- Title
- Competitive Medical Image Segmentation with the Fast Marching Method
- Author
- Hearn, Jonathan
- Degree
- Master of Sciences (Engineering), Case Western Reserve University,
Computer Engineering, 2008.
- Advisor
- M. Cenk Cavusoglu
- Pages
- 57p.
- Abstract
- Extensions to the fast marching method are introduced with the aim of further automating the segmentation of medical image data. A competitive algorithm is used to minimize the occurrence of bleeding across boundaries, and techniques including automatic starting point selection, statistical region combination, and user-influenced region expansion are introduced to support the competitive paradigm. Results show good performance in a variety of scenarios, with poorer performance in the presence of high-intensity noise and unusually obscure boundaries.
- Keywords
- Fast Marching Method; Level set Method; Image Segmentation; MRI; Competitive
Document number: case1201033602.
Bookmark this page as
<http://rave.ohiolink.edu/etdc/view?acc_num=case1201033602>.
