Table 2 Classification metrics across different models
From: Data mining polycystic ovary morphology in electronic medical record ultrasound reports
Model | Accuracy | Class label | Sensitivity | Specificity | PPV | NPV |
|---|---|---|---|---|---|---|
Gradient boosted trees | 0.961 (0.9471, 0.9721) | PCOM-absent | 0.9977 | 0.9982 | 0.9977 | 0.9982 |
Unidentifiable | 0.9474 | 0.9705 | 0.9387 | 0.9748 | ||
PCOM-present | 0.9146 | 0.9761 | 0.9259 | 0.9723 | ||
Rules-based classifier | 0.976 (0.9645, 0.9846) | PCOM-absent | 0.9885 | 0.9982 | 0.9977 | 0.9912 |
Unidentifiable | 0.966 | 0.9852 | 0.9690 | 0.9838 | ||
PCOM-present | 0.9668 | 0.9829 | 0.9472 | 0.9894 |