Unsupervised Echocardiography Registration through Patch-based MLPs and Transformers

TL;DR

This paper introduces patch-based MLP and Transformer frameworks for unsupervised echocardiography image registration, demonstrating their effectiveness and efficiency compared to CNN-based methods, with improved preservation of volume changes.

Contribution

The work provides a novel benchmark and demonstrates that patch-based MLP and Transformer models can achieve superior or comparable registration performance in echocardiography.

Findings

Patch-based models outperform CNN in preserving volume changes.

Models achieve high registration accuracy on large datasets.

Patch-based methods are efficient in time and space complexity.

Abstract

Image registration is an essential but challenging task in medical image computing, especially for echocardiography, where the anatomical structures are relatively noisy compared to other imaging modalities. Traditional (non-learning) registration approaches rely on the iterative optimization of a similarity metric which is usually costly in time complexity. In recent years, convolutional neural network (CNN) based image registration methods have shown good effectiveness. In the meantime, recent studies show that the attention-based model (e.g., Transformer) can bring superior performance in pattern recognition tasks. In contrast, whether the superior performance of the Transformer comes from the long-winded architecture or is attributed to the use of patches for dividing the inputs is unclear yet. This work introduces three patch-based frameworks for image registration using MLPs and transformers. We provide experiments on 2D-echocardiography registration to answer the former question partially and provide a benchmark solution. Our results on a large public 2D echocardiography dataset show that the patch-based MLP/Transformer model can be effectively used for unsupervised echocardiography registration. They demonstrate comparable and even better registration performance than a popular CNN registration model. In particular, patch-based models better preserve volume changes in terms of Jacobian determinants, thus generating robust registration fields with less unrealistic deformation. Our results demonstrate that patch-based learning methods, whether with attention or not, can perform high-performance unsupervised registration tasks with adequate time and space complexity. Our codes are available https://gitlab.inria.fr/epione/mlp\_transformer\_registration