Bangladesh is a vast and densely populated country with approximately 165 million people in 2020. In that same year, 360,000 estimated people developed TB, of which 30,000 were children. There were 129,919 missing people with TB, and 20,637 were children.

The current situation can be dramatically improved. The target proposed is a 95% reduction in TB deaths and a 90% reduction in new cases in 20 years.

With WHO and Stop TB Partnership funding, we, alongside IRD, set up a new approach to fight against TB in Bangladesh. The focus was on increasing early case detection, reporting and treating TB and serving poor communities in Dhaka, Bangladesh.

Under the project, in 2013, we installed an EasyDR digital X-ray system with CAD4TB, partnering with ICDDR’B. Rolled out using an innovative pay-per-use scheme, the lung screening centre charged a modest fee for every “lung screening.” A part of this fee supported the cost of the X-ray units. This system is still utilised today to help with tuberculosis screening within the country.

In 2021, under the iNTP project (funded by USAID, supported by the Stop TB Partnership project), 14 Delft Light portable backpack X-ray systems with the CAD4TB artificial intelligence software – used to detect TB on chest X-rays were delivered. As part of the project, installation and training services were also delivered to support the program in the years to come and optimise the systems’ utilisation.

In 2023, Delft Imaging delivered another 12 Delft Light portable X-ray systems, with the CAD4TB software, installation and training services to ICDDRB,B in Bangladesh.

Making a difference

Over the years, several papers have been published researching the different solutions of Delft Imaging and its utilisation in Bangladesh. A 2017 study stated that while the radiologist’s specificity was superior, CAD4TB could save Xpert tests. A study in 2020 used CAD4TB amongst tuberculosis patients in a study on the prevalence of diabetes mellitus. In 2021, a study using CAD4TB found that the artificial intelligence algorithm could reduce the number of Xpert tests required by 50% while maintaining a sensitivity above 90%. Another study in 2022 proved that CAD4TB surpassed the human radiologists’ performance with improved triage ability compared to the older version.