Effect of Robotic Exoskeleton Motion Constraints on Upper Limb Muscle Synergies: A Case Study

• Published in: IEEE transactions on neural systems and rehabilitation engineering 2021, Vol.29, p.2086-2095
• Main Authors: Mcdonald, Craig G., Fregly, Benjamin J., O'Malley, Marcia K.
• Format: Article
• Language: English
• Subjects: Case studies; Composition; Constraints; Degrees of freedom; Elbow; Electromyography; Electromyography (EMG); Exoskeleton; Exoskeletons; Factorization; Muscles; Rehabilitation; rehabilitation robotics; Rehabilitation robots; Robotics; Robots; Training; Variability; Wrist
• ISSN: 1534-4320; 1558-0210
• DOI: 10.1109/TNSRE.2021.3118591

Evidence exists that changes in composition, timing, and number of muscle synergies can be correlated to functional changes resulting from neurological injury. These changes can also serve as an indicator of level of motor impairment. As such, synergy analysis can be used as an assessment tool for robotic rehabilitation. However, it is unclear whether using a rehabilitation robot to isolate limb movements during training affects the subject's muscle synergies, which would affect synergy-based assessments. In this case study, electromyographic (EMG) data were collected to analyze muscle synergies generated during single degree-of-freedom (DoF) elbow and wrist movements performed by a single healthy subject in a four DoF robotic exoskeleton. For each trial, the subject was instructed to move a single DoF from a neutral position out to a target and back while the remaining DoFs were held in a neutral position by either the robot (constrained) or the subject (unconstrained). Four factorization methods were used to calculate muscle synergies for both types of trials: concatenation, averaging, single trials, and bootstrapping. The number of synergies was chosen to achieve 90% global variability accounted for. Our preliminary results indicate that muscle synergy composition and timing were highly similar for constrained and unconstrained trials, though some differences between the four factorization methods existed. These differences could be explained by higher trial-to-trial EMG variability for the unconstrained trials. These results suggest that using a robotic exoskeleton to constrain limb movements during robotic training may not alter a subject's muscle synergies, at least for healthy subjects.