The present invention relates to a method for generating a complete medical image data set from an incomplete image data set, comprising the method steps of: providing a first image data set which represents an image of a first region of a body, including at least a part of the surface of the body, at a first point in time; providing a second, incomplete image data set which represents an image of a second region of the body at a second point in time, wherein the first region and the second region overlap; providing a third data set which represents the contour of the body in the form of points on the surface of the body, substantially at the second point in time; adapting the first image data set to the second image data set by taking into account the third data set; and accepting the adapted first image data set as a complete image data set.