Pulmonary Tuberculosis Imaging and Diagnosis

Pulmonary tuberculosis continues to present a substantial global health burden, with radiological and advanced imaging techniques integral to case identification, management and monitoring. Chest radiography remains the cornerstone of initial evaluation, offering widespread availability and rapid assessment of lung opacities, cavitation and pleural involvement. High-resolution computed tomography (CT) provides superior spatial detail, enabling precise characterisation of nodular patterns, caseous necrosis and early detection of bronchiectasis or lymph node enlargement. Magnetic resonance imaging (MRI) has emerged as a radiation-free option for delineating soft-tissue changes, while ultrasonography proves valuable in assessing pleural effusions and chest wall collections in resource-constrained settings. Positron emission tomography combined with CT (PET-CT) affords functional insight into metabolic activity, aiding differentiation of active disease from healed scarring. Despite these advances, imaging findings alone lack absolute specificity, and microbiological confirmation remains the diagnostic gold standard. Novel computer-aided detection and deep-learning algorithms are enhancing sensitivity of screening programmes, particularly among high-risk and asymptomatic populations. Standardised scoring systems and nomograms are increasingly adopted to reduce interobserver variability, guide treatment decisions and forecast outcomes, underscoring the evolving interplay between quantitative imaging and clinical microbiology in combating tuberculosis worldwide.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Pulmonary Tuberculosis Imaging and Diagnosis publication trend

The graph below shows the total number of articles in pulmonary tuberculosis imaging and diagnosis across all publications each year (not limited to Nature Index journals).

Technical terms

Chest radiography (CXR): A two-dimensional X-ray examination of the thorax used as the first-line imaging modality to detect lung abnormalities such as opacities, cavities and pleural changes.

Computed tomography (CT): A cross-sectional imaging technique that uses X-rays and computer processing to visualise detailed internal structures, enabling precise assessment of nodules, interstitial changes and lymphadenopathy.

Radiomics: The quantitative analysis of medical images to extract large numbers of features—such as texture, shape and intensity—that may correlate with disease characteristics and outcomes.

Deep learning: A subset of artificial intelligence based on neural networks that can learn complex patterns from large datasets, here applied to automated interpretation of radiological images.

Computer-aided detection (CAD): Software tools designed to assist radiologists by highlighting suspicious areas on medical images that may warrant further review.

Nomogram: A graphical calculation tool that integrates multiple predictors to estimate the probability of a clinical event, supporting individualized diagnostic or prognostic decisions.

References

1. Development and Validation of Deep Learning–Based Infectivity Prediction in Pulmonary Tuberculosis Through Chest Radiography: Retrospective Study. Journal of Medical Internet Research (2024).
2. A CT-based radiomics predictive nomogram to identify pulmonary tuberculosis from community-acquired pneumonia: a multicenter cohort study. Frontiers in Cellular and Infection Microbiology (2024).
3. The performance of computer-aided detection for chest radiography in tuberculosis screening: a population-based retrospective cohort study. Emerging Microbes & Infections (2025).