Research Article
Prediction of Early Recurrence of Liver Cancer by a Novel Discrete Bayes Decision Rule for Personalized Medicine
Table 4
Optimal combinations of markers per number of markers and their discrimination performances obtained using training samples.
| Number of markers | Sensitivity | Specificity | Youden index | Combination of markers |
| 3 | 0.79 | 0.50 | 0.29 | Tumor number × tumor size | vp | Liver damage | | | |
| 4 | 0.80 | 0.50 | 0.30 | Tumor number × tumor size | vp | ICG | Liver damage | | |
| 5 | 0.75 | 0.50 | 0.25 | ALB | Tumor number × tumor size | vv | Degree of differentiation | Liver damage | |
| 6 | 0.74 | 0.51 | 0.24 | ALB | Tumor number × tumor size | vp | ICG | Degree of differentiation | Liver damage |
| 7 | — | — | — | | | | | | | 8 | — | — | — | | | | | | | 9 | — | — | — | | | | | | |
|
|