Virtual Assistive System for Robotic Single Incision Laparoscopic Surgery

Single Incision Laparoscopic Surgery (SILS) reduces the trauma of large wounds decreasing the post-operative infections, but introduces technical difficulties for the surgeon, who has to deal with at least three instruments in a single incision. These drawbacks can be overcome with the introduction of robotic arms inside the abdominal cavity, but still remain difficulties in the surgical field vision, limited by the endoscope field of view. This work is aimed at developing a system to improve the information required by the surgeon and enhance the vision during a robotic SILS. In the pre-operative phase, the segmentation and surface rendering of organs allow the surgeon to plan the surgery. During the intra-operative phase, the run-time information (tools and endoscope pose) and the pre-operative information (3D models of organs) are combined in a virtual environment. A point-based rigid registration of the virtual abdomen on the real patient creates a connection between reality and virtuality. The camera-image plane calibration allows to know at run-time the pose of the endoscopic view. The results show how using a small set of 4 points (the minimal number of points that would be used in a real procedure) for the camera-image plane calibration and for the registration between real and virtual model of the abdomen, is enough to provide a calibration/registration accuracy within the requirements.