Researchers at the University of Notre Dame in Indiana, US are developing a new artificial intelligence (AI) technique to improve the speed at which computed tomography (CT) scans can identify patients with Covid-19.
Difficulty accessing tests means some hospitals are using CT scans to search for visual signs of the disease, such as ground glass opacities, to diagnose the condition. Radiologists have been working overtime to screen them all, but deep learning could help drastically speed up the process and reduce this burden.
University of Notre Dame associate professor Yiyu Shi said: “We're developing a novel method inspired by Independent Component Analysis, using a statistical architecture to break each image into smaller segments, which will allow deep neural networks to target COVID-related features within large 3D images.”
The researchers are collaborating with radiologists at Guangdong Provincial People's Hospital in China and the University of Pittsburgh Medical Center, where a large number of CT images from patients with Covid-19 pneumonia are available. The team hopes to complete development by the end of the year.