Research Article
Thyroid Diagnosis from SPECT Images Using Convolutional Neural Network with Optimization
Table 2
Hashimoto disease class performance comparison of different methods (percent).
| Network | DenseNet121 | ResNet101 | InceptionV3 | VGG19 | MVGG | GoogleNet | SDAE | Proposed |
| Recall | 97.17 | 96.50 | 92.33 | 91.67 | 90.33 | 93.17 | 94.50 | 98.50 | Precision | 95.57 | 94.92 | 90.67 | 91.21 | 91.71 | 91.49 | 93.41 | 98.50 | Accuracy | 98.17 | 97.83 | 95.71 | 95.71 | 95.54 | 96.13 | 96.96 | 99.25 | Specificity | 98.50 | 98.28 | 96.83 | 97.06 | 97.28 | 97.11 | 97.78 | 99.50 | F1 score | 96.37 | 95.70 | 91.49 | 91.44 | 91.02 | 92.32 | 93.95 | 98.50 |
|
|