Score-based Self-supervised MRI Denoising

TL;DR

This paper introduces Corruption2Self, a score-based self-supervised MRI denoising framework that effectively learns from noisy data, preserving fine details and achieving state-of-the-art results without requiring high-quality labels.

Contribution

The paper proposes a novel score-based self-supervised MRI denoising method with a generalized denoising score matching loss, noise level reparameterization, and multi-contrast extension, outperforming existing self-supervised approaches.

Findings

Achieves state-of-the-art results among self-supervised MRI denoising methods.

Provides competitive performance compared to supervised methods across various noise levels.

Effectively preserves fine spatial features while reducing noise.

Abstract

Magnetic resonance imaging (MRI) is a powerful noninvasive diagnostic imaging tool that provides unparalleled soft tissue contrast and anatomical detail. Noise contamination, especially in accelerated and/or low-field acquisitions, can significantly degrade image quality and diagnostic accuracy. Supervised learning based denoising approaches have achieved impressive performance but require high signal-to-noise ratio (SNR) labels, which are often unavailable. Self-supervised learning holds promise to address the label scarcity issue, but existing self-supervised denoising methods tend to oversmooth fine spatial features and often yield inferior performance than supervised methods. We introduce Corruption2Self (C2S), a novel score-based self-supervised framework for MRI denoising. At the core of C2S is a generalized denoising score matching (GDSM) loss, which extends denoising score matching to work directly with noisy observations by modeling the conditional expectation of higher-SNR images given further corrupted observations. This allows the model to effectively learn denoising across multiple noise levels directly from noisy data. Additionally, we incorporate a reparameterization of noise levels to stabilize training and enhance convergence, and introduce a detail refinement extension to balance noise reduction with the preservation of fine spatial features. Moreover, C2S can be extended to multi-contrast denoising by leveraging complementary information across different MRI contrasts. We demonstrate that our method achieves state-of-the-art performance among self-supervised methods and competitive results compared to supervised counterparts across varying noise conditions and MRI contrasts on the M4Raw and fastMRI dataset.