In this paper we present a new metric preserving deformation method which permits to generate smoothed representations of neuroanatomical structures. These surfaces are approximated by triangulated meshes which are evolved using an external velocity field, modified by a local curvature dependent contribution. This motion conserves local metric properties since the external force is modified by explicitely including an area preserving term into the motion equation. We show its applicability by computing inflated representations from real neuroanatomical data and obtaining smoothed surfaces whose local area distortion is less than a 5 %, when comparing with the original ones. © Springer-Verlag Berlin Heidelberg 2007.
CITATION STYLE
Rueda, A., Perea, Á., Rodríguez-Pérez, D., & Romero, E. (2007). A fast mesh deformation method for neuroanatomical surface inflated representations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4872 LNCS, pp. 75–86). Springer Verlag. https://doi.org/10.1007/978-3-540-77129-6_11
Mendeley helps you to discover research relevant for your work.