Artificial intelligence can help diagnose tuberculosis in remote areas, says research
A new study appearing online in the journal Radiology shows researchers are training artificial intelligence (AI) models to identify tuberculosis (TB) on chest X-rays that may help screen and evaluate efforts in TB-prevalent areas with limited access to radiologists.
According to the co-author of the study, Paras Lakhani, M.D., from Thomas Jefferson University Hospital (TJUH) in Philadelphia, there is a tremendous interest in AI both inside and outside the field of medicine. An AI-based 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.
Lakhani and his colleague, Baskaran Sundaram, M.D., obtained 1,007 X-rays of patients with and without active TB. The cases consisted of multiple chest X-ray datasets from the National Institutes of Health, the Belarus Tuberculosis Portal, and TJUH. The datasets were split into training (68%), validation (17.1%), and test (14.9%). The cases were used to train two different DCNN models – AlexNet and GoogLeNet – which learned from TB-positive and TB-negative X-rays.
The models’ accuracy was tested on 150 cases that were excluded from the training and validation datasets. The best performing AI model was a combination of the AlexNet and GoogLeNet, with a net accuracy of 96%. The two DCNN models had disagreement in 13 of the 150 test cases. For these cases, the researchers evaluated a workflow where an expert radiologist was able to interpret the images, accurately diagnosing 100% of the cases. This workflow, which incorporated a human in the loop, had a greater net accuracy of close to 99%.
AI in healthcare is certainly growing apace. Babylon Health, a UK-based startup that offers digital healthcare app using a mixture of AI and video and text consultations with doctors and specialists, has raised $60 million in new funding which will be used to continue building out its AI capabilities, including offering diagnosis by AI, which is pegged to roll out later in 2017.
X-ray credit: ©iStock.com/stockdevil