Análise da fadiga muscular por meio de sinais eletromiográficos utilizando medidas de informação
The analysis of human movement and muscular fatigue is widely studied through the electromyo- graphy signal. For this work, we used electromyography signals that were assigned to the study and that came from a non-invasive technique of biological signal collection for a dynamic exercise, in this cas...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2017 |
| País: | Brasil |
| Institución: | Universidade Tecnológica Federal do Paraná (UTFPR) |
| Repositorio: | Repositório Institucional da UTFPR (da Universidade Tecnológica Federal do Paraná (RIUT)) |
| Idioma: | portugués |
| OAI Identifier: | oai:repositorio.utfpr.edu.br:1/3336 |
| Acceso en línea: | http://repositorio.utfpr.edu.br/jspui/handle/1/3336 |
| Access Level: | acceso abierto |
| Palabra clave: | Eletromiografia Entropia Fadiga Electromyography Entropy Fatigue CNPQ::ENGENHARIAS Engenharia Elétrica |
| Sumario: | The analysis of human movement and muscular fatigue is widely studied through the electromyo- graphy signal. For this work, we used electromyography signals that were assigned to the study and that came from a non-invasive technique of biological signal collection for a dynamic exercise, in this case cycling. This collection is done by means of electrodes placed in the muscular regions that will be studied. The test submitted submitted ten volunteer cyclists to a ciclossimulator, in which, the subjects were submitted to the test of the race of 20 km against the clock. The data from this course allowed the search for the indication of the appearance of muscular fatigue in the individuals through the analysis of the median frequency of the electromyographic signals. Also, in order to compare, two methods of information measurement are proposed: Rényi’s quadratic entropy and cross-correlation. In the first one, the process of windowing of the EMG signal is performed for the 10 individuals tested and the Rényi quadratic entropy for each window is estimated using the Gaussian nucleus. At the end, the linear interpolation of the result is calculated. In the second method, the cross-correlation of power spectra of signal segments is analyzed to evaluate the degree of similarity between the stretches of the same signal. Finally, the linear interpolation of the result is calculated and it is verified if the signal has decreased. By means of the results of the tests, it is verified that the median frequency contributed to find the regions of the signs of the muscles that have a greater descending slope. For cross-correlation, it was found that the signal resulting from the correlation of two windows indicates by means of linear interpolation that an indicative of decreasing fatigue is not always obtained, and this behavior was also evident in the tests with the entropy. Thus, the statistical results indicate that these two methodologies are auxiliary techniques in the search for recognition of muscular patterns. |
|---|