Dissertation/Thesis Abstract

Differentiating Cardiac Adipose Tissue Using Spectral Analysis of Raw Ultrasound Radiofrequency Data
by Karlapalem, Akhila, M.S., Southern Illinois University at Edwardsville, 2018, 59; 13420346
Abstract (Summary)

Obesity has become a significant concern for severe health risks and has caused millions of deaths globally. It is defined as excess accumulation of adipose tissue in the body. Cardiac adipose tissue deposited around the myocardium has been recognized as a significant contributor to several non-communicable diseases such as metabolic syndrome, cardiovascular diseases, coronary artery diseases, insulin resistance, hypertension and even some forms of cancer. As a preventive measure, quantification of the amount of adipose tissue deposited around the myocardium is vital to understand the severity of the health risk it poses. The study presented here aims to investigate spectral analysis techniques for ultrasonic tissue characterization of cardiac adipose tissue, myocardium, and blood. Autoregressive spectral estimation techniques were used in this study to compute the power spectral densities of the raw ultrasound radiofrequency data obtained from echocardiography of seven volunteer participants. Thirteen spectral features were computed from the power spectral densities of three different bandwidth ranges and random forest classifiers were generated using these spectral features. Seventy-five percent of the available data were divided into the training set to train the classifier, and the remaining data was used as a test set. The classifiers were tested on the test set data and yielded an overall average accuracy of 94.1%. This result in the preliminary study shows the capability of ultrasound radiofrequency data in tissue characterization and lays the foundation for further investigation of cardiac tissue characterization techniques.

Indexing (document details)
Advisor: Klingensmith, Jon D.
Commitee: Noble, Bradley, Wang, Yadong
School: Southern Illinois University at Edwardsville
Department: Electrical Engineering
School Location: United States -- Illinois
Source: MAI 58/04M(E), Masters Abstracts International
Subjects: Engineering, Electrical engineering
Publication Number: 13420346
ISBN: 978-0-438-82506-2
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy