Quantitative computed tomography-based finite element (QCT-FE) modeling is a computational tool for predicting bone’s response to applied load, and is used by musculoskeletal researchers to better understand bone mechanics and their role in joint health. Decisions made at the modeling stage, such as the method for assigning material properties, can dictate model accuracy. Predictions of surface strains/stiffness from QCT-FE models of the proximal tibia have been validated against experiment, yet it is unclear whether these models accurately predict internal bone mechanics (displacement). Digital volume correlation (DVC) can measure internal bone displacements and has been used to validate FE models of bone; though, its use has been limited to small specimens.
The objectives of this study were to 1) establish a methodology for high-resolution peripheral QCT (HR-pQCT) scan acquisition and image processing resulting in low DVC displacement measurement error in long human bones, and 2) apply different density-modulus relationships and material models from the literature to QCT-FE models of the proximal tibia and identify those approaches which best predicted experimentally measured internal bone displacements and related external reaction forces, with highest explained variance and least error.
Using a modified protocol for HR-pQCT, DVC displacement errors for large scan volumes were less than 19µm (0.5 voxels). Specific trabecular and cortical models from the literature were identified which resulted in the most accurate QCT-FE predictions of internal displacements (RMSE%=3.9%, R²>0.98) and reaction forces (RMSE%=12.2%, R²=0.78). This study is the first study to quantify experimental displacements inside a long human bone using DVC. It is also the first study to assess the accuracy of QCT-FE predicted internal displacements in the tibia. Our results indicate that QCT-FE models of the tibia offer reasonably accurate predictions of internal bone displacements and reaction forces for use in studying bone mechanics and joint health.