Table 3: SARLE performance for the 427 chest CT test reports.
across the 9 labels with manually obtained ground truth. “# Pos” is the number of positive examples for that label in the report test set. F = equally weighted harmonic mean of precision and recall, P = Precision, R = Recall, Acc = Accuracy.
| Label | # Pos | SARLE-Hybrid | SARLE-Rules | ||||||
|---|---|---|---|---|---|---|---|---|---|
| F-score | P | R | Acc | F-score | P | R | Acc | ||
| nodule | 341 | 0.996 | 0.991 | 1 | 0.993 | 0.996 | 0.994 | 0.997 | 0.993 |
| opacity | 213 | 0.995 | 0.995 | 0.995 | 0.995 | 0.998 | 1 | 0.995 | 0.998 |
| atelectasis | 108 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| pleural effusion | 88 | 0.978 | 0.967 | 0.989 | 0.991 | 0.977 | 0.988 | 0.966 | 0.991 |
| consolidation | 78 | 0.969 | 0.951 | 0.987 | 0.988 | 0.975 | 0.963 | 0.987 | 0.991 |
| mass | 55 | 0.915 | 0.857 | 0.982 | 0.977 | 0.956 | 0.931 | 0.982 | 0.988 |
| pericardial effusion | 44 | 0.755 | 0.685 | 0.841 | 0.944 | 0.956 | 0.935 | 0.977 | 0.991 |
| cardiomegaly | 34 | 0.919 | 0.850 | 1 | 0.986 | 0.986 | 0.971 | 1 | 0.998 |
| pneumothorax | 8 | 0.842 | 0.727 | 1 | 0.993 | 0.941 | 0.889 | 1 | 0.998 |