Accuracy and reliability of hip translations measured using upright open MRI: a validation study using a phantom.

There is increasing interest in quantifying hip translations, but few non-invasive measurement methods are available. The objective of this study was to determine the accuracy and repeatability of measuring hip positions directly from images obtained with an Upright Open MR scanner (MROpen), which allows imaging in a wide range of hip postures. A phantom was designed and 3D-printed, comprising 8 hip joints representing 4 subjects in 2 postures, one idealized hip, and 3 spheres to define a coordinate system. The phantom was scanned 5 times in the MROpen with a T1-weighted GFE sequence, which had been verified as suitable for scanning human hips on this scanner. The measured positions of the femoral head centre relative to the centre of the acetabulum were compared to the known values used to create the model. Both the mean and the RMS of the relative position error were under 0.8 mm in each direction for the four hips in two postures. The intra-class correlation coefficients (ICC) for intra- and inter-reader reliability ranged from 0.71 to 0.96. We conclude that hip positions can be directly measured with sufficient accuracy and reliability to show clinically significant differences using this upright open MR scanner.