Fast and simple segmentation of vessels in the liver using intraoperative ultrasound data

In this article we present a new method for automatically identifying vessels in intraoperative ultrasound images of the liver The algorithm is part of a novel liver surgery assistance system and will be used to register the model of a planning with the intraoperative ultrasound data. The vessels and their bifurcations within the ultrasound image will, in the future, be used as landmarks and to register the planning model. To verify the quality of the segmentation, a benchmark TUMUS-01 based on manual segmented (by an expert) ultrasound images will be used to compute the quantity and the quality of the segmentation. We describe the algorithm, the benchmark and the results.