Image Analysis and Modeling in Ophthalmology

409 Pages·2014·7.6 MB·English

by  Eddie Y. K. Ng

BIOMEDICAL ENGINEERING N
g
Digital fundus images can effectively diagnose glaucoma and diabetes retinopathy, while infrared imaging | A
can show changes in the vascular tissues. Likening the eye to the conventional camera, Image Analysis and c Image Analysis
h
a
Modeling in Ophthalmology explores the application of advanced image processing in ocular imaging. This r
y
book considers how images can be used to effectively diagnose ophthalmologic problems. It introduces a
multi-modality image processing algorithms as a means for analyzing subtle changes in the eye. It details eye | C
a
imaging, textural imaging, and modeling, and highlights specific imaging and modeling techniques. m
p
i
The book covers the detection of diabetes retinopathy, glaucoma, anterior segment eye abnormalities, lh and Modeling
o
instruments on detection of glaucoma, and development of human eye models using computational fluid | S
dynamics and heat transfer principles to predict inner temperatures of the eye from its surface temperature. u
r
i
It presents an ultrasound biomicroscopy (UBM) system for anterior chamber angle imaging and proposes an
automated anterior segment eye disease classification system that can be used for early disease diagnosis and
treatment management. It focuses on the segmentation of the blood vessels in high-resolution retinal images
and describes the integration of the image processing methodologies in a web-based framework aimed at
I in
retinal analysis.
m
The authors introduce the A-Levelset algorithm, explore the ARGALI system to calculate the cup-to-disc a
g
ratio (CDR), and describe the Singapore Eye Vessel Assessment (SIVA) system, a holistic tool which brings
e
together various technologies from image processing and artificial intelligence to construct vascular models

A
from retinal images. The text furnishes the working principles of mechanical and optical instruments for the Ophthalmology
diagnosis and healthcare administration of glaucoma, reviews state-of-the-art CDR calculation detail, and n
discusses the existing methods and databases. a
l
y
Image Analysis and Modeling in Ophthalmology includes the latest research development in the field of eye s
modeling and the multi-modality image processing techniques in ocular imaging. It addresses the differences, i
s
performance measures, advantages and disadvantages of various approaches, and provides extensive reviews
a
on related fields. n
d
E. Y. K. Ng , PhD, received his PhD from Cambridge University, UK and is an associate professor at Nanyang
EEddiitteedd bbyy
Technological University, Singapore. He serves as editor for six international journals and as Editor-in Chief for two M
SCIE indexed Js. His research interests are in thermal imaging, biomedical engineering, breast cancer detection, and
computational fluid dynamics and heat transfer. o
d EE.. YY.. KK.. NNgg
e
l
U. Rajendra Acharya, PhD, DEng, is a Visiting faculty in Ngee Ann Polytechnic, Singapore and Adjunct Professor in in UU.. RRaajjeennddrraa AAcchhaarryyaa
University of Malaya, Malaysia. He received his PhD from National Institute of Technology Karnataka, Surathkal, India
g
and D Engg from Chiba University, Japan. He has published more than 285 papers, in refereed international SCI-IF

journals (178), international conference proceedings (48), textbook chapters (62), books (15 including in Press) with i AAuurréélliioo CCaammppiillhhoo
n
h-index of 24 (h-index =22 without self-citations, Scopus). 
O
JJaassjjiitt SS.. SSuurrii
p
Aurélio Campilho, PhD, is a Professor at University of Porto, and coordinator of the Bioimaging research group, of the
h
Biomedical Engineering Institute, Portugal. His main research interest is in medical image analysis. He served as an
t
Associate Editor of IEEE Trans. on Biomedical Engineering and of Machine Vision Applications journal. He authored one h
book, co-edited 12 books, and published about 150 papers in journals and conferences.
a
l
m
Jasjit S. Suri, MS, PhD, MBA, Fellow AIMBE, is an innovator, scientist, and an internationally known world leader and o
has spent over 20 years in the field of biomedical engineering/sciences and its management. He earned his doctorate l
o
from the University of Washington, Seattle and MBA from Weatherhead, Case Western Reserve University, Cleveland.
g
Suri was awarded the President’s Gold Medal in 1980 by the
y
Directorate General National Cadet Corps and has been named a
Fellow of American Institute of Medical and Biological Engineering
for his outstanding contributions.
6000 Broken Sound Parkway, NW
Suite 300, Boca Raton, FL 33487
711 Third Avenue
New York, NY 10017
an informa business
2 Park Square, Milton Park w w w . c r c p r e s s . c o m
www.crcpress.com Abingdon, Oxon OX14 4RN, UK
Image Analysis
and Modeling
in Ophthalmology

Image Analysis
and Modeling
in
Ophthalmology
Edited by
E. Y. K. Ng
U. Rajendra Acharya
Aurélio Campilho
Jasjit S. Suri
Boca Raton London New York
CRC Press is an imprint of the
Taylor & Francis Group, an informa business
MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does not warrant the
accuracy of the text or exercises in this book. This book’s use or discussion of MATLAB® software or related products
does not constitute endorsement or sponsorship by The MathWorks of a particular pedagogical approach or particular
use of the MATLAB® software.
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2014 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Version Date: 20131108
International Standard Book Number-13: 978-1-4665-5938-7 (eBook - PDF)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been
made to publish reliable data and information, but the author and publisher cannot assume responsibility for the valid-
ity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright
holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this
form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may
rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or uti-
lized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopy-
ing, microfilming, and recording, or in any information storage or retrieval system, without written permission from the
publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://
www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923,
978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For
organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface ............................................................................................................................................vii
Editors ..............................................................................................................................................xi
Contributors .................................................................................................................................xiii
1. Ultrasound Biomicroscopic Imaging of the Anterior Chamber Angle .......................1
Maria Cecilia Aquino and Paul Chew
2. Automated Glaucoma Identification Using Retinal Fundus Images:
A Hybrid Texture Feature Extraction Paradigm ...............................................................9
Muthu Rama Krishnan Mookiah, U. Rajendra Acharya, Chandan Chakraborty,
Lim Choo Min, Eddie Y. K. Ng, and Jasjit S. Suri
3. Ensemble Classification Applied to Retinal Blood Vessel Segmentation:
Theory and Implementation...............................................................................................23
Muhammad Moazam Fraz and Sarah A. Barman
4. Computer Vision Algorithms Applied to Retinal Vessel Segmentation
and Quantification of Vessel Caliber ...............................................................................49
Muhammad Moazam Fraz and Sarah A. Barman
5. Segmentation of the Vascular Network of the Retina ..................................................85
Ana Maria Mendonça, Behdad Dashtbozorg, and Aurélio Campilho
6. Automatic Analysis of the Microaneurysm Turnover in a Web-Based
Framework for Retinal Analysis .....................................................................................111
Noelia Barreira, Manuel G. Penedo, Sonia González, Lucía Ramos, Brais Cancela,
and Ana González
7. A-Levelset-Based Automatic Cup-to-Disc Ratio Measurement for Glaucoma
Diagnosis from Fundus Image ........................................................................................129
Jiang Liu, Fengshou Yin, Damon Wing Kee Wong, Zhuo Zhang, Ngan Meng Tan,
Carol Cheung, Mani Baskaran, Tin Aung, and Tien Yin Wong
8. The Singapore Eye Vessel Assessment System ............................................................143
Qiangfeng Peter Lau, Mong Li Lee, Wynne Hsu, and Tien Yin Wong
9. Quantification of Diabetic Retinopathy Using Digital Fundus Images .................161
Hasan Mir, Hasan Al-Nashash, and U. Rajendra Acharya
10. Diagnostic Instruments for Glaucoma Detection ........................................................171
Teik-Cheng Lim, U. Rajendra Acharya, Subhagata Chattopadhyay, Jasjit S. Suri,
and Eddie Y. K. Ng
© 2008 Taylor & Francis Group, LLC v
vi Contents
11. Automated Cup-to-Disc Ratio Estimation for Glaucoma Diagnosis
in Retinal Fundus Images .................................................................................................179
Irene Fondón, Carmen Serrano, Begoña Acha, and Soledad Jiménez
12. Arteriovenous Ratio Calculation Using Image-Processing Techniques .................203
Manuel G. Penedo, Sonia González, Noelia Barreira, Marc Saez, Antonio Pose-Reino,
and María Rodríguez-Blanco
13. Survey on Techniques Used in Iris Recognition Systems .........................................227
Nagarajan Malmurugan, Shanmugam Selvamuthukumaran,
and Sugadev Shanmugaprabha
14. Formal Design and Development of an Anterior Segment Eye Disease
Classification System .........................................................................................................245
Oliver Faust, Chan Wei Yan, Muthu Rama Krishnan Mookiah, U. Rajendra Acharya,
Eddie Y. K. Ng, and Wenwei Yu
15. Modeling of Laser-Induced Thermal Damage to the Retina and the Cornea .......265
Mathieu Jean and Karl Schulmeister
16. Automatization of Dry Eye Syndrome Tests.................................................................293
Manuel G. Penedo, Beatriz Remeseiro, Lucía Ramos, Noelia Barreira,
Carlos García-Resúa, Eva Yebra-Pimentel, and Antonio Mosquera
17. Thermal Modeling of the Ageing Eye ............................................................................321
Anastasios Papaioannou and Theodoros Samaras
18. A Perspective Review on the Use of In Vivo Confocal Microscopy
in the Ocular Surface .........................................................................................................339
Sze-Yee Lee, Andrea Petznick, Shakil Rehman, and Louis Tong
19. Computational Modeling of Thermal Damage Induced by Laser
in a Choroidal Melanoma .................................................................................................367
José Duarte da Silva, Alcides Fernandes, Paulo Roberto Maciel Lyra,
and Rita de Cássia Fernandes de Lima
© 2008 Taylor & Francis Group, LLC
Preface
The human eye is a complex and important organ that works similarly to the conven-
tional camera. Many advanced image-processing algorithms have been proposed to
analyze the subtle changes in the eye to diagnose eye abnormalities efficiently. Digital
fundus images have been used efficiently for the diagnosis of diabetes retinopathy and
glaucoma. Infrared imaging provides a temperature profile that depicts changes in the
vascular tissues, which helps to study the ocular surface temperature (OST) and ocular
diseases like dry eye and cataracts. This book covers the detection of diabetes retinopa-
thy, glaucoma, and anterior segment eye abnormalities; instruments for the detection
of glaucoma; and the development of a human eye model using computational fluid
dynamics and heat transfer principles to predict inner temperatures of the eye from its
surface temperature.
Ultrasound biomicroscopy (UBM) is one of the imaging systems that allow visualization
of the anatomy and pathology of the anterior segment. It is particularly beneficial in evalu-
ating regions of the eye behind the iris such as the ciliary body, lens zonular attachment,
and lens periphery, which are obscured in other anterior scanning systems. In glaucoma
imaging, UBM plays a significant role in objective assessment of peripheral anterior cham-
ber angle morphology, which is useful in angle closure glaucoma diagnosis and manage-
ment. Chapter 1 presents a UBM system for anterior chamber angle imaging.
Chapter 2 describes both the formal design and development of an automated anterior
segment eye disease classification system. The proposed system can be used for early dis-
ease diagnosis and treatment management. The classification is done with a two-step pro-
cessing model. The first processing step extracts nine features from optical eye image data
using higher-order spectra, discrete wavelet transform, and texture techniques. The pro-
cessing step feeds these clinically significant features to a support vector machine (SVM)
algorithm for automated classification. The authors have obtained classification accuracy
of 90%, sensitivity of 93.8%, and specificity of 100%.
The segmentation and structural properties of the blood vessels of the retina such as
width, length, branching pattern, and angles are important features during the screen-
ing of diabetes, eye diseases, and cardiovascular diseases. An efficacious retinal vessel
segmentation methodology is presented in Chapter 3, which is founded on supervised
classification using an ensemble classifier of boosted and bagged decision trees. In this
chapter, a more detailed theoretical background, experimental evaluation, and analysis of
results are presented.
In Chapter 4, a detailed review, analysis, and categorization of retinal vessel segmen-
tation and caliber measurement techniques are presented. The objective is to provide a
detailed resource of the algorithms employed for vessel segmentation and width measure-
ment as a ready reference. The key points are highlighted, the differences and performance
measures are illustrated, and the advantages and disadvantages of various approaches are
discussed.
Chapter 5 focuses on the segmentation of the blood vessels in high-resolution retinal
images. Very recent methods (from the last two to three years) for retinal image analysis are
outlined first, by surveying some papers in the segmentation methodologies and their clini-
cal applications. In the next sections, methodology for blood vessel retinal segmentation,
© 2008 Taylor & Francis Group, LLC vii
viii Preface
with special emphasis on the segmentation of high-resolution fundus images and its
validation for the arteriovenous ratio (AVR) calculation, is discussed.
An automated procedure for the location of microaneurysms from retinal images as well
as a methodology for lesion registration in order to compute the microaneurysm turn-
over is proposed in Chapter 6. This chapter also describes the integration of the image-
processing methodologies in a web-based framework aimed at retinal analysis. The web
framework simplifies data management and provides a user-friendly interface to interact
with the image-processing algorithms.
To boost the performance of the level set algorithm, the A-Levelset algorithm, which
cascades the level set and active shape model, is proposed in Chapter 7. The A-Levelset-
based ARGALI system is built to automatically segment the optic cup and optic disc from
2D digital fundus images. The ARGALI system further calculates the cup-to-disc ratio
(CDR), which is an important indicator for glaucoma assessment and diagnosis. It paves
the way for automatic objective glaucoma diagnosis and screening using widely available
fundus images.
Chapter 8 describes a holistic tool, the Singapore Eye Vessel Assessment (SIVA) system,
which brings together various technologies from image processing and artificial intelligence
to construct vascular models from retinal images. Subsequently, these models of blood ves-
sels can be queried for a variety of measurements. Incorporating automated techniques
reduces manual intervention, allowing a large number of retinal images to be used for popu-
lation studies. A number of these measurements of vascular morphology have already been
shown to be correlated with certain diseases, while others are under active study.
Diabetic retinopathy is a common complication that results in impaired visual function.
Low-cost automated detection and assessment of diabetic retinopathy is an invaluable
tool to encourage regular screenings. Chapter 9 discusses a methodology for assessing
the severity of diabetic retinopathy using digital fundus images. The methodology uses
the foveal region and exudates in order to quantify the severity of diabetic retinopathy.
Results from sample digital fundus images were used to demonstrate the behavior of the
methodology as well as its potential role as a complement to the standard ophthalmologi-
cal assessment.
Glaucoma (repeated) is typically a silent but progressive illness that increases the intra-
ocular pressure causing damage to the optic nerve. Chapter 10 reviews the use of mechani-
cal and optical instruments for the diagnosis and health-care administration of glaucoma.
The working principles of these instruments are furnished.
Glaucoma cannot be cured but when early detected and treated, blindness due to
glaucoma can be prevented. Chapter 11 reviews state-of-the-art CDR calculation and its
research application in detail. Finally, a discussion on the existing methods and databases
that facilitate the research and comparison of results among researchers and future lines
of research is presented.
Chapter 12 describes the stages involved in the computation of the AVR and explains
how it can be implemented using image-processing techniques. The problems related to
each stage are analyzed and the solutions proposed to overcome the limitations are dis-
cussed. Moreover, different approaches proposed in the literature with the aim of provid-
ing alternatives for developing an automatic methodology for the AVR computation are
analyzed.
Chapter 13 discusses the research efforts and existing techniques on iris segmentation,
normalization, and the corresponding phases of iris recognition and their limitations. It
explains clearly the need to develop new algorithms in segmentation and matching phases
of iris recognition.
© 2008 Taylor & Francis Group, LLC
Preface ix
Glaucoma is one of the most common causes of blindness. Robust mass screening may
help to extend the symptom-free life of affected patients. In Chapter 14, a novel automated
glaucoma diagnosis system based on texture features and data-mining techniques is pro-
posed. Various texture features are extracted from the digital fundus images and fed to
the SVM classifier. The SVM classifier with kernel functions of polynomial order 1 and
radial basis function (RBF) correctly identified the glaucoma and normal images with an
accuracy of 93.33%, and sensitivity and specificity of 86.7% and 100%, respectively.
Chapter 15 concentrates on the investigation of thermally induced threshold damage to
the cornea and retina. It describes the fundaments of physics-based models intended for
this specific laser–tissue interaction and for reproducing experimental threshold values.
Also, it briefly reviews the optics of the eye and of layered tissues, and discusses setting up
the bioheat equation and modeling the occurrence of macroscopic damage.
Dry eye is a symptomatic disease that affects activities of daily living, adversely affect-
ing important tasks such as computer use and driving. In practice, there are several clini-
cal tests to diagnose this syndrome by means of analyzing the tear film quality. Chapter 16
describes automatic image-processing methodologies to perform two clinical tests: the
analysis of the interference lipid pattern and the tear film breakup time test.
Chapter 17 explains the calculation of numerical OST distributions for different age
groups and conditions and compares them with measurements. It was shown that, in agree-
ment with experimental results, computational eye models, which consider the physiologi-
cal and anatomical changes with ageing and are constructed accordingly, could predict
the decrease in the minimum temperature of the cornea with age, as well as the approach
of its location to the geometric corneal center (GCC). Moreover, based on the numerical
simulations, the authors have proposed that the changes in the aqueous humor (AH) flow
that occur with age inside the anterior chamber, as a result of anatomical changes, could
be another possible mechanism explaining the above effect, in addition to reduced blood
flow and the blinking rate of the older eye.
Optical microscopy is a vital tool in life science investigations with the aim of imaging
at subcellular resolution. Confocal microscopy is used for obtaining three-dimensional
images of thick objects. Chapter 18 discusses how detailed imaging of the ocular surface
structures and associated cell types can be performed using in vivo confocal microscopy.
This technology will enable in-depth understanding of ocular surface physiology and
pathology without the necessity of tissue excisions.
The efficiency of melanoma treatment by transpupillary thermotherapy (TTT) depends
largely on the amount of energy absorbed by the tumor and surrounding tissues. The cal-
culation of the amount of energy in heterogeneous tissues represents a very complex task
due to a lack of accurate information on the coefficients of absorption, on scattering of the
laser radiation, and on the tissues’ thermophysical properties. In Chapter 19, a model is
proposed for calculating the value of thermal damage at various depths within the mela-
noma. The methodology presented here can be improved by considering correlations that
take into account the variation of the physical properties with the temperature, mainly in
melanotic tissues.
Many authors have contributed immensely and made this book possible with their
hard work and precious time. We thank them heartily for their valuable contributions. In
no particular order, they are Maria Cecilia Aquino, Paul Chew, Muthu Rama Krishnan
Mookiah, Chandan Chakraborty, Lim Choo Min, Muhammad Moazam Fraz, Sarah A.
Barman, Ana Maria Mendonça, Behdad Dashtbozorg, Noelia Barreira, Manuel G. Penedo,
Sonia González, Lucía Ramos, Brais Cancela, Ana González, Jiang Liu, Fengshou Yin,
Damon Wing Kee Wong, Zhuo Zhang, Ngan Meng Tan, Carol Cheung, Mani Baskaran,
© 2008 Taylor & Francis Group, LLC