Could AI help clinicians to differentiate neurodegenerative disease from retinal scans?

University of Waterloo researchers are developing machine learning technology that could revolutionize early detection of brain diseases through routine eye examinations. The AI system analyzes how polarized light interacts with different types of protein deposits in the retina – amyloid beta deposits linked to Alzheimer’s disease and TDP-43 deposits associated with frontotemporal dementia and ALS. This breakthrough could enable optometrists and eye care professionals to identify these neurodegenerative conditions before symptoms appear, when diseases are typically already advanced. The researchers hope to develop a simple, accessible eye test that could detect and distinguish multiple brain diseases, particularly benefiting patients in underserved communities who may lack access to specialized neurological testing facilities. https://www.aop.org.uk/ot/news/2026/03/12/could-ai-help-clinicians-to-differentiate-neurodegenerative-disease-from-retinal-scans