Cancer is one of the leading causes of disease and mortality in the worldwide. Breast cancer is one of the most prevalent and severe types of cancer in women, making it a prominent research area in medical science.Since histopathological images contain sufficient phenotypic information, they play an indispensable role in the diagnosis and treatment of breast cancers. To improve the accuracy and objectivity of Breast Histopathological Image Analysis (BHIA), deep learning architecture approaches are used in the segmentation and classification tasks of breast histopathological images. In this paper, preliminary experiment was carried out using the deep learning method on Breast Cancer Histopathological Database (BreakHis). On the basis of the BreakHis dataset, a high-precision breast classification approach based on convolutional neural network is developed and the accuracy of the proposed CNN model is 92.7%. Comparing the proposed CNN model with state-of-the-art models, although it achieved lower accuracy in some of them, it achieved this result with fewer parameters. This result shows that the proposed model is much better than other models in terms of computational complexity. Considering the results obtained and the values in the literature, future studies can be carried out to increase the performance values in breast cancer classification.
Anahtar Kelimeler: Breast cancer, Histopathology images, Convolutional Neural Network, classification
|
