New calculation method of image similarity for endoscope tracking based on image registration in endoscope navigation
Review articleOpen access
2003/06/01 Full-length article DOI: 10.1016/S0531-5131(03)00331-5
Journal: International Congress Series
AbstractThis paper presents a new calculation method of image similarity for camera motion tracking in a flexible endoscope navigation system. An endoscope navigation system is a tool which provides navigation information that is acquired from pre-operative images to a medical doctor during an endoscopic examination in real time. In this system, one of the fundamental functions is to track endoscope camera motion. Since it is difficult to attach a positional sensor at the tip of a flexible endoscope, especially bronchoscope, due to space limitation, an image registration technique becomes a strong tool for camera motion tracking. The previous method used image similarity computed by summing gray-level differences up for all pixels of real and virtual endoscopic images. The method could not properly estimate the position and orientation of the endoscope due to averaging, since image similarity does not change significantly. We proposed a new image similarity measure which uses characteristic structure in computation. This method divides the original image into a set of small subblocks and selects only the subblocks in which characteristic shapes are observed. Then, an image similarity value is calculated only within the selected subblocks. We applied the proposed method to eight pairs of X-ray CT images and real bronchoscopic videos. In the experimental results, the proposed method showed much improvement in continuous tracking performance. Nearly 1000 consecutive frames were tracked correctly.
Request full text