Table 2 Summery of the classifier performance evaluation utilizing the best feature subset, i.e.[‘ENT’, ‘MAX’, ‘CON’, ‘MEA’]. The best result in each column is underlined.

From: A priori prediction of tumour response to neoadjuvant chemotherapy in breast cancer patients using quantitative CT and machine learning

Classifier/score

\({{\varvec{A}}{\varvec{U}}{\varvec{C}}}_{0.632+}\) (%)

Accuracy (%)

AUC (%)

Specificity (%)

Sensitivity (%)

Precision (%)

F-Score (%)

Time (Sec)

SVM

80.32

75.27

74.79

78.26

72.34

77.27

74.73

6.6

MLP

79.76

73.12

74.12

73.91

72.34

73.91

73.12

2,231

RF

84.15

80.65

80.23

84.78

76.60

83.72

80.00

139

Adaboost-SVM

83.11

78.49

78.03

82.16

74.47

81.40

77.78

1,493

Adaboost-DT

88.72

83.67

84.23

88.10

80.36

89.95

84.91

265

Hybrid

84.29

79.57

79.83

84.78

74.47

83.33

78.65

471

Back to article page