Abstract
Robotics and machine learning algorithms can potentially enhance upper limb rehabilitation, addressing the limitations of traditional therapy methods. This study presents a Human-Robot Interaction (HRI) platform designed to improve the occupational therapy environment in clinical settings. Eight participants engaged in grasping and reaching tasks with the assistance of an upper limb robotic system, guided by visual feedback. The tasks were structured across three levels of difficulty. Machine learning models were employed to classify task difficulty based on muscle activity patterns in the moving arm. Participant performance was evaluated using average trajectory deviation and non-dimensional squared jerk, providing insight into movement accuracy and smoothness, which corresponded with the predefined task difficulty levels. This platform represents a meaningful step toward delivering more personalized and effective rehabilitation for individuals recovering from stroke-related upper limb impairments. Future advancements in rehabilitation technology must prioritize patient-centered design, accessibility, and data-driven innovation to ensure impactful and inclusive care.
Get full access to this article
View all access options for this article.