Research Article
[Retracted] A Computer-Aided Method for Digestive System Abnormality Detection in WCE Images
Table 4
The results of classification on Kvasir dataset.
| Class /metric | Accuracy | FN rate | FP rate | Precision | Recall | F-measure |
| Angiectasia | 0.973 | 0.066 | 0.019 | 0.906 | 0.934 | 0.920 | Blood-fresh | 0.992 | 0.034 | 0.005 | 0.945 | 0.966 | 0.956 | Hematin | 0.999 | 0.167 | 0.000 | 1.000 | 0.833 | 0.909 | Erosion | 0.951 | 0.273 | 0.025 | 0.759 | 0.727 | 0.743 | Erythematous | 0.989 | 0.232 | 0.005 | 0.797 | 0.768 | 0.782 | Foreign bodies | 0.999 | 0.004 | 0.001 | 0.994 | 0.996 | 0.995 | Lymphoid hyperplasia | 0.981 | 0.081 | 0.011 | 0.916 | 0.919 | 0.917 | Polyp | 0.999 | 0.091 | 0.001 | 0.943 | 0.909 | 0.926 | Ulcer | 0.973 | 0.090 | 0.015 | 0.921 | 0.910 | 0.915 | Normal mucosa | 0.934 | 0.169 | 0.040 | 0.832 | 0.831 | 0.831 |
|
|