Medical Image Segmentation

West Virginia University Research Corporation, Morgantown WV

Investigators

Abstract

This award supports research activities under the Interdisciplinary Grants in the Mathematical Sciences (IGMS) program. The Principal Investigator, a professor of Mathematics at West Virginia University, will spend a year working in the Medical Image Processing Group (MIPG), a section within the Department of Radiology, University of Pennsylvania Medical Center. The time spent by the PI in MIPG will allow him to fully submerge in the medical image processing research. The main research goal of the project is to develop mathematical methods that will optimize algorithms used in medical image processing. The PI will focus on the algorithms concerning segmentation of the images. In particular, he will concentrate on: (1) improving the topologically motivated algorithms (i.e., so called fuzzy connectedness, FC, algorithms, based on the notion of path strength); (2) comparing FC algorithms with the algorithms based on the analysis and/or differential equations ideas (like curve propagation algorithms, implemented as level set methods of fast marching methods), underpinning the similarities and differences between both approaches, and creating new algorithms combining their strengths. More generally, the PI plans to compare all currently existing segmentation algorithms, express them in a common theoretical framework, and fuse the ideas from different approaches to create new generation and better algorithms. Practical implementation of the new methods is also planned in which the expertise of MIPG staff will be crucial. The PI will develop a set of skills that would allow him to continue his medically driven imaging research and to pass his expertise to a new generation of students. This IGMS project is jointly supported by the MPS Office of Multidisciplinary Activities (OMA) and the Division of Mathematical Sciences (DMS).