In St. Louis, researchers have developed an AI system that could transform how eye doctors diagnose retinal disease—not by replacing their expertise, but by helping them cut through the deluge of data that modern eye scans produce. The technology, called OCTCube-M, is the latest advance in a field where artificial intelligence is proving it can spot patterns faster and more reliably than human review of hundreds of images per scan.
The need for this kind of tool is urgent. At least 2.2 billion people worldwide have vision impairment, according to the World Health Organization, and early diagnosis remains critical. Modern eye imaging—specifically optical coherence tomography—generates extraordinarily detailed three-dimensional pictures of the retina and optic nerve in a single, swift scan. That precision is the problem: each scan produces hundreds of cross-sectional images that physicians must review manually, a time-consuming process vulnerable to human error and missed subtle signs of disease.
Aaron Lee, MD, the head of the Department of Ophthalmology & Visual Sciences at Washington University School of Medicine, and his colleagues at the University of Washington in Seattle and Genentech, Inc. developed OCTCube-M as a family of three AI models designed specifically to read and interpret these 3D eye images. In their study published in Nature Biomedical Engineering, the researchers demonstrated that the system more accurately identified eight different retinal diseases compared with older AI models, including age-related macular degeneration—the leading cause of blindness in people over 50—and geographic atrophy, a severe form that threatens vision loss.
What sets this work apart is the breadth of what the AI can detect. The retina's tiny blood vessels are anatomically identical to those in the kidney and developmentally similar to the vessels in the heart and brain. The researchers found that OCTCube-M could infer health risks well beyond the eye, predicting outcomes such as heart attack, stroke, and kidney failure based solely on retinal imaging. Those same disease processes that leave signatures in the eye—like plaque buildup inside blood vessel walls—also mark vessels throughout the body.
Dr. Lee emphasized the practical impact: "Our AI system has the potential to empower physicians to make faster diagnoses, tailor treatment more precisely and design clinical trials that bring new therapies to patients faster." The technology doesn't require physicians to change their practice; it simply helps them process the volume of images their own instruments generate, revealing patterns that might otherwise be missed during manual review.
The implications ripple outward. For patients at risk of geographic atrophy or other sight-threatening conditions, faster, more accurate diagnosis means earlier intervention. For those with undetected heart, kidney, or neurological disease, a routine eye exam could become a screening tool that catches serious illness before it advances. And for researchers designing clinical trials, the ability to predict disease progression more precisely could accelerate the path to new treatments.
This is AI working at its most practical: not replacing the expertise of trained doctors, but enhancing their ability to help patients see clearly and stay healthy.