Adaptive signal representation for accelerated dynamic MRI
Event details
| Date | 15.01.2013 |
| Hour | 14:00 |
| Speaker | Dr. Mathews Jacob, University of Iowa |
| Location |
AAC114
|
| Category | Conferences - Seminars |
Cardiac MRI schemes such as myocardial perfusion and viability imaging play key roles in evaluating both ischemic and non-ischemic heart disease. The need to reliably detect subtle lesions requires good in-plane spatial resolution, good spatial coverage, high temporal resolution, long breath-hold duration and good contrast to noise ratio. Since these contradictory goals are often difficult to realize, clinicians are often forced to compromise on spatio-temporal resolution and coverage
The main focus of this talk is to introduce adaptive signal representations to considerably improve the state of the art in dynamic MRI. I will start by introducing blind linear models, where the basis functions and the coefficients of the representation are estimated from highly under-sampled MRI data. The recovery is posed as an optimization problem, which is solved using a fast augmented Lagrangian algorithm. This framework is then extended to blind compressive sensing, where the dictionary basis functions and the sparse coefficients are estimated from heavily under-sampled data. I will also introduce a novel algorithm for dictionary learning from under-sampled data. Unlike classical algorithms, the proposed scheme can handle arbitrary dictionary constraints; I will also illustrate the utility in using alternate dictionary constraints to further improve the recovery.
The main focus of this talk is to introduce adaptive signal representations to considerably improve the state of the art in dynamic MRI. I will start by introducing blind linear models, where the basis functions and the coefficients of the representation are estimated from highly under-sampled MRI data. The recovery is posed as an optimization problem, which is solved using a fast augmented Lagrangian algorithm. This framework is then extended to blind compressive sensing, where the dictionary basis functions and the sparse coefficients are estimated from heavily under-sampled data. I will also introduce a novel algorithm for dictionary learning from under-sampled data. Unlike classical algorithms, the proposed scheme can handle arbitrary dictionary constraints; I will also illustrate the utility in using alternate dictionary constraints to further improve the recovery.
Practical information
- General public
- Free
Organizer
- Delgado Gonzalo Ricard <[email protected]>
Contact
- Delgado Gonzalo Ricard <[email protected]>