Abstract
Human brains lie at the core of complex neurobiological systems, where the neurons, circuits, and subsystems interact in enigmatic ways. Understanding the structural and functional mechanisms of the brain has long been an intriguing pursuit for neuroscience research and clinical disorder therapy. Mapping the connections of the human brain as a network is one of the most pervasive paradigms in neuroscience. Graph Neural Networks (GNNs) have recently emerged as a potential method for modeling complex network data. Deep models, on the other hand, have low interpretability, which prevents their usage in decision-critical contexts like healthcare. To bridge this gap, we propose an interpretable framework to analyze disorder-specific Regions of Interest (ROIs) and prominent connections. The proposed framework consists of two modules: a brain-network-oriented backbone model for disease prediction and a globally shared explanation generator that highlights disorder-specific biomarkers including salient ROIs and important connections. We conduct experiments on three real-world datasets of brain disorders. The results verify that our framework can obtain outstanding performance and also identify meaningful biomarkers. All code for this work is available at https://github.com/HennyJie/IBGNN.git.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bai, B., et al.: Why attentions may not be interpretable? In: SIGKDD (2021)
Caspers, J., et al.: Within-and across-network alterations of the sensorimotor network in Parkinson’s disease. Neuroradiology 63, 2073–2085 (2021)
Corso, G., et al.: Principal neighbourhood aggregation for graph nets. In: NeurIPS (2020)
Cui, H., et al.: BrainGB: a benchmark for brain network analysis with graph neural networks. arXiv preprint arXiv:2204.07054 (2022)
Cui, H., Lu, Z., Li, P., Yang, C.: On positional and structural node features for graph neural networks on non-attributed graphs. arXiv preprint arXiv:2107.01495 (2021)
Das, T.K., et al.: Parietal lobe and disorganisation syndrome in schizophrenia and psychotic bipolar disorder: a bimodal connectivity study. Psychiatry Res. Neuroimaging 303, 111139 (2020)
Dyrba, M., et al.: Multimodal analysis of functional and structural disconnection in Alzheimer’s disease using multiple kernel SVM. Hum. Brain Mapp. 36, 2118–2131 (2015)
van Eimeren, T., et al.: Dysfunction of the default mode network in Parkinson disease: a functional magnetic resonance imaging study. Arch. Neurol. 66, 877–883 (2009)
Fey, M., et al.: Fast graph representation learning with pytorch geometric. In: RLGM@ICLR (2019)
Figley, T.D., et al.: Probabilistic white matter atlases of human auditory, basal ganglia, language, precuneus, sensorimotor, visual and visuospatial networks. Front. Hum. Neurosci. 11, 306 (2017)
Flannery, J.S., et al.: HIV infection is linked with reduced error-related default mode network suppression and poorer medication management abilities. medRxiv.org (2021)
Herting, M.M., et al.: Default mode connectivity in youth with perinatally acquired HIV. Medicine (2015)
Jain, S., et al.: Attention is not explanation. In: NAACL-HLT (2019)
Jie, B., et al.: Sub-network based kernels for brain network classification. In: ACM BCB (2016)
Kan, X., Cui, H., Lukemire, J., Guo, Y., Yang, C.: FBNetGen: task-aware GNN-based fMRI analysis via functional brain network generation. In: MIDL (2022)
Kan, X., Dai, W., Cui, H., Zhang, Z., Guo, Y., Yang, C.: Brain network transformer. arXiv preprint (2022)
Kawahara, J., et al.: BrainNetCNN: convolutional neural networks for brain networks; towards predicting neurodevelopment. NeuroImage 146, 1038–1049 (2017)
Kendi, A.K., et al.: Altered diffusion in the frontal lobe in Parkinson disease. AJNR Am. J. Neuroradiol. 29, 501–505 (2008)
Kipf, T.N., et al.: Semi-supervised classification with graph convolutional networks. In: ICLR (2017)
Li, X., et al.: BrainGNN: interpretable brain graph neural network for fMRI analysis. Med. Image Anal. 74, 102233 (2021)
Li, Y., et al.: Structural gray matter change early in male patients with HIV. Int. J. Clin. Exp. Med 7, 3362 (2014)
Liu, Y., et al.: Multi-view multi-graph embedding for brain network clustering analysis. In: AAAI (2018)
Lu, H., et al.: MPCA: multilinear principal component analysis of tensor objects. IEEE Trans. Neural Netw. 19, 18–39 (2008)
Luo, D., et al.: Parameterized explainer for graph neural network. In: NeurIPS (2020)
Ma, Q., et al.: HIV-associated structural and functional brain alterations in homosexual males. Front. Neurol. (2021)
Maron, H., et al.: Invariant and equivariant graph networks. In: ICLR (2018)
Martensson, G., et al.: Stability of graph theoretical measures in structural brain networks in Alzheimer’s disease. Sci. Rep. 8, 1–15 (2018)
O’Bryan, R.A., et al.: Disturbances of visual motion perception in bipolar disorder. Bipolar Disord. 16, 354–365 (2014)
Paszke, A., et al.: PyTorch: an imperative style, high-performance deep learning library. In: NeurIPS (2019)
Reavis, E.A., et al.: Structural and functional connectivity of visual cortex in Schizophrenia and bipolar disorder: a graph-theoretic analysis. Schizophr. Bull. Open 1, sgaa056 (2020)
Shao, W., et al.: Clustering on multi-source incomplete data via tensor modeling and factorization. In: PAKDD (2015)
Shirer, W.R., et al.: Decoding subject-driven cognitive states with whole-brain connectivity patterns. Cereb. Cortex 22, 158–165 (2012)
Tessitore, A., et al.: Default-mode network connectivity in cognitively unimpaired patients with Parkinson disease. Neurology 79, 2226–2232 (2012)
Veličković, P., et al.: Graph attention networks. In: ICLR (2018)
Veličković, P., et al.: Deep graph infomax. In: ICLR (2019)
Vu, M.N., et al.: PGM-explainer: probabilistic graphical model explanations for graph neural networks. In: NeurIPS (2020)
Xia, M., et al.: BrainNet viewer: a network visualization tool for human brain connectomics. PLoS ONE 8, e68910 (2013)
Yang, Y., et al.: Data-efficient brain connectome analysis via multi-task meta-learning. In: KDD (2022)
Ying, Z., et al.: GNNExplainer: generating explanations for graph neural networks. In: NeurIPS (2019)
Yuan, H., et al.: Explainability in graph neural networks: a taxonomic survey. arXiv.org (2020)
Yun, S., et al.: Graph transformer networks. In: NeurIPS (2019)
Zhan, L., et al.: Comparison of nine tractography algorithms for detecting abnormal structural brain networks in Alzheimer’s disease. Front. Aging Neurosci. 7, 48 (2015)
Zhu, Y., Cui, H., He, L., Sun, L., Yang, C.: Joint embedding of structural and functional brain networks with graph neural networks for mental illness diagnosis. In: EMBC (2022)
Acknowledgement
This research was partly supported by the internal funds and GPU servers provided by the Computer Science Department of Emory University and the University Research Committee of Emory University. Xiaoxiao Li was supported by NSERC Discovery Grant (DGECR-2022-00430). Lifang He was supported by ONR N00014-18-1-2009 and Lehigh’s accelerator grant S00010293.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Cui, H., Dai, W., Zhu, Y., Li, X., He, L., Yang, C. (2022). Interpretable Graph Neural Networks for Connectome-Based Brain Disorder Analysis. In: Wang, L., Dou, Q., Fletcher, P.T., Speidel, S., Li, S. (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2022. MICCAI 2022. Lecture Notes in Computer Science, vol 13438. Springer, Cham. https://doi.org/10.1007/978-3-031-16452-1_36
Download citation
DOI: https://doi.org/10.1007/978-3-031-16452-1_36
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16451-4
Online ISBN: 978-3-031-16452-1
eBook Packages: Computer ScienceComputer Science (R0)
