A method for determining a characteristic map for an object, in particular for a living being, based on at least one first image, in particular core pin resonance image

摘要

It is a system and method (12) for determining a characteristic map (82) for an object, in particular for a living being, based on at least one first image (84), in particular a core pin resonance image, mr - image, of the object stated. In the process (12), in a first step (96) a structure for a reference pairs is defined, wherein each reference pair (16 - 26) at least two entries (62) comprises. The first entry represents a property value, in particular, an attenuation value. The second entry (62) preferably represents a group of the associated image points (67) which, in particular from mr - images (28) are extracted, which permits the value of corresponding pixel of interest comprises. In a further step (98) of the method (12), a plurality of training pairs (16 - 26) is provided. A structure of the training pairs (16 - 26) corresponds to the structure of the reference electrode pairs, and the entries of the respective training pairs (16 - 26) are known. In a further step (100) of the method (12), an association between the first entries and the further entries (62 - 66) of the training pairs (16 - 26) by means of machine learning is determined, in order to thus for an arbitrary point (90) of the first image (84) a corresponding value (88) to be able to predict.