A method and a device are disclosed for automated detection of at least one part of the central line of at least one portion of a tubular tissue structure. In at least one embodiment of the method, characteristic landmarks of the tubular tissue structure are detected in a 3D data record of the tubular tissue structure the detected characteristic landmarks of the tubular tissue structure and landmarks of a model of landmarks, which model belongs to the tubular tissue structure and takes into account the position of the landmarks relative to one another, are related to one another such that at least one portion of the tubular tissue structure is identified in the 3D data record of the tubular tissue structure. The central line of the identified portion of the tubular tissue structure is determined at least in part on the basis of at least one detected characteristic landmark and/or at least one landmark of the model of landmarks.
