Research Article
NSCR-Based DenseNet for Lung Tumor Recognition Using Chest CT Image
Table 4
Comparison of CT results for different network models and classification algorithms.
| Network model | Classification algorithm | Accuracy (%) | Sensitivity (%) | Specificity (%) | -score (%) | MCC (%) |
| AlexNet | SVM | 97.50 | 97.40 | 97.60 | 97.50 | 95.00 | SRC | 96.32 | 96.36 | 96.28 | 96.32 | 92.64 | NSCR | 98.52 | 98.44 | 98.60 | 98.52 | 97.04 |
| GoogleNet | SVM | 97.90 | 97.76 | 98.04 | 97.90 | 95.80 | SRC | 98.02 | 99.20 | 96.84 | 98.04 | 96.07 | NSCR | 98.82 | 99.20 | 98.44 | 98.82 | 97.64 |
| DenseNet-201 | SVM | 98.26 | 98.32 | 98.20 | 98.26 | 96.52 | SRC | 98.32 | 99.60 | 97.04 | 98.34 | 96.67 | NSCR | 99.10 | 99.60 | 98.60 | 99.10 | 98.20 |
|
|