Clinically, the recovery process following stroke is characterized by the emergence of stereotypic multi-joint movement patterns that reflect a loss of independent joint control. These patterns, in conjunction with our recent isometric and flexion reflex studies, provide evidence for a loss of certain muscle coactivation and joint torque patterns in the impaired arm. The general aim of our work is to elucidate the role of abnormal neural constraints in upper limb discoordination following hemiparetic stroke, in the following areas:

NeuroImaging | Biomechanics | Rehabilitation and Neurotherapeutic Training

Cortical reorganization following brain injury is being investigated in our laboratories with three techniques: multi-channel EEG recordings in conjunction with anatomical MRI, Transcranial Magnetic Stimulation in conjunction with anatomical MRI, and functional MRI. Through these investigations we seek to illustrate the difference in somatotopic organization between normal and injured brains. Significant differences will support the premise that abnormal movement constraints observed in the hemiparetic extremity of people post stroke are due to reorganization of the brain following brain injury. Investigation of possible roles of cortical reorganization in the emergence of abnormal movement patterns after stroke using EEG, fMRI, and Transcranial Magnetic Stimulation (TMS). By using multiple techniques, we hope to overcome their individual disadvantages. EEG and fMRI are used to understand brain organization during voluntary movments. EEG allows us to investigate in depth, any temporal changes in the movements and also has the great advantage of allowing large isometric exertions at the elbow and shoulder that are not possible in the MRI environment. TMS activates the arm muscles by stimulation of the brain while the subject is at rest, thereby allowing the study of the cortical connections without any additional input by other areas as would happen during voluntary movement. Both EEG and TMS are used in coordination with a 6 degree of freedom load cell and 12 channels of electromyographic (EMG) recording to accurately measure arm movement and muscle activity for a highly controlled experiment. We are currently working on ways to quantitatively measure movements and/or muscle activity in the MRI environment.

-- Cortical Activity Related to Isometric Joint Torques Measured with EEG
-- Motor planning and execution related cortical activities following stroke
-- The Role Of The Cortex In Discoordination After Stroke
-- Roles of cortex and pathways in abnormal muscle coactivation pattern after stroke

-- Development of Brain-Machine-Interface for Stroke Survivors

(back to top)

Investigation of basic mechanisms of movement disorder following stroke. This is done using a 6 degrees of freedom loadcell to measure torques generated at the elbow and shoulder during isometric contractions. In this way, we can quantitatively record movement disabilities following stroke. Spasticity quantification at the elbow and shoulder in stroke subjects using a Biodex motor.

Current therapeutic approaches to improve upper extremity function following stroke have met with limited success. Our studies serve to quantify movement deficits associated with stroke and develop and validate novel training programs to enhance functional recovery. Recent studies in our laboratory (Dewald et al., 1995; Beer et al., 1999) strongly suggest that the spatial disturbances of arm movements following stroke primarily reflect the existence of abnormal activation of shoulder and elbow muscles in the impaired limb. Accordingly, our current projects involve the development and validation of isometric and isokinetic training programs that seek to diminish abnormal constraints on torque generation in the impaired limb. Quantification of torque patterns under dynamic conditions in the upper limb following brain injury. Using a 3-D HapticMASTER robot, we can quantitatively monitor endpoint forces and calculate shoulder torques during dynamic arm movements using a 6 DOF load cell. Feedback is provided to the subject via a virtual arm displayed on a monitor in front of them. The robot creates a virtual environment for the subject to interact with, and can be programmed to provide varying levels of support (making it easier for the subject to lift their limb against gravity), or to require a greater shoulder torque to lift the limb (as if the subject were lifting an object), or prescribe alternate planes of motion. This gives us unprecedented ability to look at the expression of synergies following stroke during dynamic reach and retrieval movements, as well as a new tool to train subjects to work outside of these debilitating synergies.

-- Modifiability of Abnormal Torque Patterns in Chronic Hemiparetic Stroke

(back to top)