{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/asd-diagnet-a-hybrid-learning-approach-for","title":"ASD-DiagNet: A hybrid learning approach for detection of Autism Spectrum Disorder using fMRI data","arxiv_id":"1904.07577","date":"2019-04-16","proceeding":null,"authors":["Taban Eslami","Vahid Mirjalili","Alvis Fong","Angela Laird","Fahad Saeed"],"abstract":"Mental disorders such as Autism Spectrum Disorders (ASD) are heterogeneous\ndisorders that are notoriously difficult to diagnose, especially in children.\nThe current psychiatric diagnostic process is based purely on the behavioural\nobservation of symptomology (DSM-5/ICD-10) and may be prone to over-prescribing\nof drugs due to misdiagnosis. In order to move the field towards more\nquantitative fashion, we need advanced and scalable machine learning\ninfrastructure that will allow us to identify reliable biomarkers of mental\nhealth disorders. In this paper, we propose a framework called ASD-DiagNet for\nclassifying subjects with ASD from healthy subjects by using only fMRI data. We\ndesigned and implemented a joint learning procedure using an autoencoder and a\nsingle layer perceptron which results in improved quality of extracted features\nand optimized parameters for the model. Further, we designed and implemented a\ndata augmentation strategy, based on linear interpolation on available feature\nvectors, that allows us to produce synthetic datasets needed for training of\nmachine learning models. The proposed approach is evaluated on a public dataset\nprovided by Autism Brain Imaging Data Exchange including 1035 subjects coming\nfrom 17 different brain imaging centers. Our machine learning model outperforms\nother state of the art methods from 13 imaging centers with increase in\nclassification accuracy up to 20% with maximum accuracy of 80%. The machine\nlearning technique presented in this paper, in addition to yielding better\nquality, gives enormous advantages in terms of execution time (40 minutes vs. 6\nhours on other methods). The implemented code is available as GPL license on\nGitHub portal of our lab (https://github.com/pcdslab/ASD-DiagNet).","url_abs":"http://arxiv.org/abs/1904.07577v1","url_pdf":"http://arxiv.org/pdf/1904.07577v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"asd-diagnet-a-hybrid-learning-approach-for","repo_url":"https://github.com/pcdslab/ASD-DiagNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"diagnostic","task_name":"Diagnostic"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}