A medical image traditionally represents a "snapshot" of a portion of the patient?s anatomy at a particular moment in time. However, the appearance of image objects often varies between images taken at different times. Such changes may occur because of differences in the image acquisition techniques employed, because of shifts in the positions of organs, or because of development, progression, or regression of pathology.We have developed a method of defining vessel "trees" from 3D image data. These vessel trees can then be registered with any type of 2-or 3D image data obtained from the same patient and that show the vasculature. Changes in vascular configuration can then be used not only to clarify vascular changes over time, but also to help determine the location and/or change in ther image objects. As vessels are present throughout the body, the approach is applicable to any anatomical region.