Explainable Multimodal Deep Learning Framework for Tuberculosis Detection Using Chest X-Ray Images and Clinical Symptoms
International Journal of Drug Delivery Technology · 04.06.2026 · 16 (45s)
DOI: 10.25258/ijddt.16.45s.85 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Drug Delivery Technology, online erschienen am 04.06.2026. Verfasst von Miki Patel und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Tuberculosis (TB), a critical global health issue, is still a problem in developing countries. Early diagnosis is essential to effectively treat TB and control the spread of disease. Most of the available AI-based systems have focused on using only chest X-rays for TB diagnosis and have failed to account for other clinical factors that are needed when diagnosing TB in the real world. To overcome this limitation, we present an explainable multimodal deep learning model that fuses the radiological data from chest X-rays with the clinical data of other clinical symptoms to help improve the ability for TB diagnosis. To extract features from the chest X-ray images, we will use a fine-tuned ResNet50 model and will use an XGBoost classifier to evaluate the clinical attributes of the clinical symptoms of fever, cough, chest pain, night sweats, fatigue, and weight loss. The features will be fused and processed through an MLP to produce the final prediction. The use of Grad-CAM and SHAP provides the explainability for this model. The experimental results of the proposed multimodal deep learning model will produce more balanced and reliable predictions when compared to unimodal approaches, thus providing more robustness and real-world applicability of TB diagnosis systems.
Zitieren
APA
Patel, M., & Bhadla, M. (2026). Explainable Multimodal Deep Learning Framework for Tuberculosis Detection Using Chest X-Ray Images and Clinical Symptoms. International Journal of Drug Delivery Technology, 16(45s). https://doi.org/10.25258/ijddt.16.45s.85
Vancouver
Patel M, Bhadla M. Explainable Multimodal Deep Learning Framework for Tuberculosis Detection Using Chest X-Ray Images and Clinical Symptoms. International Journal of Drug Delivery Technology. 2026;16(45s). doi:10.25258/ijddt.16.45s.85
BibTeX
@article{Patel2026,
author = {Patel, Miki and Bhadla, Mohit},
title = {Explainable Multimodal Deep Learning Framework for Tuberculosis Detection Using Chest X-Ray Images and Clinical Symptoms},
journal = {International Journal of Drug Delivery Technology},
year = {2026},
volume = {16},
number = {45s},
doi = {10.25258/ijddt.16.45s.85}
}
EUVIMED