By using the AI method, a simple image of a chest X-ray developed on film and captured with a mobile phone could be sent to a central location for analysis, which could be particularly helpful in low-resource areas, where TB is often concentrated.

The authors of the study detailed that TB causes 1.6 million deaths per year, making it the world´s 13th leading cause of death and the second-leading cause of death from infections after Covid-19.

An estimated three million patients were undiagnosed in 2021, specialists noted.

Frauke Rudolf, a specialist at the Department of Infectious Diseases at the Danish University Hospital Aarhus, and colleagues compared the AI software to detect TB in chest X-rays with that of Ethiopian radiologists tested.

Chest X-rays are very important for diagnosing patients who cannot produce good quality sputum samples for analysis by a microbiologist.

The scientists analyzed the chest X-rays of 498 patients, 57 of whom had been diagnosed with TB.

For some components of TB diagnosis, AI performed better than radiologists combined (such as consolidation and nodule findings) and for others it achieved worse deductions (such as hilar lymphadenopathy and cavitation).

“It is useful when radiologists are not available, but it could also be useful in settings where you don’t have experienced radiologists to support decision making,” Rudolf said.

pll/rgh/lpn