 This paper proposes a novel method that uses a combination of machine learning algorithms to predict the likelihood of a patient requiring admission to the intensive care unit, ICU, and assessing their risk of in-hospital mortality due to coronavirus disease 2019, COVID-19. The authors used a convolutional neural network, CNN, to analyze chest x-rays and a random forest algorithm to identify key clinical variables. The CNN was then combined with these variables to create a predictive model that accurately predicted ICU admission area under the curve AUC equals 0.92 plus or minus 0.04 and hospital mortality AUC equals 0.81 plus or minus 0.06. This model can be used as a valuable tool to help healthcare professionals make informed decisions about which patients should receive ICU treatment or be admitted to the hospital. This article was authored by Nicholas Munera, Esteban Garcia-Gallow, Alvaro Gonzalez, and others.