Differentiating between drug-sensitive and drug-resistant tuberculosis with machine learning for clinical and radiological features

Abstract

Background: Tuberculosis (TB) drug resistance is a worldwide public health problem that threatens progress made in TB care and control. Early detection of drug resistance is important for disease control, with discrimination between drug-resistant TB (DR-TB) and drug-sensitive TB (DS-TB) still being an open problem. The objective of this work is to investigate the relevance of readily available clinical data and data derived from chest X-rays (CXRs) in DR-TB prediction and to investigate the possibility of applying machine learning techniques to selected clinical and radiological features for discrimination between DR-TB and DS-TB. We hypothesize that the number of sextants affected by abnormalities such as nodule, cavity, collapse and infiltrate may serve as a radiological feature for DR-TB identification, and that both clinical and radiological features are important factors for machine classification of DR-TB and DS-TB.Methods: We use data from the NIAID TB Portals program (https://tbportals.niaid.nih.gov), 1,455 DR-TB cases and 782 DS-TB cases from 11 countries. We first select three clinical features and 26 radiological features from the dataset. Then, we perform Pearson's chi-squared test to analyze the significance of the selected clinical and radiological features. Finally, we train machine classifiers based on different features and evaluate their ability to differentiate between DR-TB and DS-TB.Results: Pearson's chi-squared test shows that two clinical features and 23 radiological features are statistically significant regarding DR-TB vs. DS-TB. A ten-fold cross-validation using a support vector machine shows that automatic discrimination between DR-TB and DS-TB achieves an average accuracy of 72.34% and an average AUC value of 78.42%, when combing all 25 statistically significant features. Conclusions:Our study suggests that the number of affected lung sextants can be used for predicting DR-TB, and that automatic discrimination between DR-TB and DS-TB is possible, with a combination of clinical features and radiological features providing the best performance.

Generalization Challenges in Drug-Resistant Tuberculosis Detection from Chest X-rays

“…In these examples, it can be seen that for the DR-TB class, activations are more frequent. This is consistent with previous results [ 9 ], where DR-TB patients are shown to have more abnormalities in their lungs.…”

“…This study observed that the co-existence of multiple findings (multiple cavities, thick-walled cavities, disseminated lesions along the bronchi, whole-lung involvement) was indicative of DR-TB. Finally, more recent work [ 9 ] compared 1455 DR-TB and 782 DS-TB cases, using two clinical features and 23 types of radiological findings. A support vector machine was used to distinguish between DR-TB and DS-TB with an AUC of 0.78.…”

Clinical and Computed Tomography Features Associated with Multidrug-Resistant Pulmonary Tuberculosis: A Retrospective Study in China

“…Additionally, this study revealed that a retreated history and a longer period of previous anti-TB treatment were more common in patients with MDR-PTB than in those with DS-PTB, which corroborates some previous studies. [8][9][10][11][12] Anti-TB treatment times, adverse reactions, and interrupted or failed treatment have been reported many times as risk factors of MDR-PTB, which inevitably prolong the treatment period. Moreover, our results showed that CD4 + T and CD4/CD8 + T lymphocyte count ratio was lower in patients with MDR-PTB than in those with DS-PTB.…”

Drug-Resistant Tuberculosis Treatment Recommendation, and Multi-Class Tuberculosis Detection and Classification Using Ensemble Deep Learning-Based System

“…Figure 3 demonstrates the GradCAM heatmap [ 66 , 67 , 68 ] of various drug-resistant types. We divide the entire lung into six parts: left and right at the top, center left and right, and bottom left and right.…”