Wearable Sensor-driven Force Myography for Targeted Muscle Identification with Deep Learning

Quantifying muscle engagement during an activity is critical for gauging the effectiveness of rehabilitation and exercise regimens. This study introduces a new method that employs force myography (FMG) using wearable sensors to assess the effectiveness of physical activity by quantifying the involvement of a particular muscle or muscle group during an activity in real time. FMG-based muscle activity data was used to train a deep learning (DL) algorithm for this purpose. The preliminary findings for the upper and lower extremities reported a muscle activity level identification accuracy of over 97% for upper arm muscles and approximately 90% for thigh muscles.