Customizable cooperative multi-joint control to enhance walking mobility after stroke

点击翻译按钮获取中文摘要

Most stroke survivors walk slowly and are at an increased risk of falls. As a result, many adopt a sedentary lifestyle with limited functional independence that negatively impacts health and can be socially isolating. Physical therapy including advanced rehabilitation techniques have improved function, but there is no intervention available that enables stroke survivors with moderate and severe impairment to walk at speeds necessary for independent community ambulation. The long-term goal of this work is to restore stroke survivors’ ability to walk safely in the community at speeds necessary for independence. Our approach utilizes an implanted neuroprosthesis, that is a device inside the body that applies small electrical pulses to activate the nerves that cause the muscles serving multiple joints to contract in a coordinated manner for functional movement of the entire limb. The system measures volitional muscle activity and body motion and then coordinates stimulation at the different joints in the leg to produce the necessary movement for safe walking at functionally relevant speeds. The benefit of such an approach is that it is always available and does not require setup for individuals with impaired hand control. The implanted hardware also improves reliability and bypasses sensory fibers that can cause discomfort. Our team has shown in a case study that targeting muscles throughout the paretic limb can substantially improve walking speed and endurance. This study will expand this work through achieving the following Aims: 1) determining the clinical impact of an implanted multi- joint neuroprosthesis on post-stroke gait, and 2) developing and assessing an advanced neuroprosthesis cooperative control strategy. This study will implement an available neuroprosthesis that incorporates an external control unit and some external sensors in preparation for implementation of a fully implanted system that has been developed at our Center. Six participants will be implanted with devices that include 12-channels of stimulation and 2-channels for recording muscle activity. External sensors will measure limb motion. After the device is implanted, stimulation patterns will be generated and participants will undergo training to use the device. A simple triggering pattern will be created for home use and then we will implement our advanced controller in the laboratory via machine-learning techniques. Once a controller and stimulation pattern have been defined, we will determine how much faster, more safely, and easier walking is with the neuroprosthesis compared to without and confirm whether these effects are maintained over time. We will also determine if the advanced controller substantially improves walking ability over the simple triggering methods that have been previously implemented. Successful completion will confirm approaches for a post-stroke neuroprosthesis for walking and generate preliminary effect sizes for subsequent clinical trials to evaluate home and community use of a fully implanted system. This study may lead to a new clinical tool that can empower independent walking after stroke, improve quality of life, and enhance overall health of stroke survivors.