Learning a Model-Driven Variational Network for Deformable Image Registration

05/25/2021
by   Xi Jia, et al.
7

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.

READ FULL TEXT

page 1

page 3

page 6

page 10

research
06/25/2022

Non-iterative Coarse-to-fine Registration based on Single-pass Deep Cumulative Learning

Deformable image registration is a crucial step in medical image analysi...
research
12/05/2018

Enhancing Label-Driven Deep Deformable Image Registration with Local Distance Metrics for State-of-the-Art Cardiac Motion Tracking

While deep learning has achieved significant advances in accuracy for me...
research
09/26/2021

Nesterov Accelerated ADMM for Fast Diffeomorphic Image Registration

Deterministic approaches using iterative optimisation have been historic...
research
07/07/2023

Unsupervised Segmentation of Fetal Brain MRI using Deep Learning Cascaded Registration

Accurate segmentation of fetal brain magnetic resonance images is crucia...
research
09/15/2020

F3RNet: Full-Resolution Residual Registration Network for Multimodal Image Registration

Multimodal deformable image registration is essential for many image-gui...
research
04/18/2016

Most Likely Separation of Intensity and Warping Effects in Image Registration

This paper introduces a class of mixed-effects models for joint modeling...
research
07/16/2018

Iterative Residual Network for Deep Joint Image Demosaicking and Denoising

Modern digital cameras rely on sequential execution of separate image pr...

Please sign up or login with your details

Forgot password? Click here to reset