A novel graph neural network method for Alzheimer's disease classification.

Alzheimer's disease (AD) is a chronic neurodegenerative disease. Early diagnosis are very important to timely treatment and delay the progression of the disease. In the past decade, many computer-aided diagnostic (CAD) algorithms have been proposed for classification of AD. In this paper, we propose a novel graph neural network method, termed Brain Graph Attention Network (BGAN) for classification of AD. First, brain graph data are used to model classification of AD as a graph classification task. Second, a local attention layer is designed to capture and aggregate messages of interactions between node neighbors. And, a global attention layer is introduced to obtain the contribution of each node for graph representation. Finally, using the BGAN to implement AD classification. We train and test on two open public databases for AD classification task. Compared to classic models, the experimental results show that our model is superior to six classic models. We demonstrate that BGAN is a powerful classification model for AD. In addition, our model can provide an analysis of brain regions in order to judge which regions are related to AD disease and which regions are related to AD progression.

Template-based graph registration network for boosting the diagnosis of brain connectivity disorders.

Brain graphs are powerful representations to explore the biological roadmaps of the human brain in its healthy and disordered states. Recently, a few graph neural networks (GNNs) have been designed for brain connectivity synthesis and diagnosis. However, such non-Euclidean deep learning architectures might fail to capture the neural interactions between different brain regions as they are trained without guidance from any prior biological template-i.e., template-free learning. Here we assume that using a population-driven brain connectional template (CBT) that captures well the connectivity patterns fingerprinting a given brain state (e.g., healthy) can better guide the GNN training in its downstream learning task such as classification or regression. To this aim we design a plug-in graph registration network (GRN) that can be coupled with any conventional graph neural network (GNN) so as to boost its learning accuracy and generalizability to unseen samples. Our GRN is a graph generative adversarial network (gGAN), which registers brain graphs to a prior CBT. Next, the registered brain graphs are used to train typical GNN models. Our GRN can be integrated into any GNN working in an end-to-end fashion to boost its prediction accuracy. Our experiments showed that GRN remarkably boosted the prediction accuracy of four conventional GNN models across four neurological datasets.

Path analysis: A method to estimate altered pathways in time-varying graphs of neuroimaging data.

Graph-theoretical methods have been widely used to study human brain networks in psychiatric disorders. However, the focus has primarily been on global graphic metrics with little attention to the information contained in paths connecting brain regions. Details of disruption of these paths may be highly informative for understanding disease mechanisms. To detect the absence or addition of multistep paths in the patient group, we provide an algorithm estimating edges that contribute to these paths with reference to the control group. We next examine where pairs of nodes were connected through paths in both groups by using a covariance decomposition method. We apply our method to study resting-state fMRI data in schizophrenia versus controls. Results show several disconnectors in schizophrenia within and between functional domains, particularly within the default mode and cognitive control networks. Additionally, we identify new edges generating additional paths. Moreover, although paths exist in both groups, these paths take unique trajectories and have a significant contribution to the decomposition. The proposed path analysis provides a way to characterize individuals by evaluating changes in paths, rather than just focusing on the pairwise relationships. Our results show promise for identifying path-based metrics in neuroimaging data.

Joint selection of brain network nodes and edges for MCI identification.

BACKGROUND AND OBJECTIVE: Functional brain graph (FBG), by describing the interactions between different brain regions, provides an effective representation of fMRI data for identifying mild cognitive impairment (MCI), an early stage of Alzheimer's Disease (AD). Prior to the identification task, selecting features from the estimated FBG is a necessary step for reducing computational cost, alleviating the risk of overfitting, and finding potential biomarkers of brain diseases. In practice, either node-based features (e.g., local clustering coefficients) or edge-based features (e.g., adjacency weights) are generally considered in current studies. Despite their popularity, these schemes can only capture one granularity (node or edge) of information in the FBG, which might be insufficient for the classification task and the interpretation of the classification result. METHODS: To address this issue, in this paper, we propose to jointly select nodes and edges from the estimated FBGs. Specifically, we first assign the edges to different node groups. Then, sparse group least absolute shrinkage and selection operator (sgLASSO) is used to select groups (nodes) and edges in the groups towards a better classification performance. Such a technique enables us to simultaneously locate discriminative brain regions, as well as connections between these brain regions, making the classification results more interpretable. RESULTS: Experimental results show that the proposed method achieves better classification performance than state-of-the-art methods. Moreover, by exploring brain network "features" that contributed most to MCI identification, we discover potential biomarkers for MCI diagnosis. CONCLUSION: A novel method for jointly selecting nodes and edges from the estimated functional brain graphs (FBGs) is proposed.

Brain graph synthesis by dual adversarial domain alignment and target graph prediction from a source graph.

Developing predictive intelligence in neuroscience for learning how to generate multimodal medical data from a single modality can improve neurological disorder diagnosis with minimal data acquisition resources. Existing deep learning frameworks are mainly tailored for images, which might fail in handling geometric data (e.g., brain graphs). Specifically, predicting a target brain graph from a single source brain graph remains largely unexplored. Solving such problem is generally challenged with domain fracturecaused by the difference in distribution between source and target domains. Besides, solving the prediction and domain fracture independently might not be optimal for both tasks. To address these challenges, we unprecedentedly propose a Learning-guided Graph Dual Adversarial Domain Alignment (LG-DADA) framework for predicting a target brain graph from a source brain graph. The proposed LG-DADA is grounded in three fundamental contributions: (1) a source data pre-clustering step using manifold learning to firstly handle source data heterogeneity and secondly circumvent mode collapse in generative adversarial learning, (2) a domain alignment of source domain to the target domain by adversarially learning their latent representations, and (3) a dual adversarial regularization that jointly learns a source embedding of training and testing brain graphs using two discriminators and predict the training target graphs. Results on morphological brain graphs synthesis showed that our method produces better prediction accuracy and visual quality as compared to other graph synthesis methods.

A diagnostic unified classification model for classifying multi-sized and multi-modal brain graphs using graph alignment.

BACKGROUND: Presence of multimodal brain graphs derived from different neuroimaging modalities is inarguably one of the most critical challenges in building unified classification models that can be trained and tested on any brain graph regardless of its size and the modality it was derived from. EXISTING METHODS: One solution is to learn a model for each modality independently, which is cumbersome and becomes more time-consuming as the number of modalities increases. Another traditional solution is to build a model inputting multimodal brain graphs for the target prediction task; however, this is only applicable to datasets where all samples have joint neuro-modalities. NEW METHOD: In this paper, we propose to build a unified brain graph classification model trained on unpaired multimodal brain graphs, which can classify any brain graph of any size. This is enabled by incorporating a graph alignment step where all multi-modal graphs of different sizes and heterogeneous distributions are mapped to a common template graph. Next, we design a graph alignment strategy to the target fixed-size template and further apply linear discriminant analysis (LDA) to the aligned graphs as a supervised dimensionality reduction technique for the target classification task. RESULTS: We tested our method on unpaired autistic and healthy brain connectomes derived from functional and morphological MRI datasets (two modalities). CONCLUSION: Our results showed that our unified model method not only has great promise in solving such a challenging problem but achieves comparable performance to models trained on each modality independently.

Structural network topology in limbic encephalitis is associated with amygdala enlargement, memory performance and serostatus.

Limbic encephalitis (LE) forms a spectrum of autoimmune diseases involving temporal lobe epilepsy and memory impairment. Imaging features of LE are known to depend on the associated antibody and to occur on the brain network level. However, first studies investigating brain networks in LE have either focused on one distinct antibody subgroup or on distinct anatomical regions. In this study, brain graphs of 17 LE patients with autoantibodies against glutamic acid decarboxylase 65 (GAD-LE), four LE patients with autoantibodies against leucine-rich glioma-inactivated 1, five LE patients with autoantibodies against contactin-associated protein-like 2, 26 age- and gender-matched healthy control subjects, and 20 epilepsy control patients with hippocampal sclerosis were constructed based on T1-weighted structural magnetic resonance imaging scans and diffusion tensor imaging. GAD-LE showed significantly altered global network topology in terms of integration and segregation as compared to healthy controls and patients with hippocampal sclerosis (P < .01, analysis of variance with Tukey-Kramer post hoc tests). Linear regression linked global network measures with amygdala volume and verbal memory performance (P < .05). Alterations of local network topology show serotype dependence in hippocampus, amygdala, insula, and various cortical regions. Our findings reveal serotype-dependent patterns of structural connectivity and prove the relevance of in silico network measures on clinical grounds.

Brain graph super-resolution for boosting neurological disorder diagnosis using unsupervised multi-topology connectional brain template learning.

Existing graph analysis techniques generally focus on decreasing the dimensionality of graph data (i.e., removing nodes, edges, or both) in diverse predictive learning tasks in pattern recognition, computer vision, and medical data analysis such as dimensionality reduction, filtering and embedding techniques. However, graph super-resolution is strikingly lacking, i.e., the concept of super-resolving low-resolution (LR) graphs with nr nodes into high-resolution graphs (HR) with [Formula: see text] nodes. Particularly, learning how to automatically generate HR brain connectomes, without resorting to the computationally expensive MRI processing steps such as image registration and parcellation, remains unexplored. To fill this gap, we propose the first technique to super-resolve undirected fully connected graphs with application to brain connectomes. First, we root our brain graph super-resolution (BGSR) framework in learning how to estimate a centered LR population-based brain graph representation, coined as connectional brain template (CBT), acting as a proxy in the target BGSR task. Specifically, we hypothesize that the estimation of a well-representative and centered CBT would help better capture the individuality of each LR brain graph via its residual distance from the population-based CBT. This will eventually allow an accurate identification of the most similar individual graphs to a new testing graph in the LR domain for the target prediction task. Second, we leverage the estimated LR CBT (i.e., population mean) to derive residual LR brain graphs, capturing the deviation of all subjects from the estimated CBT. Third, we learn multi-topology LR graph manifolds using different graph topological measurements (e.g., degree, closeness, betweenness) by estimating residual LR similarity matrices modeling the relationship between pairs of residual LR graphs. These are then fused so we can effectively identify for each testing LR subject its most K similar training LR graphs. Last, the missing testing HR graph is predicted by averaging the HR graphs of the K selected training subjects. Predicted HR from LR functional brain graphs boosted classification results for autistic subjects by 16.48% compared with LR functional graphs.

Neural Reconstruction Integrity: A Metric for Assessing the Connectivity Accuracy of Reconstructed Neural Networks.

Neuroscientists are actively pursuing high-precision maps, or graphs consisting of networks of neurons and connecting synapses in mammalian and non-mammalian brains. Such graphs, when coupled with physiological and behavioral data, are likely to facilitate greater understanding of how circuits in these networks give rise to complex information processing capabilities. Given that the automated or semi-automated methods required to achieve the acquisition of these graphs are still evolving, we developed a metric for measuring the performance of such methods by comparing their output with those generated by human annotators ("ground truth" data). Whereas classic metrics for comparing annotated neural tissue reconstructions generally do so at the voxel level, the metric proposed here measures the "integrity" of neurons based on the degree to which a collection of synaptic terminals belonging to a single neuron of the reconstruction can be matched to those of a single neuron in the ground truth data. The metric is largely insensitive to small errors in segmentation and more directly measures accuracy of the generated brain graph. It is our hope that use of the metric will facilitate the broader community's efforts to improve upon existing methods for acquiring brain graphs. Herein we describe the metric in detail, provide demonstrative examples of the intuitive scores it generates, and apply it to a synthesized neural network with simulated reconstruction errors. Demonstration code is available.

Application of Graph Theory to Assess Static and Dynamic Brain Connectivity: Approaches for Building Brain Graphs.

Human brain connectivity is complex. Graph theory based analysis has become a powerful and popular approach for analyzing brain imaging data, largely because of its potential to quantitatively illuminate the networks, the static architecture in structure and function, the organization of dynamic behavior over time, and disease related brain changes. The first step in creating brain graphs is to define the nodes and edges connecting them. We review a number of approaches for defining brain nodes including fixed versus data-driven nodes. Expanding the narrow view of most studies which focus on static and/or single modality brain connectivity, we also survey advanced approaches and their performances in building dynamic and multi-modal brain graphs. We show results from both simulated and real data from healthy controls and patients with mental illnesse. We outline the advantages and challenges of these various techniques. By summarizing and inspecting recent studies which analyzed brain imaging data based on graph theory, this article provides a guide for developing new powerful tools to explore complex brain networks.

Comparing brain graphs in which nodes are regions of interest or independent components: A simulation study.

BACKGROUND: A key challenge in building a brain graph using fMRI data is how to define the nodes. Spatial brain components estimated by independent components analysis (ICA) and regions of interest (ROIs) determined by brain atlas are two popular methods to define nodes in brain graphs. It is difficult to evaluate which method is better in real fMRI data. NEW METHOD: Here we perform a simulation study and evaluate the accuracies of a few graph metrics in graphs with nodes of ICA components, ROIs, or modified ROIs in four simulation scenarios. RESULTS: Graph measures with ICA nodes are more accurate than graphs with ROI nodes in all cases. Graph measures with modified ROI nodes are modulated by artifacts. The correlations of graph metrics across subjects between graphs with ICA nodes and ground truth are higher than the correlations between graphs with ROI nodes and ground truth in scenarios with large overlapped spatial sources. Moreover, moving the location of ROIs would largely decrease the correlations in all scenarios. COMPARISON WITH EXISTING METHOD (S): Evaluating graphs with different nodes is promising in simulated data rather than real data because different scenarios can be simulated and measures of different graphs can be compared with a known ground truth. CONCLUSION: Since ROIs defined using brain atlas may not correspond well to real functional boundaries, overall findings of this work suggest that it is more appropriate to define nodes using data-driven ICA than ROI approaches in real fMRI data.

Covariate-assisted spectral clustering.

Biological and social systems consist of myriad interacting units. The interactions can be represented in the form of a graph or network. Measurements of these graphs can reveal the underlying structure of these interactions, which provides insight into the systems that generated the graphs. Moreover, in applications such as connectomics, social networks, and genomics, graph data are accompanied by contextualizing measures on each node. We utilize these node covariates to help uncover latent communities in a graph, using a modification of spectral clustering. Statistical guarantees are provided under a joint mixture model that we call the node-contextualized stochastic blockmodel, including a bound on the misclustering rate. The bound is used to derive conditions for achieving perfect clustering. For most simulated cases, covariate-assisted spectral clustering yields results superior both to regularized spectral clustering without node covariates and to an adaptation of canonical correlation analysis. We apply our clustering method to large brain graphs derived from diffusion MRI data, using the node locations or neurological region membership as covariates. In both cases, covariate-assisted spectral clustering yields clusters that are easier to interpret neurologically.

Building an EEG-fMRI Multi-Modal Brain Graph: A Concurrent EEG-fMRI Study.

The topological architecture of brain connectivity has been well-characterized by graph theory based analysis. However, previous studies have primarily built brain graphs based on a single modality of brain imaging data. Here we develop a framework to construct multi-modal brain graphs using concurrent EEG-fMRI data which are simultaneously collected during eyes open (EO) and eyes closed (EC) resting states. FMRI data are decomposed into independent components with associated time courses by group independent component analysis (ICA). EEG time series are segmented, and then spectral power time courses are computed and averaged within 5 frequency bands (delta; theta; alpha; beta; low gamma). EEG-fMRI brain graphs, with EEG electrodes and fMRI brain components serving as nodes, are built by computing correlations within and between fMRI ICA time courses and EEG spectral power time courses. Dynamic EEG-fMRI graphs are built using a sliding window method, versus static ones treating the entire time course as stationary. In global level, static graph measures and properties of dynamic graph measures are different across frequency bands and are mainly showing higher values in eyes closed than eyes open. Nodal level graph measures of a few brain components are also showing higher values during eyes closed in specific frequency bands. Overall, these findings incorporate fMRI spatial localization and EEG frequency information which could not be obtained by examining only one modality. This work provides a new approach to examine EEG-fMRI associations within a graph theoretic framework with potential application to many topics.

Assessing dynamic brain graphs of time-varying connectivity in fMRI data: application to healthy controls and patients with schizophrenia.

Graph theory-based analysis has been widely employed in brain imaging studies, and altered topological properties of brain connectivity have emerged as important features of mental diseases such as schizophrenia. However, most previous studies have focused on graph metrics of stationary brain graphs, ignoring that brain connectivity exhibits fluctuations over time. Here we develop a new framework for accessing dynamic graph properties of time-varying functional brain connectivity in resting-state fMRI data and apply it to healthy controls (HCs) and patients with schizophrenia (SZs). Specifically, nodes of brain graphs are defined by intrinsic connectivity networks (ICNs) identified by group independent component analysis (ICA). Dynamic graph metrics of the time-varying brain connectivity estimated by the correlation of sliding time-windowed ICA time courses of ICNs are calculated. First- and second-level connectivity states are detected based on the correlation of nodal connectivity strength between time-varying brain graphs. Our results indicate that SZs show decreased variance in the dynamic graph metrics. Consistent with prior stationary functional brain connectivity works, graph measures of identified first-level connectivity states show lower values in SZs. In addition, more first-level connectivity states are disassociated with the second-level connectivity state which resembles the stationary connectivity pattern computed by the entire scan. Collectively, the findings provide new evidence about altered dynamic brain graphs in schizophrenia, which may underscore the abnormal brain performance in this mental illness.

State-related functional integration and functional segregation brain networks in schizophrenia.

Altered topological properties of brain connectivity networks have emerged as important features of schizophrenia. The aim of this study was to investigate how the state-related modulations to graph measures of functional integration and functional segregation brain networks are disrupted in schizophrenia. Firstly, resting state and auditory oddball discrimination (AOD) fMRI data of healthy controls (HCs) and schizophrenia patients (SZs) were decomposed into spatially independent components (ICs) by group independent component analysis (ICA). Then, weighted positive and negative functional integration (inter-component networks) and functional segregation (intra-component networks) brain networks were built in each subject. Subsequently, connectivity strength, clustering coefficient, and global efficiency of all brain networks were statistically compared between groups (HCs and SZs) in each state and between states (rest and AOD) within group. We found that graph measures of negative functional integration brain network and several positive functional segregation brain networks were altered in schizophrenia during AOD task. The metrics of positive functional integration brain network and one positive functional segregation brain network were higher during the resting state than during the AOD task only in HCs. These findings imply that state-related characteristics of both functional integration and functional segregation brain networks are impaired in schizophrenia which provides new insight into the altered brain performance in this brain disorder.