DP-MDM: Detail-Preserving MR Reconstruction via Multiple Diffusion Models

TL;DR

This paper introduces a novel MRI reconstruction method that employs multiple diffusion models and a multi-scale cascade architecture to better preserve intricate details in medical images, outperforming existing techniques.

Contribution

It proposes a detail-preserving MRI reconstruction framework using multiple diffusion models in k-space and an inverted pyramid structure for multi-scale data representation.

Findings

Outperforms existing MRI reconstruction methods on clinical and public datasets.

Effectively captures intricate details in MRI images.

Utilizes a cascade refinement approach for improved image quality.

Abstract

Detail features of magnetic resonance images play a cru-cial role in accurate medical diagnosis and treatment, as they capture subtle changes that pose challenges for doc-tors when performing precise judgments. However, the widely utilized naive diffusion model has limitations, as it fails to accurately capture more intricate details. To en-hance the quality of MRI reconstruction, we propose a comprehensive detail-preserving reconstruction method using multiple diffusion models to extract structure and detail features in k-space domain instead of image do-main. Moreover, virtual binary modal masks are utilized to refine the range of values in k-space data through highly adaptive center windows, which allows the model to focus its attention more efficiently. Last but not least, an inverted pyramid structure is employed, where the top-down image information gradually decreases, ena-bling a cascade representation. The framework effective-ly represents multi-scale sampled data, taking into ac-count the sparsity of the inverted pyramid architecture, and utilizes cascade training data distribution to repre-sent multi-scale data. Through a step-by-step refinement approach, the method refines the approximation of de-tails. Finally, the proposed method was evaluated by con-ducting experiments on clinical and public datasets. The results demonstrate that the proposed method outper-forms other methods.