Tuberculosis is a multisystemic disease, being among the leading causes of infectious disease–related mortality worldwide [1].
During the twentieth century, the overall incidence and mortality rate for tuberculosis declined considerably over the years due to new treatment [2].
Unfortunately, despite advances in diagnosis and treatment, tuberculosis infection began spreading after the 1980s in developed countries, partly because of the emergence of HIV, and still has a high incidence in developing countries, which makes it necessary to improve screening methods for detection, optimal prevention and eradication of the disease [2].
Tuberculosis may mimic a large variety of diseases. The diagnosis is often delayed due to nonspecific laboratory findings and atypical clinical manifestations, resulting in complications, extensive disease and higher risk of mortality [3].
Imaging modalities
There are various imaging modalities and each plays a certain role, depending on the clinical profile of the patient:
1. Chest X-ray
2. CT
3. MRI
4. PET-CT
Chest X-ray is the basic imaging modality in the initial evaluation of patients for pulmonary tuberculosis, being a sensitive tool for screening pulmonary TB lesions [2].
Pulmonary disease is the most common presentation. It has conventionally been divided into primary and postprimary TB [2].
Fig. 1: Radiographic features of primary and post-primary TB [4].
References: McAdams HP, Erasmus J, Winter JA. Radiologic manifestations of pulmonary tuberculosis. Radiol Clin North Am. 1995;33:655–78.
The trained AI algorithm has the ability to:
-
analyse digital X-ray images;
-
detect a variety of pulmonary lesions;
-
divide the lesions into different pathology categories.
The availability of more data improved the accuracy in TB lesions detection, which makes it suitable for a screening program in a disease where early detection is essential. Moreover, the abnormalities are indicated through a heatmap, which serves as a supplementary aid for the examining radiologist.