"AI Predicts ICU Requirements for COVID-19 Patients with Deep Learning Model"

Researchers from the University of Chicago have developed a deep learning-based model that can predict the need for intensive care in COVID-19 patients by analyzing their chest X-ray images. The model, which utilizes transfer learning, achieved a good degree of accuracy in predicting the need for intensive care within 24, 48, 72, and 96 hours of the exam. It outperformed existing models and could help with clinical decision-making and resource management, improving patient care during the pandemic. The researchers are working on further improvements, including incorporating clinical variables and extending the model to include lung opacities caused by related diseases.