What is the recommended size of a Volume of Interest for cancellous bone ? A skeleton-based study

Abstract

The study of the bone fracture is an important issue for oste- oporosis and car safety. The behavior of cancellous bone is strongly linked to the micro-architecture, the strain rate (Prot et al. 2015), and the specimen size Harrison & McHugh 2010). Numerical models are used in order to simulate the viscoelas- tic behavior up to the point of fracture propagation in cancellous bone. Finite element method (FEM) models based on micro-CT scans are currently the most popular approach. However, the results are dependent on the specimen size and the mesh den- sity, in addition to which the fracture analysis is time-consum- ing (Hambli 2013). Moreover, the variable architecture within a typical specimen limits the minimum sample size that will still provide reasonable architectural parameter values in comparison with the full specimen size. Indeed, a BV/TV variation up to 20% was found in the same specimen (ϕ = 7.85 mm) (Stauber et al. 2014). Skeleton-based models have already shown a great poten- tial for the efficient simulation of bone behavior and fracture. Cancellous bone geometry is based on nodes, beams, and plates (Stauber & Müller 2006), which is straightforward to implement from a skeleton. In this study, the effect of the Volume of Interest (VOI) size, within a sample, on the evaluation of cancellous bone architec- tural parameters from the skeletonized model will be presented. The aim was to furnish recommendations for the sample size for further numerical simulations.