Optimal shrinkage denoising breaks the noise floor in high-resolution diffusion MRI

Khoi Huynh; Wei-Tang Chang; Ye Wu; Pew-Thian Yap

doi:10.1016/j.patter.2024.100954

Optimal shrinkage denoising breaks the noise floor in high-resolution diffusion MRI

Patterns (N Y). 2024 Mar 14;5(4):100954. doi: 10.1016/j.patter.2024.100954. eCollection 2024 Apr 12.

Authors

Khoi Huynh^{1

2}, Wei-Tang Chang^{1

2}, Ye Wu^{1

2}, Pew-Thian Yap^{1

2}

Affiliations

¹ Department of Radiology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA.
² Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA.

Abstract

The spatial resolution attainable in diffusion magnetic resonance (MR) imaging is inherently limited by noise. The weaker signal associated with a smaller voxel size, especially at a high level of diffusion sensitization, is often buried under the noise floor owing to the non-Gaussian nature of the MR magnitude signal. Here, we show how the noise floor can be suppressed remarkably via optimal shrinkage of singular values associated with noise in complex-valued k-space data from multiple receiver channels. We explore and compare different low-rank signal matrix recovery strategies to utilize the inherently redundant information from multiple channels. In combination with background phase removal, the optimal strategy reduces the noise floor by 11 times. Our framework enables imaging with substantially improved resolution for precise characterization of tissue microstructure and white matter pathways without relying on expensive hardware upgrades and time-consuming acquisition repetitions, outperforming other related denoising methods.

Keywords: dMRI; diffusion magnetic resonance imaging; noise removal; optimal shrinkage.

Abstract

Grants and funding