Classification of breast cancer histology images using Convolutional Neural Networks

Abstract

Breast cancer is one of the main causes of cancer death worldwide. The diagnosis of biopsy tissue with hematoxylin and eosin stained images is non-trivial and specialists often disagree on the final diagnosis. Computer-aided Diagnosis systems contribute to reduce the cost and increase the efficiency of this process. Conventional classification approaches rely on feature extraction methods designed for a specific problem based on field-knowledge. To overcome the many difficulties of the feature-based approaches, deep learning methods are becoming important alternatives. A method for the classification of hematoxylin and eosin stained breast biopsy images using Convolutional Neural Networks (CNNs) is proposed. Images are classified in four classes, normal tissue, benign lesion, in situ carcinoma and invasive carcinoma, and in two classes, carcinoma and non-carcinoma. The architecture of the network is designed to retrieve information at different scales, including both nuclei and overall tissue organization. This design allows the extension of the proposed system to whole-slide histology images. The features extracted by the CNN are also used for training a Support Vector Machine classifier. Accuracies of 77.8% for four class and 83.3% for carcinoma/non-carcinoma are achieved. The sensitivity of our method for cancer cases is 95.6%.

Fig 1. Examples of microscopy image patches from the used dataset [3].

Nuclei and cytoplasm appear purple and pinkish, respectively, due to the hematoxylin and eosin staining. A normal tissue; B benign abnormality; C malignant carcinoma in situ; D malignant invasive carcinoma.

Fig 2. Histology image normalization.

A and C original images; B and D images after normalization.

Fig 3. Convolutional Neural Network architecture, according to Table 1.

The original image has 512 × 512 pixels and 3 RGB channels. Orange and purple squares represent the convolutional and max-pooling kernels, respectively.

Fig 4. Activation examples for the first (A, B) and second (C) layers of the Convolutional Neural Network.

Different structures with diagnostic relevance are analyzed.

Fig 5. 2D projection of the training patches and their activations on different layers of the CNN using t-SNE [29].

A training patches; B last convolutional layer; C second fully-connected layer. Diamond shapes represent test images.