Related Experiment Video
Updated: Jan 6, 2026
09:29
Author Spotlight: Improving Percutaneous Vertebroplasty Surgical Accuracy and Efficiency Through Advanced Puncture Techniques
Published on: August 9, 2024
566
Vertebral rotation estimation from frontal X-rays using a quasi-automated pedicle detection method
Shahin Ebrahimi1, Laurent Gajny2, Claudio Vergari1
1Institut de Biomécanique Humaine Georges Charpak, Arts et Métiers Institute of Technology, Paris, France.
Summary
This study introduces a quasi-automated method for estimating vertebral axial rotation (VAR) in scoliosis assessment. The new approach offers robust pedicle localization and accurate VAR calculation with minimal user input.
Area of Science:
- Radiology
- Medical Imaging
- Orthopedics
Background:
- Vertebral axial rotation (VAR) measurement is crucial for scoliosis assessment.
- The Stokes method estimates VAR from frontal X-rays by analyzing pedicle and vertebral body positions.
- Manual landmark identification for the Stokes method is time-consuming.
Purpose of the Study:
- To develop a quasi-automated method for pedicle detection and VAR estimation.
- To reduce user intervention in VAR calculation for scoliosis assessment.
Main Methods:
- A retrospective study included 149 healthy and adolescent idiopathic scoliosis (AIS) subjects.
- Frontal X-rays were used to develop an automated pedicle detector based on image analysis and machine learning.
- VAR was calculated using the Stokes method with both automated and manual pedicle localizations for comparison.
Main Results:
- Automated pedicle localization achieved 84% precision with a mean difference of 1.2 ± 1.2 mm compared to manual identification.
- VAR values calculated using automated pedicle localization showed a mean difference of -0.2 ± 3.4° compared to manual methods.
- Uncertainty in pedicle location was less than 2 mm along each image axis.
Conclusions:
- The proposed quasi-automated method enables robust pedicle localization and VAR calculation from frontal radiographs.
- This method significantly reduces user intervention, improving efficiency in scoliosis assessment.
