Enhanced Graph Convolutional Networks for ASD Diagnosis: A Multimodal Transfer Learning Based Approach Using Chebyshev Spectral Graph and Attention Mechanism
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur, Bangladesh.
Abstract
Autism Spectrum Disorder (ASD) is a complicated neurodevelopmental disorder
marked by variation in symptom presentation and neurological underpinnings,
making early and objective diagnosis extremely problematic. This paper offers a
Graph Convolutional Network (GCN) model, incorporating Chebyshev Spectral
Graph Convolution and Graph Attention Networks (GAT), to increase the
classification accuracy of ASD utilizing multi-modal neuroimaging and phenotypic
data.
Alongside, a transfer learning approach has been utilized in separate
experiments to explore the possibility of knowledge transfer in GCN based
architectures.
Leveraging the ABIDE I dataset, which contains resting-state
functional MRI (rs-fMRI), structural MRI (sMRI), and phenotypic variables from
870 patients, the model leverages a multi-branch architecture that processes each
modality individually before merging them via concatenation. Graph structure is
encoded using site-based similarity to generate a population graph, which helps in
understanding relationship connections across individuals. Chebyshev polynomial
filters provide localized spectral learning with lower computational complexity,
whereas GAT layers increase node representations by attention-weighted
aggregation of surrounding information. Also, transfer learning leads to a more
robust feature generalizations across domains. Extensive trials illustrate the model’s
superiority, reaching a test accuracy of 74.82% and an AUC of 0.82 on the whole
dataset, and a transfer learning based approach achieved an accuracy of 72.89% and
an AUC of 0.79, surpassing multiple state-of-the-art baselines, including
conventional GCNs, autoencoder-based deep neural networks, and multi-modal
CNNs. Ablation studies further demonstrate the incremental influence of each
architectural component, emphasizing the usefulness of the methods using graph
attention, hyper-parameter adjustment, and transfer learning. The results validate
the effectiveness of graph-based deep learning for ASD classification, notably in
combining multi-modal data and capturing complicated inter-subject interactions.
This work takes a crucial step toward building reliable, scalable, and objective
computer-aided diagnostic methods for ASD
Description
Supervised by
Prof. Dr. Md. Hasanul Kabir,
Department of Computer Science and Engineering (CSE),
Islamic University of Technology (IUT),
Board Bazar, Gazipur, Bangladesh.
This thesis is submitted in partial fulfillment of the requirement for the degree of Master of Science in Computer Science and Engineering, 2025
Keywords
Citation
[1] S. Q. Abbas, L. Chi, and Y.-P. P. Chen, “Deepmnf: Deep multimodal neuroimaging framework for diagnosing autism spectrum disorder”, Artificial Intelligence in Medicine, vol. 136, p. 102475, 2023. [2] S. Q. Abbas, L. Chi, and Y.-P. P. Chen, “Deepmnf: Deep multimodal neuroimaging framework for diagnosing autism spectrum disorder”, Artificial Intelligence in Medicine, vol. 136, p. 102475, Feb. 2023, issn: 09333657. doi: 10 . 1016 / j . artmed . 2022 . 102475. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S0933365722002275. [3] A. Abraham, M. P. Milham, A. Di Martino, et al., “Deriving reproducible biomarkers from multi-site resting-state data: An autism-based example”, NeuroImage, vol. 147, pp. 736–745, 2017. [4] F. Almuqhim and F. Saeed, “Asd-saenet: A sparse autoencoder, and deep-neural network model for detecting autism spectrum disorder (asd) using fmri data”, Frontiers in Computational Neuroscience, vol. 15, p. 654315, 2021. [5] J. Ashburner, “A fast diffeomorphic image registration algorithm”, Neuroimage, vol. 38, no. 1, pp. 95–113, 2007. [6] R. O. Bahado-Singh, S. Vishweswaraiah, B. Aydas, et al., “Artificial intelligence analysis of newborn leucocyte epigenomic markers for the prediction of autism”, Brain research, vol. 1724, p. 146457, 2019. [7] J. Bathelt, J. Holmes, D. E. Astle, et al., “Data-driven subtyping of executive function–related behavioral problems in children”, Journal of the American Academy of Child & Adolescent Psychiatry, vol. 57, no. 4, pp. 252–262, 2018. [8] M. A. Bayram,. Özer, and F. Temurta, “Deep learning methods for autism spectrum disorder diagnosis based on fmri images”, Sakarya university journal of computer and information sciences, vol. 4, no. 1, pp. 142–155, 2021. [9] C. Cameron, B. Yassine, C. Carlton, et al., “The neuro bureau preprocessing initiative: Open sharing of preprocessed neuroimaging data and derivatives”, 56 Frontiers in Neuroinformatics, vol. 7, 2013. doi: 10.3389/CONF.FNINF.2013. 09.00041/EVENT_ABSTRACT. [10] F. Campo, A. Retico, S. Calderoni, and P. Oliva, “Multi-site mri data harmonization with an adversarial learning approach: Implementation to the study of brain connectivity in autism spectrum disorders”, Applied Sciences, vol. 13, no. 11, p. 6486, 2023. [11] C. Chang and G. H. Glover, “Time–frequency dynamics of resting-state brain connectivity measured with fmri”, Neuroimage, vol. 50, no. 1, pp. 81–98, 2010. [12] C. Chlebowski, J. A. Green, M. L. Barton, and D. Fein, “Using the childhood autism rating scale to diagnose autism spectrum disorders”, Journal of autism and developmental disorders, vol. 40, p. 787, 7 Jul. 2010, issn: 01623257. doi: 10.1007/S10803-009-0926-X. [Online]. Available: https://pmc.ncbi.nlm. nih.gov/articles/PMC3612531/. [13] A. M. Daniels and D. S. Mandell, “Explaining differences in age at autism spectrum disorder diagnosis: A critical review”, Autism : the international journal of research and practice, vol. 18, pp. 583–597, 5 2014, issn: 1461-7005. doi: 10 . 1177 / 1362361313480277. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/23787411/. [14] M. Defferrard, X. Bresson, and P. Vandergheynst, “Convolutional neural networks on graphs with fast localized spectral filtering”, Advances in neural information processing systems, vol. 29, 2016. [15] Y. Dong, D. Batalle, and M. Deprez, “A framework for comparison and interpretation of machine learning classifiers to predict autism on the abide dataset”, Human brain mapping, vol. 46, 5 Apr. 2025, issn: 1097-0193. doi: 10 . 1002 / HBM . 70190. https://pubmed.ncbi.nlm.nih.gov/40095417/. [Online]. Available: [16] F. Doshi-Velez and B. Kim, “Towards a rigorous science of interpretable machine learning”, arXiv preprint arXiv:1702.08608, 2017. [17] J. Du, S. Wang, R. Chen, and S. Wang, “Improving fmri-based autism severity identification via brain network distance and adaptive label distribution learning”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 33, pp. 162–174, 2025. doi: 10.1109/TNSRE.2024.3516216. [18] Y. Du, X. He, P. Kochunov, et al., “A new multimodality fusion classification approach to explore the uniqueness of schizophrenia and autism spectrum disorder”, Human Brain Mapping, vol. 43, pp. 3887–3903, 12 Aug. 2022, issn: 1097-0193. doi: 10 . 1002 / HBM . 25890. [Online]. Available: https://onlinelibrary.wiley.com/doi/full/10.1002/hbm.25890. 57 [19] C. Ecker, S. Y. Bookheimer, and D. G. Murphy, “Neuroimaging in autism spectrum disorder: Brain structure and function across the lifespan”, The Lancet Neurology, vol. 14, no. 11, pp. 1121–1134, 2015. [20] F.EDITION,“Diagnosticandstatisticalmanualofmentaldisorders”,American Psychiatric Association, Washington, DC, pp. 205–224, 1980. [21] F. Edition et al., “Diagnostic and statistical manual of mental disorders”, Am Psychiatric Assoc, vol. 21, no. 21, pp. 591–643, 2013. [22] B.Fischl, “Freesurfer”, Neuroimage, vol. 62, no. 2, pp. 774–781, 2012. [23] A. El-Gazzar, M. Quaak, L. Cerliani, P. Bloem, G. van Wingen, and R. Mani Thomas, “A hybrid 3dcnn and 3dc-lstm based model for 4d spatio-temporal fmri data: An abide autism classification study”, in International Workshop on OR 2.0 Context-Aware Operating Theaters, Springer, 2019, pp. 95–102. [24] K.Gupta, A. Aly, and E. Ifeachor, “Cross-domain transfer learning for domain adaptation in autism spectrum disorder diagnosis”, in 18th International Conference on Health Informatics, 2025. [25] J.Gutiérrez,Z.Che,G.Zhai,andP.LeCallet,“Saliency4asd:Challenge,dataset and tools for visual attention modeling for autism spectrum disorder”, Signal Processing: Image Communication, vol. 92, p. 116092, 2021. [26] J. Hashemi, G. Dawson, K. L. Carpenter, et al., “Computer vision analysis for quantification of autism risk behaviors”, IEEE Transactions on Affective Computing, vol. 12, no. 1, pp. 215–226, 2018. [27] M.He,Z. Wei, and J.-R. Wen, “Convolutional neural networks on graphs with chebyshev approximation, revisited”, Advances in neural information processing systems, vol. 35, pp. 7264–7276, 2022. [28] M.He, Z. Wei, H. Xu, et al., “Bernnet: Learning arbitrary graph spectral filters via bernstein approximation”, Advances in neural information processing systems, vol. 34, pp. 14239–14251, 2021. [29] A.S.Heinsfeld, A.R.Franco,R.C.Craddock,A.Buchweitz,andF.Meneguzzi, “Identification of autism spectrum disorder using deep learning and the abide dataset”, NeuroImage: Clinical, vol. 17, pp. 16–23, 2018, issn: 22131582. doi: 10.1016/j.nicl.2017.08.017. [30] S.-J. Hong, S. L. Valk, A. Di Martino, M. P. Milham, and B. C. Bernhardt, “Multidimensional neuroanatomical subtyping of autism spectrum disorder”, Cerebral Cortex, vol. 28, no. 10, pp. 3578–3588, 2018. 58 [31] Y. Huang and A. C. Chung, “Edge-variational graph convolutional networks for uncertainty-aware disease prediction”, in Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part VII 23, Springer, 2020, pp. 562–572. [32] V. Jain, C. T. Rakshe, S. S. Sengar, M. Murugappan, and J. F. A. Ronickom, “Age-and severity-specific deep learning models for autism spectrum disorder classification using functional connectivity measures”, Arabian Journal for Science and Engineering, vol. 49, no. 5, pp. 6847–6865, 2024. [33] N. A. Khan and X. Shang, “Asd-gannet: A generative adversarial network-inspired deep learning approach for the classification of autism brain disorder”, Brain Sciences, vol. 14, no. 8, p. 766, 2024. [34] M. Khosla, K. Jamison, A. Kuceyeski, and M. R. Sabuncu, “3d convolutional neural networks for classification of functional connectomes”, in international workshop on deep learning in medical image analysis, Springer, 2018, pp. 137–145. [35] L. Li, H. Jiang, G. Wen, et al., “Te-hi-gcn: An ensemble of transfer hierarchical graph convolutional networks for disorder diagnosis”, Neuroinformatics, vol. 20, pp. 353–375, 2 Apr. 2022, issn: 15590089. doi: 10 . 1007 / S12021- 021- 09548- 1 / METRICS. [Online]. Available: https://link.springer.com/article/10.1007/s12021-021-09548-1. [36] M. V. Lombardo, M.-C. Lai, and S. Baron-Cohen, “Big data approaches to decomposing heterogeneity across the autism spectrum”, Molecular psychiatry, vol. 24, no. 10, pp. 1435–1450, 2019. [37] C. Lord, T. S. Brugha, T. Charman, et al., “Autism spectrum disorder”, Nature reviews. Disease primers, vol. 6, 1 Jan. 2020, issn: 2056-676X. doi: 10.1038/ S41572-019-0138-4. [Online]. Available: https://pubmed.ncbi.nlm.nih. gov/31949163/. [38] C.Lord, T. Charman, A. Havdahl, et al., “The lancet commission on the future of care and clinical research in autism”, The Lancet, vol. 399, no. 10321, pp. 271–334, 2022. [39] C. Lord, M. Rutter, S. Goode, et al., “Autism diagnostic observation schedule: A standardized observation of communicative and social behavior”, Journal of autism and developmental disorders, vol. 19, pp. 185–212, 2 Jun. 1989, issn: 0162-3257. doi: 10 . 1007 / BF02211841. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/2745388/. 59 [40] C. Lord, M. Rutter, and A. Le Couteur, “Autism diagnostic interview-revised: A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders”, Journal of autism and developmental disorders, vol. 24, no. 5, pp. 659–685, 1994. [41] Y. Ma, X. Mu, T. Zhang, and Y. Zhao, “Maft-so: A novel multi-atlas fusion template based on spatial overlap for asd diagnosis”, Journal of Biomedical Informatics, vol. 157, p. 104714, Sep. 2024, issn: 1532-0464. doi: 10.1016/J.JBI.2024.104714. [42] M. J. Maenner, K. A. Shaw, A. V. Bakian, et al., “Prevalence of autism spectrum disorder among children aged 8 yearsautism and developmental disabilities monitoring network, 11 sites, united states, 2020”, MMWR Surveillance Summaries, vol. 72, no. 2, p. 1, 2023. [43] S. Majumder, F. Apicella, F. Muratori, and K. Das, “Detecting autism spectrum disorder using topological data analysis”, in ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, 2020, pp. 1210–1214. [44] A.D.Martino,C.G.Yan,Q.Li,etal.,“Theautismbrainimagingdataexchange: Towards a large-scale evaluation of the intrinsic brain architecture in autism”, Molecular Psychiatry 2014 19:6, vol. 19, pp. 659–667, 6 Jun. 2013, issn: 1476 5578. doi: 10.1038/mp.2013.78. [Online]. Available: https://www.nature. com/articles/mp201378. [45] A. M. Pagnozzi, E. Conti, S. Calderoni, J. Fripp, and S. E. Rose, “A systematic review of structural mri biomarkers in autism spectrum disorder: A machine learning perspective”, International Journal of Developmental Neuroscience, vol. 71, pp. 68–82, 2018. [46] S. J. Pan and Q. Yang, “A survey on transfer learning”, IEEE Transactions on knowledge and data engineering, vol. 22, no. 10, pp. 1345–1359, 2009. [47] S. Parisot, S. I. Ktena, E. Ferrante, et al., “Disease prediction using graph convolutional networks: Application to autism spectrum disorder and alzheimers disease”, Medical image analysis, vol. 48, pp. 117–130, 2018. [48] S. Rajaprakash, C. B. Basha, C. S. Ram, I. Ameethbasha, V. Subapriya, and R. Sofia, “Using convolutional network in graphical model detection of autism disorders with fuzzy inference systems”, Intelligence-Based Medicine, vol. 11, p. 100213, 2025, issn: 26665212. doi: 10.1016/j.ibmed.2025.100213. [49] M. Raki, M. Cabezas, K. Kushibar, A. Oliver, and X. Lladó, “Improving the detection of autism spectrum disorder by combining structural and functional mri information”, NeuroImage: Clinical, vol. 25, p. 102181, 2020. 60 [50] A. Rathore, S. Palande, J. S. Anderson, B. A. Zielinski, P. T. Fletcher, and B. Wang, “Autism classification using topological features and deep learning: A cautionary tale”, in International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer, 2019, pp. 736–744. [51] J. Rehg, G. Abowd, A. Rozga, et al., “Decoding children’s social behavior”, in Proceedings of the IEEE conference on computer vision and pattern recognition, 2013, pp. 3414–3421. [52] K. Riddle, C. J. Cascio, and N. D. Woodward, “Brain structure in autism: A voxel-based morphometry analysis of the autism brain imaging database exchange (abide)”, Brain imaging and behavior, vol. 11, no. 2, pp. 541–551, 2017. [53] G. Riva and E. Riva, “De-enigma: Multimodal human–robot interaction for teaching and expanding social imagination in autistic children.”, CyberPsychology, Behavior & Social Networking, vol. 23, no. 11, 2020. [54] S. Saponaro, A. Giuliano, R. Bellotti, et al., “Multi-site harmonization of mri data uncovers machine-learning discrimination capability in barely separable populations: Anexamplefromtheabidedataset”,NeuroImage:Clinical,vol.35, p. 103082, 2022. [55] S. Saponaro, F. Lizzi, G. Serra, et al., “Deep learning based joint fusion approach to exploit anatomical and functional brain information in autism spectrum disorders”, Brain Informatics, vol. 11, pp. 1–13, 1 Dec. 2024, issn: 21984026. doi: 10 .1186/S40708- 023- 00217- 4/FIGURES/4. [Online]. Available: https://link.springer.com/articles/10.1186/s40708-023-00217-4. [56] I. Shahzad, S. U. R. Khan, A. Waseem, Z. U. Abideen, and J. Liu, “Enhancing asd classification through hybrid attention-based learning of facial features”, Signal, Image and Video Processing, vol. 18, no. Suppl 1, pp. 475–488, 2024. [57] K. A. Shaw, “Prevalence and early identification of autism spectrum disorder among children aged 4 and 8 yearsautism and developmental disabilities monitoring network, 16 sites, united states, 2022”, MMWR. Surveillance Summaries, vol. 74, 2025. [58] S. Shrivastava, U. Mishra, N. Singh, A. Chandra, and S. Verma, “Control or autism-classification using convolutional neural networks on functional mri”, in 2020 11th International Conference on Computing, Communication and Networking Technologies (ICCCNT), IEEE, 2020, pp. 1–6. [59] T. Song, Z. Ren, J. Zhang, and M. Wang, “Multi-view and multimodal graph convolutional neural network for autism spectrum disorder diagnosis”, 61 Mathematics 2024, Vol. 12, Page 1648, vol. 12, p. 1648, 11 May 2024, issn: 2227-7390. doi: 10 . 3390 / MATH12111648. [Online]. Available: https://www.mdpi.com/2227-7390/12/11/1648. [60] J. Stember, D. Stember, L. Pasquini, J. Merhnaz, A. Holodny, and H. Shalu, “Deep reinforcement learning for fmri prediction of autism spectrum disorder”, arXiv preprint arXiv:2206.11224, 2022. [61] K. A. Stigler, B. C. McDonald, A. Anand, A. J. Saykin, and C. J. McDougle, “Structural and functional magnetic resonance imaging of autism spectrum disorders”, Brain research, vol. 1380, pp. 146–161, Mar. 2011, issn: 1872-6240. doi: 10 . 1016 / J . BRAINRES . 2010 . 11 . 076. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/21130750/. [62] Y. Tao and M.-L. Shyu, “Sp-asdnet: Cnn-lstm based asd classification model using observer scanpaths”, in 2019 IEEE International conference on multimedia & expo workshops (ICMEW), IEEE, 2019, pp. 641–646. [63] P. M. Thompson, E. R. Sowell, N. Gogtay, et al., “Structural mri and brain development”, International review of neurobiology, vol. 67, pp. 285–323, 2005. [64] T. Tommasi, F. Orabona, and B. Caputo, “Safety in numbers: Learning categories from few examples with multi model knowledge transfer”, in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE, 2010, pp. 3081–3088. [65] B. G. Travers, N. Adluru, C. Ennis, et al., “Diffusion tensor imaging in autism spectrum disorder: A review”, AutismResearch, vol. 5, no. 5, pp. 289–313, 2012. [66] L. Q. Uddin, K. Supekar, and V. Menon, “Reconceptualizing functional brain connectivity in autism from a developmental perspective”, Frontiers in human neuroscience, vol. 7, p. 458, 2013. [67] G. Varoquaux, “Cross-validation failure: Small sample sizes lead to large error bars”, NeuroImage, vol. 180, pp. 68–77, 2018. [68] P.Velikovi, G.Cucurull, A.Casanova,A.Romero,P.Lio,andY.Bengio,“Graph attention networks”, arXiv preprint arXiv:1710.10903, 2017. [69] M. Wang, J. Guo, Y. Wang, M. Yu, and J. Guo, “Multimodal autism spectrum disorder diagnosis method based on deepgcn”, IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 31, pp. 3664–3674, 2023, issn: 15580210. doi: 10.1109/TNSRE.2023.3314516. [70] Z. Wang, Y. Xu, D. Peng, J. Gao, and F. Lu, “Brain functional activity-based classification of autism spectrum disorder using an attention-based graph 62 neural network combined with gene expression”, Cerebral Cortex, vol. 33, no. 10, pp. 6407–6419, 2023. [71] I. H. Witten and E. Frank, “Data mining: Practical machine learning tools and techniques with java implementations”, Acm Sigmod Record, vol. 31, no. 1, pp. 76–77, 2002. [72] W. Yan, G. Qu, W. Hu, et al., “Deep learning in neuroimaging: Promises and challenges”, IEEE Signal Processing Magazine, vol. 39, no. 2, pp. 87–98, 2022. [73] B. Yu, H. Yin, and Z. Zhu, “Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting”, arXiv preprint arXiv:1709.04875, 2017. [74] R. Zhang, F. Sheng, L. Wang, and W. Zeng, “Research on multi-center assisted diagnosis of asd based on multimodal feature fusion”, International Journal of Imaging Systems and Technology, vol. 34, no. 3, e23110, 2024. doi: https : / / doi . org / 10 . 1002 / ima . 23110. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1002/ima.23110. [75] R. Zhang, F. Sheng, L. Wang, and W. Zeng, “Research on multi-center assisted diagnosis of asd based on multimodal feature fusion”, International Journal of Imaging Systems and Technology, vol. 34, e23110, 3 May 2024, issn: 1098-1098. doi: 10.1002/IMA.23110. [Online]. Available: https://onlinelibrary. wiley.com/doi/full/10.1002/ima.23110. [76] S. Zhang, H. Tong, J. Xu, and R. Maciejewski, “Graph convolutional networks: A comprehensive review”, Computational Social Networks, vol. 6, no. 1, pp. 1–23, 2019. [77] A.Zia, A. Khamis, J. Nichols, et al., “Topological deep learning: A review of an emerging paradigm”, Artificial Intelligence Review, vol. 57, no. 4, p. 77, 2024. [78] A. Zunino, P. Morerio, A. Cavallo, et al., “Video gesture analysis for autism spectrum disorder detection”, in 2018 24th International Conference on Pattern Recognition (ICPR), 2018, pp. 3421–3426. doi: 10.1109/ICPR.2018.8545095. [79] L. Zwaigenbaum, M. L. Bauman, W. L. Stone, et al., “Early identification of autism spectrum disorder: Recommendations for practice and research”, Pediatrics, vol. 136 Suppl 1, S10–S40, Suppl 1 Oct. 2015, issn: 1098-4275. doi: 10 . 1542 / PEDS . 2014- 3667C. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/26430168/.