Skip to content

MungoMeng/Registration-AAN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

61 Commits
 
 
 
 
 
 
 
 

Repository files navigation

AAN: Appearance Adjustment Networks for medical image registration

Deformable image registration is fundamental for many medical image analyses. A key obstacle for accurate image registration is the variations in image appearance such as the variations in textures, intensities, and noises. These variations are highly apparent in medical images, especially in brain images where registration is frequently used. Recently, deep learning-based registration methods (DLRs), using deep neural networks, have gained computational efficiency that is several orders of magnitude faster than traditional optimization-based registration methods (ORs). DLRs have degraded adaptability to appearance variations and perform poorly when image pairs (fixed/moving images) have large appearance variations. In this study, we propose an Appearance Adjustment Network (AAN) to enhance DLRs’ adaptability to appearance variations. Our AAN, when integrated into a DLR, provides appearance transformations to reduce the appearance variations during registration. In addition, we propose an anatomy-constrained loss function through which our AAN can generate anatomy-preserving transformations. Our AAN has been purposely designed to be readily inserted into a wide range of DLRs and can be trained cooperatively in an unsupervised and end-to-end manner.
For more details, please refer to our paper. [Elsevier] [arXiv]

Workflow

workflow

Publication

If this repository helps your work, please kindly cite our paper:

  • Mingyuan Meng, Lei Bi, Michael Fulham, David Dagan Feng, Jinman Kim, "Enhancing Medical Image Registration via Appearance Adjustment Networks," NeuroImage, vol. 259, pp. 119444, 2022, doi: 10.1016/j.neuroimage.2022.119444. [Elsevier] [arXiv]

About

AAN: Appearance Adjustment Networks to enhance deep learning-based image registration

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages