Paramétrisation des potentiels d'action d'une unité motrice détectés non-invasivement

Abstract

Muscular contraction is very vital for our life as it allows all kind of movement and physical activities. The motor unit (MU) is the smallest functional element of the muscular contraction and thus, it is very important to study and investigate. In this work, MUs have been studied and investigated throughout their action potentials (MUAP) detected via multi-electrodes placed on the surface of the muscle. The idea of ‘besides the MU properties and depending on the recording techniques, MUAPs can have unique pattern’ was adopted. The aim of this work was to recognise whether a Laplacian-detected MUAP extracted from the surface electromyographyc signal (sEMG) is isolated or overlapped basing on novel morphological features using fuzzy classifier. This recognition can be helpful to investigate abnormal MUAPs and to decompose the SEMG signal. The overlapping between MUAPs can leads to ambiguous situation in which, overlapped MUAPs seen as isolated ones. Moreover, by the noise effect, isolated MUAPs can be seen as overlapped ones. Additionally, the overlapping is a matter of a degree. From these traits of overlapping, fuzzy logic seems to be able to undertake those situations where the ambiguity and uncertainty are present. A fuzzy classifier based on ‘If…then’ rules was elaborated using learning examples and optimized by genetic algorithm (GA) to improve the classification rate. Training data set was constructed using signals provided by three muscles: the abductor pollicis brevis (APB), the first dorsal interosseous (FDI) and the biceps brachii (BB) muscles of 11 healthy subjects. The proposed fuzzy classier recognized automatically the isolated MUAPs with a performance of 95.03% which was improved to 97.8% by adjusting the certainty grades of rules using GA. Synthetic signals were used as reference to further evaluate the performance of the elaborated classifier. The recognition of the isolated MUAPs depends largely on noise level and is acceptable down to the signal to noise ratio of 20 dB with a detection probability of 0.96. The recognition of overlapped MUAPs depends slightly on the noise level with a detection probability of about 0.8. The corresponding misrecognition is caused principally by the synchronisation and the small overlapping degree.