13 Biofeedback for Rehabilitation: Development of Muscle Activation Feedback App to Retrain Muscles After Stroke

Faculty Mentor: Jacob George (Electrical and Computer Engineering, University of Utah)

The long-term goal of this research is to rehabilitate healthy muscle activation patterns in stroke patients through a portable muscle biofeedback system. Approximately 795,000 people experience a stroke each year, and 80% of survivors struggle with upper-limb hemiparesis. Hemiparesis is associated with delayed muscle activation and delayed muscle relaxation that makes it difficult to perform activities of daily living. To address this challenge, we developed a software application to provide real-time biofeedback of muscle activity that patients can use to practice faster muscle activations and relaxations at home. The app automatically detects and logs muscle activations by thresholding electromyography (EMG) data sampled at 1 kHz from the forearm. Time constants are used to quantify the speed of muscle activation and relaxation and are displayed in real-time with dynamic color changes indicating improvement relative to a running average over the past three activations. Preliminary data from one stroke patient hinted at delayed muscle relaxation during hand extension on the paretic side relative to the non-paretic side (0.411 s vs 0.223 s; p = 0.074, Wilcoxon signed rank test). The time constant of muscle relaxation during hand extension trended towards improving after 10 minutes of biofeedback (from 0.411 s to 0.154 s; p = 0.097, Wilcoxon signed rank test). The participant also expressed interest in using the app for at-home rehabilitation. At-home use of this app may further improve muscle activation through real-time biofeedback, muscle strengthening, and high-repetition training. This accessible rehabilitation platform could also be used to rehabilitate muscle function after other neurological conditions (e.g., spinal cord injury or traumatic brain injury).