A Region-Based Statistical Shape Modeling on the First Trapezoid-Metacarpal.

Area of Science:

• Orthopedics
• Medical Imaging
• Biomechanical Engineering

Background:

• Osteoarthritis (OA) detection and prevention are crucial.
• Understanding bone morphology variations is key to early OA diagnosis.
• Statistical Shape Models (SSM) offer advanced analysis of anatomical structures.

Purpose of the Study:

• To compare region-based (rSSM) and non-region-based (nrSSM) statistical shape models for analyzing the first trapezoid-metacarpal (TMC) bone.
• To identify specific morphological differences and shape variations for early OA detection.
• To enhance the accuracy and specificity of OA risk assessment.

Main Methods:

• Utilized CT image data from 31 healthy volunteers to create 3D TMC models.
• Applied Principal Component Analysis (PCA) to both rSSM and nrSSM for shape variation analysis.
• Employed Gaussian curvature analysis for automatic region identification in rSSM.

Main Results:

• Identified four dominant shape variations using both nrSSM and rSSM.
• rSSM demonstrated more variations with fewer principal components than nrSSM.
• The first principal component indicated size variation; significant differences in the second and fourth components highlighted distal first metacarpal torsion in rSSM.

Conclusions:

• Region-based SSM (rSSM) provides more specific and sensitive morphological variation analysis for the TMC bone compared to nrSSM.
• Torsion in the distal head of the first metacarpal, identified by rSSM, is a potential indicator for increased osteoarthritis risk.
• This approach can aid in early detection and prevention strategies for osteoarthritis.