Fig. 4 | Scientific Reports

Fig. 4

From: The value of intratumoral and peritumoral ultrasound radiomics model constructed using multiple machine learning algorithms for non-mass breast cancer

Fig. 4

Performance of combined models in the validation group. The optimal radiomic models for the validation group, including the intratumoral and peritumoral areas at 1 mm, 2 mm, 3 mm, 4 mm, and 5 mm, were combined with independent clinical and ultrasound predictors to obtain the ROC curve (A), P–R curve (B), DCA curve (C), and calibration curve (D). ROC: receiver operating characteristic, AUC: area under curve, CI: confidence interval, P-R: precision- recall, DCA: decision curve analysis.

Back to article page