23 October 2020 in IEEE Sensors Journal
Musculoskeletal disorders are the biggest cause of disability worldwide and wearable mechatronic rehabilitation devices have been proposed as a potential tool for providing treatment; however, before such devices can be widely adopted, improvements in reliability are necessary. Changes in system dynamics caused by user actions, such as picking up a weight, can greatly affect control stability; hence, detecting these changes can lead to improved system performance. It is difficult to integrate conventional sensing technologies for completing this task into a wearable device in an unobtrusive way, therefore an alternative solution using bioelectrical signals, such as electroencephalography (EEG) and electromyography (EMG), to detect task weight is proposed. In this study, EEG and EMG signals were collected during dynamic elbow flexion–extension motion at different speeds, while holding different weights. These biosignals were used to develop different EEG–EMG fusion models to classify the weight the user was holding while moving. It was found that using a Weighted Average fusion method, and incorporating speed information into the model, provided the best performance, with an accuracy of 83.01 ± 6.04% when classifying three task weights. This work demonstrated the feasibility of using EEG–EMG fusion for classification of task weight during dynamic motion, which can be used to improve the adaptability and robustness of wearable mechatronic rehabilitation devices.
15 April 2025 in IEEE Sensors Journal
The rise in popularity of wearable robotic devices has brought many opportunities in the area of assistive devices for rehabilitation. However, despite their numerous advantages, these wearable mechatronic devices are not widely adopted due to a poor interaction between the device and the wearer. To solve this issue, several studies have proposed the use of sensor fusion of electroencephalography (EEG) and electromyography (EMG) signals using Convolutional Neural Networks (CNN) to detect the motion intention of the user and control the wearable device. Although normalization techniques can be applied during data pre-processing in order to improve performance, the correct normalization methods to apply for combined EEG and EMG data are unknown, hindering the potential improvement of the resulting CNN models. Therefore, in this study, no normalization, subject-wise, speed-wise, image-wise, and channel-wise normalization methods were compared. The effectiveness of these methods was tested on the overall and speed specific accuracies of CNN models trained using a database of combined EEG–EMG data collected during elbow flexion–extension motions at different speeds. The results of this study showed that no normalization improved the performance of the overall accuracy of CNN models, with grouped spectrogram-based models achieving an accuracy of 81.57 ± 7.11%. A similar trend was found for speed specific accuracies, in which no normalization showed to be a slightly better option during slow motions. Overall, these results provide information on the different tools that can be used to improve the performance of CNN-based models used for the control of wearable robotic devices.