Keywords: Analysis/Processing, Heart, fine-tuning, SAM, segmentation, heart
Motivation: Cardiac image segmentation is essential for accurately assessing heart structure in conditions like heart failure. However, general-purpose models lack the specialized focus needed for cardiac MRI.
Goal(s): This study fine-tunes the Foundational Segment Anything Model (SAM) to enhance segmentation accuracy on ACDC cardiac dataset, targeting the left ventricle (LV), right ventricle (RV), and myocardium (MYO).
Approach: We performed fine-tuning of adapter blocks on SAM’s decoder using sam-vit-base model with annotated cine MRI and evaluated segmentation on an unseen test set.
Results: The fine-tuned SAM achieved an average Dice Score of 0.951±0.039 and IoU of 0.909±0.065, improving over state-of-the-art models.
Impact: Fine-tuning of the foundational model SAM for cardiac MRI offers enhanced segmentation accuracy, enabling precise assessments of heart structure and function. This advancement supports improved diagnostic workflows and potential early detection in conditions like heart failure and cardiomyopathy.
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