Artificial Intelligence To Now Help Diagnose Tuberculosis

The prevalence of artificial intelligence technology is growing day by day in our lives. Now, this technology can also be used to diagnose the infectious disease, Tuberculosis. A recent study revealed that it may be possible to use artificial intelligence to diagnose tuberculosis in developing as well as underdeveloped countries.

It is said scientists are training artificial intelligence (AI) models to identify tuberculosis on chest X-rays, which in turn helps to screen and evaluating efforts in TB prevalent areas where the number of radiologists is limited.

The researchers from Thomas Jefferson University in Philadelphia used 1,007 x-rays of people who are both unaffected and those who are confirmed TB patients, to train artificial intelligence models to identify TB on X-rays.

Dr. Paras Lakhani, co-author and assistant professor of Radiology, Thomas Jefferson University, said: “There is a tremendous interest in artificial intelligence, both inside and outside the field of medicine, “An artificial intelligence solution that could interpret radiographs for presence of TB in a cost-effective way could expand the reach of early identification and treatment in developing nations.”

The best performing AI model was a combination of the AlexNet and GoogleNet with an accuracy of 96 percent. Artificial Intelligence is a growing branch of computer science in which machines perform tasks that usually require human intelligence.

The research team is now planning to improve the models with more training cases and with other deep learning methods.