Evaluation of MRI Denoising Methods Using Unsupervised Learning

Moreno López, Marc and Frederick, Joshua M. and Ventura, Jonathan (2021) Evaluation of MRI Denoising Methods Using Unsupervised Learning. Frontiers in Artificial Intelligence, 4. ISSN 2624-8212

[thumbnail of pubmed-zip/versions/2/package-entries/frai-04-642731.pdf] Text
pubmed-zip/versions/2/package-entries/frai-04-642731.pdf - Published Version

Download (3MB)

Abstract

In this paper we evaluate two unsupervised approaches to denoise Magnetic Resonance Images (MRI) in the complex image space using the raw information that k-space holds. The first method is based on Stein’s Unbiased Risk Estimator, while the second approach is based on a blindspot network, which limits the network’s receptive field. Both methods are tested on two different datasets, one containing real knee MRI and the other consists of synthetic brain MRI. These datasets contain information about the complex image space which will be used for denoising purposes. Both networks are compared against a state-of-the-art algorithm, Non-Local Means (NLM) using quantitative and qualitative measures. For most given metrics and qualitative measures, both networks outperformed NLM, and they prove to be reliable denoising methods.

Item Type: Article
Subjects: Journal Eprints > Multidisciplinary
Depositing User: Managing Editor
Date Deposited: 14 Mar 2023 09:12
Last Modified: 24 Jun 2024 04:15
URI: http://repository.journal4submission.com/id/eprint/947

Actions (login required)

View Item
View Item