Mount Sinai develops AI for diagnosing sleep disorders

Researchers at Mount Sinai have developed an AI algorithm to diagnose REM sleep behavior disorder (RBD) more accurately. RBD is often an early indicator of Parkinson's or dementia, but diagnosing it has been challenging. The new algorithm analyzes video recordings from sleep tests to detect movement patterns during REM sleep. It achieved a detection accuracy of 91.9%, significantly improving upon traditional methods that often miss cases of RBD. This advancement allows for better diagnosis in clinical settings and could lead to earlier interventions for patients at risk of developing Parkinson's or dementia. The approach uses standard 2-D cameras, making it feasible for widespread implementation.