Dictionary Learning and Time Sparsity for Dynamic MR Data Reconstruction
- Submitting institution
-
King's College London
- Unit of assessment
- 12 - Engineering
- Output identifier
- 131259719
- Type
- D - Journal article
- DOI
-
10.1109/TMI.2014.2301271
- Title of journal
- Ieee Transactions on Medical Imaging
- Article number
- N/A
- First page
- 979
- Volume
- 33
- Issue
- 4
- ISSN
- 0278-0062
- Open access status
- Out of scope for open access requirements
- Month of publication
- January
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Accelerating MRI remains critical challenge. Undersampling offers the ultimate scope for this, but leads to errors. A dictionary approach to reconstructing under-sampled cardiac cine data is described, achieving enhanced performance using an overcomplete dictionary of patches trained specifically for the task at hand. It introduced a compact dictionary that used a shared real dictionary to represent both real and imaginary values. This allows a sparser representation to be achieved, yielding improved reconstruction performance compared to prior methods. The paper pioneered an adaptive approach, helping to pave the way for a flood of papers in the literature describing trained reconstruction methods.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -