Learning a Model-Driven Variational Network for Deformable Image Registration

doi:10.1109/TMI.2021.3108881

Learning a Model-Driven Variational Network for Deformable Image Registration

2022-01-01

20

OA

AI

DOI：10.1109/TMI.2021.3108881

Original

Original request for help

Save

Abstract

En 中文

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 issue and meanwhile retain the fast inference speed of deep learning, we propose VR-Net, a novel cascaded variational network for unsupervised deformable image registration. Using a 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 and the other one being 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 (termed generalized denoising layer) to formulate the denoising problem. Finally, we cascade the three neural layers multiple times to form our 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, whilst maintaining the fast inference speed of deep learning and the data-efficiency of variational models.

Keywords:

Iterative methods

Strain

Image registration

Deformable models

Optimization

Deep learning

Noise reduction

Convolutional neural network

image registration

unsupervised learning

variationalmodel

variational neural network

Learning a Model-Driven Variational Network for Deformable Image Registration

Abstract

Journal

Organization