Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198417,
        author = {Rushikesh Londhe and Nishant Mahajan and Piyush Marathe and Anand Ingle},
        title = {Tuberculosis Detection using Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {13210-13215},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198417},
        abstract = {Tuberculosis (TB) continues to pose a severe public health challenge globally, with the WHO’s 2023 Global TB Report identifying India as the country carrying the single largest share of new cases worldwide, at approximately 26–27 percent of the global total [10]. Conventional diagnosis through sputum culture and radiological interpretation is time-intensive, resource-dependent, and subject to substantial inter-observer variability, particularly in low-resource clinical settings. This paper presents an end-to-end intelligent TB detection system that integrates a fine-tuned DenseNet-121 convolutional neural network with a four-stage severity classification framework, clinical decision support output, and full-stack deployment across a React-based frontend, Spring Boot REST backend, and MySQL database. The model is trained on the publicly available Kaggle Tuberculosis Chest X-ray Dataset comprising 700 TB-positive and 3,500 normal radiographs, achieving 98.10 percent classification accuracy, 99.99 percent AUC-ROC, 100 percent sensitivity, and zero false negatives on the held-out test set. Gradient-weighted Class Activation Mapping (Grad-CAM) is incorporated to provide spatial explainability of model predictions, enabling clinician trust and interpretability. The system further augments detection output with stage-specific precautionary recommendations, hospital referral guidance, and specialist physician information, addressing the critical gap between AI-based screening and patient-actionable clinical guidance.},
        keywords = {Tuberculosis Detection; Chest X-ray; DenseNet-121; Transfer Learning; Grad-CAM; Severity Classification; Clinical Decision Support; Medical Image Analysis},
        month = {April},
        }