Learning a Model-Driven Variational Network for Deformable Image Registration
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2021
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913563910602752 |
|---|---|
| author | Jia, Xi Thorley, Alexander Chen, Wei Qiu, Huaqi Shen, Linlin Styles, Iain B Chang, Hyung Jin Leonardis, Ales de Marvao, Antonio O'Regan, Declan P. Rueckert, Daniel Duan, Jinming |
| author_facet | Jia, Xi Thorley, Alexander Chen, Wei Qiu, Huaqi Shen, Linlin Styles, Iain B Chang, Hyung Jin Leonardis, Ales de Marvao, Antonio O'Regan, Declan P. Rueckert, Daniel Duan, Jinming |
| contents | Data-driven deep learning approaches to image registration can be less accurate than conventional iterative approaches, especially when training data is limited. To address this whilst retaining the fast inference speed of deep learning, we propose VR-Net, a novel cascaded variational network for unsupervised deformable image registration. Using the variable splitting optimization scheme, we first convert the image registration problem, established in a generic variational framework, into two sub-problems, one with a point-wise, closed-form solution while the other one is a denoising problem. We then propose two neural layers (i.e. warping layer and intensity consistency layer) to model the analytical solution and a residual U-Net to formulate the denoising problem (i.e. generalized denoising layer). Finally, we cascade the warping layer, intensity consistency layer, and generalized denoising layer to form the VR-Net. Extensive experiments on three (two 2D and one 3D) cardiac magnetic resonance imaging datasets show that VR-Net outperforms state-of-the-art deep learning methods on registration accuracy, while maintains the fast inference speed of deep learning and the data-efficiency of variational model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2105_12227 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Learning a Model-Driven Variational Network for Deformable Image Registration Jia, Xi Thorley, Alexander Chen, Wei Qiu, Huaqi Shen, Linlin Styles, Iain B Chang, Hyung Jin Leonardis, Ales de Marvao, Antonio O'Regan, Declan P. Rueckert, Daniel Duan, Jinming Computer Vision and Pattern Recognition Image and Video Processing Data-driven deep learning approaches to image registration can be less accurate than conventional iterative approaches, especially when training data is limited. To address this whilst retaining the fast inference speed of deep learning, we propose VR-Net, a novel cascaded variational network for unsupervised deformable image registration. Using the variable splitting optimization scheme, we first convert the image registration problem, established in a generic variational framework, into two sub-problems, one with a point-wise, closed-form solution while the other one is a denoising problem. We then propose two neural layers (i.e. warping layer and intensity consistency layer) to model the analytical solution and a residual U-Net to formulate the denoising problem (i.e. generalized denoising layer). Finally, we cascade the warping layer, intensity consistency layer, and generalized denoising layer to form the VR-Net. Extensive experiments on three (two 2D and one 3D) cardiac magnetic resonance imaging datasets show that VR-Net outperforms state-of-the-art deep learning methods on registration accuracy, while maintains the fast inference speed of deep learning and the data-efficiency of variational model. |
| title | Learning a Model-Driven Variational Network for Deformable Image Registration |
| topic | Computer Vision and Pattern Recognition Image and Video Processing |
| url | https://arxiv.org/abs/2105.12227 |