EMG Biofeedback with AMES

Period of Performance: 09/01/2007 - 08/31/2008


Phase 1 SBIR

Recipient Firm

Ames Technology, Inc.
Principal Investigator


DESCRIPTION (provided by applicant): The long-range goal of this project is to be able to treat effectively plegic stroke victims using a novel therapeutic regimen and robotic device called "AMES," an acronym for Assisted Movement with Enhanced Sensation. In AMES, the patient assists the motion of the robotic device using biofeedback of voluntary joint torque, while the sensation of motion is enhanced by tendon vibration. In a pre-clinical trial, we showed AMES to be effective at restoring functional movement in both upper and lower extremities in a majority of profoundly disabled chronic stroke patients (>2 yr event). At enrollment, these subjects ranked <30th percentile of normal limb strength. Despite our success at treating profoundly disabled stroke victims, AMES treatment did not restore functional movement at joints (e.g., fingers, wrist, and ankle) rendered completely plegic by the stroke. Using EMG recordings, however, we found that most plegic stroke patients retain the ability to activate voluntarily the 'plegic' muscles. Because the activity is so weak and it competes with antagonistic spasticity and co-contraction, these individuals are unable to move the joint in one or the other direction. The goal of the proposed project is to incorporate EMG biofeedback into the AMES treatment, as an alternative for torque biofeedback, in order to treat effectively profoundly plegic stroke patients. There are 2 specific aims proposed in this Phase I application. The first specific aim is to convert EMG signals into useful biofeedback on a graphics display. Two tasks must be accomplished: (1) to develop and test software to low-pass filter the EMG signals so the bandwidth (~0-2 Hz) is matched to that of the patient's efforts and (2) to present on a graphics display the activity from 2 agonist and 2 antagonist muscles in a way that it is intuitive to the patients how to correct their dyssynergia. The second specific aim is necessitated by the proposed EMG acquisition in the presence of tendon vibration-mechanical and electrical interference produced by the vibrators must be reduced in the EMG recording to levels where it does not reduce the information content of the biofeedback. Specific Aim 2 is to reduce this interference down to =2% of the overall signal amplitude. We propose to explore a number of alternative solutions to the mechanical artifact and electrical artifact problems and to implement solutions that are effective while minimizing cost. Once we have implemented useful EMG biofeedback with the AMES device, Phase II of this project will test the methodology in a controlled clinical trial. This project addresses a sub-group of chronic stroke patients with complete plegia at one or more joints, but who retain some ability to activate voluntarily the 'plegic' muscles. Our preliminary studies indicate that a sizable proportion of the 5 million chronic stroke patients in the US are plegic at one or more joints while retaining some voluntary activation of the 'plegic' muscles. The proposed technology is designed to permit the reversal of dyssynergia at plegic joints and, with further therapy, to restore functional movement. A separate pre-clinical trial indicates that AMES treatment is effective at restoring functional movement in individuals who chronically disabled by traumatic brain injury (TBI). There are current about 1 million chronically disabled TBI patients in the US.