Training data independent image registration using generative adversarial networks and domain adaptation

作者：

Highlights：

• A dataset independent deep learning based approach for image registration is described.

• Unsupervised domain adaptation techniques through latent space feature encoding is described.

• Registration network trained on lung X-ray images is used to register brain MRI and retinal images.

摘要

•A dataset independent deep learning based approach for image registration is described.•Unsupervised domain adaptation techniques through latent space feature encoding is described.•Registration network trained on lung X-ray images is used to register brain MRI and retinal images.

论文关键词：Registration,Domain adaptation,GANs,X-ray,MRI

论文评审过程：Received 6 November 2018, Revised 17 October 2019, Accepted 13 November 2019, Available online 19 November 2019, Version of Record 26 November 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107109