Frameless anatomy-based 2d/3d image registration

Abstract

The present invention relates to the calculation of a 3D representation of a vascular malformation of a vascular structure of an anatomy and to the registration of 2D and 3D medical images for said calculation. The advantages of the used 2D X-ray image data are the high resolution and the fact that differences between arteries and veins are highly visible for the medical practitioner, in particular with respect to the distribution evolvement of a contrast agent within the vascular structure. These advantages of the X-ray image data are transferred into the 3D image data of the same vascular structure when applying the method of the present invention. In particular, the present invention relates to a computer-implemented medical method of 2D/3D image co-registration, a corresponding computer program, a non-transitory program storage medium storing such a program, a computer for executing the program, as well as a medical image analysing system and a radiotherapy treatment system. The method comprises the steps of providing 2D X-ray image data of a vascular structure containing a contrast agent, step S1. Therein the vascular structure contains a vascular malformation. Furthermore, the steps of providing 3D image data of the vascular structure, step S2, and transforming the 3D image data into transformed 2D image data, step S3, are comprised as well. Further, the 2D X-ray image data and the transformed 2D image data are overlaid and a correspondence between the 2D X-ray image data and the transformed 2D image data are calculated, thereby determining/achieving a co-registration between the 2D X-ray image data and the 3D image data, i.e. step S4. Moreover, the steps of determining in the co-registered 2D X-ray image data the vascular malformation, step S5, and calculating the 3D representation of the vascular malformation based on the determined vascular malformation of the co-registered 2D X-ray image data, step S6, are comprised by the method.

Dependencies are inferred from the claim text.