Comparison of two numerical approaches for bone remodelling.

Area of Science:

• Computational mechanics
• Biomaterials science
• Numerical analysis

Background:

• The checkerboard phenomenon is a common issue in numerical simulations of bone remodeling.
• Element-based finite element analyses have been traditionally used but suffer from this artifact.
• The suitability of element-based approaches for bone remodeling requires investigation.

Purpose of the Study:

• To investigate the suitability of element-based versus node-based finite element analyses for bone remodeling simulations.
• To identify methods to mitigate or eliminate the checkerboard phenomenon.
• To optimize computational efficiency in bone remodeling simulations.

Main Methods:

• Implementation of both element-based and node-based finite element analyses using ABAQUS.
• Comparison of numerical results to identify the occurrence of the checkerboard phenomenon.
• Introduction and evaluation of the first-order Adams-Bashforth integration method to reduce computational cost.

Main Results:

• The checkerboard phenomenon was observed exclusively in element-based finite element analyses.
• The node-based approach successfully eliminated the checkerboard phenomenon but increased computational time.
• The first-order Adams-Bashforth method demonstrated enhanced accuracy and reduced computational cost compared to Euler's forward method.

Conclusions:

• Node-based finite element analysis is recommended for computational bone remodeling to avoid the checkerboard artifact.
• The first-order Adams-Bashforth integration scheme is effective in reducing computational cost while maintaining accuracy.
• Enforcing bone density continuity across element boundaries is crucial for accurate bone remodeling simulations.