Researchers from Northwestern University in Evanston, Illinois, have recently applied machine learning tools to Electronic Health Record (EHR) data to identify secondary bacterial pneumonia as the major cause of death in seriously COVID-19 ill patients. The research, which was conducted by a team of scientists at Northwestern Memorial Hospital in Chicago, monitored 585 patients who were admitted to the hospital’s intensive care unit (ICU) and were suffering from severe pneumonia and respiratory failure. Of these patients, 190 had COVID-19.
Using a newly developed machine learning approach called CarpeDiem, the researchers were able to group similar ICU patient days into individual clinical states. The study findings refute the prevailing opinion that inflammation induced by a so-called “cytokine storm” is the main cause of death among COVID-19 patients. As the study’s co-author, Catherine Gao MD, commented in the study release: “The application of machine learning and artificial intelligence to clinical data can be used to develop better ways to treat diseases like COVID-19 and to assist ICU physicians managing these patients.”
Northwestern University is a private research university founded in 1851. Located in Evanston, Illinois, Northwestern is home to 12 colleges and schools and serves around 21,000 students and comprises over 4,400 faculty members worldwide. Renowned for its successful medical, law, and business programs and research, the university is regarded as a leading higher education institution in the United States and abroad.
Catherine Gao MD is one of the authors of the study conducted by Northwestern University on the analysis of EHR data in order to identify the main cause of death in severely ill COVID-19 patients. Dr. Gao is affiliated with Northwestern Medicine and is an Assistant Professor at Northwestern University’s Feinberg School of Medicine. Her research interests include healthcare analytics, patient care decision support, predictive analysis, and machine learning.
