Research On Dynamic Functional Connectivity In Irritable Bowel Syndrome Based On Resting State Functional Magnetic Resonance Imaging

Irritable bowel syndrome(IBS)is one of the most prevalent functional gastrointestinal disorders.With unclear causes and no effective treatment available,IBS brings a great negative impact on the patients.It has been reported that the bidirectional interaction of the brain-gut axis is crucial in the pathological mechanism of IBS.The brain is the most advanced organ of human beings,leading all physiological activities in the body.The development of resting-state functional magnetic resonance imaging(rs-f MRI)provides an opportunity to explore the abnormal brain mechanisms behind the disease.However,most of the existing IBS-related rs-f MRI studies were based the hypothesis that the functional connectivity(FC)of the brain is stationary,which ignores the dynamic characteristics in temporal dimension.How to effectively mine the dynamic characteristics of brain network,capture the aberrant pathological patterns of IBS and help with clinical diagnosis and treatment has become an urgent problem to be solved.Consequently,the present research consists of two parts as follows:1.We explored the abnormality of dynamic functional brain network and the relationship between brain network dynamics and abnormalities of gut microbiome in patients with IBS.A total of 33 patients with IBS and 32 age-,gender-matched healthy controls(HCs)were included.Using group independent component analysis,the whole brain was divided into 47 independent components which belong to 7 resting-state networks(RSNs).Sliding window analysis was performed to construct dynamic functional brain networks,and K-means clustering was used to extract FC states that spontaneously recurred in the temporal dimension.Group differences of temporal properties and FC variability in each state and association with clinical characteristics were explored.In addition,the gut microbiome was analyzed by 16 S r RNA gene sequencing of stool samples,and the relationship with abnormal brain measures was also explored.The present study identified four recurring transient FC states.The patients with IBS had significantly higher fraction time and mean dwell time in State 4and lower number of transitions from State 3 to State 1.In addition,decreased FC variability in IBS group was observed in both State 1 and State 3.In State 1,the variability of FC that connects middle temporal gyrus and supplementary motor area was negatively correlated with visceral sensitivity index,and IBS-quality of life score was positively correlated with the FC variability between hippocampus and anterior cingulate and paracingulate gyri.Microbial analysis showed that IBS patients had a lower alpha-diversity index.Additionally,we identified 9 significantly differential abundances in microbial composition.We also found that IBS-related microbiota was significantly associated with decreased FC variability.Our results provide new insights into the dysconnectivity and disrupted brain-gut axis interactions in IBS from a dynamic perspective.2.We used the auto-regressive dynamic conditional correlation(AR-DCC)model to construct time-varying functional connectivity and employed K-means clustering to explore the dynamic characteristics of time-varying brain networks.AR-DCC model not only overcomes the shortcoming of sliding window analysis,but also makes the DFC can achieve the best balance of sensitivity and specificity.In DFC analysis,four recurring FC states were identified.We found that the patients with IBS had abnormal temporal properties in multiple states,and some of these abnormalities were significantly associated with depression scores.In addition,we observed decreased FC variability between multiple networks in State 4,which may be related to the abnormality of the “perception-transmission-processing-regulation” circuit of pain experience.We also found abnormal dynamic graph theory properties in the middle frontal gyrus and precuneus,some of which showed significant correlation with clinical characteristics such as Hamilton depression scale.Our results indicated that these brain regions may play a key role in the abnormal brain mechanism of IBS.Compared with sliding window analysis,dynamic functional connectivity constructed by AR-DCC model could reveal more IBS-related differences,and these differences could more effectively distinguish patients from controls.In summary,mining the dynamic information in the temporal dimension of f MRI data and exploring the dynamic brain network in IBS patients would help understand the neurophysiological mechanism of IBS and provide assistance for the clinical diagnosis and treatment.