Robotic Assistance for Upper Limbs May Induce Slight Changes in Motor Modules Compared With Free Movements in Stroke Survivors: A Cluster-Based Muscle Synergy Analysis

Abstract

Background: Robot-assisted rehabilitation is a technique for achieving motor recovery, whose efficacy is still matter of debate. Usually, the effects of robot assistance are evaluated with standard clinical assessments; few studies investigated the value of human-centered instrumental analysis for assessing the process of interaction of post-stroke patients with robotic set-ups. In this paper, the muscle synergies analysis was used to describe the effect of human-robot interaction on the spatial and temporal features and directional tuning of motor modules during robot-assisted movements. Methods: Twenty-two post-stroke patients performed a session composed of a sequence of hand-to-mouth movements, performed freely and with robot assistance. Patients were evaluated through an instrumental assessment, by recording kinematics and EMG to extract spatial muscle synergies and their temporal components. Patients’ motor synergies were grouped by the means of cluster analysis, pairwise matched across conditions, and compared in terms of spatial and temporal features, and directional tuning. Results: Motor synergies were successfully extracted on all 22 patients in both conditions. Seven clusters were identified, both in free and robot assisted movements. Interaction with the robot slightly altered, at different extent, synergies spatial features, as well as temporal components and synergies tuning. Conclusions: Slight alterations of the modules composition, temporal components and directional tuning were observed when considering post-stroke free movements in comparison to robot assisted. Such effects are worth being investigated in the framework of a modular description of the neuro-muscle-skeletal system to increase insights on human-robot interaction and effects induced with robotic rehabilitation.