UC San Diego recently announced that its health radiologists and other physicians are now leveraging artificial intelligence (AI) to augment lung-imaging analysis in a clinical research study aimed at COVID-19 lung imaging analysis.
The cause of death for most COVID-19 patients is pneumonia, which often requires long hospital stays in intensive care units and assistance breathing with ventilators.
Last year, Albert Hsiao, MD, PhD, associate professor of radiology at University of California San Diego School of Medicine and radiologists at UC San Diego Health, and his team, developed a machine learning algorithm that allowed radiologists to use AI to enhance their own abilities to spot pneumonia on chest X-rays.
The algorithm was trained with 22,000 notations by radiologists and overlays X-rays with color-coded maps that indicate pneumonia probability, researchers explained.
“Pneumonia can be subtle, especially if it’s not your average bacterial pneumonia, and if we could identify those patients early, before you can even detect it with a stethoscope, we might be better positioned to treat those at highest risk for severe disease and death,” Hsiao said.
The AI algorithm was most recently applied to 10 chest X-rays from five patients treated in China and the US for COVID-19.
Although the images were taken at several different hospitals and varied significantly in technique, contrast, and resolution, the algorithm consistently localized areas of pneumonia.
The study, enabled by Amazon Web Services (AWS), allows for Hsiao’s AI method to be deployed across UC San Diego Health in a clinical research study that allows any physician and radiologist to get a quick estimate regarding a patient’s likelihood of having pneumonia.
“AWS has partnered with us on multiple projects in the past,” Michael Hogarth, MD, professor of biomedical informatics at UC San Diego School of Medicine and clinical research information officer at UC San Diego Health, said in a statement.
“Once COVID-19 became a crisis, AWS reached out to us and asked if there was anything they could do to help. My mind immediately went to a presentation I’d seen Albert give on their initial AI tests for pneumonia,” he added.“AWS helped our Clinical Research IT team get the study up and running system-wide in just 10 days.”
So far, the new AI capabilities have provided UC San Diego Health physicians with insights to more than 2,000 images.
Image Credit: Thinkstock
Thanks to Heinz V. Hoenen. Follow him on twitter: @HeinzVHoenen