An efficient framework for lesion segmentation in ultrasound images using global adversarial learning and region-invariant loss

Van Manh, Xiaohong Jia, Wufeng Xue, Wenwen Xu, Zihan Mei, Yijie Dong, Jianqiao Zhou, Ruobing Huang, Dong Ni

Abstract: Highlights•A segmentation framework that focuses on global information and invariant features.•A global adversarial learning mechanism with a self-attention-based discriminator.•Embedding lesion texture information into a region-invariant loss.