Dilated and soft attention‐guided convolutional neural network for breast cancer histology images classification. Issue 4 (3rd December 2021)

• Record Type: Journal Article
• Authors: Zhong, Yutong; Piao, Yan; Zhang, Guohui
• Is Part Of: Microscopy research and technique. Volume 85:Issue 4(2022)
• Page Start: 1248
• Page End: 1257
• Publication Date: 2021-12-03
• Subjects: attention mechanism -- breast cancer -- classification -- dilated convolution -- histopathological microscopic images -- deep learning; Electron microscopy -- Technique -- Periodicals; Microscopy -- Periodicals; Microscopy -- Technique -- Periodicals; 502.825
• DOI: 10.1002/jemt.23991
• Languages: English
• ISSNs: 1059-910X

Abstract

Abstract: Breast cancer is one of the most common types of cancer in women, and histopathological imaging is considered the gold standard for its diagnosis. However, the great complexity of histopathological images and the considerable workload make this work extremely time‐consuming, and the results may be affected by the subjectivity of the pathologist. Therefore, the development of an accurate, automated method for analysis of histopathological images is critical to this field. In this article, we propose a deep learning method guided by the attention mechanism for fast and effective classification of haematoxylin and eosin‐stained breast biopsy images. First, this method takes advantage of DenseNet and uses the feature map's information. Second, we introduce dilated convolution to produce a larger receptive field. Finally, spatial attention and channel attention are used to guide the extraction of the most useful visual features. With the use of fivefold cross‐validation, the best model obtained an accuracy of 96.47% on the BACH2018 dataset. We also evaluated our method on other datasets, and the experimental results demonstrated that our model has reliable performance. This study indicates that our histopathological image classifier with a soft attention‐guided deep learning model for breast cancer shows significantly better results than the latest methods. It has great potential as an effective tool for automatic evaluation of digital histopathological microscopic images for computer‐aided diagnosis. Abstract : We propose a neural network model guided by the soft attention mechanism that can realize multiple‐classification tasks for pathology images of patients with breast cancer Compared with attention guidance, which focus only on the spatial domain or the channel domain, our proposed hybrid domain attention module improves the correlation between feature maps and leads to effective feature representation We introduce a dilated convolution kernel during feature extraction that can enlarge the receptive field while maintaining the image resolution, which greatly enhances the particularity of medical images