Robotic techniques for upper limb evaluation and rehabilitation of stroke patients

• Published in: IEEE transactions on neural systems and rehabilitation engineering 2005-09, Vol.13 (3), p.311-324
• Main Authors: Colombo, R., Pisano, F., Micera, S., Mazzone, A., Delconte, C., Carrozza, M.C., Dario, P., Minuco, G.
• Format: Article
• Language: English
• Subjects: Admittance; Aged; Assessment scales; biomechatronics; Chronic Disease; computerized analysis; Condition monitoring; Diagnosis, Computer-Assisted - methods; Female; Humans; Male; Manipulators; Measurement standards; Medical treatment; Middle Aged; Motion Therapy, Continuous Passive - methods; Motor Skills; neurorehabilitation; Paresis - diagnosis; Paresis - etiology; Paresis - rehabilitation; Patient monitoring; Physical Examination - methods; Physical Stimulation - methods; Rehabilitation robotics; Robotics - methods; Robots; stroke; Stroke - complications; Stroke - diagnosis; Stroke Rehabilitation; Therapy, Computer-Assisted - methods; Upper Extremity - physiopathology; Velocity measurement; Wrist
• ISSN: 1534-4320; 1558-0210
• DOI: 10.1109/TNSRE.2005.848352

This paper presents two robot devices for use in the rehabilitation of upper limb movements and reports the quantitative parameters obtained to characterize the rate of improvement, thus allowing a precise monitoring of patient's recovery. A one degree of freedom (DoF) wrist manipulator and a two-DoF elbow-shoulder manipulator were designed using an admittance control strategy; if the patient could not move the handle, the devices completed the motor task. Two groups of chronic post-stroke patients (G1 n=7, and G2 n=9) were enrolled in a three week rehabilitation program including standard physical therapy (45 min daily) plus treatment by means of robot devices, respectively, for wrist and elbow-shoulder movements (40 min, twice daily). Both groups were evaluated by means of standard clinical assessment scales and a new robot measured evaluation metrics that included an active movement index quantifying the patient's ability to execute the assigned motor task without robot assistance, the mean velocity, and a movement accuracy index measuring the distance of the executed path from the theoretic one. After treatment, both groups improved their motor deficit and disability. In G1, there was a significant change in the clinical scale values (p