Answers are grounded in this paper and cite their sources.
Aim The authors have developed an automated algorithm-based technique to identify implant position based on post-operative CT scans for total knee arthroplasty. To analyse the performance of this algorithm we compared the implant orientation against two trained radiologists. This was an exploratory secondary analysis from CT scans collected as part of an RCT. Methods To assess this algorithm we used Bayesian hierarchical modeling, which has the unique benefits of being able to accept the null hypothesis and allow interpretation of concordance between the radiologists and our algorithm. Results Results show this automated algorithm is effective for deriving post-operative implant locations from CT scans. Lack of concordance in certain parameters are likely driven by the challenges associated with locating certain implant landmarks due to the geometry of the components. Such limitations will not impact the accuracy of our automated algorithm. With the increase in 3D pre-surgical planning and robotic assistance, these techniques can be used in future studies, such as assessing implant alignment philosophies. Conclusion Overall, this automated algorithmic approach for locating post-operative implant position provides good accuracy, valuable insights in surgical execution and saves resources by reducing the time demands on trained medical specialists.
No takes yet. Share an insight, caveat, or question.
Motesharei et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: