Russian Scientists Develop Neural Network for Early Parkinson’s Diagnosis

Researchers at Sechenov University in Moscow have created an AI system capable of detecting Parkinson’s disease at its earliest stages with up to 97% accuracy—long before clinical symptoms appear.

Specialists at Sechenov University have developed an artificial intelligence tool that can identify Parkinson’s disease at its preclinical stage—before symptoms are visible. The system demonstrated remarkable accuracy of up to 97% when analyzing electroencephalogram (EEG) data, according to Russian media reports.

The neural network was trained on medical research data, with neurologists labeling EEG records of both healthy individuals and patients with confirmed diagnoses. Testing showed the algorithm’s effectiveness in detecting brain pathologies at the earliest possible stage.

While Parkinson’s remains incurable, early detection can slow its progression and help preserve patients’ quality of life. Global forecasts warn that by 2050, the number of people with Parkinson’s could surpass 25 million, making early screening methods critically important.

Researchers plan to expand the dataset and develop a user-friendly digital platform for physicians. This will simplify preliminary diagnosis and speed up treatment decisions.

As the project’s authors emphasize, the findings demonstrate that AI can detect changes in brain activity long before symptoms appear—a breakthrough in preventive neurology.