Abstract #2391
A Synthetic Generator of Myocardial Blood-Oxygen-Level-Dependent MRI Timeseries with Structural Sparse Decomposition Modeling
Cristian Rusu 1 , Rohan Dharmakumar 2,3 , and Sotirios A. Tsaftaris 1,4
1
IMT Institute for Advanced Studies Lucca,
Lucca, Italy,
2
Biomedical
Imaging Research Institute, Cedars-Sinai Medical Center,
Los Angeles, CA, United States,
3
Medicine,
University of California, Los Angeles, CA, United
States,
4
Electrical
Engineering and Computer Science, Northwestern
University, Evanston, IL, United States
Cardiac Phase resolved Blood-Oxygen-Level-Dependent
(CP-BOLD) MRI has been recently demonstrated for the
identification of ischemic territories under resting
conditions. Lack of accurate registration, necessary to
provide pixel-to-pixel correspondences in the cardiac
cycle, causes the majority of analysis to rely on
segmental definitions of the myocardium and to use a few
cardiac phases, decreasing the potential diagnostic
power of the technique. To accelerate the development of
methods that could potentially yield pixel-level
characterization of ischemia, we propose and validate a
synthetic CP-BOLD timeseries generator based on a
composite dictionary model that learns to represent
efficiently patterns under healthy and ischemic
conditions.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
