Bio-mimetic Control of Functional Electrical Stimulation

Muscle paralysis is among the leading causes that impair an individual’s self-sufficiency, compromising her/his social integration and ability to carry out Activities of Daily Living (ADL). One of the most widely used techniques in physical therapy is Functional Electrical Stimulation (FES), which, through the non-invasive application of low-energy pulses, allows muscle contractions to be triggered even if the brain control is inhibited, thus preventing muscle atrophy and promoting patient recovery.

In recent years, our research group has been active in controlling FES by using the bio-inspired Average Threshold Crossing (ATC) technique, which has proven to be valid for the real-time modulation of bio-mimetic FES patterns. The typical application scenario involves two subjects: a therapist and a patient. Muscle activity is recorded from the therapist by means of our acquisition devices, and the ATC data of each device are processed to generate the FES pattern to be applied to the patient, in order to make her/him simultaneously replicate the movement performed by the therapist. The generated FES pattern is multichannel, to account for muscle synergies, and bio-mimetic, since it is derived from the muscle activity of the therapist, instead of being a binary on/off mechanism or limited to a few levels of intensity. This therapist-patient configuration is particularly effective thanks to the Action Observation Treatment (AOT) approach, which promotes neural plasticity by actively involving the patient in the rehabilitation session. A variant telerehabilitation-oriented approach consists of the use of pre-recorded patterns to be applied to the patient while s/he is watching a video of the therapist performing the exercise, and monitoring the patient’s movements using Inertial Measure Units (IMU) to further adjust the stimulation parameters in real-time, thus allowing for the execution of functional movements without direct supervision of a therapist.