Table 3.
Performance of diagnostic support models constructed using combinations of candidate biomarkers with various classifiers
| Features | Classifier | Accuracy | Sensitivity | Precision | AUC |
|---|---|---|---|---|---|
| PTPRC+ASUN | Decision tree | 91.49 % | 91.5 % | 97.7 % | 0.943 |
| PTPRC+ASUN+DHX29 | Random Forest | 93.62 % | 93.6 % | 93.6 % | 0.982 |
| PTPRC+ASUN+DHX29 | SVM | 95.74 % | 95.7 % | 96.2 % | 0.969 |
| PTPRC+ASUN+DHX29 | Naïve Bayes | 97.87 % | 97.9 % | 98 % | 0.979 |
Sensitivity: TP/(TP+FN); Precision: TP/(TP+FP); performance was evaluated by 5-fold cross-validation