CT colonography is an emerging technique for colorectal cancer screening. Using this technique virtual reality models of the human colon are generated from an abdominal CT study of a suitably prepared patient. Manual navigation through these models is a slow and tedious process. It is possible to automate navigation by calculating the centerline of the human colon. There are numerous well documented approaches for centerline calculation, most of these have been developed as alternatives to 3D topological thinning which is regarded as being far too computationally intensive for use with large volumetric datasets. This paper describes a fully automated, optimised version of 3D topological thinning that has been specifically developed for calculating the centerline of the human colon. This algorithm is fast, portable and robust. Initial tests have been performed on real patient data and the results are promising.
