Morphological Research Of Brain Structures Based On Magnetic Resonance Imaging

The human brain is the most complex system in the world. Early studies on the brain are mainly based on autopsy. Recently, with the development of brain imaging technology, mathematics, computer science, informatics and physics, brain morphological analysis has become one of the most investigated interests of neuroscience research.This study mainly focused on application of the existing methods to detect abnormalities in the cerebral cortex and on improving traditional brain imaging analysis methods based on MRI. Our main contributions are as follows:1. Using gray matter density and cortical thickness measurements, we investigated the structural abnormalities in major depressive disorder (MDD). We revealed that patients with MDD showed significant morphological abnormalities in prefrontal cortex, putamen, posterior cingulate and precuneus regions, compared with normal controls. These results indicate that MDD is a disease that involves multiple brain regions.2. Using cortical thickness measurement, structural networks were constructed for patients with schizophrenia and normal controls respectively. Network analysis showed abnormal connectivities between prefrontal cortex and the temporal poles. In addition, patients with schizophrenia showed higher local efficiency and lower global efficiency, compared with controls. Moreover, we investigated the difference of the hub nodes in the two networks. We found that the abnormal hub nodes in schizophrenia are mainly distributed in the default mode network, which is consistent with previous functional studies. These abnormalities in structural network might be the substrates of abnormal functional connectivities.3. We proposed an adaptive pixon represented segmentation algorithm for 3D magnetic resonance brain images. Different from traditional method, an adaptive mean shift algorithm was adopted to adaptively smooth the query image and create a pixon-based image representation. Then an expectation-maximization (EM) iterations composed of bias correction, a priori digital brain atlas information, and Markov random field (MRF) segmentation were processed. The adoption of bias correction and brain atlas made the current method more suitable for brain image segmentation than the previous pixon based segmentation algorithm. The proposed method was validated on both simulated normal brain images from BrainWeb and real brain images from the IBSR public dataset. Compared with some other popular MRI segmentation methods, the proposed method exhibited a higher degree of accuracy in segmenting both simulated and real 3D MRI brain data.