An Optimized Lightweight CNN Architecture for Automated Pneumonia Classification from Chest X-Ray Images

DOI

10.2991/978-94-6239-789-7_49How to use a DOI?

Keywords

Pneumonia Detection; Convolutional Neural Network (CNN); Chest X-ray Image; Deep learning; Medical Image analysis

Abstract

Chest X-ray images are usually interpreted by expert radiologists for the diagnosis of pneumonia, which is a time-consuming and subjective process. We propose a light-weight and computationally efficient convolutional neural network (CNN) for automated pneumonia detection to deal with these challenges. The goal is to achieve high diagnostic accuracy and low computational requirements. This makes the approach suitable for practical use. The dataset of the proposed framework includes 5,863 chest X-ray images of pneumonia and normal cases. Before training, the images are processed with standardization and enhancement techniques and data augmentation methods to better equip the model to generalize over diverse samples. The CNN model is trained on these prepared inputs and is designed to balance performance and efficiency. The experimental results show that the model can reach an accuracy of 95.8% and has a strong ability to distinguish between pneumonia and healthy cases. Such a performance indicates that the model could be used for a reliable and prompt diagnosis. Overall, the study provides an AI based solution which can help healthcare professionals in early detection of pneumonia. The proposed approach is lightweight, making it especially useful in scenarios with limited computational resources, which can enable wider adoption of intelligent diagnostic systems.

Copyright

© 2026 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  - Princi Saini
AU  - Mridula Singh
PY  - 2026
DA  - 2026/10/08
TI  - An Optimized Lightweight CNN Architecture for Automated Pneumonia Classification from Chest X-Ray Images
BT  - Proceedings of theInternational Conference on Innovation and Practices in Artificial Intelligence (ICIPAI 2026)
PB  - Atlantis Press
SP  - 730
EP  - 746
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-789-7_49
DO  - 10.2991/978-94-6239-789-7_49
ID  - Saini2026
ER  -