Radiography is the predominant imaging modality used for the in-vivo analysis of orthopaedic implants. A major disadvantage of radiography is that the articulating joint components that are composed of radiolucent polyethylene cannot be directly visualized. Current strategies attempt to circumvent this limitation by estimating component positions and simplifying the joint system, however, these approaches lead to a number of associated errors. Thus, this study provides a method to enable the visualization of the polyethylene component of total knee replacements in radiographic images. This was achieved through the repeatable insertion of markers and accompanying registration process, which were evaluated in this study for reproducibility and accuracy. An insertion guide was developed to insert tantalum beads into polyethylene tibial surface liners. The bead-inserted liners were micro-CT scanned to obtain 3D surface geometries. An in-vivo mimicking phantom RSA experiment was then used to test the 3D to 2D registration process. The guide positioned the beads consistently to ±0.21 mm. The 3D to 2D registration demonstrated a repeatability of -0.014 ± 0.008 mm. Registration of different bead-inserted tibial liners to the phantom revealed an average error of 0.026 ± 0.047 mm for this visualization method. This visualization approach provides greatly improved registration and inter-component measurements than current alternative strategies. This process is suitable for a number of other joints and would greatly benefit procedures that analyze component interactions and implant performance over time. © 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 35:2017-2022, 2017.
Read full abstract