CT imaging-based machine learning model: a potential modality for predicting low-risk and high-risk groups of thymoma: “Impact of surgical modality choice”

Table 3 ROC results with six machine-learning classifiers of validation set

Risk groups	Statistical measures	KNN	SVM	XGBoost	RF	LR	DT
Low	AUC	0.943	0.857	0.8	0.693	0.943	0.436
	95% CI	0.85–1	0.66–1	0.58–1	0.45–0.93	0.74–1	0.18–0.69
	Sensitivity	0.9	0.8	0.7	0.8	0.8	0.3
	Specificity	0.86	0.86	0.71	0.43	0.86	0.57
High	AUC	0.943	0.857	0.8	0.693	0.943	0.436
	95% CI	0.85–1	0.66–1	0.58–1	0.45–0.93	0.74–1.00	0.18–0.69
	Sensitivity	0.86	0.86	0.71	0.43	0.86	0.57
	Specificity	0.9	0.8	0.7	0.8	0.8	0.3

KNN k-nearest neighbor, SVM support vector machine, XGBoost eXtreme Gradient Boosting, RF random forest, LR logistic regression, DT decision tree

ISSN: 1477-7819