An Rs-fMRI Study On Functional Connectivity Of Hippocampal Subregion And Classification Of Cognitive Impairment In Obstrcutive Sleep Apnea

With the development of medical imaging technology,apnea caused obstruction syndrome clinical complications-cognitive dysfunction is more and more attention.OSA causes of cognitive impairment of patients with individuals,families and society are caused a heavy burden.This article focuses on OSA patients resting state of brain dysfunction and the relationship between the cognitive and emotional damage.This paper is divided into two parts,specific content as follows:Section One: A Resting-State Functional Magnetic Resonance Imaging Study on Abnormal Functional Connectivity of Hippocampal Subdivisions in Obstructive Sleep ApneaPurpose:The hippocampus is involved in various cognitive function,including memory,execution,and more.Hippocampal structural and functional abnormalities have been observed in patients with obstructive sleep apnea(OSA),but the functional connectivity(FC)patterns among hippocampal subdivisions in OSA patients remain unclear.The purpose of this study was to investigate the changes in FC between hippocampal subdivisions and their relationship with neurocognitive function in male patients with OSA.Methods : Resting-state fMRI were obtained from 46 male patients with untreated severe OSA and 46 male good sleepers.We used the seed point functional connectivity based on the previous hippocampal subregions method,the hippocampus was divided into anterior,middle,and posterior parts,and the differences in FC between hippocampal subdivisions and other brain regions were determined.Correlation analysis was used to explore the relationships between abnormal FC of hippocampal subdivisions and clinical characteristics in patients with OSA.Results : In the aspect of brain function connection,Our results revealed increased FC in the OSA group between the left anterior hippocampus and left middle temporal gyrus;between the left middle hippocampus and the left inferior frontal gyrus,right anterior central gyrus,and left anterior central gyrus;between the left posterior hippocampus and right middle frontal gyrus;between the right middle hippocampus and left inferior frontal gyrus;and between the right posterior hippocampus and left middle frontal gyrus.Correlation analysis: The FC values between the left anterior hippocampus and left middle temporal gyrus were negatively correlated with N1(r = 0.315,p = 0.033)and positively correlated with orientation(r = 0.343,p = 0.019).The FC value between the left middle hippocampus and left inferior frontal gyrus was negatively correlated with N1(r = 0.326,p = 0.027)and positively correlated with N3(r = 0.309,p =0.036).The FC value between the left middle hippocampus and right precentral gyrus was negatively correlated with the Mo CA scores(r = 0.429,p = 0.003),orientation(r = 0.394,p = 0.007),and total sleep time(r = 0.352,p = 0.016).The FC value between the left posterior hippocampus and right middle frontal gyrus was negatively correlated with total sleep time(r = 0.368,p = 0.012)and attention(r =0.344,p = 0.019)and positively correlated with an Sa O2 of <90%(r = 0.314,p =0.033).The FC value between the left posterior hippocampus and left middle frontal gyrus was positively correlated with an Sa O2 of <90%(r = 0.307,p = 0.038).Finally,the FC value between the right middle hippocampus and left inferior frontal gyrus was positively correlated with naming(r = 0.309,p = 0.036).Conclusion : These FC abnormalities predominantly manifested in the sensorimotor network,fronto-parietal network,and semantic/default mode network,which are closely related to the neurocognitive impairment observed in OSA patients.This study advances our understanding of the potential pathophysiological mechanism of cognitive impairment in OSA.Section Two: Classification of Obstructive Sleep Apnea with Cognitive Impairment Using Degree Centrality: A Machine Learning AnalysisPurpose:Patients with obstructive sleep apnea(OSA)have abnormal resting state brain function.However,it is not known whether there are differences in brain function between patients with OSA accompanied by mild cognitive impairment(MCI)and those without(no-mild cognitive impairment(n MCI).This research adopts the Degree Centrality(DC)combined with machine learning methods,to distinguish whether with MCI OSA patients.Methods : Ninety-nine OSA patients were recruited for rs-MRI scanning,including 51 MCI patients and 48 participants with no mild cognitive impairment.Based on the Automated Anatomical Labeling(AAL)brain atlas,the DC features of all participants were calculated and extracted.The most relevant DC features were screened out by deleting variables with high pin-correlation and minimum absolute contraction and performing selective operator lasso regression.Finally we use support vector machines(SVM)and random forests(RF),logistic regression(LR)method to establish classification model to distinguish whether OSA patients with MCI.Results:After feature selection program,we eventually got the 10 DC features,including the olfactory cortex,cingulate gyrus,globus pallidus,transverse temporal gyrus,cerebellum,etc.The support vector machine method had the best classification efficiency(AUC = 0.78),followed by random forest(AUC = 0.71)and logistic regression(AUC = 0.77).Conclusion:These abnormal DC characteristics mainly relates to the default network(DMN),basal ganglia area network,and cerebellar network.These findings demonstrate an effective machine learning approach for differentiating OSA patients with and without MCI and provide potential neuroimaging evidence for cognitive impairment caused by OSA.