Table 2.
Diagnostic performance of classifiers
Tested population | General population | ||||||
Model | Accuracy | Sensitivity | Specificity | PPV | NPV | PPV | NPV |
SVM multiprotein index | |||||||
Learning population | 92.86 | 93.33 | 92.31 | 93.33 | 92.31 | 1.50 | 99.99 |
Testing population | 87.50 | 83.33 | 83.33 | 83.33 | 83.33 | 0.62 | 99.97 |
Logistic regression multiprotein index | |||||||
Learning population | 96.43 | 93.33 | 100.00 | 100.00 | 92.86 | 100.00 | 99.99 |
Testing population | 91.67 | 83.33 | 100.00 | 100.00 | 87.71 | 100.00 | 99.98 |
Positive Predictive Value (PPV) and Negative Predictive Value (NPV) were estimated for the tested population and for the general population (with a prevalence of 1/800).