Computational anatomy: an emerging discipline (Q4497937)

This paper studies mathematical methods in the emerging new discipline of ``Computational Anatomy''. Herein the Brown/Washington University model of anatomy following the global pattern theory is formalized. In this model, anatomies are represented as deformable templates, collections of 0, 1, 2, 3-dimensional manifolds. A typical structure is carried by the template with the variabilities accommodated via the application of random transformations to the background manifolds. The group of diffeomorphic transformations is chosen to be rich enough so that a large family of shapes may be generated with the topologies of the template maintained. For normal anatomy, one deformable template is studied corresponding to a homogeneous space. For diseases, a family of templates is introduced of perhaps varying dimensional transformation classes. There are three principal components to computational anatomy studied herein: (1) Computation of large deformation maps; (2) Computation of empirical probability laws; (3) Inferences and disease testing. Computational Anatomy is considered as belonging to the discipline of geometry, governed by pattern-theory principles, and whose kinematics is described in terms of concepts borrowed from continuum mechanics.