Deep Learning Based Severity Prediction of Autism Spectrum Disorder Through Face Image Detection

DOI

10.2991/978-94-6463-738-0_36How to use a DOI?

Keywords

Autisum Spectrum Disorder (ASD); Deep Learning; Convolutional Neural Networks (CNNs); Transfer Learning; Xception Model; VGG16 Model

Abstract

Most screening tools for Autisum Spectrum disorder are questionnaire-based and rely on subjective responses by caregivers. While behavioral observation provides an objective approach, these usually come quite pricey, very time-consuming, and technically demanding. That is why there is a great demand for systems that are efficient, scale appropriately, and reliably identify the risk behaviors related to ASD. Even though the causes of autism have yet to be discovered, early detection and intervention have become good solutions that can make a huge difference in the behavior of ASD individuals. Recent advancements in AI actually found this opportunity for early detection that could change lives. ASD, primarily a neurodevelopmental disorder, is related to brain development and can be observed or detected in images of biological nature, especially in facial aspects. In this research paper, the use of convolutional neural networks with transfer learning is proposed for classifying ASD from facial images. This was done through the Xception and VGG16 pretrained models to classify an image. Utilizing the online-based platform Kaggle, 2,940 facial images were tested with these models based on performance metrics, such as accuracy, sensitivity, and specificity standards. It is then followed by VGG16 with accuracy of 75%, and Xception scored 98%. The result obtained from these experiments shows that deep learning really has the potential to enhance accuracy in the recognition of ASD from facial features.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - J. Navinkumar
AU  - S. Madalvizhi
AU  - C. Manimegalai
AU  - K. Kalaiselvi
PY  - 2025
DA  - 2025/06/22
TI  - Deep Learning Based Severity Prediction of Autism Spectrum Disorder Through Face Image Detection
BT  - Proceedings of the International Conference on Advances and Applications in Artificial Intelligence (ICAAAI 2025)
PB  - Atlantis Press
SP  - 445
EP  - 457
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-738-0_36
DO  - 10.2991/978-94-6463-738-0_36
ID  - Navinkumar2025
ER  -