The non-stationary property of electromyography (EMG) signals in real life settings usually hinders the clinical application of the myoelectric pattern recognition for prosthesis control. The classical EMG pattern recognition approach consists of two separate steps: training and testing, without considering the changes between training and testing data induced by electrode shift, fatigue, impedance changes and psychological factors, and often results in performance degradation. The aim of this study was to develop an adaptive myoelectric pattern recognition system, aiming to retrain the classifier online with the testing data without supervision, providing a self-correction mechanism for suppressing misclassifications. This paper presents an adaptive unsupervised classifier based on support vector machine (SVM) to improve the classification performance. Experimental data from 15 healthy subjects were used to evaluate performance. Preliminary study on intra-session and inter-session EMG data was conducted to verify the performance of the unsupervised adaptive SVM classifier. The unsupervised adaptive SVM classifier outperformed the conventional SVM by 3.3% and 8.0% for the combination of time-domain and autoregressive features in the intra-session and inter-session tests, respectively. The proposed approach is capable of incorporating the useful information in testing data to the classification model by taking into account the overtime changes in the testing data with respect to the training data to retrain the original classifier, therefore providing a self-correction mechanism for suppressing misclassifications.

In this study, we used muscle and motor unit indices, derived from convenient surface electromyography (EMG) measurements, for examination of paretic muscle changes post stroke. For 12 stroke subjects, compound muscle action potential and voluntary surface EMG signals were recorded from paretic and contralateral first dorsal interosseous, abductor pollicis brevis, and abductor digiti minimi muscles. Muscle activation index (AI), motor unit number index (MUNIX), and motor unit size index (MUSIX) were then calculated for each muscle. There was a significant AI reduction for all the three muscles in paretic side compared with contralateral side, providing an evidence of muscle activation deficiency after stroke. The hand MUNIX (defined by summing the values from the three muscles) was significantly reduced in paretic side compared with contralateral side, whereas the hand MUSIX was not significantly different. Furthermore, diverse changes in MUNIX and MUSIX were observed from the three muscles. A major feature of the present examinations is the primary reliance on surface EMG, which offers practical benefits because it is noninvasive, induces minimal discomfort and can be performed quickly.

The objective was to re-evaluate the controversial reports of EMG-torque relation between impaired and non-impaired sides using linear electrode array EMG recordings. Ten subjects with chronic stroke performed a series of submaximal isometric elbow flexion tasks. A 20-channel linear array was used to record surface EMG of the biceps brachii muscles from both impaired and non-impaired sides. M-wave recordings for bilateral biceps brachii muscles were also made. Distribution of the slope of the EMG-torque relations for the individual channels showed a quasi-symmetrical “M” shaped pattern. The lowest value corresponded to the innervation zone (IZ) location. The highest value from the slope curve for each side was selected for comparison to minimize the effect of electrode placement and IZ asymmetry. The slope was greater on the impaired side in 4 of 10 subjects. There were a weak correlation between slope ratio and strength ratio and a moderate to high correlation between slope ratio and M-wave ratio between two sides. These findings suggest that the EMG-torque relations are likely mediated and influenced by multiple factors. Our findings emphasize the importance of electrode placement and suggest the primary role of peripheral adaptive changes in the EMG-torque relations in chronic stroke.

This study investigates the effect of the feature dimensionality reduction strategies on the classification of surface electromyography (EMG) signals toward developing a practical myoelectric control system. Two dimensionality reduction strategies, feature selection and feature projection, were tested on both EMG feature sets, respectively. A feature selection based myoelectric pattern recognition system was introduced to select the features by eliminating the redundant features of EMG recordings instead of directly choosing a subset of EMG channels. The Markov random field (MRF) method and a forward orthogonal search algorithm were employed to evaluate the contribution of each individual feature to the classification, respectively. Our results from 15 healthy subjects indicate that, with a feature selection analysis, independent of the type of feature set, across all subjects high overall accuracies can be achieved in classification of seven different forearm motions with a small number of top ranked original EMG features obtained from the forearm muscles (average overall classification accuracy >95% with 12 selected EMG features). Compared to various feature dimensionality reduction techniques in myoelectric pattern recognition, the proposed filter-based feature selection approach is independent of the type of classification algorithms and features, which can effectively reduce the redundant information not only across different channels, but also cross different features in the same channel. This may enable robust EMG feature dimensionality reduction without needing to change ongoing, practical use of classification algorithms, an important step toward clinical utility.