Research On The Mechanisms Of Nervous System Disease Based On Brain Functional Connectivity

The human brain has been so far believed to be the most complex and delicate system, which consists of vast numbers of neurons connected closely through electrochemical action. The brain is constructed by parallel, cooperating systems of anatomically linked areas. Currently, brain connectivity research has been widely used in the field of neuroscience, which can explore the mechanism and underlying organization of the brain from the perspective of system. Recently, the rapid development of neuroimaging technologies provide powerful tool to observe brain activities, while brain connectivity analysis based on complex network theory provides an effective technological means for brain functions research. Based on brain network analysis, this dissertation employs electroencephalogram (EEG) and function magnetic resonance imaging (fMRI) to explore synchronized firings of related motor neurons pool, functional connectivity between brain regions, hubs of functional brain network, brain network topological properties, and dynamics of brain network evolution after neurological damage.In order to explore the pathophysiological mechanism of nervous system disease, the static and dynamic approaches have been applied to investigate brain functional network at many levels. The main works and contributions of this dissertation can be summarized as follows:Patients always show locomotor and sensory disabilities after SCI. According to varying degrees of injuries, we designed two kinds of lower limbs rhythmic movement, that is walking and gait-like movement. During the performance of walking task, the EEG signals of related brain region were analysized by the time-frequency diagram of EEG and event-related synchronization/desynchronization (ERS/ERD), and the characteristic of synchronization of related motor neurons pool were also explored. We first found that the EEG amplitude decreased in all frequency bands, especially in lower frequency band. These findings are important to understand the mechanism of brain work during lower limbs rhythmic movement. Otherwise, we also found that the ERS amplitude of the related brain region increased in higher frequency band which indicated the increase in synchronization of the related motor neurons pool in the same frequency band. This finding, which is inconsistent with the previous results during voluntary foot movement, suggested that SCI patients need more motor cortex when performing rhythmic movement compared to voluntary movement. These findings may be a feature of neural activity after SCI, which could become a biological marker of early diagnosis for SCI.Combining the proposed method of analysis synchronization between brain regions based on time-frequency scale and inverse algorithm based on standardized low resolution brain electromagnetic tomography (sLORETA) to investigate the synchronization between related brain regions of patients with spinal cord injury (SCI) during lower limbs rhythmic movement. The results showed that the synchronization between related brain regions mainly decreased in lower-frequency, and increased in higher-frequency. Moreover, some related motor regions showed varying brain functions in different frequency bands. This phenomenon indicated that varying neural network existed in brain along with various oscillation frequencies of brain activity, and verified our proposed method of analysis synchronization between brain regions based on time-frequency scale was appropriate and necessary. In addition, we found that the thalamus showed different synchronization with related brain regions between patients and controls in passive movement condition but not in attempted/active movement condition. The result suggested that the modulation of the nervous system is different between passive rhythmic movement condition and attempted/active rhythmic movement condition. This new finding is a verification of previous study on primates during voluntary movement tasks. In order to explore the trend of brain activity after spinal cord injury, our study revealed the synchronization phenomenon between motor neurons during rhythmic lower limb movement from two levels. The research on the relationship between this trend and SCI may contribute to understand the pathological mechanism of SCI deeply, which has some clinical merit in the diagnosis, treatment and rehabilitation of SCI patients.GAMP algorithm based on brain anatomical structure was proposed to explore the modularity structure of functional brain network. Afterwards, participation coefficients in the modularity structure were used to identify hubs in functional brain network. Relative to the previous method which adopted degree to identify hubs, our proposed approach will not affect by the module size and can discriminate hubs more accurately. The proposed definition of Hub was applied to analyze the functional brain network after stroke. We found that the consistency of hubs in functional brain network of stroke was lower than that of normal, which suggested that the generalization of brain functions occurred in the brain of stroke. Moreover, the common hubs of patient group were focused on brain regions which have a close relationship with motor execution. This finding indicated that the patterns of information integration, information transfer and processing among brain regions changed after stroke. Our results revealed that the efficiency of information transfer and processing in brain network of patients suffered from stroke were decreased as compared to the able-bodied subjects. This study verified the proposed definition of Hub was more appropriate for functional brain network and investigated the patterns of brain collaboration and neural plasticity in brain functions after stroke from the perspective of information transfer and processing in brain network. Our research offered a way to reveal the internal pathological mechanism of stroke.By comparing the functional brain network topological properties between controls and strokes, a probabilistic model, in which the distance between brain regions were served as control factor, was first proposed to simulate the dynamic evolvement process of brain networks in the acute period of stroke based on a computational experiment platform, in an effort to address the changing mechanism of brain functions after brain injury. An independent fMRI dataset was also used to test the rationality of the proposed model. Simulate results shows that the evolution model can be effectively applied to simulate the evolution of stroke-affected brain networks in our study. More importantly, we found that the dynamic of functional brain network after stroke had some relationship with anatomical distance between brain regions. The anatomical distance between a pair of brain regions inhibited the occurrence of link between the pair of brain regions, and directly affected the efficiency of information transfer and processing in functional brain network. Our study, through the use of evolution model, may contribute to investigate a comprehensive dynamic process self-organization mechanism of brain networks and aid in further knowledge of the internal pathological mechanism of stroke. Furthermore, the proposed approaches can be extended to investigate pathological mechanism of other nervous system disease due to neurodegeneration or neurodevelopmental impairments.