Explainable AI Based Fully Fine-Tuned Data-efficient Image Transformer (DeiT-B) Model for Multi Class Chest X-Ray Image Classification

DOI

10.2991/978-94-6239-664-7_5How to use a DOI?

Keywords

Chest X-ray; Data-efficient Image Transformer; Explainable AI; LIME; Deep Learning

Abstract

Chest X-ray imaging is utilized significantly in the diagnosis of respiratory disorders like COVID-19, viral pneumonia, and lung opacities. Deep learning has evolved computerized classification systems that are able to assist radiologists in making more accurate and rapid diagnoses. In this paper, we propose a fully fine-tuned Data-efficient Image Transformer (DeiT-B) model with Explainable AI (XAI) techniques, including LIME and Attention maps, for chest X-ray image classification. The method leverages DeiT-B’s attention mechanism to focus on relevant regions of the X-ray images and provide visual explanations of its predictions. The model was trained and tested on 4,800 chest X-ray images from a Kaggle dataset. Experimental outcomes demonstrate that the model achieves a test accuracy of 95.21%, its weighted precision, recall, and F1-score values of 95.46%, 95.21%, and 95.26%, and its Cohen’s Kappa value is 0.9361, superior to baseline CNN models such as VGG16, ResNet18, ResNet50, EfficientNetB3, DenseNet121, and even compared with transformer-based models such as ViT-B/16. XAI integration, in the sense of LIME and Attention maps, ensures interpretability and reliability of the model, thereby making the model suitable for real-world clinical application.

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  - Md Parvez Kabir
AU  - Md Jahidul Islam Mozumdar
AU  - Md Rezaul
AU  - Rasedul Islam
AU  - Sourav Ghosh
AU  - Md Toha Hayder
PY  - 2026
DA  - 2026/06/08
TI  - Explainable AI Based Fully Fine-Tuned Data-efficient Image Transformer (DeiT-B) Model for Multi Class Chest X-Ray Image Classification
BT  - Proceedings of the International Conference on Intelligent Data Analysis and Applications (IDAA 2025)
PB  - Atlantis Press
SP  - 50
EP  - 61
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-664-7_5
DO  - 10.2991/978-94-6239-664-7_5
ID  - Kabir2026
ER  -