MRI-Based Network Analysis Method In Applied Research In Epilepsy

The human brain can efficiently isolate and integrate multiple sources of information, and create new knowledge, which is one of the most complex systems in the nature. Combined magnetic resonance imaging(MRI) technology and network analyses method is used to explore brain structural and functional connectivities, which makes progress of cognitive neuroscience, and then improve the research, diagnosis and treatment of psychological or neurological disorders. Epilepsy is a central nervous system disorder caused by neurons abnormal synchronous discharges. Repeated epileptic seizures seriously affect the patients’ quality of life, learning and work, and bring a heavy psychological and economic burden for their families. In the paper, electroencephalogram(EEG), functional magnetic resonance imaging(fMRI) and diffusion MRI, and graph theory method were used to study hemodynamic response function(HRF) in epilepsy patients, cognitive control network(CCN), the default mode network(DMN) and large-scale whole-brain anatomical network, then to discuss the pathophysiological mechanisms of epilepsy.First, the HRF describes the dynamic process of the blood oxygen level dependent(BOLD) signal evoked by a very short stimulus of unit intensity. Understanding HRF is a key issue for exploring the underlying dynamics of nervous system activation. Some researches suggested that the shape of HRF varied with different brain regions and different subjects, even at different time point in the same brain region of a subject. In this work, epilepsy discharges were adopted as markers of short stimuli, and then we used EEG-fMRI method to investigate HRF in temporal lobe epilepsy with interictal regional slow wave. Five patients with idiopathic temporal lobe epilepsy were enrolled in this work, the peak delay and amplitude of the estimated HRF was compared with Gamma, Glover and the SPM canonical HRF, and the results showed that the delay of the estimated HRF was different from either one of the three theoretical model, thus further study on both theoretical model and real data estimation was needed.Second, EEG and fMRI were widely utilized to investigate brain functional connectivity in various types of epilepsy patients, and much progress had been made. However, no research explored deeply how the number and amplitude of epileptic discharges influence on functional connectivity. In this work, various numbers and magnitude of epileptic discharges were simulated, and then BOLD signal caused by the discharges was superimposed to brain regions. Correlation coefficients between regions were calculated to assess the functional connectivity. The results showed that the number and amplitude of epileptic discharge had a significant impact on brain functional connectivity, so the future study of brain function network in epilepsy patients should consider the influence of epileptic discharges.Third, the altered functional connectivity in CCN has been shown in previous studies in patients with childhood absence epilepsy. In order to explore the structural connectivity in CCN, nine patients with childhood absence epilepsy(CAE) and 12 healthy controls were recruited. The fiber bundles among regions of CCN were tracked by diffusion tensor imaging(DTI) in each subject. The parameters(including count, length, weight, mean of fractional anisotropic(FA) and mean diffusivity(MD)) of the each fiber bundle were assessed between two groups by using two-sample T-test. The significantly decreased FA and increased MD in main fiber bundles were found in patient group. Furthermore, the mean FA value in fiber bundles between left prefrontal cortex and thalamus was negatively related to the epilepsy duration. These findings might provide potential structural evidence for the altered functional connectivity in CCN in absence epilepsy. It implied that the abnormalities of structural connections might result in cognitive control dysfunction in the patients with childhood absence epilepsy.Fourth, abnormal functional connectivity in DMN has been uncovered in EEG and fMRI studies, which suggests that the abnormality might be related to the cognitive mental impairment and unconsciousness during absence seizures. However, so far, little is known about the structural connectivity in DMN about CAE. In the present study, we hypothesized that the structural connectivity in DMN should be disrupted to respond to the altered brain function in CAE. To test the hypothesis, 11 patients with CAE and 12 age- and gender- matched healthy controls were recruited. We utilized diffusion tensor imaging tractography to map the anatomical structural connectivity of DMN. The fiber bundles among regions of DMN were built for each subject. Then, mean length, FA, MD and connection strength on fibers linking two brain regions were calculated. Further, these parameters were executed two-sample t-test between CAE group and health control group. Finally, we used Pearson’s correlation coefficient to evaluate the relationship between these parameters and epilepsy duration. Both CAE and healthy control groups showed similar structural connectivity patterns in DMN. Among these fiber bundles, three were identified in all subjects, with one linking posterior cingulate cortex/precuneus to medial prefrontal cortex, and another two linking posterior cingulate cortex/precuneus to bilateral medial temporal lobes. Furthermore, the significantly decreased FA and connection strength, and increased MD in fiber bundles linking posterior cingulated cortex/precuneus to medial prefrontal cortex, were found in patients compared with the cases in healthy controls. Predominantly, the FA values in fiber bundles linking posterior cingulated cortex/precuneus to medial prefrontal cortex were negatively correlated with the epilepsy duration. These findings might reflect the structural basis of the altered functional connectivity in DMN about absence epilepsy. Given that functional connectivity abnormality in our previous work, it implied that the abnormal fiber connectivity from posterior cingulated cortex/precuneus to medial prefrontal cortex plays an important role in absence epilepsy. This may open up new avenues to understand the pathophysiological mechanisms of absence epilepsy.Finally, although abnormal functional connectivity has been uncovered in CAE in previous EEG and fMRI studies, little is known regarding the structural connectivity in CAE. We hypothesized that the structural connectivity would be disrupted in response to the decreased brain function in CAE. Diffusion tensor imaging tractography was utilized to map the white matter(WM) structural network, composed of 90 cortical and sub-cortical regions, in the CAE and healthy controls. Graph theoretical methods were applied to investigate the alterations in the topological and nodal properties of the networks in these patients. Both the CAE and the controls showed small-world properties in their WM networks. However, the network connection strength, absolute clustering coefficient, and global/local efficiency were significantly decreased, but characteristic path length was significantly increased in the CAE compared with the controls. Significantly decreased WM connections, nodal properties, and impaired sub-networks were found in the sub-cortical structures, orbitofrontal area, and limbic cortex in the CAE. The present study preliminarily demonstrated the disrupted topological organization of WM networks in CAE. The decreased connectivity and efficiency in the orbitofrontal and sub-cortical regions might serve as anatomical evidence to support the functional abnormalities related to the epileptic discharges observed in CAE. Moreover, the orbitofrontal sub-network might play a key role in CAE. These findings open up new avenues toward the understanding of absence epilepsy.