Abu Dhabi-based healthcare technology company M42 highlighted a study that has validated its AI model for triaging chest x-rays for tuberculosis (TB).
Conducted at the Capital Health Screening Centre in Abu Dhabi, the study involved over 1 million chest x-rays and assessed the efficacy and scalability of a model called AI Radiology in Screening TB (AIRIS-TB). The model is designed to automate the reporting of all x-rays without any findings.
According to the results, AIRIS-TB achieved an area under the operating curve of 98.5% and an overall false negative rate (FNR) of 1.57%, outperforming radiologists (1.85%) while maintaining a 0% TB-FNR, the company said.
“This large-scale validation demonstrates AIRIS-TB’s safety and efficiency in high-volume TB screening programs, reducing radiologist workload without compromising diagnostic accuracy,” the article’s authors wrote.
The full article is available in npj Digital Medicine.
