Medical Imaging: Translating 2 Dimensional Mri Scans of the Human Forearm into 3 Dimensional Dielectric Phantoms

By Esuabom David Dijemeni

Author Esuabom Dijemeni takes readers on an easy, simple and interesting scientific literary journey through his new book, Human Forearm into 3 Dimensional Dielectric Phantoms project conducted at the University of Bristol to understand and model the differences in the dielectric properties between an osteoporotic bone and a healthy bone.

The International Osteoporosis Foundation reveals that at least one in three women and one in five men over the age of 50 will suffer a fracture caused by weak bones. As osteoporosis threatens the health of aging individuals, studies related to the subject have become a continuous endeavor. The research presented in this book by the author was a foundational step to model dielectric properties of the forearm. The global aim of the research was to understand and model the differences in the dielectric properties between an osteoporotic bone and a healthy bone. The research proposed a new computational method of classifying an osteoporotic bone from a healthy bone.

Packed with knowledge and in-depth informationlike image segmentation, boundary detection and Dijemeni algorithm, to name a few MRI Scans of the Human Forearm into 3 Dimensional Dielectric Phantoms read. The knowledge acquired is unbelievably fascinating.

• Language: English
• Publisher: Xlibris UK
• Release date: Dec 20, 2012
• ISBN: 9781479705863

Esuabom David Dijemeni

Esuabom Dijemeni is a young Christian writer who currently studies Bachelors in Electronics and Communication Engineering at the University of Bristol. He was the Secretary and a student leader of Agape Student life at Bristol. His desire to know more about God led him to start studying External Certificate in Bible Studies at Mattersey Hall. During his College days in Cambridge, he experienced God's supernatural transforming power. He is passionate to see both young and old experience the abundant joy, boundless grace and unending love of God. His very popular saying is 'God is the most amazing friend you can ever have'.

Book preview

Copyright © 2012 by Esuabom Dijemeni. 304057-DIJE

Library of Congress Control Number: 2012915646

ISBN: Hardcover 978-1-4797-0585-6

Softcover 978-1-4797-0584-9

Ebook 978-1-4797-0586-3

All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the copyright owner.

To order additional copies of this book, contact:

Xlibris Corporation

0-800-644-6988

www.xlibrispublishing.co.uk

Orders@ Xlibrispublishing.co.uk

CONTENTS

Aknowledgements

Dedication

Credits

Abstract

Table Of Figures

1 Introduction

2 Literature Review

3 Image Segmentation

4 Translating 2D Segmented Image To 3D Uniform Dielectric Phantom At 2 Ghz

5 Further Development

6 Project Evaluation

7 Future Work

8 Conclusion

9 References

10 Appendix

AKNOWLEDGEMENTS

I would like to thank the project supervisors, Professor Ian Craddock and Dr Tommy Henriksson, for all their guidance, continuous encouragement, and motivational support throughout the project.

DEDICATION

To my Father,

Steve Dijemeni

CREDITS

Publishing Consultant

Fey Gano

Author Services Representative

Naomi Orleans

Manuscript Services Representative

James Calonia

Marketing Services Consultant

Casey Diaz

Project Motivator

Tomiwa Adewunmi

Sponsor

NOTOS Ventures

ABSTRACT

Understanding the causes and effects of osteoporosis on the human forearm requires knowledge of bone imaging and modelling the dielectric properties of the human forearm. The primary aim of the project was to translate 2 dimensional (2D) MRI scans of the human forearm into 3 dimensional (3D) uniform dielectric phantoms. The techniques used for image segmentation were edge detection, boundary detection, and the Dijemeni algorithm.

The Dijemeni algorithm was able to segment the image into 18 different regions to form nine different parts of the bone. The dielectric values at 2 GHz were directly mapped to the 18 different regions to create 2D dielectric phantoms of the forearm. The 2D dielectric phantoms were reproduced in the z-plane to create a uniform 3D dielectric phantom of the human forearm at 2 GHz.

A modal averaging technique was used to interpolate the 2D dielectric phantoms in order to reduce the image resolution for microwave finite difference time domain (FDTD) reconstruction of the 2D and 3D dielectric phantoms. Afterwards, homomorphic filtering was applied to the MRI scans to suppress the illumination (low frequency) and enhance the reflectance (high frequency) component of the MRI scans. The image histogram of the segmented MRI scans before and after homomorphic filtering showed different variations in the bone structure and revealed there was no significant improvement in the segmentation process when homomorphic filtering was applied.

A linear interpolation technique was applied to the three MRI scans of the forearm segmented by the Dijemeni algorithm to create a more realistic phantom of the human forearm. The Cole-Cole equation was used to model the dielectric spectrum of the different parts of the bone from 10 Hz to 100 GHz. The result of the Cole-Cole equation was applied to the 3D varying phantoms to create 3D varying dielectric phantoms from 10 Hz to 100 GHz.

TABLE OF FIGURES