id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
10_1101-2021_01_08_425976	Ahuja, Yuri	Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET) Using Electronic Health Record Data	2021	23	.pdf	application/pdf	7047	861	50	Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET) Using Electronic Health Record Data yielding highly biased disease risk estimators if used as event time labels (Cipparone and others, 2015; Uno and SCORNET utilizes current status labels while also employing a robust semisupervised imputation approach on the extensive unlabeled set to maximize survival estimation efficiency. further illustrate the utility of SCORNET in clinical applications, we apply it to a real-world EHR study estimating the risk of heart failure among rheumatoid arthritis patients in Section 4. existing survival function estimators with current status data: 1) parametric Weibull Accelerated Failure Time confidence intervals constructed with the bootstrap (red) and plug-in (blue) standard error estimators in various simulated settings with = = 200 observed current status labels. set with observed current status labels, the SCORNET estimator serves as a robust and efficient alternative	./cache/10_1101-2021_01_08_425976.pdf	./txt/10_1101-2021_01_08_425976.txt