Toward A Quantitative Evaluation of the Fall Risk Using the Fusion of Inertial Signals and Electromyography with Wearable Sensors

Freezing of Gait (FOG) is an unpredictable gait disorder typical of Parkinson's Disease (PD). The main goals of this work are detecting FOG episodes, classifying FOG subtypes and analyzing the leg muscles activity toward a deeper insight into the disorder pathophysiology and in the associated risk of fall. Fusion of inertial and electromyographic signals in our wearable system allows distinguishing correctly 98.4% of FOG episodes and monitoring in free-living conditions the activity type and intensity of leg antagonist muscles involved in FOG. This is an advancement in the state-of-art knowledge of PD pathophysiology, possibly allowing the implementation of current therapeutic strategies.