Khan et al. Res. Trends Int. J. Technol. Innov., July - September 2026, 1 (3) : 50-57
1Department of Biomedical Engineering, Manipal Institute of Technology, Manipal, India; 2Department of Computer Science and Engineering, Indian Institute of Technology Delhi, New Delhi, India
Article History
Accepted : 03 Jul 2026
Published : 10 Jul 2026
Publication Issue
Volume 1, Issue 3
July - September 2026
Page Number50–57
Chest X-ray interpretation for pneumonia screening in resource-limited settings is often bottlenecked by radiologist availability. This paper adapts a compact vision transformer, MobileViT, for binary pneumonia classification and evaluates it on a combined dataset of 12,800 paediatric and adult chest radiographs. The model achieved a sensitivity of 95.6 percent and specificity of 91.3 percent at the operating threshold selected to prioritise recall, with an inference time of 41 milliseconds per image on a standard laptop CPU, supporting deployment in low-resource radiology workflows.
Keywords - pneumonia detection, vision transformer, chest X-ray, medical imaging, MobileViT
Pneumonia remains a leading cause of childhood mortality in regions with limited access to radiologists, and automated triage of chest radiographs can help prioritise cases for urgent specialist review where reporting turnaround is otherwise measured in days.
A MobileViT-XS architecture, combining lightweight convolutional stages with transformer blocks, was fine-tuned on 12,800 frontal chest radiographs drawn from combined paediatric and adult public datasets, labelled as pneumonia-positive or normal, with the classification threshold selected via a precision-recall curve to prioritise sensitivity given the screening use case.
At the selected operating threshold, the model achieved 95.6 percent sensitivity and 91.3 percent specificity on a held-out test set, with an AUC of 0.96. Inference averaged 41 milliseconds per image on a standard laptop CPU without GPU acceleration, and the full model occupied 9.7MB, supporting deployment on modest clinic hardware.
A compact vision transformer architecture can deliver radiologist-comparable screening sensitivity for pneumonia while running efficiently on non-specialist hardware. Future work will evaluate performance across a wider range of X-ray equipment vendors to assess generalisation.
[1] Mehta S. and Rastegari M., MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer, ICLR, 2022. [2] Rajpurkar P. et al., CheXNet: Radiologist-level pneumonia detection, arXiv, 2017.
© 2026 The Author(s). Published by IJEIA Editorial Office. This is an open access article under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
Fatima Zahra Khan, Rajat Bindal (2026). Automated Detection of Pneumonia from Chest X-Rays Using a Lightweight Vision Transformer. IJEIA, 1(3), 50-57.
Nikhil Deshpande, Swathi Reddy
Emerging Engineering TechnologiesAyesha Siddiqui, Rohit Bhatnagar