An adaptive hand exoskeleton rehabilitation training system integrating virtual reality and an AI-based assessment engine

Results validate all three hypotheses: (i) multimodal fusion exceeded technical targets ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:msup><mml:mi>R</mml:mi> <mml:mn>2</mml:mn></mml:msup> <mml:mo>=</mml:mo> <mml:mn>0.72</mml:mn> <mml:mo>></mml:mo> <mml:mn>0.65</mml:mn></mml:math> , latency 38 ms <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mo><</mml:mo> <mml:mn>50</mml:mn></mml:math> ms); (ii) clinical efficacy surpassed MCID with large effect sizes (FMA-UE <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mo>+</mml:mo> <mml:mn>9.1</mml:mn> <mml:mo>></mml:mo> <mml:mn>6</mml:mn></mml:math> points, <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mi>d</mml:mi> <mml:mo>=</mml:mo> <mml:mn>0.98</mml:mn> <mml:mo>></mml:mo> <mml:mn>0.8</mml:mn></mml:math> ), exceeding published spontaneous recovery rates (2-4 points); (iii) ablation experiments demonstrated <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mo>≥</mml:mo> <mml:mn>15</mml:mn> <mml:mi>%</mml:mi></mml:math> degradation when removing any component, confirming non-additive synergistic effects of bio-AI-VR integration. Compared to recent brain-computer interface systems using EEG-based motor imagery, this approach achieves paradigm shift toward execution-based rehabilitation with direct motor intent capture and real-time physical feedback. The AAN control law elevates from low-level motion control to high-level rehabilitation strategy, spanning multiple temporal scales (window to course-level) and dual channels (physical assistance <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mo>+</mml:mo></mml:math> cognitive challenge). Limitations include single-arm design limiting causal inference, small sample size ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mi>n</mml:mi> <mml:mo>=</mml:mo> <mml:mn>24</mml:mn></mml:math> ), short intervention period (4 weeks), FMA-UE proxy construction via linear interpolation, and controlled clinical setting vs. real-world deployment. Future work requires larger RCTs with active control arms, extended follow-up (3-6 months), dense longitudinal assessments, exploration of deep learning architectures for temporal modeling, and validation in home-based telerehabilitation settings. The bio-AI-VR system demonstrates feasibility of data-driven, multimodal closed-loop rehabilitation, offering a wearable, low-latency, and personalized solution for post-stroke hand recovery that bridges the gap between laboratory innovation and clinical translation.