Citation

J. Monani, Udita and Maity, Ritu and Pattnaik, Prasant Kumar and Anbananthen, Kalaiarasi Sonai Muthu and Muthaiyah, Saravanan and Sain, Mangal (2026) Evaluating Deep Learning Models for Autism Detection in Children Using Facial Images. Journal of Human, Earth, and Future, 7 (1). pp. 48-60. ISSN 2785-2997

Text 3 (1).pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

This study develops and evaluates a comprehensive deep-learning framework for early detection of Autism Spectrum Disorder (ASD) through facial image analysis. Five state-of-the-art convolutional neural network (CNN) architectures, VGG16, VGG19, ResNet50, InceptionV3, and MobileNet, were systematically assessed using a balanced dataset of 5,000 images (2,500 ASD, 2,500 non-ASD). Transfer learning and data augmentation enhanced model generalization. VGG19 achieved the highest overall accuracy (77.89%) and F1-score (0.7962), ResNet50 attained the best precision (82.53%), and InceptionV3 produced the highest recall (99.67%), indicating strong screening potential. The findings confirm that deep CNNs can capture subtle facial morphological cues linked to ASD, supporting their feasibility as noninvasive diagnostic tools. This work provides a benchmark for future multimodal, explainable, and clinically validated AI systems for autism detection.

Item Type:	Article
Uncontrolled Keywords:	Autism spectrum disorder, deep learning
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	04 May 2026 04:38
Last Modified:	04 May 2026 04:38
URII:	http://shdl.mmu.edu.my/id/eprint/15868

Downloads

Edit (login required)