Detección y reconocimiento de activación en datos 4D de resonancia magnética funcional por análisis multiresolución y multivariable

In this thesis, we analyze functional magnetic resonance imaging fMRI and discusses some existing techniques to identify brain activity. We propose a methodology based on multivariate techniques: Principal Component Analysis (PCA) and Independent Component Analysis (ICA) in conjunction with wavelet...

Descripción completa

Detalles Bibliográficos
Autor: EMMANUEL MORALES FLORES
Tipo de recurso: tesis de maestría
Estado:Versión aceptada para publicación
Fecha de publicación:2010
País:México
Institución:Instituto Nacional de Astrofísica, Óptica y Electrónica
Repositorio:Repositorio Institucional del INAOE
Idioma:español
OAI Identifier:oai:inaoe.repositorioinstitucional.mx:1009/542
Acceso en línea:http://inaoe.repositorioinstitucional.mx/jspui/handle/1009/542
Access Level:acceso abierto
Palabra clave:info:eu-repo/classification/Transformada de Wavelet/Wavelet transform
info:eu-repo/classification/Transformada discreta de Wavelet/Discrete wavelet transform
info:eu-repo/classification/Biomedical MRI/Biomedical MRI
info:eu-repo/classification/cti/1
info:eu-repo/classification/cti/22
info:eu-repo/classification/cti/2203
Descripción
Sumario:In this thesis, we analyze functional magnetic resonance imaging fMRI and discusses some existing techniques to identify brain activity. We propose a methodology based on multivariate techniques: Principal Component Analysis (PCA) and Independent Component Analysis (ICA) in conjunction with wavelet decomposition, to identify brain areas involved in mental processes and perform classification of images based on the hemodynamic response. For the classification task, three known classifiers are tested: Mahalanobis, k-NN and Support Vector Machines. We report results of the proposed methodology applied to fMRI data obtained from the public repository of the fMRI Data Center. A MATLAB-based virtual instrument, which incorporates the developed algorithms, as well as utilities for fMRI images study, was developed as part of this thesis, and it is also described in this report.