Comparison of Multiplanar Reconstruction versus Computer-Aided 3D Measurements from Computed Tomography Scans of Normal Canine Radii

Introduction: Computer-aided design software (CADS) measurements have not been related to previously accepted multiplanar reconstruction (MPR). The purpose of this study was to compare CADS to MPR for radial measurements in normal dogs. We hypothesized that measurements would be different between methods and that the measurements made using CADS would be more reliable than those using MPR.

Materials and Methods: Computed tomography scans of the forelimbs of dogs with normal morphology were included. Paired measurements were made using MPR and CADS methodology. Statistical analysis included the Wilcoxon signed rank test for measurement comparisons of measurements (p < 0.05) and intraclass correlation coefficients (ICC) to determine measurement reliability.

Results: Ten right-sided antebrachii have been evaluated. The median distal frontal measurement was 5.3 degrees valgus for MPR and 8.3 degrees varus for CADS (p < 0.01). CADS provided excellent reliability for sagittal and distal frontal measurements whereas MPR provided moderate (sagittal) or poor (distal frontal) reliability.

Discussion/Conclusion: We partially accepted our hypotheses. The distal frontal angle was different between methods, which is likely due to differences in landmarks. The reliability of CADS was equal to or better than MPR. Determining the relationship between CADS and MPR measurements should help surgeons use CADS measurements. The eventual goal is to train neural networks for automated measurements of bone and limb angulation to improve accuracy and reduce surgeon time commitments to measurements.

Acknowledgment

This project was funded using faculty discretionary funds.

 

MissingLinkText

Publication History

Article published online:
15 July 2025