Citation

Ehsan, Khafsa and Sultan, Kashif and Fatima, Abreen and Sheraz, Muhammad and Chuah, Teong Chee (2025) Early Detection of Autism Spectrum Disorder Through Automated Machine Learning. Diagnostics, 15 (15). p. 1859. ISSN 2075-4418

Text diagnostics-15-01859.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Autism spectrum disorder (ASD) is a neurodevelopmental disorder distinguished by an extensive range of symptoms, including reduced social interaction, communication difficulties and tiresome behaviors. Early detection of ASD is important because it allows for timely intervention, which significantly improves developmental, behavioral, and communicative outcomes in children. However, traditional diagnostic procedures for identifying autism spectrum disorder (ASD) typically involve lengthy clinical examinations, which can be both time-consuming and costly. This research proposes leveraging automated machine learning (AUTOML) to streamline the diagnostic process and enhance its accuracy. Methods: In this study, by collecting data from various rehabilitation centers across Pakistan, we applied a specific AUTOML tool known as Tree-based Pipeline Optimization Tool (TPOT) for ASD detection. Notably, this study marks one of the initial explorations into utilizing AUTOML for ASD detection. The experimentations indicate that the TPOT provided the best pipeline for the dataset, which was verified using a manual machine learning method. Results: The study contributes to the field of ASD diagnosis by using AUTOML to determine the likelihood of ASD in children at prompt stages of evolution. The study also provides an evaluation of precision, recall, and F1-score metrics to confirm the correctness of the diagnosis. The propose TPOT-based AUTOML framework attained an overall accuracy 78%, with a precision of 83%, a recall of 90%, and an F1-score of 86% for the autistic class. Conclusions: In summary, this research offers an encouraging approach to improve the detection of autism spectrum disorders (ASD) in children, which could lead to better results for affected individuals and their families.

Item Type:	Article
Uncontrolled Keywords:	Autism spectrum disorder, AUTOML, TPOT
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Engineering (FOE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	27 Aug 2025 03:48
Last Modified:	04 Sep 2025 03:58
URII:	http://shdl.mmu.edu.my/id/eprint/14433

Downloads

Edit (login required)