REACHING AND HAND-TO-MOUTH MOTION PRIMITIVES FOR THE FUNCTIONAL ASSESSMENT OF MOTOR ABILITIES IN ROBOT-ASSISTED REHABILITATION PRACTICE

INTRODUCTION Multi joint (shoulder and elbow) control is a prerequisite for the use of the hand in ADLs. The use of robotic devices to reduce Upper Limb (UL) impairment at proximal joints is a promising rehabilitation technique. RehaMSR is a robotic platform developed for the assessment and rehabilitation of functional movements. Two gestures were selected, the Reaching Movement (RM) against gravity and the Hand-to-Mouth Movement (HtMM), as they are representative of ADLs such as eating and reaching for objects. First studies to determine the effects of velocity profiles and paths on the EMG pattern activity, shoulder torque and movement perception were carried out on healthy subjects executing assisted RMs. Experiments on neurological patients to test the applicability of the robotic treatment and to verify the results in the short term are ongoing. EXPERIMENTAL SETUP AND GESTURES Subjects - 6 Healthy subjects (23 - 35 years, 5 males) 8 Neurological Patients (Stroke, TBI, PCI and SCI; 16 - 75 years; 6 males) Experimental setup -Serial robot equipped with a force sensor - Marker-based tracking system - Wireless EMG Robot control highlights - Spline-based paths and motion laws for complex functional ADL movements - Haptic feedbacks with VR integration - Easy networking and interface RESULTS - Free movements at physiological velocities minimize the exerted shoulder torque thanks to the UL inertia optimum exploitation. - The VAS score is higher for bell-shaped velocity profiles. - Muscular cocontractions decrease with subjects' participation and natural bell-shaped velocity profiled trajectories. - Patients' movements are in general slow figuring higher shoulder torques because of the absence of UL inertia. - Muscular activation is induced even during passive mobilization. -The EMG activation pattern in patients is strongly dependent by the mobilization velocity, the type/level of the patients interaction and by the robot control modality CONCLUSIONS The robotic treatment was applicable to all patients, independently of the pathology and the residual abilities (from low to moderate). A high treatment program customization was indeed needed. Results, from both healthy and pathological subjects, suggest that the trajectories (path and velocities) and the robot control modes should be accurately tuned during customization of the robotic treatment. Evaluation of the effect in the short term of the robotic therapy are still ongoing. REFERENCES [1] Caimmi 2012. 4th IEEE RAS and EMBS BioRob Intl. Conf. [2] Caimmi 2011. 20th annual Meeting of the ESMAC.