Abstract
Background: Autism Spectrum Disorder (ASD) is a multifaceted neurodevelopmental condition with significant heterogeneity in its clinical presentation. Timely and precise identification of ASD is crucial for effective intervention and assistance. Recent advances in deep learning techniques have shown promise in enhancing the accuracy of ASD detection.
Objective: This comprehensive review aims to provide an overview of various deep learning methods employed in detecting ASD, utilizing diverse neuroimaging modalities. We analyze a range of studies that use resting-state functional Magnetic Resonance Imaging (rsfMRI), structural MRI (sMRI), task-based fMRI (tfMRI), and electroencephalography (EEG). This paper aims to assess the effectiveness of these techniques based on criteria such as accuracy, sensitivity, specificity, and computational efficiency.
Methods: We systematically review studies investigating ASD detection using deep learning across different neuroimaging modalities. These studies utilize various preprocessing tools, atlases, feature extraction techniques, and classification algorithms. The performance metrics of interest include accuracy, sensitivity, specificity, precision, F1-score, recall, and area under the curve (AUC).
Results: The review covers a wide range of studies, each with its own dataset and methodology. Notable findings include a study employing rsfMRI data from ABIDE that achieved an accuracy of 80% using LeNet. Another study using rsfMRI data from ABIDE-II achieved an impressive accuracy of 95.4% with the ASGCN deep learning model. Studies utilizing different modalities, such as EEG and sMRI, also reported high accuracies ranging from 74% to 95%.
Conclusion: Deep learning-based approaches for ASD detection have demonstrated significant potential across multiple neuroimaging modalities. These methods offer a more objective and data-driven approach to diagnosis, potentially reducing the subjectivity associated with clinical evaluations. However, challenges remain, including the need for larger and more diverse datasets, model interpretability, and clinical validation. The field of deep learning in ASD diagnosis continues to evolve, holding promise for early and accurate identification of individuals with ASD, which is crucial for timely intervention and support.
Keywords: Deep learning, autism spectrum disorder (ASD), comprehensive review, neuroimaging modalities, resting-state functional magnetic resonance imaging (rsfMRI), structural MRI (sMRI), task-based fMRI (tfMRI), electroencephalography (EEG), preprocessing tools, atlases, ABIDE dataset, LeNet, ASGCN (adaptive spatiotemporal, graph convolution network).
[http://dx.doi.org/10.3389/fnsys.2022.904770] [PMID: 36817947]
[http://dx.doi.org/10.2147/PRBM.S117499] [PMID: 28883746]
[http://dx.doi.org/10.1016/j.nicl.2017.08.017] [PMID: 29034163]
[http://dx.doi.org/10.1109/IJCNN.2019.8852002]
[http://dx.doi.org/10.3389/fnins.2019.01325] [PMID: 32009868]
[http://dx.doi.org/10.1016/j.clinph.2020.11.037] [PMID: 33450566]
[http://dx.doi.org/10.3389/fnins.2023.1132231] [PMID: 36968494]
[http://dx.doi.org/10.1016/j.compbiomed.2021.104963] [PMID: 34700253]
[http://dx.doi.org/10.3389/fnins.2022.844851] [PMID: 35937896]
[PMID: 4880460]
[http://dx.doi.org/10.1016/j.jaac.2013.10.013] [PMID: 24472258]
[http://dx.doi.org/10.3389/fnins.2023.1101071] [PMID: 37694110]
[http://dx.doi.org/10.3389/fpsyt.2022.970611] [PMID: 36440386]
[http://dx.doi.org/10.1017/S003329171400172X] [PMID: 25108395]
[http://dx.doi.org/10.15585/mmwr.ss6904a1] [PMID: 32214087]
[http://dx.doi.org/10.1542/peds.2019-3447]
[http://dx.doi.org/10.1016/j.chc.2017.02.008] [PMID: 28577609]
[PMID: 33090020]
[http://dx.doi.org/10.3389/fnmol.2022.922840] [PMID: 35726297]
[http://dx.doi.org/10.1378/chest.13-1940] [PMID: 24297139]
[http://dx.doi.org/10.15585/mmwr.ss7011a1] [PMID: 34855725]
[http://dx.doi.org/10.3389/fnins.2022.1046268] [PMID: 36483179]
[http://dx.doi.org/10.1352/1944-7558-117.6.478] [PMID: 23167487]
[http://dx.doi.org/10.1007/s12098-015-1894-0] [PMID: 26411730]
[http://dx.doi.org/10.3389/fnins.2021.828512] [PMID: 35185454]
[http://dx.doi.org/10.1016/S0140-6736(18)31129-2] [PMID: 30078460]
[http://dx.doi.org/10.3389/fpsyg.2021.667359] [PMID: 34335378]
[http://dx.doi.org/10.1007/s10803-005-0050-5] [PMID: 16477517]
[http://dx.doi.org/10.1007/s10803-012-1695-5] [PMID: 23114566]
[http://dx.doi.org/10.1017/S0954579414000479] [PMID: 24915526]
[http://dx.doi.org/10.1016/j.rasd.2016.08.004]
[http://dx.doi.org/10.1016/j.neubiorev.2011.12.008] [PMID: 22212588]
[http://dx.doi.org/10.1037/xge0000550] [PMID: 30652891]
[http://dx.doi.org/10.1037/bul0000310] [PMID: 33104376]
[http://dx.doi.org/10.3389/fped.2021.598276] [PMID: 34604128]
[http://dx.doi.org/10.1111/j.1469-8749.2008.03242.x] [PMID: 19207298]
[http://dx.doi.org/10.1016/j.rasd.2013.07.002]
[http://dx.doi.org/10.1002/aur.2230] [PMID: 31625694]
[http://dx.doi.org/10.1111/j.1469-7610.2006.01531.x] [PMID: 16712640]
[http://dx.doi.org/10.1177/1362361311402230] [PMID: 21610184]
[http://dx.doi.org/10.3389/fnins.2021.756868] [PMID: 34712116]
[http://dx.doi.org/10.1176/ajp.152.8.1228]
[http://dx.doi.org/10.3389/fninf.2021.635657] [PMID: 34248531]
[http://dx.doi.org/10.1177/1362361316646559] [PMID: 27231337]
[http://dx.doi.org/10.1038/ncomms11254] [PMID: 27075704]
[http://dx.doi.org/10.1109/RBME.2018.2886237] [PMID: 30561351]
[http://dx.doi.org/10.1007/s10803-006-0314-8] [PMID: 17160456]
[http://dx.doi.org/10.3109/08039488.2012.748824] [PMID: 23234539]
[http://dx.doi.org/10.3389/fncom.2021.654315] [PMID: 33897398]
[http://dx.doi.org/10.1038/s41586-020-2314-9] [PMID: 32483374]
[http://dx.doi.org/10.1016/j.cortex.2014.08.011] [PMID: 25243989]
[http://dx.doi.org/10.1016/j.nicl.2014.12.013] [PMID: 25685703]
[http://dx.doi.org/10.3389/fnins.2018.00491] [PMID: 30087587]
[http://dx.doi.org/10.1007/s11042-018-5625-1]
[http://dx.doi.org/10.1109/ACCESS.2019.2936639]
[http://dx.doi.org/10.1038/s41598-019-40427-7] [PMID: 30846796]
[http://dx.doi.org/10.1093/cercor/bhl006] [PMID: 16772313]
[http://dx.doi.org/10.1002/mp.14692] [PMID: 33378589]
[http://dx.doi.org/10.1515/revneuro-2020-0043] [PMID: 32866134]
[http://dx.doi.org/10.1109/TCYB.2014.2379621] [PMID: 25576588]
[http://dx.doi.org/10.1109/TNNLS.2020.3007943] [PMID: 32692687]
[http://dx.doi.org/10.1109/BIBM.2018.8621472]
[http://dx.doi.org/10.1109/InC457730.2023.10262873]
[http://dx.doi.org/10.1016/j.eswa.2020.113513]
[http://dx.doi.org/10.1155/2020/1357853]
[http://dx.doi.org/10.1109/JBHI.2020.2998603] [PMID: 32750917]
[http://dx.doi.org/10.1016/j.nicl.2020.102181] [PMID: 31982680]
[http://dx.doi.org/10.1016/j.neuroimage.2020.117012] [PMID: 32526386]
[http://dx.doi.org/10.1016/j.neunet.2020.03.017] [PMID: 32259762]
[http://dx.doi.org/10.1016/j.jneumeth.2019.108538] [PMID: 31794776]
[http://dx.doi.org/10.3389/fpsyt.2020.00440] [PMID: 32477198]
[http://dx.doi.org/10.1016/j.icte.2022.08.004]
[http://dx.doi.org/10.1007/s00521-020-05514-1]
[http://dx.doi.org/10.1002/int.22586]
[http://dx.doi.org/10.1007/s42979-023-02439-z]
![Wayfinder Image](/images/wayfinder.jpg)