Table 3.
Threshold | Case set A |
Case set B |
||||
---|---|---|---|---|---|---|
ML method | EN algorithm | Naive logit | ML method | EN algorithm | Naive logit | |
0.1 | 0.85 | 0.85 | 0.73 | 0.89 | 0.86 | 0.77 |
0.2 | 0.79 | 0.74 | 0.60 | 0.85 | 0.80 | 0.63 |
0.3 | 0.74 | 0.70 | 0.49 | 0.83 | 0.71 | 0.56 |
0.4 | 0.70 | 0.64 | 0.35 | 0.80 | 0.64 | 0.47 |
0.5 | 0.66 | 0.59 | 0.28 | 0.78 | 0.61 | 0.41 |
0.6 | 0.62 | 0.55 | 0.26 | 0.72 | 0.57 | 0.34 |
0.7 | 0.56 | 0.48 | 0.21 | 0.66 | 0.55 | 0.30 |
0.8 | 0.52 | 0.42 | 0.15 | 0.60 | 0.49 | 0.24 |
0.9 | 0.41 | 0.40 | 0.10 | 0.54 | 0.45 | 0.16 |
EN: Elkan and Noto; ML: maximum likelihood; PA: primary aldosteronism; TPR: true positive rate.