Liver Segmentation and its Application to Hepatic Interventions

Abstract

The thesis addresses the development of an intuitive and accurate liver segmentation approach, its integration into software prototypes for the planning of liver interventions, and research on liver regeneration. The developed liver segmentation approach is based on a combination of the live wire paradigm and shape-based interpolation. Extended with two correction modes and integrated into a user-friendly workflow, the method has been applied to more than 5000 data sets. The combination of the liver segmentation with image analysis of hepatic vessels and tumors allows for the computation of anatomical and functional remnant liver volumes. In several projects with clinical partners world-wide, the benefit of the computer-assisted planning was shown. New insights about the postoperative liver function and regeneration could be gained, and most recent investigations into the analysis of MRI data provide the option to further improve hepatic intervention planning.