Development of a simple numerical model for trabecular bone structures.

Area of Science:

• Biomedical Engineering
• Materials Science
• Medical Imaging

Background:

• Additive manufacturing enables creating patient-specific bone structures.
• High-resolution anthropomorphic phantoms require accurate bone surrogates.
• Trabecular bone microarchitecture presents complex statistical distributions.

Purpose of the Study:

• To propose a simple numerical model for generating trabecular bone microarchitecture.
• To create surrogate bone structures with statistical distributions similar to real bone.
• To investigate the model's applicability with a two-material approximation for CT simulations.

Main Methods:

• A numerical model was developed using two complex functions for trabecular thickness and spacing.
• Human bone (humerus, radius, ulna, vertebrae) was scanned at the Australian Synchrotron.
• Simulations incorporated a two-material approximation for absorption- and phase-contrast CT.

Main Results:

• The synthetic structures closely matched real trabecular microarchitecture, with mean thickness errors as low as 2 μm.
• The model accurately reproduced bone structures at 19 axial locations.
• Simulated phase-contrast CT data demonstrated successful reconstruction and phase-contrast effects.

Conclusions:

• The model can generate trabecular distributions for creating phantoms for quality assurance and validation.
• A two-material approximation simplifies additive manufacturing and synthetic data generation.
• The generated synthetic data can be used for training machine learning models in medical imaging.