Abstract In recent years, dynamic functional connectivity based on functional magnetic resonance imaging data has shown great potentials in studying psychiatric disorders. Traditional clustering-based dynamic functional connectivity analysis methods, such as K-means, are sensitive to the influence of class numbers, initial values, and noise, which may result in unreliable functional connectivity states (FCSs). This paper proposed a new dynamic functional connectivity analysis method based on ensemble clustering. First, K-means was performed with multiple different class numbers (k values) to generate diverse clusters. Then, the similarity between different clusters was mined based on Jaccard coefficient and random walk to construct a weighted graph reflecting the relationship between clusters. Finally, reliable meta-clusters were obtained by community detection on the weighted graph, and each functional connectivity window was grouped into different meta-clusters by voting, and its centroid was calculated as the functional connectivity states. Based on fMRI data from 105 healthy controls (HC) and 70 schizophrenia (SZ) patients, we comprehensively compared the performance of the proposed method with the commonly used K-means based method for dynamic functional network analysis. Compared with K-means method, the average inter class similarity of the proposed method decreased from 83.2% to 81.1% on FCS 2, and decreased from 76.8% to 73.5% on FCS 3. The Davies Bouldin index decreased from 6.74 to 6.44, and the Silhouette coefficacy index increased from 0.018 to 0.031. In terms of group differences, results obtained from our proposed method showed that the differences between the HC and SZ groups were mainly concentrated in FCS 2. In contrast, group differences resulted from K-means method were dispersed across FCS 2, FCS 3, and FCS 4, primarily due to subpar clustering metrics and high average inter-cluster similarity. In summary, the proposed method in this study was able to automatically determine the number of FCSs and exhibit better clustering quality and more reliable FCSs. Additionally, the method revealed more meaningful inter-group differences than k-means, indicating the potentials of exploring biomarkers for mental disorders.
Fang Songke,Du Yuhui. A New Ensemble Clustering Dynamic Functional Connectivity Analysis Method for ExploringBrain Connectivity Variation in Schizophrenia[J]. Chinese Journal of Biomedical Engineering, 2024, 43(3): 257-266.
[1] Buckner RL,DiNicola LM. The brain′s default network: updated anatomy, physiology and evolving insights [J]. Nature Reviews Neuroscience, 2019, 20(10): 593-608. [2] Du Yuhui,Fu Zening,Calhoun VD. Classification and Prediction of Brain Disorders Using Functional Connectivity: Promising but Challenging [J]. Frontiers in Neuroscience, 2018, 12: 525. [3] Allen EA,Damaraju E,Plis SM,et al. Tracking whole-brain connectivity dynamics in the resting state [J]. Cerebral Cortex, 2014, 24(3): 663-676. [4] Abrol A,Damaraju E,Miller RL,et al. Replicability of time-varying connectivity patterns in large resting state fMRI samples [J]. NeuroImage, 2017, 163: 160-176. [5] Lurie DJ,Kessler D,Bassett DS,et al. Questions and controversies in the study of time-varying functional connectivity in resting fMRI [J]. Network Neuroscience, 2020, 4(1): 30-69. [6] Du Yuhui,Fryer SL,Fu Zening,et al. Dynamic functional connectivity impairments in early schizophrenia and clinical high-risk for psychosis [J]. NeuroImage, 2018, 180: 632-645. [7] Sendi MSE,Pearlson GD,Mathalon DH,et al. Multiple overlapping dynamic patterns of the visual sensory network in schizophrenia [J]. Schizophrenia Research, 2021, 228: 103-111. [8] Li Weicheng,Wang Chengyu,Lan Xiaofeng,et al. Aberrant dynamic functional connectivity of posterior cingulate cortex subregions in major depressive disorder with suicidal ideation [J]. Frontiers in Neuroscience, 2022, 16: 937145. [9] Shappell HM,Duffy KA,Rosch KS,et al. Children with attention-deficit/hyperactivity disorder spend more time in hyperconnected network states and less time in segregated network states as revealed by dynamic connectivity analysis [J]. NeuroImage, 2021, 229: 117753. [10] 王英杰,沈辉. 脑静息功能连接自发波动的可变性分析 [J]. 中国生物医学工程学报, 2017, 36(1): 20-27. [11] Damaraju E,Allen EA,Belger A,et al. Dynamic functional connectivity analysis reveals transient states of dysconnectivity in schizophrenia [J]. NeuroImage: Clinical, 2014, 5: 298-308. [12] Yaesoubi M,Miller RL,Calhoun VD. Mutually temporally independent connectivity patterns: a new framework to study the dynamics of brain connectivity at rest with application to explain group difference based on gender [J]. NeuroImage, 2015, 107: 85-94. [13] Miller RL,Yaesoubi M,Turner JA,et al. Higher dimensional meta-state analysis reveals reduced resting fMRI connectivity dynamism in schizophrenia patients [J]. PLoS ONE, 2016, 11: e0149849. [14] Leonardi N,Richiardi J,Gschwind M,et al. Principal components of functional connectivity: a new approach to study dynamic brain connectivity during rest [J]. NeuroImage, 2013, 83: 937-950. [15] Fu Zening,Caprihan A,Chen Jiayu,et al. Altered static and dynamic functional network connectivity in Alzheimer's disease and subcortical ischemic vascular disease: shared and specific brain connectivity abnormalities [J]. Human Brain Mapping, 2019, 40(11): 3203-3221. [16] Salman MS,Du Yuhui,Calhoun VD. Identifying fMRI dynamic connectivity states using affinity propagation clustering method: application to schizophrenia [C] // International Conference on Acoustics, Speech and Signal Processing. New Orleans: IEEE, 2017: 904-908. [17] Vergara VM,Salman M,Abrol A,et al. Determining the number of states in dynamic functional connectivity using cluster validity indexes [J]. Journal of Neuroscience Methods, 2020, 337: 108651. [18] Huang Dong,Wang Changdong,Peng Hongxing,et al. Enhanced ensemble clustering via fast propagation of cluster-wise similarities [J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2021, 51(1): 508-520. [19] Yan Chaogan,Zang Yufeng. DPARSF: a MATLAB Toolbox for "pipeline" data analysis of resting-state fMRI [J]. Frontiers in Systems Neuroscience, 2010, 4: 13. [20] Fox MD,Snyder AZ,Vincent JL,et al. The human brain is intrinsically organized into dynamic, anticorrelated functional networks [J]. Proceedings of the National Academy of Sciences of the United States of America, 2005, 102(27): 9673-9678. [21] Tzourio-Mazoyer N,Landeau B,Papathanassiou D,et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain [J]. NeuroImage, 2002, 15(1): 273-289. [22] Leonardi N,Van De Ville D. On spurious and real fluctuations of dynamic functional connectivity during rest [J]. NeuroImage, 2015, 104: 430-436. [23] Du Yuhui,Pearlson GD,Lin Dongdong,et al. Identifying dynamic functional connectivity biomarkers using GIG-ICA: application to schizophrenia, schizoaffective disorder, and psychotic bipolar disorder [J]. Human Brain Mapping, 2017, 38: 2683-2708. [24] Friedman J,Hastie T,Tibshirani R. Sparse inverse covariance estimation with the graphical lasso [J]. Biostatistics, 2008, 9(3): 432-441. [25] Smith SM,Miller KL,Salimi-Khorshidi G,et al. Network modelling methods for FMRI [J]. NeuroImage, 2011, 54(2): 875-891. [26] Lai Darong,Lu Hongtao,Nardini C. Enhanced modularity-based community detection by random walk network preprocessing [J]. Physical review. E, Statistical, nonlinear, and soft matter physics, 2010, 81(6 Pt 2): 066118. [27] Khosla M,Jamison K,Ngo GH,et al. Machine learning in resting-state fMRI analysis [J]. Magnetic Resonance Imaging, 2019, 64: 101-121. [28] Spencer A,Goodfellow M. Using deep clustering to improve fMRI dynamic functional connectivity analysis [J]. NeuroImage, 2022, 257: 119288. [29] Xie Hua,Calhoun VD,Gonzalez-Castillo J,et al. Whole-brain connectivity dynamics reflect both task-specific and individual-specific modulation: A multitask study [J]. NeuroImage, 2018, 180: 495-504. [30] Rubinov M,Sporns O. Complex network measures of brain connectivity: uses and interpretations [J]. NeuroImage, 2010, 52(3): 1059-1069. [31] Newman MEJ. Modularity and community structure in networks [J]. Proceedings of the National Academy of Sciences, 2006, 103(23): 8577-8582. [32] Reichardt J,Bornholdt S. Statistical mechanics of community detection [J]. Physical Review E, 2006, 74(1): 016110. [33] Davies DL,Bouldin DW. A cluster separation measure [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1979, PAMI-1: 224-227. [34] Rousseeuw PJ. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis [J]. Journal of Computational and Applied Mathematics, 1987, 20: 53-65. [35] Iraji A,Faghiri A,Lewis N,et al. Tools of the trade: estimating time-varying connectivity patterns from fMRI data [J]. Social Cognitive and Affective Neuroscience, 2021, 16(8): 849-874. [36] Filippi M,Spinelli EG,Cividini C,et al. Resting state dynamic functional connectivity in neurodegenerative conditions: a review of magnetic resonance imaging findings [J]. Frontiers in Neuroscience, 2019, 13: 657. [37] Wang Min,Abrams ZB,Kornblau SM,et al. Thresher: determining the number of clusters while removing outliers [J]. BMC Bioinformatics, 2018, 19(1): 9. [38] Saha DK,Damaraju E,Rashid B,et al. A classification-based approach to estimate the number of resting functional magnetic resonance imaging dynamic functional connectivity states [J]. Brain Connectivity, 2021, 11(2): 132-145. [39] Mukherjee P,Whalley HC,McKirdy JW,et al. Altered amygdala connectivity within the social brain in schizophrenia [J]. Schizophrenia Bulletin, 2014, 40(1): 152-160. [40] Jeong B,Wible CG,Hashimoto R,et al. Functional and anatomical connectivity abnormalities in left inferior frontal gyrus in schizophrenia [J]. Human Brain Mapping, 2009, 30(12): 4138-4151. [41] Palaniyappan L,Liddle PF. Dissociable morphometric differences of the inferior parietal lobule in schizophrenia [J]. European Archives of Psychiatry and Clinical Neuroscience, 2012, 262(7): 579-587. [42] Kraguljac NV,White DM,Hadley N,et al. Aberrant hippocampal connectivity in unmedicated patients with schizophrenia and effects of antipsychotic medication: a longitudinal resting state functional MRI study [J]. Schizophrenia Bulletin, 2016, 42(4): 1046-1055. [43] Fitzsimmons J,Rosa P,Sydnor VJ,et al. Cingulum bundle abnormalities and risk for schizophrenia [J]. Schizophrenia Research, 2020, 215: 385-391.
