DCE-MRI Pharmacokinetic-Based Phenotyping of Invasive Ductal Carcinoma: A Radiomic Study for Prediction of Histological Outcomes

Table 3

Results of multivariable analysis. For each classification task, the model with the higher AUC was chosen and its order, AUC, sensitivity, specificity, and accuracy were reported together with the standard error on a 95% confidence interval over all bootstrap sample.

	Order	AUC	Sensitivity	Specificity	Accuracy
ER+/ER−	4	0.826 ± 0.006	0.833 ± 0.004	0.587 ± 0.016	0.804 ± 0.003
PR+/PR−	8	0.875 ± 0.009	0.895 ± 0.005	0.730 ± 0.019	0.882 ± 0.005
HER2+/HER2−	4	0.838 ± 0.006	0.623 ± 0.014	0.825 ± 0.005	0.785 ± 0.004
TN/NTN	10	0.876 ± 0.007	0.660 ± 0.022	0.896 ± 0.004	0.881 ± 0.004
Ki67+/Ki67−	2	0.811 ± 0.005	0.641 ± 0.006	0.736 ± 0.007	0.677 ± 0.004
lowGrade/highGrade	5	0.895 ± 0.006	0.735 ± 0.012	0.865 ± 0.006	0.807 ± 0.004