Research Article
Is Intensity Inhomogeneity Correction Useful for Classification of Breast Cancer in Sonograms Using Deep Neural Network?
Table 2
Classification results of the BUSIS database (the displayed value: original/corrected).
| ā | SDAE | AlexNet | Inception v3 | ResNet | DenseNet |
| TP | 17/18 | 10/16 | 13/15 | 16/16 | 15/17 | TN | 13/15 | 12/14 | 14/16 | 13/10 | 14/13 | FP | 7/5 | 8/6 | 6/4 | 7/10 | 6/7 | FN | 3/2 | 10/4 | 7/5 | 4/4 | 5/3 | Precision | 0.71/0/78 | 0.56/0.73 | 0.68/0.79 | 0.69/0.61 | 0.71/0.71 | Recall | 0.85/0.9 | 0.5/0.8 | 0.65/0.75 | 0.8/0.8 | 0.75/0.85 | Specificity | 0.65/0.75 | 0.6/0.7 | 0.7/0.8 | 0.65/0.5 | 0.7/0.65 | Accuracy | 0.75/0.83 | 0.55/0.75 | 0.68/0.78 | 0.73/0.65 | 0.73/0.75 | F-measure | 0.77/0.83 | 0.53/0.76 | 0.67/0.77 | 0.74/0.70 | 0.73/0.77 |
|
|