New AI Tool in Development for Quicker COVID-19 Diagnosis

June 3, 2020

The novel AI tool can help frontline health care workers quickly distinguish between COVID-19 pneumonia and non-COVID-19 as they deal with the rising number of patients.

Researchers in Canada are developing a new artificial intelligence (AI) tool to help frontline health care workers expedite the time they spend distinguishing between coronavirus disease 2019 (COVID-19) pneumonia and non-COVID-19 cases as they deal with an increasing number of patients.

Researchers from Simon Fraser University (SFU) and Providence Health Care (PHC) are collaborating on the novel AI tool, which is in now in the validation phase at St. Paul’s Hospital in Vancouver, Canada, according to an SFU press release.

“This multidisciplinary collaboration provided much needed validation of our models and expedited our development process, thereby directly assisting Providence Health Care’s efforts to plan for the urgent and ever evolving nature of the COVID-19 crisis,” said Soyean Kim, director, Digital Products, PHC, in the SFU press release.

The AI tool allows a physician to feed a patient’s chest x-ray image into a computer, run a bio-image detection analysis, and determine a positive COVID-19 pneumonia case. Researchers used x-ray images of both COVID-19 and non-COVID-19 patients to identify the unique characteristics of the virus to refine the AI system.

The novel tool is not a standalone clinical diagnosis solution, but it is designed to help physicians and other frontline health care workers quickly confirm their suspicion in combination with other tools such as CT scans.

For residents and less experienced physicians, the AI tool can help them review a data set and prepare to present a diagnosis when a senior physician is able to step in.

Once the AI tool receives approval, it will be made available for free with support from the United Nations and an ongoing multinational collaboration with further improve efficacy and provide authentication.

For more COVID-19 coverage for primary care, visit our COVID-19 Resource Page.