Table 2 Quantitative performance of five different tissue region classifiers on validation dataset of 105 patches.

From: A Novel Digital Score for Abundance of Tumour Infiltrating Lymphocytes Predicts Disease Free Survival in Oral Squamous Cell Carcinoma

Classifiers

Accuracy

Sensitivity

Specificity

F1-Score

AUC

TRC-1 (ResNet5030)

94.04

88.19

96.03

88.08

97.88

TRC-2 (DenseNet31)

95.40

90.80

96.94

90.78

98.56

TRC-3 (Inception-v332)

95.84

91.91

97.25

91.70

98.86

TRC-4 (Xception33)

95.94

92.16

97.32

91.88

98.83

TRC-5 (MobileNet34)

96.31

92.66

97.55

92.62

98.91

Back to article page