Presenting a New Muscle Synergy Analysis based Mechanism to Design a Trackable Visual Biofeedback Signal: Applicable to Arm Movement Recovery after Ischemic Stroke

This study aimed to evaluate the effect of a novel underlining mechanism of visual biofeedback based on muscle synergy pattern on upper extremity motor functions for subacute stroke patients. The experimental studies were conducted on 12 participants in the control group and 24 subjects with ischemic stroke. In the first step, a visual biofeedback trajectory designed for rehabilitation was produced using the patterns extracted from the muscle synergy of arm movement using the hierarchical alternating least squares (HALS) method and with the help of nonlinear autoregressive with exogenous inputs (NARX) from the family of recurrent neural networks and was evaluated on healthy participants. In the second step, all patients received conventional therapy for the upper extremity, 2 times per week for 5 weeks. The interventional group additionally received training with the proposed visual biofeedback system for 30 minutes per session. The evaluations were performed regarding modeling performance, trackability, and clinical efficacy. In terms of modeling performance, the results showed that the NARX method has the best performance compared to other conventional models. Regarding trackability, the analyses based on computing the correlation coefficient showed significant improvement in the trackability from baseline to post-treatment in both the interventional and the control groups. In terms of clinical efficacy and based on analyzing the NIHSS, Fugl-Meyer, and MRS scores, the findings showed that the proposed visual biofeedback mechanism along with the conventional therapy may have supplemental benefits for stroke survivors.