Graphic by Tai Michaels.
The coronavirus pandemic has led to the deaths of hundreds of thousands of people worldwide. Because COVID-19 can manifest itself through a wide range of symptoms and complications, predicting patient mortality has been very difficult.
To circumvent this challenge, researchers at the Icahn School of Medicine at Mount Sinai developed a mortality prediction model using machine learning techniques. They used a database of information from patients treated for COVID-19 in the Mount Sinai Health System to design a computational model based on patient age, the minimum oxygen levels over the course of their care, and the type of patient care.When tested, the model showed high accuracy in predicting mortality both for patients who had already been treated and for patients who were in the process of being treated for COVID-19.
An accurate model such as this one could have positive implications in clinical settings, where it could help guide medical professionals in patient prognosis and management. However, this specific model requires further testing in a variety of populations.