Premium
Atlas‐based knee cartilage assessment
Author(s) -
CarballidoGamio Julio,
Majumdar Sharmila
Publication year - 2011
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.22836
Subject(s) - osteoarthritis , atlas (anatomy) , cartilage , magnetic resonance imaging , landmark , medicine , nuclear medicine , computer science , orthodontics , artificial intelligence , anatomy , radiology , pathology , alternative medicine
Magnetic resonance imaging analysis of knee cartilage properties at corresponding anatomic locations could be a valuable tool in studies of knee osteoarthritis by enabling accurate comparisons at practically any region. A technique of this kind is presented in this study. The proposed technique is based on gray‐level bone matching using affine transformations and free‐form deformations thus eliminating the need of bone segmentations and landmark matching. Sixteen subjects of the osteoarthritis initiative with knee osteoarthritis (10 from baseline; 6 from 24‐month follow‐up) were included in this study. Baseline subjects were used to create a gray‐level atlas of the patella with its corresponding mean cartilage thickness and T 2 maps. Follow‐up subjects were used to validate atlas‐based point‐to‐point cartilage comparisons. All registrations were qualitatively evaluated with fused gray‐level images of registered patellas. Quantitative evaluation was performed based on mean values of minimum Euclidean distances between matched bone‐cartilage interfaces. A mean distance of 0.554 mm was obtained between the subjects used to build the atlas, and a mean distance of 0.633 mm was found between the atlas and validation subjects. The technique can be applied to other anatomical regions and with other cartilage measures. Qualitative and quantitative results demonstrate the accuracy of the technique and warrant its application in larger cross‐sectional and longitudinal studies of osteoarthritis. Magn Reson Med, 2011. © 2011 Wiley‐Liss, Inc.