Research Article
Thyroid Diagnosis from SPECT Images Using Convolutional Neural Network with Optimization
Table 4
Normal class performance comparison of different methods (percent).
| Network | DenseNet121 | ResNet101 | InceptionV3 | VGG19 | MVGG | GoogleNet | SDAE | Proposed |
| Recall | 100.00 | 100.00 | 99.50 | 99.17 | 96.83 | 100.00 | 96.33 | 100.00 | Precision | 100.00 | 100.00 | 95.52 | 96.75 | 94.93 | 97.24 | 97.14 | 100.00 | Accuracy | 100.00 | 100.00 | 98.71 | 98.96 | 97.92 | 99.29 | 98.38 | 100.00 | Specificity | 100.00 | 100.00 | 98.44 | 98.89 | 98.28 | 99.06 | 99.06 | 100.00 | F1 score | 100.00 | 100.00 | 97.47 | 97.94 | 95.87 | 98.60 | 96.74 | 100.00 |
|
|