Modelling and remodelling are the processes by which bone adapts its shape and internal structure to external influences. However, the cellular mechanisms triggering osteoclastic resorption and osteoblastic formation are still unknown. In order to investigate current biological theories, in silico models can be applied. In the past, most of these models were based on the continuum assumption, but some questions related to bone adaptation can be addressed better by models incorporating the trabecular microstructure. In this paper, existing simulation models are reviewed and one of the microstructural models is extended to test the hypothesis that bone adaptation can be simulated without particular knowledge of the local strain distribution in the bone. Validation using an experimental murine loading model showed that this is possible. Furthermore, the experimental model revealed that bone formation cannot be attributed only to an increase in trabecular thickness but also to structural reorganization including the growth of new trabeculae. How these new trabeculae arise is still an unresolved issue and might be better addressed by incorporating other levels of hierarchy, especially the cellular level. The cellular level sheds light on the activity and interplay between the different cell types, leading to the effective change in the whole bone. For this reason, hierarchical multi-scale simulations might help in the future to better understand the biomathematical laws behind bone adaptation.

This study explores a novel method to quantify in vivo soft tissue biomechanics from endoscopic confocal fluorescence microscope images of externally loaded biological tissues. A custom algorithm based on normalized cross-correlation is used to track fluorescently labeled cells within soft tissue structures as they deform. Cellular displacements are subsequently reduced to tissue strains by deriving the spatial gradient of the spline smoothed cellular displacement field. The relative performance of the tracking method is verified using a synthetic dataset with known underlying deformation. In biological application of the method, tissue strains are measured in the Achilles tendon of an anesthetized mouse. Over repeated trials, structural strain in the tendon (i.e., the relative change in distance between cells located at view field extremes) is 20.3+/-3.1%, thus establishing the reproducibility of the loading protocol. Analysis of local tendon tissue strains reveal primary engineering strains in the tissue to range from 5 to 55%, signifying a highly inhomogeneous strain state, with complex relative motions of neighboring tendon substructures. In summary, the current work establishes a baseline for a promising experimental method, and demonstrates its technical feasibility.

No abstract