Hierarchical Graph Convolutional Network Built by Multiscale Atlases for Brain Disorder Diagnosis Using Functional Connectivity

Knowledge Distillation Guided Interpretable Brain Subgraph Neural Networks for Brain Disorder Exploration

The human brain is a highly complex neurological system that has been the subject of continuous exploration by scientists. With the help of modern neuroimaging techniques, there has been significant progress made in brain disorder analysis. There is an increasing interest about utilizing artificial intelligence techniques to improve the efficiency of disorder diagnosis in recent years. However, these methods rely only on neuroimaging data for disorder diagnosis and do not explore the pathogenic mechanism behind the disorder or provide an interpretable result toward the diagnosis decision. Furthermore, the scarcity of medical data limits the performance of existing methods. As the hot application of graph neural networks (GNNs) in molecular graphs and drug discovery due to its strong graph-structured data learning ability, whether GNNs can also play a huge role in the field of brain disorder analysis. Thus, in this work, we innovatively model brain neuroimaging data into graph-structured data and propose knowledge distillation (KD) guided brain subgraph neural networks to extract discriminative subgraphs between patient and healthy brain graphs to explain which brain regions and abnormal functional connectivities cause the disorder. Specifically, we introduce the KD technique to transfer the knowledge of pretrained teacher model to guide brain subgraph neural networks training and alleviate the problem of insufficient training data. And these discriminative subgraphs are conducive to learn better brain graph-level representations for disorder prediction. We conduct abundant experiments on two functional magnetic resonance imaging datasets, i.e., Parkinson's disease (PD) and attention-deficit/hyperactivity disorder (ADHD), and experimental results well demonstrate the superiority of our method over other brain graph analysis methods for disorder prediction accuracy. The interpretable experimental results given by our method are consistent with corresponding medical research, which is encouraging to provide a potential for deeper brain disorder study.

MHNet: Multi-view High-Order Network for Diagnosing Neurodevelopmental Disorders Using Resting-State fMRI

Deep learning models have shown promise in diagnosing neurodevelopmental disorders (NDD) like ASD and ADHD. However, many models either use graph neural networks (GNN) to construct single-level brain functional networks (BFNs) or employ spatial convolution filtering for local information extraction from rs-fMRI data, often neglecting high-order features crucial for NDD classification. We introduce a Multi-view High-order Network (MHNet) to capture hierarchical and high-order features from multi-view BFNs derived from rs-fMRI data for NDD prediction. MHNet has two branches: the Euclidean Space Features Extraction (ESFE) module and the Non-Euclidean Space Features Extraction (Non-ESFE) module, followed by a Feature Fusion-based Classification (FFC) module for NDD identification. ESFE includes a Functional Connectivity Generation (FCG) module and a High-order Convolutional Neural Network (HCNN) module to extract local and high-order features from BFNs in Euclidean space. Non-ESFE comprises a Generic Internet-like Brain Hierarchical Network Generation (G-IBHN-G) module and a High-order Graph Neural Network (HGNN) module to capture topological and high-order features in non-Euclidean space. Experiments on three public datasets show that MHNet outperforms state-of-the-art methods using both AAL1 and Brainnetome Atlas templates. Extensive ablation studies confirm the superiority of MHNet and the effectiveness of using multi-view fMRI information and high-order features. Our study also offers atlas options for constructing more sophisticated hierarchical networks and explains the association between key brain regions and NDD. MHNet leverages multi-view feature learning from both Euclidean and non-Euclidean spaces, incorporating high-order information from BFNs to enhance NDD classification performance.

mGNN-bw: Multi-Scale Graph Neural Network Based on Biased Random Walk Path Aggregation for ASD Diagnosis

In recent years, computationally assisted diagnosis for classifying autism spectrum disorder (ASD) and typically developing (TD) individuals based on neuroimaging data, such as functional magnetic resonance imaging (fMRI), has garnered significant attention. Studies have shown that long-range functional connectivity patterns in ASD patients exhibit significant abnormalities, and individual brain networks display considerable heterogeneity. However, current graph neural networks (GNNs) used in ASD research have failed to adequately capture long-range connectivity and have overlooked individual differences. To address these limitations, this study proposes a novel multi-scale graph neural network based on biased random walks (mGNN-bw). The model introduces a co-optimization strategy between sub-models and the main model, leveraging node pooling scores from sub-models to guide biased random walks, effectively capturing long-range connectivity. By constructing high-order brain networks through path encoding and aggregation, and integrating them with low-order brain networks based on Pearson correlation, the model achieves a robust multi-scale feature representation. Experimental results on the publicly available ABIDE I dataset demonstrate the superior performance of our approach, achieving accuracy rates of 74.8% and 73.2% using CC200 and AAL atlases, respectively, outperforming existing methods. Additionally, the model identifies key ASD-associated brain regions, including the frontal lobe, insula, cingulate, and calcarine, supported by existing research. The proposed method significantly contributes to the clinical diagnosis of ASD.

Signed Curvature Graph Representation Learning of Brain Networks for Brain Age Estimation

Graph Neural Networks (GNNs) play a pivotal role in learning representations of brain networks for estimating brain age. However, the over-squashing impedes interactions between long-range nodes, hindering the ability of message-passing mechanism-based GNNs to learn the topological structure of brain networks. Graph rewiring methods and curvature GNNs have been proposed to alleviate over-squashing. However, most graph rewiring methods overlook node features and curvature GNNs neglect the geometric properties of signed curvature. In this study, a Signed Curvature GNN (SCGNN) was proposed to rewire the graph based on node features and curvature, and learn the representation of signed curvature. First, a Mutual Information Ollivier-Ricci Flow (MORF) was proposed to add connections in the neighborhood of edge with the minimal negative curvature based on the maximum mutual information between node features, improving the efficiency of information interaction between nodes. Then, a Signed Curvature Convolution (SCC) was proposed to aggregate node features based on positive and negative curvature, facilitating the model's ability to capture the complex topological structures of brain networks. Additionally, an Ollivier-Ricci Gradient Pooling (ORG-Pooling) was proposed to select the key nodes and topology structures by curvature gradient and attention mechanism, accurately obtaining the global representation for brain age estimation. Experiments conducted on six public datasets with structural magnetic resonance imaging (sMRI), spanning ages from 18 to 91 years, validate that our method achieves promising performance compared with existing methods. Furthermore, we employed the gaps between brain age and chronological age for identifying Alzheimer's Disease (AD), yielding the best classification performance.

Learnable Brain Connectivity Structures for Identifying Neurological Disorders

Brain networks/graphs have been widely recognized as powerful and efficient tools for identifying neurological disorders. In recent years, various graph neural network models have been developed to automatically extract features from brain networks. However, a key limitation of these models is that the inputs, namely brain networks/graphs, are constructed using predefined statistical metrics (e.g., Pearson correlation) and are not learnable. The lack of learnability restricts the flexibility of these approaches. While statistically-specific brain networks can be highly effective in recognizing certain diseases, their performance may not exhibit robustness when applied to other types of brain disorders. To address this issue, we propose a novel module called Brain Structure Inference (termed BSI), which can be seamlessly integrated with multiple downstream tasks within a unified framework, enabling end-to-end training. It is highly flexible to learn the most beneficial underlying graph structures directly for specific downstream tasks. The proposed method achieves classification accuracies of 74.83% and 79.18% on two publicly available datasets, respectively. This suggests an improvement of at least 3% over the best-performing existing methods for both tasks. In addition to its excellent performance, the proposed method is highly interpretable, and the results are generally consistent with previous findings.

Functional Networks of Reward and Punishment Processing and Their Molecular Profiles Predicting the Severity of Young Adult Drinking

Alcohol misuse is associated with altered punishment and reward processing. Here, we investigated neural network responses to reward and punishment and the molecular profiles of the connectivity features predicting alcohol use severity in young adults. We curated the Human Connectome Project data and employed connectome-based predictive modeling (CPM) to examine how functional connectivity (FC) features during wins and losses are associated with alcohol use severity, quantified by Semi-Structured Assessment for the Genetics of Alcoholism, in 981 young adults. We combined the CPM findings and the JuSpace toolbox to characterize the molecular profiles of the network connectivity features of alcohol use severity. The connectomics predicting alcohol use severity appeared specific, comprising less than 0.12% of all features, including medial frontal, motor/sensory, and cerebellum/brainstem networks during punishment processing and medial frontal, fronto-parietal, and motor/sensory networks during reward processing. Spatial correlation analyses showed that these networks were associated predominantly with serotonergic and GABAa signaling. To conclude, a distinct pattern of network connectivity predicted alcohol use severity in young adult drinkers. These "neural fingerprints" elucidate how alcohol misuse impacts the brain and provide evidence of new targets for future intervention.

A Multi-Graph Cross-Attention-Based Region-Aware Feature Fusion Network Using Multi-Template for Brain Disorder Diagnosis

Functional connectivity (FC) networks based on resting-state functional magnetic imaging (rs-fMRI) are reliable and sensitive for brain disorder diagnosis. However, most existing methods are limited by using a single template, which may be insufficient to reveal complex brain connectivities. Furthermore, these methods usually neglect the complementary information between static and dynamic brain networks, and the functional divergence among different brain regions, leading to suboptimal diagnosis performance. To address these limitations, we propose a novel multi-graph cross-attention based region-aware feature fusion network (MGCA-RAFFNet) by using multi-template for brain disorder diagnosis. Specifically, we first employ multi-template to parcellate the brain space into different regions of interest (ROIs). Then, a multi-graph cross-attention network (MGCAN), including static and dynamic graph convolutions, is developed to explore the deep features contained in multi-template data, which can effectively analyze complex interaction patterns of brain networks for each template, and further adopt a dual-view cross-attention (DVCA) to acquire complementary information. Finally, to efficiently fuse multiple static-dynamic features, we design a region-aware feature fusion network (RAFFNet), which is beneficial to improve the feature discrimination by considering the underlying relations among static-dynamic features in different brain regions. Our proposed method is evaluated on both public ADNI-2 and ABIDE-I datasets for diagnosing mild cognitive impairment (MCI) and autism spectrum disorder (ASD). Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art methods. Our source code is available at https://github.com/mylbuaa/MGCA-RAFFNet.

An Umbrella Review of the Fusion of fMRI and AI in Autism

The role of functional magnetic resonance imaging (fMRI) is assuming an increasingly central role in autism diagnosis. The integration of Artificial Intelligence (AI) into the realm of applications further contributes to its development. This study's objective is to analyze emerging themes in this domain through an umbrella review, encompassing systematic reviews. The research methodology was based on a structured process for conducting a literature narrative review, using an umbrella review in PubMed and Scopus. Rigorous criteria, a standard checklist, and a qualification process were meticulously applied. The findings include 20 systematic reviews that underscore key themes in autism research, particularly emphasizing the significance of technological integration, including the pivotal roles of fMRI and AI. This study also highlights the enigmatic role of oxytocin. While acknowledging the immense potential in this field, the outcome does not evade acknowledging the significant challenges and limitations. Intriguingly, there is a growing emphasis on research and innovation in AI, whereas aspects related to the integration of healthcare processes, such as regulation, acceptance, informed consent, and data security, receive comparatively less attention. Additionally, the integration of these findings into Personalized Medicine (PM) represents a promising yet relatively unexplored area within autism research. This study concludes by encouraging scholars to focus on the critical themes of health domain integration, vital for the routine implementation of these applications.

A common spectrum underlying brain disorders across lifespan revealed by deep learning on brain networks

Brain disorders in the early and late life of humans potentially share pathological alterations in brain functions. However, the key neuroimaging evidence remains unrevealed for elucidating such commonness and the relationships among these disorders. To explore this puzzle, we build a restricted single-branch deep learning model, using multi-site functional magnetic resonance imaging data (N = 4,410, 6 sites), for classifying 5 different early- and late-life brain disorders from healthy controls (cognitively unimpaired). Our model achieves 62.6 ± 1.9% overall classification accuracy and thus supports us in detecting a set of commonly affected functional subnetworks, including default mode, executive control, visual, and limbic networks. In the deep-layer representation of data, we observe young and aging patients with disorders are continuously distributed, which is in line with the clinical concept of the "spectrum of disorders." The relationships among brain disorders from the revealed spectrum promote the understanding of disorder comorbidities and time associations in the lifespan.