MATHEMATICAL TEXTBOOK OF DEFORMABLE NEUROANATOMIES

Mathematical techniques are presented for the transformation of digita l anatomical textbooks from the ideal to the individual, allowing for the representation of the variabilities manifest in normal human anato mies. The ideal textbook is constructed on a fixed coordinate system t o contain all of the information currently available about the physica l properties of neuroanatomies. This information is obtained via senso r probes such as magnetic resonance, as well as computed axial and emi ssion tomography, along with symbolic information such as white- and g ray-matter tracts, nuclei, etc. Human variability associated with indi viduals is accommodated by defining probabilistic transformations on t he textbook coordinate system, the transformations forming mathematica l translation groups of high dimension. The ideal is applied to the in dividual patient by finding the transformation which is consistent wit h physical properties of deformable elastic solids and which brings th e coordinate system of the textbook to that of the patient. Registrati on, segmentation, and fusion all result automatically because the text book carries symbolic values as well as multisensor features.