AI Detects Early Parkinson’s Through Breathing Patterns

A 2024 study from the MIT Jameel Clinic revealed a neural network capable of detecting Parkinson’s disease with 95% accuracy by analyzing breathing patterns during sleep.
The system, trained on over 7,500 hours of polysomnography data, detects early neural anomalies invisible to clinical symptoms.
According to Nature Medicine, the model could enable non-invasive home diagnostics, reducing misdiagnosis by up to 60%.
Ethical oversight for medical validation is ongoing at Massachusetts General Hospital’s IRB...