Kaichen Tang, MS (2025)
Primary advisor: Xiaoqian Jiang, PhD
Committee members: Yejin Kim, PhD and Shayan Shams, PhD
PhD thesis: McWilliams School of Biomedical Informatics at UTHealth Houston.
ABSTRACT
Stroke is a leading cause of long-term disability, often requiring intensive rehabilitation and frequent clinical assessments such as the Fugl-Meyer Assessment. However, traditional in-person evaluations are resource-heavy and difficult to scale, limiting access for many patients. Additionally, monitoring vital signs like blood pressure and oxygen saturation typically depends on specialized equipment, leading to fragmented care and incomplete recovery insights. Even when assessments and monitoring are available, patients often struggle to maintain high-quality, intensive exercise routines without supervision, further limiting rehabilitation outcomes.
To address these challenges, this dissertation proposes an AI-driven, smartphone-based framework that integrates automated motor function assessment, non-invasive vital sign estimation, and patient engagement. First, a deep learning pipeline is developed to automate Fugl-Meyer Assessment upper extremity scoring using smartphone video. Combining top-down pose estimation and a 3D convolutional network, the system achieves an average F1-score of 0.883 across FMA-UE tasks. Second, a transformer-based model predicts ABP and SpO? from PPG signals with clinically acceptable accuracy, validated on subjects in the MIMIC III dataset. Third, an iOS application (MusicRehab) was developed to enable home-based rehabilitation by combining real-time motion tracking with adaptive music feedback. Built on co-adaptive human-machine interaction principles, the system adjusts tempo based on motor performance features. A simulation framework with multiple controllers and a harmonic gain metric was used to evaluate adaptation strategies, and validation confirmed the accuracy of iPad-based motion tracking.
This work illustrates the feasibility of a unified, informatics-driven stroke rehabilitation ecosystem that consolidates motor assessment, vital sign estimation, and engagement strategies into a low-cost, privacy-conscious mobile solution. By enabling home-based care with minimal hardware requirements, the proposed platform lays the groundwork for a scalable and integrative approach to long-term stroke rehabilitation.