تعداد نشریات | 20 |
تعداد شمارهها | 1,133 |
تعداد مقالات | 10,350 |
تعداد مشاهده مقاله | 43,912,300 |
تعداد دریافت فایل اصل مقاله | 9,863,165 |
A Multithreading Queuing based sEMG Data Acquisition System for Rehabilitation with Real-Time Visualization | ||
Journal of Rehabilitation Sciences & Research | ||
مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 03 اردیبهشت 1403 | ||
نوع مقاله: Original Articles | ||
شناسه دیجیتال (DOI): 10.30476/jrsr.2024.98627.1368 | ||
نویسندگان | ||
Rajendra Kachhwaha* 1؛ Ajay P Vyas2؛ Rajesh Bhadada2؛ Rajendra Kachhwaha3 | ||
1Department of Computer Science and Engineering, MBM University, Jodhpur, India | ||
2Department of Electronics and Communication, MBM University, Jodhpur, India | ||
3Department of Physiotherapy, Narayan Hrudayalaya Institute of Physiotherapy, Bengaluru, 560099, India. | ||
چکیده | ||
Background: Myoelectric signals have been used in various applications during the past two decades, most notably in rehabilitation technology and hybrid human-machine interfaces. A critical issue encountered in the fabrication of self-engineered and economical devices is the acquisition of a reliable and accurate myoelectric signal. Furthermore, identifying the optimal anatomical site for signal detection on the muscle presents a considerable challenge that this study aims to address. Method: This study is an applied research, specifically a technological development study with experimental elements. This research proposes a multithreading queuing-based technology to display and record muscle activity in real-time when applied to a low-cost multi-channel surface electromyography (sEMG) signal-collection system. This technique has been compared in terms of categorization results utilizing raw (R) dataset and feature (F) dataset via specialized classifiers to categorize sEMG signals of the silent utterance of English vowels from the three facial muscles of a single healthy volunteer. Results: Using the low-cost hardware with the proposed sEMG data acquisition technique, significantly surpasses previous techniques in English vowel classification accuracy, achieving 0.91 mean accuracy for both R (raw) and F (feature) datasets, a substantial increase over past research. Model4, when used with low-cost hardware, attains a notable 0.94 mean accuracy, far exceeding traditional methods, from 14.6% to 74.07% improvement compared to prior studies. Conclusion: The results we obtained with MTQ technique are significantly improved to those of existing hardware configurations, which means that low-cost sEMG data collection devices could be used instead of commercial hardware configurations in the rehabilitation device and related human-machine interface domains. | ||
کلیدواژهها | ||
Surface Electromyography (sEMG)؛ Silent Speech Recognition؛ Signal Processing؛ Rehabilitation Technology؛ Low-Cost Hardware | ||
آمار تعداد مشاهده مقاله: 114 |