COVID-19: Vancouver researchers lead international study to predict novel coronavirus presence

    1 of 1 2 of 1

      As experts from around the world race to develop or improve medical treatments and tools for addressing the COVID-19 pandemic, a team of researchers from several local institutions are working together on a project that will improve diagnoses of the coronavirus.

      A group from Vancouver General Hospital (VGH), the University of British Columbia (UBC), and the Vancouver Coastal Health Research Institute are leading an international study to predict the presence of COVID-19 based on computerized tomography (CT) scans.

      Radiologists, residents, fellows, and UBC medical students are collecting, analyzing, and labelling CT scans and some chest X-rays from international COVID-19 patients, including from Canada, South Korea, Italy, and the Middle East.

      VGH emergency and trauma radiology director Dr. Savvas Nicolaou is leading the project with Dr. William Parker, a radiology resident at VGH and UBC.

      "We've seen patients present in the emergency department with non-typical symptoms such as severe abdominal pain, stroke and acute chest pain, and upon reviewing their CT scans for those conditions, we see the tell-tale haziness of COVID-19 in their lungs,” Dr. Nicolaou stated in a news release.

      Dr. Nicolaou stated that the lungs of COVID-19 patients appear “white and hazy, like a white-out or blizzard” and that they can’t currently predict the severity of the disease or its impact upon different patient populations.

      However, information from the scans will be used for an open-source artificial intelligence (AI) model to predict the presence, severity, and complications of COVID-19 on CT scans and to help improve patient care (such as whether a patient should be treated at home or in hospital).

      "The model will also assist in detecting similarities and differences in variations of patterns across different cultural and ethnic groups, and help us understand early and late stages of patterns of disease," VGH emergency physician and UBC Cloud Innovation Centre academic director Dr. Kendall Ho stated in a news release.

      The model could also help to identify patients with the potential to develop permanent lung damage or fibrosis.

      The new AI model will be piloted at Vancouver General Hospital with the goal of using it in routine COVID-19 diagnostic procedures. (It won’t replace current testing.)