Análisis EEG de presencia de dolor en voluntarios orientado a la detección en pacientes no-comunicativos

In this work, a complete process of EEG signal analysis was carried out to detect the presence of pain in volunteers, setting up a system of robust bio-markers with the aim to apply in the future to the detection of pain in non-communicative patients. The development of this process involved the exe...

Descripción completa

Detalles Bibliográficos
Autor: Diego Andres Blanco Mora
Tipo de recurso: tesis doctoral
Estado:Versión aceptada para publicación
Fecha de publicación:2017
País:México
Institución:Instituto Nacional de Astrofísica, Óptica y Electrónica
Repositorio:Repositorio Institucional del INAOE
Idioma:español
OAI Identifier:oai:inaoe.repositorioinstitucional.mx:1009/847
Acceso en línea:http://inaoe.repositorioinstitucional.mx/jspui/handle/1009/847
Access Level:acceso abierto
Palabra clave:info:eu-repo/classification/EEG/EEG
info:eu-repo/classification/Dolor en voluntarios/Pain in volunteers
info:eu-repo/classification/Detección/Detection
info:eu-repo/classification/Electroencefalografía/Electroencephalography
info:eu-repo/classification/cti/1
info:eu-repo/classification/cti/22
info:eu-repo/classification/cti/2203
info:eu-repo/classification/cti/330790
Descripción
Sumario:In this work, a complete process of EEG signal analysis was carried out to detect the presence of pain in volunteers, setting up a system of robust bio-markers with the aim to apply in the future to the detection of pain in non-communicative patients. The development of this process involved the execution of each of the steps in a research work with EEG signals, such as: the acquisition of signals, which was carried out in the INAOE facilities in the Bio-signals laboratory, with volunteers under resting state and pain stimulation by immersion of the left hand in cold water; pre-processing, applying the pre-processing tool based on EEGLab provided by COMA Science Group of Belgium; the extraction of descriptors or bio-markers that could highlight characteristics of the EEG signals that allow differentiating between rest and pain state, including the proposal of a modified phase index (TPLI) and the application of indices that had not been tested for the detection of pain in the state of the art; the evaluation of the performance of the tested descriptors, by means of statistical tools such as ROC curves and p-values, applying the Bonferroni correction for multiple comparisons; interpretation of the results of the selected descriptors of better performance, which were PSD, Spectral Entropy and PLI and WPLI Phase indices; settlement of an inference system that would allow to detect the presence of pain with EEG signals, which obtained a 90% pain detection percentage applied in a new set of data with pain stimulation. Additionally, a set of exploratory works of bio-marker applications to study the resting state in non-communicative patients were developed and included in this work.