{"title":"fMRI-Based Multi-class DMDC Model Efficiently Decodes the Overlaps between ASD and ADHD.","authors":"Zahra Zolghadr, Seyed Amir Hossein Batouli, Hamid Alavi Majd, Lida Shafaghi, Yadollah Mehrabi","doi":"10.32598/bcn.2023.4302.1","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>Neurodevelopmental disorders comprise a group of neuropsychiatric conditions. Presently, behavior-based diagnostic approaches are utilized in clinical settings, but the overlapping features among these disorders obscure their recognition and management. Attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) have common characteristics across various levels, from genes to symptoms. Designing a computational framework based on the neuroimaging findings could provide a discriminative tool for ultimate more efficient treatment. Machine learning approaches, specifically classification methods are among the most applied techniques to reach this goal.</p><p><strong>Methods: </strong>We applied a novel two-level multi-class data maximum dispersion classifier (DMDC) algorithm to classify the functional neuroimaging data (utilizing datasets: ADHD-200 and autism brain imaging data exchange (ABIDE)) into two categories: Neurodevelopmental disorders (ASD and ADHD) or healthy participants, based on calculated functional connectivity values (statistical temporal correlation).</p><p><strong>Results: </strong>Our model achieved a total accuracy of 62% for healthy controls. Specifically, it demonstrated an accuracy of 51% for healthy subjects, 61% for autism spectrum disorder, and 84% for ADHD. The support vector machine (SVM) model achieved an accuracy of 46% for both the healthy control and ASD groups, while the ADHD group classification accuracy was estimated to be 84%. These two models showed similar classification indices for the ADHD group. However, the discrimination power was higher in the ASD class.</p><p><strong>Conclusion: </strong>The method employed in this study demonstrated acceptable performance in classifying disorders and healthy conditions compared to the more commonly used SVM method. Notably, functional connections associated with the cerebellum showed discriminative power.</p>","PeriodicalId":8701,"journal":{"name":"Basic and Clinical Neuroscience","volume":null,"pages":null},"PeriodicalIF":1.0000,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11470894/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Basic and Clinical Neuroscience","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.32598/bcn.2023.4302.1","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"NEUROSCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Introduction: Neurodevelopmental disorders comprise a group of neuropsychiatric conditions. Presently, behavior-based diagnostic approaches are utilized in clinical settings, but the overlapping features among these disorders obscure their recognition and management. Attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) have common characteristics across various levels, from genes to symptoms. Designing a computational framework based on the neuroimaging findings could provide a discriminative tool for ultimate more efficient treatment. Machine learning approaches, specifically classification methods are among the most applied techniques to reach this goal.
Methods: We applied a novel two-level multi-class data maximum dispersion classifier (DMDC) algorithm to classify the functional neuroimaging data (utilizing datasets: ADHD-200 and autism brain imaging data exchange (ABIDE)) into two categories: Neurodevelopmental disorders (ASD and ADHD) or healthy participants, based on calculated functional connectivity values (statistical temporal correlation).
Results: Our model achieved a total accuracy of 62% for healthy controls. Specifically, it demonstrated an accuracy of 51% for healthy subjects, 61% for autism spectrum disorder, and 84% for ADHD. The support vector machine (SVM) model achieved an accuracy of 46% for both the healthy control and ASD groups, while the ADHD group classification accuracy was estimated to be 84%. These two models showed similar classification indices for the ADHD group. However, the discrimination power was higher in the ASD class.
Conclusion: The method employed in this study demonstrated acceptable performance in classifying disorders and healthy conditions compared to the more commonly used SVM method. Notably, functional connections associated with the cerebellum showed discriminative power.
期刊介绍:
BCN is an international multidisciplinary journal that publishes editorials, original full-length research articles, short communications, reviews, methodological papers, commentaries, perspectives and “news and reports” in the broad fields of developmental, molecular, cellular, system, computational, behavioral, cognitive, and clinical neuroscience. No area in the neural related sciences is excluded from consideration, although priority is given to studies that provide applied insights into the functioning of the nervous system. BCN aims to advance our understanding of organization and function of the nervous system in health and disease, thereby improving the diagnosis and treatment of neural-related disorders. Manuscripts submitted to BCN should describe novel results generated by experiments that were guided by clearly defined aims or hypotheses. BCN aims to provide serious ties in interdisciplinary communication, accessibility to a broad readership inside Iran and the region and also in all other international academic sites, effective peer review process, and independence from all possible non-scientific interests. BCN also tries to empower national, regional and international collaborative networks in the field of neuroscience in Iran, Middle East, Central Asia and North Africa and to be the voice of the Iranian and regional neuroscience community in the world of neuroscientists. In this way, the journal encourages submission of editorials, review papers, commentaries, methodological notes and perspectives that address this scope.