Artificial intelligence 96% accurate in detecting tuberculosis

Tuberculosis (TB) remains one of the top global causes of death due to a lack of screening in remote areas with limited access to radiologists. A recent study published in Radiology discussed how researchers have utilized artificial intelligence (AI) to identify TB on chest x-rays.

Using a deep convolutional neural network (DCNN), which evaluates multiple layers within x-ray images, researchers believe the technology could improve screening and treatments of TB in areas where medical professionals are scarce.

"There is a tremendous interest in artificial intelligence, both inside and outside the field of medicine," said study co-author Paras Lakhani, MD, from the 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."

Researchers collected 1,007 chest x-rays of patients with active TB from the National Institutes of Health, the Belarus Tuberculosis Portal and TJUH. They then split datasets into training, validation and test. Two DCNN models, AlexNet and GoofLeNet, were trained to discriminate between positive and negative TB x-rays. Accuracy was tested on 150 cases not included in the training process to properly test the programs detection ability.

Results showed the combination of the two DCNN models achieved the highest accuracy at 96 percent. Of the 150 test cases, the models disagreed on 13 cases where an expert radiologist intervened to achieve 100 percent accuracy on each case. The involvement of a radiologist raised the net accuracy rate to 99 percent.

"Application of deep learning to medical imaging is a relatively new field," Lakhani said. "In the past, other machine learning approaches could only get to a certain accuracy level of around 80 percent. However, with deep learning, there is potential for more accurate solutions, as this research has shown. 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."

""
Cara Livernois, News Writer

Cara joined TriMed Media in 2016 and is currently a Senior Writer for Clinical Innovation & Technology. Originating from Detroit, Michigan, she holds a Bachelors in Health Communications from Grand Valley State University.

Trimed Popup
Trimed Popup