Institute for Pure and Applied Mathematics (IPAM)
Random Shapes Workshop IV: Image Processing for Random Shapes: Applications to Brain Mapping, Geophysics and Astrophysics
University of California, Los Angeles, CA
May 21-25, 2007


Poster

Automatic sulcal classification using geometric shape descriptors

Christian Laing, Juan B. Gutierrez, Deborah A. Smith, De Witt Sumners, Monica K. Hurdal
Department of Mathematics, Florida State University

Presented by J.B. Gutierrez

We have developed a computational method based on a family of geometric measures for the purpose of classification and identification of families of sulcal curves from human brain surfaces. Geometric measures involving combinations of writhe, average crossing number, ropelength and thickness of sulcal curves are computed to obtain a set of feature vectors in a high dimensional vector space. We then reduce the dimensionality of these vectors to find an optimal planar projection in order to identify significant clusters. Human brain surfaces were extracted from MRI scans of human brains. In our preliminary results, an automatic differentiation between sulcal paths from the left or right hemispheres, an age differentiation and a male-female classification were achieved.



Copyright 2007 by Monica K. Hurdal. All rights reserved.