Nesterov Accelerated ADMM for Fast Diffeomorphic Image Registration

@inproceedings{Thorley2021NesterovAA,
  title={Nesterov Accelerated ADMM for Fast Diffeomorphic Image Registration},
  author={Alexander Thorley and Xi Jia and Hyung Jin Chang and Boyang Liu and Karina V Bunting and Victoria M. Stoll and Antonio de Marvao and Declan P. O’Regan and Georgios V. Gkoutos and Dipak Kotecha and Jinming Duan},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  year={2021}
}
Deterministic approaches using iterative optimisation have been historically successful in diffeomorphic image registration (DiffIR). Although these approaches are highly accurate, they typically carry a significant computational burden. Recent developments in stochastic approaches based on deep learning have achieved sub-second runtimes for DiffIR with competitive registration accuracy, offering a fast alternative to conventional iterative methods. In this paper, we attempt to reduce this… 

Learn2Reg: comprehensive multi-task medical image registration challenge, dataset and evaluation in the era of deep learning

Image registration is a fundamental medical image analysis task, and a wide variety of approaches have been proposed. However, only a few studies have comprehensively compared medical image

U-Net vs Transformer: Is U-Net Outdated in Medical Image Registration?

It is shown that U-Net can outperform transformer-based architectures on inter-subject and atlas-based 3D medical image registration and also outperforms TransMorph on a MICCAI Learn2Reg 2021 challenge dataset.

Structure Unbiased Adversarial Model for Medical Image Segmentation

A simple, yet powerful Structure-Unbiased Adversarial (SUA) network which accounts for both intensity and structural differences between the training and test sets when performing image segmentation is proposed.

FedMed-ATL: Misaligned Unpaired Brain Image Synthesis via Affine Transform Loss

A novel federated self-supervised learning (FedMed) for brain image synthesis is proposed with better stability than other GAN-based algorithms and reduces the demand for deformable registration while encouraging to leverage the misaligned and unpaired data.

FedMed-ATL: Misaligned Unpaired Cross-Modality Neuroimage Synthesis via Affine Transform Loss

A novel federated self-supervised learning (FedMed) for brain image synthesis is proposed with better stability than other GAN-based algorithms and reduces the demand for deformable registration while encouraging to leverage the misaligned and unpaired data.

Fourier-Net: Fast Image Registration with Band-limited Deformation

The FourierNet is proposed, replacing the expansive path in a U-Net style network with a parameter-free model-driven decoder, allowing the unsupervised Fourier-Net to contain fewer parameters and computational operations, resulting in faster inference speeds.

References

SHOWING 1-10 OF 28 REFERENCES

Fast Symmetric Diffeomorphic Image Registration with Convolutional Neural Networks

A novel, efficient unsupervised symmetric image registration method which maximizes the similarity between images within the space of diffeomorphic maps and estimates both forward and inverse transformations simultaneously.

VoxelMorph: A Learning Framework for Deformable Medical Image Registration

VoxelMorph promises to speed up medical image analysis and processing pipelines while facilitating novel directions in learning-based registration and its applications and demonstrates that the unsupervised model’s accuracy is comparable to the state-of-the-art methods while operating orders of magnitude faster.

Diffeomorphic 3D Image Registration via Geodesic Shooting Using an Efficient Adjoint Calculation

The key contribution of this work is to provide an accurate estimation of the so-called initial momentum, which is a scalar function encoding the optimal deformation between two images through the Hamiltonian equations of geodesics.

DeepFLASH: An Efficient Network for Learning-Based Medical Image Registration

  • Jian WangMiaomiao Zhang
  • Computer Science
    2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2020
Experimental results show that the DeepFLASH method is significantly faster than the state-of-the-art deep learning based image registration methods, while producing equally accurate alignment.

Automatic 3D bi-ventricular segmentation of cardiac images by a shape-refined multi-task deep learning approach

A multi-task deep learning approach with atlas propagation is combined to develop a shape-refined bi-ventricular segmentation pipeline for short-axis CMR volumetric images that is robust and capable of producing accurate, high-resolution, and anatomically smooth bi- ventricular 3D models, despite the presence of artifacts in input CMR volumes.

Curvature Based Image Registration

A fully automated, non-rigid image registration algorithm that not only produces accurate and smooth solutions but also allows for an automatic rigid alignment and an implementation based on the numerical solution of the underlying Euler-Lagrange equations.

A Duality Based Approach for Realtime TV-L1 Optical Flow

This work presents a novel approach to solve the TV-L1 formulation, which is based on a dual formulation of the TV energy and employs an efficient point-wise thresholding step.