Image Understanding And Registration Of Positron Emission Tomography (PET) And Magnetic Resonance (MR) Images

In this paper we present the use of a priori knowledge in the registration of PET and MR brain images and in the construction of a knowledge base of brain image data. The registration is achieved through the use of brain anatomical landmarks as fiduciary marks. The interhemispheric fissure, a surface landmark, is segmented automatically. Further segmentation could be used in building a knowledge base system. To ensure efficient use of the knowledge, a suitable organization for the image and associated data base is proposed. This type of approach could be used to construct an expert system for the correlation of structure and function of the brain and other anatomical components.