DC Field | Value | Language |
---|---|---|
dc.contributor.author | Chung, Hyungjin | ko |
dc.contributor.author | Lee, Eun Sun | ko |
dc.contributor.author | Ye, Jong Chul | ko |
dc.date.accessioned | 2023-05-13T05:01:48Z | - |
dc.date.available | 2023-05-13T05:01:48Z | - |
dc.date.created | 2023-05-12 | - |
dc.date.issued | 2023-04 | - |
dc.identifier.citation | IEEE TRANSACTIONS ON MEDICAL IMAGING, v.42, no.4, pp.922 - 934 | - |
dc.identifier.issn | 0278-0062 | - |
dc.identifier.uri | http://hdl.handle.net/10203/306810 | - |
dc.description.abstract | Patient scans from MRI often suffer from noise, which hampers the diagnostic capability of such images. As a method to mitigate such artifacts, denoising is largely studied both within the medical imaging community and beyond the community as a general subject. However, recent deep neural network-based approaches mostly rely on the minimum mean squared error (MMSE) estimates, which tend to produce a blurred output. Moreover, such models suffer when deployed in real-world situations: out-of-distribution data, and complex noise distributions that deviate from the usual parametric noise models. In this work, we propose a new denoising method based on score-based reverse diffusion sampling, which overcomes all the aforementioned drawbacks. Our network, trained only with coronal knee scans, excels even on out-of-distribution in vivo liver MRI data, contaminated with a complex mixture of noise. Even more, we propose a method to enhance the resolution of the denoised image with the same network. With extensive experiments, we show that our method establishes state-of-the-art performance while having desirable properties which prior MMSE denoisers did not have: flexibly choosing the extent of denoising, and quantifying uncertainty. | - |
dc.language | English | - |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | - |
dc.title | MR Image Denoising and Super-Resolution Using Regularized Reverse Diffusion | - |
dc.type | Article | - |
dc.identifier.wosid | 000964765000003 | - |
dc.identifier.scopusid | 2-s2.0-85141598809 | - |
dc.type.rims | ART | - |
dc.citation.volume | 42 | - |
dc.citation.issue | 4 | - |
dc.citation.beginningpage | 922 | - |
dc.citation.endingpage | 934 | - |
dc.citation.publicationname | IEEE TRANSACTIONS ON MEDICAL IMAGING | - |
dc.identifier.doi | 10.1109/TMI.2022.3220681 | - |
dc.contributor.localauthor | Ye, Jong Chul | - |
dc.contributor.nonIdAuthor | Lee, Eun Sun | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Noise reduction | - |
dc.subject.keywordAuthor | Noise measurement | - |
dc.subject.keywordAuthor | Magnetic resonance imaging | - |
dc.subject.keywordAuthor | Mathematical models | - |
dc.subject.keywordAuthor | Training | - |
dc.subject.keywordAuthor | Diffusion processes | - |
dc.subject.keywordAuthor | Numerical models | - |
dc.subject.keywordAuthor | Diffusion model | - |
dc.subject.keywordAuthor | stochastic contraction | - |
dc.subject.keywordAuthor | denoising | - |
dc.subject.keywordAuthor | MRI | - |
dc.subject.keywordPlus | NOISE-LEVEL ESTIMATION | - |
dc.subject.keywordPlus | DOMAIN | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.