Clicky

AI technology can assess COVID-19 severity with a promising degree of accuracy

Artificial intelligence (AI) technology developed by researchers at the University of Waterloo is capable of assessing the severity of COVID-19 cases with promising accuracy.

Researchers from Waterloo and the spin-off start-up DarwinAI as well as radiologists from the Stony Brook School of Medicine and Montefiore took part in a study that is part of the open source initiative COVID-Net, which was launched more than a year ago . involved medical center in New York.

Deep learning AI was trained to analyze the extent and cloudiness of the infection in the lungs of COVID-19 patients using chest x-rays. The results were then compared to assessments of the same radiographs by experienced radiologists.

In terms of both extent and opacity, important indicators of the severity of infections, the predictions of the AI software were in good agreement with the values of the human experts.

Alexander Wong, professor of systems design engineering and co-founder of DarwinAI, said the technology could give doctors an important tool to help them manage cases.

Assessing the severity of a patient with COVID-19 is a critical step in the clinical workflow in determining the best course of action for treatment and care, whether it is admitting the patient to the ICU, oxygenating a patient, or embedding one Patient on a mechanical ventilator. “

Alexander Wong, Systems Design Engineering Professor and Co-Founder, DarwinAI

“The promising results of this study show that artificial intelligence has strong potential to be an effective tool to support frontline healthcare workers in making decisions and improve clinical efficiency, which is particularly important as the ongoing Pandemic places a heavy burden on health systems in the area. “The world.”

Source:

Journal reference:

Wong, A., et al. (2021) On the way to the computer-aided assessment of the degree of severity via deep neural networks for the assessment of the geographical and the degree of opacity of SARS-CoV-2 chest X-rays. Scientific reports. doi.org/10.1038/s41598-021-88538-4.

-