Article: Explainable ensemble machine learning model for autism identification using EEG and optimised feature selection Journal: International Journal of Biomedical Engineering and Technology (IJBET) 2025 Vol.49 No.3 pp.202 - 232 Abstract: Autism spectrum disorder (ASD) is a complex neurodevelopmental condition that remains challenging to diagnose due to its reliance on subjective clinical assessments. To address this limitation, an automated electroencephalography (EEG)-based ASD identification framework is proposed, aiming to enhance identification accuracy. Temporal dynamics of EEG signals analysed through three-phase feature selection approach involving independent samples t-test elimination, followed by a novel binary search-driven mutual information (BiS-MI) and recursive feature elimination (BiS-RFE). Six ensemble models trained on the selected features, and feature importance and classification predictions interpreted using explainable artificial intelligence (XAI). CatBoost achieved the highest performance, with an accuracy of 0.9923 and recall of 0.9864 using BiS-MI, and an accuracy of 0.9936 and recall of 0.9911 using BiS-RFE. SHAP analysis identified features from frontal and central EEG electrodes as the most significant contributors. These results highlight the potential for developing interpretable, non-invasive and improved diagnostic tools for ASD identification. Inderscience Publishers - linking academia, business and industry through research

Title: Explainable ensemble machine learning model for autism identification using EEG and optimised feature selection

Authors: Anamika Ranaut; Padmavati Khandnor; Trilok Chand

Addresses: Department of Computer Science and Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, India ' Department of Computer Science and Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, India ' Department of Computer Science and Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, India

Abstract: Autism spectrum disorder (ASD) is a complex neurodevelopmental condition that remains challenging to diagnose due to its reliance on subjective clinical assessments. To address this limitation, an automated electroencephalography (EEG)-based ASD identification framework is proposed, aiming to enhance identification accuracy. Temporal dynamics of EEG signals analysed through three-phase feature selection approach involving independent samples t-test elimination, followed by a novel binary search-driven mutual information (BiS-MI) and recursive feature elimination (BiS-RFE). Six ensemble models trained on the selected features, and feature importance and classification predictions interpreted using explainable artificial intelligence (XAI). CatBoost achieved the highest performance, with an accuracy of 0.9923 and recall of 0.9864 using BiS-MI, and an accuracy of 0.9936 and recall of 0.9911 using BiS-RFE. SHAP analysis identified features from frontal and central EEG electrodes as the most significant contributors. These results highlight the potential for developing interpretable, non-invasive and improved diagnostic tools for ASD identification.

Keywords: autism; electroencephalography; ensemble models; feature selection; feature extraction; mutual information; recursive feature elimination; statistical test.

DOI: 10.1504/IJBET.2025.150082

International Journal of Biomedical Engineering and Technology, 2025 Vol.49 No.3, pp.202 - 232

Received: 27 Apr 2025
Accepted: 07 Jul 2025
Published online: 28 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Explainable ensemble machine learning model for autism identification using EEG and optimised feature selection

Keep up-to-date