(Subscribe to our Today's Cache newsletter for a quick snapshot of top 5 tech stories. Click here to subscribe for free.)
Researchers at the University Of Notre Dame in US are developing Artificial Intelligence-based technology to speed up the process of CT screening images of lungs for faster COVID detection. The team's research could provide some relief to overworked radiologists.
The pandemic has led to an influx of patients admitted with COVID-19. They are CT scanned for visual signs of the disease, including ground glass opacities -- a condition that consists of abnormal lesions.
Most patients with coronavirus show signs of COVID-related pneumonia. And radiologists are working overtime to screen them. The new technique will reduce burden on the radiologists tasked with screening each image.
“We have shown that we can use deep learning — a field of AI — to identify those signs, drastically speeding up the screening process and reducing the burden on radiologists,” Yiyu Shi, associate professor at Notre Dame, and lead researcher on the project said.
The University Of Notre Dame is collaborating with radiologists at Guangdong Provincial People’s Hospital in China and the University of Pittsburgh Medical Center, to access a large number of COVID-related chest images from CT scans.
Shi and his team of researchers are working to analyse 3D data from the CT scans to spot visual features of COVID-19-related pneumonia.
They are aiming to integrate the analysis software into off-the-shelf hardware for a light-weight mobile device that can then be used across clinics in the world.
However, 3D CT scans are large, making it impossible to identify particular features and extract them accurately on mobile devices, Shi said.
The team is developing a new method using statistical architecture to break each image into smaller segments, and then analyse its independent components. This will help spot COVID-related features within large 3D images, Shi said.
The research is being funded by the National Science Foundation and the development is expected to complete by the end of the year.