Research Article
Chest X-Ray Images to Differentiate COVID-19 from Pneumonia with Artificial Intelligence Techniques
Table 4
The classification performance for COVID-19 with SURF features.
| Classifier model | TPR/recall (%) | FNR (%) | PPV/precision (%) | FDR (%) | F1-score (%) |
| Fine tree | 90.00 | 10.00 | 85.70 | 14.30 | 87.80 | Linear discriminant | 97.50 | 2.50 | 97.50 | 2.50 | 97.50 | Logistic regression | 97.50 | 2.50 | 97.50 | 2.50 | 97.50 | Linear SVM | 97.50 | 2.50 | 98.70 | 1.30 | 98.10 | Coarse Gaussian SVM | 98.80 | 1.20 | 97.50 | 2.50 | 98.15 | Cosine kNN | 78.80 | 21.30 | 98.40 | 1.60 | 87.51 | Ensemble subspace discriminant | 97.50 | 2.50 | 97.50 | 2.50 | 97.50 |
|
|