0715 Machine Learning Model to Predict Isolated REM Sleep Behavior Disorder Phenoconversion Time and Subtype using EEG

Abstract Introduction More than 80% of patients isolated rapid eye movement (REM) sleep behavior disorder (iRBD), a prodromal disease α-synucleinopathies, progress to neurological like Parkinson's (PD), dementia with Lewy bodies (DLB), and multiple system atrophy (MSA). Resting-state EEG measurements taken at baseline have been related the phenoconversion. The timing conversion which it will convert are crucial issues in iRBD. This work used iRBD create prediction model for phenoconversion time subtype α-synucleinopathy. Methods assessments were performed on spectral power, Shannon entropy weighted phase lag index employed as features. Four models predict subtypes PD-MSA DLB groups, three survival. External validation was also performed. Results 29 out 143 who followed up nine years (mean 3.4 years) later developed α-synucleinopathies (14 PD, 9 DLB, 6 MSA). With concordance 0.8130 an integrated Brier score 0.0921, random survival forest best predicting For analysis, highest accuracy, extreme gradient boosting, had accuracy 86.52%. Both indicated high importance slowing Conclusion It is possible using machine learning biomarkers. To confirm our findings, further study required, including large sample data from various countries. Support (if any)