Table 4.

Summary of LSTM prediction performance on different imputation methods

	Precision	Recall	F1 score	AUROC
Median	0.8184 ± 0.0085a	0.7865 ± 0.0058	0.8023 ± 0.0020	0.9369 ± 0.0038a
Mean	0.8183 ± 0.0073	0.7851 ± 0.0036	0.8014 ± 0.0031	0.8977 ± 0.0048
MICE	0.8128 ± 0.0082	0.8017 ± 0.0070	0.8072 ± 0.0037a	0.9002 ± 0.0042
KNN	0.7984 ± 0.0094	0.8034 ± 0.0073a	0.8009 ± 0.0039	0.8854 ± 0.0032

AUROC: area under the receiver-operating characteristic curve.

^aBest result for the metric.