Table 3 Discrimination of different models in predicting blastocyst formation
From: Does conventional morphological evaluation still play a role in predicting blastocyst formation?
Training set | Validation set | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
Model | AUC (95%CI) | Cutoff | Sensitivity(95%CI) | Specificity (95%CI) | PPV(95%CI) | NPV(95%CI) | Accuracy(95%CI) | AUC (95%CI) | PPV(95%CI) | NPV(95%CI) | Accuracy(95%CI) |
GEEa | 0.771(0.768–0.774) | 0.51 | 0.804(0.801–0.808) | 0.625(0.62–0.629) | 0.743(0.739–0.746) | 0.703(0.698–0.708) | 0.728(0.725–0.731) | 0.779(0.77–0.787) | 0.761(0.751–0.77) | 0.721(0.707–0.735) | 0.747(0.739–0.755) |
LASSOb | 0.775(0.772–0.778) | 0.55 | 0.793(0.789–0.796) | 0.636(0.631–0.64) | 0.746(0.742–0.749) | 0.695(0.69–0.699) | 0.726(0.723–0.729) | 0.78(0.771–0.789) | 0.77(0.761–0.78) | 0.699(0.686–0.713) | 0.744(0.736–0.752) |
Xgboostb | 0.783(0.78–0.785) | 0.57 | 0.8(0.796–0.803) | 0.635(0.63–0.639) | 0.747(0.743–0.75) | 0.701(0.697–0.706) | 0.729(0.726–0.732) | 0.754(0.745–0.763) | 0.773(0.763–0.783) | 0.645(0.632–0.659) | 0.719(0.711–0.727) |