Skip to main content

Table 4 Performance comparison of different models for predicting pathological grades in the training cohort

From: Exploration of the relationship between tumor-infiltrating lymphocyte score and histological grade in breast cancer

Model

AUC (95%)

SEN (95%)

SPE (95%)

PPV (95%)

NPV (95%)

CLI

0.746

(0.696–0.792)

70.49

(64.9–75.7)

76.60

(62.0-87.7)

94.86

(91.6–96.9)

29.75

(25.0–35.0)

TILs-WG

0.613 (0.559–0.665)

40.63

(34.9–46.5)

78.72 (64.3–89.3)

92.1

(86.9–95.4)

17.8

(15.3–20.5)

TILs-score

0.747

(0.697–0.793)

79.17

(74.0-83.7)

63.83

(48.5–77.3)

93.06

(90.1–95.2)

33.33

(26.8–40.6)

Nomogram

0.802

(0.755–0.843)

61.46

(55.6–67.1)

87.23

(74.3–95.2)

96.72

(93.3–98.4)

26.97

(23.5–30.7)

  1. Abbreviations AUC, area under the curve; SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive value