Leveraging Physics-Based Synthetic MR Images and Deep Transfer Learning for Artifact Reduction in Echo-Planar Imaging
The current study demonstrates the feasibility of using a physics-based method for synthesizing a large data set of images with realistic artifacts and the effectiveness of utilizing this synthetic data set in a "stacked transfer learning" approach to training a GAN for reduction of EPI-based artifacts.
Authors
- Noriko Salamon (ORCID: https://orcid.org/0000-0002-3520-9467)
- Richard G. Everson (ORCID: https://orcid.org/0000-0003-2809-3061)
- Benjamin M. Ellingson (ORCID: https://orcid.org/0000-0002-2764-6640)
- Thorsten Feiweier (ORCID: https://orcid.org/0009-0005-3537-9844)
- Jingwen Yao (ORCID: https://orcid.org/0000-0002-8828-1841)
- Xiaodong Zhong (ORCID: https://orcid.org/0000-0001-8355-5279)
- Bryan Clifford (ORCID: https://orcid.org/0000-0002-1966-0581)
- Timothy F. Cloughesy (ORCID: https://orcid.org/0000-0002-8656-7483)
- Sonoko Oshima (ORCID: https://orcid.org/0000-0002-1453-3807)
- Catalina Raymond (ORCID: https://orcid.org/0000-0003-2757-439X)
- Donatello Telesca (ORCID: https://orcid.org/0000-0003-0773-451X)
- Heiko Meyer (ORCID: https://orcid.org/0000-0002-2631-1910)
Institutions
- Siemens (Germany) (DE)
- University of California, Los Angeles (US)
- Neurobehavioral Systems (US)
- Medical Solutions (GB)
- Samueli Institute (US)
- Siemens Healthineers (Germany) (DE)
Publication Details
- Journal
- American Journal of Neuroradiology
- Published
- 2025-02-13
- DOI
- https://doi.org/10.3174/ajnr.a8566
- Citations
- 1
- Primary Topic
- Advanced MRI Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 1.36
Funders
- Siemens
- National Cancer Institute
- Congressionally Directed Medical Research Programs