Medical images from multicentres often suffer from the domain shift problem, which makes the deep learning models trained on one domain usually fail to generalize well to another. One of the potential solutions for the problem is the generative adversarial network (GAN), which has the capacity to translate images between different domains. Nevertheless, the existing GAN-based approaches are prone to fail at preserving image-objects in image-to-image (I2I) translation, which reduces their practicality on domain adaptation tasks. In this regard, a novel GAN (namely IB-GAN) is proposed to preserve image-objects during cross-domain I2I adaptation. Specifically, we integrate the information bottleneck constraint into the typical cycle-consistency-based GAN to discard the superfluous information ( e.g., domain information) and maintain the consistency of disentangled content features for image-object preservation. The proposed IB-GAN is evaluated on three tasks—polyp segmentation using colonoscopic images, the segmentation of optic disc and cup in fundus images and the whole heart segmentation using multi-modal volumes. We show that the proposed IB-GAN can generate realistic translated images and remarkably boost the generalization of widely used segmentation networks ( e.g., U-Net).

中文翻译：

---

超越互信息：使用信息瓶颈约束进行域适应的生成对抗网络


来自多中心的医学图像经常遇到域转移问题，这使得在一个域上训练的深度学习模型通常无法很好地推广到另一个域。该问题的潜在解决方案之一是生成对抗网络（GAN），它能够在不同领域之间转换图像。然而，现有的基于 GAN 的方法很容易无法在图像到图像（I2I）转换中保留图像对象，这降低了它们在域适应任务中的实用性。在这方面，提出了一种新颖的 GAN（即 IB-GAN）来在跨域 I2I 适应期间保留图像对象。具体来说，我们将信息瓶颈约束集成到典型的基于循环一致性的 GAN 中，以丢弃多余的信息（例如领域信息）并保持解开的内容特征的一致性，以实现图像对象保存。所提出的 IB-GAN 在三个任务上进行评估：使用结肠镜图像进行息肉分割、眼底图像中视盘和视杯的分割以及使用多模态体积的整个心脏分割。我们表明，所提出的 IB-GAN 可以生成逼真的翻译图像，并显着促进广泛使用的分割网络（例如 U-Net）的泛化。