A Wearable Biofeedback System to Improve Mobility of Above-Knee Amputees

Rafael Nunez 2; Jan Andrysek 2

 1. Institute of Biomaterials and Biomedical Engineering, University of Toronto; 2. Holland Bloorview Kids Rehabilitation Hospital

Lower limb amputees (LLA) typically have reduced sensory-motor control, resulting in gait and balance deficits and diminished mobility function. Biofeedback (BFB) control systems have the potential to improve overall prosthetic function and walking balance by providing real-time biomechanical sensory feedback, using visual, auditory and haptic cues. For LLA, specially at above-knee level, adequate control of the prosthetic knee plays a key role in gait stability; therefore, the aim of the proposed research is to develop and clinically validate a wearable vibrotactile BFB system, which is pretend to be used as a training device for above-knee amputees to potentially alter specific gait parameters, such as prosthetic knee angle, making gait safer by improving prosthetic limb control while walking. To achieve this objective, a wearable vibrotactile BFB system was developed to clinically validate its performance in six (N=6) above-knee amputees. The wearable vibrotactile BFB system was developed to measure the prosthetic knee flexion angle in real-time, as well as to detect two specific gait events (i.e., heel strike (HS) and toe-off (TO)). These gait events were used to define the initiation of the stance and swing phase, respectively, at each gait cycle. Thus, a control algorithm (embedded into a microcontroller) drives a vibrating motor (located in the prosthetic socket) to provide continuous (CNST) and discrete (CORR) feedback to the users, depending on the corrective action that needs to be taken at each gait cycle. For instance, CNST feedback is provided during the swing phase (i.e., right after TO) to encourage LLA to achieve an adequate prosthetic knee flexion while walking. An adequate knee flexion is important to achieve sufficient foot clearance when the opposite leg is brought from behind into the front to execute the next step in order to avoid falls. CORR feedback is provided during the stance phase (i.e., right before HS) to encourage LLA to fully extend their prosthetic limb for weight bearing support, otherwise the prosthetic knee will buckle deriving in a fall. In terms of the clinical testing, four experimental conditions will be evaluated (i.e., Baseline, No feedback, CNST, and CORR feedback) to evaluate the performance of the BFB system on each participant. We expect to show that this augmented training experience using the wearable vibrotactile BFB system could be sufficient to trigger gains in gait performance (i.e. gait retraining and motor learning to promote effective use of the prosthesis and targeting enhanced human movement) in individuals with LLA. The proposed research is expected to provide new insights into innovative interventions, and technologies that can help to drive the efficient utilization of BFB systems in gait rehabilitation, to improve patient outcomes and decrease the burden of limb amputations and ultimately other physical and neurological disabilities that affect nearly 10% of all Canadians.