FARFAN Fernando Daniel
congresos y reuniones científicas
Evaluación de estacionariedad en señales electromiográficas dinámicas
GABRIEL ALFREDO RUIZ; ALVARO GABRIEL PIZÁ; FERNANDO DANIEL FARFÁN
Congreso; XXII Congreso Argentino De Bioingenieria y XI Jornadas De Ingenieria Clinica; 2020
Sociedad Argentina de Bioingeniería
This work proposes a new analysis methodology to determine the degree of stationarity of electromyographic signals recorded in dynamic conditions and under different states of static charge. The proposed method is compared with three specific statistical tests: Run test (RT), Reverse Arrangements Test (RAT) and Modified Reverse Arrangements Test (MRAT). In order to prove the accuracy of the statistical tests for stationarity, two signals were generated, a stationary random signal with normal distribution (mean 0 and variance 1) and another non-stationary signal consisting of the sum of a sine wave with exponentially decreasing amplitude and random noise with distribution normal Gaussian. The three tests shown to be inaccurate for stationarity of the test signals, i.e. although the three tests classified the stationary signal as stationary, all of them classified incorrectly the non-stationary signal. Similarly, inconsistencies were detected when RT, RAT and MRAT were tested with real signals. The results of the RT test indicated the trend presence for all the EMG segmentations studied, contrary to the results obtained with the RAT and MRAT tests. On the other hand, it was also found that the lengths of EMG stationary segments, determined with the proposed method, matches with those obtained by other authors, and that the degree of stationarity depends strongly on the kinematic conditions of the movements and to a lesser extent on the kinetic conditions.