Semester

Fall

Date of Graduation

2004

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Susan K. Lemieux

Committee Co-Chair

Mark Jerabek

Abstract

Magnetic Resonance Imaging (MRI) is a widely used medical technology for diagnosis and detection of various tissue abnormalities, tumor detection, and in evaluation of either residual or recurrent tumors. This thesis work exploits MRI information acquired on brain tumor structure and physiological properties and uses a novel image segmentation technique to better delineate tissue differences.;MR image segmentation will be important in distinguishing between boundaries of different tissues in the brain. A segmentation software tool was developed that combines the different types of clinical MR images and presents them as a single colored image. This technique is based on the fuzzy c-means (FCM) clustering algorithm. The MR data sets are used to form five-dimensional feature vectors. These vectors are segmented by FCM into six tissue classes for normal brains and nine tissue classes for human brains with tumors. The segmented images are then compared with segmentation performed using Statistical Parametric Mapping (SPM2)---software that is commonly used for brain tissue segmentation. The results from segmenting the whole volume MRI using FCM show better distinction between tumor tissues than SPM2.

Download

Share

COinS