Research Of Network Functional Feature Of Tumor Patients Before And After Surgery Based On DTI

Brain tumor patients usually show different degrees and features of cognitive impairments,which is difficult to be explained by traditional clinical anatomy theory. On the one hand, due to local lesions secondarily affect the surrounding tissue, and thus affect the characteristics of the whole brain functional network. On the other hand, because of the brain plasticity and recombination of brain functional networks, some patients will emerge functional compensatory phenomenon soon before or after surgery, and the cognitive impairment will recover in a very short time. By seeking for the characteristic parameters to evaluate the cognitive impairment of preoperative and postoperative changes, which is helpful in making a surgery planning and improving the living quality of the patients after operation.This study was aimed to construct anatomical networks of brain tumor patients based on the tracking of white-matter fiber bundles, and apply graph theory approaches to investigate the brain functional properties pre-operation and post-operation. The main contents and innovations of this thesis are as follows:1. Diffusion tensor imaging(DTI) was utilized to construct the whole-brain and hemispheric-brain anatomical networks of healthy volunteers, and the topological properties of these networks were described quantitatively. The analysis results showed that there were significant gender differences in the brain regions involved in language, emotional, motor skills as well as orientation. At the same time, the differences between men and women on the function of the brain were consistent with the differences of the brain regions related to some functions, which indicated that the function researches based on the brain structure is feasible.2. The brain anatomical networks of tumor patients and healthy volunteers were constructed, and the topological properties and characteristic parameters of these networks were analyzed quantitatively. Compared to the networks of healthy volunteers, the brain structure of patients showed a significant increase in small-world feature and clustering coefficient caused by brain’s plasticity. The increase of connectivity density between paralimbic cortex and the default network of tumor patients manifested the existence of the brain’s compensatory and re-organization mechanism.3. The properties and parameters of whole-brain and hemispheric-brain anatomical networks of preoperative and postoperative group were analyzed, and found that the tumor itself and surgicaltreatment have changed the brain anatomical network, and the parameters of preoperative group are superior to postoperative group. Tumor patients(both preoperative group and postoperative group)showed stronger small-world properties than controls. These results can be applied in prediction and evaluation of surgery effect.