Table 6.
Performance indices of forecasting models when using the validating dataset (n = 171) to predict recurrence within 10 years after breast cancer surgery (unit: %) *.
| Models | Sensitivity | Specificity | PPV | NPV | Accuracy | AUROC |
|---|---|---|---|---|---|---|
| ANN | 88.90 | 95.52 | 84.21 | 96.97 | 94.12 | 97.62 |
| (86.69–91.11) | (94.87–96.17) | (82.67–85.75) | (95.57–98.38) | (93.01–95.23) | (96.83–98.41) | |
| KNN | 46.15 | 76.66 | 46.15 | 76.67 | 67.44 | 61.40 |
| (35.14–57.16) | (71.48–81.84) | (35.24–57.06) | (65.69–87.65) | (61.12–73.76) | (55.41–67.39) | |
| SVM | 70.37 | 94.35 | 75.50 | 93.13 | 89.79 | 82.40 |
| (65.67–75.07) | (93.24–95.46) | (70.49–80.51) | (92.01–94.25) | (88.12–91.46) | (81.12–83.68) | |
| NBC | 100.00 | 0.00 | 19.01 | 0.00 | 19.01 | 50.00 |
| (99.94–100.00) | (0.00–0.00) | (9.98–28.04) | (0.00–0.00) | (9.57–28.45) | (39.68–60.32) | |
| COX | 20.93 | 0.60 | 6.62 | 0.23 | 5.29 | 10.50 |
| (7.93–33.95) | (0.00–1.01) | (4.32–8.92) | (0.10–0.36) | (3.37–7.21) | (4.78–16.22) | |
| p value | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 |
ANN: artificial neural network; KNN: k-nearest neighbor; SVM: support vector machine; NBC: naïve Bayesian classifier; PPV: positive predictive value; NPV: negative predictive value; AUROC: area under the receiver operating characteristic curve. * 1000 pairs of forecasting models with bootstrapping methods were compared in terms of accuracy in predicting recurrence within 10 years after breast cancer surgery.