Risk analysis for intraoperative liver surgery

Hepatic vessel structure is very important to ensure the blood supply of the liver tissue. Therefore the knowledge of the hepatic vessel system is indispensable in liver surgery planning, for example before performing a liver resection. The purpose of this paper is to present an easy to use and fast method concerning hepatic vessel segmentation and risk analysis, which is intended to be applicable in clinical routine. Using CT scans vessels cannot be easily distinguished from other liver tissues. The segmentation algorithm used in this approach is mainly based on the arboreal structure of the hepatic vessel system. It is fully automatic and a prerequisite for the performance of a risk analysis concerning the minimal distances between tumors and vessels. A set of 20 oncological patient datasets was used to evaluate the segmentation algorithm and the risk analysis relating to their speed performance and ease of use, respectively. Segmentation algorithm was always performed in less than 1 minute and risk analysis even in less than 10 seconds. Each step was performed fully automatical. The obtained results show, that both segmentation algorithm and risk analysis are easy to use because no user interaction is required. In combination with the speed performance it is possible for the surgeon to accomplish a preoperative and intraoperative liver surgery planning on his own, respectively.