Interpretable Deep Learning Framework for Chest X-Ray Classification of Pneumonia and Lung Abnormalities

DOI

10.2991/978-94-6239-674-6_12How to use a DOI?

Keywords

Chest X-ray classification; pneumonia detection; deep learning; explainable AI; Grad-CAM; SHAP; Integrated Gradients; model interpretability; medical image analysis; pulmonary abnormalities

Abstract

Pneumonia and other lung abnormalities continue to be significant health issues of global concern and any diagnostic support to aid clinical decision-making must be quick and precise. The Chest X-ray imaging has been extensively utilized in respiratory assessment, but the manual interpretation is time-consuming and is likely to be intra-clinician variability. This research paper suggests a deep learning model that can be interpreted to classify pneumonia and related pulmonary abnormalities using chest radiographs. The model combines a convolutional neural network framework that is optimized to extract features with explainable AI algorithms like Grad-CAM, SHAP, and Integrated Gradients to visualize and confirm the behavior of the model. The interpretability tools underscore meaningful lung areas that lead to predictions and, therefore, make sure that such decisions are based on clinical significance, and not on arbitrary trends. High accuracy, sensitivity, and specificity have been shown by experimental evaluation in the ability to differentiate between normal and pathological images and explainability results indicate consistency with radiologically significant appearances like opacities and consolidation patterns. Model reliability is enhanced by the inclusion of transparent decision mechanisms and thus its use as a reliable diagnostic tool. All in all, the framework facilitates the gap between performance and interpretability, which leads to safer and more responsible AI-based medical imaging systems to detect respiratory diseases.

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  - Anjul Singh
AU  - Akanksha Kapoor
AU  - Shilpa Gupta
AU  - Kumud Dixit
AU  - Sujeet Kumar
PY  - 2026
DA  - 2026/05/28
TI  - Interpretable Deep Learning Framework for Chest X-Ray Classification of Pneumonia and Lung Abnormalities
BT  - Proceedings of the International Conference on Sustainable Computing and Artificial Intelligence (ICSCAI 2025)
PB  - Atlantis Press
SP  - 129
EP  - 142
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-674-6_12
DO  - 10.2991/978-94-6239-674-6_12
ID  - Singh2026
ER  -