MRI denoising using Deep Learning and Non-local averaging

TL;DR

This paper introduces a two-stage MRI denoising method combining deep learning and non-local means filtering, achieving fast, automatic, and effective noise reduction across various MRI types with spatially varying noise.

Contribution

A novel two-stage MRI denoising approach that integrates deep learning feature regression with non-local averaging, capable of blindly handling different noise patterns.

Findings

Competitive denoising performance compared to state-of-the-art methods.

Significantly faster processing than comparable filters.

Effective on both stationary and spatially varying noise patterns.

Abstract

This paper proposes a novel method for automatic MRI denoising that exploits last advances in deep learning feature regression and self-similarity properties of the MR images. The proposed method is a two-stage approach. In the first stage, an overcomplete patch-based convolutional neural network blindly removes the noise without specific estimation of the local noise variance to produce a preliminary estimation of the noise-free image. The second stage uses this preliminary denoised image as a guide image within a rotationally invariant non-local means filter to robustly denoise the original noisy image. The proposed approach has been compared with related state-of-the-art methods and showed competitive results in all the studied cases while being much faster than comparable filters. We present a denoising method that can be blindly applied to any type of MR image since it can automatically deal with both stationary and spatially varying noise patterns.