Keywords: Machine Learning/Artificial Intelligence, Low-Field MRI
Motivation: Low-field MRI offers low-cost imaging but suffers from prolonged, low-SNR scans.
Goal(s): To explore a novel future personalized-healthcare strategy, combining high- and low-field MRI. We suggest harnessing data from baseline high-field scans to boost speed and quality of follow-up low-field scans. To our knowledge, this is the first study to explore this approach.
Approach: We introduce ViT-Fuser, a novel multi-head vision transformer that fuses high- and low-field features and a hybrid loss function that facilitates high-quality reconstructions.
Results: Our ViT-Fuser and hybrid loss outperform other methods, e.g. MoDL and vision-transformers, significantly improving image quality in accelerated low-field scans.
Impact: We introduce ViT-Fuser and the hybrid loss, a promising solution for accelerating and enhancing low-field MRI by leveraging high-field reference scans from the same subject. Our approach defines a personalized imaging strategy and outperforms state-of-the-art reconstruction methods and losses.
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