Cunningham comments on predictive algorithm for COVID-19 mortality
Illinois ECE Professor Brian T Cunningham, Donald Biggar Willett Professor in Engineering, was recently featured in an article from Chemistry World, providing his personal insight on their new technology from a team of American and Chinese researchers. In the study, the researchers developed a severity scoring system for COVID-19 by analyzing the biomarkers present in a blood sample.
Considering other risk factors such as age and gender, the researchers used a statistical learning algorithm to predict the likelihood of a COVID-19 patient dying from the disease. This is the first tool of its kind to predict the mortality of individual COVID-19 patients and does not require laboratory work.
The test employs a "single-use microfluidic cartridge placed within a biosensor platform, which generates immunofluorescent signals that correlate to antigen concentrations." Biomarkers were chosen based on linkage to poor outcomes in patients. Biomolecules including D-dimer, C-reactive protein, and procalcitonin were dramatically increased in patients who died compared to patients who recovered. Using AI and outcomes of COVID-19 patients in Wuhan, the model continues to learn.
According to Chemistry World, Cunningham stated that the model could "never be fully predictive, even when it is used to evaluate a group of patients with similar characteristics." However, he did acknowledge that this tool could potentially identify high-risk patients, allowing for further intensive care.
Read more from Chemistry World here.