Related Experiment Video
Updated: May 28, 2026

06:31
Erosion Identification in Metacarpophalangeal Joints in Rheumatoid Arthritis using High-Resolution Peripheral Quantitative Computed Tomography
Published on: October 6, 2023
Localized component analysis for arthritis detection in the trapeziometacarpal joint
Martijn van de Giessen1, Sepp de Raedt, Maiken Stilling
1Quantititative Imaging Group, Delft University of Technology, The Netherlands.
Summary
A new statistical shape model accurately identifies osteoarthritis (OA) in the trapeziometacarpal joint. This method localizes shape changes, improving detection of pathological trapezium bones compared to traditional techniques.
Area of Science:
- Biomedical Engineering
- Orthopedics
- Medical Imaging
Background:
- The trapeziometacarpal joint is crucial for thumb prehension.
- Osteoarthritis (OA) commonly affects this joint, altering trapezium shape.
- Existing methods may not precisely capture localized pathological changes.
Purpose of the Study:
- To develop and validate a novel local statistical shape model for trapezium analysis.
- To differentiate between healthy and OA-affected trapezia using localized shape variations.
- To improve the identification of pathological changes in trapeziometacarpal joint OA.
Main Methods:
- A novel local statistical shape model utilizing a differentiable locality measure.
- Weighted variance of point coordinates per mode to determine localized components.
- Application to a dataset of 60 trapezia (38 healthy, 22 with OA).
Main Results:
- Localized components effectively modeled regions affected by OA.
- The new method showed distinct variations compared to Principal Component Analysis (PCA).
- Pathological trapezium identification was significantly improved over PCA.
Conclusions:
- The local statistical shape model offers a precise method for analyzing trapezium morphology in OA.
- This approach enhances the detection and understanding of trapeziometacarpal joint osteoarthritis.
- The model provides a valuable tool for orthopedic research and clinical diagnostics.
