Skip to main content
. 2022 Oct 5;12:979358. doi: 10.3389/fonc.2022.979358

Table 6.

Predictive performances of the models predicting the Ki-67 status in patients with BC.

Model Set AUC SEN (%) SPE (%) ACC (%) PPV (%) NPV (%) TP FP FN TN
Clinical Training 0.578 73.1% 45.3% 64.1% 73.7% 44.6% 98 35 36 29
Test 0.665 85.5% 54.2% 76.7% 82.8% 59.1% 53 11 9 13
Rad-Score Training 0.793 74.6% 76.6% 75.3% 87.0% 59.0% 100 15 34 49
Test 0.798 88.7% 70.8% 83.7% 88.7% 70.8% 55 7 7 17
Nomogram Training 0.790 77.6% 73.4% 76.3% 86.0% 61.0% 104 17 30 47
Test 0.808 90.3% 70.8% 84.9% 88.9% 73.9% 56 7 6 17

AUC, area under the curve; SEN, sensitivity; SPE, specificity; ACC, accuracy; PPV, positive predictive value; NPV, negative predictive value; Rad-Score, radiomics score; TP, true positive; FP, false positive; FN, false negative; TN, true negative.