Innovative Control Strategy Enhances Stroke Rehabilitation With Casia-exo Exoskeleton

Stroke is 1 of nan starring causes of non-traumatic disablement worldwide, affecting much than 15 cardinal group each year, pinch astir three-quarters experiencing semipermanent functional impairments. This makes it important to create semipermanent rehabilitation programs that tin beforehand centrifugal relearning, heighten neural plasticity, and reconstruct regular centrifugal function. Robot-assisted rehabilitation, which combines neuroscience, biomechanics, and precocious power systems, is emerging arsenic a highly promising approach.

In caller years, exoskeleton-type rehabilitation robots that alteration distributed relationship crossed aggregate joints person gained fame owed to further improvements successful upper-limb centrifugal abilities. Some salient examples of specified robots see ArmeoPower, ANYexo, and EXO-UL8. They employment power strategies that optimize robotic involution and maximize progressive diligent participation. However, fewer existing systems simultaneously incorporated mobility volition detection, desired trajectory generation, and assistance level personalization for effective and businesslike neurorehabilitation.

In an innovation, a squad of researchers from China, led by Professor Zeng-Guang Hou from nan State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences, person developed CASIA-EXO, an upper-limb exoskeleton, and projected an breathtaking power strategy for centrifugal learning successful post-stroke patient-in-the-loop neurorehabilitation. Their caller findings were published successful Volume 12, Issue 8 of nan IEEE/CAA Journal of Automatica Sinica connected 20 August 2025. The squad includes researchers Chen Wang and Liang Peng from MAIS.

According to Prof. Hou, "CASIA-EXO is simply a 5 degree-of-freedom biomimetic exoskeleton that comprises 3 rotational joints adopting an oblique statement and 2 rotational joints co-locating successful a serial chain. Its dynamics is modelled and linearized to place chartless parameters embedded successful nan shoulder, elbow and wrist mechanisms."

Following nan modelling, nan researchers propose a caller patient-in-the-loop power strategy for rehabilitation training for post-stroke centrifugal recovery.

"It consists of nan intention-based trajectory readying and performance-based involution adaptation. While an oscillator-based volition estimator quantifies nan time-varying training request and integrates nan invariant laws successful normal mobility patterns into nan multi-joint trajectory generation, nan performance-based adaptive algorithm alters nan assistance and guidance levels during nan robot-aided rehabilitation, which tin heighten progressive information of subjects pinch various impairment levels," explains Dr. Wang.

The investigation squad conducted various experiments to show the efficacy of their projected system, wherein 10 patient subjects sat successful a chair pinch their ascendant limb coupled pinch CASIA-EXO and passed nan woody boxes complete and past brought them backmost utilizing a virtual reality display. They recovered that their caller power strategy steadily individualized nan training trajectory and involution level arsenic per nan subject's changing requirements and centrifugal abilities, facilitating closed-loop robot-aided rehabilitation. Consequently, nan adjacent practice betwixt trajectory readying and involution adjustment tin assistance centrifugal relearning and functional betterment successful patients pinch centrifugal impairments arsenic people arsenic possible.

This breakthrough underscores nan imaginable of CASIA-EXO to present safer, smarter, and much personalized rehabilitation for changeable survivors.