Related Experiment Video
Updated: Mar 22, 2026
Author Spotlight: An Economic and Efficient Method for Quantitative Evaluation of Bone Microarchitecture in a Murine Osteoporosis Model
Published on: September 8, 2023
Machine learning based analytics of micro-MRI trabecular bone microarchitecture and texture in type 1 Gaucher disease
Gulshan B Sharma1, Douglas D Robertson2, Dawn A Laney3
1University of Calgary, Department of Mechanical and Manufacturing Engineering, Calgary, Alberta, Canada; Emory University, Department of Radiology and Imaging Sciences, Division of Musculoskeletal Imaging, Department of Orthopaedic Surgery, Emory Spine and Orthopaedics Center, Atlanta, GA, USA.
Type 1 Gaucher disease (GD) impacts bone health. Micro-MRI and machine learning can quantify bone changes, distinguishing GD patients from healthy individuals and identifying disease subtypes.
Area of Science:
- Biomedical imaging
- Bone microarchitecture analysis
- Machine learning in rare diseases
Background:
- Type 1 Gaucher disease (GD) is a genetic disorder affecting bone metabolism.
- Current methods like DXA and MRI scoring have limitations in assessing bone fragility in GD.
- There is a need for more precise biomarkers to evaluate bone disease severity in GD.
Purpose of the Study:
- To measure trabecular bone microarchitecture in type 1 GD patients.
- To use machine learning to differentiate GD patients from healthy controls.
- To assess GD genotypes and their impact on bone structure.
Main Methods:
- Micro-MR imaging of the distal radius in 20 type 1 GD patients and 10 healthy controls.
- Calculation of 15 stereological and textural measures (STM) from MR images.
- Application of principal component analysis (PCA) and support vector machine (SVM) for data analysis.
Main Results:
- Significant differences in stereological and textural measures between GD patients and healthy controls.
- Bone microarchitecture measures differed across GD genotypes and correlated with DXA scores.
- PCA and SVM achieved 73% accuracy in discriminating GD patients from controls.
Conclusions:
- Micro-MRI combined with machine learning can quantify trabecular bone microarchitecture in type 1 GD.
- This approach can differentiate GD patients from healthy individuals and identify disease subtypes.
- Further research aims to evaluate GD disease burden and treatment efficacy using this method.
