Citas bibligráficas
Cabezas, F., (2023). Identificación de potenciales eléctricos relacionados a eventos seleccionados por pacientes con enfermedades neurodegenerativas empleando transformada Wavelet [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/27040
Cabezas, F., Identificación de potenciales eléctricos relacionados a eventos seleccionados por pacientes con enfermedades neurodegenerativas empleando transformada Wavelet [Tesis]. PE: Universidad Nacional de Ingeniería; 2023. http://hdl.handle.net/20.500.14076/27040
@mastersthesis{renati/712950,
title = "Identificación de potenciales eléctricos relacionados a eventos seleccionados por pacientes con enfermedades neurodegenerativas empleando transformada Wavelet",
author = "Cabezas Huerta, Franklin Alfredo",
publisher = "Universidad Nacional de Ingeniería",
year = "2023"
}
A Brain - Computer Interface is a technology that allows people to communicate, actívate and control electronic devices, using brain signals, which are acquired and processed by a computer, mainly helping patients with neurological, neuromuscular or neurodegenerative diseases, who due to the disease, have lost the ability to speak, walk, or perform muscle movements. In this thesis, a method of extraction of characteristics and classification of signals is proposed, applied to a database containing brain signals from 4 people (2 healthy and 2 with neurodegenerative diseases), the first sick person suffered from Multiple Sclerosis and the second sick person suffered from Amyotrophic Lateral Sclerosis. These signals contain in certain sections P300 waves not observable with the naked eye, which arose in response to unexpected desired stimuli. The proposed method consists in the use of the Discrete Wavelet Transform for the analysis and extraction of characteristcs of the signals measured by 3 electrodes (1 active and 2 reference), using the Daubechies family of order 4 as the mother wavelet function with 5 levels of decomposition and Support Vector Machines, using a radial basis function (RBF) kernel to generate an optimal hyperplane that allows the classification of brain signals as containers or not containers of P300 waves. With the detection of the P300 waves, the images chosen by the participants have been identified, that is, the detection of these waves allowed knowing the wills of the participants through brain signals. When applying the proposed method, it was evidenced that the obtained wavelet coefficients represent the analyzed brain signals and the brain signals that contain or do not contain P300 waves were identified with averages values of 75.42% accuracy, 76.67% sensitivity and 74.17% specificity.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons