Accuracy of pelvic measurements on virtual radiographic projections based on computed tomography scans compared to conventional radiographs pre- and postoperatively

Dominik Kaiser, Armando Hoch, Christoph Stern, Stefan Sommer, Reto Sutter, Patrick O. ZinggArchives 2022 · DOI

TL;DR

This paper demonstrates that virtual X-ray-like projections generated from low-dose CT scans can accurately measure key hip joint parameters, matching the reliability of traditional X-rays. This suggests a potential to reduce radiation exposure and streamline pre- and post-operative assessments for hip pathologies.

What Problem Does This Paper Address?

Diagnosing and monitoring hip joint pathologies, such as developmental dysplasia of the hip (DDH) and femoroacetabular impingement (FAI), heavily relies on anteroposterior (AP) pelvic radiographs. These standard X-rays are used to assess various quantitative and semi-quantitative parameters like the lateral center edge angle (LCEA) and acetabular index (AI). While computed tomography (CT) offers 3D insights, it has traditionally involved higher radiation doses, making it less ideal for young patients who are often affected by these conditions and face increased lifetime cancer risk. The challenge is to leverage the advanced imaging capabilities of CT while minimizing radiation exposure, ideally by replacing or supplementing conventional radiographs.

Methods

The researchers developed a method to generate "radiograph-like projections" from CT scan data. This involved using a customized 3D cone beam projection algorithm, based on existing implementations, to reconstruct virtual AP pelvic radiographs from CT datasets. These virtual projections were then compared to actual conventional AP pelvic radiographs. The study involved two independent examiners (an orthopedic surgeon and a musculoskeletal radiologist) who measured several standard radiographic parameters on both the real and virtual images, both pre- and post-operatively. Statistical analysis, including Intraclass Correlation Coefficients (ICC), was used to assess the agreement and reliability between the measurements from the two imaging modalities.

Datasets

The study utilized data from a consecutive series of 16 hips (from 15 patients) who underwent periacetabular osteotomy (PAO) for symptomatic developmental dysplasia of the hip. This included both preoperative and postoperative conventional pelvic radiographs and CT scans. No new benchmark datasets were introduced; the study focused on evaluating existing imaging modalities and a novel reconstruction technique.

Key Insights and Ideas

  • CT as a Viable Radiograph Alternative: The core idea is that CT, with advancements in dose reduction, can provide not just 3D anatomical information but also generate 2D projections that are diagnostically equivalent to traditional X-rays for specific hip measurements.
  • Radiation Dose Reduction Potential: By potentially replacing conventional radiographs with CT-derived projections, a significant reduction in radiation exposure can be achieved, especially crucial for the young patient population.
  • Streamlining Workflow: Integrating CT-based projections could simplify the imaging workflow, potentially reducing the need for multiple imaging sessions and associated administrative burdens.
  • 3D Planning Advantage: CT inherently offers 3D data, enabling more comprehensive surgical planning (like for PAO) and precise assessment of osseous consolidation post-surgery, which traditional 2D radiographs cannot fully provide.

Key Results

The study found no statistically significant differences in the measured parameters (LCEA, MCEA, acetabular arc, extrusion index, AI) between conventional radiographs and the radiograph-like projections generated from CT data, for both examiners and across both pre- and postoperative assessments. The Intraclass Correlation Coefficients (ICC) demonstrated good to excellent reliability (0.78–0.99) for most parameters when comparing the two imaging modalities. Inter-examiner reliability was also good to excellent for most parameters. A notable exception was the extrusion index, which showed moderate to poor reliability between examiners, a finding consistent with existing literature. The crossover sign and posterior wall sign also showed minimal variation.

Limitations and Future Work

The authors acknowledge that the number of participants in the study was relatively small, although they believe this did not impact the clear findings. The study focused primarily on parameters derived from the AP pelvic radiograph, which are considered most clinically relevant. Future work could involve developing automated algorithms for generating these radiographic projections to further accelerate the workflow. The authors suggest that the implementation of ultra-low-dose CT with calculated radiographic projections could positively influence the overall quality of care by reducing radiation dose and potentially streamlining the examination process.

Summary updated

Abstract

Radiograph-like projections show comparable results to radiographs with regard to the important investigated parameters (lateral center edge angle, medial center edge angle, acetabular index, acetabular arc, extrusion index, crossover sign and posterior wall sign). Thus, ultra-low-dose CT scans may reduce the need for conventional radiographs in pre- and postoperative analyses of 3-dimensional hip pathologies in the future, as the advantages increasingly outweigh the disadvantages.
BibTeX
@inproceedings{kaiser2022accuracy,
  author = {Dominik Kaiser and Armando Hoch and Christoph Stern and Stefan Sommer and Reto Sutter and Patrick O. Zingg},
  title = {{Accuracy of pelvic measurements on virtual radiographic projections based on computed tomography scans compared to conventional radiographs pre- and postoperatively}},
  booktitle = {Archives},
  year = {2022},
  doi = {10.1007/s00402-022-04476-4},
}
Figure