Classification of Breast Cancer Lesions in Ultrasound Images by Using Attention Layer and Loss Ensemble in Deep Convolutional Neural Networks

Abstract

The reliable classification of benign and malignant lesions in breast ultrasound images can provide an effective and relatively low-cost method for the early diagnosis of breast cancer. The accuracy of the diagnosis is, however, highly dependent on the quality of the ultrasound systems and the experience of the users (radiologists). The use of deep convolutional neural network approaches has provided solutions for the efficient analysis of breast ultrasound images. In this study, we propose a new framework for the classification of breast cancer lesions with an attention module in a modified VGG16 architecture. The adopted attention mechanism enhances the feature discrimination between the background and targeted lesions in ultrasound. We also propose a new ensembled loss function, which is a combination of binary cross-entropy and the logarithm of the hyperbolic cosine loss, to improve the model discrepancy between classified lesions and their labels. This combined loss function optimizes the network more quickly. The proposed model outperformed other modified VGG16 architectures, with an accuracy of 93%, and also, the results are competitive with those of other state-of-the-art frameworks for the classification of breast cancer lesions. Our experimental results show that the choice of loss function is highly important and plays a key role in breast lesion classification tasks. Additionally, by adding an attention block, we could improve the performance of the model.

Ensemble Deep-Learning-Enabled Clinical Decision Support System for Breast Cancer Diagnosis and Classification on Ultrasound Images

“…U-net is a type of CNN framework that was developed for the segmentation of life science images containing constrained trained data. Yousef Kalaf et al [13] presented an architecture for the classification of breast cancer with an attention mechanism in an adapted VGG16 framework. The adapted attention model distinguishes between features of the background and targeted lesions in ultrasound image.…”

CTG-Net: Cross-task guided network for breast ultrasound diagnosis

“…In this approach, a unified model is adopted to learn feature representations and classify BUS image visual patterns, such as combining DCNNs with image processing techniques to help classification [15]. The latest tweaks to neural networks, such as deep transfer learning [16][17][18][19] and the attention model [8,20,21], have also been employed. However, the limited availability of data with annotations has been hindering progress.…”

Application of Artificial Intelligence to Ultrasonography

“…Wu et al reviewed breast ultrasound AI technology to detect breast nodule (15). Using a deep learning model (GoogleNet model of CNN), Kalafi et al analyzed ultrasound breast images to identify benign and malignant tumors with an accuracy of 93% (16). This method can classify malignant lesions in a short amount of time and aids the radiologists' diagnosis of malignant lesions.…”