New screening tool for Tuberculosis

Automated chest-radiography as a new screening tool for Tuberculosis in resource-constrained settings

Almost nine million people around the world become ill with tuberculosis (TB) each year. More than one-third of them fail to gain access to accurate diagnosis or effective treatment and are needlessly suffering and dying from this curable disease. Many of these people live in poverty-stricken areas and have very limited or non-existent access to health services.

Chest radiography has always been widely used to find TB. Digital chest radiography has made X-rays cheaper and easier to use. No films, chemicals and water are needed to produce a digital radiograph and automatic exposure control ensures images of a high quality that are instantly available. One issue remains: the lack of human experts to read the chest radiograph in research constrained settings.

To address this issue Delft Imaging Systems (well known as the inventor of the Odelca camera from the 1960s, one of the most widely used systems in X-ray TB screening in the world) and Radboud University developed the CAD4TB software. With this software, computers can automatically analyse chest radiographs for signs of tuberculosis.

In this project we have expanded the CAD4TB software to include other TB indicators, like HIV status, so that a TB score indicating the likelihood of a person having TB can be generated. The rationale behind this is that new molecular diagnostics for TB hold great potential for tuberculosis (TB) diagnosis, but are costly and time consuming. Radiography can therefore act as a filter: TB suspects undergo symptom screening and chest radiography, and only those with a high score undergo further testing.

A TB clinical decision system (CAD4TB+) has been developed for rapid screening of TB suspects. The system automatically analyses a chest radiography and by combining the result with other TB indicators - like HIV status and age - gives a nominal score between 0-100, the higher the score the more likely it is a TB suspect actually has the disease.
Technology Readiness Level (TRL)
7 - 9
Time period
31 months