Down syndrome Detection in Children With Deep Learning and Using Multi-Model

DOI

10.2991/978-94-6463-858-5_240How to use a DOI?

Keywords

Down syndrome; deep learning; Convolutional Neural Networks; multi-modal detection; VGG16; facial image analysis; early diagnosis; machine learning; automated detection; healthcare technology

Abstract

One out of every 700 live newborns worldwide is affected by Down syndrome (DS), a genetic disease brought on by an extra copy of chromosome 21. For impacted children to receive the right interventions and have their quality of life improved, early diagnosis is crucial. Traditional techniques for identifying Down syndrome, such as noninvasive prenatal testing and karyotyping, can be expensive, time-consuming, and interpretatively complex. On the other hand, new developments in deep learning, namely in the area of image analysis, have demonstrated potential for automating the detection of Down syndrome. This paper offers a novel method for identifying children with Down syndrome from facial photos using multimodal deep learning, specifically Convolutional Neural Networks (CNNs). The dataset used comprises 2,100 training images, 899 validation images, and 40 test images from Kaggle, with two classes: Down syndrome and non-Down syndrome. Three CNN-based architectures—InceptionV3 (81.85% accuracy), ResNet (60.85% accuracy), and VGG16—were compared. Among these, VGG16 demonstrated the highest performance with an accuracy of 83.95%. A user-friendly web interface was also developed, allowing users to upload images and receive instant diagnostic results. This multi-modal deep learning approach offers a cost-effective, efficient, and accessible solution for early Down syndrome detection, which could enhance diagnostic practices and early intervention outcomes.

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. Kishore
AU  - Md. Sadiya
AU  - N. Maheshwari
AU  - N. Nikhitha
PY  - 2025
DA  - 2025/11/04
TI  - Down syndrome Detection in Children With Deep Learning and Using Multi-Model
BT  - Proceedings of International Conference on Computer Science and Communication Engineering (ICCSCE 2025)
PB  - Atlantis Press
SP  - 2854
EP  - 2868
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-858-5_240
DO  - 10.2991/978-94-6463-858-5_240
ID  - Kishore2025
ER  -