|
The paper employs deep learning to classify breast cancer histopathological image into normal, benign and malignant subclasses in situ carcinoma and invasivecarcinoma categories. The classification is mainly based on cells' density, variability, and organization along with overall tissue structure and morphology. Smaller and larger patches of histological images are extracted that includes cell-level and tissue-level features. Here, Patches are screened by Clustering algorithm and CNN is used to select the discriminative patches. The proposed approach is applied to the multi-class classification of breast cancer histology images.It achieves initial test achieves of 95% accuracy and on the overall test,88.89% accuracy.
|
