Keywords: Biology, Models, Methods, Preclinical
Motivation: Preclinical cardiac MRI is a crucial research tool in gaining understanding of various heart diseases. However, for comprehensive cardiac MRI protocols, acceleration is needed without loss of image quality. Although compressed sensing enables acceleration, deep learning may offer superior performance.
Goal(s): To accelerate preclinical cardiac CINE MRI with deep learning using a dynamic recurrent inference machine (DRIM).
Approach: Our DRIM was trained to reconstruct undersampled mouse cardiac CINE MRI data at four different accelerations up to 6x. Results were compared to optimized compressed sensing reconstructions.
Results: We successfully reconstructed accelerated dynamic cardiac MRI using the DRIM, outperforming CS for every acceleration factor.
Impact: Deep learning was able to accelerate cardiac MRI further than compressed sensing, by maintaining better image quality with higher acceleration factors. This enables more comprehensive MRI protocols for gaining a better understanding of heart diseases and effects of treatments.
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