Artificial Intelligence Predicts COVID-19 Patients Needing ICU Care

According to researchers, artificial intelligence could assist doctors in allocating resources during the COVID-19 pandemic by identifying patients who require intensive care unit (ICU) treatment.

A research team at the University of Waterloo and DarwinAI, an alumni-founded startup company, developed the system that uses AI to predict the necessity of ICU admission based on over 200 clinical data points, such as vital signs, blood test results, and medical history.

“That is a very important step in the clinical decision support process for triaging patients and developing treatment plans,” professor of systems design engineering and Canada Research Chair in AI and Medical Imaging at Waterloo, Alexander Wong, said in a press release.

The AI technology was trained using data from around 400 cases at Hospital Sirio-Libanes in Sao Paulo, Brazil, in which doctors decided which COVID-19 patients should be admitted for intensive care.

Although this study uses Sao Paulo, Brazil, as the sample, due to the universal nature of COVID-19 spread, it could provide key insights to public health officials around the globe. COVID-19 treatment insights remain valuable as new variants continue to emerge.

According to the research team, the neural network can predict the need for ICU admission in new COVID-19 cases with more than 95 percent accuracy. Additionally, the system identifies the key factors that drive predictive analytics, providing supporting evidence for clinicians.

Rather than replace doctors with technology, the system is meant to assist them with tools to make faster and more informed decisions, supporting augmented intelligence. The technology will ensure that patients most in need of intensive care will receive it.

“The goal is to help clinicians make faster, more consistent decisions based on past patient cases and outcomes,” said Wong, a director of the Vision and Image Processing (VIP) Lab at Waterloo. “It’s all about augmenting their expertise to optimize the use of medical resources and individualize patient care.”

In a global collaboration effort, the research team made the technology freely available so engineers and scientists worldwide can work to improve the AI system. The team is also incorporating the technology into a more extensive clinical decision support system to develop their ongoing COVID-Netopen-source initiative.

The initiative helps doctors identify COVID-19 and determine the severity of the infection using AI analysis of medical images.

Wong collaborated on the ICU admission work with DarwinAI researchers Audrey Chung and Mahmoud Famouri and Andrew Hryniowski, an engineering PhD student in the VIP Lab.