Research Article
Automatic Prediction of Meningioma Grade Image Based on Data Amplification and Improved Convolutional Neural Network
Table 5
Impact of improved network structure on the error rate of meningioma classification.
| Network layer | Feature map | Convolution kernel | Filter size | Iteration rate | Error rate | Conv1 | Conv2 | Conv3 | Conv1 | Conv2 | Conv3 | Pool1 | Pool2 | Pool3 |
| 7 | 4 | 8 | 16 | 5 × 5 | 5 × 5 | 4 × 4 | 2 × 2 | 2 × 2 | 2 × 2 | 0.003 | 83.33% | 7 | 5 | 10 | 20 | 5 × 5 | 5 × 5 | 4 × 4 | 2 × 2 | 2 × 2 | 2 × 2 | 0.0001 | 53.33% | 7 | 5 | 10 | 20 | 5 × 5 | 5 × 5 | 2 × 2 | 4 × 4 | 3 × 3 | 2 × 2 | 0.0005 | 60.00% | 7 | 6 | 12 | 24 | 5 × 5 | 5 × 5 | 2 × 2 | 4 × 4 | 3 × 3 | 2 × 2 | 0.0001 | 23.33% | 7 | 6 | 12 | 24 | 5 × 5 | 5 × 5 | 4 × 4 | 2 × 2 | 2 × 2 | 2 × 2 | 0.0001 | 16.67% | 7 | 6 | 12 | 24 | 5 × 5 | 5 × 5 | 4 × 4 | 2 × 2 | 2 × 2 | 2 × 2 | 0.00005 | 16.67% | 7 | 8 | 16 | 32 | 5 × 5 | 5 × 5 | 4 × 4 | 2 × 2 | 2 × 2 | 2 × 2 | 0.0001 | 26.67% | 7 | 8 | 16 | 32 | 9 × 9 | 5 × 5 | 2 × 2 | 4 × 4 | 2 × 2 | 2 × 2 | 0.0001 | 56.67% |
|
|