Year of Publication

2019

Season of Publication

Spring

Paper Type

Master's Thesis

College

College of Computing, Engineering & Construction

Degree Name

Master of Science in Computer and Information Sciences (MS)

Department

Computing

NACO controlled Corporate Body

University of North Florida. School of Computing

First Advisor

Dr. Sherif Elfayoumy

Second Advisor

Dr. Roger Eggen

Third Advisor

Dr. Sanjay Ahuja

Department Chair

Dr. Sherif Elfayoumy

College Dean

Dr. William Klostermeyer

Abstract

The research presented here supports the ongoing need for automatic heart volume calculation through the identification of the left and right ventricles in MRI images. The need for automated heart volume calculation stems from the amount of time it takes to manually processes MRI images and required esoteric skill set. There are several methods for region detection such as Deep Neural Networks, Support Vector Machines and Ant Colony Optimization. In this research Ant Colony Optimization (ACO) will be the method of choice due to its efficiency and flexibility. There are many types of ACO algorithms using a variety of heuristics that provide advantages in different environments and knowledge domains. All ACO algorithms share a foundational attribute, a heuristic that acts in conjunction with pheromones. These heuristics can work in various ways, such as dictating dispersion or the interpretation of pheromones. In this research a novel heuristic to disperse and act on pheromone is presented. Further, ants are applied to more general problem than the normal objective of finding edges, highly qualified region detection. The reliable application of heuristic oriented algorithms is difficult in a diverse environment. Although the problem space here is limited to MRI images of the heart, there are significant difference among them: the topology of the heart is different by patient, the angle of the scans changes and the location of the heart is not known. A thorough experiment is conducted to support algorithm efficacy using randomized sampling with human subjects. It will be shown during the analysis the algorithm has both prediction power and robustness.

Share

COinS