id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-313294-ffgo56gl	Bertsimas, D.	Personalized Prescription of ACEI/ARBs for Hypertensive COVID-19 Patients	2020-11-04		.txt	text/plain	7314	393	44	We couple electronic medical record (EMR) and registry data of 3,643 patients from Spain, Italy, Germany, Ecuador, and the US with a machine learning framework to personalize the prescription of ACEIs and ARBs to hypertensive COVID-19 patients. Since machine learning estimates a binary or continuous outcome of interest from large, high-dimensional datasets, a common approach involves training separate prediction models for the treatment and the control group, and recommending the alternative with the best outcome [18, 41] . In this paper, we propose a machine learning-based approach for personalized prescription of ACEI/ARBs for hospitalized hypertensive patients with COVID-19. One ensemble of various machine learning models is trained to predict mortality/morbidity risk with ACEI/ARBs, and another ensemble is trained to predict the risk when patients are not given ACEI/ARBs. We then employ a voting scheme to aggregate the risk scores of the individual methods and give a final prescription and estimated benefit of treatment.	./cache/cord-313294-ffgo56gl.txt	./txt/cord-313294-ffgo56gl.txt