Keywords: Other AI/ML, AI/ML Software
Motivation: Accurate morphometric assessment of cartilage—such as thickness and volume—via MRI is essential for monitoring knee osteoarthritis. Segmenting cartilage remains challenging and dependent on extensive expert-annotated datasets which are heavily subjected to inter-reader variability. Recent advancements in Visual Foundational Models(VFM), especially memory-based approaches, offer opportunities for improving generalizability and robustness.
Goal(s): Leverage memory-based VFMs as an accurate method for segmenting soft tissue in 3DMRIs of the knee joint.
Approach: Trained 2DVFM, 3DVFM, and 3DCNN on 500MRIs; proposed method, SAMRI-2, evaluated against external datasets and inter-reader variability.
Results: SAMRI-2 outperformed other models, achieving 5 DSC points higher and three times lower error in cartilage thickness.
Impact: By leveraging memory-based 3D-VFM for morphometric assessment of cartilage through 3DMRIs, we significantly enhance the accuracy and generalizability of the challenging knee soft tissue segmentation, paving the way for more precise measurements of osteoarthritis progression.
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