Multiple formulations of bone adaptation models are compared on the case of implanted rat tibiae with external activation.
Bone adaptation threshold is determined through a histogram analysis of the stimulus in physiological homeostatic conditions.
The strain based stimulus was shown the most appropriate to implement a robust adaptation model at the continuum scale.
The use of a ‘zone of influence’ filter larger that 3–5 trabecular spaces was found mandatory to obtain reliable results.
Other modeling parameters were found to have a limited influence and their influence is discussed.