Diagnosis System using Data Mining approach for

Glaucoma (A Social Threat).

Mrs. Kinjan Chauhan Dr Prashant Chauhan Dr. Ravi Gulati
Shree Ramkrishna Institute of Ophthalmologist Department of Computer Science,
Computer Education Arvind Eye Hospital, Veer Narmad South Gujarat
and Applied Sciences, Surat, Gujarat,India. University,
Surat, Gujarat, India. drprashant614@rediffmail.com Surat, Gujarat, India.
kinjanchauhan99@gmail.com rmgulati@gmail.com

Abstract: Glaucoma is the second leading cause of blindness after

cataracts and its detection is essential to prevent visual damage.
Glaucoma is the third leading cause of blindness in India. The
II. GLAUCOMA DIAGNOSIS
framework designed, will be able to help in early detection and
diagnosis of glaucoma using perimetry data and Optical Glaucoma is the third leading cause of blindness in India.
Coherence Tomography (OCT) images. The co-relation between
perimetry and OCT data will be evaluated and the resultant data
Glaucoma diagnosis is greatly researched topic. Various
will be used as an input for the diagnostic system. The system will Methods such as Image processing, Artificial Neural Network,
be able to identify the normal and glaucomatous eye, it will also Neuro–Fuzzy etc have been applied for successful diagnosis of
be able to predict the progression of glaucoma. Thus, assisting Glaucoma. Images obtained from OCT and fundus have been
ophthalmologist in earlier detection of glaucoma which will help used widely for developing diagnosis [11][12].
the society in turn for fighting the silent killer of vision.

Keywords: Glaucoma, Socio-economic effect, OCT image,

Perimetry, Data mining techniques.

I. INTRODUCTION
Glaucoma is the name for a number of conditions that
damage the optic nerve, usually as a result of increased
pressure within the eye that results when the naturally
occurring fluid (aqueous humor) in the eye does not drain
properly out of the eye. Slow drainage may occur with normal
eye anatomy (open-angle glaucoma) or with structural
problems in the drainage mechanism (angle-closure glaucoma).
The optic nerve damage of glaucoma usually affects your
peripheral vision first, leading to tunnel vision and moves
progressively to involve your central vision.
Glaucoma is the second leading cause of blindness after
cataracts and its detection is essential to prevent visual damage.
Glaucoma causes 12.8 per cent of the total blindness in the
country. In India, more than 90 per cent of glaucoma in the Fig -1 Glaucoma – Figures in India
community is undiagnosed. In India, it is estimated that Source : http://www.glaucomaindia.com/
glaucoma affects 12 million people and by 2020, this is
expected to be 16 million. Statistics say one in eight persons
above the age of 40 years in India is either suffering from To facilitate early detection and help preserve vision in
glaucoma or is at risk of the disease [1]. glaucoma individuals at risk -
In India, glaucoma is present in 2.5% to 5% of people older • age over 40 years of age
than 40 years, adding that angle-closure glaucoma is
significantly more common in India than predominantly • glaucoma in the family
Caucasian areas.[16]
• those with diabetes/ thyroid disease / hypertension
• have received steroid containing preparations :
copyright 978-1-4673-2070-2/12/$31.00 ©2012 IEEE tablets/ drops/ ointments/ puffs/ injections
 • see rainbow coloured rings around bright light
• have a rapid change of glasses
• take medication for sleep/ anxiety / depression /
asthma/ parkinsonism
• have had an injury to the face /eye myopia.
loss. The diagnostic performance of both OCT and SAP in
glaucoma as well as the correlation between SAP and OCT
measurements has been investigated [2].
The proposed system deals with the Image obtained from
Stratus OCT and report generated from Humphrey Standard
Automated Perimetry.

• Have headaches
III. SOCIO-ECONOMIC EFFECT OF GLAUCOMA
Healthcare resources in India, though ample, are
inadequate. There were just 12,000 ophthalmologists, i.e., 1 per
100,000, with very few glaucoma specialists in 2001. [16, 17]
Most ophthalmologists in India (70%) are located in urban
areas and cater to only 23% of its population.
Many of the newer diagnostic modalities for early
diagnosis and monitoring progression of glaucoma are
available in the country. The spectrum of antiglaucoma
medications is readily available. Indian manufacturers have
made a number of drugs available at affordable prices. Several
public and private hospitals are equipped to provide state-of-
the-art care to the patients and high quality training to Fig. 2 Proposed Framework
residents. Yet, more than 90% of the glaucoma remains
undiagnosed contrary to 40–60% in developed countries. [16]
The system is divided into 4 stages
A large percentage of blindness in our country stems from
the population living in the rural areas where medical facilities 1] Image Pre-processing
are not easily available. Glaucoma is an asymptomatic disease
with no appreciable benefit to the patient with therapy. The 2] Diagnosis of the Glaucomatous eye from the image
side effects of medicines may lead to a decreased quality of 3] Comparison of the Generated Graphical Perimetric
life. Most of the patients are not motivated due to lifelong Graph, Actual Perimetric Graph and OCT Graph
treatment. Nearly 35% of the Indian population falls below the
international poverty line. So, in spite of the availability of 4] Data mining Process.
medications at lower prices, they are still not affordable
considering the lifelong need. This makes glaucoma a cost- A. Image Pre -Processings
intensive disease, with a low socioeconomic status having a Image of the patient’s eye obtained from OCT test will be
negative impact. The added expense of prolonged, regular anti- pre processed for the removal of noise.
glaucoma medications is a major hurdle. [19]
Image pre processing will be carried out in 2 steps:
Our System will try to minimize the negativity of
socioeconomic status. • Pre processing
• Image based feature extraction
IV. PROPOSED FRAMEWORK
Pre-processing:- This process will eliminate disease
independent variations from the input image. The Process
Perimetry and OCT are the widely performed tests due to is aimed at removal of vessels and illumination corrections.
their high sensitivity and specificity. OCT is an imaging Image based Feature Extraction: - This step will transform
technique based on interferometry that provides three- the pre processed data to characteristic and compact
dimensional imaging of the retina at a very low cost. representation of the data.
Polarization Sensitive Optical Coherence Tomography (PS-
OCT) provides additional imaging modalities. Clinically, these Existing computer aided analysis of retina images are based
high-resolution images can be used for better diagnosis and on segmentation which is mostly done manually or by semi-
monitoring of disease. automated methods. Segmentation based techniques have one
major drawback like small errors in segmentation may lead to
Similarly, Standard Automated white-on-white Perimetry significant change in the measurements and thus the diagnosis.
(SAP) is the standard for examining the visual field. Perimetric [14]
tests are able to provide quantitative measurements of
differential light sensitivity at many test point locations in the B. Diagnosis of the Glaucomatous eye image
visual field, and commercially available statistical analysis
packages help clinicians in identifying significant visual field Classification of the glaucomatous eye will depend on the
features selected and then the decision rules will be applied.
The rules will help classify the image into glaucomatous and
normal eye image. Data mining Classifier such as Support
Vector Machine (SVM), Machine Learning Classifiers (MLC)
will be tested during this phase for classifying the eyes as
NORMAL or GLAUCOMATOUS eye.
C. Comparison of Generated Graphical Perimetric Graph
,Actual Perimetric Graph and OCT Graph
Co-relation between the RNFL thickness and Actual
Perimetric Data will need to be interpreted. The Glaucomatous
eye image will be used to predict proposed Graphical
Perimetry report.
D. Data Mining Process
Data mining process is an interactive and iterative process.
Pre-processing of the data is a very important stage in any data
mining process. The field of data mining addresses the question
of how best to use this data to discover new knowledge and
improve the process of decision making.
Mining problems can be grouped in three categories:
identifying classifications, finding sequential patterns, and
discovering associations. Although data mining is a powerful
knowledge discovery technique, there are constraints in the
way it can be applied: it is application dependent, different
applications usually require different mining techniques, and
data must be of a certain size and format. Most fundamental
mining algorithms (rule-based learning systems, neural
networks, decision trees, Bayesian networks, logistic
regressions, and so on), which have been used with great
success in medicine.
It is important, therefore, to consider how to pre process
images so that we can transform them to data representations
which are amenable to data mining techniques. Although there
are algorithms for classifying images, there is a lack of
effective algorithms for learning from images directly.
Machine learning classifiers such as Linear discriminant
analysis (LDA), Support Vector Machine (SVM) and recursive
partitioning and regression tree (RPART) have been used
extensively for diagnosis of glaucoma,[7][12]
Trained classifiers can be used to predict the diagnosis of
new cases.
Thus, Proposed Graphical perimetry data, Actual Graphical
Perimetry data and the data obtained from OCT will be tested
for appropriate data mining algorithm like SVM, Naive
Bayesian. The algorithm giving higher sensitivity and
specificity for detecting and diagnosing the progression of
glaucoma will be used in the data mining process.
III. CONCLUSION
Early detection and in time management will help the patient in
decreasing the chances of vision loss in glaucomatous patient.
Also, the cost of treatment can be reduced if glaucoma is
detected in the earlier stage. And the vision expectancy will be
good if the progression of the glaucoma is prevented.
The system aims to help the ophthalmologist in diagnosing the
glaucoma, with better success rate compared to present method
of diagnosis, which can provide a detailed analysis of the
progression of glaucoma with the help of graphical
representation of the Perimetric as well as OCT data. Thus,
preventing the vision loss in the patient due to glaucoma and
benefiting the society in return.
REFERENCES
[1] http://www.dnaindia.com/health/report_glaucoma-silently-blights-light-
in-the-eyes_1519673
[2] Relationship between Visual Field Sensitivity and Retinal Nerve Fiber
Layer Thickness as Measured by Optical Coherence Tomography -
Csilla Ajtony,Zsolt Balla,Szabolcs Somoskeoy,and Balint Kovacs .
IOVS, January 2007, Vol. 48, No. 1 pgno 258- 263
[3] http://www.revoptom.com/content/d/technology/c/19985/
[4] Rule Extraction for Glaucoma Detection with Summary Data from
StratusOCT -Mei-Ling Huang,1 Hsin-Yi Chen,2 and Jian-Cheng Lin1
IOVS, January 2007, Vol. 48, No. 1pg no 244-247
[5] Integration and fusion of standard automated perimetry and optical
coherence tomography data for improved automated glaucoma
diagnostics -Dimitrios Bizios*, Anders Heijl and Boel Bengtsson Bizios
et al. BMC Ophthalmology 2011, 11:20
http://www.biomedcentral.com/1471-2415/11/20
[6] Regional Relationship Between Retinal Nerve Fiber Layer Thickness
and Corresponding Visual Field Sensitivity in Glaucomatous Eyes-
Akiyasu Kanamori, MD, PhD; Maiko Naka, MD; Azusa Nagai-
Kusuhara, MD, PhD;Yuko Yamada, MD, PhD; Makoto Nakamura, MD,
PhD; Akira Negi, MD, PhD ARCH OPHTHALMOL/VOL 126 (NO. 11),
NOV 2008 WWW.ARCHOPHTHALMOL.COM
[7] Optical Coherence Tomography Machine Learning Classifiers for
Glaucoma Detection: A Preliminary Study – Zvia Burgansky-Eliash,
Gadi Wollstein,Tianjiao Chu, Joseph D. Ramsey Investigative
Ophthalmology & Visual Science, November 2005, Vol. 46, No. 11 Pg no 4147-
4152
[8] http://www.eyeworld.org/article.php?sid=4477
[9] Factors for Glaucoma Progression and the effect of Treatment –
M.Cristine Leske ,Anders Heijl ,Mohammed Hussain ,Bo Bengtsson
,Leslie Heymann ARCH OPHTHALMOL/VOL 126 / JAN 2003
[10] . Evaluation of the structure-function relationship in glaucoma. Gardiner
SK, Johnson CA, Cioffi GA Invest Ophthalmol Vis Sci.
2005;46(10):3712-3717.
[11] An Efficient Automated System for Detection using Fundus image -
K.NARASIMHAN, K.VIJAYAREKH A Journal of Theoretical and
Applied Information Technology15th November 2011. Vol. 33 No.1
[12] An Efficient Decision Support System for Detection for Galucoma in
Findus Images Using ANFIS -S.Kavitha1, K.Duraiswamy International
Journal of Advances in Engineering & Technology, Jan 2012.©IJAET
ISSN: 2231-1963 227 Vol. 2, Issue 1, pp. 227-240
[13] Identifying Visual Field Progression with Fuzzy Logic Model -DC
Hoffman1, DE Gaasterland2 and J Caprioli , Invest Ophthalmol Vis Sci
2002;43: E-Abstract 2169.© 2002 ARVO
[14] Effects of Pre processing Eye Fundus Images on Appearance Based
Glaucoma Classification - Jorg Meier1, R udiger Bock1, Georg
Michelson2, L_aszl_o G. Ny_ul1, and Joachim Hornegger
[15] Thomas R. Glaucoma in India: Current status and the road ahead. Indian
J Ophthalmol 2011;59:3-4 [Thomas R. Glaucoma in India: Current
status and the road ahead. Indian J Ophthalmol [serial online] 2011
[cited 2012 Sep 18];59:3-4. Available from:
http://www.ijo.in/text.asp?2011/59/7/3/73678]
[16] Rao GN. Ophthalmology in India. Arch Ophthalmol. 2000;118:1431–2.
[PubMed]
[17] 17. George R, Ramesh SV, Vijaya L. Glaucoma in India: Estimated
burden of disease. J Glaucoma. 2010;19:391–97. [PubMed]
[18] Glaucoma society hopes to improve low detection rate throughout India
Ocular Surgery News India Edition, December 2007. Vijaya L, George
R, et al. Prevalence of primary angle-closure disease in an urban south
 Indian population and comparison with a rural population: The Chennai
Glaucoma Study. Ophthalmology. 2007 Sept. 13;
[19] Surgical management of glaucoma: An Indian perspective – R.
Ramakrishnan and Mona Khurana.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3038496/