Robust outlier detection in high-density surface electromyographic signals

• Published in: 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Vol. 2010; pp. 4850 - 4853
• Main Authors: Marateb, H R, Rojas-Martinez, M, Mananas Villanueva, M A, Merletti, R
• Format: Conference Proceeding Journal Article
• Language: English
• Published: United States IEEE 01.01.2010
• Subjects: Algorithms; Arrays; Diagnosis, Computer-Assisted - methods; Electromyography; Electromyography - methods; Feature extraction; Humans; Muscle Contraction - physiology; Muscle, Skeletal - physiology; Muscles; Pattern Recognition, Automated - methods; Reproducibility of Results; Robustness; Sensitivity; Sensitivity and Specificity; Tuning
• ISBN: 1424441234; 9781424441235
• ISSN: 1094-687X; 1557-170X; 1558-4615
• DOI: 10.1109/IEMBS.2010.5627280

High Density surface Electromyography (HDsEMG) has been applied in both research and clinical applications for non-invasive neuromuscular assessment in several different fields using 2-D array. Proper interpretation of HDsEMG signals requires identifying "good" channels (where there is no short-circuit or bad-contact or major power line interference problem). Recording with many channels usually implies bad-contacts (that introduces large power line interference) and short-circuits (when using gels). In addition to online monitoring the electrode-contact quality, it is necessary to identify "bad" channels, or outliers, prior to the analysis of HDsEMG signal. In this paper we introduce a robust method to identify outliers in a set of monopolar HDsEMG signals recorded from Biceps and Triceps Brachii, Anconeus, Brachioradialis and Pronator Teres. The sensitivity and precision of this method show that this approach is promising.