Lesion Segmentation in Ultrasound Using Semi-Pixel-Wise Cycle Generative Adversarial Nets

Abstract

Breast cancer is the most common invasive cancer with the highest cancer occurrence in females. Handheld ultrasound is one of the most efficient ways to identify and diagnose the breast cancer. The area and the shape information of a lesion is very helpful for clinicians to make diagnostic decisions. In this study we propose a new deep-learning scheme, semi-pixel-wise cycle generative adversarial net (SPCGAN) for segmenting the lesion in 2D ultrasound. The method takes the advantage of a fully convolutional neural network (FCN) and a generative adversarial net to segment a lesion by using prior knowledge. We compared the proposed method to a fully connected neural network and the level set segmentation method on a test dataset consisting of 32 malignant lesions and 109 benign lesions. Our proposed method achieved a Dice similarity coefficient (DSC) of 0.92 while FCN and the level set achieved 0.90 and 0.79 respectively. Particularly, for malignant lesions, our method increases the DSC (0.90) of the fully connected neural network to 0.93 significantly (p <0.001). The results show that our SPCGAN can obtain robust segmentation results. The framework of SPCGAN is particularly effective when sufficient training samples are not available compared to FCN. Our proposed method may be used to relieve the radiologists' burden for annotation.

A Multi-Task Learning Framework for Automated Segmentation and Classification of Breast Tumors From Ultrasound Images

“…Over the years, there were many Convolutional Neural Network (CNN) based models designed for ultrasound image segmentation. 19 – 22 Xing et al 19 developed a semi-pixel-wise cycle model using Generative Adversarial Network (GAN) and CNN for tumor segmentation. The anatomy based image segmentation is also very important in ultrasound image analysis to reduce the false positives in image segmentation.…”

A Brief Survey on Breast Cancer Diagnostic With Deep Learning Schemes Using Multi-Image Modalities

“…Recent developments in computational techniques, significant advancement in image-processing technology, and prevalence of DMG images have opened the opportunity to resolve the early diagnosing of breast abnormalities using DL schemes [23], [115], [116]. The existing ML approaches are imperfect for precise detection of breast densities; however, the DL approaches to deliver the auspicious development in mass segmentation to overcome the false-positive ratio (FPR).…”

Ultrasound tissue classification: a review

“…Quite often, the deep learning techniques are combined with post-processing or other traditional techniques to boost the classification performance. In ultrasound images, both patch-based approach (Kaizhi et al 2014;Gustavo et al 2012;Smistad et al 2017;Lekadir et al 2017;Feng et al 2018;Jang et al 2017;Patra and Alison 2020;Zhu et al 2017;Ravishankar et al 2016;Zyuzin et al 2018;Jabbar et al 2016;Mishra et al 2018) and FCN Yap 2017;Oktay et al 2018;Milletari et al 2017;Sundaresan et al 2017;Wang et al 2019;Chiang et al 2019;Azzopardi et al 2020;Zhang et al 2020;Fujioka et al 2019;Liao et al 2019;Xing et al 2020;Andreassen et al 2019) are applied in various applications. In , Liu et al (2017a), the CNN was used with shape modeling to achieve better performance.…”