A smooth and stable fly-through scheme for virtual navigation of 3D CT colonography

This paper presents an efficient algorithm for virtual fly-through of CT colonography. The algorithm first segments colon using a fuzzy based region growing technique. Then, a skeletonization method based on a voxel coding scheme to define an optimum path for virtual camera is applied on segmented colon. The optimum path is a connected and one-voxel-thin centerline of a 3D object, which provides the location of a virtual camera for the fly-through. Finally, a smooth and stable virtual camera view direction on the optimum path is recursively calculated based on updating the view direction of the current site by adding a deviation vector that is determined by the weighted average of the deviation vector of the next n voxels with respect to the current location. The proposed system was applied on several CT colonography data. The experimental result was visually inspected by a qualified radiologist and it demonstrates the efficiency of the algorithm.