Design And Implementation Of The Navigation System Computer-assisted Minimal Invasive Cardiac Surgery

Computer-assisted surgical navigation technology taking advantage of many kinds of medical images not only can help surgeon doing the preoperative plan and minimizing the wound, but also can bring the surgery high quality and convenience. Computer-assisted surgical navigation fuses the real world scenario and the virtual models, which are painted in the computer and created based on the medical data obtained from the medical imaging equipment such as CT and MRI, so that the user can get extra information from the virtual models and the real environment can be augmented. Under the assistance of the navigation system, the surgeon can gain enough information which can guide the surgeon to achieve the surgery. The work of this thesis is to apply the computer-assisted surgical navigation technology into the minimal invasive cardiac surgery.This thesis designed and implemented a navigation system for computer-assisted minimal invasive cardiac surgery. Before the discussion of the system design, we introduced the knowledge of image segmentation, medical data visualization and the relative algorithms which are used in the system. Then we described the preoperative port placement planning module of the system. In this part, we focused on the definition of surgical target, computation of candidate ports and the selection of optimum ports for the endoscope and instruments. The selection of ports were based on four optimization criterions proposed in the thesis. In the surgical navigation chapter, we described the navigation module which can locate the endoscope and surgical instruments accurately, show the relative positions between the endoscope, surgical instruments and the patient's anatomical structures real-timely and build an augmented reality environment consisting of endoscope image, virtual endoscope image and ultrasound image.After that we commented the design of the system. To testify the proposed system, several experiments were done. In order to verify our preoperative port placement planning algorithm, several clinical cases were performed to compare our port placement algorithm to traditional method. The phantom was employed to do the test and the result reveals that the registration error is 1.08±0.16mm. And the distance measure test of skeleton model shows that the error is 1.04±0.43mm which means that the system can locate the surgical target, surgical instruments and endoscope precisely. An animal study demonstrating the validity of the whole system was also performed, which reveals that the system error is 5. 05±0.67mm.