A cGAN-based tumor segmentation method for breast ultrasound images

Abstract

Objective. This paper proposes a conditional GAN (cGAN)-based method to perform data enhancement of ultrasound images and segmentation of tumors in breast ultrasound images, which improves the reality of the enhenced breast ultrasound image and obtains a more accurate segmentation result. Approach. We use the idea of generative adversarial training to accomplish the following two tasks: (1) In this paper, we use generative adversarial networks to generate a batch of samples with labels from the perspective of label-generated images to expand the dataset from a data enhancement perspective. (2) In this paper, we use adversarial training instead of postprocessing steps such as conditional random fields to enhance higher-level spatial consistency. In addition, this work proposes a new network, EfficientUNet, based on U-Net, which combines ResNet18, an attention mechanism and a deep supervision technique. This segmentation model uses the residual network as an encoder to retain the lost information in the original encoder and can avoid the gradient disappearance problem to improve the feature extraction ability of the model, and it also uses deep supervision techniques to speed up the convergence of the model. The channel-by-channel weighting module of SENet is then used to enable the model to capture the tumor boundary more accurately. Main results. The paper concludes with experiments to verify the validity of these efforts by comparing them with mainstream methods on Dataset B. The Dice score and IoU score reaches 0.8856 and 0.8111, respectively. Significance. This study successfully combines cGAN and optimized EfficientUNet for the segmentation of breast tumor ultrasound images. The conditional generative adversarial network has a good performance in data enhancement, and the optimized EfficientUNet makes the segmentation more accurate.

Chest x-ray diagnosis via spatial-channel high-order attention representation learning

“…Deep learning networks have been widely employed on many visual learning tasks, such as object detection (Elakkiya et al 2021, Wang et al 2023b, scene-text spotting (Ronen et al 2022), etc. Meanwhile, they have also been used for many medical image representation and analysis tasks, such as radiation report generation (Chen et al 2022), medical image segmentation (You et al 2022(You et al , 2023, etc. Since the large-scale chest x-ray images are usually available (Wang et al 2017, Irvin et al 2019, various CNN based deep neural network methods (Kim et al 2021 have been successfully applied for x-ray image learning and recognition problems.…”

Automatic breast ultrasound (ABUS) tumor segmentation based on global and local feature fusion

“…Breast cancer is a common cancer among women worldwide and is now considered as a leading cause of death in many countries (Xie et al 2020). Due to convenience, safety, and low-cost, ultrasound imaging has become a popular modality for breast cancer detection and diagnosis (You et al 2023). Compared with handheld ultrasound, automatic breast ultrasound (ABUS) employs a standardized process to capture the 3D volume, which reduces subjectivity and provides more information, being a promising technology for breast cancer examinations (Ma et al 2022).…”