Automatic classification of autism spectrum disorder in children using cortical thickness and support vector machine

Abstract

Objective: Autism spectrum disorder (ASD) is a neurodevelopmental condition with a heterogeneous phenotype. The role of biomarkers in ASD diagnosis has been highlighted; cortical thickness has proved to be involved in the etiopathogenesis of ASD core symptoms. We apply support vector machine, a supervised machine learning method, in order to identify specific cortical thickness alterations in ASD subjects.Methods: A sample of 76 subjects (9.5 ± 3.4 years old) has been selected, 40 diagnosed with ASD and 36 typically developed subjects. All children underwent a magnetic resonance imaging (MRI) examination; T1-MPRAGE sequences were analyzed to extract features for the characterization and parcellation of regions of interests (ROI); average cortical thickness (CT) has been measured for each ROI. For the classification process, the extracted features were used as input for a classifier to identify ASD subjects through a "learning by example" procedure; the features with best performance was then selected by "greedy forward-feature selection." Finally, this model underwent a leave-one-out cross-validation approach.Results: From the training set of 68 ROIs, five ROIs reached accuracies of over 70%.After this phase, we used a recursive feature selection process in order to identify the eight features with the best accuracy (84.2%). CT resulted higher in ASD compared to controls in all the ROIs identified at the end of the process. Conclusion:We found increased CT in various brain regions in ASD subjects, confirming their role in the pathogenesis of this condition. Considering the brain development curve during ages, these changes in CT may normalize during development. Further validation on a larger sample is required.

Brain volumetric and white matter structural connectivity alterations in autistic children: case–control study

“…But Squarcina et al found that CT increased in various brain regions in ASD subjects, confirming their role in the pathogenesis of this condition. Considering the brain development curve during ages, these changes in CT may normalize during development [46].…”

Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

“…Existing methodologies for reducing the dimensionality of features include feature extraction, feature selection, and sparse learning methods. F-score ( Devika and Oruganti, 2020 ; Fu et al, 2021 ), Recursive Feature Elimination ( Ali et al, 2021 ; Fu et al, 2021 ), PCA ( Irimia et al, 2018 ), greedy selection ( Squarcina et al, 2021 ), and AEs ( Sen et al, 2018 ) are examples of feature selection techniques.…”

“…Several researchers have developed ML algorithms employing sMRI ( Moradi et al, 2017 ) or multimodal data ( Eill et al, 2019 ) to diagnose ASD or uncover novel biomarkers for it ( Table 2 , summarizing all ML-based studies). SVM has been extensively evaluated, with ACCs ranging from 45 to 94% on various ASD datasets ( Morris and Rekik, 2017 ; Squarcina et al, 2021 ). In addition, various additional conventional ML models, such as NB ( Xiao et al, 2017 ), KNN ( Yassin et al, 2020 ), AdaBoost ( Bilgen et al, 2020 ), and RF ( Devika and Oruganti, 2020 ), have been examined.…”

Can Autism Be Diagnosed with Artificial Intelligence? A Narrative Review

“…Conventional method: The most popular is the SVM. It is in many neuroimaging tasks Chen et al [2016], Squarcina et al [2021], Jahedi [2020]. However, SVM is not recommended when the samples less than the features number due to the overfitting.…”