A data-physics hybrid generative model for patient-specific post-stroke motor rehabilitation using wearable sensor data

2025-12-16

AI Summary

Post-stroke motor rehabilitation assessment lacks dynamic task prediction capability, hindering clinical judgment of patients’ ability to safely perform real-world locomotor tasks (e.g., ramp walking, stair climbing). To address this, we propose a data–physics hybrid generative model that reconstructs subject-specific neuromuscular control and predicts gait performance on complex terrains using only 20-meter level-ground walking data from wearable sensors. Our approach innovatively integrates target-conditioned deep reinforcement learning with adversarial imitation learning, jointly coupling a healthy movement atlas and a proportional-derivative (PD) physics-based controller to enable interpretable and generalizable personalized gait generation. In a cohort of 11 stroke survivors, the model improved prediction accuracy for joint angles and endpoint positions by 4.73% and 12.10%, respectively, while reducing training time to 25.56% of that required by pure physics-based models. Multicenter validation demonstrated that clinical decision support enabled by our model increased Fugl-Meyer Assessment scores by 6.0 points versus 3.7 points in the control group.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: State Estimation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Dynamic prediction of locomotor capacity after stroke is crucial for tailoring rehabilitation, yet current assessments provide only static impairment scores and do not indicate whether patients can safely perform specific tasks such as slope walking or stair climbing. Here, we develop a data-physics hybrid generative framework that reconstructs an individual stroke survivor's neuromuscular control from a single 20 m level-ground walking trial and predicts task-conditioned locomotion across rehabilitation scenarios. The system combines wearable-sensor kinematics, a proportional-derivative physics controller, a population Healthy Motion Atlas, and goal-conditioned deep reinforcement learning with behaviour cloning and generative adversarial imitation learning to generate physically plausible, patient-specific gait simulations for slopes and stairs. In 11 stroke survivors, the personalized controllers preserved idiosyncratic gait patterns while improving joint-angle and endpoint fidelity by 4.73% and 12.10%, respectively, and reducing training time to 25.56% relative to a physics-only baseline. In a multicentre pilot involving 21 inpatients, clinicians who used our locomotion predictions to guide task selection and difficulty obtained larger gains in Fugl-Meyer lower-extremity scores over 28 days of standard rehabilitation than control clinicians (mean change 6.0 versus 3.7 points). These findings indicate that our generative, task-predictive framework can augment clinical decision-making in post-stroke gait rehabilitation and provide a template for dynamically personalized motor recovery strategies.

Problem

• Predicting dynamic locomotor capacity after stroke using patient-specific data
• Generating personalized gait simulations for slopes and stairs rehabilitation scenarios
• Augmenting clinical decision-making with task-conditioned locomotion predictions

Innovation

• Hybrid generative model combines wearable sensor data with physics
• Uses deep reinforcement learning for patient-specific gait simulations
• Predicts locomotion across tasks to guide personalized rehabilitation strategies