Improving Quality Control of MRI Images Using Synthetic Motion Data
MRI quality control (QC) is challenging due to unbalanced and limited datasets, as well as subjective scoring, which hin-der the development of reliable automated QC systems. To address these issues, we introduce an approach that pretrains a model on synthetically generated motion artifacts before applying transfer learning for QC classification. This method not only improves the accuracy in identifying poor-quality scans but also reduces training time and resource requirements compared to training from scratch. By leveraging syn-thetic data, we provide a more robust and resource-efficient solution for QC automation in MRI, paving the way for broader adoption in diverse research settings.
Authors
- Kiho Cho (ORCID: https://orcid.org/0000-0001-5250-3564)
- Sylvain Bouix (ORCID: https://orcid.org/0000-0003-1326-6054)
- June Kane
- Samira Ebrahimi Kahou
- Michael P. Harms (ORCID: https://orcid.org/0000-0001-7945-036X)
- Ofer Pasternak (ORCID: https://orcid.org/0000-0002-7078-6446)
- Barnaby Nelson (ORCID: https://orcid.org/0000-0002-6263-2332)
- Patrick D. McGorry (ORCID: https://orcid.org/0000-0002-3789-6168)
- Charles F Bearden
- S.W. Woods
- Martha E. Shenton (ORCID: https://orcid.org/0000-0003-4235-7879)
- R.S. Kahn
- Charles Bricout
Institutions
- Northwell Health (US)
- Mount Sinai Hospital (US)
- University of Calgary (CA)
- UCLA Health (US)
- Orygen (AU)
- Département d'Informatique (FR)
- Mass General Brigham (US)
Publication Details
- Published
- 2025-04-14
- DOI
- https://doi.org/10.1109/isbi60581.2025.10981056
- Primary Topic
- Medical Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Institutes of Health