A Statistical Test to Assess Tissue Characterization of Abdominal Organs By Color-Fusion MRI

Date of Award

2013

Degree Type

Thesis

Degree Name

Master of Science in Biomedical Sciences

Department

Basic Sciences

First Advisor

H Keith Brown, PhD

Second Advisor

Kimberly Deadmond, PhD

Third Advisor

Lori Hardy, PhD

Abstract

Color fusion MRI is being investigated for its value in automatic segmentation of tissues. An existing color fusion MRI data set of the liver, pancreas, and kidney of a normal male volunteer was analyzed both visually and statistically. Automatic tissue segmentation can allow better differentiation of abdominal pathologies, as well as pathologies associated with other organs. My research hypothesis is that fuzzy c-means clustering can be used to quantify the confidence levels of the correct classification of renal, pancreatic, and hepatic tissues visualized by the color fusion MRI method. Results from data show that fuzzy c-means clustering can be used to validate the correctness of classification of abdominal tissues that are visualized by color fusion MRI.

This document is currently not available here.

Share

COinS