AI For Early Detection of ASD: A VGG16-Based Child Facial Image Classification System
Applied Computer Science and Software Engineering · 28.03.2026 · 2 (1)
DOI: 10.63254/acsse.v2i1.31 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Computer Science and Software Engineering, online erschienen am 28.03.2026. Verfasst von Aftina Zahrayani und 1 weiteren Autor:innen.
Abstract (Originalsprache)
This study develops an Autism Spectrum Disorder (ASD) classification system based on children’s facial images using a Convolutional Neural Network (CNN) with VGG16 transfer learning. The model was trained using the Piosenka Autism Image Dataset, which was divided into training, validation, and testing sets. Experimental results show that the model achieved an accuracy of 84%, with balanced precision, recall, and F1-score across both classes. Performance evaluation using a confusion matrix and ROC–AUC indicates good discriminative capability. The model was also implemented in a Streamlit-based application, demonstrating its potential for early ASD detection support.
Zitieren
APA
Zahrayani, A., & Tikaningsih, A. (2026). AI For Early Detection of ASD: A VGG16-Based Child Facial Image Classification System. Applied Computer Science and Software Engineering, 2(1). https://doi.org/10.63254/acsse.v2i1.31
Vancouver
Zahrayani A, Tikaningsih A. AI For Early Detection of ASD: A VGG16-Based Child Facial Image Classification System. Applied Computer Science and Software Engineering. 2026;2(1). doi:10.63254/acsse.v2i1.31
BibTeX
@article{Zahrayani2026,
author = {Zahrayani, Aftina and Tikaningsih, Ades},
title = {AI For Early Detection of ASD: A VGG16-Based Child Facial Image Classification System},
journal = {Applied Computer Science and Software Engineering},
year = {2026},
volume = {2},
number = {1},
doi = {10.63254/acsse.v2i1.31}
}
EUVIMED