Outlier detection in high-density surface electromyographic signals

• Published in: Medical & biological engineering & computing Vol. 50; no. 1; pp. 79 - 89
• Main Authors: Marateb, Hamid R., Rojas-Martínez, Monica, Mansourian, Marjan, Merletti, Roberto, Mañanas Villanueva, Miguel A.
• Format: Journal Article Publication
• Language: English
• Published: Berlin/Heidelberg Springer-Verlag 01.01.2012; Springer Nature B.V
• Subjects: Adult; Algorithms; Arrays; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Channels; Computer Applications; Data mining; Elbow; Electrodes; Electromiografia; Electromyography; Electromyography - methods; Electrònica biomèdica; Enginyeria biomèdica; Focusing; Human Physiology; Humans; Image processing systems; Imaging; Male; Medical equipment; Methods; Multivariate analysis; Muscle, Skeletal - physiology; Muscles; Muscular system; Original Article; Radiology; Recording; Regression analysis; Sensitivity and Specificity; Signal processing; Signal Processing, Computer-Assisted; Skin; Studies; Test sets; Young Adult; Àrees temàtiques de la UPC; Robust statistics; Multichannel surface electromyography; Logistic regression; Feature extraction; Detection theory; Multivariate outlier detection
• ISSN: 0140-0118; 1741-0444; 1741-0444
• DOI: 10.1007/s11517-011-0790-7

Recently developed techniques allow the analysis of surface EMG in multiple locations over the skin surface (high-density surface electromyography, HDsEMG). The detected signal includes information from a greater proportion of the muscle of interest than conventional clinical EMG. However, recording with many electrodes simultaneously often implies bad-contacts, which introduce large power-line interference in the corresponding channels, and short-circuits that cause near-zero single differential signals when using gel. Such signals are called ‘outliers’ in data mining. In this work, outlier detection (focusing on bad contacts) is discussed for monopolar HDsEMG signals and a new method is proposed to identify ‘bad’ channels. The overall performance of this method was tested using the agreement rate against three experts’ opinions. Three other outlier detection methods were used for comparison. The training and test sets for such methods were selected from HDsEMG signals recorded in Triceps and Biceps Brachii in the upper arm and Brachioradialis, Anconeus, and Pronator Teres in the forearm. The sensitivity and specificity of this algorithm were, respectively, 96.9 ± 6.2 and 96.4 ± 2.5 in percent in the test set (signals registered with twenty 2D electrode arrays corresponding to a total of 2322 channels), showing that this method is promising.