Modelo y simulación de actividad eléctrica neuronal para generación sintética de EEG
Archivos
Autores
Erazo Erazo, Andrés Mauricio
Director
Espinosa Oviedo, Jairo José (Thesis advisor)
Tipo de contenido
Trabajo de grado - Maestría
Idioma del documento
EspañolFecha de publicación
2012
Título de la revista
ISSN de la revista
Título del volumen
Documentos PDF
Resumen
Con el objetivo de formar una imagen tridimensional que ayude a identificar y caracterizar las fuentes eléctricas que generan las señales adquiridas en un electroencefalograma (EEG) se requiere una previa solución al problema directo. En este, el principal objetivo es la búsqueda de un modelo de cabeza que explique matemáticamente la propagación de la señal generada al interior del cerebro hasta el cuero cabelludo, en donde se encuentran los sensores. Para encontrar una solución se analiza la representación matemática de las señales generadas por la actividad neuronal y la manera como estas se propagan a través de modelos esféricos, esféricos multicapa y realistas que representan la cabeza, luego se analiza como esta representación se resuelve por medio del método de elementos de frontera BEM. Finalmente, se presenta una comparación de esta metodología con métodos utilizados actualmente y se desarrolla un programa computacional con interfaz gráfica que permite encontrar una solución para el problema directo en EEG
Abstract : In order to form a three-dimensional image that help identify and characterize the electric sources that generate the acquired signal in an electroencephalogram (EEG), the prior forward problem solution is required. Here, the goal is the search for a mathematical head model that explain the propagation of the signal generated within the brain until the scalp, where the sensors are located. To find a solution the mathematical representation of the signals generated by neuronal activity is analyzed and how they propagate through spherical models, multilayered spherical models and realistic models representing the head, and then how this representation is achieved by the Border Element Method BEM is analized. Finally, a comparison of this methodology with currently used methods is presented and a computer program with a graphical interface is developed which allows find a solution to the forward problem in EEG
Abstract : In order to form a three-dimensional image that help identify and characterize the electric sources that generate the acquired signal in an electroencephalogram (EEG), the prior forward problem solution is required. Here, the goal is the search for a mathematical head model that explain the propagation of the signal generated within the brain until the scalp, where the sensors are located. To find a solution the mathematical representation of the signals generated by neuronal activity is analyzed and how they propagate through spherical models, multilayered spherical models and realistic models representing the head, and then how this representation is achieved by the Border Element Method BEM is analized. Finally, a comparison of this methodology with currently used methods is presented and a computer program with a graphical interface is developed which allows find a solution to the forward problem in EEG