Development and validation of a clinicoradiomic nomogram to assess the HER2 status of patients with invasive ductal carcinoma

Table 2 Performances of the six machine learning classifiers for predicting HER2 status in the validation cohort

Radiomics classifier	Accuracy	True positive rate (Sensitivity)	True negative rate (Specificity)	Threshold	AUC(95%CI)
LR	0.750	0.893	0.621	0.514	0.810 (0.709–0.905)
LDA	0.733	0.829	0.718	0.547	0.801 (0.701–0.901)
SVM	0.756	0.821	0.724	0.545	0.840 (0.758–0.922)
RF	0.744	0.857	0.689	0.546	0.826 (0.738–0.914)
NB	0.755	0.786	0.741	0.527	0.788 (0.694–0.882)
XGB	0.710	0.857	0.637	0.494	0.790 (0.688–0.891)

LR Logistic Regression, LDA Linear Discriminant Analysis, SVM Support Vector Machine, RF Random Forest, NB Naive Bayesian, XGB XGBoost

ISSN: 1471-2407