Analysis of Fmri Data for Statistical Activation Mapping

摘要

Functional Magnetic Resonance Imaging (fMRI) is a non-invasive technique which shows great promise in providing neurological information on healthy subjects and clinical patients by mapping functional activation within the brain. The functional structure of human brain, the correlation between neural activities and the Blood-Oxygen-Level Dependent (BOLD) signal and fMRI experimental design techniques were studied in this work. The 2D and 3D anatomical high resolution and 3D + Time series (4D) low resolution functional images have been reconstructed and normalized. Spin echo-Echo Planar Imaging (EPI) technique has been used for obtaining fMRI data acquisition with spatially high resolution activation map overlaid on EPI image with the reduction of image acquisition time. The steps in the analysis of fMRI data were described and two statistical techniques, e.g., t-statistic and correlation analysis for data from single events have been proposed. The effect of Hemodynamic Response (HDR) on the neural activation has been observed for square, delayed square, visual cortex, averaged visual cortex and ideal HD response using correlation technique. It has been realized that ideal HDR generates the relevant activation map with correlation coefficient, r > 0.20. This fMRI analysis offers possibilities for improved neurological research and clinical neurosurgical applications.