Motion Tracking for Automatically Controlled Functional Electrical Stimulation System in Mirror Therapy Configuration: The Enhanced Lazarus Solution

In this work, we present the potential application of an automatic control system for adjusting the parameter values of Functional Electrical Stimulation (FES). FES is a key enabling technology in the functional recovery of patients undergoing neuro-muscular rehabilitation. In this application, successful stimulation outcomes depend on the accurate setting of patient-specific parameters - such as pulse width and stimulation frequency - which are typically determined by physiotherapists based on their clinical experience. The proposed system is implemented in a Mirror Therapy setup, where an injured limb is controlled via FES to replicate the movement of a healthy limb - either belonging to the same patient or to a physiotherapist. The FES activation signal is derived from surface Electromyography (sEMG) recordings of the healthy limb and transmitted to the FES unit after compression using an Average Threshold Crossing (ATC) algorithm. Optimal FES parameters are adjusted through a closed-loop feedback system based on motion tracking of the injured limb, which is continuously compared to the movement of the healthy limb. Additionally, we explore the possibility of replacing the ATC algorithm used for compressing the sEMG signal from the healthy limb with a Compressed Sensing-based approach. This alternative would enable full reconstruction of the sEMG signal-unlike ATC-at the cost of only a minimal increase in computational effort. A prototype was tested on 22 healthy subjects under five different parameter configurations, confirming the feasibility and adaptability of the proposed system.