基于适应性超图神经网络的抑郁症诊断
Adaptive Hypergraph Neural Network for the Diagnosis of Depression
摘要: 近年来的研究越来越关注基于深度学习的抑郁症自动诊断。然而，以往基于图神经网络(GNN)的抑郁症fMRI数据分类方法并未充分考虑脑区间的高级交互作用以及其动态变化。为了解决这一挑战，本研究提出了适应性超图神经网络(AHGNN)来进行抑郁症的诊断。该模型采用了超图神经网络(HGNN)来提取脑区间的高阶交互信息，并使用可学习策略适应性地在模型训练过程中更新大脑超图结构。我们在抑郁症脑影像大数据联盟(REST-meta-MDD)数据集上对该模型进行了训练和评估。实验结果显示，研究所提出的架构在抑郁症诊断上取得了显著效果，模型的分类准确率达到了71.41%，实验结果还表明超图神经网络和自适应超图构建模块的运用显著增强了模型的分类性能。
Abstract: Depression classification has emerged as a popular research topic in recent years. Previous graph neural network (GNN) based methods for Depression fMRI data classification have given less con-sideration to the high-order interactions among brain regions and the dynamic variations in func-tional connectivity. To address this issue, we propose an Adaptive Hypergraph Neural Network (AHGNN). We utilize HGNN, coupled with the learnable adaptive hypergraph construction (AHC) module, to extract intricate interplay among brain regions. The model is trained and evaluated on the REST-meta-MDD dataset. Our proposed architecture achieves a commendable depression diag-nostic performance, realizing a classification accuracy rate of 71.41%. The employment of HGNN and the AHC module significantly enhances the classification efficacy of the model.
文章引用：刘永顺. 基于适应性超图神经网络的抑郁症诊断[J]. 交叉科学快报, 2024, 8(1): 1-8. https://doi.org/10.12677/ISL.2024.81001

参考文献

[1] 刘威, 李海江, 邱江. 抑郁症的脑网络失调: 来自图论分析的证据[J]. 心理科学进展, 2015, 23(1): 85-92.
[2] Biswal, B.B., Mennes, M., Zuo, X.N., Gohel, S., Kelly, C., Smith, S.M. and Milham, M.P. (2010) Toward Discovery Science of Human Brain Function. Proceedings of the National Academy of Sciences, 107, 4734-4739. [Google Scholar] [CrossRef] [PubMed]
[3] Greicius, M.D., Krasnow, B., Reiss, A.L. and Menon, V. (2003) Functional Connectivity in the Resting Brain: A Network Analysis of the Default Mode Network. Neuron, 38, 657-668.
[4] Fox, M.D. and Raichle, M.E. (2007) Spontaneous Fluctuations in Brain Activity Observed with Func-tional Magnetic Resonance Imaging. Nature Reviews Neuroscience, 8, 700-711. [Google Scholar] [CrossRef] [PubMed]
[5] Saeidi, M., Karwowski, W., Farahani, F.V., Fiok, K., Hancock, P., Sawyer, B.D., Christov-Moore, L. and Douglas, P.K. (2022) Decoding Task-Based fMRI Data with Graph Neural Networks, Considering Individual Differences. Brain Sciences, 12, 1094. [Google Scholar] [CrossRef] [PubMed]
[6] Kan, X., Cui, H., Lukemire, J., Guo, Y. and Yang, C. (2022) Fbnetgen: Task-Aware Gnn-Based Fmri Analysis via Functional Brain Network Generation. Proceedings of the International Conference on Medical Imaging with Deep Learning, PMLR, Zurich, 8 June 2022, 618-637.
[7] Zhang, Y., Tetrel, L., Thirion, B. and Bellec, P. (2021) Functional Annota-tion of Human Cognitive States Using Deep Graph Convolution. NeuroImage, 231, Article ID: 117847. [Google Scholar] [CrossRef] [PubMed]
[8] Battiston, F., Amico, E., Barrat, A., Bianconi, G., Ferraz de Arruda, G., Franceschiello, B., Iacopini, I., Kéfi, S., Latora, V., Moreno, Y., et al. (2021) The Physics of High-er-Order Interactions in Complex Systems. Nature Physics, 17, 1093-1098. [Google Scholar] [CrossRef] 
[9] Jiang, J., Wei, Y., Feng, Y., Cao, J. and Gao, Y. (2019) Dy-namic Hypergraph Neural Networks. IJCAI, Macao, 16 October 2019, 2635-2641. [Google Scholar] [CrossRef] 
[10] Ma, J., Wang, Y., Liu, B. and Liu, W. (2021) Accurately Modeling the Human Brain Functional Correlations with Hypergraph Laplacian. Neurocomputing, 428, 239-247. [Google Scholar] [CrossRef] 
[11] Giusti, C., Ghrist, R. and Bassett, D.S. (2016) Two’s Compa-ny, Three (or More) Is a Simplex: Algebraic-Topological Tools for Understanding Higher-Order Structure in Neural Data. Journal of Computational Neuroscience, 41, 1-14. [Google Scholar] [CrossRef] [PubMed]
[12] Wang, X., Xin, J., Wang, Z., Li, C. and Wang, Z. (2022) An Evolving Hypergraph Convolutional Network for the Diagnosis of Alzheimer’s Disease. Diagnostics, 12, 2632. [Google Scholar] [CrossRef] [PubMed]
[13] Zhang, Z., Liu, J., Li, B. and Gao, Y. (2021) Diagnosis of Childhood Autism Using Multi-Modal Functional Connectivity via Dynamic Hypergraph Learning. CICAI, Hangzhou, 5 June 2021, 123-135. [Google Scholar] [CrossRef] 
[14] Feng, Y., You, H., Zhang, Z., Ji, R. and Gao, Y. (2019) Hy-pergraph Neural Networks. Proceedings of the AAAI Conference on Artificial Intelligence, Honolulu, 27 January 2019, 3558-3565. [Google Scholar] [CrossRef] 
[15] Yan, C., Chen, X. and Li, L. (2022) Data of the REST-Meta-MDD Project from DIRECT Consortium. V2. Science Data Bank.
[16] Kawahara, J., Brown, C.J., Miller, S.P., Booth, B.G., Chau, V., Grunau, R.E. and Hamarneh, G. (2017) BrainNetCNN: Convolutional Neural Networks for Brain Networks, towards Predicting Neurodevelopment. NeuroImage, 146, 1038-1049. [Google Scholar] [CrossRef] [PubMed]
[17] Zhang, Y., Li, Y., Wang, Y., Yang, Y. and Wei, X. (2021) Adaptive Spatio-Temporal Graph Information Fusion for Remaining Useful Life Prediction. IEEE Sensors Journal, 22, 3334-3347. [Google Scholar] [CrossRef]