MBD: Multi b-value Denoising of Diffusion Magnetic Resonance Images

TL;DR

This paper presents MBD, a neural network method that denoises diffusion MRI images by leveraging multiple b-values, improving image quality especially in low-redundancy or single-shot acquisitions.

Contribution

The paper introduces a novel CNN-based denoising technique that exploits multi-b-value data, addressing limitations of existing methods in low-redundancy diffusion MRI scenarios.

Findings

Effective noise reduction in high-noise dMRI images

Applicable to low-redundancy and single-shot acquisitions

Prevents blurring while denoising

Abstract

We propose a novel approach to denoising diffusion magnetic resonance images (dMRI) using convolutional neural networks, that exploits the benefits of data acquired at multiple b-values to offset the need for many redundant observations. Denoising is especially relevant in dMRI since noise can have a deleterious impact on both quantification accuracy and image preprocessing. The most successful methods proposed to date, like Marchenko-Pastur Principal Component Analysis (MPPCA) denoising, are tailored to diffusion-weighting repeated for many encoding directions. They exploit high redundancy of the dataset that oversamples the diffusion-encoding direction space, since many directions have collinear components. However, there are many dMRI techniques that do not entail a large number of encoding directions or repetitions, and are therefore less suited to this approach. For example, clinical dMRI exams may include as few as three encoding directions, with low or negligible data redundancy across directions. Moreover, promising new dMRI approaches, like spherical b-tensor encoding (STE), benefit from high b-values while sensitizing the signal to diffusion along all directions in just a single shot. We introduce a convolutional neural network approach that we call multi-b-value-based denoising (MBD). MBD exploits the similarity in diffusion-weighted images (DWI) across different b-values but along the same diffusion encoding direction. It allows denoising of diffusion images with high noise variance while avoiding blurring, and using just a small number input images.