Keywords: Myocardium, Machine Learning/Artificial Intelligence
Motivation: Deep learning approaches have been proposed for the analysis of 3D tagged cardiac MRI, but they are mostly limited to 2D cases due to the challenge of annotating 3D datasets.
Goal(s): To train a deep-learning framework on synthetic data for the analysis of in-vivo 3D cardiac tagged MRI.
Approach: Synthetic data was generated starting from a biophysical models of left-ventricular function, coupled with a 3D tagged MRI signal model. This data was used to train two UNets to infer myocardial displacement fields.
Results: Strain predictions from the displacement fields obtained on in-vivo datasets showed good agreement with those derived from manual annotations.
Impact: This work demonstrates that synthetic datasets can be used to train neural networks for the analysis of 3D tagged MRI, reducing the burden of annotating in-vivo data. The trained models and synthetic data will be made available.
How to access this content:
For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.
After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.
After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.
Keywords