AUTHOR=Han Tingting , Wu Jun , Luo Wenting , Wang Huiming , Jin Zhe , Qu Lei 

TITLE=Review of Generative Adversarial Networks in mono- and cross-modal biomedical image registration

JOURNAL=Frontiers in Neuroinformatics

VOLUME=Volume 16 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/neuroinformatics/articles/10.3389/fninf.2022.933230

DOI=10.3389/fninf.2022.933230

ISSN=1662-5196

ABSTRACT=Biomedical image registration refers to aligning corresponding anatomical structures among different images, which is critical to many tasks such as brain atlas building, tumor growth monitoring, and image fusion based medical diagnosis. However, due to intrinsic variations of the intensity, texture, and anatomy resulted from different imaging modalities, different sample preparation methods, or different developmental stages of the imaged subject, the high-throughput biomedical image registration still remains a challenging task. Recently, the Generative Adversarial Networks (GAN) have attracted increasing interest in both the mono- and the cross-modal biomedical image registrations due to their special ability to eliminate the modal variance and their adversarial training strategy. In this paper, we provide a comprehensive survey on the GAN based mono- and cross-modal biomedical image registration methods. According to the different implementation strategies, we organize the GAN based mono- and cross-modal biomedical image registration methods into four categories: the modality translation, the symmetric learning, the adversarial strategies, and the joint training. The key concepts, the main contributions, and the advantages and disadvantages of the different strategies are summarized and discussed. Finally, we analyze the statistics of all the cited works from different points of view and reveal future trends of the GAN-based biomedical image registration studies.