US2Mask: Image-to-mask generation learning via a conditional GAN for cardiac ultrasound image segmentation

Gang Wang, Mingliang Zhou, Xin Ning, Prayag Tiwari, Haobo Zhu, Guang Yang, Choon Hwai Yap

Abstract: Highlights•We propose an image-to-mask generation framework for cardiac US segmentation.•We propose a new approach based on a Conditional Generative Adversarial Network.•We design a paired image scheme to enhance the discriminator’s ability.•Extensive experiments on three large cardiac US datasets are conducted.