AUTHOR=Moreno López Marc , Frederick Joshua M. , Ventura Jonathan 

TITLE=Evaluation of MRI Denoising Methods Using Unsupervised Learning

JOURNAL=Frontiers in Artificial Intelligence

VOLUME=Volume 4 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2021.642731

DOI=10.3389/frai.2021.642731

ISSN=2624-8212

ABSTRACT=In this paper we evaluate two unsupervised approaches to denoise Magnetic Resonance Images (MRI) in k-space. 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 k-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 all given metrics, both networks outperformed NLM and based on qualitative measures, they prove to be reliable denoising methods.