Keywords: Analysis/Processing, Osteoarthritis, AI, Accelerated Imaging
Motivation: Long MRI acquisition times limit T1ρ/T2 mapping for assessing cartilage health, necessitating faster and reliable quantification methods.
Goal(s): This study evaluates the use of accelerated qMRI and deep learning-based segmentation for reliable T1ρ/T2 mapping without high-resolution morphological images.
Approach: A pretrained segmentation model, with and without transfer learning, was used to segment six cartilage and meniscal regions from GRAPPA2-MAPSS and compressed sensing accelerated (CS-AF8)-MAPSS echo images. These segmentations, overlaid on maps, provided T1ρ/T2 quantification.
Results: CS-AF8, with transfer learning, achieved comparable segmentation performance and T1ρ/T2 quantification to the standard MAPSS with DESS segmentation, although it showed higher variation in T1ρ/T2 values.
Impact: This research offers a clinically feasible accelerated qMRI and deep learning-based approach for faster and accurate qMRI-based cartilage assessment, allowing T1ρ and T2 mapping in under 3 minutes.
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