Keywords: Susceptibility/QSM, Quantitative Susceptibility mapping
Motivation: The research addresses limitations in QSM susceptibility maps, where over-regularization suppresses high-frequency details critical for accuracy, and deep learning models frequently produce artifacts due to inconsistent data handling.
Goal(s): The study aims to recover high-frequency components in susceptibility maps, improving data consistency and accuracy with minimal computational demand.
Approach: By initializing with an over-regularized map, the study applies a few gradient descent iterations to impose data consistency, balancing noise reduction with detail preservation.
Results: NDI with limited gradient descent iterations effectively recovers high-frequency details, improving structural clarity and consistency, enhancing susceptibility map quality in simulated and in vivo contexts.
Impact: This study’s refinement method for QSM could empower clinicians with higher-quality susceptibility maps, improving diagnostic accuracy in neurological assessments. It encourages further exploration of minimal-iteration techniques, promoting efficient approaches to enhance imaging fidelity without intensive computational resources.
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