Research Article
A Novel Algorithm for Breast Mass Classification in Digital Mammography Based on Feature Fusion
Algorithm 1
Mammography mass benign-malignant classification algorithm.
| (1) | Input: mammography mass image | | (2) | Output: diagnosis results matrix | | (3) | Calculate the deep feature matrix | | (4) | Get textons by convoluting MR8 filter bank with the | | (5) | Calculate the textons-based feature matrix through LBP | | (6) | if use the direct fusion approach then | | (7) | Fusion feature , | | (8) | Train softmax classifier with and predict | | (9) | Return | | (10) | end if | | (11) | if use the fusion after reduction approach then | | (12) | Calculate the subset of by random forest | | (13) | Calculate the subset of by random forest | | (14) | Fusion feature , | | (15) | Train SVM classifier with and predict | | (16) | Return | | (17) | end if |
|
