| ObjectiveFrom the perspective of traditional Chinese medicine(TCM),type 2 diabetes mellitus(T2DM)is divided into deficiency syndrome and excess syndrome to study whether there are differences in brain function between them.To further subdivide deficiency syndrome of T2 DM,the yin deficiency of T2 DM was studied.According to montreal cognitive assessment(MoCA),yin deficiency of T2 DM was subdivided into yin deficiency T2 DM with cognitive impairment(T2DM-CI)and yin deficiency T2 DM without cognitive impairment(T2DM-noCI).A method of dynamic high-order functional connectivity(dHOFC)network construction method was used to capture the changes of the brain functional network in T2 DM with deficiency and excess syndromes,T2DM-CI,as well as T2DM-noCI.We used machine learning based multivariable pattern recognition method to classify individual T2 DM with deficiency and excess syndromes,yin deficiency T2DM-CI and healthy controls(HC),as well as yin deficiency T2DM-noCI and HC.The contributing features in the classification model could constitute a homologous abnormality pattern for better understanding of the diabetic encephalopathy.Methods23 T2 DM with deficiency syndrome,27 T2 DM with excess syndrome,50 yin deficiency T2DM(including 23 yin deficiency T2DM-CI,27 yin deficiency T2DM-noCI),and 50 HC were enrolled in the First Affiliated Hospital of Guangzhou University of Chinese Medicine from October 2018 to December 2020.We divided the participants into three experiments.One experiment consisted of 23 T2 DM with deficiency syndrome and27 T2 DM with excess syndrome,one experiment constituted of 23 yin deficiency T2DM-CI and 25 HC,while one experiment constituted of 27 yin deficiency T2DM-noCI and 25 different HC.All participants were right-handed.In the last two experiments,participants completed MoCA,auditory verbal learning test(AVLT),trail making test(TMT),clocking drawing test(CDT),digit span test(DST),and other cognitive function scales.All data were collected using a 3.0-T GE magnetic resonance imaging(MRI)scanner.All participants underwent conventional MRI,resting-state functional magnetic resonance imaging(rs-f MRI)and high-resolution 3D-T1 WI whole-brain structural imaging.Network-based classification toolbox(Brain Net Class v1.1)was used to construct brain dHOFC networks for all participants and classify the participants in the three experiments.dHOFC features,absolute and relative involvements of each functional network were also calculated.Appropriate statistical methods were selected to compare the demographic and clinical data between T2 DM with deficiency syndrome and excess syndrome.The differences of demographic,clinical and neurocognitive function data of yin deficiency T2DM-CI and HC as well as yin deficiency T2DM-noCI and HC were compared.The correlations between dHOFC features and MoCA scores were analyzed for yin deficiency T2DM-CI and HC with age,gender,and education as covariables.The correlations between dHOFC features and auditory verbal learning test(immediate recall)(AVLT-IR)were analyzed for yin deficiency T2DM-noCI and HC with age,gender,and education as covariables.The results with P < 0.05 were considered as statistical significance.Results1.The dHOFC-based classification performance in three experiments classification were better than that of LOFC with respect to all the performance metrics.2.We identified one dHOFC node with their local clustering coefficients as discriminative feature in the classification of T2 DM with deficiency syndrome and excess syndrome.Local clustering coefficient of dHOFC node encompassed sensorimotor network(SMN),default mode network(DMN),and ventral attention network(VAN).It contained32 inter-network connections,among which 16 were inter-network connections of VAN,12 were inter-network connections of SMN and 4 were inter-network connections of DMN.Specifically,local clustering coefficient of dHOFC node showed highly synchronized temporal functional connections among bilateral rolandic operculum,bilateral superior frontal gyrus,medial orbital,bilateral rectus gyrus,bilateral insula,bilateral heschl gyrus,and bilateral superior temporal gyrus.According to the relative involvement and absolute involvement degrees,VAN was mostly involved,followed by SMN and DMN.3.Statistical comparisons between T2 DM with deficiency syndrome and excess syndrome showed no significant differences in the age,gender,education,cognition,and clinical characteristics.Statistical comparisons between yin deficiency T2DM-CI and HC as well as between yin deficiency T2DM-noCI and HC revealed no significant differences in age,gender,or education.Some clinical characteristics such as blood pressure showed significant differences both between T2DM-CI and HC and between T2DM-noCI and HC.Body mass index(BMI)of yin deficiency T2DM-noCI and HC was statistically significant.MoCA scores and AVLT-IR showed significant differences between yin deficiency T2DM-CI and their matched HC.For other neuropsychological characteristics,they showed no significant differences between yin deficiency T2DM-noCI and the matched HC.4.We identified two dHOFC nodes with their local clustering coefficients as discriminative features in the classification of yin deficiency T2DM-CI from HC.Specifically,the node 1 in dHOFC encompassed SMN,DMN,and VAN.It contained 50 links,22 out of which were the intra-network connections in the SMN.The involved connections were among the bilateral median cingulate and paracingulate gyri,bilateral rolandic operculum,bilateral supramarginal gyrus,bilateral superior temporal gyrus,right supplementary motor area,bilateral insula,right precentral gyrus,bilateral postcentral gyrus,bilateral heschl’s gyrus.In the node 1,SMN accounted for 75% relative involvement with an absolute involvement degree of 68,while the involvements of the other two high-level systems(DMN and VAN)were relatively small(12.5% relative involvement and their sum of absolute involvement degree was 16,respectively).The node 2 in dHOFC encompassed more functional systems than the node 1,including both primary(SMN and visual network(VN))and high-level(DMN,frontoparietal network(FPN),dorsal attention network(DAN),and VAN)systems.Among 48 links,46 were inter-network connections.The connections involved the bilateral paracentral lobule,bilateral calcarine fissure and surrounding cortex,bilateral cuneus,bilateral lingual gyrus,bilateral superior occipital gyrus,bilateral middle occipital gyrus,bilateral inferior occipital gyrus,bilateral fusiform gyrus,bilateral precuneus,bilateral inferior temporal gyrus,bilateral cerebellum crus1,bilateral cerebellum6,vermis45,and vermis6.According to the relative involvement,VN(visual network,53.85%)was mostly involved,followed by DAN(15.38%),SMN(11.54%),DMN(7.69%),VAN(7.69%),and FPN(3.85%).However,SMN had the highest absolute involvement degree(50),followed by VN(28),DAN(8),DMN(4),VAN(4),and FPN(2).5.We identified one dHOFC node with their local clustering coefficients as discriminative feature in the classification of yin deficiency T2DM-noCI from HC.The regions and connections of T2DM-noCI encompassed SMN and FPN.It contained 16 links,half of which were the intra-network connections in the FPN,while others were inter-network connections.The involved connections were among the bilateral superior frontal gyrus,orbital part,bilateral middle frontal gyrus,orbital part,bilateral cerebellum7,and bilateral cerebellum8.According to the relative involvement and absolute involvement degrees,FPN(frontoparietal network,75%,24)was mostly involved,while SMN was relatively small(25%,8).6.In the experiment of yin deficiency T2DM-CI and HC,the correlation between the local clustering coefficient of the dHOFC node 1 and MoCA scores was significant.The correlation between the local clustering coefficient of the dHOFC node 2 and MoCA scores was nearly significant.MoCA scores were positively correlated with education level.In the experiment of yin deficiency T2DM-noCI and HC,the correlation between the local clustering coefficient of the dHOFC node and AVLT-IR was significant.AVLT-IR was positively correlated with education level.Conclusion1.The dHOFC could well simulate the complex interactions between brain regions,measure the long-term functional interactions in a time-varying way,and capture their synchronization,so as to be more sensitive to the subtle changes of cognitive state.2.The dHOFC changes in VAN,DMN,and SMN in the experiment of T2 DM with deficiency syndrome and excess syndrome.3.The changes of dHOFC in SMN,VN,DMN,FPN,DAN,and VAN in yin deficiency T2DM-CI might be widespread impaired brain networks caused by cognitive impairment.4.Yin deficiency T2DM-noCI involved brain regions and connections of SMN and FPN,it might not be separated from the HC as successfully as yin deficiency T2DM-CI did due to largely intact cognitive functions and less affected brain networks.5.The local clustering coefficient of the dHOFC node 1 of yin deficiency T2DM-CI was positively correlated with MoCA scores while the local clustering coefficient of the dHOFC node of yin deficiency T2DM-noCI was negatively correlated with AVLT-IR.6.Education level was positively correlated with MoCA scores and AVLT-IR,which might indicate that yin deficiency T2 DM with higher education level had better cognitive function. |
PDF Full Text Request