Machine learning classification of texture features of MRI breast tumor and peri-tumor of combined pre- and early treatment predicts pathologic complete response

Table 4 MRI texture analysis using combined tp1 + tp2 MRI data and molecular subtypes using different machine learning classifiers

Method	Sens.	Spec.	PPV	NPV	Accuracy	AUC	P-value
Rusboosted Tree	94.59	92.31	97.22	85.71	94	0.98 (0.94, 1.0)	0.00029
Decision Tree	1.00	0	74.00	NA	90	0.92 (0.81, 1.0)	0.00459
SVM coarse Gaussian	94.59	0	72.92	0	74	0.72 (0.55, 0.88)	0.5738
Kernel Naïve Bayes	70.27	69.23	86.66	45.00	70	0.70 (0.55, 0.85)	0.7925
KNN	94.59	0	72.92	0	70	0.60 (0.43, 0.76)	0.7924

The number in parenthesis showed the 95% confidence intervals. These data were tumor contour without morphological dilation

ISSN: 1475-925X