Multi-scale rigid registration to detect damage in micro-CT images of progressively loaded bones

We present a method to detect damage in time-lapsed, micro-CT images of progressively loaded bone. The method we have developed splits the image into regions and performs registration on each region individually. The procedure is repeated with progressively smaller regions until either a minimum size or a maximum number of levels is reached. The regions are then classified as damaged or undamaged. This method has been successfully applied to three sets of images and tested with simulated damage. It will allow us to determine the characteristics of trabecular architecture that provide strength and to predict regions that are most likely to be damaged based on their structure.