A clustering-based method to detect functional connectivity differences.
ABSTRACT: Recently, resting-state functional magnetic resonance imaging (R-fMRI) has emerged as a powerful tool for investigating functional brain organization changes in a variety of neurological and psychiatric disorders. However, the current techniques may need further development to better define the reference brain networks for quantifying the functional connectivity differences between normal and diseased subject groups. In this study, we introduced a new clustering-based method that can clearly define the reference clusters. By employing group difference information to guide the clustering, the voxels within the reference clusters will have homogeneous functional connectivity changes above predefined levels. This method identified functional clusters that were significantly different between the amnestic mild cognitively impaired (aMCI) and age-matched cognitively normal (CN) subjects. The results indicated that the distribution of the clusters and their functionally disconnected regions resembled the altered memory network regions previously identified in task fMRI studies. In conclusion, the new clustering method provides an advanced approach for studying functional brain organization changes associated with brain diseases.
Project description:Functional magnetic resonance imaging (fMRI) is one of the most useful methods to generate functional connectivity networks of the brain. However, conventional network generation methods ignore dynamic changes of functional connectivity between brain regions. Previous studies proposed constructing high-order functional connectivity networks that consider the time-varying characteristics of functional connectivity, and a clustering method was performed to decrease computational cost. However, random selection of the initial clustering centers and the number of clusters negatively affected classification accuracy, and the network lost neurological interpretability. Here we propose a novel method that introduces the minimum spanning tree method to high-order functional connectivity networks. As an unbiased method, the minimum spanning tree simplifies high-order network structure while preserving its core framework. The dynamic characteristics of time series are not lost with this approach, and the neurological interpretation of the network is guaranteed. Simultaneously, we propose a multi-parameter optimization framework that involves extracting discriminative features from the minimum spanning tree high-order functional connectivity networks. Compared with the conventional methods, our resting-state fMRI classification method based on minimum spanning tree high-order functional connectivity networks greatly improved the diagnostic accuracy for Alzheimer's disease.
Project description:Various machine-learning classification techniques have been employed previously to classify brain states in healthy and disease populations using functional magnetic resonance imaging (fMRI). These methods generally use supervised classifiers that are sensitive to outliers and require labeling of training data to generate a predictive model. Density-based clustering, which overcomes these issues, is a popular unsupervised learning approach whose utility for high-dimensional neuroimaging data has not been previously evaluated. Its advantages include insensitivity to outliers and ability to work with unlabeled data. Unlike the popular k-means clustering, the number of clusters need not be specified. In this study, we compare the performance of two popular density-based clustering methods, DBSCAN and OPTICS, in accurately identifying individuals with three stages of cognitive impairment, including Alzheimer's disease. We used static and dynamic functional connectivity features for clustering, which captures the strength and temporal variation of brain connectivity respectively. To assess the robustness of clustering to noise/outliers, we propose a novel method called recursive-clustering using additive-noise (R-CLAN). Results demonstrated that both clustering algorithms were effective, although OPTICS with dynamic connectivity features outperformed in terms of cluster purity (95.46%) and robustness to noise/outliers. This study demonstrates that density-based clustering can accurately and robustly identify diagnostic classes in an unsupervised way using brain connectivity.
Project description:We employed a multi-scale clustering methodology known as "data cloud geometry" to extract functional connectivity patterns derived from functional magnetic resonance imaging (fMRI) protocol. The method was applied to correlation matrices of 106 regions of interest (ROIs) in 29 individuals with autism spectrum disorders (ASD), and 29 individuals with typical development (TD) while they completed a cognitive control task. Connectivity clustering geometry was examined at both "fine" and "coarse" scales. At the coarse scale, the connectivity clustering geometry produced 10 valid clusters with a coherent relationship to neural anatomy. A supervised learning algorithm employed fine scale information about clustering motif configurations and prevalence, and coarse scale information about intra- and inter-regional connectivity; the algorithm correctly classified ASD and TD participants with sensitivity of 82.8% and specificity of 82.8%. Most of the predictive power of the logistic regression model resided at the level of the fine-scale clustering geometry, suggesting that cellular versus systems level disturbances are more prominent in individuals with ASD. This article provides validation for this multi-scale geometric approach to extracting brain functional connectivity pattern information and for its use in classification of ASD.
Project description:Functional magnetic resonance imaging (fMRI) studies have shown that anatomically distinct brain regions are functionally connected during the resting state. Basic topological properties in the brain functional connectivity (BFC) map have highlighted the BFC's small-world topology. Modularity, a more advanced topological property, has been hypothesized to be evolutionary advantageous, contributing to adaptive aspects of anatomical and functional brain connectivity. However, current definitions of modularity for complex networks focus on nonoverlapping clusters, and are seriously limited by disregarding inclusive relationships. Therefore, BFC's modularity has been mainly qualitatively investigated. Here, we introduce a new definition of modularity, based on a recently improved clustering measurement, which overcomes limitations of previous definitions, and apply it to the study of BFC in resting state fMRI of 53 healthy subjects. Results show hierarchical functional modularity in the brain.
Project description:Background: Functional connectivity detected by resting-state functional MRI (R-fMRI) helps to discover the subtle changes in brain activities. Patients with end-stage renal disease (ESRD) on hemodialysis (HD) have impaired brain networks. However, the functional changes of brain networks in patients with ESRD undergoing peritoneal dialysis (PD) have not been fully delineated, especially among those with preserved cognitive function. Therefore, it is worth knowing about the brain functional connectivity in patients with PD by using R-fMRI. Methods: This case-control study prospectively enrolled 19 patients with ESRD receiving PD and 24 age- and sex- matched controls. All participants without a history of cognitive decline received mini-mental status examination (MMSE) and brain 3-T R-fMRI. Comprehensive R-fMRI analyses included graph analysis for connectivity and seed-based correlation networks. Independent t-tests were used for comparing the graph parameters and connectivity networks between patients with PD and controls. Results: All subjects were cognitively intact (MMSE > 24). Whole-brain connectivity by graph analysis revealed significant differences between the two groups with decreased global efficiency (Eglob, p < 0.05), increased betweenness centrality (BC) (p < 0.01), and increased characteristic path length (L, p < 0.01) in patients with PD. The functional connections of the default-mode network (DMN), sensorimotor network (SMN), salience network (SN), and hippocampal network (HN) were impaired in patients with PD. Meanwhile, in DMN and SN, elevated connectivity was observed in certain brain regions of patients with PD. Conclusion: Patients with ESRD receiving PD had specific disruptions in functional connectivity. In graph analysis, Eglob, BC, and L showed significant connectivity changes compared to the controls. DMN and SN had the most prominent alterations among the observed networks, with both decreased and increased connectivity regions. Our study confirmed that significant changes in cerebral connections existed in cognitively intact patients with PD.
Project description:The amygdala plays an important role in emotional functions and its dysfunction is considered to be associated with multiple psychiatric disorders in humans. Cytoarchitectonic mapping has demonstrated that the human amygdala complex comprises several subregions. However, it's difficult to delineate boundaries of these subregions in vivo even if using state of the art high resolution structural MRI. Previous attempts to parcellate this small structure using unsupervised clustering methods based on resting state fMRI data suffered from the low spatial resolution of typical fMRI data, and it remains challenging for the unsupervised methods to define subregions of the amygdala in vivo. In this study, we developed a novel brain parcellation method to segment the human amygdala into spatially contiguous subregions based on 7T high resolution fMRI data. The parcellation was implemented using a semi-supervised spectral clustering (SSC) algorithm at an individual subject level. Under guidance of prior information derived from the Julich cytoarchitectonic atlas, our method clustered voxels of the amygdala into subregions according to similarity measures of their functional signals. As a result, three distinct amygdala subregions can be obtained in each hemisphere for every individual subject. Compared with the cytoarchitectonic atlas, our method achieved better performance in terms of subregional functional homogeneity. Validation experiments have also demonstrated that the amygdala subregions obtained by our method have distinctive, lateralized functional connectivity (FC) patterns. Our study has demonstrated that the semi-supervised brain parcellation method is a powerful tool for exploring amygdala subregional functions.
Project description:Important functional connections within the default mode network (DMN) are disrupted in Alzheimer's disease (AD), likely from amyloid-beta (Abeta) plaque-associated neuronal toxicity. Here, we sought to determine if pathological effects of Abeta amyloid plaques could be seen, even in the absence of a task, by examining functional connectivity in cognitively normal participants with and without preclinical amyloid deposition.Participants with Alzheimer's disease (AD) (n = 35) were compared with 68 cognitively normal participants who were further subdivided by positron emission tomography (PET) Pittsburgh Compound-B (PIB) imaging into those without evidence of brain amyloid (PIB-) and those with brain amyloid (PIB+) deposition.Resting state functional magnetic resonance imaging (fMRI) demonstrated that, compared with the PIB- group, the PIB+ group differed significantly in functional connectivity of the precuneus to hippocampus, parahippocampus, anterior cingulate, dorsal cingulate, gyrus rectus, superior precuneus, and visual cortex. These differences were in the same regions and in the same direction as differences found in the AD group.Thus, before any manifestations of cognitive or behavioral changes, there were differences in resting state connectivity in cognitively normal subjects with brain amyloid deposition, suggesting that early manifestation of Abeta toxicity can be detected using resting state fMRI.
Project description:The caudate nucleus plays important roles in cognition and affect. Depending on associated connectivity and function, the caudate can be further divided into dorsal and ventral aspects. Dorsal caudate, highly connected to dorsolateral prefrontal cortex (DLPFC), is implicated in executive function and working memory; ventral caudate, more interconnected with the limbic system, is implicated in affective functions such as pain processing. Clinically, certain brain disorders are known to differentially impact dorsal and ventral caudate. Thus, precise parcellation of caudate has both basic and clinical neuroscience significance. In young adults, past work has combined resting-state fMRI functional connectivity with clustering algorithms to define dorsal and ventral caudate. Whether the same approach is effective in older adults and how to validate the parcellation results have not been considered. We addressed these problems by obtaining resting-state fMRI data from 56 older non-demented adults (age: 69.07 ± 5.92 years and MOCA: 25.71 ± 2.46) along with a battery of cognitive and clinical assessments. Connectivity from each voxel of caudate to the rest of the brain was computed using cross correlation. Applying the K-means clustering algorithm to the connectivity patterns with K = 2 yielded two substructures within caudate, which agree well with previously reported dorsal and ventral divisions of caudate. Furthermore, dorsal-caudate-seeded functional connectivity was shown to be more strongly associated with working memory and fluid reasoning composite scores, whereas ventral-caudate-seeded functional connectivity more strongly associated with pain and fatigue severity. These results demonstrate that dorsal and ventral caudate can be reliably identified by combining resting-state fMRI and clustering algorithms in older adults.
Project description:Task-free functional magnetic resonance imaging (TF-fMRI) has great potential for advancing the understanding and treatment of neurologic illness. However, as with all measures of neural activity, variability is a hallmark of intrinsic connectivity networks (ICNs) identified by TF-fMRI. This variability has hampered efforts to define a robust metric of connectivity suitable as a biomarker for neurologic illness. We hypothesized that some of this variability rather than representing noise in the measurement process, is related to a fundamental feature of connectivity within ICNs, which is their non-stationary nature. To test this hypothesis, we used a large (n = 892) population-based sample of older subjects to construct a well characterized atlas of 68 functional regions, which were categorized based on independent component analysis network of origin, anatomical locations, and a functional meta-analysis. These regions were then used to construct dynamic graphical representations of brain connectivity within a sliding time window for each subject. This allowed us to demonstrate the non-stationary nature of the brain's modular organization and assign each region to a "meta-modular" group. Using this grouping, we then compared dwell time in strong sub-network configurations of the default mode network (DMN) between 28 subjects with Alzheimer's dementia and 56 cognitively normal elderly subjects matched 1:2 on age, gender, and education. We found that differences in connectivity we and others have previously observed in Alzheimer's disease can be explained by differences in dwell time in DMN sub-network configurations, rather than steady state connectivity magnitude. DMN dwell time in specific modular configurations may also underlie the TF-fMRI findings that have been described in mild cognitive impairment and cognitively normal subjects who are at risk for Alzheimer's dementia.