Caroline D. Jordan1, 2, Uche D. Monu1, 3, Emily J. McWalter1, Ronald D. Watkins1, Weitian Chen4, Neal K. Bangerter5, Brian Andrew Hargreaves1, Garry E. Gold1, 6
1Radiology, Stanford University, Stanford, CA, United States; 2Bioengineering, Stanford University, Stanford, CA, United States; 3Electrical Engineering, Stanford University, Stanford, CA, United States; 4Global Applied Science Laboratory, GE Healthcare, Menlo Park, CA, United States; 5Electrical & Computer Engineering, Brigham Young University, Provo, UT, United States; 6Orthopaedic Surgery, Stanford University, Stanford, CA , United States
Quantitative MRI parameters such as T1&[rho], T2 and sodium have been shown to vary in osteoarthritis patients. We measured the reproducibility of the 3D quantitative MRI techniques of CubeQuant T1&[rho], qDESS T2 and cones sodium measurements in knee cartilage. Intra-observer, inter-observer, adjacent slice, and short-term/long-term intra-subject variability were measured by computing the coefficient of variation (CV). The short-term and long-term intra-subject reproducibility amongst different regions of cartilage, expressed by the root-mean-square CV, ranged between 2.7-8.6% and 4.6-9.0% for CubeQuant T1&[rho], between 2.4-9.8% and 4.4-13.8% for qDESS T2, and between 5.2-13.8% and 4.6-16.0% for 3D cones sodium MRI, respectively.