Multimodal Deep Learning For Covid-19 Prognosis Prediction In The Emergency Department: A Bi-centric Study

This study developed an algorithm that used textual and tabular data to predict 30-day mortality and ICU admission in patients in the ED for COVID-19. The main finding was that combining tabular and textual variables led to better prediction of 30-day mortality compared to models using only tabular variables. The methodology for analyzing textual data, i.e. the NLP, is commonly used for entity recognition, literature-based discovery, and question answering, but its potential for predicting outcomes in COVID-19 patients has not been fully explored.