Optimal configuration of wearable sensors for gait analysis in parkinson's disease patients

The development of wearable inertial sensors has provided the possibility to assess and diagnose Parkinson’s disease (PD) based on the patients’ gait pattern. The goal of this study was to compare the classification performances from fifteen different combinations of sensor configurations using three different machine learning techniques among the group of eighteen subjects. The best configuration was achieved by the sensors placed on the chest, wrist, and shank together. The optimal single sensor configuration was produced by chest sensor.