Table 3 Performance of AI classification of texture and morphological features utilizing five different classifiers of COVID-19 (N = 130) vs viral pneumonia (N = 145)
Classifier | Sensitivity | Specificity | PPV | NPV | Accuracy | AUC (LB, UP) | p value |
|---|---|---|---|---|---|---|---|
XGB-L | 97.44 | 97.67 | 97.44 | 97.67 | 97.56 | 0.98 (0.96, 1.00) | 2.00e−16 |
XGB-Tree | 94.87 | 95.35 | 94.87 | 94.87 | 95.12 | 0.98 (0.95, 1.00) | 2.00e−16 |
CART (DT) | 94.87 | 93.02 | 92.50 | 95.24 | 93.90 | 0.94 (0.89, 0.99) | 2.00e−16 |
KNN | 69.63 | 88.37 | 84.38 | 76.00 | 79.27 | 0.85 (0.77, 0.94) | 4.38e−07 |
Naïve Bayes | 64.10 | 95.35 | 92.59 | 74.55 | 80.49 | 0.93 (0.88, 0.98) | 1.20e−07 |