DiffuseReg: Denoising Diffusion Model for Obtaining Deformation Fields in Unsupervised Deformable Image Registration

TL;DR

DiffuseReg introduces a diffusion-based deformable image registration method that enhances transparency, control, and accuracy by denoising deformation fields with a Swin Transformer network, outperforming existing methods on medical datasets.

Contribution

The paper presents a novel diffusion-based registration approach that denoises deformation fields, incorporating a Swin Transformer and a regularization technique for improved interpretability and performance.

Findings

Outperforms existing diffusion registration methods by 1.32 in Dice score.

Enables real-time observability and adjustment during registration.

Uses a Swin Transformer-based denoising network for better image integration.

Abstract

Deformable image registration aims to precisely align medical images from different modalities or times. Traditional deep learning methods, while effective, often lack interpretability, real-time observability and adjustment capacity during registration inference. Denoising diffusion models present an alternative by reformulating registration as iterative image denoising. However, existing diffusion registration approaches do not fully harness capabilities, neglecting the critical sampling phase that enables continuous observability during the inference. Hence, we introduce DiffuseReg, an innovative diffusion-based method that denoises deformation fields instead of images for improved transparency. We also propose a novel denoising network upon Swin Transformer, which better integrates moving and fixed images with diffusion time step throughout the denoising process. Furthermore, we enhance control over the denoising registration process with a novel similarity consistency regularization. Experiments on ACDC datasets demonstrate DiffuseReg outperforms existing diffusion registration methods by 1.32 in Dice score. The sampling process in DiffuseReg enables real-time output observability and adjustment unmatched by previous deep models.