id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-346712-ky5gt3pu	Soltan, A. A.	Artificial intelligence driven assessment of routinely collected healthcare data is an effective screening test for COVID-19 in patients presenting to hospital	2020-07-08		.txt	text/plain	5759	315	46	In this study, we develop two early-detection models to identify COVID-19 using routinely collected data typically available within one hour (laboratory tests, blood gas and vital signs) during 115,394 emergency presentations and 72,310 admissions to hospital. The results are presented as percentages for categorical data and as median and interquartile range for age Table 3 shows a summary of the relative performance of models trained using each independent feature set at identifying presentations due to COVID-19, reported in terms of AUROC achieved during stratified 10-fold cross validation alongside standard deviations (SDs). . https://doi.org/10.1101/2020.07.07.20148361 doi: medRxiv preprint Table 5 : Assessment of performance (SD) of (a) our ED and (b) Admissions models, calibrated to 70, 80 and 90% sensitivities during training, at identifying COVID-19 amongst patients presenting to or admitted hospital emergency departments in a heldout test set with 50% assumed prevalence.	./cache/cord-346712-ky5gt3pu.txt	./txt/cord-346712-ky5gt3pu.txt