Improved post-stroke walking balance through adaptive multisensory augmentation

About This Grant

Veterans who experience a stroke are often unable to return to their prior levels of activity participation, due in part to limited independent mobility. A major contributor to mobility deficits is walking imbalance, which increases the risk of falls and fear of falling, either of which can decrease quality of life. While several common rehabilitation methods (e.g., locomotor training, strengthening) can improve some aspects of function, they have failed to sufficiently address post-stroke deficits in walking balance and the related increase in fall risk. In part, this lack of success is likely due to current interventions not being targeted toward the specific mechanisms that underlie changes in the control of walking balance after a stroke. A long-term goal of this research is to create a toolbox of mechanism-based interventions for post-stroke walking balance, from which clinicians can select the most promising option for individual Veteran patients. As a step toward this goal, this project focuses on one potential approach to improving walking balance – sensory augmentation. The specific objective of this research is to develop a novel augmentation approach that targets the somatosensory deficits that can limit walking function in people with stroke. The central hypothesis of this work is that the benefits of sensory augmentation will be enhanced through multisensory, personalized methods that adapt to walking context. The proposed research will build on prior foundational work by expanding beyond a focus on augmenting singular sensory sources, moving from a one-size-fits-all approach to personalization, and integrating automatic adaptability into the augmentation control. This will be accomplished through three Specific Aims. The first Specific Aim is to quantify the effects of multisensory augmentation on post-stroke walking balance. This investigation of several possible sensory targets for augmentation will reveal which targets are most promising, as well as how they interact. The second Specific Aim is to assess the feasibility, reliability, and effectiveness of human-in-the-loop optimization to personalize multisensory augmentation for individual participants. These experiments will use a model-based optimization strategy to account for the high level of variability between participants in terms of the responses to sensory stimulation. Finally, the third Specific Aim is to determine whether adaptive multisensory augmentation improves performance of time-varying gait tasks. The results of this Aim will be a first step toward implementing somatosensory augmentation in real-world walking contexts, which are substantially more complex than simply walking at a constant speed. The proposed project is based on a combination of a mechanistic understanding of post-stroke walking balance and a strong foundation in the established neurophysiological effects of somatosensory stimulation. Accomplishing the Aims of this project would have the potential to open the door for new intervention methods able to improve post-stroke walking balance, and to be targeted to the patients most likely to benefit.