A robotic gait training system integrating split-belt treadmill, footprint sensing and synchronous EEG recording for neuro-motor recovery

This paper presents a robotic gait training system for neuro-motor rehabilitation of hemiplegic stroke survivors. The system is composed of a treadmill consisting of two separated belts, footprint array sensor attached below each belt for gait data acquisition, and an electroencephalography (EEG) device for monitoring brain activities during gait training. The split belt treadmill allow physical therapists to set different treadmill belt velocities to modify physical workload of the patients during walking, thus being able to better improve the symmetry of gait phases between affected and unaffected (sound) legs in comparison with conventional treadmills where there is only one single belt. In contrast to in-shoe pressure sensors, the under-belt footprint sensor array designed in this study not only reduces the preparation complexity of gait training but also collects more gait data for motion analysis. Recorded EEG is segmented synchronously with gait-related events. The processed EEG data can be used for monitoring brain-activities during gait training, providing a neurological approach for motion assessment. One subject with simulated stroke using an ankle-foot orthosis participated in this study. Preliminary results indicate the feasibility of the proposed system to improve gait function and monitor neuro-motor recovery.