Relationship between fractal dimension and firing rate slopes in quadriceps muscle during fatiguing contractions

Cescon, Corrado and Beretta Piccoli, Matteo and Clijsen, Ron and Barbero, Marco (2018) Relationship between fractal dimension and firing rate slopes in quadriceps muscle during fatiguing contractions. UNSPECIFIED. In: ISEK 2018 Proceedings ISEK 2018 Congress - XXII Congress of the International Society of Electrophysiology and Kinesiology, 29th June - 1st July 2018, Dublin, Ireland.

[img] Text
ISEK 2018 Poster Cescon FD-Decomp.pdf

Download (492kB)


BACKGROUND AND AIM:. Myoelectric fatigue has been studied extensively using different parameters extracted from EMG signals (e.g. muscle fiber conduction velocity, mean power frequency, high order statistics). Recently fractal dimension of EMG signals was proposed as an index of central fatigue, as it is related to the syncrhonization of the firing patterns of the active MUs. The aim of the present study was to investigate if the slope of FD during fatiguing contractions was correlated with the increase of firing rate of the active motor units. METHODS. A dataset of EMG signals collected in previous experiments was used for this study. The database included EMG data from 70 subjects. Subjects were sitting on an ergometer were asked to perform isometric leg extension with the knee flexed at 120 degrees. EMG signals were detected in monopolar configuration from Vastus Medialis and Vastus Lateralis muscles of left and right leg using bidimensional arrays of 32 electrodes with 8mm IED (Spes Medica, Italy). Signals were decomposed using the decomposition tool implemented in the OTBiolab software (OTBioelettronica, Torino, Italy) (see figure a). The parameters set for the algorithm were the same for each signal and were selected on preliminary tests on a subportion of the dataset. The instantaneous firing rate of the identified motor units was extracted and linear regression was applied to identify initial value and slope (see figure c). Fractal dimension was computed for each channel and time epoch of 1s (see figure b). The average value among all channels was computed and linear regression was applied to obtain initial value and slope. Pearson correlation coefficient was applied to evaluate the correlation between slope of FD and firing rate. RESULTS. The decomposition algorithm identified in average 3.5 motor units in 122 signals, with an average number of 562 firings for each identified MU. The firing patterns were visually inspected in order to remove duplicates or incorrect results. The total number of analysed signals was 84. A significant correlation was observed between slope of FD and slope of firing rate of the active motor units (see figure d). CONCLUSIONS. Fractal dimension of EMG signals is sensitive to changes of firing rate during fatiguing contractions, thus it could be used as a tool to evaluate central nervous system changes, as it does not require large bidimensional arrays or decomposition algorithms

Actions (login required)

View Item View Item