Automatic Registration for Superimposed Display of Organs for a Development of Laparoscopic Surgery Navigation

To support laparoscopic surgery which requires a high level of skill from the surgeon, we have developed a system that superimposes endoscopic images and kidney models by using augmented reality (AR) technology and considers intraoperative organ deformation. In this paper, we propose a real-time automatic alignment method for the superimposed display of endoscopic images and kidney models. This method uses feature points (markers) around the organ as reference points to align the endoscopic image in real space and the CG model in virtual space. To verify the accuracy of the proposed method, we evaluate the error between the real space and the virtual space on the superimposed display screen.