International Journal of Recent Technology and Engineering (IJRTE)

ISSN: 2277-3878, Volume-1, Issue-3, August 2012

Analyzing the Severity of the Diabetic

Retinopathy and Its Preventive Measures by
Maintaining Database Using Gui In Matlab
Anil Kumar Neelapala, Mehar Niranjan Pakki
Abstract-Diabetic-related eye disease is a major cause of
blindness in the world. It is a complication of diabetes which can
also affect various parts of the body. When the small blood vessels
have a high level of glucose in the retina, the vision will be blurred
and can cause blindness eventually, which is known as diabetic
retinopathy. Regular screening is essential to detect the early stages
of diabetic retinopathy for timely treatment and to avoid further
deterioration of vision. This project aims to detect the presence of
abnormalities in the retina such as the structure of blood vessels,
micro aneurysms and exudates using image processing techniques by
automating the detection of Diabetic retinopathy (DR). This Process
is achieved by the fundus images using morphological processing
techniques to extract features such as blood vessels, micro
aneurysms and exudates and then we calculate the area of each
extracted feature. Depending on the area of each feature we classify
the severity of the disease.
Then finally by knowing the severity of the disease corresponding
treatment measures can be analyzed. In addition to this, well
established database have been developed regarding the disease
analysis of patients which is implemented using GUI in MATLAB.
It will surely help to reduce the risk and increase efficiency for
ophthalmologists.
Keywords: Diabetic Retinopathy, Exudates, Fundus Camera,
Micro-aneurysms, Morphological Operations, Segmentation.

I. INTRODUCTION
Diabetic retinopathy (DR) is one of the complications
resulted from prolonged diabetic condition usually after ten to
fifteen years of having diabetes. In the case of DR, the high
glucose level or hyperglycemia causes damage to the tiny
blood vessels inside the retina. These tiny blood vessels will
leak blood and fluid on the retina, forming features such as
micro aneurysms, hemorrhages, hard exudates, cotton wool
spots, or venous loops U.R.Acharya (2007) et al [1].
DR affects about 60% of patients having diabetes for 15
years or more and a percentage of these are at risk of
developing blindness dicussed in [2]. Despite these
intimidating statistics, research indicates that at least 90% of
these new cases could be reduced if there was proper and
vigilant treatment and monitoring of the eyes [3]. Laser

photocoagulation is an example of surgical method that can

reduce the risk of blindness in people who have proliferative
retinopathy [4]. However, it is of vital importance for diabetic
patients to have regular eye checkups. Current examination
methods use to detect and grade retinopathy include
ophthalmoscope (indirect and direct) James L. Kinyoun et al
[5], photography (fundus images) and fluorescein
angiography. The objective of this project is to implement an
automated detection of diabetic retinopathy (DR) using digital
fundus images. By using MATLAB to extract and detect the
features such as blood vessels, micro-aneurysms and exudates
which will determine classifications: normal or abnormal (DR)
eye. An early detection of diabetic retinopathy enables
medication or laser therapy to be performed to prevent or delay
visual loss.
II. ALGORITHM DESIGN
The automated disease identification system is not a single
process. This system consists of various modules the success
rate of each and every step is highly important to ensure the
overall high accurate outputs. The rest of the work is organized
as follows (a) Read the input image from Fundus camera, (b)
Image Pre-Processing, (c) Anatomical Structure Extraction, (d)
Feature Extraction, (e) Disease Severity, (f) Corresponding
Treatment and (g) Maintaining Database and GUI. All these
techniques are explained in the following sections.

Manuscript received on August, 2012

Neelapala Anil Kumar, Electronics & Communication Engineering,
Vignans Institute of Information Technology, Visakhapatnam,India.
Mehar Niranjan Pakki, Electronics & Communication Engineering,
Vignans Institute of Information Technology, Visakhapatnam,India.

131

Analyzing the Severity of the Diabetic Retinopathy and Its Preventive Measures by Maintaining Database Using Gui In Matlab
the pre-processed. The second category is the indirect method
in which various anatomical structures are initially segmented
from the preprocessed images and then features are extracted
from these anatomical structures. These anatomical structures
include macula, border formation and the optical disk.
We extract the anatomical structures like optical disk and
macula using morphological opening function Daniel Welfer
(2010), Cemal Kose (2008) et al [8, 9]. And using two border
detection techniques for effective extraction, we extract the
image border by applying morphological dilate and erode then
subtracting dilate with erode image.
V. FEATURE EXTRACTION
The third step is to extract the features to detect the required
features such as exudates, micro-aneurysms and blood vessels.
In this process we use morphological operations such as dilate,
erode, opening and closing Dietrich Paulus (2005), Ana Maria
Mendonca (2006) et al [10, 11]. After applying these
operations we convert image into binary image then using
logical operations AND and OR, filters like colfilt and
segmentation we segment the exudates, micro-aneurysms and
blood vessels Jagadish Nayak, Akara Sopharak (2008) et al
[12, 13].
After extracting the features, by using loops we will
calculate the area of each feature to know the severity of the
disease.
VI. DISEASE SEVERITY
Fig.1 Algorithm processing steps
III. IMAGE PRE-PROCESSING
The preliminary step in automated retinal pathology
diagnosis is image preprocessing. This includes various
techniques such as contrast enhancement, gray/green
component, image de-noising etc. Initially we convert the RGB
image into gray color image/green channel image Salvatelli A.
(2007) et al [6] to further process the image. Then Image
enhancement techniques are designed to improve the quality of
an image as perceived by human being. Adaptive Histogram
equalization is a constant enhancement technique which
provides a sophisticated method for modifying the dynamic
range and contrast of an image by altering that image such that
its intensity histogram has a desired shape Andrea Anzalone
(2008) et al [7]. The median filter is a non linear digital filter
technique often used to remove noise from images or other
signals. We use bwareaopen also to remove the small area
of pixels considered to be noise after applying morphological
operations.
IV. ANATOMICAL STRUCTURE EXTRACTION
The second step is the feature extraction technique in which
suitable feature set is extracted from the enhanced retinal
images. The feature extraction techniques for retinal images
are broadly divided into two classes. The first category is the
direct method in which the textural features are extracted from

The next step is to know the severity of the disease which

will be calculated depending on the area calculated from the
above step. Depending on the severity we will categorize the
disease into three categories such as Mild, Moderate and
Severe by using ANOVA test [14].
And in the final stage depending on the severity of the
disease we will give the corresponding treatment to the patient.
The treatments for the diabetic retinopathy are Vitrectomy,
Scatter laser treatment, Focal laser treatment and Laser
photocoagulation, treatments depending upon severity are
gathered from Visakhapatnam government eye hospital.
VII. MAINTAINING DATABASE
After doing all the Necessary operations the disease severity
and corresponding treatment, of all patients and along with
patient details are Saved and maintained in the database
[15,16,17]. The following table shows the patients details, in
that we maintain patients name, and for uniqueness we used
patients phone number. Then as an input we give the patients
input image for processing, after processing we get patients
corresponding segmented images of blood vessels,
micro-aneurysms and exudates. Then in the processing of
image we calculate area of all the extracted features also. Then
by depending on that area we should notify the severity of the
disease and corresponding treatment details to the patient. All
the patients disease severity and treatment details for 10
patients are shown in the following table 1.

132

International Journal of Recent Technology and Engineering (IJRTE)

ISSN: 2277-3878, Volume-1, Issue-3, August 2012
Table I: The table shows all the patient details, patients disease severity and corresponding treatment.

S.No

Patient
Name

Patient
Phone
Number

1.

Ganesh

2.

Patient
Input
Image

Blood
Vessels

Disease
Severity

Corresponding
Treatment

7396069927

Mild

Laser PRP
Treatment

Srinu

8885384318

Moderate

Vitrectomy

3.

Mahesh

9700370745

Severe

Laser
Photocoagulation

4.

Lalitha

9985536700

Moderate

Vitrectomy

5.

Surya

9908026375

Severe

Laser
Photocoagulation

6.

Rahim

9908026375

Mild

Laser PRP
Treatment

7.

John

8483864467

Severe

Laser
Photocoagulation

8.

Williams

7726798340

Moderate

Vitrectomy

9.

Rohit

9848693260

Severe

Laser
Photocoagulation

10.

Chan Lee

9497921875

Moderate

Vitrectomy

Exudate

133

Micro-aneur
ysms

Study of Wind Power Generation Using Slip Ring Induction Generator

VIII. IMPLEMENTING IN GUI

All the details are implemented in GUI in MATLAB[18]
for all patients. While implementing in GUI first we create a
Main GUI which consists of some important push buttons
like Add Patient, Patient Database, Search Patient, Delete
Patient, Clean Database, Main Page and Exit. When we click
on any button then corresponding GUI will be called from
Main GUI except Exit push button. If we click on Exit Push
button the dialogue box open whether to exit or not. The
Main GUI is shown in the following figure 2.

Fig.3 GUI for entering patient details and analyzing disease

severity

Fig.2 GUI for Main Page to add patient details, analyzing the
disease severity, searching, deleting and cleaning patient
details
When we click on the Add Patient push button in the main
page GUI then immediately Add Patient GUI will be called.
In this Add patient GUI we two main push buttons one for
entering patients details like patients name, patients phone
number and read the patients image. Then another push
button to analyze the disease, extraction of the features,
calculating area, to know disease severity and to indicate
corresponding treatment. All these information will be
displayed in the text fields. And patients input image,
extracted features like exudates, blood vessels and
micro-aneurysms are shown in figures. Along with the above
said push buttons we have three additional buttons to return
to the Main GUI, to clean the text fields and axis and Exit
button. After we enter and analyze one patient, we can add
another patient and like that we can add and analyze any
number of patients. Then if we want to return to the Main
GUI we can click on the Main page push button then
immediately this GUI calls Main Page GUI. Otherwise we
can directly exit from this GUI by clicking on Exit push
button. This GUI is shown in the following figure3.

Then after entering patients details, analyzing disease and

return to Main page GUI we have another button i.e., Patients
Database if we click on this button then immediately the
Main page GUI calls the Patients Database GUI. In this GUI
we have three main push buttons to see the patients details,
Next and Previous push buttons to see all the patients
database. When we click on the Patient Database push button
we get the first entered patients details like patient name,
patient phone number, disease severity and corresponding
treatment. And if we click on the Next push button we get the
details of next patient like that if we click on Previous push
button we get previous patient details. As like before GUI we
have another three push buttons to return to main page, to exit
from this GUI and to clean the text fields, all of them works
as in previous GUI which are explained in the above fig.3.
This GUI is shown in the following figure 4.

Fig.4 Patient Database which contains all the patients details

After we see all the patients details and return to the main
page GUI we have another push button i.e., Search Patient
button. This is used to search the particular patient details
whether the patient is there or not. In this GUI we have one
important push button to search patient details i.e., Enter
Patient Details when we click on this button a dialogue box

134

International Journal of Recent Technology and Engineering (IJRTE)

ISSN: 2277-3878, Volume-1, Issue-3, August 2012
will be open it asks for the patient name then immediately it
asks the patient enrolled phone number because there may be
the patient with same name so for uniqueness we ask for
patient phone number also. Then if both name and phone
numbers are matched with anyone of the patient who are
present in the database the those patients corresponding
disease and treatment will be dispayed on the text fields. If
both name and phone number are not matched with any
patient then a dialogue box will display NoPatient. Like
that we can search any number of patients. As like in the
above GUI we have three common push buttons to return
main page, to clean text fields and to exit. This GUI is shown
in the following figure5.

And another push button is to delete all the patients details

i.e. to clean all patients database in Main GUI. This push
button is used to remove all the patients details when we click
on this button and it asks for conformity whether to delete
permanently or not then if we click on the yes button it will
clean whole database. This GUI is shown in the following
figure7.

Fig.7 GUI to clean Database of all patients

Fig.5 GUI to search patient details

IX. CONCLUSION

In Main GUI we have another important push button i.e.,

Delete Patient button, which is used to delete the required or
unwanted patients details. When we click on the Delete
Patient push button then immediately Main GUI calls the
Delete Patient GUI, in this GUI we have Delete Patient push
button when we click on this button we get a dialogue box to
enter the patient name to delete but along with patient name
we ask for phone number also because same name patients
may exist. After checking both the name and the phone
number successfully it delete the patient if exists otherwise it
shows the dialogue box that patient not there. In this GUI we
have two additional push buttons to return to main page and
to exit from this GUI directly. This GUI is shown in the
following figure6.

In summary, a medical system for the automatic diagnosis

of the primary signs of Diabetic Retinopathy (DR) has been
developed by maintaining the database which addresses the
severity and preventive measures. The results demonstrated
with the various samples of DR eye and this algorithm
proven to be well suited in complement the screening of DR
helping the ophthalmologists in their daily practice.
X. ACKNOWLEDGMENT
I have taken efforts in this project. However, it would not
have been possible without the kind support and help of
many individuals and organizations. I would like to extend
my sincere thanks to all of them.
I would like to express my special gratitude and thanks to
T.
JYOTHIRMAYI,
M.SC
(OPTHAMOLOGIST),
Visakhapatnam government hospital.
REFERENCES
[1]

[2]
[3]
[4]
[5]

[6]

Fig.6 GUI to delete unwanted Patient details

135

U R Acharya, C M Lim, E Y K Ng, C Chee and T Tamura. Computerbased detection of diabetes retinopathy stages using digital fundus
images.
Singapore
Association
of
the
Visually
Handicapped.
http://www.savh.org.sg/info_cec_diseases.php.
What is Diabetic Retinopathy? http://www.news-medical.net/health/
What-is-Diabetic-Retinopathy.aspx.
Diabetic Retinopathy. http://www.hoptechno.com/book45.htm.
James L. Kinyoun, Donald C. Martin, Wilfred Y. Fujimoto, Donna L.
Leonetti. Opthalmoscopy Versus Fundus Photographs for Detecting
and Grading Diabetic Retinopathy.
Salvatelli A., Bizai G., Barbosa G.Drozdowicz and Delrieux (2007),
A comparative analysis of pre-processing techniques in colour retinal
images, Journal of Physics: Conference series 90.

Study of Wind Power Generation Using Slip Ring Induction Generator

[7]

Andrea Anzalone, Federico Bizzari, Mauro Parodi, Marco Storace

(2008), A modular supervised algorithm for vessel segmentation in
red-free retinal images, Computers in Biology and Medicine, Vol. 38,
pp. 913-922.
[8] Daniel Welfer, Jacob Schacanski, Cleyson M.K., Melissa M.D.P.,
Laura W.B.L., Diane Ruschel Marinho (2010), Segmentation of the
optic disc in color eye fundus images using an adaptive morphological
approach, Journal on Computers in Biology and Medicine, Vol. 40,
pp. 124-137.
[9] Cemal Kose, Ugur Sevik, Okyay Gencalioglu (2008), Automatic
segmentation of age-related macular degeneration in retinal fundus
images, Computers in Biology and Medicine,Vol.38, pp. 611-619.
[10] Dietrich Paulus and Serge Chastel and Tobias Feldmann (2005),
Vessel segmentation in retinal images, Proceedings of SPIE, Vol.
5746, No.696.
[11] Ana Maria Mendonca and Aurelio Campilho (2006), Segmentation of
Retinal Blood Vessels by Combining the Detection of centerlines and
Morphological Reconstruction, IEEE Transaction on Medical
Imaging, Vol. 25, No. 9, pp. 1200-1213.
[12] Jagadish Nayak, Subbanna Bhat (2008), Automated identification of
diabetic retinopathy stages using digital fundus images, Journal of
medical systems, Vol.32, pp. 107-115.
[13] Akara Sopharak, Bunyarit Uyyanonvara, Sarah Barman, Thomas
H.Williamson (2008), Automatic detection of diabetic retinopathy
exudates from non-dilated retinal images using mathematical
morphology methods, Computerized Medical Imaging and Graphics,
Vol. 32, pp. 720-727.
[14] ANOVA test for severity of disease. http://afni.nimh.nih.gov/pub/dist/
HOWTO/howto/ht05_group/html/background_ANOVA.shtml
[15] Save function in matlab.
http://www.mathworks.in/help/techdoc/ref/save.html
[16] Load function in matlab.
http://www.mathworks.in/help/techdoc/ref/load.html
[17] strvcat function in matlab.
http://www.mathworks.in/help/techdoc/ref/strvcat.html
[18] Graphical User Interfaces (GUIs) Matlab, http://www.mathworks.com

Neelapala. Anil Kumar has Obtained B. Tech. in ECE

Department from JNT University, Hyderabad and ME in
Electronic Instrumentation (EI) from Andhra University,
Visakhapatnam. He has Seven years of teaching
experience, presently working at Vignan's Institute of
Information Technology, Visakhapatnam, as Assistant
Professor in Department of ECE. His Areas of interests are bio medical
instrumentation and image processing.

Mehar Niranjan Pakki is pursuing his M. Tech degree

in the Department of Electronics & Communications,
Vignan's institute of Information and Technology,
Duvvada. His Areas of interests are bio-medical and
image processing.

136