SURFACE ELECTROMYOGRAM DURING VOLUNTARY CONTRACTION - PROCESSING TOOLS AND RELATION TO PHYSIOLOGICAL EVENTS

Surface electromyography (SEMG) has been used extensively in the last years in a variety of applications, including muscle function assessme nt, pathology identification, ergonomics, pattern analysis, or populat ion characterization. Advanced processing methods, especially in the s pectral domain, provide the research worker with more and more precise and user-friendly tools for signal characterization, analysis, and cl assification. The use of such sophisticated tools requires many assump tions on signal characteristics, and the wide variety of computing opt ions related to each processing method makes it difficult to compare t he results of different works when these options are omitted in the re ports or improperly applied. This work first aims at taking stock of t he various processing methods which have emerged in the last years aro und surface electromyography: signal acquisition, random feature extra ction, time and spectral parameter determination, statistical tests ap plication. The main methods are briefly explained and discussed, then variations between apparently equivalent methods are pointed out, nece ssary hypotheses are underlined, and the use of such methods in SEMG p rocessing is shown with respect to the more recent works. A second sec tion shows how authors deal with parameters extracted from SEMG in ord er to relate them to physiological modifications (force, fiber type, f iber environment).