Abstract:Diffusion models with continuous stochastic differential equations (SDEs) have shown superior performances in image generation. It can serve as a deep generative prior to solving the inverse problem in magnetic resonance (MR) reconstruction. However, low-frequency regions of $k$-space data are typically fully sampled in fast MR imaging, while existing diffusion models are performed throughout the entire image or $k$-space, inevitably introducing uncertainty in the reconstruction of low-frequency regions. Additionally, existing diffusion models often demand substantial iterations to converge, resulting in time-consuming reconstructions. To address these challenges, we propose a novel SDE tailored specifically for MR reconstruction with the diffusion process in high-frequency space (referred to as HFS-SDE). This approach ensures determinism in the fully sampled low-frequency regions and accelerates the sampling procedure of reverse diffusion. Experiments conducted on the publicly available fastMRI dataset demonstrate that the proposed HFS-SDE method outperforms traditional parallel imaging methods, supervised deep learning, and existing diffusion models in terms of reconstruction accuracy and stability. The fast convergence properties are also confirmed through theoretical and experimental validation. Our code and weights are available at <a class="link-external link-https" href="https://github.com/Aboriginer/HFS-SDE" rel="external noopener nofollow">this https URL</a>.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to solve the problems faced by fast imaging techniques in magnetic resonance imaging (MRI). Specifically, the paper focuses on the following two main challenges: 1. **Consistency problem in the low - frequency region**: - In fast MRI imaging, the low - frequency region of k - space data is usually fully sampled, while the existing diffusion models operate on the entire image or k - space, which may lead to increased uncertainty in the reconstruction of the low - frequency region. Therefore, how to ensure that the low - frequency region of the reconstructed image is consistent with the actually collected low - frequency region is a key issue. 2. **Problem of slow convergence**: - Existing diffusion models usually require a large number of iterations to converge, resulting in a long - lasting reconstruction process. This is an important bottleneck for fast MRI imaging in clinical applications. To solve these problems, the authors propose a new stochastic differential equation (SDE), called the high - frequency frequency - domain diffusion model (HFS - SDE). This method ensures the determinacy of the low - frequency region by performing the diffusion process in the high - frequency frequency - domain and accelerates the sampling process of the reverse diffusion. Experimental results show that HFS - SDE is superior to traditional parallel imaging methods, supervised deep learning methods, and existing diffusion models in terms of reconstruction accuracy and stability. In addition, theoretical and experimental verifications also confirm that HFS - SDE has a faster convergence speed.

High-Frequency Space Diffusion Models for Accelerated MRI

High-Frequency Space Diffusion Model for Accelerated MRI

CDiffMR: Can We Replace the Gaussian Noise with K-Space Undersampling for Fast MRI?

Correlated and Multi-frequency Diffusion Modeling for Highly Under-sampled MRI Reconstruction

Sub-DM:Subspace Diffusion Model with Orthogonal Decomposition for MRI Reconstruction

Multifold Acceleration of Diffusion MRI Via Deep Learning Reconstruction from Slice-Undersampled Data.

SPIRiT-Diffusion: Self-Consistency Driven Diffusion Model for Accelerated MRI

Diffusion model based on generalized map for accelerated MRI

Highly Accelerated MRI via Implicit Neural Representation Guided Posterior Sampling of Diffusion Models

K-space Cold Diffusion: Learning to Reconstruct Accelerated MRI without Noise

Reconstructing High-Quality Diffusion MRI Data from Orthogonal Slice-Undersampled Data Using Graph Convolutional Neural Networks

Towards performant and reliable undersampled MR reconstruction via diffusion model sampling

Autoregressive Image Diffusion: Generation of Image Sequence and Application in MRI

Physics-Informed DeepMRI: k-Space Interpolation Meets Heat Diffusion

Learning to reconstruct accelerated MRI through K-space cold diffusion without noise

Fast Controllable Diffusion Models for Undersampled MRI Reconstruction

Zero-shot Dynamic MRI Reconstruction with Global-to-local Diffusion Model

Score-based Diffusion Models With Self-supervised Learning For Accelerated 3D Multi-contrast Cardiac Magnetic Resonance Imaging

Clinically Feasible Diffusion Reconstruction for Highly-Accelerated Cardiac Cine MRI

Diffusion Modeling with Domain-conditioned Prior Guidance for Accelerated MRI and qMRI Reconstruction

Spatiotemporal Diffusion Model with Paired Sampling for Accelerated Cardiac Cine MRI