVO:

    • Diffuse retinal thickening
    • Cystoid macular edema (localized cysts deforming normal retinal architecture)
    • Serous neurosensory detachment (indicative of blood-retinal barrier breakdown)
    • Disruption of EZ and ELM (photoreceptor involvement, critical for final visual acuity)

    These capabilities make OCT an integral part of modern RVO monitoring.

    rvo 2

    4. Top 3 Challenges in RVO OCT Analysis

    Despite its power, OCT assessment of RVO has significant limitations:

    1. Need for normative comparison
      Interpretation requires comparison with the patient’s contralateral eye or established normal values. Systemic vascular anomalies can affect both eyes, limiting standardization.
    2. Complexity with comorbidities
      Many RVO patients have systemic (hypertension, diabetes) or ophthalmic comorbidities (diabetic retinopathy, AMD, glaucoma, epiretinal membrane), complicating interpretation. It can be difficult to distinguish RVO-related changes from combined pathology.
    3. Requirement to consider the clinical context
      OCT provides only part of the clinical picture. Accurate interpretation requires integration of symptoms, medical history, systemic factors, fundoscopic findings, and other diagnostic tests. Anatomical variations, comorbidities (glaucoma, cataract), and individual treatment response also necessitate a personalized approach.

    5. Treatment of RVO: Modern Approaches

    Currently, no treatment restores normal retinal venous circulation. Therefore, therapy focuses on controlling complications, primarily macular edema and preventing neovascularization (retinal, iris/optic disc, neovascular glaucoma, hemorrhages, and tractional changes).

    All RVO patients should receive systemic management, ideally in collaboration between an ophthalmologist and a cardiologist or internist. Monitoring of blood pressure, lipids, glucose, and coagulation factors is essential, as RVO often signals systemic vascular risk.

    Treatment decisions must be individualized, considering:

    • RVO subtype (CRVO vs. BRVO)
    • Edema severity
    • Clinical and OCT findings
    • Risk of adverse effects
    • Patient status (comorbidities, ability for regular follow-up)

    Anti-VEGF Therapy as First-Line Treatment

    Intravitreal anti-VEGF injections are the first-line therapy for macular edema associated with RVO. These drugs reduce vascular endothelial growth factor (VEGF) expression, lowering vascular permeability, fluid leakage, edema, and inhibiting pathological neovascularization.

    Commonly used agents:

    • Ranibizumab, Aflibercept, Faricimab: proven safe and effective for CRVO and BRVO-related macular edema brvo vs crvo oct; studies show significant improvements in best-corrected visual acuity (BCVA) and central macular thickness (CMT).
    • Bevacizumab: used off-label for macular edema and neovascularization.

    Long-term studies indicate anti-VEGF therapy provides sustained visual improvement for many patients, with injection frequency often decreasing over time.

    Advantages:

    • High efficacy for macular edema
    • Good tolerability and safety (systemic complications are rare)
    • Personalized treatment possible

    Limitations / Challenges:

    • Some patients respond insufficiently
    • Requires frequent injections (clinic visits, financial burden, potential complications, patient discomfort)
    • Chronic or refractory edema may require alternative or combination approaches

    Steroid Implants and Injections: Second-Line Therapy

    Dexamethasone intravitreal implant (OZURDEX) is approved for RVO-related macular edema, particularly when:

    • Anti-VEGF therapy is insufficient
    • Frequent injections are impractical (distance, transportation, cost)

    Steroids reduce inflammation, vascular permeability, and fluid accumulation, useful in chronic or resistant edema.

    Risks / Limitations:

    • Cataract (especially with repeated or long-term use)
    • Increased intraocular pressure (IOP), potential steroid-induced glaucoma

    Laser Therapy

    • Panretinal photocoagulation is effective for neovascularization.
    • Its use has declined with anti-VEGF availability, which offers strong anatomical and functional results.

    Surgical Approaches

    • Vitrectomy may be considered in selected cases.
    • Surgery carries risks and is reserved for situations where other treatments fail or are inappropriate.

    Combination Strategies

    • In practice, clinicians often combine anti-VEGF therapy with steroid implants or laser treatment, depending on disease course.
    • This can reduce total injection burden, minimize side effects, and improve outcomes in chronic or recurrent edema.

    Monitoring Frequency

    • Active macular edema or ongoing treatment requires regular OCT follow-up to evaluate therapeutic response and adjust injection intervals.
    • OCT schedule:
      • Monthly at treatment initiation
      • Individualized intervals using Treat-and-Extend protocols
      • Structural monitoring to prevent atrophic changes
    • Ischemic RVO patients have the highest neovascularization risk within the first 90 days; monthly monitoring during the first 6 months is critical.

    Conclusions and Recommendations

    RVO is a complex, multifactorial vascular disorder that can cause sudden and severe vision loss, particularly in patients with systemic risk factors. Modern management aims not only to address acute complications but also to control long-term structural retinal changes.

    OCT has transformed RVO care by providing:

    • Early detection of edema, subclinical ischemia, and architectural changes
    • Dynamic monitoring of treatment response, allowing timely adjustments and optimization
    • Improved long-term prognostication through evaluation of macular thickness, outer retinal layers, and fluid volume

    OCT helps identify edema type and secondary changes—atrophy, photoreceptor damage, inner retinal thinning—allowing a more accurate visual prognosis, especially in ischemic RVO.

    When combined with modern anti-VEGF agents, long-acting steroid implants, and personalized dosing regimens, OCT enables:

    • Reduction of unnecessary injections via interval optimization
    • Maximized treatment efficacy based on morphological findings
    • Prevention of recurrence and progression through early detection of edema

    Thus, OCT is not merely a visualization tool but a core element of clinical decision-making, improving patient management, preventing complications, and enabling more complete and stable visual recovery.

    Clinical Recommendation: Integrate regular OCT assessments into RVO management, with attention to macular thickness dynamics and outer retinal layer integrity for precise disease control and optimized therapeutic outcomes.

    References:

    1. https://pubmed.ncbi.nlm.nih.gov/38714470/
    2. https://www.rcophth.ac.uk/wp-content/uploads/2015/07/Retinal-Vein-Occlusion-Guidelines-Executive-Summary-2022.pdf
    3. https://www.mdpi.com/2077-0383/14/4/1183
    4. https://www.auctoresonline.org/article/clinical-therapeutic-orientation-in-retinal-venous-obstruction
    5. https://www.mdpi.com/2077-0383/10/3/405
    6. https://pmc.ncbi.nlm.nih.gov/articles/PMC10801953
    7. https://www.mdpi.com/2075-4418/13/19/3100
    8. https://karger.com/oph/article-abstract/242/1/8/255831/Microvascular-Retinal-and-Choroidal-Changes-in?redirectedFrom=fulltext
    9. https://link.springer.com/article/10.1007/s40123-024-01077-9
    10. https://pubmed.ncbi.nlm.nih.gov/39717563/
    11. https://provider-rvo.vision-relief.com/introduction/management/

     

  • Key Trends in Ophthalmology and Optometry in 2026

    trends
    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    3 min.

    Introduction

    The year 2026 in ophthalmology will not be defined by a single “major breakthrough,” but rather by key trends in Ophthalmology and optometry in 2026, and the maturation of several directions whose discoveries and innovations are now transitioning into everyday clinical practice. While just a few years ago innovations were often perceived as isolated technologies far removed from real-world care (a new drug, device, or piece of equipment), today entire ecosystems are being formed: from early detection to long-term monitoring, from the ophthalmologist’s office to optometric screening, from a single consultation to a longitudinal patient journey supported by digital tools.

    The core logic of 2026 is a shift from reactive to proactive ophthalmology. Increasingly, the goal is to prevent disease at the stage of risk-factor modification, intervene in the earliest pathological changes, and track preclinical markers. This shift is visible across several dimensions: the growing role of telemedicine and portable diagnostics; autonomous AI becoming a public health tool; and oculomics, which enables ocular image analysis to serve as a source of early biomarkers for systemic conditions. At the same time, the treatment paradigm is evolving: where repeated procedures once dominated (for example, frequent intravitreal injections), 2026 brings a move toward extended-duration regimens, implant-based drug delivery platforms, and disease control with fewer clinic visits.

    Another important axis is the alignment of patient expectations. Some new approaches (for example, in the management of dry AMD and geographic atrophy) do not promise to “restore vision,” but rather to buy time—slowing structural retinal damage and functional vision loss. As a result, in 2026, risk–benefit communication and shared decision-making become almost as important as the choice of molecule or device itself.

    Below, we outline the key eye care trends of 2026: what is changing, why it matters, and how it will shape ophthalmic and optometric practice.

    trend pol

    1. New Approaches to Treatment

    1.1. Geographic Atrophy (GA): The Introduction of Active Treatment in eye care trends 2026

    1.1.1. Injectable Therapies as Ophthalmology Trends 2026

    Following the key trends in Ophthalmology and Optometry in 2026 , development of injectable therapies for geographic atrophy, clinical practice is entering a “second wave” phase—where the main questions are no longer whether therapy is possible for a disease historically considered untreatable, but how that therapy should be practically implemented. In 2026, the focus will be on patient selection, treatment initiation, dosing frequency and duration, as well as monitoring.

    Currently, the FDA has approved the following injectable therapies for GA:

    • Izervay (avacincaptad pegol) — a C5 complement inhibitor.
    • Syfovre (pegcetacoplan) — a C3 complement inhibitor.

    Their mechanism of action involves reducing chronic inflammation and cellular damage in the retina and—most importantly—slowing the rate of GA lesion expansion.

    Because most available data focus on slowing atrophy progression (an anatomical endpoint) rather than guaranteed improvements in visual acuity, properly managing patient expectations becomes particularly critical in 2026. Clear discussions about therapeutic goals and limitations are emphasized in review publications addressing the first approved GA treatments.

    ga injections

    1.1.2. Multiwavelength Photobiomodulation

    Multiwavelength photobiomodulation is one of the most promising emerging approaches and key trends in Ophthalmology and Optometry in 2026 aimed at halting or slowing the progression of dry AMD through modulation of mitochondrial activity. The use of specific wavelengths (red and near-infrared light, approximately 590–850 nm) may reduce oxidative stress in retinal cells, inflammation, and apoptosis of retinal pigment epithelium cells.

    Its appeal is clear: a non-invasive procedure with significantly better acceptability for some patients compared with regular injections.

    Until recently, its effectiveness remained debated, with studies showing only temporary functional improvement and reduction in drusen volume. At ARVO 2025, updated results from the LIGHTSITE III study demonstrated that photobiomodulation can significantly slow visual acuity decline and reduce the rate of GA expansion.

    In 2025, the FDA approved photobiomodulation for AMD, creating strong prospects for broader clinical adoption in 2026.

    The 2026 trend is correct positioning and stratification:

    • Use of photobiomodulation based on clear indications for specific dry AMD stages and patient profiles.
    • Transparent communication of expectations, with goals focused on functional support and slowing GA progression rather than guaranteed vision restoration.

    photobiomodulation

    1.2. Extended Anti-VEGF Treatment Regimens

    Another major trend is the shift toward regimens with reduced injection frequency. This is not merely about comfort, but primarily about preventing missed visits: patients with AMD and diabetic retinopathy with DME often fall out of treatment due to visit burden. Thus, 2026 reinforces the principle that treatment must be effective in real-world conditions, not only under ideal adherence.

    The ranibizumab port delivery system (Susvimo, Port Delivery System) has become emblematic of this trend. In 2025, the FDA also approved Susvimo for the treatment of diabetic retinopathy.

    1.3. Gene Therapy for Macular Telangiectasia Type 2 (MacTel 2)

    MacTel 2 is a chronic, progressive neurodegenerative retinal disease that previously lacked active treatment.

    In 2025, the first implantation of ENCELTO (revakinagene taroretcel)—the first and currently only FDA-approved gene therapy for MacTel 2—was performed in the United States. ENCELTO enables a shift from observation to active intervention, with the potential to preserve visual function in early-stage patients.

    The device is based on encapsulated cell therapy technology: a capsule containing genetically modified cells that continuously secrete recombinant human ciliary neurotrophic factor (CNTF), acting as a neuroprotective agent that slows photoreceptor degeneration.

    In 2026, the focus will move from “innovation storytelling” to routine clinical implementation, including defining early selection criteria, monitoring protocols (OCT biomarkers, functional testing), and accumulating real-world long-term data on photoreceptor preservation and visual function.

    1.4. Gene Therapy for Neovascular AMD: Closest to Real Transformation

    For neovascular AMD, gene therapy remains one of the most anticipated eye care trends 2026 directions, as it has the potential to fundamentally change treatment logic—from repeated injections to a single vector administration enabling long-term therapeutic protein expression. Reviews published in 2025 highlight active programs such as RGX-314, ADVM-022 (Ixo-vec), 4D-150, and others.

    In 2026, the key questions shift from “does it work?” to “how does it work across different patient groups?” including:

    • Stability and duration of expression;
    • Inflammatory and immune response profiles;
    • Need for supplemental anti-VEGF therapy;
    • Patient selection criteria;

    Injection centers and post-procedure monitoring standards.

    2. Oculomics: The Eye as a “Window to the Body” and a Source of Digital Biomarkers

    Oculomics is one of the most compelling trends of 2026, as it reshapes ophthalmology’s role within medicine as a whole. The concept is simple: the eye is the only structure where microvasculature, neurons, and signs of metabolic and inflammatory processes can be visualized non-invasively at high resolution. As a result, fundus and OCT/OCTA data may serve as biomarkers for systemic conditions—from cardiovascular risk to neurodegenerative diseases.

    oculomics

    In contemporary research, oculomics is described as an approach that uses retinal images to assess systemic risks and conditions, with potential scalability for screening. In 2026, this “scale” becomes critical: data may originate not only from ophthalmology clinics, but also from optometric practices, mobile screening programs, and telemedicine.

    What truly changes in 2026:

    • A transition from “interesting correlations” to clinical utility, with models expected to demonstrate actionable impact on patient management.
    • Data verification and management of false-positive risk, including the communication of systemic risk to patients.
    • Integration with AI, as multidimensional patterns often exceed human interpretive capacity.

    A major risk in 2026 is over-marketing, reinforcing the need for externally validated models with clear clinical context that do not generate unnecessary “medical noise.”

    3. AI Technologies: From Decision Support to Autonomous Screening and Managed Patient Pathways

    votes

    3.1. Autonomous Diabetic Retinopathy Screening as a Scalable Standard

    In 2026, diabetic retinopathy remains the most studied use case for autonomous AI. In the United States, three FDA-approved autonomous DR screening systems are already described (LumineticsCore/IDx-DR, EyeArt, AEYE-DS). This positions AI as a practical tool capable of influencing large-scale screening programs, particularly in primary care, endocrinology clinics, and mobile settings.

    The FDA approval of AEYE-DS as a fully autonomous solution (portable camera plus algorithm) underscores that in 2026, AI increasingly “works where the patient is,” not only where an ophthalmologist is present.

    3.2. 2026 as the Year of Integration

    Successful projects in 2026 will be distinguished by:

    • Image quality standards and quality control;
    • Clear referral rules and urgency levels;
    • Mechanisms to ensure patient follow-through (scheduling, reminders, visit tracking);
    • Transparent documentation for clinicians, patients, and audit purposes.

    3.3. AI as “Invisible Infrastructure”

    In 2026, AI increasingly functions as invisible infrastructure: highlighting high-risk cases, prioritizing queues, generating structured reports, and standardizing interpretation. The impact is reduced variability, faster routing, and fewer missed cases.

    4. Telemedicine: From Video Calls to Retinal Screening and Remote Management

    By 2026, telemedicine in ophthalmology is no longer synonymous with video consultations. Its foundation is tele-imaging: transmission and assessment of retinal images (fundus photos, sometimes OCT) with structured referral protocols.

    At the same time, limitations become more openly discussed. Certain conditions and components of assessment may be less accurately captured remotely, requiring clear protocols to define which patients can be managed remotely and which require in-person examination.

    The 2026 trend is a shift from “tool” to “pathway”:

    • Tele-screening as the first step;
    • Automated or semi-automated reporting;
    • Referral and follow-up control;

    Remote reassessment for ongoing risk monitoring.

    5. New Devices and Portable Diagnostics: Closer, Faster, More Scalable Care

    trend vote

    5.1. Portable Diagnostics as the Foundation of Coverage

    Portable fundus cameras and compact diagnostic systems represent one of the most practical changes of 2026. Their value lies not only in technology, but in enabling large-scale screening in locations without full ophthalmic infrastructure.

    Synergy with autonomous AI (such as AEYE-DS) is especially strong here, supporting new partnership models:

    • Endocrinology and primary care clinics;
    • Optical stores and optometric practices;
    • Mobile programs for workplaces or regions.

    5.2. Devices Deliver Value Only with Quality Protocols

    Success depends not just on acquiring devices, but on defined protocols:

    • Staff training in image acquisition;
    • Minimum quality criteria;
    • Retake rules;
    • Handling ungradable cases.

    In 2026, image quality becomes decisive, as AI and telemedicine depend on it.

    5.3. Home and Remote Monitoring for Extended Treatment Regimens as eye care trends 2026

    As treatment intervals lengthen, the risk of between-visit deterioration increases. Thus, 2026 strengthens the role of:

    • Home functional monitoring;
    • Digital questionnaires and symptom trackers;

    Remote checkpoints signaling the need for earlier recall.

    6. 2026 as the Year of Standardized Myopia Control and Greater Risk Awareness

    By 2026, myopia control is no longer debated but formalized, grounded in consensus documents and systematic reviews. Myopia is increasingly recognized as a chronic disease with stages, phenotypes, and potentially blinding complications.

    Implications for practice:

    1. Focus on preventing progression to high myopia.
    2. Combined strategies integrating behavioral, optical, and pharmacologic interventions with monitoring.
    3. A shared language between optometrists and ophthalmologists, with coordinated patient pathways.
    4. Support from AI and telemedicine for risk detection and personalized care.

    Myopia control in 2026 becomes a structured, long-term risk-reduction process.

    7. Optogenetics: Expanding the Evidence Horizon in Inherited Retinal Degenerations

    In 2026, optogenetics moves beyond concept into longer-term observation. Publications from 2025 highlight functional stabilization or improvement in retinitis pigmentosa, emphasizing pragmatic success criteria.

    For patients with severe vision loss, meaningful outcomes extend beyond visual acuity charts to spatial orientation, object recognition, and contrast sensitivity. In 2026, discussions increasingly focus on realistic endpoints and honest communication of limitations.

    8. Less Invasive Interventions and Patient Comfort as Components of Clinical Effectiveness

    Another key eye care trends 2026 is less traumatic technology that preserves efficacy while improving patient experience. A notable example is the FDA approval of Epioxa (epi-on) for keratoconus in 2025, preserving corneal epithelium and potentially reducing pain and recovery time.

    This trend spans refractive surgery, ocular surface disease, and chronic condition management, reinforcing that patient experience is integral to adherence and clinical outcomes.

    trends summary

    Conclusion

    The ophthalmology trends 2026 clearly demonstrate that ophthalmology and optometry are entering a phase of mature transformation, where success is driven not by isolated innovations but by their integration into coherent clinical pathways. The focus is shifting from treating consequences to early detection, slowing progression, and long-term management of chronic eye disease.

    Active treatment of geographic atrophy, photobiomodulation, extended anti-VEGF regimens, and the emergence of gene therapies for MacTel 2 and neovascular AMD fundamentally reshape patient management—from observation or frequent procedures to strategies aimed at preserving retinal structure and function with minimal procedural burden. These approaches require careful patient stratification and responsible expectation management, as the goal increasingly becomes slowing neurodegeneration rather than restoring vision.

    At the diagnostic level, 2026 reinforces decentralization: portable devices, telemedicine, and autonomous AI bring screening closer to patients and enable coverage of much broader populations. Oculomics and AI transform ocular images into sources of digital biomarkers that may influence not only ophthalmic but also general clinical management. At the same time, it becomes clear that technological value is defined not by algorithms or devices, but by data quality, model validation, and clearly structured patient pathways—from screening to treatment.

     

Recently Posted

  • glaucoma

    Glaucoma OCT Monitoring Guide: From Detection to Long-Term Care

    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min

    Glaucoma OCT Monitoring Guide: From Detection to Long-Term Care

    Table of Contents

    1. Glaucoma detection: why early diagnosis is critical
    2. How to detect glaucoma in early stages: key approaches
    3. Advanced imaging for glaucoma: OCTA
    4. OCT glaucoma monitoring after diagnosis
    5. Additional tools for monitoring glaucoma treatment
    6. Glaucoma OCT: the foundation of long-term glaucoma care

    Optical Coherence Tomography (OCT) has fundamentally changed glaucoma diagnostics over the past two decades. It enables non-invasive, micron-level imaging of retinal microstructures and provides objective measurements of the retinal nerve fibre layer (RNFL), ganglion cell complex (GCC), and optic nerve head (ONH) parameters. Moreover, the advent of OCT angiography (OCTA) has introduced a new dimension in assessing microcirculation—complementing structural analysis and potentially predicting glaucoma progression.

    Today,  OCT is the standard for early detection, monitoring, and risk stratification of glaucoma progression, as recognised in international clinical guidelines. When combined with functional tests, tonometry, and anterior chamber angle assessment, OCT becomes the foundation for personalised glaucoma management.

    Glaucoma detection: why early diagnosis is critical

    Early glaucoma diagnosis is vital, as optic nerve damage caused by the disease is irreversible. Many patients seek care only after significant vision loss has occurred, at which point treatment may slow progression but cannot restore lost function. This is why ophthalmologists emphasise the importance of glaucoma detection at preclinical or pre-perimetric stages.

    How does OCT help in early glaucoma detection?

    OCT provides high-resolution imaging of the retina and optic nerve head. Unlike subjective functional tests, OCT delivers objective, quantitative data on ganglion cells, nerve fibre layers, and the neuroretinal rim, enabling recognition of even subtle structural changes.

    Recent OCT models go further, allowing detailed visualisation of the lamina cribrosa, a structure known to be altered in glaucoma. Today, OCT is recognised as a key diagnostic tool in the guidelines of both the European Glaucoma Society and the American Academy of Ophthalmology.

    How to detect glaucoma in early stages: key approaches

    Early glaucoma detection relies on evaluating structural and functional parameters of the eye, supported by advanced imaging techniques. The three main parameters assessed with glaucoma OCT are:

    • Ganglion Cell Complex (GCC) thickness and asymmetry
    • Retinal Nerve Fibre Layer (RNFL) thickness
    • Optic nerve head parameters with the DDLS scale

    In addition, OCT Angiography (OCTA) provides complementary insights into ocular microvasculature that may indicate early glaucomatous damage.

    Glaucoma detection parameter 1: GCC thickness and asymmetry

    One of the most sensitive preclinical biomarkers of glaucomatous damage is thinning of the ganglion cell complex (GCC), which includes the ganglion cell layer (GCL), inner plexiform layer (IPL), and macular RNFL (mRNFL). It is assessed through macular OCT scans. Damage in this area is particularly critical, as 50–60% of all ganglion cells are concentrated within the central 6 mm zone.

    Assessing asymmetry between the superior and inferior halves of the macula within the GCC is a key diagnostic indicator. Studies show that minimum GCC thickness and FLV/GLV indices (Focal Loss Volume / Global Loss Volume) are predictors of future RNFL thinning or emerging visual field defects. Asymmetry maps significantly ease clinical interpretation.

    A newer approach—vector analysis of GCC loss—also allows clinicians to visualise the direction of damage, which often correlates with future visual field defects.

    Measuring Ganglion Cell Complex (GCC) Thickness and GCC Asymmetry

    Glaucoma detection parameter 2: RNFL thickness analysis

    RNFL analysis is among the most widely used glaucoma diagnostic methods. The RNFL reflects the axons of the ganglion cells and is readily measured in optic nerve scans. Temporal sectors are the most sensitive and often show the earliest changes.

    Even when the overall thickness appears normal, localised defects should raise suspicion. Sectoral thinning of ≥5–7 μm is considered statistically significant. Age-related RNFL decline (~0.2–0.5 μm/year) must also be considered.

    Glaucoma detection parameter 3: optic nerve head parameters and the DDLS scale

    Evaluating the optic nerve head (ONH) is essential. OCT enables automated assessment of optic disc area, cup-to-disc ratio (C/D), cup volume, rim area, and the lamina cribrosa.

    The Disc Damage Likelihood Scale (DDLS) classifies glaucomatous ONH changes based on the thinnest radial rim width or, if absent, the extent of rim loss. Unlike the C/D ratio, DDLS adjusts for disc size. When combined with OCT, DDLS significantly enhances objective clinical assessment.

    In high myopia, automatic ONH segmentation often misclassifies anatomy. Here, newer deep learning–based segmentation models improve accuracy.

    Evaluating the optic nerve head (ONH)

    Advanced imaging for glaucoma: OCTA

    OCT Angiography (OCTA), an advanced glaucoma OCT technique, provides unique insights into ocular circulation. It enables evaluation of:

    • Vessel density in the peripapillary region
    • Optic nerve and macular vascularisation
    • Retinal versus ONH perfusion in both eyes

    OCTA for early glaucoma detection

    Studies confirm that reduced vessel density correlates with RNFL loss and visual field deterioration, and often precedes both.

    OCT glaucoma monitoring after diagnosis

    Glaucoma can progress even with stable intraocular pressure (IOP), making regular structural assessment of the optic nerve and inner retina crucial for therapy adjustment.

    Glaucoma OCT is not only a diagnostic tool but also the primary method for monitoring glaucomatous damage. Unlike functional tests, OCT can detect even minimal RNFL or GCL thinning months or even years before visual field loss appears. With serial measurements and built-in analytics, OCT allows clinicians to track glaucoma progression rates and identify high-risk patients.

    Methods for glaucoma progression monitoring

    There are two main approaches to monitoring glaucoma progression with OCT:

    Method 1: event-based analysis

    This method compares current scans with a reference baseline, identifying whether RNFL or GCL thinning exceeds expected variability.

    ? Example: Heidelberg Eye Explorer (HEYEX) highlights suspicious areas in yellow (possible loss) or red (confirmed loss).

    Limitations include sensitivity to artifacts, image misalignment, and segmentation quality. A high-quality baseline scan is essential.

    Method 2: trend-based analysis

    This approach accounts for time. The software plots RNFL/GCL thickness trends over time in selected sectors or globally and calculates the rate of progression.

    Examples:

    • RNFL thinning >1.0 μm/year is clinically significant.
    • Thinning >1.5 μm/year indicates active progression.

    It also accounts for age-related changes, helping differentiate physiological vs. pathological decline.

    Visual assessment in glaucoma OCT

    Qualitative analysis also plays an important role in detecting glaucoma progression. Key aspects include:

    • Focal RNFL thinning (localised defects)
    • Changes in the neuroretinal rim
    • Alterations in ONH cupping
    • GCL/GCIPL comparison (superior vs. inferior) on macular maps
    • New segmentation artifacts (may mimic progression)

    Visual glaucoma OCT analysis

    OCT glaucoma findings that indicate true progression

    Five OCT findings suggest true glaucomatous progression:

    1. RNFL thinning >10 μm in one sector or >5 μm in several sectors
    2. New or worsening GCL asymmetry (yellow to red colour shift)
    3. Emerging or expanding RNFL defects on colour maps
    4. Increasing C/D ratio with concurrent rim thinning
    5. New localised areas of vessel density loss on OCTA

    Particular attention should be paid to the inferotemporal and superotemporal RNFL sectors, where 80% of early changes occur.

    Frequency of glaucoma OCT monitoring

    According to the AAO and EGS, the recommended frequency for OCT glaucoma monitoring is:

    • High-risk patients: every 6 months
    • Stable patients: once a year
    • For trend analysis: at least 6–8 scans over 2 years to ensure statistical reliability

    Looking ahead, broader use of AI for glaucoma is expected to support earlier and more accurate detection, while also reducing false positives.

    Additional tools for monitoring glaucoma treatment

    While glaucoma OCT is essential for detecting structural changes, a comprehensive glaucoma assessment requires a multimodal approach. Additional tools include perimetry, tonometry, optic disc fundus photography, and gonioscopy.

    Perimetry (visual field testing)

    Functional assessment of the optic nerve remains crucial. Standard Automated Perimetry (SAP), most often performed with Humphrey Visual Field Analyzer protocols (24-2, 30-2, 10-2), is the most widely used method.

    Key indices:

    • MD (mean deviation): average deviation from normal values
    • PSD (pattern standard deviation): highlights localised defects
    • VFI (visual field index): summarises global visual function; useful for tracking glaucoma progression
    • GHT (glaucoma hemifield test): automated analysis of field asymmetry

    ? Important: In 30–50% of cases, structural changes such as RNFL thinning on OCT precede visual field defects; in others, functional loss appears first. Best practice relies on integrated OCT and perimetry to correlate damage location and monitor glaucoma progression more precisely.

    Combined OCT and perimetry remains the gold standard for progression monitoring.

    Tonometry

    Intraocular pressure (IOP) is the only clearly modifiable risk factor associated with both glaucoma onset and progression.

    • Goldmann applanation tonometry remains the gold standard.
    • A single IOP reading is insufficient — diurnal fluctuations are an independent risk factor, particularly in normal-tension glaucoma.

    Optic disc fundus photography

    Although subjective, fundus imaging remains valuable for documenting glaucomatous changes, especially in borderline cases. Unlike OCT, it does not provide quantitative data but helps visualise morphology over time.

    What to assess:

    • Progressive disc cupping
    • Changes in neuroretinal rim shape or colour
    • Disc margin haemorrhages (linked to faster RNFL thinning and visual field loss)
    • Inter-eye comparisons

    Gonioscopy

    Gonioscopy evaluates the anterior chamber angle and helps exclude angle-closure, pigmentary, or pseudoexfoliative glaucoma. It also identifies:

    • Neovascularisation
    • Trabecular meshwork abnormalities
    • Other angle anomalies

    Patient education: a key to successful glaucoma management

    Accurate glaucoma detection and therapy are not enough; adherence to monitoring and treatment is equally critical.

    The challenge:

    • Early-stage glaucoma is asymptomatic.
    • Many patients underestimate its seriousness, leading to poor compliance, missed follow-ups, and discontinuation of therapy.

    The goals of patient education:

    • Explain that glaucoma progresses silently but can lead to irreversible blindness if untreated.
    • Use real-life examples (before/after OCT scans, visual field comparisons) to illustrate progression.
    • Teach patients to recognise warning signs (vision changes, eye pain).
    • Visualise disease progression with AI tools showing RNFL loss and future risk.

    Educational resources may include:

    • Printed brochures in patient-friendly language
    • Videos featuring OCT images with explanations
    • Doctor–patient in-clinic discussions
    • Telemedicine platforms with reminders and follow-up prompts

    According to the AAO, patients who understand glaucoma are 2.5 times more likely to adhere to treatment and attend check-ups.

    Glaucoma OCT: the foundation of long-term glaucoma care

    Glaucoma OCT now plays a central role in both diagnosis and monitoring. Its ability to detect subtle structural changes before measurable functional loss makes early intervention possible and increases the chances of preserving vision.

    But technology alone is not enough. Accurate interpretation, combined with strong patient education, is essential. When patients understand their disease and the role of glaucoma OCT in treatment, adherence improves and outcomes are better.

    OCT is not just a diagnostic device; it is the cornerstone of an integrated glaucoma management strategy, from initial screening to long-term monitoring and treatment optimisation.

    Disclaimer: USA FDA 510(k) Class II; Altris Image Management System (Altris IMS); AI/ML models and components intended to use for research purposes only, not for clinical diagnosis purposes.

  • Dry AMD Treatment: Modern Ways to Slow Progression

    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    5 min.

    Dry AMD Treatment: Modern Ways to Slow Progression

    Table of Contents

    1. What are the dry macular degeneration treatment breakthroughs?
    2. How to monitor dry AMD progression with OCT?
    3. What are the challenges of dry age-related macular degeneration monitoring?
    4. How do I organize efficient dry AMD monitoring in my clinic?
    5. Why are optometrists on the front line of early AMD detection?
    6. How can OCT insights help support patients emotionally?
    7. Conclusion

    For many years, dry or non-exudative AMD was seen as untreatable. Most research focused on wet AMD and anti-VEGF therapy.

    Today, this paradigm is shifting. Around 30% of patients with age-related macular degeneration are affected by the dry form, which makes finding effective therapies critical. Recently, the first FDA-approved drugs for dry macular degeneration injections have appeared, offering hope to patients with geographic atrophy (GA). Alongside, new physiotherapeutic methods, such as multi-wavelength photobiomodulation, are showing promising results.

    Geographic atrophy (GA) is an advanced, irreversible form of dry AMD. It occurs when parts of the retina undergo cell death, leading to progressive vision loss. But even the best dry AMD treatment is incomplete without objective measurement. That’s where modern tools for macular degeneration monitoring come in, and optical coherence tomography (OCT) is now at the core of this process.

     

    What are the dry macular degeneration treatment breakthroughs?

    The latest dry macular degeneration treatment breakthroughs include:

    • Multiwavelength photobiomodulation
    • FDA-approved injectable drugs
    • AREDS 2-based supplements

    In the past, recommendations focused only on reducing risks — quitting smoking, managing blood pressure, and eating a healthy diet.
    Now, new approaches to dry AMD treatment combine prevention with active therapies to slow AMD progression and especially the advance of GA.

    1. Dry AMD treatment using multiwavelength photobiomodulation

    Multiwavelength photobiomodulation for AMD is a promising new treatment. It uses specific red and near-infrared light wavelengths (~590–850 nm) and helps reduce oxidative stress, inflammation, and pigment epithelial cell death.

    One of the best-known systems is Valeda Light Therapy, which delivers controlled multiwavelength light directly to the retina.

    The LIGHTSITE III clinical trial showed that photobiomodulation can slow the decline in visual acuity and reduce the rate of GA expansion.

    Limitations:

    • Only 3–5 years of long-term data available
    • Requires costly equipment and training
    • Effectiveness in late-stage GA remains unclear

    Dry Macular Degeneration Treatment Breakthroughs: Multiwavelength photobiomodulation

    2. Dry AMD treatment using FDA-approved injectable drugs

    AMD injection drugs approved by the FDA include Izervay and Syfovre.

    • Izervay (avacincaptad pegol): A C5 complement protein inhibitor that targets the complement cascade involved in chronic retinal inflammation and damage. Izervay, approved for geographic atrophy secondary to dry AMD, has demonstrated a reduced rate of GA progression in clinical trials.
    • Syfovre (pegcetacoplan): A C3 complement inhibitor that blocks the central component of the complement system to reduce inflammation. Syfovre is the first FDA-approved treatment for GA that targets complement component C3, showing a clinically meaningful slowing of GA progression.

    Both dry macular degeneration injections have shown the ability to slow GA progression compared to placebo. Although they do not restore vision, slowing vision loss is a meaningful clinical outcome.

    Key considerations for injections:

    • Administered intravitreally, usually monthly or every other month
    • Require doctor training and patient education on risks (e.g., endophthalmitis, increased intraocular pressure)
    • Cost and access may limit use

    Dry macular degeneration injections

    3. Dry AMD treatment using AREDS 2-based supplements

    AREDS 2 supplements are antioxidant supplements containing lutein, zeaxanthin, vitamins C and E, zinc, and copper. They can reduce the risk of progression to late-stage AMD by around 25% over five years, according to the AREDS 2 study.

    Pros:

    • Widely available
    • Safe, with low side effect risk
    • Supported by strong clinical evidence

    Cons:

    • Do not directly treat GA
    • Cannot replace active therapies such as dry macular degeneration injections or photobiomodulation

    How to monitor dry AMD progression with OCT

    Effective macular degeneration monitoring relies on OCT. It is the gold standard for tracking retinal changes and predicting GA development.
    Without OCT, clinicians are essentially “flying blind” when assessing AMD progression.

    Key monitoring parameters of AMD progression

    The key monitoring parameters of AMD progression include GA area, drusen, and distance to fovea.

    1. GA area

    This is the main metric when using intravitreal eye injections. Modern OCT systems provide GA measurements in mm², allowing doctors to objectively track changes over time.

    Even if patients don’t notice symptoms, a growing GA area signals disease progression. In FDA trials for Syfovre and Izervay, the GA area was the primary endpoint.

    2. Drusen

    Drusen vary in number, size, and shape. A reduction or disappearance of drusen on OCT may seem like an improvement, but could actually indicate a transition to the atrophic stage. Regular monitoring helps detect this early.

    3. Distance to fovea

    The closer GA is to the fovea, the greater the risk of sudden vision loss.

    Early detection enables:

    • Referral to an ophthalmologist
    • Timely conversations about potential vision loss

    OCT outputs for AMD progression monitoring and communication

    Useful OCT outputs for AMD progression monitoring and communication are heat maps and progress charts.

    1. Heat maps

    Modern OCT systems use color-coded heat maps to show pigment epithelium thickness and drusen distribution. This visual format helps in several ways:

    • Makes interpretation easier for clinicians
    • Helps patients better understand their condition
    • Encourages patients to stay engaged with treatment

    In clinical practice, it serves as a highly effective communication tool.

     

    2. Progress charts

    Most OCT systems can compare results across visits

    • For doctors: Helps guide treatment decisions
    • For patients: Provides visual proof of stabilization or worsening

     

    The role of objective evidence in patient treatment

    Patients may question the value of long-term treatments or costly procedures.

    OCT is the gold standard for patient motivation. When patients see actual changes, they’re more likely to agree to treatment.

    What are the challenges of macular degeneration monitoring?

    Monitoring dry AMD presents technical, organizational, and psychological challenges. Doctors of all levels of experience should be aware of them.

    1. Invisible microchanges

    Early atrophy or drusen changes may be subtle. Patients may not notice them due to eccentric fixation or slow adaptation.

    Without OCT, doctors may miss early GA, delaying treatment.

    It is necessary to perform OCT even when there are only minor changes in visual acuity or if the patient reports image distortion (metamorphopsia).

    2. Subjective assessment

    Ophthalmoscopy reveals only obvious changes. Subtle drusen or early atrophy might be missed.

    Relying on patients’ complaints is risky — many don’t notice issues until it’s too late.

    That’s why even small optical practices should establish clear referral pathways for OCT exams.

    3. Unnecessary referrals

    Optometrists or primary care doctors often refer patients to ophthalmologists “just in case,” because they don’t have access to OCT or lack experience interpreting it.

    This puts unnecessary strain on specialists. In many cases, nothing new is done after the exam because there are no previous images for comparison.

    4. Limitations of OCT devices

    Not all OCT devices measure GA or track drusen equally well. Older models may lack automated measurements of atrophy area.

    In some cases, referral to a center with advanced OCT is necessary.

    OCT devices used to monitor AMD progression

    How do I organize efficient dry AMD monitoring in my clinic?

    Practical tips:

    1. Create a baseline chart with OCT images during the first visit.

    2. Monitor regularly:

    • Every 6–12 months in the early stages
    • Every 3–6 months with GA
    • Before each intravitreal injection

    3. Standardise scanning protocols to minimise variability.

    4. Use OCT software tools for image comparison, GA calculation, heat maps.

    5. Communicate clearly with patients about drusen, atrophy, and treatment goals.

    Why are optometrists on the front line of early AMD detection?

    Optometrists play a key role in spotting the early signs of AMD, as they are often the first point of contact in eye care.

    They perform initial screenings, provide guidance on lifestyle and supplements, and ensure regular OCT monitoring.

    If drusen, pigment epithelial changes, or signs of GA are present, they refer patients to ophthalmologists for confirmation and treatment planning.

    How can OCT insights help support patients emotionally?

    Patients with dry AMD often ask: “Why bother if it can’t be cured?”
    Here, OCT plays an emotional as well as clinical role. Showing OCT scans can:

    • Prove the value of slowing AMD progression
    • Emphasise patients’ role in preserving sight
    • Reassure them that long-term care makes a difference

     

    Dry macular degeneration treatment breakthroughs: key takeaways for slowing AMD progression

    Modern dry macular degeneration treatment breakthroughs, including FDA-approved injections, photobiomodulation, and AREDS 2 supplements,  have changed the outlook for patients.
    Yet treatment alone is not enough. Without consistent macular degeneration monitoring using OCT, the benefits of these therapies may be lost.

    The future of dry AMD treatment lies in a partnership between optometrists, ophthalmologists, and patients. Together, with breakthrough therapies and precise monitoring, we can slow AMD progression and give patients the best chance of preserving vision.

    Disclaimer: USA FDA 510(k) Class II; Altris Image Management System (Altris IMS); AI/ML models and components intended to use for research purposes only, not for clinical diagnosis purposes.

  • AItris for Buchanan Optometrists

    AI Ophthalmology and Optometry | Altris AI Mark Braddon
    3 min.

    Disclaimer: In the USA, Altris Image Management System (Altris IMS) has USA FDA 510(k) Class II clearance; AI/ML models and components are intended to use for research purposes only, not for clinical diagnosis purposes.

    Buchanan Optometrists and Audiologists is no ordinary eye-care center.

    The Association of Optometrists (AOP) estimates 17,500 registered optometrists working across roughly 6,000 practices in the UK. The UK Optician Awards recognise the best in the UK Optical industry.  To even make the top 5 is our equivalent of an Oscar nomination! They are the only practice in the UK to consistently make the top 5 since 2008. Buchanan Optometrists describe themselves as innovators who “continually push boundaries.”

    Their list of awards speaks for itself:

    • 2012 – National Optician Award for Premium Lens Practice of the Year
    • 2013 – Luxury Eyewear Retailer of the Year and Premium Lens Practice of the Year
    • 2013 – Winner at the UK Optician Awards
    • 2015–2016 – Best UK Independent Practice
    • 2017–2018 – Optometrist of the Year, with Alisdair Buchanan named the top optometrist in the UK
    • 2023–2024 – Best Independent Optician and Best Technology Practice

    And this list is not finished, as Alisdair Buchanan, the Owner and the Director of the center, is investing in their growth continuously.

    Buchanan Optometrists are being recognized for their achievements

    With a track record like this, it’s no surprise that Buchanan Optometrists was among the first to adopt AI for Decision Support in OCT. AI is rapidly becoming a vital part of modern eye care, and leading centers are already embracing it.

    Mark Braddon, Altris AI VP of Clinical Sales, sat down with Alisdair Buchanan, the owner and director of the practice, to talk about his experience with AI and what it means for the future of optometry.

    Mark Braddon: You’ve been working with OCT for years. What changed in your practice after bringing in Altris AI Decision Support for OCT?

    Alisdair Buchanan, Owner: As someone already confident in interpreting scans, I didn’t need help understanding OCT—but Altris provides something even more valuable: a kind of second opinion. It supports my clinical decisions and offers an added layer of reassurance, particularly in borderline or complex cases. That’s not just helpful—it’s powerful.

    I didn’t think our OCT assessments could improve much—until we started using Altris AI. It’s not just an upgrade; it’s become an indispensable part of delivering modern, high-quality eye care. Altris AI has significantly enhanced the way we interpret OCT scans. What used to require prolonged focus and cross-referencing now takes moments, without sacrificing accuracy or depth. The system analyses images with incredible precision, highlighting subtle pathological changes that are often time-consuming to detect, especially during a busy clinic day.

    Mark Braddon: What was the first real benefit you noticed after bringing  Altris AI into your day-to-day routine?

    Alisdair Buchanan, Owner: One of the most immediate benefits has been in patient communication. The platform generates clear, colour-coded visuals that make explaining findings effortless. Instead of trying to talk patients through grainy greyscale images, we can now show them precisely what we’re seeing. It’s improved understanding, reduced anxiety, and increased trust in the care we’re providing.

    Mark Braddon: Was it easy to fit AI Decision Support into your OCT workflow? How easy did you find integrating Altris AI?

    Alisdair Buchanan, Owner: Integration was seamless—no faff, no friction. It fits naturally into our existing workflow, with scans uploaded and analysed within seconds. It’s helped us work more efficiently, without compromising the thoroughness our patients expect.

    In short, Altris AI has sharpened our clinical edge and strengthened the service we offer. It doesn’t replace experience—it enhances it. And that, for me, is the real value.

    Mark Braddon: In your experience, where has AI been the most helpful in clinical work?

    Alisdair Buchanan, Owner: The main area where it shines is in picking up early macular changes, particularly dry AMD. Things like drusen or subtle changes in the outer retinal layers, which could easily be missed at a glance, are brought to the surface immediately.

    It’s also been handy with diabetic patients. Just having that extra layer of input to flag microstructural changes helps us stay ahead of progression.

    We’ve also started using it with glaucoma suspects. While our Heidelberg Spectralis remains our go-to for structural monitoring, having the RNFL analysis from Altris adds a checkpoint. I’d never base a referral purely on it, but it’s nice to have a second opinion—even if it’s an AI one.

    Mark Braddon: Has AI Decision Support changed how you handle borderline or difficult-to-call cases?

    Alisdair Buchanan, Owner: I’d say it’s given us more confidence, particularly in the grey areas—those borderline cases where you’re not quite sure if it’s time to refer or just monitor a bit more closely. With AMD, for example, it has helped us catch early signs of progression and refer patients before things become urgent.

    And for glaucoma, again, it’s not replacing anything we do—it’s just another tool we can lean on. Sometimes it confirms what we already thought, and other times it nudges us to look again more carefully.

    Mark Braddon: How has using AI impacted your conversations with patients during consultations?

    Alisdair Buchanan, Owner: One of the unexpected benefits has been how much it helps with patient conversations. We show the scans on-screen during the consultation, and the colour overlays make things much easier to explain, especially with older patients. They can see what we’re talking about, which makes the whole thing feel more real and less abstract.

    They often say, “Ah, now I understand,” or “So that’s what you’re looking at.” It’s not about dazzling them with tech—it just helps make the discussion more transparent and more reassuring.

    Mark Braddon: Some professionals worry that AI might replace human judgment. How do you see its role in clinical decision-making?

    Alisdair Buchanan, Owner: I don’t see Altris —or any AI—as a threat to what we do. It’s not here to replace us. We still make the decisions, take responsibility, and guide our patients. But it does help.

    For me, it’s like having a quiet assistant in the background. It doesn’t get everything right, and I certainly wouldn’t act on it blindly—but it prompts me to pause, double-check, and sometimes spot something I might have missed otherwise. That can only be a good thing.

    In short, Altris has sharpened our clinical edge and strengthened the service we offer. It doesn’t replace experience—it enhances it. And that, for me, is the real value.

  • AI for Decision Support for OCT

    AI for Decision Support with OCT: “Altris Gave Me More Certainty in My Clinical Decisions”

    AI Ophthalmology and Optometry | Altris AI Maria Martynova
    2 minutes

    Disclaimer: In the USA, Altris Image Management System (Altris IMS) has USA FDA 510(k) Class II clearance; AI/ML models and components are intended to use for research purposes only, not for clinical diagnosis purposes.

    AI for Decision Support with OCT: An Interview with Clara Pereira, Optometrist from Franco Oculista

    About Franco Oculista Optometry in Portugal.

    Franco Oculista is the optometry center with a 70-year-old history: its roots date back to the mid-1950s in Luanda, where it was founded by Gonçalo Viana Franco. Having left behind a career in pharmacy, Gonçalo pursued his entrepreneurial vision by opening an optician’s bearing his name in the heart of the Angolan capital. Driven by a thirst for knowledge and a deep sense of dedication, he turned his dream into reality. With a commitment to professionalism and a forward-thinking approach, he integrated the most innovative technologies available at the time. This blend of passion, expertise, and innovation established Franco Oculista as a benchmark for quality and excellence in the field. In 1970s, the family returned to Portugal and opened the new FRANCO OCULISTA space on Avenida da Liberdade.

    How do Franco Oculista describe their mission?

    “Through individualized and segmented service, we seek to respond to the needs of each client. We combine our knowledge with the most sophisticated technical equipment and choose quality and reliable brands. We prioritize the evolution of our services and, for this reason, we work daily to satisfy and retain our customers with the utmost professionalism.”

    Clara Pereira is one of the optometrists at Franco Oculista and has been an optometrist for nearly two decades. Based in a private clinic in Portugal, she brings years of experience and calm confidence to her consultations. We talked with her to learn how her clinical practice has evolved, particularly since integrating OCT and, more recently, Altris AI – AI for Decision Support with OCT.

    Altris AI: Clara, can you tell us a bit about your daily work?

    Clara: “Of course. I’ve been working as an optometrist for 19 years now. My practice is quite comprehensive—I assess refractive status, binocular vision, check the anterior segment with a slit lamp, measure intraocular pressure, and always examine the fundus.

    Clara: “In Portugal, we face limitations. We’re not allowed to prescribe medication or perform cycloplegia, so imaging becomes crucial. I rely heavily on fundus photography and OCT to guide referrals and detect early pathology.”

    Altris AI: How central is OCT diagnostics to your workflow?
    Clara: “OCT is substantial. I perform an OCT exam on nearly every patient, on average, eight OCT exams per day. It’s an essential part of how I gather information. With just one scan, I can learn so much about eye health.”

    Altris AI: What kind of conditions do you encounter most frequently?
    Clara: “The most common diagnosis is epiretinal membrane—fibrosis. But I also manage patients with macular degeneration and other retinal pathologies. Having the right tools is key.”

    Altris AI: And what OCT features do you use the most?
    Clara: “I regularly use the Retina, Glaucoma, and Macula maps. But if I had to choose one, the Retina Map gives me the most complete picture. It’s become my go-to.”

    Altris AI: You’ve recently started using Altris AI. What has that experience been like?
    Clara: “At first, I didn’t know much about it. But when Optometron introduced Altris AI to me—a company I trust—I didn’t hesitate. And I’m glad I didn’t. From the beginning, it felt like a natural extension of my clinical reasoning.

    Clara: “Altris AI gives me an extra layer of certainty. It helps me extract more from the OCT images. I usually interpret the scan myself first, and then I run it through the platform. That way, I validate my thinking while also learning something new.”

    Altris AI: Have any standout cases where Altris AI made a difference?

    Clara: “Yes. I’ve had a few. One was a case of advanced macular degeneration, in which the AI visualization really helped me explain the condition to the patient. Another was using anterior segment maps for fitting scleral lenses—Altris was incredibly useful there, too. I do a lot of specialty lens fittings, so that was a big advantage.”

    Altris AI: Would you recommend Altris AI to your colleagues?

    Clara: “I would recommend Altris AI to my colleagues. For me, it’s about more than just the diagnosis. It’s about feeling confident that I’m seeing everything clearly and giving my patients the best care possible. Altris AI helps me do exactly that.”

    Why This Matters: Altris AI in Real Practice

    Clara’s story reflects the real value of AI in optometry—not as a replacement for clinical judgment, but as a powerful companion. With every OCT scan, she strengthens her expertise, improves diagnostic accuracy, and gives her patients the reassurance they deserve.

    Whether identifying early signs of fibrosis, supporting complex scleral lens fittings, or acting as a second opinion, Altris AI seamlessly fits into the modern optometrist’s workflow, making every scan more meaningful.

    AI for Decision Support with OCT: Transforming Retinal Diagnostics

    Artificial Intelligence (AI) is revolutionizing the field of ophthalmology, particularly through its integration with Optical Coherence Tomography (OCT). OCT is a non-invasive imaging technique that captures high-resolution cross-sectional images of the retina, enabling early detection and monitoring of various ocular conditions. However, interpreting these scans requires time, expertise, and consistency—factors that AI-based decision support systems are uniquely positioned to enhance.

    Altris AI (AI for OCT decision support platform) analyzes thousands of data points across B-scans, automatically detecting retinal pathologies, quantifying biomarkers, and identifying patterns that may be subtle or overlooked by the human eye. By providing objective, standardized assessments, Altris AI reduces diagnostic variability and improves clinical accuracy, especially in busy or high-volume practices.

    For optometrists and ophthalmologists, AI acts as a second opinion, flagging early signs of diseases such as age-related macular degeneration (AMD), diabetic retinopathy, and glaucoma. It streamlines workflows by highlighting areas of concern, prioritizing cases that require urgent attention, and offering visual explanations that are easy to communicate to patients.

    Moreover, Altris AI enableS longitudinal tracking of pathology progression. By comparing OCT scans over time ( even from various OCT devices), clinicians can monitor subtle changes in drusen volume, retinal thickness, supporting timely clinical decisions and tailored treatment strategies. The integration of AI into OCT interpretation not only enhances diagnostic confidence but also supports evidence-based care, early intervention, and improved patient outcomes. As AI continues to evolve, it will play a vital role in advancing precision medicine in ophthalmology, empowering eye care professionals with tools that are fast, reliable, and scalable.

    In essence, AI for OCT decision support is not replacing clinical expertise; it is augmenting it, elevating the standard of care through speed, accuracy, and actionable insights.

  • future of ophthalmology

    Future of Ophthalmology: 2025 Top Trends

    AI Ophthalmology and Optometry | Altris AI Maria Znamenska
    13.03.2025
    12 min read

    Future of Ophthalmology: 2025 Top Trends

    In a recent survey conducted by our team, we asked eye care specialists to identify the most transformative trends in ophthalmology by 2025. The results highlighted several key areas, with artificial intelligence (AI) emerging as the clear frontrunner, cited by 78% of respondents.

    future of Ophthalmology

    However, the survey also underscored the significant impact of optogenetics, novel AMD/GA therapies, and the continuing evolution of anti-VEGF treatments. This article will explore the practical implications of these advancements, providing an overview of how they are poised to reshape diagnosis, treatment, research, and, ultimately, patient outcomes in ophthalmology.

    In this article, we will also discuss Oculomics, a very promising field that is gaining momentum.

    Top AI Technology for Detecting Eye-related Health Risks 2025

    Building upon the survey’s findings, we begin with the most prevalent trend: top AI technology for detecting eye-related health risks in 2025

    future of opthalmology

    AI in Clinical Eye Care Practice

    With the increasing prevalence of conditions like diabetic retinopathy and age-related macular degeneration, there is a growing need for efficient and accurate screening tools. And AI is already valuable for eye-care screening: algorithms can analyze retinal images and OCT scans to identify signs of these diseases, enabling early detection and timely intervention.

    future of ophthalmology

    Source

    AI-powered screening tools can also help identify rare inherited retinal dystrophies, such as Vitelliform dystrophy and Macular telangiectasia type 2. These conditions can be challenging to diagnose, but AI algorithms can analyze retinal images to detect subtle signs that human observers may miss.

    AI also starts to play a crucial role in glaucoma management. Early detection of glaucoma demands exceptional precision, as the early signs are often subtle and difficult to detect. Another significant challenge in glaucoma screening is the high rate of false positive referrals, which can lead to unnecessary appointments in secondary care and cause anxiety for patients, yet delayed or missed detection of glaucoma results in irreversible vision loss for millions of people worldwide. So, automated AI-powered glaucoma analysis can offer transformative potential to improve patient outcomes.

    This OD module evaluates optic disc parameters using OCT, providing personalized assessments by accounting for individual disc sizes and angle of rim absence. Such a tailored approach eliminates reliance on normative databases, making evaluations more accurate and patient-specific.

    Furthermore, it enables cross-evaluation across different OCT systems, allowing practitioners to analyze macula and optic disc pathology, even when data originates from multiple OCT devices. Key parameters evaluated by Altris AI’s Optic Disc Analysis include disc area, cup area, cup volume, minimal and maximum cup depth, cup/disc area ratio, rim absence angle, and disc damage likelihood scale (DDLS).

    future of ophthalmology

     

    AI for Clinical Trials and Research

    AI is revolutionizing clinical trials and research in ophthalmology. One such key application of AI is biomarker discovery and analysis. Algorithms can analyze large datasets of medical images, such as OCT scans, to identify and quantify biomarkers for various eye diseases. These biomarkers can be used to assess disease progression, monitor treatment response, and predict clinical outcomes.

    AI is also being used to improve the efficiency and effectiveness of clinical trials. By automating the process of identifying eligible patients for clinical trials, AI can help researchers recruit participants more quickly and ensure that trials include appropriate patient populations, accelerating the development of new treatments.

    future of ophthalmology

    Algorithms can analyze real-world data (RWD) collected from electronic health records and other sources to generate real-world evidence (RWE). RWE provides valuable insights into disease progression, treatment patterns, and long-term outcomes in everyday clinical settings, complementing the findings of traditional randomized controlled trials.

    Oculomics

    Integrating digitized big data and computational power in multimodal imaging techniques has presented a unique opportunity to characterize macroscopic and microscopic ophthalmic features associated with health and disease, a field known as oculomics. To date, early detection of dementia and prognostic evaluation of cerebrovascular disease based on oculomics has been realized. Exploiting ophthalmic imaging in this way provides insights beyond traditional ocular observations.

    future of ophthalmology

    For example, the NeurEYE research program, led by the University of Edinburgh, is using AI to analyze millions of anonymized eye scans to identify biomarkers for Alzheimer’s disease and other neurodegenerative conditions. This research can potentially revolutionize early detection and intervention for these devastating diseases.

    Another effort spearheaded by researchers from Penn Medicine, Penn Engineering is exploring the use of AI to analyze retinal images for biomarkers indicative of cardiovascular risk. AI systems are being trained on fundus photography to detect crucial indicators, such as elevated HbA1c levels, a hallmark of high blood sugar, and a significant risk factor for both diabetes and cardiovascular diseases.

    future of ophthalmology

    Source

    AI analysis of retinal characteristics, such as retinal thinning, vascularity reduction, corneal nerve fiber damage, and eye movement, has shown promise in predicting Neurodegenerative diseases. Specifically, decreases in retinal vascular fractal dimension and vascular density have been identified as potential biomarkers for early cognitive impairment, while reductions in the retinal arteriole-to-venular ratio correlate with later stages.

    Moving from AI, we now turn to another significant trend identified in our survey:

    Optogenetics

    Optogenetics represents a significant leap forward in ophthalmic therapeutics, offering a potential solution for vision restoration in patients with advanced retinal degenerative diseases, where traditional gene therapy often falls short. While gene replacement therapies are constrained by the need for viable target cells and the complexity of multi-gene disorders like retinitis pigmentosa (RP), optogenetics offers a broader approach.

    future of ophthalmology

    This technique aims to circumvent the loss of photoreceptors by introducing light-sensitive proteins, known as opsins, into the surviving inner retinal cells and optic nerve, restoring visual function through light modulation. This method is particularly advantageous as it is agnostic to the specific genetic cause of retinal degeneration.

    By delivering opsin genes to retinal neurons, the technology enables the precise manipulation of cellular activity, essentially transforming these cells into new light-sensing units. This approach can bypass the damaged photoreceptor layer, transmitting visual signals directly to the brain.

    Several companies are pioneering advancements in this field. RhyGaze, for example, has secured substantial funding to accelerate the development of its lead clinical candidate, a novel gene therapy designed for optogenetic vision restoration. Their efforts encompass preclinical testing, including pharmacology and toxicology studies, an observational study to define clinical endpoints, and a first-in-human trial to assess safety and efficacy. The success of RhyGaze’s research could pave the way for widespread clinical applications, significantly impacting the treatment of blindness globally.

    future of ophthalmology

    Source

    Nanoscope Therapeutics is also making significant strides with its MCO-010 therapy. This investigational treatment, administered through a single intravitreal injection, delivers the Multi-Characteristic Opsin (MCO) gene, enabling remaining retinal cells to function as new light-sensing cells. Unlike earlier optogenetic therapies that required bulky external devices, MCO-010 eliminates the need for high-tech goggles, simplifying the treatment process and enhancing patient convenience. The ability to restore light sensitivity without external devices represents a major advancement, potentially broadening the applicability of optogenetics to a wider patient population.

    future of ophthalmology

    Source

    Another critical area of innovation highlighted in our survey is the advancement of treatments for AMD and GA.

    New AMD/GA Treatment

    Age-related macular degeneration (AMD) and geographic atrophy (GA) represent a significant challenge in ophthalmology, demanding innovative therapeutic strategies beyond the established anti-VEGF paradigm.

    future of ophthalmology

    Source

    Gene Correction

    Gene editing is emerging as a powerful tool in the fight against AMD and GA, potentially correcting the underlying genetic errors that contribute to these diseases. Essentially, it allows us to make precise changes to a patient’s DNA.

    Traditional gene editing techniques often rely on creating ‘double-strand breaks’ (DSBs) in the DNA at specific target sites, which are like precise cuts in the DNA strand. These cuts are made using specialized enzymes, like CRISPR-Cas9, which act as molecular scissors. While effective, these methods can sometimes introduce unwanted changes at the cut site, such as small insertions or deletions.

    After a DSB is made, the cell’s natural repair mechanisms kick in. There are two main pathways:

    • Non-Homologous End Joining (NHEJ): This is the cell’s quick-fix method. It essentially glues the broken ends back together. However, this process can sometimes introduce errors, leading to small insertions or deletions that can disrupt the gene’s function.
    • Homology-Directed Repair (HDR): This is a more precise repair method. It uses a ‘donor’ DNA template to guide the repair process, ensuring accuracy. However, HDR is more complex and less efficient, especially in non-dividing cells.

    To overcome these limitations of traditional gene editing, researchers have developed more precise techniques:

    • Base Editing: This technique allows scientists to change a single ‘letter’ in the DNA code without creating DSBs.
    • Prime Editing: This advanced technique builds upon CRISPR-Cas9, allowing for a wider range of precise DNA changes. It can correct most disease-causing mutations with enhanced safety and accuracy.
    • CASTs (CRISPR-associated transposases): This method enables larger DNA modifications without creating DSBs, offering a safer approach to genetic correction.

    Why does this matter for AMD and GA? These advancements in gene editing are crucial for addressing the genetic roots of these pathologies. We can potentially develop more effective and targeted therapies by precisely correcting the faulty genes that contribute to these diseases. The technologies are still being researched, but they hold great promise for the future of ophthalmology.

    Cell Reprogramming

    Cell reprogramming offers a novel approach to regenerative medicine, with the potential to replace damaged retinal cells. This technique involves changing a cell’s fate, either in vitro or in vivo. In vitro reprogramming involves extracting cells, reprogramming them in a laboratory, and then transplanting them back into the patient. In vivo reprogramming, which directly reprograms cells within the body, holds particular promise for retinal diseases. This approach has succeeded in preclinical studies, demonstrating the potential to restore vision in conditions like congenital blindness.

    future of ophthalmology

    Vectors and Delivery Methods

    The success of gene therapy relies on efficiently delivering therapeutic genes to target retinal cells. Vectors are essentially delivery vehicles, designed to carry therapeutic genes into cells. These vectors can be broadly classified into two categories: viral and non-viral. Vectors, both viral and non-viral, are crucial for this process.

    Viral vectors are modified viruses that have been engineered to remove their harmful components and replace them with therapeutic genes. They are highly efficient at delivering genes into cells, as they have evolved to do just that. Adeno-associated viruses (AAVs) are the most commonly used viral vectors in ocular gene therapy due to their safety profile and cell-specificity. The diversity of AAV serotypes allows for tailored gene delivery to specific retinal cell types.

    Non-viral vectors, on the other hand, are synthetic systems that don’t rely on viruses. They can be made from lipids, polymers, or even DNA itself. While they may be less efficient than viral vectors, they offer safety and ease of production advantages.

    Advances in vector design, whether viral or non-viral, are focused on enhancing gene expression, cell-specificity, and carrying capacity.

    Now, let’s examine the ongoing evolution of anti-VEGF treatments, a cornerstone of modern retinal care.

    New Anti-VEGF drugs

    The landscape of ophthalmology has undergone a dramatic transformation since the early 1970s when Judah Folkman first proposed the concept of tumor angiogenesis. His idea sparked research that ultimately led to the identification of vascular endothelial growth factor (VEGF) in 1989 and the development of anti-VEGF therapies, revolutionizing the treatment of neovascular eye diseases, dramatically improving outcomes for patients with wet AMD, diabetic retinopathy, and retinal vein occlusions.

    Population-based studies have shown a substantial reduction (up to 47%) in blindness due to wet AMD since the introduction of anti-VEGF therapies. However, significant gaps remain despite this progress, especially regarding treatment durability. Anti-VEGF drugs require frequent intravitreal injections, which can be difficult for patients due to time commitments, financial costs, and potential discomfort. Although newer agents have extended treatment intervals, patient adherence and undertreatment challenges persist in real-world settings. Innovative approaches are being investigated to address these unmet needs to increase drug durability and reduce the treatment burden.

    Tyrosine Kinase Inhibitors

    One approach to increasing treatment durability is using tyrosine kinase inhibitors (TKIs). TKIs are small-molecule drugs that act as pan-VEGF blockers by binding directly to VEGF receptor sites inside cells, offering a different action mechanism than traditional anti-VEGF drugs that target circulating VEGF proteins.

    Currently, TKIs are being investigated as maintenance therapy, primarily in conjunction with sustained-release delivery systems. Two promising TKIs for retinal diseases are axitinib and vorolanib. In a bioresorbable hydrogel implant, Axitinib is being studied for neovascular AMD and diabetic retinopathy. Vorolanib, in a sustained-release delivery system, is also being investigated for neovascular AMD. These TKIs offer the potential for less frequent dosing, reducing the treatment burden for patients.

    Port Delivery System

    The Port Delivery System (PDS) is a surgically implanted, refillable device that provides continuous ranibizumab delivery for up to 6 months. While it’s FDA-approved for neovascular AMD, it’s also being investigated for other retinal diseases, such as diabetic macular edema and diabetic retinopathy.

    future of ophthalmologySource

    Although the PDS faced a voluntary recall due to issues with septum dislodgment, it has returned to the market with modifications. The PDS offers the potential for significantly reduced treatment frequency for a subset of patients. However, challenges remain, including the need for meticulous surgical implantation and the risk of endophthalmitis.

    Nanotechnology

    Nanotechnology offers promising solutions to overcome limitations of current ocular drug delivery. The unique structure of the eye, with its various barriers, poses challenges for drug delivery. Topical administration often fails to achieve therapeutic concentrations, while frequent intravitreal injections carry risks. Nanotechnology can improve drug solubility, permeation, and bioavailability through nanoparticles, potentially extending drug residence time and reducing the need for frequent injections. Several nanoparticle systems, lipid and polymeric, are being studied for ocular drug delivery, offering hope for more effective and less invasive treatments.

    Summing up

    The advancements discussed in this article, encompassing AI, optogenetics, novel AMD/GA therapies, and refined anti-VEGF treatments, collectively signal a transformative era for ophthalmology. As highlighted by the survey results, AI probably encompasses most of the changes by redefining diagnostic and clinical workflows through its capacity for image analysis, biomarker identification, and personalized patient management.

    Optogenetics offers a distinct pathway to vision restoration, bypassing limitations of traditional gene therapy. The progress in AMD/GA treatments, particularly gene editing and cell reprogramming, presents opportunities for targeted interventions. Finally, the evolution of anti-VEGF therapies, with innovations in drug delivery and sustained-release mechanisms, addresses persistent challenges in managing neovascular diseases.

    These developments, driven by technological innovation and clinical research, promise to enhance patient outcomes and reshape the future of ophthalmic care.

     

     

    Disclaimer: USA FDA 510(k) Class II; Altris Image Management System (Altris IMS); AI/ML models and components intended to use for research purposes only, not for clinical diagnosis purposes.

  • ML Applied to 3D Optic Disc Analysis for Glaucoma Risk Assessment Across Different OCT Scan Protocols Without a Normative Database

    AI Ophthalmology and Optometry | Altris AI Angelina Hramatik
    14.02.2025
    20 min read

    Machine Learning Applied to 3D Optic Disc Analysis for Glaucoma Risk Assessment Across Different OCT Scan Protocols Without a Normative Database

    1. Introduction

    Glaucoma is one of the leading causes of irreversible blindness worldwide, affecting millions of people annually. The disease is often asymptomatic in its early stages, making timely diagnosis particularly challenging. Early detection of glaucomatous changes is crucial for preventing vision loss and improving long-term patient outcomes. 

    One well-established method for assessing glaucoma is the Disc Damage Likelihood Scale (DDLS), which evaluates structural changes in the optic nerve head (ONH) based on the extent of neuroretinal rim loss. This method categorizes glaucomatous damage severity by analyzing the relationship between the optic cup and neural rim, while also accounting for optic disc size without relying on a normative database. 1, 2, 3, 4. 

    While DDLS is recognized for its reliability and utility in clinical practice, it is not a standalone diagnostic tool. Rather, it is one of several methods used to identify signs of glaucoma, and its implementation is often limited to specific imaging modalities or scan protocols, such as 3D optic disc-only scans or fundus images. 

    In this article, we introduce an enhanced approach to DDLS analysis that overcomes these limitations. We want to present a solution, which is capable of performing DDLS analysis on any OCT scan protocol that captures the optic nerve, including 3D optic disc scans (which provide the most detailed view of the nerve), as well as OCT horizontal and vertical 3D wide scans. By leveraging advanced machine learning models, we achieve unprecedented flexibility and accuracy, ensuring reliable analysis across different scanning protocols and OCT systems. 

    Unlike traditional systems restricted to specific devices or data formats, our solution processes scans from multiple OCT systems. Moreover, it excels in challenging scenarios, providing clinicians with a robust and versatile tool for analyzing potential signs of glaucoma. 

    A Brief Theoretical Overview 

    Optical coherence tomography (OCT) scans vary in the anatomical regions they capture. One specific type is the optic disc OCT scan (Figure 2), which provides high-resolution imaging of the optic disc and the surrounding optic nerve head (ONH) structures. This scan type is commonly used in glaucoma assessment, as it allows for the evaluation of the optic nerve’s structure, including the neuroretinal rim, optic cup, and surrounding peripapillary retinal nerve fiber layer (RNFL) — key areas affected in glaucomatous damage. 

    disc likelihood damage oct

    Figure 1. Photograph of the retina of the human eye, with overlay diagrams showing the positions and sizes of the macula, fovea, and optic disc (Reference). 

    disc likelihood damage oct

    Figure 2. 6 mm OCT b-scan of the optic nerve head (ONH) region. 

    In contrast, macular OCT scans (Figure 3) focus on the central retina, providing detailed visualization of structures such as the foveal center, retinal layers, and macular biomarkers (such as drusen, hypertransmission, fluids etc). Since the macula is anatomically distinct from the optic nerve head, standard macular scans do not capture the ONH comprehensively. 

    ai oct optic disc analysis

    Figure 3. 6 mm OCT b-scan of the macular region, showing the foveal pit and retinal layers. 

    A more comprehensive scanning approach is 12 mm wide scan OCT (Figure 4), which captures both the macular region and optic nerve head in a single scan. This broader field of view allows for the simultaneous assessment of central retinal structures and optic nerve-related changes, making it valuable for detecting and monitoring conditions that affect both regions, such as glaucoma and other neurodegenerative or vascular retinal diseases. 

    3d wide glaucoma report

    Figure 4. 12 mm wide scan OCT b-scan, which captures both the macular region and part of the optic nerve head.

    2. Results

    2.1. Experiment Setup 

    Brief Method Overview 

    To evaluate the effectiveness of DDLS analysis in assessing glaucoma severity, we designed an experiment comparing results obtained from processing 3D Optic Disc OCT scans and 3D Wide scan OCT scans with the corresponding reports generated by the OCT system. Our method follows four key steps:  

    1. Detecting optic nerve landmarks like Bruch’s Membrane Opening (BMO) points (Eye Keypoints Retrieval / OCT Keypoint Detector Model); 
    2. Segmenting the inner limiting membrane (ILM) (Retina Layers Segmentation Model); 
    3. Reconstructing the neuroretinal rim geometry; 
    4. Applying the Disc Damage Likelihood Scale (DDLS) for classification.  

    The dataset below was used to validate the algorithm. 

    Dataset Used for Validating the Entire Algorithm 

    For validation, we compared our algorithm’s DDLS measurements with the DDLS values generated by the built-in algorithms of the Optopol REVO NX 130 OCT system. This provided a baseline for assessing accuracy and consistency. 

    To validate our approach, we conducted an experiment comparing DDLS metrics derived from: 

    • 3D Optic Disc OCT scans, which are traditionally used for DDLS analysis. 
    • 3D Wide scans, which capture both the macular and optic nerve regions, providing a more comprehensive dataset for analysis. 

    The dataset includes imaging data from 37 patients examined using the Optopol REVO NX 130 OCT system, with each patient undergoing the following protocols on the same day: 

    • 3D Optic Disc OCT (6mm zone): 168 scans 
    • 3D Wide scan (horizontal protocol, 12mm): 128 scans 

    A report was obtained from the 3D Optic Disc OCT scans, containing all parameters calculated by the device. 

    Since no manual annotations are available for these data, our comparison is conducted directly against the device-generated results. 

    The distribution of data was as follows: 

    • Glaucomatous Optic Disc: 21 cases; 
    • Normal Optic Disc: 16 cases. 

    2.2. Final Validation Results: DDLS Accuracy and Error Metrics 

    To evaluate the performance of our DDLS analysis method, we compared its results with the corresponding DDLS values generated by the OCT device’s built-in algorithms. The device reports serve as a reference point for all calculations, meaning the accuracy, MAE/STD values presented below indicate the level of agreement between our method and the device’s measurements. 

    The parameters compared below are the key indicators for glaucoma stage assessment. 

    • The rim-to-disc ratio (RDR) represents the thinnest neuroretinal rim width relative to the vertical optic disc diameter. A lower RDR indicates a more advanced stage of rim thinning, as glaucoma leads to progressive narrowing of the neuroretinal rim due to the loss of ganglion cells axons. 
    • The rim absence angle (RAA) quantifies the extent of neuroretinal rim loss in degrees. It defines the angle where the rim is completely absent, exposing the optic cup. A wider RAA suggests a more severe stage of glaucoma, as it indicates greater rim loss across the disc circumference. 

    Both RDR and RAA provide complementary perspectives on structural optic nerve damage: 

    • RDR measures the smallest remaining rim thickness in proportion to the disc. 
    • RAA evaluates how much of the disc circumference has lost its rim. 

    By considering both parameters together, a more comprehensive assessment of glaucoma severity can be achieved. Based on RDR and RAA, a DDLS stage is assigned, allowing for standardized classification of glaucoma progression. 

    ai oct optic disc analysis

    Table 1. Validation Results of DDLS Analysis on 3D Optic Disc and 3D Wide Scan OCT Scans 

    The table presents validation results comparing 3D Optic Disc OCT scan and 3D Wide scan OCT in DDLS analysis, focusing on Mean Absolute Error (MAE) and Standard Deviation (STD) for key parameters, along with overall DDLS staging accuracy. These metrics are calculated for the rim-to-disc ratio and rim absence angle by comparing their respective values from 3D Optic Disc OCT scans and 3D Wide scans against those from the device reports, providing a precise assessment of deviations from the reference values. 

    Key Observations

    1. Our Goal: Consistency with Device Reports, Not Outperformance

    The experiment does not aim to surpass the device’s accuracy but rather to demonstrate that our method produces results in alignment with the device-generated DDLS reports. 

    The device report serves as a reference, helping to interpret the figures we present, but this does not mean the device’s output is always the absolute truth. 

    2. High DDLS Staging Accuracy for Both Scan Types

    3D Optic Disc OCT scan: 97.3% accuracy in determining DDLS glaucoma stage. 

    3D Wide scan OCT: 94.59% accuracy, demonstrating strong reliability despite a broader scan area and fewer scans capturing the nerve, leading to less available information. 

    Conclusion: 

    • Both types of scans allow the production of clinically reliable DDLS results, but as expected, 3D optic disc scans provide slightly better accuracy due to their higher resolution of the optic nerve head (ONH). 
    • The small accuracy gap and close values for key parameters between the two suggests that 3D wide scan OCT can still be a viable option for glaucoma assessment, despite offering less detailed information about the optic nerve compared to optic disc scans. 

    3. RD Ratio and Rim Absence Angle: High Precision Within Clinical Margins

    RD Ratio (rim-to-disc ratio): 

    • Step size between DDLS stages: 0.1. 
    • Mean Absolute Error (3D Optic Disc OCT scan): 0.008 (significantly smaller than step size). 
    • Mean Absolute Error (3D Wide scan OCT): 0.024 (still relatively small). 

    Conclusion: 

    • Both 3D Optic Disc OCT scan and 3D Wide scan analysis provide high precision in RD ratio calculations. 
    • The small error ensures that stage classification remains reliable, especially in optic disc scans. 

    Rim Absence Angle: 

    • Step size between DDLS stages: Minimum 45°. 
    • Mean Absolute Error (3D Optic Disc OCT scan): 2.2° (very small compared to step size). Mean Absolute Error (3D Wide scan OCT): 4.2° (still well below stage transition threshold). 

    Conclusion: 

    • The method’s margin of error is far smaller than the clinical threshold for stage differentiation, confirming high accuracy in rim loss assessment. 
    • 3D Optic Disc scans again show better precision, reinforcing that they remain the preferred scan type for DDLS.

    4. Our Advantage: Ability to Perform DDLS on Both Scan Types

    • Unlike traditional DDLS implementations, which work only with 3D Optic Disc scans, our method can perform DDLS analysis on both 3D Wide scan and 3D Optic Disc OCTs. 
    • However, 3D Optic Disc OCT remains the preferred method for maximum precision, as it provides a higher-resolution view of the optic nerve. 

    Key Conclusions 

    1. Our method is unique in its ability to process multiple scan types, while still maintaining high accuracy in both cases. 
    2. On 3D Optic Disc scans, we achieve maximum precision, while on 3D Wide scans, we still maintain clinically reliable accuracy. 
    3. Consistency: Across all glaucoma stages, our method produced stable results that closely matched ground truths provided by medical experts. 
    4. Universal Compatibility: The algorithm performed equally well with scans from other manufacturers, demonstrating its versatility and robustness. 

    2.3. Patient Case Studies: DDLS Analysis in Real-World Scenarios 

    Accurate assessment of glaucoma severity relies on precise measurements of optic nerve parameters, such as disc area, rim-to-disc ratio, and rim absence angle. In the following examples, we analyzed four patient cases, including both normal optic discs and glaucomatous eyes, using 3D Optic Disc OCT scan, 3D Wide scan OCT, and device-generated reports as a reference standard. 

    By consolidating individual patient cases into a single comparative table, we can examine the consistency of DDLS analysis across different scan types and highlight key variations that may arise due to differences in scan coverage, segmentation accuracy, and anatomical structure. The following table summarizes the key optic nerve parameters measured for each patient and scan type. 

    AI OCT Optic Disc Analysis

    Table 2. Comparative DDLS Evaluation Across Multiple Patient Cases 

    Key Findings & Interpretation 

    1. High Consistency Between Our Method and Device Reports

    • Across all cases, the DDLS stage remains identical (4 for normal eyes, 7 or 8 for glaucomatous cases) regardless of whether the input scan was 3D Optic Disc OCT or wide scan, and this result corresponds to the device-generated report. 
    • Key optic nerve parameters such as disc area, cup area, and rim area closely align with the device reference, demonstrating strong algorithm performance. 

    2. Minor Variations in Cup and Rim Measurements

    • Cup and rim area values show slight deviations between 3D Optic Disc OCT scans and 3D Wide scan scans, which is expected due to differences in scan coverage and segmentation sensitivity. 
    • For example, in Patient 3 (Glaucoma, Stage 8): 
    • Cup area was 1.86 mm² (3D Optic Disc OCT scan), 1.88 mm² (3D Wide scan), and 1.81 mm² (Device Report). 
    • Rim area was 0.55 mm² (3D Optic Disc OCT scan), 0.53 mm² (3D Wide scan), and 0.58 mm² (Device Report). 
    • These small variations do not affect final DDLS staging but highlight how scan type can introduce subtle segmentation differences.

    3. Rim Absence Angle Varies Slightly but Remains Within Expected Tolerances

    • The rim absence angle shows minor fluctuations across scan types, especially in glaucomatous cases. 
    • Example: In Patient 3 (Stage 8 Glaucoma), the device reported a rim absence angle of 162°, while our algorithm calculated 155° (3D Optic Disc OCT scan) and 151° (3D Wide scan). 
    • Since DDLS categories for severe glaucoma are defined in large increments (e.g., 45°+ thresholds), these small differences do not impact staging accuracy.

    4. 3D Wide scan OCT Provides Comparable Results to 3D Optic Disc OCT scan

    • Despite covering a larger field of view, wide scans produced DDLS staging results consistent with 3D Optic Disc OCT scans and device reports. 
    • In patients with coexisting macular pathologies, 3D Wide scan OCT may provide additional clinical insights while still maintaining high reliability for glaucoma staging. 

    Conclusion: Reliable DDLS Analysis Across Different Scan Types 

    This unified case study analysis confirms that our DDLS analysis algorithm produces highly consistent results across different scan protocols and patient conditions. 

    1. DDLS stage assignment is identical to device reports across all scan types, ensuring high agreement with clinically validated reference values. 
    2. Key optic nerve measurements (disc area, cup area, rim area) are closely aligned across 3D Optic Disc OCT scan, 3D Wide scan, and device reports, reinforcing algorithm accuracy. 
    3. Minor variations in rim absence angle and segmentation metrics do not affect final glaucoma staging, highlighting the algorithm’s robustness. 
    4. 3D Wide scan OCT offers a viable alternative for 3D Optic Disc OCT scans, particularly in cases where both macular and optic nerve regions need simultaneous evaluation. 

    5. Visual Comparison Shows Strong Similarity to Device Reports

    1. The disk and cup boundaries detected by our algorithm closely match those in the device-generated reports, maintaining consistent shapes and anatomical alignment across both 3D Optic Disc and 3D Wide scan OCT scans. 
    2. However, wide scan-based segmentations tend to be slightly rougher, as less structural information is available compared to dedicated optic disc scans. This trade-off is expected due to the broader field of view in wide scans. 

    These findings validate our algorithm’s flexibility, adaptability, and clinical reliability, demonstrating its potential for seamless integration into real-world ophthalmic workflows. 

    2.4. Why Our Approach Stands Out: Key Advantages Over Traditional DDLS Systems 

    While the previous patient case studies demonstrated the accuracy and consistency of our DDLS analysis across different optic disc conditions, another critical advantage of our method is its ability to work seamlessly across various scanning protocols. Unlike traditional device-restricted solutions, our approach supports DDLS assessment on both standard 3D Optic Disc OCT scans and 3D Wide scans with different orientations. 

    The following table illustrates the same patient’s optic nerve head analyzed using three different scanning protocols: 3D Optic Disc OCT scan, 3D Wide scan Horizontal, and 3D Wide scan Vertical. This comparison highlights the method’s adaptability to different scan formats, ensuring reliable DDLS analysis regardless of the scanning protocol used. This example is taken from a Topcon Maestro 2 OCT system, providing an additional reference for processing across different OCT systems. 

    AI OCT Optic Disc Analysis

    Table 3. Comparative DDLS Analysis Across Different Scanning Protocols: 3D Optic Disc OCT, 3D Wide scan Horizontal, and 3D Wide scan Vertical. 

    This capability significantly enhances clinical applicability, allowing our algorithm to process data from various scanning protocols and devices while maintaining high accuracy. The ability to analyze both 3D Optic Disc and 3D Wide scan OCT scans — across different orientations and machine types — ensures comprehensive glaucoma assessment even in cases where scan availability or quality may vary. 

    Key advantages over traditional DDLS analysis methods 

    1. Device Independence

    1. While most existing solutions are restricted to proprietary OCT data formats, our algorithm processes scans from any OCT system, ensuring broad compatibility across devices. 

    2. Consistent Accuracy Across Different Scan Types 

    1. Our algorithm closely matches device-generated DDLS reports, achieving 97.3% accuracy for 3D Optic Disc OCT scans and 94.59% for 3D Wide scan OCTs. 
    2. Patient cases confirm this consistency, with both normal and glaucomatous eyes correctly classified, even when analyzed with different scan types. 

    3. Robust Performance in Edge Cases 

    1. Unlike traditional device-based DDLS assessments, which may struggle with low-quality images or atypical anatomical features, our approach maintains high accuracy in challenging clinical scenarios. 
    2. Patient examples with small optic discs and advanced-stage glaucoma demonstrated that our algorithm successfully identified key DDLS indicators even when scan quality or nerve structure was less distinct. 

    4. Expanded Assessment Through 3D Wide scan OCT 

    1. The ability to perform DDLS analysis on Horizontal and Vertical 3D Wide scans allows for a more comprehensive evaluation by incorporating both macular and optic nerve data. 
    2. In patients with coexisting macular pathologies, wide scans enabled earlier detection of glaucomatous changes that would have been missed if only optic disc scans were used. 

    3. Detailed Approach Description

    To assess glaucoma stage on OCT scans using DDLS analysis, the following steps should be performed: 

    1. Optic Nerve Landmarks Detection – Localization of the optic nerve in the b-scan view of each scan by identifying key anatomical landmarks. 
    2. ILM DetectionSegmentation of the inner limiting membrane (ILM) in the b-scan view of each scan to establish a reference for neuroretinal rim measurement. 
    3. Neuroretinal Rim Reconstruction – Construction of the neuroretinal rim geometry based on detected nerve landmarks and ILM segmentation. 
    4. DDLS Analysis – Application of the Disc Damage Likelihood Scale (DDLS) to assess glaucoma severity based on neuroretinal rim measurements. This includes assigning a DDLS stage according to rim width and optic disc size, with a focus on detecting localized thinning and asymmetry. 

    3.1. Keypoint Annotation Process / Nerve Detection 

    The foundation of our approach lies in a high-quality, annotated dataset meticulously labeled by a team of four expert ophthalmologists. The annotation process focused on identifying key anatomical landmarks in both the macular region and the optic disc nerve zones, both of which are critical for detecting glaucomatous changes and performing Disc Damage Likelihood Scale (DDLS) analysis. 

    These keypoints serve as essential data for evaluating disease progression and training machine learning models. The dataset was carefully selected based on key clinical features, such as the presence or absence of nerve fibers, foveal pits, and other pathological markers, ensuring a comprehensive representation of various conditions and scan types. 

    The annotated dataset consists of approximately 370 unique OCT examinations with more than 56,000 b-scans, covering a range of physical scanning areas, pathology types, and optic nerve conditions to enhance the model’s robustness. The scans are categorized as follows: 

    • Optic Disc with no excavation: ~15 examinations; 
    • Glaucomatous Optic Disc: ~105 examinations; 
    • Normal Optic Disc: ~105 examinations; 
    • Wide scans (covering both the macular and optic nerve regions): ~60 examinations; 
    • Normal Retina Scans: ~40 examinations; 
    • Pathological Retina Scans: ~45 examinations. 

    This detailed annotation process ensures high precision and reliability, enabling the algorithm to generalize across diverse cases while maintaining clinical accuracy in real-world scenarios. 

    3.2. Eye Keypoints Retrieval / OCT Keypoint Detector 

    Our keypoint detection model represents a logical evolution of the model for exam center detection, designed to efficiently and accurately identify key anatomical landmarks in OCT scans. The architecture integrates elements from UNet 5 and CenterNet 6, incorporating YOLO-inspired 7 techniques for keypoint prediction. Additionally, the backbone has been adapted to a transformer-based model 8, enhancing feature extraction capabilities. 

    Training Process 

    The training process follows a multi-stage approach, ensuring robustness, accuracy, and efficiency: 

    1. Stage 1: Detects general keypoints, establishing a foundation for precise landmark localization. 
    2. Stage 2: Groups and refines the identification of specific keypoints, progressively improving the model’s understanding of anatomical structures. 

    This structured approach enhances the model’s reliability across different scan types while maintaining computational efficiency. 

    Key Features 

    Data Preprocessing 

    • The data is augmented using unsupervised techniques, leveraging libraries such as Albumentations 9 to introduce variations such as rotations, scaling, and noise addition. 
    • This ensures the model encounters a wider variety of real-world scenarios during training, improving its generalization capability. 

    Training Process 

    • The model is trained using supervised learning techniques, optimizing a loss function through backpropagation and gradient descent. 
    • This approach allows for continuous refinement and adaptation to complex variations in OCT scans. 

    Parameterization & Tuning 

    • The model includes millions of adjustable parameters (weights), which are fine-tuned to increase accuracy. 
    • Key hyperparameters such as learning rate, batch size, and network depth are carefully selected to maximize performance. 
    • Advanced optimization techniques, including grid search, random search, and Bayesian optimization, are used to find the best hyperparameter configuration. 

    3.3. Retina Layers Segmentation Model 

    The Retina Layers Segmentation Model is our production-stage model, actively used within the Altris AI platform. It was incorporated into this experiment without modifications, ensuring that the results reflect real-world performance as seen in our deployed system. 

    Our Retina Layers Segmentation Model enables precise segmentation of key retinal layers in OCT scans, crucial for detecting structural changes linked to glaucoma and other retinal diseases. The model identifies: 

    • ILM, RNFL, GCL, IPL, INL, OPL, ONL, ELM, MZ, EZ, OS, RPE, BM 

    The training dataset consists of 5,000 expert-annotated OCT b-scans, covering a diverse range of patient demographics, including different ages and ethnic backgrounds. The segmentation model is designed to detect and delineate key retinal layers with high accuracy. 

    Training & Architecture 

    The model is based on U-Net with a ResNet backbone, optimized for OCT images. Training includes: 

    • Expert Annotation: Medical specialists labeled layers for ground truth. 
    • Augmentation: Albumentations-based transformations enhance robustness. 
    • Supervised Learning: Predicts segmentation masks using backpropagation. 
    • Hyperparameter Optimization: Grid search, random search, and Bayesian tuning maximize performance. 

    Model Validation & Performance 

    • The model was validated using a holdout validation approach, with separate validation and test sets that were not exposed during training. 
    • Real-world testing was conducted using scans from various clinical settings to ensure robustness. 
    • Performance was evaluated using the Mean Dice Coefficient across all layers, achieving a score of 0.80, with layer-specific scores ranging from 0.63 to 0.92, confirming high segmentation accuracy. 
    • Cross-domain testing demonstrated consistent performance across different OCT systems, and stability was confirmed over scans collected across different time periods. 

    This efficient, accurate, and generalizable model strengthens DDLS analysis and enhances AI-driven retinal diagnostics. 

    3.4. DDLS Algorithm 

    The DDLS algorithm evaluates glaucomatous changes by analyzing the geometric relationship between the neural rim and optic cup in the optic nerve head. Key steps include: 

    1. Localization: Identifying boundaries of the optic cup and neuroretinal rim by reconstructing geometry on a b-scan view using disc landmarks and an inner limiting membrane.

    3d wide glaucoma report

    Figure 5. B-scan Geometry Visualization. 

    1. Measurement: Calculating the DDLS stage based on the ratio between the rim and disc boundaries.
    2. Cross-Scan Application: Adapting the analysis for 3D Wide scans (both Horizontal and Vertical protocols) as well as 3D Optic Disc-specific scans.

    Our implementation enhances this traditional method by leveraging wide scans, enabling a more comprehensive assessment of glaucomatous changes. 

    3.5. Evaluation 

    To ensure the reliability and effectiveness of our DDLS algorithm, we conducted a rigorous evaluation process, adhering to best practices in data usage, ethics, and performance validation. 

    Data Integrity 

    • Measures were implemented to prevent data leakage, ensuring that scans from the same patient did not appear in both training and testing sets. 

    Ethical Considerations 

    • The analysis strictly relies on OCT-related data (e.g., scan zone size, laterality, pixel spacing) without incorporating any personal patient information. 

    Performance Metrics 

    • Keypoint detection accuracy was evaluated using Mean Squared Error (MSE), comparing model-predicted keypoints with expert annotations. 
    • Additional metrics included correctness of scan center-related landmarks and accuracy in the optic nerve region, ensuring precision in clinical applications. 

    The evaluation results confirmed the algorithm’s robustness, demonstrating significant performance gains, particularly in edge cases, where traditional methods often struggle. 

    Discussion 

    Our DDLS analysis method represents a significant advancement in glaucoma detection. Key benefits include: 

    1. Universal Compatibility: The ability to process data from various devices ensures broad applicability. 
    2. Enhanced Accuracy: By incorporating data from both macular and optic nerve regions, our approach captures more subtle glaucomatous changes. 
    3. Edge Case Performance: Advanced machine learning techniques enable accurate analysis even in challenging scenarios. 

    Compared to traditional methods, our system provides a more flexible, reliable, and comprehensive solution for early glaucoma detection. 

    Disclaimer: USA FDA 510(k) Class II; Altris Image Management System (Altris IMS); AI/ML models and components intended to use for research purposes only, not for clinical diagnosis purposes.

    Conclusion 

    By integrating 3D Wide scans and state-of-the-art machine learning models, we have enhanced DDLS analysis for glaucoma detection, ensuring high accuracy, broad compatibility, and robustness across diverse clinical scenarios. 

    Unlike traditional solutions, our algorithm: 

    1. Works across multiple OCT devices, eliminating the constraints of proprietary data formats. 
    2. It closely matches device-generated DDLS reports, achieving 97.3% accuracy for 3D Optic Disc OCT scans and 94.59% for 3D Wide scans. 
    3. Performs reliably in edge cases, such as small optic discs and advanced-stage glaucoma, where traditional methods may struggle. 
    4. Supports both Horizontal and Vertical 3D Wide scans, enabling more comprehensive assessments that incorporate both macular and optic nerve data. 
    5. Enhances early glaucoma detection, particularly in patients with coexisting macular pathologies, where wide scans provide additional clinical insights. 

    By delivering consistently accurate DDLS staging, regardless of scan type or manufacturer, our system establishes a new benchmark for universal glaucoma assessment. This technology has the potential to significantly improve early detection and management, ultimately preserving vision and enhancing patient outcomes. 

    References 

    1. Spaeth, G. L. (2005). The Disc Damage Likelihood Scale. Glaucoma Today. https://glaucomatoday.com/articles/2005-jan-feb/0105_18.html 
    2. Cheng, K. K. W., & Tatham, A. J. (2021). Spotlight on the Disc-Damage Likelihood Scale (DDLS). Clinical Ophthalmology, 15, 4059–4071. https://pmc.ncbi.nlm.nih.gov/articles/PMC8504474/ 
    3. Zangalli, C., Gupta, S. R., & Spaeth, G. L. (2011). The disc as the basis of treatment for glaucoma. Saudi Journal of Ophthalmology, 25(4), 381-387. https://www.sciencedirect.com/science/article/pii/S1319453411000993 
    4. Review of Optometry Staff. (2023, January 23). Optic disc staging systems effective in grading advanced glaucoma. Review of Optometry. https://www.reviewofoptometry.com/article/optic-disc-staging-systems-effective-in-grading-advanced-glaucoma 
    5. Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. [Preprint]. Posted May 18, 2015. https://arxiv.org/abs/1505.04597 
    6. Duan K, Bai S, Xie L, et al. CenterNet: Keypoint Triplets for Object Detection. [Preprint]. Posted April 17, 2019. https://arxiv.org/abs/1904.08189 
    7. Redmon J, Divvala S, Girshick R, Farhadi A. You Only Look Once: Unified, Real-Time Object Detection. [Preprint]. Posted June 8, 2015. https://arxiv.org/abs/1506.02640 
    8. Dosovitskiy A, Beyer L, Kolesnikov A, et al. An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale. [Preprint]. Posted October 22, 2020. https://arxiv.org/abs/2010.11929 
    9. Buslaev A, Iglovikov V, Khvedchenya E, et al. Albumentations: Fast and Flexible Image Augmentations. [Preprint]. Posted September 18, 2018. https://arxiv.org/abs/1809.06839