Research Article
Automated Detection and Characterization of Colon Cancer with Deep Convolutional Neural Networks
Table 3
Using our DCNN model to compare its performance with other transfer learning models.
| Model | Precision | Recall | Fl score | Training accuracy | Testing accuracy |
| VGG-16 | 88.80 | 89.33 | 90.02 | 98.95 | 92.90 | ResNet101V2 | 86.89 | 88.09 | 83.47 | 97.93 | 90.45 | EfficientNetB0 | 81.45 | 87.90 | 73.50 | 96.80 | 91.80 | DcnscNctl21 | 81.32 | 64 | 71.69 | 98.03 | 91.10 | MobileNetV2 | 80.89 | 78.90 | 69.02 | 92.46 | 81.26 | ResNet5O | 86.02 | 83.80 | 88.20 | 96.90 | 89.80 | Proposed DCNN | 100 | 99.59 | 99.80 | 99.87 | 99.80 |
|
|