Statistical shape analysis of the human spleen geometry for probabilistic occupant models

Area of Science:

• Biomechanics
• Medical Imaging
• Computational Anatomy

Background:

• Statistical shape models (SSMs) enable computational modeling of human organs, capturing geometric variations between individuals.
• Accurate anatomical models are crucial for understanding injury mechanisms and improving safety systems.

Purpose of the Study:

• To create statistical mean and boundary models of the human spleen in an occupant posture.
• To establish computational models that represent spleen shape variation for injury risk assessment.

Main Methods:

• Applied principal component analysis (PCA) to fifteen human spleen scans.
• Utilized a landmark sliding approach for precise landmark placement and shape contour refinement.
• Developed statistical mean and boundary models from the analyzed spleen data.

Main Results:

• The primary mode of variation identified was overall spleen volume, explaining 69% of the shape variability.
• Generated statistical mean and boundary models representing the human spleen's geometry in an occupant posture.
• Established a foundation for probabilistic finite element (FE) models.

Conclusions:

• The developed spleen models can be integrated into finite element models for injury prediction.
• These models offer potential for enhancing automotive safety by identifying spleen injury risks during vehicle collisions.