International

INTERNATIONALJournal of Advanced
JOURNAL Research in Engineering RESEARCH
OF ADVANCED and Technology IN(IJARET), ISSN 0976 –
ENGINEERING
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 2, February (2014), pp. 95-100, © IAEME
AND TECHNOLOGY (IJARET)

ISSN 0976 - 6480 (Print)

ISSN 0976 - 6499 (Online)
IJARET
Volume 5, Issue 2, February (2014), pp. 95-100
© IAEME: www.iaeme.com/ijaret.asp
Journal Impact Factor (2014): 4.1710 (Calculated by GISI) ©IAEME
www.jifactor.com

A REVIEW ON DIFFERENT GLAUCOMA DETECTION METHODS

Saja Usman1, Dimple Shajahan2

1
(Computer Science and Engineering, TKM College of Engineering, Kollam, India)
2
(Computer Science and Engineering, TKM College of Engineering, Kollam, India)

ABSTRACT

Glaucoma is one of the leading cause of blindness worldwide. It is due to the decrease in intra
ocular pressure within the eyes. The detection and diagnosis of glaucoma is very important. There
are manual and automatic detection methods available. In this paper a survey is conducted on
different glaucoma detection methods in image processing such as Scanning Laser Polarimetry,
optical coherence tomography, mfreg, Wavelet Fourier transform etc and also their advantages and
disadvantages. This paper also discusses a new method to classify the glaucomatous images which is
the proposed method.

Keywords- Glaucoma, Cup to disc ratio, Wavelet Transform, Classifiers, Artificial Neural Network,
Segmentation.

I. INTRODUCTION

Glaucoma is one of the main reason of blindness worldwide. This is caused due to the
increase in intra ocular pressure within the eyes. The intra ocular pressure increases due to the
malfunction or malformation of the drainage system of the eye. Aqueous humor- a clear fluid flow in
and out of the chamber of the eye. This fluid nourishes the nearby tissues. Aqueous humor controls
the intra ocular pressure of the eye. The pressure within the eye is maintained by providing a small
amount of aqueous humor while the same amount of fluid flows out of the eye through a drainage
system. Pressure within the eye increases because of the insufficient fluid flow between the iris and
the cornea. The increased intra ocular pressure in the eye damages the optical nerve through which
retina sends light to the brain where they are recognized as images and make vision possible. The
major reason for glaucoma is the increased pressure variation in the eye.
One technique to asses patients who are suffering from glaucoma is to analyse the ultrasound
images of the eye in order to detect the changes that reduces the flow of fluid out of the eye through
the drainage system. Usually the analysis of this images are done manually. The features that are

95
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 2, February (2014), pp. 95-100, © IAEME

analysed by the clinicians while testing the eye are the sclera- a dense fibrous opaque white outer
coat enclosing the eye ball, sclera spur- a minute triangular area in a meridional section of the sclera
tissue with its base along the inner surface of the sclera; the anterior chamber, the region surrounded
by the latter surface of the cornea and the central part of the lens; and, the trabecular-iris recess, the
top point between the sclera region and the iris.
Manual analysis of the eye is time consuming and the accuracy of the parameter
measurements also varies with different clinicians. To overcome this problems with the manual
analysis, the objective of this paper is to introduce a method to automatically analyse the ultrasound
images of the eye. Automatic analysis is much more effective than manual analysis. Since glaucoma
is the second leading a use to blindness its very important to detect it and diagonise.
The remaining of this paper is organized as follows. Section II discusses about the different
methods and techniques by which glaucoma can be detected. Section III gives a brief explanation
about the proposed system, and section IV concludes the paper.

II. TECHNIQUES INVOLVED

Glaucoma is a disease characterized by degeneration of optic nerves(optic disc). So the fall in

blood flow to the optic nerve give to the visual field defects associated with glaucoma. Drug therapy
to control the elevated intraocular pressure and serial evaluation of the optical nerves are the
principal method of curing the disease. Standard methods of evaluation of the optic nerve using
ophthalmology or stereo photography or evaluation of visual fields.
There has been importance in rising more objective, reproducible techniques to document
optic nerve damage as well as to perceive early on changes in the optic nerve and retinal nerve fiber
layer (RNFL) before the development of eternal visual field deficits. Particularly, evaluating
changes in the thickness of the retinal nerve fiber layer have been investigated as a method to make a
diagnosis and observe glaucoma. In addition, there has been importance in measuring ocular blood
flow as a diagnostic and a supervision tool for glaucoma.
This section briefly describes some of the techniques that are used for the detection of
glaucoma.

1.1. Confocal Scanning Laser Ophthalmology

In [1] Alexandrescus Dascular AM introduced a Confocal scanning laser
ophthalmoscopy(CSLO), a laser based image gaining which is proposed to improve the quality of the
examination compared to ordinary ophthalmologic examination. A laser id scanned crossways the
retina along with a detector system. Once a single spot on the retina is illuminated at any time,
ensuing in a high-contrast image of great reproducibility that can be used to estimate the width of the
RNFL. In addition, this technique does not need maximal mydriasis, which may be a problem in
patients having glaucoma. The Heidelberg Retinal Tomography is possibly the most common
example of this technology.

1.2. Scanning Laser Polarimetry

In [2] Ferreri F,Aragona P introduced the SLP. RNFL is birefringent, which causes a change
in the state of divergence of a laser beam as it passes. It uses a 780-nm diode to illuminate optic
nerve. The polarization state of the light emerging from the eye is then evaluated and linked with
RNFL thickness. Unlike CSLO, scanning laser polarimetry (SLP) can unswervingly measure the
thickness of the RNFL. GDx® is an ordinary example of a scanning laser polarimeter. GDx®
contain a normative database and statistical software package to permit comparison to age-matched
normal subjects of the same racial origin. The advantages of this system are that images can be
obtained without pupil dilation, and evaluation can be done roughly in 10 minutes. Modern

96
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 2, February (2014), pp. 95-100, © IAEME

instruments have added improved and erratic corneal compensation technology to account for
corneal polarization.

1.3. Optical Coherence Tomography

In[3] Optical coherence tomography (OCT) uses near-infrared light to provide direct cross-
sectional measurement of the RNFL. The principles employed are alike to those used in B-mode
ultrasound except light, not sound, is used to create the 2-dimensional images. The light source can
be directed into the eye through a conservative slit-lamp biomicroscope and focused on the retina
through a distinctive 78-diopter lens. This system requires dilation of the patient’s pupil. OCT® is an
example of this technology.

1.4. mfREG
J. M. Miquel-Jimenez[4] et al proposed the Glaucoma detection by wavelet-based analysis of
the global flash multifocal electroretinogram. Existing clinical analysis of the multifocal
electroretinography (mfERG) recording for detecting glaucoma is based on standard signal
morphology, that measures the amplitudes and latencies. This analysis is not sensitive enough for
detection of minute changes in the multifocal electroretinogram signals. Here an another method for
the detection of open angle glaucoma based on the categorization of global flash mfERG signals is
given. The digital signal processing technique is based on wavelets for the detection of advanced-
stage glaucoma. Two markers were obtained from the recorded signals by applying the discrete
wavelet transform, which help discriminate healthy from glaucomatous signals.

1.5. Pulsatile Ocular Blood Flow

The pulsatile disparity in ocular pressure is due to the flow of blood into the eye during
cardiac systole. Pulsatile ocular blood flow can detect by the permanent monitoring of intraocular
pressure. The detected pressure pulse can be changed into a volume measurement using the
identified relationship between ocular pressure and ocular volume. Pulsatile blood flow is principally
determined by the choroidal vessels, mainly applicable to patients with glaucoma, since the optic
nerve is supplied in large part by choroidal circulation.
The optical coherence tomography and multifocal electro retinograph (mfERG) are well-
known methods working in order to analyze functional abnormality of the eye especially glaucoma.

1.6. Computer Based Diagnosis of Glaucoma using Digital Fundus Images.

Archana Nandibewoor[5] introduced the Computer Based Diagnosis of Glaucoma using
Digital Fundus Images. Objective of this paper is to map the person’s eye color image with database
having images of normal person as well as images of people with glaucoma. Images having different
color variation inside the eye is compared by using images taken by high definition laser camera.
These are called as fundus images. MATLAB software tool is used to extract features from these
fundus images. We can know whether a person is suffering from glaucoma by measuring the color
pixels at the affected areas. To identify whether the person is suffering from Glaucoma, a test is
made using the image of a normal person which is kept as reference (say zero) and then compared
with the clinical observations of the person’s image.
Simonthomas.S, Thulasi.N[6] presented a method for glaucoma detection using an Haralick
Texture Features from digital fundus images. These features were chosen because they can be made
invariant to translations and rotations, because they describe more spontaneous aspects of the images
(e.g. coarse versus smooth, directionality of the pattern, image complexity, etc.) using statistics of the
gray-level co-occurrence matrix. For each image. K Nearest Neighbors (KNN) classifiers are used to
perform supervised classification. Haralick Texture Features has Database and classification parts, in
Database the image has been loaded and Gray Level Co-occurrence Matrix (GLCM) and thirteen

97
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 2, February (2014), pp. 95-100, © IAEME

haralick features are combined to extract the image features, performs better than the other classifiers
and correctly identifies the glaucoma images.
CDSS[7] are used in ophthalmology to generate a decision support system that identify
disease pathology in human eyes. In CDSS two types of features are extracted from the images,
structural and texture features. The disk area, rim area, cup to disc ratio and topographical features
are the extracted structural features. Glaucoma can be automatically diagnosed by calculating the cup
to disc ratio. The CDR (Cup-to-Disc Ratio) is defined as the ratio of the vertical cup height to the
vertical disc height. A CDR value that is greater than 0.65 indicates high glaucoma risk. The
glaucoma diagnosis can be improved by the enhancement of optic cup to disc ratio[8].
M.Balasubramanian [9] proposed a method known as Proper orthogonal decomposition
(POD) which is a technique that uses structural features to identify glaucomatous progression. In
POD the pixel-level information is used to estimate significant changes across samples that are
location or region specific. Structural features are location or region specific, so the detection of the
disease is limited to that region only.
The measurement of texture features is roughly defined as the spatial variation of pixel
intensity (gray-scale values) across the image. Textural features are, thus, not bound to specific
locations on the image techniques, including spectral techniques, are available to determine texture
features.

1.7. Wavelet Fourier Analysis

The texture-based techniques have been proven successful but its still a challenge to generate
features that retrieve generalized structural and textural features from retinal images. To overcome
the generalization of features wavelet transforms[10] in image processing are used to extract the
texture features from images.. In WT, the image is represented in terms of the frequency of content
of local regions over a range of scales. This representation helps the analysis of image features,
which are independent in size and can often be characterized by their frequency domain properties.
The use of WFA for the categorization of neuroanatomic distraction in glaucoma has achieved
substantial success. WFA is used as an exact model to analyze and parameterize the temporal,
superior nasal, inferior, and temporal (TSNIT) shapes. Two types of wavelet transforms are used,
discrete wavelet transforms and continuous wavelet transforms in image processing. In this method,
discrete wavelet transform (DWT) using a fourth-order symlets wavelet is used to extract features
and analyze discontinuities and rapid changes contained in signals.
DWT is a multiscale analysis method where analysis can be performed on a range of scales.
Each level of the transformation provides an analysis of the source image at a different resolution,
resulting in its independent rough calculation and detailed coefficients. In the WFA, the fast Fourier
transform (FFT) is applied to the detailed coefficients. The resultant Fourier amplitudes are
combined with the normalized approximation coefficients of the DWT to create a set of features.
Sumeet Dua, U. Rajendra Acharya, Pradeep Chowriappa, S. Vinitha Sree introduced another
method in[10] to classify normal eye images and glaucoma affected eye images based on the
distribution of average texture features obtain from three prominent wavelet families. Their objective
is to evaluate and select important features for improved specificity and sensitivity of glaucomatous
image classification. Quantitatively examine the effectiveness of different wavelet filters on a set of
curated glaucomatous images by employing the standard 2-D-DWT. Use three well-known wavelet
filters, the daubechies (db3), the symlets (sym3), and the biorthogonal (bio3.3, bio3.5, and bio3.7)
filters. Then calculate the averages of the detailed horizontal and vertical coefficients and wavelet
energy signature from the detailed vertical coefficients and subject the extracted features to abundant
of feature ranking and feature selection schemes to determine the best combination of features to
maximize interclass similarity and assist in the union of classifiers, such as the support vector

98
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 2, February (2014), pp. 95-100, © IAEME

machine (SVM), sequential minimal optimization (SMO), random forest, and na¨ıve Bayes
techniques.

III. PROPOSED SYSTEM

Our objective is to classify the images of retina as normal and glaucoma affected and the
segment the affected part from the glaucomatous images. The energy features extracted from the
images using the three different wavelet families are given to the artificial neural network for
classification. Apply any clustering models to the affected images to segment the detected part. The
Modules includes are the following:

i. Discrete Wavelet Transforms

ii. Energy feature Extraction
iii. Artificial neural network training and classification
iv. Segmentation

Overview
• The system is designed for automatic retina glaucomatous image classification and
segmentation of the affected part.
• The image classification module includes the feature extraction by using three prominent
wavelet families and classification by artificial neural network.
• The DWT will be used to decompose the image into four subbands. The energy features are
extracted from the high frequency coefficients and these subbands are contained the detailed
coefficients.
• The energy features obtained from the images are given to the ANN for classification. The
trained network will classify the images into either normal or glaucomatous.
• In case of glaucomatous images a clustering model for segmentation is applied to detect the
affected parts.

IV. CONCLUSION

Glaucoma is the second leading cause of blindness in the world. Hence its detection and
diagnosis are very essential. Different manual and automatic glaucoma detection methods are
discussed here. A new method is introduced to classify the images into glaucomatous and normal by
ANN and to detect the affected parts. Overview of the proposed system also discussed.

V. REFERENCES

[1] Alexandrescus C, Dascular AM “Confocal Scanning Laser Ophthalmoscopy in Glaucoma

Diagnosis and Management”,J Med Life 2010 Jul-Sep;3(3):229-34.
[2] Ferreri F,Aragona P “Scanning LaserPolarimetry and Confocal Scanning Laser
Ophthalmology:Technical notes on their use in Glaucoma” Prog Brain Res 2008;173:125-38.
doi: 10.1016/S0079-6123(08)01109-6.
[3] K. R. Sung et al., “Imaging of the retinal nerve fiber layer with spectral domain optical
coherence tomography for glaucoma diagnosis,” Br. J. Ophthalmol., 2010.
[4] J. M. Miquel-Jimenez et al., “Glaucoma detection by wavelet-based analysis of the global
flash multifocal electroretinogram,” Med. Eng. Phys., vol. 32, pp. 617–622, 2010.

99
International Journal of Advanced Research in Engineering and Technology (IJARET), ISSN 0976 –
6480(Print), ISSN 0976 – 6499(Online) Volume 5, Issue 2, February (2014), pp. 95-100, © IAEME

[5] Archana Nandibewoor S B Kulkarni Sridevi Byahatti Ravindra Hegadi” Computer Based
Diagnosis of Glaucoma using Digital Fundus Images”, Proceedings of the World Congress
on Engineering 2013 Vol III, WCE 2013, July 3 - 5, 2013, London, U.K.
[6] Simonthomas.S, Thulasi.N Automated Diagnosis of Glaucoma using Haralick Texture
Features IOSR Journal of Computer Engineering (IOSR-JCE) e-ISSN: 2278-0661, p- ISSN:
2278-8727Volume 15, Issue 1 (Sep. - Oct. 2013), PP 12-17.
[7] S.Weiss, C.A. Kulikowski, and A. Safir, “Glaucoma consultation by computer,” Comp. Biol.
Med, vol. 8, pp. 24–40, 1978.
[8] A.Murthi and M.Madheswaran, ”Enhancement of optic cup to disc ratio detection in
glaucoma diagnosis,” International Conference on Computer Communication and
Informatics (ICCCI -2012), Jan. 10 – 12, 2012, Coimbatore, INDIA978-1-4577- 1583-
9/12/$26.00 © 2012 IEEE.
[9] M. Balasubramanian et al., “Clinical evaluation of the proper orthogonal decomposition
framework for detecting glaucomatous changes in human subjects,” Invest. Ophthalmol. Vis.
Sci., vol. 51, pp. 264–271, 2010.
[10] E. A. Essock,Y. Zheng, and P.Gunvant, “Analysis of GDx-VCC polarimetry data by wavelet-
Fourier analysis across glaucoma stages,” Invest. Ophthalmol. Vis. Sci., vol. 46,
pp. 2838–2847, Aug. 2005.
[11] Sumeet Dua, U. Rajendra Acharya, Pradeep Chowriappa, S. Vinitha Sree”, Wavelet-Based
Energy Features for Glaucomatous Image Classification” IEEE transaction on information
technology in biomedicine, vol 16, No.1, January 2012.
[12] Nataraj A.Vijapur, Dr.Rekha Mudhol and Dr.R.Shrinivasa Rao Kunte, “An Approach for
Detection of Primary Open Angle Glaucoma”, International Journal of Advanced Research in
Engineering & Technology (IJARET), Volume 4, Issue 6, 2013, pp. 185 - 194, ISSN Print:
0976-6480, ISSN Online: 0976-6499.

100