Research Article
Gastric Pathology Image Classification Using Stepwise Fine-Tuning for Deep Neural Networks
Table 2
Performances of the proposed two-stage fine-tuning using tissue-wise data.
| Data size | Scheme | CNN architecture | VGG-16 | AlexNet | GoogLeNet (Inception V3) | AUC | ACC | Precision | Recall | AUC | ACC | Precision | Recall | AUC | ACC | Precision | Recall |
| Small | One stage | 0.879 | 0.793 | 0.863 | 0.695 | 0.828 | 0.723 | 0.74 | 0.72 | 0.838 | 0.753 | 0.75 | 0.75 | Two stage (tissue) | 0.914 | 0.829 | 0.865 | 0.781 | 0.844 | 0.761 | 0.76 | 0.76 | 0.877 | 0.772 | 0.79 | 0.77 |
| Large | One stage | 0.936 | 0.836 | 0.957 | 0.703 | 0.867 | 0.794 | 0.80 | 0.79 | 0.881 | 0.779 | 0.79 | 0.78 | Two stage (tissue) | 0.963 | 0.881 | 0.869 | 0.898 | 0.920 | 0.837 | 0.84 | 0.84 | 0.934 | 0.862 | 0.86 | 0.86 |
|
|