Method for determining a three-dimensional model data set of a blood vessel system comprising at least one vessel segment

摘要

Method for determining a three-dimensional model data set of a blood vessel system of a patient comprising at least one vascular segment from a plurality of projection images of the blood vessel system, which were recorded at different imaging angles, comprising the steps of: a) dividing the projection images into image areas comprising at least one pixel each, b) determining a feature vector, for each of the image regions, c) determining a classification information that defines a membership of the respective image region to a according to anatomical default data, in particular according to an anatomical atlas defined at least the image data of at least the image area associated with the feature vector and / or in accordance with this image data Vascular segment of the blood vessel system or a non-affiliation to the blood vessel system describes, for each of the image areas by applying a classification function on the feature vector associated with the image area, the classification function being trained by training data records annotated with classification information on at least one person, d) segmenting the blood vessel system in the projection images by grouping image areas with the same classification information, e) calculating the three-dimensional model data set depending on the segmented projection images and the classification information.