NSF Award
2404476
Stroke is a leading cause of disability in the United States. In the wake of a stroke, patients are often left with arm impairments that current rehabilitation methods struggle to effectively treat. This research project aims to develop innovative methods in artificial intelligence (AI) and high-dimensional motion capture technology to analyze movement. This project will develop new methods of analyzing multimodal data with contrasting images of patient movements to better assess patient capabilities in rehabilitation tasks. It also develops new statistical methods to identify anomalies in the movements (e.g., errors in performing the rehabilitation task). This research could substantially improve our ability to utilize multimodal data, as well as enhance stroke rehabilitation, therefore providing significant social and scientific impacts.<br/><br/>This research project focuses on three key contributions to improve stroke rehabilitation. Firstly, the project will develop a unique method to amalgamate multimodal data into a shared representation space, allowing for a precise count of training movements. Secondly, the project will introduce a novel framework for quantifying anomalies from high-dimensional data, enabling accurate measurement of motion impairment. Lastly, the project will create an AI-powered rehabilitation system, integrating the first two contributions, and assess its effectiveness using an independent cohort's data. This research holds transformative potential in signal processing, statistics, and neurorehabilitation, enhancing the rigor of therapeutic trials and the delivery of evidence-based regimens. In addition, the research will advance human action recognition technology and contribute to the development of healthcare-savvy data scientists through new curricula and early-career researcher training at the intersection of data science and healthcare.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
