Inter-classifier comparison for upper extremity EMG signal at different hand postures and arm positions using pattern recognition

Abstract

The utilization of surface EMG and intramuscular EMG signals has been observed to create significant improvement in pattern recognition approaches and myoelectric control. However, there is less data of different arm positions and hand postures available. Hand postures and arm positions tend to affect the combination of surface and intramuscular EMG signal acquisition in terms of classifier accuracy. Hence, this study aimed to find a robust classifier for two scenarios: (1) at fixed arm position (FAP) where classifiers classify different hand postures and (2) at fixed hand posture (FHP) where classifiers classify different arm positions. A total of 20 healthy male participants (30.62 ± 3.87 years old) were recruited for this study. They were asked to perform five motion classes including hand grasp, hand open, rest, hand extension, and hand flexion at four different arm positions at 0°, 45°, 90°, and 135°. SVM, KNN, and LDA classifier were deployed. Statistical analysis in the form of pairwise comparisons was carried out using SPSS. It is concluded that there is no significant difference among the three classifiers. SVM gave highest accuracy of 75.35% and 58.32% at FAP and FHP respectively for each motion classification. KNN yielded the highest accuracies of 69.11% and 79.04% when data was pooled and was classified at different arm positions and at different hand postures respectively. The results exhibited that there is no significant effect of changing arm position and hand posture on the classifier accuracy.

Intramuscular EMG feature extraction and evaluation at different arm positions and hand postures based on a statistical criterion method

“…The execution of windowing was followed by extracting sixteen-time domain features as demonstrated in Table 1. 33 X is wavelet coefficient, L is length of coefficient, T is threshold value, and w is weight window function. Features were extracted at a threshold value of 0.1, as no exact scientific methods are available for selecting the threshold value.…”

“…Sixteen-time domain features extracted. 33 Serial number Feature Formula 1 Mean absolute value (MAV) MAV = 1 L P L i = 1 x i j j where x is the wavelet coefficient and L is the length of coefficient. 2Waveform length (WL) WL = P L i = 2 x i À x iÀ1 j j where x is the wavelet coefficient and L is the length of coefficient.…”

Simple recognition of hand gestures using single-channel EMG signals

“…2 Electromyography (EMG) is a method for recording muscle electrical activity, 12 the acquisition of EMG signals from a person's skin surface is known as surface electromyography (SEMG), whereas the acquisition of EMG signals from a person's muscles is known as intramuscular electromyography. 13 EMG signals have shown positive potential for prosthesis control in a variety of sectors, including assistive technology and rehabilitation sciences, 14 actually, they are becoming vital biological characteristics, with applications in human-machine interface, prosthetic device development, and rehabilitation equipment. 12 Moreover, EMG provides valuable insights into neuromuscular function and provides detailed motor control data that can be beneficial for both diagnostic purposes and the development of prosthetic appliances.…”

“…2 Electromyography (EMG) is a method for recording muscle electrical activity, 12 the acquisition of EMG signals from a person’s skin surface is known as surface electromyography (SEMG), whereas the acquisition of EMG signals from a person’s muscles is known as intramuscular electromyography. 13…”