Knee cartilage topography, thickness, and contact areas from MRI: in-vitrocalibration and in-vivo measurements

Objective: This study assessed the three-dimensional accuracy of magnetic r esonance imaging (MRI) for measuring articular surface topographies and car tilage thicknesses of human cadaveric knee joints, by comparison with the c alibrated stereophotogrammetric (SPG) method. Methods: Six fresh frozen cadaveric knees and the knees of four volunteers were imaged with a three-dimensional spoiled gradient-recalled acquisition with fat suppression using a linear extremity coil in a 1.5 T superconducti ng magnet. The imaging voxel size was 0.47 x 0.47 x 1.0 mm. Both a manual a nd a semi-automated segmentation method were employed to extract topographi c measurements from MRI. Following MRI, each of the six cadaveric knees was dissected and its articular surfaces quantified using stereophotogrammetry . The MRI surface measurements were compared numerically with the SPG; meas urements. Results: For six cadaveric knees, the average accuracies of cartilage and s ubchondral bone surface measurements were found to be 0.22 mm and 0.14 mm r espectively and the thickness measurements demonstrated an average accuracy of 0.31 mm. It was found that while most of the error may be attributed to random measurement error, the accuracy was somewhat affected by systematic errors. For each bone of the knee, accuracies were most favorable in the p atella, followed by the femur and then the tibia. The more efficient semi-a utomated method provided equally good and sometimes better accuracies than manual segmentation. Conclusions: This study demonstrates that clinical MRI can provide accurate measurements of cartilage topography, thickness, contact areas and surface curvatures of the knee.