Artificial intelligence (AI) could be used to tackle one of the world’s leading causes of death, new research suggests.
Diagnosing tuberculosis (TB) in developing regions is challenging, with access to radiologists limited. But new research by scientists in America shows that AI could potentially be used to do that job in a cost-effective way.
And with treatments available for the disease – listed by the World Health Organization (WHO) as one of the world’s top 10 killers – there is huge potential to save lives. In 2016 alone, WHO estimates nearly two million people worldwide died from the disease.
“There is a tremendous interest in artificial intelligence, both inside and outside the field of medicine,” said study co-author Dr Paras Lakhani, from Thomas Jefferson University Hospital (TJUH) in Philadelphia. “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 study, led by Dr Lakhani and his colleague, Dr Baskaran Sundaram, revealed that by combining two deep learning AI models they could achieve a 96% accuracy in diagnosing TB.
Dr Lakhani added: “The relatively high accuracy of the deep learning models is exciting. The applicability for TB is important because it’s a condition for which we have treatment options. It’s a problem that can be solved.
“We hope to prospectively apply this in a real world environment. An artificial intelligence solution using chest imaging can play a big role in tackling TB.”
