Table 3.

Performances of logistic regression and deep neural network models for main outcomes.

Variable	CVD Hospital				CVD Death				CVD Total (Hospital + Death)
	Missing Value (+)		Missing Value (−)		Missing Value (+)		Missing Value (−)		Missing Value (+)		Missing Value (−)
	LR	DNN	LR	DNN	LR	DNN	LR	DNN	LR	DNN	LR	DNN
Accuracy	0.646	0.824	0.655	0.863	0.777	0.886	0.780	0.925	0.673	0.843	0.659	0.860
F1 score	0.648	0.811	0.656	0.854	0.782	0.889	0.783	0.924	0.672	0.834	0.658	0.852
Precision	0.645	0.877	0.654	0.912	0.764	0.870	0.773	0.935	0.675	0.879	0.662	0.903
Recall	0.650	0.754	0.658	0.803	0.800	0.908	0.793	0.912	0.669	0.794	0.654	0.807
AUC	0.646	0.907	0.655	0.932	0.777	0.959	0.780	0.979	0.673	0.923	0.660	0.933

CVD: cardiovascular disease, DNN: deep neural network, LR: logistic regression.