The analysis of cognitive function of temporal lobe tumor patients based on the anatomical sub-networks

This study aims to apply graph theory to investigate the influence of anatomical networks and recognition function on temporal lobe tumor patients. Diffusion tensor imaging (DTI) was utilized to construct brain anatomical networks of patients and healthy controls, then the whole brain network was divided into three sub networks on the basis of fiber length, after that, the parameters of global efficiency and local efficiency were compared and analyzed, furthermore, a simple analysis of regeneration and disappearance of fiber bundles in temporal lobe brain tumor patients was given. In addition, the cognitive assessments of two groups were examined, afterwards, the correlation between cognitive parameters and nodal efficiency were analyzed, thus we can obtain the function distribution of patients. The analysis of parameters of the whole-brain networks reveals that there is no significant difference between healthy controls and patients, besides, the parameters of controls are better than patients. The analysis of sub-networks reveals that there were significant differences only in short-range sub-networks, the global efficiency of patients is superior to healthy controls significantly. In terms of local efficiency, the healthy controls are better than patients, cognitive function of patients is lower than the normal group. We can come to the conclusion that the regeneration of fiber bundles inside the brain region and among different brain regions with long distance make the reorganization of brain function, temporal lobe tumor patients are significantly lower than healthy controls in terms of attention, memory and learning ability. The frontotemporal-parietal brain regions are involved in social cognitive function, the alteration of information processing efficiency of these special brain regions can reflect the changes of cognition indirectly.