Artificial intelligence extension of the OSCAR-IB criteria.

This review explores how artificial intelligence (AI) can analyse images of the retina, taken using optical coherence tomography (OCT), to help identify or monitor neurological conditions such as Alzheimer’s disease, multiple sclerosis, Parkinson’s disease, stroke and epilepsy.

Because changes in the retina can be very subtle, high-quality images are essential for reliable AI results. The authors recommend extending existing image-quality guidelines to create OSCAR-AI, a framework to help ensure retinal images are suitable for AI analysis. They also highlight the importance of addressing data ownership and involving patients and the public in research.

Petzold A, Albrecht P, Balcer L, Bekkers E, Brandt AU, Calabresi PA, Deborah OG, Graves JS, Green A, Keane PA, Nij Bijvank JA, Sander JW, Paul F, Saidha S, Villoslada P, Wagner SK, Yeh EA; IMSVISUAL, ERN-EYE Consortium. Artificial intelligence extension of the OSCAR-IB criteria. Ann Clin Transl Neurol. 2021 Jul;8(7):1528-1542. doi: 10.1002/acn3.51320. Epub 2021 May 19. PMID: 34008926; PMCID: PMC8283174.