سامانه همایش های دانشگاه آزاد اسلامی واحد نجف آباد

صفحه اصلی / اولین همایش بین المللی تحول خدمات سلامت: تلفیق مراقبت، درمان، بازتوانی و امیدبخشی

Neuromuscular Interfaces and Cyborg Systems in the Recovery of Atrophied Muscles: A Systematic Review of Adaptive EMG Signal Processing Approaches

نویسندگان :

Faezeh Mahmoudi¹ Alireza Zarei²

1- موسسه آموزش عالی آپادانا 2- دانشگاه بین المللی امام رضا علیه السلام

کلمات کلیدی :

Muscular Atrophy،Electromyography،Bionics،Man-Machine Systems،Exoskeleton Device،Rehabilitation

چکیده :

Background: Muscular atrophy severely impairs motor function, necessitating advanced rehabilitation interventions. The convergence of neuromuscular interfaces and adaptive surface electromyography (sEMG) signal processing has catalyzed the development of cyborg and bionic systems. However, integrating these engineering advancements into robust clinical rehabilitation pipelines remains challenging. This systematic review aims to synthesize current adaptive sEMG processing approaches and evaluate their efficacy in bionic systems for the recovery of atrophied muscles. Method: Conducted in strict adherence to the PRISMA 2020 guidelines, a comprehensive search strategy utilizing MeSH terms (including Muscular Atrophy, Bionics, and Signal Processing) was executed across major databases. From an initial pool of 600 records, rigorous screening and quality assessment using the MINORS criteria were applied. Ultimately, 21 studies between 2025 and 2026, met the final inclusion criteria. Due to clinical and algorithmic heterogeneity, a narrative synthesis was performed to map the intersections of interface technologies, signal processing frameworks, and clinical outcomes. Finding: The synthesized evidence highlights a pivotal transition toward hybrid, adaptive algorithms that mitigate the inherent non-stationarity of sEMG signals in atrophied muscles. While advanced processing pipelines demonstrate enhanced motion intention decoding and robust physical human-robot interaction (pHRI), significant limitations persist. Specifically, computational overhead, vulnerability to real-world noise, and the “black-box” nature of complex deep learning models constrain clinical translation. Conclusion: Neuromuscular interfaces powered by adaptive sEMG signal processing hold profound potential for cyborg-assisted muscle recovery. To bridge the gap between engineering testbeds and clinical rehabilitation, future paradigms must integrate lightweight architectures, foundation models, and Explainable AI (XAI) to enable personalized, closed-loop therapeutic interventions.