Research Article
Combined CNN and Pixel Feature Image for Fatty Liver Ultrasound Image Classification
Table 2
The comparison of the proposed method with other three methods.
| | Methods | Layers | Filters | Filter size | FLOPs (M) | Params (M) | Accuracy | Specificity | Sensitivity |
| | Proposed method | 3 | 198 | | 2448 | 2376 | 92% | 95% | 83% | | VggNet [18] | 16 | 512 | | 1055 | 134 | 82% | 88% | 78% | | Inception-v3 [36] | 22 | 192 | | 234 | 21 | 85% | 90% | 79% | | GLCM-svm [24] | — | — | — | — | — | 79% | 81% | 46% | | Gabor’s filters-svm [26] | — | — | — | — | — | 54% | 55% | 46% | | Wavelet-svm [28] | — | — | — | — | — | 80% | 82% | 79% |
|
|
