A prediction model based on DNA methylation biomarkers and radiological characteristics for identifying malignant from benign pulmonary nodules

Table 3 Accuracy and predictive value between four models

Cross Validation	Model	Sensitivity	Specificity	PPV	NPV	Accuracy	AUC
4-fold on training cohort	KNN	0.83	0.86	0.90	0.80	0.83	0.84
	SVM	0.89	0.85	0.89	0.86	0.87	0.92
	RF	0.88	0.85	0.89	0.85	0.87	0.91
	RL	0.91	0.83	0.88	0.89	0.87	0.93
Validated in an independent cohort	KNN	0.93	0.84	0.9	0.9	0.89	0.88
	SVM	0.93	0.93	0.91	0.91	0.93	0.96
	RF	0.91	0.93	0.89	0.89	0.92	0.95
	RL	0.91	0.88	0.88	0.91	0.9	0.96

KNN K-nearest neighbors, SVM support vector machine, RF random forest, RL logistic regression, AUC area under the curve, PPV positive predictive value, NPV negative predictive value

ISSN: 1471-2407