Minimally invasive surgery---with its reduced patient morbidity---has created new challenges for visualization during surgical procedures. However, the introduction of surgical navigation and high-definition digital endoscopy offers the opportunity to apply methods from computer vision to provide visualization enhancements such as registration, motion tracking, and augmented reality. These enhancements have particular value to surgeries of the skull base, which require high accuracy since surgeons work within millimeters of critical neurovascular structures. Since the accuracy of today's navigation systems is limited by the indirect relationship between the navigation system, the image and the patient, computer vision methods could potentially provide higher-accuracy solutions.;In this dissertation, we propose a method to directly track the position of the endoscope using video data acquired from the endoscope camera. Our method tracks image-feature points in the video and reconstructs the image-feature points to produce three-dimensional (3D) points. It then registers the reconstructed point cloud to the preoperative computed tomography (CT) data and continues to track. The system is further enhanced by registration to intraoperative C-arm cone-beam CT (CBCT) that captures any anatomical changes that may have occurred. This in turn provides high-accuracy registration of video to the surgical scene. The resulting registration and tracking enable the visualization of planning data directly in the endoscopic video. We validate our system with phantom and cadaver studies and demonstrate significant improvement over current navigation systems.
展开▼
