As RetinAI evolves under its new parent company, Ikerian AG, it is expanding its expertise beyond retinal conditions to include the detection and monitoring of other diseases, such as Multiple Sclerosis, Alzheimer’s and Parkinson’s. We embrace the concept that the “eyes are the window to the brain and body” and therefore we are set on finding ways to easily monitor these diseases with available imaging in Ophthalmology, and the application of advanced software technology, like RetinAI’s Discovery platform and our AI models for advanced segmentation of retinal layers. This article will explore the role of retinal biomarkers in neurodegenerative diseases.
12 minutes read

‍

The human body is a vast web of interconnected systems through which chronic systemic diseases often initiate their pathological pathways years, if not decades, before the manifestation of clinical symptoms. This oftentimes stealth-like progression accentuates a clinical need: the development of precise and reliable diagnostic instruments capable of detecting the earliest disease stages.

The golden opportunity lies within this critical time frame, or window of treatment opportunity, offering a chance for interventions that might halt or significantly slow disease progression.
‍

To this aim, recent advances in specialised imaging modalities and Artificial Intelligence (AI) show great promise in fast-tracking the identification and quantification of relevant disease-specific biomarkers.

‍

In this article, we will discuss the transformative role of retinal biomarkers in diagnosing and monitoring chronic, systemic neurodegenerative diseases. We will highlight the impact of Optical Coherence Tomography (OCT) analysis and how AI, with its computational capabilities, paves the way for fast and reliable biomarker detection. 

‍

Why is the retina an ideal target for studying systemic diseases? 
‍

The retina, a small piece of tissue situated at the back of the eye, serves as an exemplary medium for studying systemic diseases that impact both the nervous and cardiovascular systems. Anatomically and developmentally, the retina is an extension of the central nervous system (CNS)¹.

It is directly connected to the brain through the optic nerve, which is formed by the axons of the retinal ganglion cells. This makes it a perfect target to study neurodegenerative disorders that can extend to the retina, even at early stages, such as multiple sclerosis (MS), Alzheimer's disease (AD), and Parkinson's disease (PD).

In addition, the retina contains an extensive network of blood vessels that share similar physiological, embryological, and anatomical features with the cerebral¹ and coronary circulation².

Changes in the retinal vasculature are indicators of cardiovascular conditions (e.g., hypertension) since they parallel the occurrence of systemic processes such as inflammation, oxidative stress, and endothelial dysfunction, among others³.
‍
Finally, the retina shows a major advantage: it is easily accessible through routine eye examination using non-invasive imaging methods (for example: fundus examination and optical coherence tomography).

This makes it a convenient and cost-effective target for disease screening and monitoring, potentially enabling large-scale population studies.

‍

Optical coherence tomography (OCT): a rising star across the medical universe
‍

Optical Coherence Tomography (OCT) is a non-invasive imaging technique commonly used in ophthalmology to visualise the structure of the retina. In recent years, the medical need for capturing even subtle changes in the retina has positioned OCT technology at centre stage.

OCT not only provides high-resolution and detailed images of the retina but also the possibility of quantification of specific biomarkers (e.g., retinal layer thickness and fluid volume). 

For these reasons, OCT has progressively become a valuable tool for diagnosing and monitoring various chronic diseases affecting primarily the posterior segment of the eye (for example, age-related macular degeneration -AMD- or glaucoma).

And this is not only true for Ophthalmology but extends to other medical fields. OCT imaging has been included in routine practice for the screening and monitoring of neurodegenerative and cardiovascular diseases, such as Multiple Sclerosis (MS) and hypertension, respectively, which have been associated with severe ocular complications.

In these cases, OCT provides in-depth insights into the retina status, facilitating the management of these conditions even if they are not limited to the eye.

‍

Retinal OCT: a promising tool for monitoring neurodegeneration progression

What is neurodegeneration? 

Neurodegeneration entails the progressive deterioration of neuronal structure and function, which accompany the dysfunction of the central nervous system⁴ (CNS).

The causes associated with neuronal degeneration remain a popular matter of research as they are poorly understood. It is now known that most of these diseases are multifactorial and result from interactions between environmental and genetic risk factors that accumulate over the years⁴.

Of these, ageing has been consistently identified as the main factor for most neurodegenerative diseases⁵. This means that the incidence of neurodegeneration increases with age and primarily affects elderly people. 

‍

Why is early detection important? 

Because the brain oversees many aspects of how our bodies work, diseases affecting the brain function can disrupt a wide range of abilities over time. These include basic functions like speaking, moving, and staying balanced, as well as more complex tasks like controlling the bladder, thinking clearly, and handling pain. Unfortunately, most neurodegenerative diseases progress irreversibly⁶.

In some cases, available treatments can ease symptoms, relieve pain, or help people regain some functionalities such as balance and mobility.

Therefore, early diagnosis combined with neuroprotective therapy for people at high risk of neurodegeneration allows for postponing its development. This will entail an improvement in the quality of life, as well as saving resources for the healthcare system and society as a whole.

In this scenario, the retina represents an easily accessible window to study the rest of the brain.

‍
Increasing evidence has demonstrated that inflammatory and neurodegenerative changes in the retina mirror brain alterations and, in some cases, even precede clinical symptoms⁷.

For these reasons, retinal biomarkers are progressively gaining more and more popularity for the timely detection of diseases that also trespass the limits of the eye. 

‍

OCT biomarkers for the screening of neurodegenerative disorders 

OCT provides in-depth alterations in the thickness and structure of individual retinal layers and the optic nerve caused by alterations in neuronal and glial cells in both systemic and localised neurodegenerative diseases⁸. The most popular retinal biomarkers include:

• The ganglion cell complex status, including the retinal nerve fibre layer (RNFL), the ganglion cell layer (GCL), and the inner plexiform layer (IPL) 
• Hyper-reflective foci (HRF)
• Disorganisation of the inner retinal layers (DRIL) 
• Thickness and integrity of the outer retinal layers and choroidal thickness

‍

Even though these biomarkers are not disease-specific, they can definitely help physicians screen the patient and monitor disease activity in a non-invasive way. The level of clinical evidence supporting the role of retinal biomarkers in diagnosis and monitoring varies among the different neurodegenerative diseases. This means that for some of them, retinal biomarkers have already been introduced for supporting patient clinical management (e.g., Multiple Sclerosis), whereas for others, they remain in the experimental phase (e.g., Alzheimer’s and Parkinson’s).  

‍

Multiple Sclerosis

Multiple Sclerosis (MS) is an autoimmune inflammatory disease targeting components of the myelin sheath. Optic neuritis is a typical manifestation of MS dissemination, which was demonstrated to be an early sign of the disease in 25% of patients⁹. In recent years, OCT has emerged as a valuable tool for diagnosing and monitoring MS by detecting retinal atrophy as a result of MS-associated optic neuritis^10,11. It typically presents as a reduced thickness of the peripapillary retinal nerve fibre layer (pRNFL) and macular ganglion cell and inner plexiform layers (GCL-IPL) resulting from neuroaxonal damage¹².

Recently, a clinical study¹³ has shown compelling evidence that the assessment of the optic nerve by means of OCT improves the diagnostic accuracy of MS by increasing sensitivity without compromising specificity. Specifically, the study supports the measurement of the RNFL and GCL-IPL thickness in OCT as a reliable biomarker of MS activity. In addition, retinal microglial activation at baseline -evidenced by increased HRF in the inner retinal layers in OCT scans- was associated with subsequent inflammatory events in patients with relapse-onset MS¹⁴. This suggests that HRF might be a good biomarker candidate with prognostic value but needs further investigation. 

The progressive retinal thinning was also associated with cognitive decline, suggesting that quantifying OCT biomarkers such as RFNL thickness may aid in identifying MS patients at risk of cognitive dysfunction¹⁵.

In summary, OCT technology enables the understanding of neuroaxonal damage degree with excellent reproducibility and concordance with Magnetic Resonance Imaging (MRI) detection (gold standard) ^10,16. Moreover, OCT presents considerable advantages over MRI: it is non-invasive, inexpensive, easy and fast to perform, and, more importantly, it allows for reliable quantitative measures of specific biomarkers ¹⁷. 

‍

Alzheimer's Disease 

Alzheimer's Disease (AD) is a highly deleterious neurodegenerative condition that significantly contributes to global morbidity and mortality ¹⁸. AD is the most common form of dementia (60–70% of cases), which is currently the seventh leading cause of death and one of the major causes of disability and dependency among older people globally ¹⁹. In this scenario, early diagnosis and treatment implementation urge as health and long-term care costs are lower when AD is detected at early stages of functional impairment. 

AD was first considered as a solely brain disease, but recently mounting evidence has shown that AD extends also to the retina ^20,21. In fact, AD-specific biomarkers, including amyloid β-protein (Aβ) deposits, hyperphosphorylated Tau proteins (pTau), and plaques and neurofibrillary tangles (NFTs), have been identified in the neuroretina even at early stages ^22,23. 

In addition, non-specific retinal changes and visual impairment have been associated with hallmark biomarkers in these patients. Most significant retinal alterations include neuronal degeneration (evidenced by macular volume loss and the thinning of the RNFL and GCL), optic nerve degeneration, vascular abnormalities, and inflammation^22,23. 

In this context, OCT can easily provide in-depth information about retinal structural changes that accompany AD patients and correlate with brain disease activity ²⁴. Moreover, OCT may also have predictive value since recent studies have demonstrated an association between thinner RNFL at baseline and cognitive decline and dementia development over time ^25–27. 

Using non-invasive, advanced high-resolution imaging techniques such as OCT to identify AD-related retinal biomarkers has not only the potential to tailor individual patient care but also to enable large-scale screening and continuous monitoring of populations at risk. Efforts are now focused on validating and reproducing these retinal biomarkers in longitudinal studies and on finding standardised measurements to be used for disease assessment in routine practice ²⁸. 

‍

Parkinson's Disease 

Parkinson's disease (PD) is a progressive neurodegenerative disorder primarily affecting the CNS, characterised by the depletion of dopaminergic neurons in the midbrain's substantia nigra. In addition to its impact on the brain, PD manifests in the retina, where a subpopulation of amacrine cells releases dopamine. In support, the presence of alpha-synuclein, a key PD hallmark, has been detected in the inner retina ²⁹. 

Advanced imaging techniques like OCT have revealed thinner retinas in PD patients, predominantly affecting inner retinal layers, such as the GCL-IPL and RNFL ³⁰. Very recently, a large cross-sectional study in the UK showed that prevalent PD was also associated with thinner INL, which represents the hub of dopaminergic activity in the neurosensory retina. The authors also found evidence that reduced GCL-IPL and INL thicknesses were associated with an increased chance of developing PD, supporting the prognostic clinical relevance of these retinal layers ³¹. More recently, it was shown that outer retinal alterations such as photoreceptor dysfunction and degeneration paralleled the inner retinal changes at differentiating PD from healthy age-matched adults ³². 

This investigation raises the possibility of using retinal OCT scans for detecting individuals who may be prone to developing PD, facilitating timely intervention and monitoring ³³. Moreover, the advantages of OCT being non-invasive and already integrated into medical practice, offer a unique opportunity of easily implementing this approach in routine clinical workflow. 

‍

How Artificial Intelligence can pave the way for retinal OCT analysis in neurodegenerative disorders? 

Artificial Intelligence (AI) is revolutionising the field of ophthalmology, particularly in the realm of OCT imaging. This advanced technology offers compelling solutions to the inherent limitations of traditional retinal OCT analysis, such as the complexity of interpretation and time-consuming analysis. 

AI algorithms are exceptionally proficient in processing and interpreting complex OCT images. They are capable of identifying subtle shifts in retinal layer thicknesses or fluid volume changes that might be missed after a rapid manual assessment. This heightened precision not only amplifies the accuracy of diagnoses but also offers time efficiency ^34,35.

‍

‍

The applications of AI-based analysis extend across the medical landscape. Within ophthalmology, the automation of retinal image assessments for detecting, screening, and monitoring widespread retinal diseases like diabetic retinopathy, AMD, glaucoma, and others, is already seamlessly integrated into both routine clinical practice and ongoing clinical studies ^36–40.

However, the integration of AI in Ophthalmology isn’t without its challenges. Predominant among these are issues related to the quality and precision of the datasets used to train the algorithms, the generalisability of the algorithm, the occurrence of false negatives, and the broader societal apprehension towards this technology³⁵.

Given that certain structural retinal biomarkers in neurodegeneration are common to those found in other retinal conditions, AI can also efficiently support the analysis of retinal OCT images in the context of neurodegenerative disorders.

Additionally, AI’s inherent capability to swiftly process and automate vast amounts of OCT data streamlines what was once a labour-intensive endeavour.  Such efficiency is pivotal for the prompt detection and monitoring of neurodegenerative diseases on a large scale (e.g., population screening programs).

‍

RetinAI Discovery: a multi-disease platform 

RetinAI’s Discovery is not just a tool, but a comprehensive platform tailored for clinicians across different medical fields.

By capitalising on advanced AI-based algorithms, the platform streamlines the intricate process of patient OCT data analysis, deploying powerful management tools that ensure precision and reliability. 

With an intuitive interface and sophisticated analytics capabilities, Discovery can help you efficiently navigate the complexities of OCT data, providing a robust solution that seamlessly complements human expertise when assessing retinal involvement across a spectrum of ophthalmic and non-ophthalmic diseases.  

‍
During patient consultations,  Discovery empowers clinicians by providing accurate and detailed segmentation and quantification of retinal layers and fluids. This real-time analysis not only fosters a comprehensive understanding of diseases affecting the retina, but also paves the way for timely diagnosis and treatment planning. 
‍

Furthermore, the platform’s capabilities in monitoring retinal morphological changes over time are pivotal in the dynamic landscape of patient care. By offering insights into disease activity and the consequent impact of therapeutic interventions on retinal morphology, Discovery becomes an indispensable ally. This not only propels the customization of treatment regimens, but also reshapes the entire clinical workflow, elevating the overall patient experience through confidence, clarity, and care.
‍

In an era where precision and speed are paramount, RetinAI Discovery stands as a beacon, guiding clinicians towards a future of enhanced retinal care. 

‍

The synergy of OCT & AI: sharper tools, clearer insights

OCT has cemented its reputation in the medical field, particularly for diagnosing and monitoring diseases affecting the retina. The recent inclusion of AI is not just a fleeting trend;  it’s redefining how image analysis is approached in every day practice.

It’s crucial, however, to note that while OCT provides valuable biomarkers for neurodegenerative diseases, a holistic diagnosis often requires a combination of clinical examination, medical history, and additional imaging modalities. By integrating regular OCT monitoring, clinicians are better poised to detect early signs of disease progression, thus tailoring treatment plans to individual patients. 

Empowered by AI models, RetinAI Discovery augments the prowess of OCT image analysis, enhancing both its diagnostic and prognostic capabilities when examining chronic systemic diseases affecting the retina and beyond. 

‍

Thinking of harnessing the combined force of OCT and AI to enhance diagnostic precision in chronic neurodegenerative conditions? Connect with our experts now. 

‍

This article was written by Romina Lasagni Vitar, PhD

‍

‍

‍

‍

‍

References

1. London A, Benhar I, Schwartz M. The retina as a window to the brain—from eye research to CNS disorders. Nat Rev Neurol. 2013;9(1):44-53. doi:10.1038/nrneurol.2012.227

2. Monteiro-Henriques I, Rocha-Sousa A, Barbosa-Breda J. Optical coherence tomography angiography changes in cardiovascular systemic diseases and risk factors: A Review. Acta Ophthalmol. 2022;100(1):e1-e15. doi:10.1111/aos.14851

3. Donati S, Maresca AM, Cattaneo J, et al. Optical coherence tomography angiography and arterial hypertension: A role in identifying subclinical microvascular damage? European Journal of Ophthalmology. 2021;31(1):158-165. doi:10.1177/1120672119880390

4. Chen WW, Zhang X, Huang WJ. Role of neuroinflammation in neurodegenerative diseases (Review). Molecular Medicine Reports. 2016;13(4):3391-3396. doi:10.3892/mmr.2016.4948

5. Hou Y, Dan X, Babbar M, et al. Ageing as a risk factor for neurodegenerative disease. Nat Rev Neurol. 2019;15(10):565-581. doi:10.1038/s41582-019-0244-7

6. Lamptey RNL, Chaulagain B, Trivedi R, Gothwal A, Layek B, Singh J. A Review of the Common Neurodegenerative Disorders: Current Therapeutic Approaches and the Potential Role of Nanotherapeutics. International Journal of Molecular Sciences. 2022;23(3):1851. doi:10.3390/ijms23031851

7. Colligris P, Perez de Lara MJ, Colligris B, Pintor J. Ocular Manifestations of Alzheimer’s and Other Neurodegenerative Diseases: The Prospect of the Eye as a Tool for the Early Diagnosis of Alzheimer’s Disease. Journal of Ophthalmology. 2018;2018(1):8538573. doi:10.1155/2018/8538573

8. Vujosevic S, Parra MM, Hartnett ME, et al. Optical coherence tomography as retinal imaging biomarker of neuroinflammation/neurodegeneration in systemic disorders in adults and children. Eye. 2023;37(2):203-219. doi:10.1038/s41433-022-02056-9

9. Toosy AT, Mason DF, Miller DH. Optic neuritis. The Lancet Neurology. 2014;13(1):83-99. doi:10.1016/S1474-4422(13)70259-X

10. Petzold A, Balcer LJ, Calabresi PA, et al. Retinal layer segmentation in multiple sclerosis: a systematic review and meta-analysis. The Lancet Neurology. 2017;16(10):797-812. doi:10.1016/S1474-4422(17)30278-8

11. Villoslada P, Sanchez-Dalmau B, Galetta S. Optical coherence tomography: A useful tool for identifying subclinical optic neuropathy in diagnosing multiple sclerosis. Neurology. 2020;95(6):239-240. doi:10.1212/WNL.0000000000009840

12. Gabilondo I, Martínez-Lapiscina EH, Fraga-Pumar E, et al. Dynamics of retinal injury after acute optic neuritis. Annals of Neurology. 2015;77(3):517-528. doi:10.1002/ana.24351

13. Bsteh G, Hegen H, Altmann P, et al. Diagnostic Performance of Adding the Optic Nerve Region Assessed by Optical Coherence Tomography to the Diagnostic Criteria for Multiple Sclerosis. Neurology. 2023;101(8):e784-e793. doi:10.1212/WNL.0000000000207507

14. Pengo M, Miante S, Franciotta S, et al. Retinal Hyperreflecting Foci Associate With Cortical Pathology in Multiple Sclerosis. Neurology Neuroimmunology & Neuroinflammation. 2022;9(4):e1180. doi:10.1212/NXI.0000000000001180

15. Alba-Arbalat S, Solana E, Lopez-Soley E, et al. Predictive value of retinal atrophy for cognitive decline across disease duration in multiple sclerosis. J Neurol Neurosurg Psychiatry. 2024;95(5):419-425. doi:10.1136/jnnp-2023-332332

16. Pemp B, Kardon RH, Kircher K, Pernicka E, Schmidt-Erfurth U, Reitner A. Effectiveness of averaging strategies to reduce variance in retinal nerve fibre layer thickness measurements using spectral-domain optical coherence tomography. Graefes Arch Clin Exp Ophthalmol. 2013;251(7):1841-1848. doi:10.1007/s00417-013-2337-0

17. London F, Zéphir H, Drumez E, et al. Optical coherence tomography: a window to the optic nerve in clinically isolated syndrome. Brain. 2019;142(4):903-915. doi:10.1093/brain/awz038

18. 2022 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia. 2022;18(4):700-789. doi:10.1002/alz.12638

19. Dementia. Accessed June 13, 2024. https://www.who.int/news-room/fact-sheets/detail/dementia

20. Mirzaei N, Shi H, Oviatt M, et al. Alzheimer’s Retinopathy: Seeing Disease in the Eyes. Front Neurosci. 2020;14. doi:10.3389/fnins.2020.00921

21. Hart NJ, Koronyo Y, Black KL, Koronyo-Hamaoui M. Ocular indicators of Alzheimer’s: exploring disease in the retina. Acta Neuropathol. 2016;132(6):767-787. doi:10.1007/s00401-016-1613-6

22. Blanks JC, Schmidt SY, Torigoe Y, Porrello KV, Hinton DR, Blanks RHI. Retinal pathology in Alzheimer’s disease. II. Regional neuron loss and glial changes in GCL. Neurobiology of Aging. 1996;17(3):385-395. doi:10.1016/0197-4580(96)00009-7

23. Koronyo Y, Biggs D, Barron E, et al. Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer’s disease. JCI Insight. 2017;2(16):e93621. doi:10.1172/jci.insight.93621

24. van der Heide FCT, Steens ILM, Limmen B, et al. Thinner inner retinal layers are associated with lower cognitive performance, lower brain volume, and altered white matter network structure—The Maastricht Study. Alzheimers Dement. 2023;20(1):316-329. doi:10.1002/alz.13442

25. Ko F, Muthy ZA, Gallacher J, et al. Association of Retinal Nerve Fiber Layer Thinning With Current and Future Cognitive Decline: A Study Using Optical Coherence Tomography. JAMA Neurol. 2018;75(10):1198-1205. doi:10.1001/jamaneurol.2018.1578

26. Sekimitsu S, Shweikh Y, Shareef S, et al. Association of retinal optical coherence tomography metrics and polygenic risk scores with cognitive function and future cognitive decline. Br J Ophthalmol. 2024;108(4):599-606. doi:10.1136/bjo-2022-322762

27. Mutlu U, Colijn JM, Ikram MA, et al. Association of Retinal Neurodegeneration on Optical Coherence Tomography with Dementia: A Population-Based Study. JAMA Neurology. 2018;75(10):1256-1263. doi:10.1001/jamaneurol.2018.1563

28. Foster PJ, Atan D, Khawaja A, et al. Cohort profile: rationale and methods of UK Biobank repeat imaging study eye measures to study dementia. BMJ Open. 2023;13(6):e069258. doi:10.1136/bmjopen-2022-069258

29. Bodis-Wollner I, Kozlowski PB, Glazman S, Miri S. α-synuclein in the inner retina in parkinson disease. Annals of Neurology. 2014;75(6):964-966. doi:10.1002/ana.24182

30. Chrysou A, Jansonius NM, van Laar T. Retinal layers in Parkinson’s disease: A meta-analysis of spectral-domain optical coherence tomography studies. Parkinsonism & Related Disorders. 2019;64:40-49. doi:10.1016/j.parkreldis.2019.04.023

31. Wagner SK, Romero-Bascones D, Cortina-Borja M, et al. Retinal Optical Coherence Tomography Features Associated With Incident and Prevalent Parkinson Disease. Neurology. 2023;101(16):e1581-e1593. doi:10.1212/WNL.0000000000207727

32. Tran KKN, Lee PY, Finkelstein DI, et al. Altered Outer Retinal Structure, Electrophysiology and Visual Perception in Parkinson’s Disease. J Parkinsons Dis. 2024;14(1):167-180. doi:10.3233/JPD-230293

33. Poveda S, Arellano X, Bernal-Pacheco O, Valencia López A. Structural changes in the retina as a potential biomarker in Parkinson’s disease: an approach from optical coherence tomography. Front Neuroimaging. 2024;3:1340754. doi:10.3389/fnimg.2024.1340754

34. Daich Varela M, Sen S, De Guimaraes TAC, et al. Artificial intelligence in retinal disease: clinical application, challenges, and future directions. Graefes Arch Clin Exp Ophthalmol. 2023;261(11):3283-3297. doi:10.1007/s00417-023-06052-x

35. Dahrouj M, Miller JB. Artificial Intelligence (AI) and Retinal Optical Coherence Tomography (OCT). Semin Ophthalmol. 2021;36(4):341-345. doi:10.1080/08820538.2021.1901123

36. Riedl S, Vogl WD, Mai J, et al. The Effect of Pegcetacoplan Treatment on Photoreceptor Maintenance in Geographic Atrophy Monitored by Artificial Intelligence–Based OCT Analysis. Ophthalmology Retina. 2022;6(11):1009-1018. doi:10.1016/j.oret.2022.05.030

37. Holz FG, Abreu-Gonzalez R, Bandello F, et al. Does real-time artificial intelligence-based visual pathology enhancement of three-dimensional optical coherence tomography scans optimise treatment decision in patients with nAMD? Rationale and design of the RAZORBILL study. Br J Ophthalmol. Published online August 6, 2021:bjophthalmol-2021-319211. doi:10.1136/bjophthalmol-2021-319211

38. Gallardo M, Munk MR, Kurmann T, et al. Machine Learning Can Predict Anti-VEGF Treatment Demand in a Treat-and-Extend Regimen for Patients with Neovascular AMD, DME, and RVO Associated Macular Edema. Ophthalmol Retina. 2021;5(7):604-624. doi:10.1016/j.oret.2021.05.002

39. Martin-Pinardel R, Izquierdo-Serra J, De Zanet S, et al. Artificial intelligence-based fluid quantification and associated visual outcomes in a real-world, multicentre neovascular age-related macular degeneration national database. Br J Ophthalmol. Published online January 10, 2023:bjo-2022-322297. doi:10.1136/bjo-2022-322297

40. Grzybowski A, Brona P, Lim G, et al. Artificial intelligence for diabetic retinopathy screening: a review. Eye. 2020;34(3):451-460. doi:10.1038/s41433-019-0566-0

‍