Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Empowering Tuberculosis Screening with Explainable Self-Supervised Deep Neural Networks

Neel Patel, Alexander Wong, Ashkan Ebadi

TL;DR

This work tackles TB screening from chest X-rays in settings with limited radiology resources by introducing an explainable self-supervised self-train network (DISTL) built on a Vision Transformer backbone pre-trained on CheXpert. The approach uses lung segmentation, multi-crop inputs, and a teacher–student knowledge distillation framework with self-supervised and supervised losses to learning from both labeled and unlabeled data. It achieves high diagnostic performance (overall accuracy 98.14%; TB precision 99.44%; TB recall 95.72%) and provides attention-based explanations that align with radiologic TB indicators. The study demonstrates potential for reducing labeled-data requirements while delivering clinically meaningful explanations, though it acknowledges dataset limitations and calls for radiologist validation and broader, multi-center evaluation before real-world deployment.

Abstract

Tuberculosis persists as a global health crisis, especially in resource-limited populations and remote regions, with more than 10 million individuals newly infected annually. It stands as a stark symbol of inequity in public health. Tuberculosis impacts roughly a quarter of the global populace, with the majority of cases concentrated in eight countries, accounting for two-thirds of all tuberculosis infections. Although a severe ailment, tuberculosis is both curable and manageable. However, early detection and screening of at-risk populations are imperative. Chest x-ray stands as the predominant imaging technique utilized in tuberculosis screening efforts. However, x-ray screening necessitates skilled radiologists, a resource often scarce, particularly in remote regions with limited resources. Consequently, there is a pressing need for artificial intelligence (AI)-powered systems to support clinicians and healthcare providers in swift screening. However, training a reliable AI model necessitates large-scale high-quality data, which can be difficult and costly to acquire. Inspired by these challenges, in this work, we introduce an explainable self-supervised self-train learning network tailored for tuberculosis case screening. The network achieves an outstanding overall accuracy of 98.14% and demonstrates high recall and precision rates of 95.72% and 99.44%, respectively, in identifying tuberculosis cases, effectively capturing clinically significant features.

Empowering Tuberculosis Screening with Explainable Self-Supervised Deep Neural Networks

TL;DR

This work tackles TB screening from chest X-rays in settings with limited radiology resources by introducing an explainable self-supervised self-train network (DISTL) built on a Vision Transformer backbone pre-trained on CheXpert. The approach uses lung segmentation, multi-crop inputs, and a teacher–student knowledge distillation framework with self-supervised and supervised losses to learning from both labeled and unlabeled data. It achieves high diagnostic performance (overall accuracy 98.14%; TB precision 99.44%; TB recall 95.72%) and provides attention-based explanations that align with radiologic TB indicators. The study demonstrates potential for reducing labeled-data requirements while delivering clinically meaningful explanations, though it acknowledges dataset limitations and calls for radiologist validation and broader, multi-center evaluation before real-world deployment.

Abstract

Tuberculosis persists as a global health crisis, especially in resource-limited populations and remote regions, with more than 10 million individuals newly infected annually. It stands as a stark symbol of inequity in public health. Tuberculosis impacts roughly a quarter of the global populace, with the majority of cases concentrated in eight countries, accounting for two-thirds of all tuberculosis infections. Although a severe ailment, tuberculosis is both curable and manageable. However, early detection and screening of at-risk populations are imperative. Chest x-ray stands as the predominant imaging technique utilized in tuberculosis screening efforts. However, x-ray screening necessitates skilled radiologists, a resource often scarce, particularly in remote regions with limited resources. Consequently, there is a pressing need for artificial intelligence (AI)-powered systems to support clinicians and healthcare providers in swift screening. However, training a reliable AI model necessitates large-scale high-quality data, which can be difficult and costly to acquire. Inspired by these challenges, in this work, we introduce an explainable self-supervised self-train learning network tailored for tuberculosis case screening. The network achieves an outstanding overall accuracy of 98.14% and demonstrates high recall and precision rates of 95.72% and 99.44%, respectively, in identifying tuberculosis cases, effectively capturing clinically significant features.
Paper Structure (11 sections, 3 figures, 1 table)

This paper contains 11 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Data and Methods
  3. Data Collection and Preparation
  4. Network Architecture
  5. Network Training and Performance Evaluation
  6. Explainability Analysis
  7. Results
  8. Performance Analysis
  9. Explainability Analysis
  10. Discussion and Conclusion
  11. Limitations and Future Work

Figures (3)

  • Figure 1: Sample images in the dataset.
  • Figure 2: The high-level conceptual flow of the framework that encompasses three primary components for segmentation, self-supervision, and self-training.
  • Figure 3: Two sample tuberculosis-positive cases, correctly classified by the proposed network, with identified critical regions.