Table 2 Predicting endometriosis using the RF algorithm for each feature data
From: Application of machine learning techniques in the diagnosis of endometriosis
Feature | Accuracy(%) | Sensitivity(%) | Specificity (%) | AUC | P | 95%CI |
|---|---|---|---|---|---|---|
ca125 | 75.80 | 79.30 | 74.10 | 0.822 | 0.041 | 0.655, 0.844 |
ca125 + ca199 | 79.31 | 86.21 | 75.86 | 0.841 | 0.006 | 0.6929,0.8725 |
ca125 + APTT | 78.16 | 75.86 | 79.31 | 0.789 | 0.0132 | 0.6802,0.8631 |
ca125 + NLR | 78.16 | 86.21 | 74.14 | 0.850 | 0.0132 | 0.6802,0.8631 |
ca125 + HB | 74.17 | 93.10 | 65.50 | 0.841 | 0.067 | 0.6425,0.8342 |