Russian Researchers Train AI to Diagnose Brain Disorders Through the Eyes
Scientists in Russia have developed an AI algorithm that can detect neurological disorders from retinal scans, offering a faster, cheaper, and more accessible path to early diagnosis.
Researchers at Ural Federal University (UrFU) have created an innovative method that uses artificial intelligence to diagnose neurological conditions — including autism, ADHD, and Parkinson’s disease — from simple, low-cost retinal imaging.
The technology promises a breakthrough in global healthcare by speeding up diagnosis and easing the workload on physicians. Its introduction into Russian clinics could make AI-driven diagnostics a routine part of medical practice.
Future plans include expanding the system to detect glaucoma and other retinal diseases.
According to the university’s press office, the foundation of the method is the electroretinogram — a test that records the retina’s electrical response to flashes of light. UrFU’s Artificial Intelligence Center trained the algorithm to recognize subtle patterns in these signals that point to serious brain disorders.
Unlike many opaque AI systems, this Russian technology is built on transparent and explainable models. Instead of a simple yes-or-no result, physicians can see exactly which parts of the signal triggered the algorithm’s concern. That visibility helps doctors understand the AI’s logic and make informed decisions about further patient care.
The breakthrough comes at a time when neurological disorders are increasing worldwide, affecting billions of people. Early and precise diagnosis is key to timely treatment and improved quality of life.
