Body Worn EMG Controlling of Assistive Robotic Arm for Stroke Rehabilitation

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Stroke is the leading cause of adult disability and the third leading cause of death in the United States. Approximately 15 million people in the world, and more than 700,000 people in the United States, experience a stroke each year. Following a hemispheric stroke, motor control of extremities on one side of the body is usually affected. Many patients suffer a variety of disabling physical symptoms on the contralesional side of the body. In particular, upper limb (arm, hand, finger/thumb) dexterity is often affected, limiting fundamental activities of daily living (ADL) such as eating, dressing, writing or typing. A number of mechatronic devices have been designed as assistive tools for robot-aided stroke rehabilitation to improve upper limb function. Most of the devices can assist users to perform exercises which involve repetitive movement of their paretic limb in a passive way (as they relax), or in an active way (as they intend to contribute to the movement). Surface electromyogram (EMG) signals contain rich motor control information, from which the user's intention can be detected. Due to the upper-limb dexterity, however, most functional tasks are generally accomplished through complex temporal and spatial coordination of multiple muscles. It is unfeasible to realize the control of such multiple DOFs via one-to-one mapping (between a muscle and a DOF). Pattern recognition techniques have recently attracted increasing attention in the development of myoelectric control systems. Recently, we have presented a novel framework for stroke survivors using high density surface EMG recording and pattern recognition analysis. Our research demonstrates that high accuracies can be obtained in classification of up to 20 arm, hand, finger/thumb movements involving the affected limb, suggesting that with myoelectric pattern recognition techniques substantial motor control information can be extracted from the paretic muscles of stroke subjects. Such information will potentially enable volitional control of assistive devices, thereby facilitating the functional restoration for the affected limb. In phase I, we will demonstrate the feasibility of a high densit EMG system for robot control to allow stroke subjects volitional control of assistive tools. In phase II, the myoelectric control system will be fully integrated with the assistive robot for implementing improved stroke rehabilitation and tested on a patient population.