A novel pipeline using statistical parametric mapping for assessing differences in three- and four-dimensional...

Area of Science:

• Biomechanics
• Medical Imaging
• Computational Anatomy

Background:

• Four-dimensional computed tomography (4DCT) enables quantification of joint mechanics during dynamic tasks.
• Analyzing complex arthrokinematic data across participants and motion cycles requires advanced analytical techniques.
• Statistical parametric mapping (SPM) is under-explored for 4DCT applications despite its utility in biomechanical and imaging datasets.

Purpose of the Study:

• To develop and validate a computational pipeline for analyzing 4DCT-derived arthrokinematics.
• To explore the relationship between wrist position and distal radioulnar joint (DRUJ) interosseous proximities.
• To demonstrate the pipeline's applicability to both static and dynamic joint motion analysis.

Main Methods:

• A statistical shape model was used to create a canonical joint template.
• Non-linear morphing predicted participant-specific joint surfaces from the template.
• SPM regression and generalized linear models analyzed relationships between wrist position, injury status, and interosseous proximities.

Main Results:

• The pipeline successfully quantified position-related differences in DRUJ interosseous proximities.
• Increased pronation was significantly associated with increased proximities at the sigmoid notch volar margin.
• The analysis demonstrated the pipeline's capability for 3D and 4D computed tomography-derived arthrokinematics.

Conclusions:

• The proposed pipeline effectively quantifies dynamic joint arthrokinematics using 4DCT data.
• This method allows for the analysis of relationships between joint motion and continuous or categorical variables.
• The pipeline is adaptable for studying various joints, tasks, conditions, and injury states.