HealthDay News — An artificial intelligence (AI) system can predict deterioration of COVID-19 patients in the emergency department, according to a study published online May 12 in npj Digital Medicine.
Farah E. Shamout, Ph.D., from NYU Abu Dhabi in the United Arab Emirates, and colleagues conducted a retrospective study to predict deterioration risk for COVID-19 using a deep neural network that learns from chest X-ray images and a gradient boosting model that learns from routine clinical variables. The AI prognosis system was trained using data from 3,661 patients.
The researchers found that the system was able to accurately predict the deterioration risk on a test set of new patients, achieving an area under the receiver operating characteristic curve (AUC) of 0.786 when predicting deterioration within 96 hours. The system performed comparably to two radiologists in a reader study. A preliminary version of the deep neural network was silently deployed at New York University Langone Health during the first wave of the pandemic and produced accurate predictions in real time (AUC, 0.717).
“Emergency room physicians and radiologists need effective tools like our program to quickly identify those COVID-19 patients whose condition is most likely to deteriorate quickly so that health care providers can monitor them more closely and intervene earlier,” Shamout said in a statement.