Related Experiment Video
Updated: May 20, 2026
10:35
Precision Measurements and Parametric Models of Vertebral Endplates
Published on: September 17, 2019
Quantitative vertebral morphometry based on parametric modeling of vertebral bodies in 3D
1Faculty of Electrical Engineering, University of Ljubljana, Tržaška 25, 1000, Ljubljana, Slovenia, darko.stern@fe.uni-lj.si
Summary
Quantitative vertebral morphometry (QVM) in three dimensions (3D) accurately identifies vertebral fractures. This 3D analysis provides objective discrimination between normal and fractured vertebrae, aiding in diagnosis and prediction for at-risk patients.
Area of Science:
- Medical Imaging
- Biomechanical Engineering
- Radiology
Background:
- Two-dimensional (2D) identification of vertebral fractures is challenging due to image projection and shape variability.
- Computed tomography (CT) provides detailed 3D anatomical images for precise measurement of vertebral deformations and fractures.
Purpose of the Study:
- To develop and validate a quantitative vertebral morphometry (QVM) method using 3D parametric modeling for accurate fracture detection.
- To objectively discriminate between normal and fractured vertebral bodies using 3D shape analysis.
Main Methods:
- Parametric 3D models were automatically aligned to vertebral bodies in CT images to obtain detailed 3D shape representations.
- Clinically relevant morphometric features were extracted and compared to statistical values for QVM in 3D.
- Statistical classification analysis identified optimal parameters and thresholds for differentiating normal from fractured vertebrae.
Main Results:
- The 3D QVM method was applied to 454 normal and 228 fractured vertebral bodies.
- Achieved classification sensitivity of 92.5% with 7.5% specificity, and 92.5% accuracy.
- Key discriminating features included vertebral body height and endplate inclination/concavity.
Conclusions:
- 3D QVM efficiently and objectively discriminates between normal and fractured vertebral bodies.
- The method identifies fracture morphology (wedge, (bi)concavity, crush) and grades (1-3).
- 3D QVM shows potential value in diagnosing and predicting vertebral fractures, especially in osteoporosis patients.
