Machine learning-based evaluation of prognostic factors for mortality and relapse in patients with acute lymphoblastic leukemia: a comparative simulation study

Table 5 Performance criteria of machine learning algorithms for predicting relapse of childhood acute lymphoblastic leukemia

Dataset	Method	Sensitivity		Specificity		PPV		NPV		Accuracy		AUC
Dataset	Method	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Train	SVM	0.90	0.033	0.84	0.046	0.85	0.031	0.90	0.026	0.87	0.017	0.87	0.018
	LS-SVM	0.98	0.011	0.76	0.082	0.81	0.054	0.98	0.015	0.87	0.043	0.87	0.044
	ANN	0.89	0.044	0.88	0.032	0.88	0.029	0.89	0.037	0.89	0.031	0.88	0.031
	NB	0.92	0.028	0.67	0.047	0.73	0.026	0.90	0.028	0.79	0.023	0.79	0.022
	Bagging	0.97	0.019	0.93	0.024	0.93	0.020	0.97	0.017	0.95	0.014	0.95	0.014
	Boosting	1.00	0.00	1.00	0.00	1.00	0.00	1.00	0.00	1.00	0.00	1.00	0.00
	DT	0.94	0.031	0.88	0.048	0.89	0.039	0.94	0.027	0.91	0.023	0.91	0.022
	RF	1.00	0.00	1.00	0.00	1.00	0.00	1.00	0.00	1.00	0.00	1.00	0.00
	LR	0.72	0.048	0.67	0.043	0.69	0.029	0.70	0.030	0.69	0.028	0.69	0.028
	LDA	0.74	0.058	0.67	0.049	0.69	0.025	0.73	0.033	0.71	0.026	0.71	0.025
Test	SVM	0.87	0.052	0.68	0.094	0.74	0.069	0.83	0.068	0.77	0.046	0.77	0.045
	LS-SVM	0.96	0.030	0.62	0.098	0.72	0.063	0.93	0.042	0.79	0.051	0.79	0.049
	ANN	0.82	0.060	0.65	0.094	0.70	0.075	0.78	0.073	0.73	0.053	0.74	0.052
	NB	0.88	0.052	0.59	0.086	0.69	0.063	0.82	0.068	0.74	0.045	0.74	0.043
	Bagging	0.91	0.041	0.66	0.091	0.73	0.062	0.87	0.056	0.79	0.044	0.79	0.045
	Boosting	0.97	0.019	0.71	0.081	0.77	0.058	0.96	0.029	0.84	0.042	0.84	0.041
	DT	0.86	0.060	0.63	0.088	0.70	0.062	0.82	0.073	0.74	0.045	0.74	0.044
	RF	0.64	0.083	0.96	0.025	0.73	0.070	0.94	0.037	0.80	0.050	0.81	0.083
	LR	0.70	0.067	0.59	0.096	0.63	0.073	0.66	0.073	0.64	0.047	0.64	0.046
	LDA	0.71	0.076	0.59	0.087	0.64	0.064	0.67	0.079	0.65	0.040	0.65	0.039

ISSN: 1472-6947