Heterogeneous digital biomarker integration out-performs patient self-reports in predicting Parkinson’s disease

Abstract

Deng et al. develop deep learning methods that identify Parkinson’s Disease (PD) patients using public accelerometer and position data with higher accuracy than when using gait/rest and voice-based models. Their study demonstrates the complementary predictive power of tapping, gait/rest and voice data and establishes integrative deep learning-based models for identifying PD.