This chapter consists of a review of
relevant literature drawn from several disciplines and sources. The literature
review mainly focuses on concepts theories, principles and previous researches
related to this study.
defined as the formation of a cloudy area on the lens where they occurred when
there is damaged or clumps of protein in the lens, which then limits the
passing of light through the lens into the retina resulting the person’s vision
to become poor (Foreman, et al., 2016).
Referring to (“Cataracts,” n.d.), for a normal people, they generally have a pair of normal eyes. A
normal eye allows light to penetrate through the lens and causes the retina to
be able to focus. As lens remain clear, a sharp images can be formed.
Scientifically, lens consist of water and protein. Cataract start to develop
when there are protein clumps together resulting the retina to be blocked from
receiving the light (Figure ?220.127.116.11) and
thus causing the vision to interfere. Cataract may not causes a serious problem
at early stage. However a cataract may thicken and forming more cloudiness to
the lens over period of time.
Figure ?18.104.22.168 Blockage of light by cataract
Causes of cataract
various circumstances that can lead to the existence of cataract, however among
all the circumstances, age related by far becomes the most common reason of
existence of cataract. As people getting older, their lens will be experiencing
some changes which by the time most of them tend to expose to cataract
formation (RNIB, 2014). Apart of age related factor, there are other common causes
contributing to the formation of cataract, which are:
influences such as Diabetes.
consumption of medication such as Steroids.
surgery for other eye conditions.
exposure to sunlight.
Symptoms of cataract
The growth of
cataracts are usually slow. During the evolution of cataracts, symptoms may be
vary. However, there are some common symptoms that most people experienced (RNIB, 2014).
symptoms are listed as below (National Eye Institute, 2015):
or blurry vision. (Figure ?22.214.171.124)
of colours in vision.
Headlights, lamps or sunlight may appear too bright. A halo may appear around
lights. (Figure ?126.96.36.199)
double vision or multiple images in one eye. This symptom may clear as cataract
prescription changes in your eyeglasses or contact lenses.
Figure ?188.8.131.52 Comparison of normal vision and
Figure ?184.108.40.206 Halo around the lights
Cataract awareness in
cataract in Malaysia are still at a minimal level as public are still not aware
of the threat of cataract and the importance to treat them.
A National Eye
Survey (2014) stated that lack of awareness in treating cataract has caused
about 216,000 Malaysian delaying their cataract surgery and thus resulting them
to become blind.
According to Dr
Mohamad Aziz Salowi (2016), an ophthalmologist at Selayang’s Hospital, he
stated that one of the factor that contributing to the increase of number of
cataract cases in Malaysia is due to the hesitant attitude of not taking heed
of necessity to treat their cataract.
In addition, a
Malaysian Doctors Club President, Dr Hakim Nordin (2015), also stated that
because of the effects of cataract at initial stage just causing a slight blur
to the vision, most people tend to ignore the effects and refuse to seek for
From the above
studies and surveys, the phenomena of cataract still unfamiliar among the Malaysians.
The awareness of treating cataract still at the minimal level. It is very
crucial for the researcher to identify the level of awareness of cataract among
the public because it can assist the researcher to develop an effective project
and it is hoped that this project can facilitate in improving the awareness to
Techniques being used for
several techniques applied by the medical field in diagnosing and detecting eye
diseases such as cataract. Most of these techniques capable in performing eye
diagnosing. However, they may have some flaws in certain aspects that may
affect the efficiency of the diagnosed result. Listed in Table ?2.3.1 below
are common techniques used in detecting cataract.
Table ?2.3.1 Techniques used in detection of
Used to determine
the smallest letters can be read on a standardized chart, Snellen chart or a
card held at various distances.
used to widen the pupil allowing the eye care professionals to observe more
of the lens and retina. Observe of the lens and retina with the aid of
intraocular pressure (IOP), pressure inside of the eyes are measured during
the tonometry test. This test used in detecting eye diseases such as glaucoma
which causes blindness because of damages of the nerve at the back of the
eye. Tonometry measures the pressure by recording the resistance of the
is performed when there is suspicion of posterior globe pathology but
visualisation of the back of the eye is obscured by the opaque lens.
The CT scan
is a form of x-ray. While a regular x-ray usually involves taking one or
several pictures from a certain angle, the CT scanner moves around the body
to take several images. The patient lies on a table, and may have some dye
injected to increase the visibility of abnormal tissue.
processing is a technique using computer algorithms for detecting
abnormalities in the eye automatically. This technique requires a clear set
of images of the eye for the extraction and analysing of the features of the
images. The abnormal area of the eyes are detected after images undergo a few
processes and analysis.
Even though several techniques have
been used specifically in diagnosing and detecting cataract, nevertheless,
there may consist several flaws in certain aspects which may affecting the
efficiency and reliability of the diagnosed result to be concerned. This
underlines the need of developing a reliable cataract detection system that is
more efficient in detecting cataract.
processing is the study of any algorithm that takes an image as input and
returns an image as output. This technique has been widely used in various
fields such as face recognition in forensics, medical image analysing and
forecasting. Most of the fields converting from using analog imaging into using
of digital imaging systems which are more affordable and editable. Image
processing basically processing two dimensional (2-D) image using computer. The
outcome can be in an image or a result of the extracted features and
characteristic of the image (Dewangan, 2016).
processing, there are various phases of process of analysis undergo by the
image before delivering the result. Figure ?220.127.116.11
shows the block diagram of digital image processing.
Figure ?18.104.22.168 Digital Image Processing System
is the description of the general function of different block stages:
The initial step of the image processing. Under this process, a
digital format is given to an image. Generally, a pre-processing such as
scaling has already involved during this stage of image acquisition.
Technologies such as scanner, digital cameras or aerial cameras can be used to
input the images for the process. It is recommended to use a high quality image
with higher resolution as it can be useful in producing a better image
During image acquisition stage, a partial pre-processing operations
are required to be performed. With pre-processing techniques, an input image is
improved by eliminating the unwanted distortions and enhancing the image’s
features. When processing a high resolution images, a few requirements need to
be fulfil for a better effect of processing. One of the requirement is the
image size needed to be adjusted as processing a larger size images with high
resolution may consume longer time of process. Image is then transformed into a
grey scale image. The grey colour in the components of red, green and blue
(RGB) contain equal intensities and thus, the intensity level for each pixel is
essential to be specified with a single value.
Detection and Segmentation
In edge detection, changes and events in the properties of image
can be verified by identifying some points on the image. While, multiple
segments of the image is identified in image segmentation. In these segments
form, image becomes more meaningful and easier to analyse. A segmentation is
completed when each pixel of the image has been scanned and labelled depending
on the value of grey level than the threshold value.
The appearance of images is enhanced during the process of image
restoration. Mathematical models and probabilistic analysis of image are
applied in this techniques. An image quality can be enhanced by implementing
various types of filter that are readily available or can be designed for the
image with extracted features and characteristics is obtained.
It is essential
to have good understanding and knowledge of image processing technique which
will be useful in facilitating the researcher to develop a correct system.
Advantages of Image
Processing in Medical Field
processing has been widely applied in various fields and greatly benefit them.
As detection of cataract categorized in medical field, below listed the
advantages of image processing in the medical field:
of digital data are retained even duplicated or reproduced for numerous number
the physician researching the representative images.
displayed immediately after acquiring.
are enhanced making the images easier to be analyse and interpret by physician.
modifications over time.
of images can be used for teaching and demonstrating samples of diseases or
features in any images.
characteristics and advantages of image processing in medical field have
encourage the researcher to develop a cataract detection system using image
processing technique to facilitate the treatment of cataract.
Snake Algorithm in Image
According to (Kass, Witkin, & Terzopoulos,
algorithm is an active contour models which they were able to perform
localization accurately by locking onto nearby edges. A Snake is known as an energy-minimizing
spline where the energy is guided by the existed external constraint forces and
also influenced by the image forces that pull the energy toward the lines and
edges features in the image.
model consist of the influenced of image forces and external constraint forces
in controlling the continuity of spline. The internal spline forces imposed the
smoothness of constraint. Snake is push towards the image features such as
lines, edges or subjective contour by the image forces. Snake then is located
to the desired local minimum with the aid of external constraint forces. These
forces generally obtained from user interface, automatic attentional mechanisms
or even a high-level interpretations. Basically, snake is trying to match a
deformable model to an image by means of energy minimization.
modelled as a parametric curve where minimizing the internal energy and
locating to desired local minimum is the aims of the curve. The parametric
curve represent the position of a snake and it is parametrically present by
v(s) = (x(s), y(s)).
snake energy, Kass et al. (1988) has introduced an energy functional
There are three
terms within the snake energy.
presents as the internal energy due to bending
which the purpose of the energy is to imposed smoothness constraint on the
is the image forces that forces the snake to
be push toward image features such as lines or edges. While
represents the external constraint forces that
are responsible in localizing the snake towards the desired local minimum.
vision, active contour models describe the boundaries of shapes in an image.
Snakes in particular are designed to solve problems where the approximate shape
of the boundary is known. By being a deformable model, snakes can adapt to
differences and noise in stereo matching and motion tracking. Additionally, the
method can find illusory contour in the image by ignoring missing
classical feature extraction techniques, snakes have multiple advantages:
autonomously and adaptively search for the minimum state.
image forces act upon the snake in an intuitive manner.
Gaussian smoothing in the image energy function introduces scale sensitivity.
can be used to track dynamic objects.
The key drawbacks of the traditional
are sensitive to local minima states, which can be counteracted by simulated
features are often ignored during energy minimization over the entire contour.
accuracy depends on the convergence policy.
Related work of cataract
detection using image processing
has been carried out in developing a cataract detection system using image
has developed an automated cataract detection system with the addition of
Support Vector Machine (SVM) classifier. Image processing processed the input
image, extracting the abnormal features on the image before passing to the SVM
classifier which then used the extracted features to classify the images into
three categories: normal, cataract and post cataract. It was found that the SVM
classifier able to classified the categories of cataract accurately. The
researcher stated that the SVM classifier capable in diagnosing the
effectiveness of cataract operations using post cataract images. However, the
accuracy of the systems can be improved by increasing the size and quality of
training set, rigor of training imparted and parameters chosen to represent
In a research
done by Shashwat Pathak and Basant Kumar (2016), a texture information based
automated algorithm for detection of cataracts from digital images are
proposed. The addition of standard deviation parameter combining with the mean
intensity and uniformity features enhance the robustness to decision made by
the algorithm. The algorithm proposed capable in detecting presence of cataract
by analysing texture information from circular pupil. The researcher also
developed a GUI to add simplicity in operation of the system. However, this
system still require improvement in certain criteria to become more efficient
used a novel digital image processing technique in diagnosing cataract for the
developed system. Input images transform into greyscale images and then process
to a binary formats in order to find the relative variations in pixel
intensities between healthy and cataract eyes. The researcher improve the
system by adding a repeatability feature and a 2-D and 3-D contours generator.
It was found that image generates from the generator facilitate the
ophthalmologist and the patient to be able to visualize the cataract in
studies above, clearly indicate that a cataract detection system using image
processing can be developed in various forms. Different algorithms, methods,
even software and hardware can be implemented together with image processing
technique in order to increase the efficiency and reliability of image
processing in detection of cataract. By studying these research papers, a few
new ideas can be identified. In addition, the shortcoming in these researches also
can be improved in the future project in order to make a system that is more
efficient and reliable in detecting cataract.
provides an overview of all the information and details about the development
of the project. The study of related articles and research about cataract and
image processing is added to provide a better understanding. The understanding
of this topic is essential in order to ensure a better development of the