Health IT, Health Tech

Algorithm that predicts Covid-19 complications gets EUA

The FDA granted an emergency use authorization to an algorithm developed by Dascena to predict which Covid-19 patients might face unstable blood pressure or respiratory decline.

An algorithm to predict which patients might experience complications from Covid-19 received an emergency use authorization from the Food and Drug Administration. It identifies patients that are likely to face unstable blood pressure or respiratory decline based on vitals data.

Dascena, an Oakland, Calif.-based startup, developed the tool. It had previously developed other algorithms for early intervention, such as one to predict the onset of sepsis, a life-threatening complication from infections.

The decision support tool, called COViage, pulls in information from a patient’s medical record about their age, gender, heart rate, blood pressure, temperature and respiratory rate. It then provides a one-time notification if patients are determined to be at risk for deterioration for either of these complications.

“Covid-19 remains a significant public health emergency both in the U.S. and around the globe, and we are encouraged that by receiving this EUA, our machine learning algorithm can help caregivers diagnose critical conditions resulting from Covid-19 earlier and more accurately,” Dascena President and CEO Ritankar Das said in a news release. “The early identification of patients at risk of respiratory decompensation or hemodynamic instability would enable physicians to more aggressively monitor these patients in a controlled environment and provide earlier treatment.”

In its letter granting the EUA, the FDA said it could be used as a diagnostic aid, given that there is no cleared alternative, and the benefits likely outweigh the risks. The system does not replace patient monitoring.

Dascena also released the results of a small trial testing the efficacy of its decision support tool compared to the standard of care. The current standard, the Modified Early Warning Score, is used to calculate whether patients should be transferred to the ICU based on their blood pressure, heart rate, respiratory rate, temperature and alertness.

The study included 197 patients that were admitted to five hospitals between late March and early May. They had a confirmed Covid-19 diagnosis, and their vital signs were taken within two hours of arrival to the emergency room or hospital admission.

Researchers checked whether patients needed endotracheal tube or mechanical ventilation 24 hours after COViage made its predictions. The algorithm predicted which patients would need ventilation with 36% more accuracy than MEWS, according to the study. It was published in Computers in Biology and Medicine in September.

Photo credit: mrspopman, Getty Images

Shares1
Shares1