The classification of movement intention through machine learning models: the identification of significant time-domain EMG features

Abstract

Electromyography (EMG) signal is one of the extensively utilised biological signals for predicting human motor intention, which is an essential element in human-robot collaboration platforms. Studies on motion intention prediction from EMG signals have often been concentrated on either classification and regression models of muscle activity. In this study, we leverage the information from the EMG signals, to detect the subject’s intentions in generating motion commands for a robot-assisted upper limb rehabilitation platform. The EMG signals are recorded from ten healthy subjects’ biceps muscle, and the movements of the upper limb evaluated are voluntary elbow flexion and extension along the sagittal plane. The signals are filtered through a fifth-order Butterworth filter. A number of features were extracted from the filtered signals namely waveform length (WL), mean absolute value (MAV), root mean square (RMS), standard deviation (SD), minimum (MIN) and maximum (MAX). Several different classifiers viz. Linear Discriminant Analysis (LDA), Logistic Regression (LR), Decision Tree (DT), Support Vector Machine (SVM) and k-Nearest Neighbour (k-NN) were investigated on its efficacy to accurately classify the pre-intention and intention classes based on the significant features identified (MIN and MAX) via Extremely Randomised Tree feature selection technique. It was observed from the present investigation that the DT classifier yielded an excellent classification with a classification accuracy of 100%, 99% and 99% on training, testing and validation dataset, respectively based on the identified features. The findings of the present investigation are non-trivial towards facilitating the rehabilitation phase of patients based on their actual capability and hence, would eventually yield a more active participation from them.

Classification of biological signals and time domain feature extraction using capsule optimized auto encoder‐electroencephalographic and electromyography

“…Here, under three movements, the proposed model attains the best accuracy, precision, recall, F ‐measure and specificity. Moreover, the performance of the optimized motion classifier model is compared with several existing approaches such as k ‐nearest neighbor (KNN), SVM, LDA, logistic regression and DT 35 . SVM, an LDA classifier, was introduced to classify upper limb motions using EMG signals.…”

“…Moreover, the performance of the optimized motion classifier model is compared with several existing approaches such as k-nearest neighbor (KNN), SVM, LDA, logistic regression and DT. 35 SVM, an LDA classifier, was introduced to classify upper limb motions using EMG signals. The kernel employed in these studies were the radial basis, linear, polynomial and sigmoid.…”

Neuromechanical Biomarkers for Robotic Neurorehabilitation

“…In addition, hybrid time-frequency features are proposed to overcome the limitation of time features, which relies in stationary properties of the EMG signal. These features are less applied due to computation costs, and are on time-frequency methods such as Discrete Wavelet Transform and Wavelet Packet Transform (Phinyomark et al, 2012;Nazmi et al, 2016) Comment on current/potential applications: Currently, EMG features are being used to the enhancement of robot-assisted upper limb rehabilitation platforms, by means of using the subject's intentions to generate proper feedback for the robotic system (Cahyadi et al, 2018b;Bouteraa et al, 2020;Khairuddin et al, 2021). In particular, due to their relative low computational cost, their potential combination with machine learning algorithms and other technologies such as virtual reality (Meng et al, 2019) could be the key to develop dynamic rehabilitation devices that can boost the personalization of motor training (Abdallah et al, 2017;Arteaga et al, 2020;Samuel et al, 2021) 4.…”

Deep learning analysis based on multi-sensor fusion data for hemiplegia rehabilitation training system for stoke patients

“…1) Support vector machine [36,37,38]. SVM is a type of generalized linear classifier that classifies data by supervised learning, which is widely used to realize motion intention recognition.…”