Periodic Leg Movements during Sleep Associated with REM Sleep Behavior Disorder: A Machine Learning Study

Area of Science:

• Neurology
• Sleep Medicine
• Artificial Intelligence

Background:

• Idiopathic REM sleep behavior disorder (iRBD) patients often exhibit periodic leg movements (PLMS), complicating clinical diagnosis.
• Differentiating iRBD phenotypes with and without PLMS is challenging without polysomnography.

Purpose of the Study:

• To develop a novel Machine Learning (ML) approach for distinguishing between iRBD phenotypes based on heart rate variability (HRV).

Main Methods:

• Collected HRV data from 42 iRBD patients (23 with PLMS, 19 without PLMS).
• Trained ML models (Logistic Regression, Support Vector Machine, Random Forest, eXtreme Gradient Boosting) on HRV data.
• Assessed classification performance using Leave-One-Out cross-validation.

Main Results:

• The Random Forest model achieved the highest accuracy (86%), sensitivity (96%), and specificity (74%).
• Support Vector Machine and eXtreme Gradient Boosting models also demonstrated good performance.
• Logistic Regression showed lower classification accuracy (71%).

Conclusions:

• ML algorithms can effectively differentiate iRBD phenotypes based on HRV.
• Artificial intelligence shows promise in supporting the diagnosis of clinically indistinguishable iRBD subtypes.