Table 1.

| Comparison of the performance of multiple prediction models

Methods		Accuracy	Precision	Recall	F1	AUC
Random Forest	Training	0.851	1.000	0.238	0.384	0.619
	Test	0.808	0.909	0.068	0.127	0.533
Logistic Regression	Training	0.825	0.629	0.256	0.364	0.610
	Test	0.808	0.567	0.260	0.357	0.605
Lasso Regression	Training	0.825	0.762	0.148	0.248	0.568
	Test	0.813	0.710	0.151	0.249	0.567
Radial SVM [40]	Training	0.515	0.970	0.491	0.652	0.701
	Test	0.337	0.896	0.204	0.333	0.586
	Val	0.806	0.849	0.920	0.883	0.642
Gradient boosting [40]	Training	0.851	0.934	0.899	0.916	0.690
	test	0.718	0.822	0.816	0.819	0.574
	Val	0.828	0.885	0.905	0.895	0.682
Bayes [40]	Training	0.567	0.965	0.553	0.703	0.649
	Test	0.465	0.861	0.405	0.551	0.562
	Val	0.828	0.891	0.895	0.893	0.713
Linear regression [40]	Training	0.801	0.943	0.835	0.886	0.599
	Test	0.679	0.828	0.763	0.794	0.541
	Val	0.788	0.885	0.842	0.863	0.689
Linear SVM [40]	Training	0.337	0.896	0.205	0.333	0.586
	Test	0.467	0.861	0.407	0.553	0.586
	Val	0.818	0.873	0.906	0.889	0.676
Sofa Score [13] DCQMFF (proposed)	All data	0.752	0.371	0.327	0.348	0.807
	Training	0.822	0.822	0.821	0.822	0.896
	Test	0.821	0.812	0.812	0.812	0.885
	Val	0.775	0.764	0.754	0.759	0.849
CNN (Proposed)	Training	0.928	0.924	0.856	0.888	0.953
	Test	0.924	0.887	0.845	0.865	0.947
	Val	0.834	0.825	0.818	0.821	0.909