SNRAware: Improved Deep Learning MRI Denoising with Signal-to-Noise Ratio Unit Training and G-Factor Map Augmentation
Authors:
- Hui XueHXHui Xue
- Sarah M. HooperSHSarah M. Hooper
- Iain PierceIPIain Pierce
- Rhodri H. DaviesRDRhodri H. Davies
- John StairsJSJohn Stairs
- Joseph NaegeleJNJoseph Naegele
- Adrienne E. Campbell-WashburnACAdrienne E. Campbell-Washburn
- James C. MoonJMJames C. Moon
- Michael S. HansenMHMichael S. Hansen
- Peter KellmanPKPeter Kellman
- 1 November 2025
- journal article
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (6) , e250227
Abstract
SNRAware, a model-agnostic approach for training MRI denoising models that leverages information from the image reconstruction process, improved performance and enhanced generalization to unseen imaging applications.Keywords
All Related Versions
This publication has 22 references indexed in Scilit:
- MRI denoising with a non‐blind deep complex‐valued convolutional neural networkNMR in Biomedicine, 2024
- Routine CMR at 0.55T with Standard Spatial Resolution Using an Imaging TransformerJournal of Cardiovascular Magnetic Resonance, 2024
- Low‐field MRI: A report on the 2022 ISMRM workshopMagnetic Resonance in Medicine, 2023
-
⊥
-loss: A symmetric loss function for magnetic resonance imaging reconstruction and image registration with deep learningMedical Image Analysis, 2022
- Deep learning on image denoising: An overviewNeural Networks, 2020
- Parallel MR imagingJournal of Magnetic Resonance Imaging, 2012
- Image reconstruction in SNR units: A general method for SNR measurement†Magnetic Resonance in Medicine, 2005
- Dynamic autocalibrated parallel imaging using temporal GRAPPA (TGRAPPA)Magnetic Resonance in Medicine, 2005
- Generalized autocalibrating partially parallel acquisitions (GRAPPA)Magnetic Resonance in Medicine, 2002
- SENSE: Sensitivity encoding for fast MRIMagnetic Resonance in Medicine, 1999