CLASSIFICATION OF LIVER CANCER HISTOPATHOLOGY

IMAGES USING OTSU'S METHOD

A PROJECT REPORT

Submitted by

AKSHAYA R 180301004

JEEVITHA A 180301025

SWETHA R 180301092

in partial fulfilment for the award of the degree

of

BACHELOR OF ENGINEERING

In

BIOMEDICAL ENGINEERING

RAJALAKSHMI ENGINEERING COLLEGE(AUTONOMOUS)

CHENNAI 602 105.

ANNA UNIVERSITY: CHENNAI 600 025

APRIL-2022
 BONAFIDE CERTIFICATE

It is certified that this PROJECT REPORT “CLASSIFICATION OF

LIVER CANCER HISTOPATHOLOGY IMAGES USING OSTU'S
METHOD’’ is the bonafide work of AKSHAYA R (180301005)
JEEVITHA A (180301025) SWETHA R (180301092) of B.E Biomedical
Engineering who carried out the project work under my Supervision

SIGNATURE SIGNATURE

Dr. S. RAJKUMAR Mrs. NEELA.M

Professor and Head Assistant Professor

Biomedical Engineering Biomedical Engineering

Rajalakshmi Engineering College Rajalakshmi Engineering College

Thandalam, Chennai 602105 Thandalam, Chennai 602105

Submitted for Project Report held on at Rajalakshmi

Engineering College, Thandalam.

Internal Examiner External Examiner

ii
 ACKNOWLEDGEMENT

We would like to express our immense sense of gratitude to our Head of the
Department Dr. S. Rajkumar for their valuable suggestions in improving the
project. We express our gratitude to our internal guide Mrs. M. Neela, Assistant
Professor for her support in bringing out this project successfully.

We would also like to thank our Project Coordinator Dr. M. C. Jobin Christ and
Dr. A. N. Nithya, Associate Professor for their encouragement and support
throughout the course of the project.

We express our gratitude to all who have helped us in effecting this project work.
We would like to place our sincere thanks to our Chairman Mr. S. Meganathan,
and Dr. Mrs. Thangam Meganathan our beloved chairperson. We wish to express
our intense sense of idolization and affluence to our Principal Dr. S. N.Murugeasan.
We are very grateful to all faculty and supporting staff members of Biomedical
Engineering Department, at Rajalakshmi Engineering College for their constant
help. Finally, we would like to express our gratitude to our Parents and friends who
have been very patient and supportive to us during the entire time we spent on this
project

iii
 ABSTRACT

Liver Cancer is one of the most serious disease in human body because it cannot
detect in earlier stage so the diagnosis of the liver cancer is difficult for prediction
therefore detection in earlier stage of liver cancer is the main problem. If it is
detected earlier then it can be helpful for the medical treatment, but it is a challenging
task due to the cancer cell structure. To reduce such problems, segmentation and
detection techniques for scar tissue and normal tissue in liver image is used.
Computed Tomography (CT) is highly accurate for liver cancer diagnosis. Manual
identification of tumour done by trained physicians is a time consuming task and can
be subjective depending on the skill, expertise and experience of the physician. In
this paper, the sophisticated hybrid systems are proposed which is capable to
segment liver tumour from a liver CT image and detect the scar tissue and normal
tissue in liver tumour automatically.

iv
 TABLE OF CONTENT

CHAPTER TITLE PAGE NO

NO
ABSTRACT iii
LIST OF TABLES xii
LIST OF FIGURES xiii
LIST OF ABBRIVATIONS xiv
1 INTRODUCTION 1
1.1 TYPES OF LIVER CANCER 3
1.2 SYMPTOMS OF LIVER CANCER 4
1.3 CAUSES OF LIVER CANCER 4
2 LITERATURE SURVEY 6
3 METHODOLOGY 13
3.1 EXISTING SYSTEM 13
3.1.1 DISADVANTAGES 13
3.2 PROPOSED SYSTEM 13
3.2.1 ADVANTAGES 14
3.3 BLOCK DIAGRAM 15
3.4 MONITORING UNIT 16
4 MODULE DESCRPITION 17
4.1 IMAGE ACQUISITION 17
4.1.1 2D IMAGE INPUT 17

v
 4.2 GRAY IMAGE 17

4.2.1 GRAYSCALE AS SINGLE 18

CHANNELS OF MULTICHANNEL
COLOR IMAGESH
4.3 GABOR FILTER 19
4.4 GRAY LEVEL CO-OCCURCENCE 20

MATRIX(GLCM)
4.4.1 CORRELATION 20
4.4.2 CONTRAST 21
4.4.3 ENERGY 21
4.4.4 HOMOGENITY 22
4.5 ADAPTIVE HOSTOGRAM 23

EQUILIZATION
4.6 OTSU’S THRESHOLD 24
4.7 WATERSHED ALGORITHM 24
5 SYSTEM IMPLEMENTATION 26
5.1 DATA ANALYSIS AND 26

VISUALIZATION
5.2 ACQUIRING DATA 26
5.3 ANALYZING DATA 26
5.4 VISUALIZING DATA 27
5.5 PROGRAMMING AND ALGORITHM 27
DEVELOPMENT

vi
 5.5.1 THE MATLAB LANGUAGE 27
5.6 DEVELOPMENT TOOLS 28
5.6.1 COMMAND WINDOW 28
5.6.2 MATLAB EDITOR 28
5.6.3 CODE ANALYZER 29
5.6.4 MATLAB PROFILER 29
6 SOFTWARE DESCRIPTION 30
6.1 MATLAB 30
6.2 MATLAB’S POWER OF 30

COMPUTATIONAL MATHEMATICS
6.3 FEATURES OF MATLAB 31
6.4 USE OF MATLAB 31
6.5 DIGITAL IMAGE PROCESSING 32
6.5.1 WORKING 32
6.6 SIGNAL PROCESSING 33
6.7 ANALOG IMAGE PROCESSING 33
6.8 DIGITAL IMAGE PROCESSING 34
6.9 IMAGE 34
6.10 HUMAN VOICE 34
7 OUTPUT 35
7.1 OUTPUT AND FEATURE VALUES 39
8 CONCLUSION 39
REFERENCE 40

vii
 LIST OF TABLES

TABLE NO TITLE NAME PAGE NO

1.1 Stages and feature extraction 38
values of the CT images

viii
 LIST OF FIGURES

FIGURE NO TITLE NAME PAGE NO

1.1 Structure of Liver 2
3.1 Proposed System unit – 8
MATLAB
4.2 Conversion RGB to Gray 19

4.4 GLCM Output 23

6.5 Workflow of DIP 32

6.11.1 Machine Vision System 36
6.11.2 Computer Graphics System 36

6.12 Analog Signals 38

ix
 LIST OF ABBREVIATIONS

WHO World Health Organization

HCC Hepatic Cell Carcinoma

CT Computed Tomography

CAD Computer Aided Diagnosis

SRADF Speckle Reducing Anisotropic

Reducing Filter

MSE Mean Square Error

PLGA Poly (D,L-lactide–co– glycolide

AHE Adaptive histogram equalization

CLAHE Contrast Limited Adaptive Histogram

Equalization

FFT Fast Fourier Transforms

OOP Object-Oriented Programming

MATLAB Matrix Laboratory

GLCM Gray Level Co-occurrence Matrix

x
 CHAPTER-1

1.1 INTRODUCTION

Liver cancer is one of the most frequent diseases in the world, and it is the
leading cause of cancer-related death says WHO.
Hepatic cell carcinoma (HCC), which accounts for roughly 80% of all
primary liver malignancies and most individuals with chronic liver disease,
claimed the lives of 745,000 people globally in 2012.Early detection of HCC
can considerably enhance the cure rate for patients. The great spatial resolution
and fast scanning speed of computed tomography (CT) play an important role in
the detection and diagnosis of liver cancer. Surgical resection, interventional
therapy, and loco regional ablation are some of the most common therapeutic
options. In order to build a fine treatment programme, these treatment methods
require detailed information about tumours, such as their size, shape, and
location, prior to therapy. Carmelo Militello was the associate editor who
coordinated the review of this paper and approved it for publication in regular
clinical procedures. Radiologists with sufficient expertise and experience can
manually diagnose liver cancer. However, this is a time-consuming operation
that requires the radiologist to sift through hundreds of slices and many lesions
in a 3D CT scan. Automatic liver tumour segmentation, on the other hand, is a
tough task due to a variety of image acquisition techniques, contrast agents, and
contrast enhancement levels. Furthermore, differing scanner resolutions result in
varying intensities, and many distinct types of lesions, particularly tumour
subtypes, can arise in the liver. As a result, automatic tumour segmentation is
hampered by these many types of tumours with changing contrast levels (hyper-

1
/hypo-intense tumours). With the advancement of computer-aided diagnosis
(CAD) in recent decades, numerous machine-learning-based approaches for
automatic liver tumour segmentation on CT have been developed.

Figure 1.1 Structure of Liver

Courtesy: (https://images.app.goo.gl/LDXjjHCxCeXsinAy6)

2
1.1 TYPES OF LIVER CANCERS

1. Hepatocellular carcinoma (HCC): The most frequent type of liver cancer,

also known as hepatoma, accounts for around 75% of all liver malignancies.
HCC begins in the hepatocellular cells, which are the most common form of
liver cells. Hepatitis B or C infection, or cirrhosis of the liver induced by
drinking, are the most common causes of HCC.

2. Fibrolamellar HCC: is a rare form of HCC that responds to treatment

better than other types of liver cancer. Fibrolamellar cancer is treated in the
same way that HCC is.
3. Cholangiocarcinoma (bile duct cancer) develops in the tiny, tube-like bile
ducts that convey bile to the gallbladder within the liver.
Cholangiocarcinomas are responsible for 10-20% of all liver cancers. Cancer
of the bile ducts in the liver is called intrahepatic bile duct cancer. Cancer of
the bile ducts outside the liver is known as extrahepatic bile duct cancer.

4. AngiosarcomaHemangiocarcinoma, commonly known as

hepatocarcinoma, accounts for around 1% of all liver cancers.
Angiosarcomas start in the liver's blood veins and spread swiftly.They are
usually discovered when they have progressed to an advanced stage.

5. Secondary liver cancer, When primary cancer from another region of

the body spreads to the liver, it's called a liver metastasis. According to a
2016 study published in SAGE Journals, secondary liver cancers are more
prevalent than initial liver tumours. Hepatoblastoma is the most prevalent
liver cancer in children, despite its rarity. It develops in the liver's lobes.

3
 1.1.1 LIVER CANCER SYMPTOMS

Most people don't have signs and symptoms in the early stages of primary liver
cancer. When signs and symptoms do appear, they may include:

• Losing weight without trying

• Loss of appetite

• Upper abdominal pain

• Nausea and vomiting

• General weakness and fatigue

• Abdominal swelling

• Yellow discoloration of your skin and the whites of your eyes (jaundice)

• White, chalky stools

1.1.2 CAUSES OF LUNG CANCER

Hepatocellular carcinoma (primary liver cancer) develops in livers that have

been damaged by birth abnormalities, alcohol abuse, or chronic infection with
diseases such hepatitis B and C, hemochromatosis (a hereditary condition
characterised by an excess of iron in the liver), and cirrhosis. Cirrhosis, a scarring
disorder of the liver typically induced by alcohol consumption, affects more than half
of all persons diagnosed with primary liver cancer. Hepatitis B and C, as well as
hemochromatosis, can result in lifelong liver damage and failure. Obesity and fatty

4
liver disease have been associated to liver cancer. Primary liver cancer has been
linked to a number of cancer-causing compounds, including herbicides and chemicals
like vinyl chloride and arsenic. Smoking, especially when combined with alcohol
addiction, raises your risk. Aflatoxins, Cancer-causing compounds have also been
linked to a type of plant mould. Aflatoxin contamination can occur in wheat, peanuts,
rice, corn, and soybeans. These are unusual difficulties in most industrialised
countries, such as the United States. Other influences include androgen and oestrogen
hormones, as well as the dye thorotrast, which was originally used in medical tests.
Other variables that may increase your chances of getting liver cancer include:
• Your sex. Men are more likely to get hepatocellular carcinoma than
women.

• Your weight.Obesity can increase the risk for hepatocellular

carcinoma.

• Your race. In the U.S., liver cancer is most common in Asian

Americansand Pacific Islanders.
• Anabolic steroid use. Male hormones abused by athletes to
increase musclecan slightly increase liver cancer risk with long-
term use.
• History of diabetes. Studies have suggested a link between
diabetes and liver cancer. This is likely due to the link between
diabetes and fatty liverdisease.
• Inherited metabolic diseases. Diseases that disrupt the normal
metabolism of the body have been shown to increase your risk of
liver cancer.

5
 CHAPTER 2
LITERATURE SURVEY
TITLE: Detection of Liver Cancer using Image Processing Techniques

AUTHOR: Atrayee Dutta and Aditya Dubey

YEAR: 2019

Image processing is a processing technique with the help of mathematical

operations. It uses any of the form of signal processing. Here the input is an image
or video and the output is also an image or a set of image. This technique is also used
in medical applications for various detection and treatment. In this paper, it has been
used to detect cancer cell of the liver. Here otsu’s method is used for enhancing the
MRI image and watershed method is used to segment the cancer cell from the image.

TITLE: A Comparison for Liver Cancer Treatment Alternatives

AUTHOR: Musa Sani Musa, Dilber Uzun Ozsahin, Ilker Ozsahin
YEAR: 2019
The major liver cancer therapy techniques are tumour ablation, tumour
embolization, radiotherapy, surgery, chemotherapy, immunotherapy and targeted
therapy. The choice of a particular therapy technique depends on certain disease-
related factors such as the stage of the disease, it's severity, the extent of the damage,
patient health status etc. However, other factors are related to the therapy technique
and they greatly influence the choice of treatment option. This study focuses on the
therapy related factors which include survival rate, side effects, treatment duration,
cost of treatment and comfortability. The various primary liver cancer therapy
techniques were evaluated based on the above-listed factors using Fuzzy
PROMETHEE decision making theory. The ranking result showed that Tumour
6
Ablation is the best treatment alternative for primary liver cancer upon selected
criteria and weights based on desired application. The outcome of this study will
benefit both the oncologist and the patients while trying to choose from a variety of
alternatives in different scenarios.

TITLE: Evaluating the Effect of Various Speckle Reduction Filters on Ultrasound

Liver Cancer Images
AUTHOR: Mohamed Yaseen Jabarulla and Heung-No Lee

YEAR: 2018

This paper describes the evaluation and performance analysis of five image filtering
techniques, namely Kuan, Frost, Mean, Median and Speckle reducing anisotropic
diffusion filter (SRAD) from the spatial filtering process for liver US data.
An application of US hepatic liver cancer image was chosen and selected denoising
algorithms are applied to estimate the impact on the US speckle image signal.
Experiments are investigated based on Peak Signal-to-Noise Ratio (PSNR), Mean
Structural Similarity (MSSIM) and Mean square error (MSE). The result shows that
SRAD filter performs better than other denoising filters with a PSNR =31.11 dB,
MSE=31.07, MSSIM=0.895.

TITLE: Impact of Enhancement Features on Image Registration for Liver Cancer

Interventions using CT Images
AUTHOR: LuuManhHa , Hoang Hong Son

YEAR: 2018

In this paper, we investigate how the contrast-enhanced features in computed

7
tomography (CT) images i.e. the tumours and the vessels, help the registration. For
this, we de-enhance the contrast-enhanced CT image (iDECT) of the liver acquired
during the intervention; and then we compare the accuracy of the registration
between the diagnostic contrast enhance CT image (dCECT) and the original
interventional contrast-enhanced image (iCECT) versus the dCECT image and the
de-enhanced image (iDECT). In addition, we use a rigidity term to improve the
registration using the de-enhanced image.

TITLE: Genetic Algorithm Optimization Of Convolutional Neural Network For

Liver Cancer CT Image Classification.
AUTHOR: ZiqiLi1, Huibin Ma*,1,2, Diankui Li1,2, Rui Fan

YEAR: 2018

In order to classify the liver CT images and make the initial weight of the
convolution neural network to the optimal state. We propose a method of combine
the genetic algorithm and convolution neural network to classify the liver CT tumour
images which can be divided these liver CT images into two parts, have cancer and
no cancer. This method used the global optimization and survival of fittest features
of genetic algorithms and generated initial weight for convolution neural network by
the operation of selection and crossover and mutation. After improvement, learning
performance is better than traditional convolution neural network. According to
simulation, the method of combining the genetic algorithm and convolution neural
network has higher classification accuracy than traditional convolution neural
network and support vector machine. It can help medical-aided diagnosis better.

8
TITLE: Prediction of liver cancer using Conditional probability Bayes theorem

AUTHOR: Mr.N.Ramkumar, Dr.S.Prakash

YEAR: 2017
The liver cancer which begins in the liver apart from moving from other part
of the body is called as a primary liver cancer. Cancer which spreads all other part
of the body and finally it reaches liver is called as secondary liver cancer. Liver is
one of the important part of the human. WHO surveys say out of 100,000 people,
around 30 people are suffered from liver cancer and mostly it affects the African and
Asian countries earlier. Nowadays it became a popular disease The most common
kind of a liver cancer is called as hepatocellular carcinoma, this particular affects
male rather than female. The liver cancer occurs mainly due to the more alcohol
consumption. Many data mining algorithms, Artificial intelligence concepts are used
to predict the liver cancer. The probability of predicting the liver cancer is performed
using the Bayes theory with the WEKA tool.

TITLE: Design of Ultrawideband Applicator for Microwave Ablation Aimed at

Thermal Therapy in Liver Cancer

AUTHOR: Faisal Narpati and Basaril

YEAR: 2017

In this paper, we design an applicator, which has the characteristics of UWB

by modifying the ground plane of the applicator by a hexagonal-sectioned coil. The
applicator is designed on a FR4 substrate and numerically simulated with liver
phantom by using CST Microwave Studio. The proposed applicator is able to have
wide bandwidth by 8.1 GHz, with lower frequency (fl) 1.84 GHz and higher
9
frequency (fh) 10 GHz
TITLE:A High Throughput Cantilever Array Sensor for Multiple Liver Cancer
Biomarkers Detection.
AUTHOR: Jingjing Wang, Yinfang Zhu, Xing Wang, Shuaipeng Wang, Jinling
Yang and Fuhua Yang
YEAR: 2016

A novel microcantilever biosensor was batch-fabricated with integrated

circuit (IC) compatible micro-electro-mechanical system (MEMS) technology for
joint detection of liver cancer biomarkers with high sensitivity, high throughput,
high specification, and good precision. A micro-cavity was designed in the free end
of the cantilever for local antibody-immobilization using micro printing system,
which can dramatically reduce the effect of adsorption-induced stiffness coefficient
k variation. A linear relationship between the resonance frequency shift and the
antigen concentration was observed for three liver cancer biomarkers, alpha-
fetoprotein
(AFP), γ-glutamyl transpeptidase II (GGT-2), and hepatocyte growth factor (HGF).
In addition, the presented immune sensing method has little cross-reactivity to
different antigen, paving the
way to a highly specific technique. These approaches will promote clinical
application of the cantilever sensors in early cancer diagnosis

TITLE: A Pragmatic approach for Detecting Liver Cancer using Image Processing
and Data Mining Techniques.
AUTHOR: Anisha P R, Kishor Kumar Reddy C

YEAR: 2015
Among the various cancers, liver cancer stands in the third position. Liver
10
cancer is generally diagnosed by three different test like blood test, image test and
biopsy. To make the task of detecting the liver cancer simpler, less time consuming,
an effective and efficient approach is adopted for the same. In this research a
computer aided diagnostic system for detecting liver cancer is put forward. The
proposed detection methodology makes use of MRI, CT and USG scan imagery. K
means clustering technique is adopted so as to segment the images in order to capture
the region of interest. Later, Haar wavelet transform is considered to compute the
threshold values for the region of interest. The experiment put forth gave an average
accuracy of 82% besides reducing the time complexity and computational
complexity of the test.

TITLE: Nanomechanics of Fosbretabulin A4 Polymeric Nanoparticles in Liver

Cancer Cells
AUTHOR: Radhika Poojari, Rohit Srivastava, Dulal Panda

YEAR: 2015

The purpose of this study was to develop a chemotherapeutic drug

Fosbretabulin A4 (F tub A4) encapsulated in pegylated polymeric poly (D,L-lactide–
co– glycolide) (PLGA) nanoparticles (100-200 nm) for microtubule targeted
delivery to human liver cancer cells. The effect of cell particle molecular interactions
was also evaluated. Physicochemical properties of these nanoparticles were
characterized using dynamic light scattering, zeta-potential, transmission electron
microscopy and scanning electron microscopy. The cellular internalization of
nanoparticles was observed by confocal laser scanning microscopy and transmission
electron microscopy. Potent antiproliferative and apoptotic activities of F tub A4
nanoparticle treatment in comparison to the free drug were observed in liver cancer
cells. In conclusion, F tub A4 delivery modality greatly sensitized liver cancer cells
11
as well as perturbed the microtubule assembly leading to enhanced antitumour
effects. The results indicated that the nano formulation has strong therapeutic
potential inmetastatic liver malignancies.

12
 CHAPTER 3
METHODOLOGY

3.1 EXISTING SYSTEM

The cancer will be predicted using the present method by using the Region of
Interest. Artificial intelligence research has progressed in recent years, allowing for
the creation of cancer-removal technologies in a variety of medical fields.
Although there have been advancements in the development of support equipment
for evaluating liver cancer, nothing has been done to reduce evaluator error and data
management.
3.1.1 DISADVANTAGES

➢ Liver cancer can’t be predicted exactly.

➢ Tones are predicted as cancer.

3.2 PROPOSED SYSTEM

This research aims to be the first link in using digital image processing
techniques to improve the evaluation of liver cancer. It was feasible to apply an
algorithm to a series of trial photos that yielded positive findings for wound and
location recognition thanks to the use of advanced object segmentation techniques
and a parameter that changes the system's sensitivity until the required results were
obtained. Re-dimensioning, space colour conversion, Gabor filter, and clustering
delimitation are all part of this stage. The contribution is a liver evaluation system that
can be a beneficial tool for doctors because it allows for the automatic detection of
cancer.
13
3.2.1 ADVANTAGES

➢ It is a non-invasive method and it allows an easy handling of the obtained

results.
➢ Liver cancer is predicted.

3.3.BLOCK DIAGRAM - MATLAB UNIT:

Image enhancement
Image preprocessing
Image Acquistion (histogram
(gray image)
equalisation)

Feature Extraction Otsu’s threshold Segmentation (Edge

(GLCM method) method detection)

Watershed
Result
method

Fig 4.1.1 Proposed System unit – MATLAB

14
3.4 MONITORING UNIT

Fig 4.2.1 Receiver unit

15
 CHAPTER 4
MODULE DESCRIPTION
4.1IMAGE ACQUISITION:

Digital imaging or digital image acquisition is the creation of a digitally

encoded representation of the visual characteristics of an object, such as a physical
scene or the interior structure of an object.

4.1.1 2D Image Input

The basic two-dimensional image is a monochrome (greyscale) image which

has been digitised. Describe image as a two-dimensional light intensity function
f(x,y) where x and y are spatial coordinates and the value of f at any point (x, y) is
proportional to the brightness or grey value of the image at that point. A digitised
image is one whereas partial and greyscale values have been made discrete. Intensity
measured across a regularly spaced grid in x and y directions intensities sampled to
8 bits (256 values).

4.3GRAY IMAGE:

In digital photography, computer-generated imagery, and colorimetry, a

grayscale or greyscale image is one in which the value of each pixel is a single
sample representing only an amount of light, that is, it carries only intensity
information. Grayscale images, a kind of black-and-white or gray monochrome, are
composed exclusively of shades of gray. The contrast ranges from black at the
weakest intensity to white at the strongest.

16
 Grayscale images are distinct from one-bit bi-tonal black-and-white images
which, in the context of computer imaging, are images with only two colours: black
and white (also called bilevel or binary images). Grayscale images have many shades
of gray in between.

Grayscale images can be the result of measuring the intensity of light at each
pixel according to a particular weighted combination of frequencies (or
wavelengths), and in such cases they are monochromatic proper when only a single
frequency (in practice, a narrow band of frequencies) is captured. The frequencies
can in principle be from anywhere in the electromagnetic spectrum (e.g. infrared,
visible light, ultraviolet, etc.).

A colorimetric (or more specifically photometric) grayscale image is an image

that has a defined grayscale colour space, which maps the stored numeric sample
values to the achromatic channel of a standard colour space, which itself is based on
measured properties of human vision.

If the original colour image has no defined colour space, or if the grayscale
image is not intended to have the same human-perceived achromatic intensity as the
colour image, then there is no unique mapping from such a colour image to a
grayscale image.

4.3.1 Grayscale As Single Channels Of Multichannel Colour Images

Colour images are often built of several stacked colour channels, each of them
representing value levels of the given channel. For example, RGB images are
composed of three independent channels for red, green and blue primary colour
components; CMYK images have four channels for cyan, magenta, yellow and black
ink plates, etc.

17
 Here is an example of colour channel splitting of a full RGB colour image.
The column at left shows the isolated colour channels in natural colours, while at
right there are their grayscale equivalences:

Fig 4.3.1.1 Conversion RGB to Gray

Courtesy: https://images.app.goo.gl/6WRZSowkS2QbNJGj7

The reverse is also possible: to build a full colour image from their separate
grayscale channels. By mangling channels, using offsets, rotating and other
manipulations, artistic effects can be achieved instead of accurately reproducing the
original image.

18
4.4 GABOR FILTER

In image processing, a Gabor filter, named after Dennis Gabor, is a linear filter
used for texture analysis, which means that it basically analyzes whether there
areany specific frequency content in the image in specific directions in a
localizedregion around the point or region of analysis.

4.5Gray Level Co-Occurrence Matrix (GLCM):

Feature extraction involves simplifying the amount of resources required to

describe a large set of data accurately. When performing analysis of complex data
one of the major problems stems from the number of variables involved. Analysis
with many variables generally requires a large amount of memory and computation
power or a classification algorithm which over fits the training sample and
generalizes poorly to new samples. Feature extraction is a general term for
methods of constructing combinations of the variables to get around these problems
while still describing the data with sufficient accuracy. Texture tactile or visual
characteristic of a surface. Texture analysis aims in finding a unique way of
representing the underlying characteristics of textures and represent them in some
simpler but unique form, so they can be used for robust, accurate classification and
segmentation of objects. Though texture plays a significant role in image analysis
and pattern recognition, only a few architectures implement onboard textural
feature extraction. In this paper, Gray level cooccurrence matrix is formulated to
obtain statistical texture features. Several texture features may be extracted from the
GLCM. Only four second order features namely angular second moment,
correlation, inverse difference moment, and entropy are computed. These four
measures provide high accuracy required for motion picture estimation.

19
4.5.1 Correlation:

It passes the calculation of the correlation of a pixel and its neighbour over
the whole image means it figures out the linear dependency of gray levels on those
of neighbouring pixels. On behalf a perfectly positively or negative correlated
image, the correlation value is 1 and-1.On behalf of constant image its value is
NaN..Range=[-1,1] and the formula is

4.5.2Contrast:

It passes the calculation of the correlation of a pixel and its neighbour over
the whole image means it figures out the linear dependency of gray levels on those
of neighbouring pixels. On behalf a perfectly positively or negative correlated
image, the correlation value is 1 and-1. On behalf of constant image its value is
NaN..Range=[-1,1] and the formula is

4.5.3Energy:

Since energy is used for doing work, Thus order lines. It makes use for the
texture that calculates orders in an image. It gives the sum of square elements in
GLCM. It is fully different from entropy. When the window is proficient orderly,
energy value is high. The square root of ASM(Angular Second Moment) texture
20
character is used as Energy. Its range is [0 1]. Since constant image its value is
1.Theequation of energy is

4.5.4Homogenity:

In short term it is going by the name of HOM. It passes the value that
calculates the tightness of distribution of the elements in the GLCM to the GLCM
diagonal. For diagonal GLCM its value is 1 and its range is [0,1]. Opposite of
contrast weight is homogeneity weight values, withweight decreases exponentially
loose from thediagonal. The weight employed in contrast is (i-j)^2and in
homogeneity ,it is 1/1+(i-j)^2.The equation is

where:

Pij - Element i,j of the normalized symmetrical GLCM

N - Number of gray levels in the image as specified by Number of levels in under

Quantization on the GLCM texture page of the Variable Properties dialog box.
C - The Correlation feature
sgn(x) - Sign of a real number
21
 4.5.4.1 x = -1 for x < 0
4.5.4.2 x = 0 for x = 0
4.5.4.3 x = 1 for x > 0
σ 2 - The variance of the intensities of all reference pixels in the relationships

that contributed to the GLCM, calculated as:

Note: This may approximate, but is not I dentical to, the variance of the
intensities of all the pixels in the data window W (as defined by the GLCM
algorithm), and it is dependent upon the choice of spatial relationship in that
algorithm.

μ -the GLCM mean (being an estimate of the intensity of all pixels in the
relationships that contributed to the GLCM), calculated as:

Note: This also approximates, but is not identical to, the mean of all the pixels in the
data window W (as defined by the GLCM algorithm), and it is dependent upon the
choice of spatial relationship in that algorithm.

22
 Fig 5.4.1 GLCM Output
4.6ADAPTIVE HISTOGRAM EQUALIZATION:

Adaptive histogram equalization (AHE) is a computer image processing

technique used to improve contrast in images. It differs from ordinary histogram
equalization in the respect that the adaptive method computes several histograms,
each corresponding to a distinct section of the image, and uses them to redistribute
the lightness values of the image. It is therefore suitable for improving the local
contrast and enhancing the definitions of edges in each region of an image.

However, AHE has a tendency to over amplify noise in relatively

homogeneous regions of an image. A variant of adaptive histogram equalization
called contrast limited adaptive histogram equalization (CLAHE) prevents this by
limiting the amplification.

4.6 OTSU THERSHOLD

Thresholding, we convert an image from color or grayscale into a binary

image, i.e., one that is simply black and white. Most frequently, we use
thresholding as a way to select areas of interest of an image, while ignoring

23
 the parts we are not concerned with Otsu method is a clustering based image
thresholding. It works when the histogram is bimodal. The method basically
tries to minimize the within class variance and at the same time it maximizes
the between class variance.
Total variance=Within class variance +Between Class Variance.

It is used to perform automatic image thresholding. In the simplest form, the

algorithm returns a single intensity threshold that separate pixels into two
classes, foreground and background.

4.7 WATERSHED ALGORITHM

The watershed transform finds "catchment basins" or "watershed ridge lines" in an

image by treating it as a surface where light pixels represent high elevations and dark
pixels represent low elevations. The watershed transform can be used to segment
contiguous regions of interest into distinct objects.
The watershed is a classical algorithm used for segmentation, that is, for separating
different objects in an image. Here a marker image is built from the region of low
gradient inside the image. In a gradient image, the areas of high values provide
barriers that help to segment the image.

In watershed segmentation an image is regarded as a topographic landscape with

ridges and valleys. The elevation values of the landscape are typically defined by the
gray values of the respective pixels or their gradient magnitude. Based on such a 3D
representation the watershed transform decomposes an image into catchment
basins. For each local minimum, a catchment basin comprises all points whose path
of steepest descent terminates at this minimum. Watersheds separate basins from
each other. The watershed transform decomposes an image completely and thus

24
assigns each pixel either to a region or a watershed. With noisy medical image data,
a large number of small regions arises. This is known as the “over-segmentation”.

Marker-based Watershed Often however, no examined flooding level is sufficient to

segment target structures. Therefore, the user may specify image locations that
belong to the target structure (include points), or that do not belong the target
structure (exclude points). If the user specifies an include point and an exclude point,
an additional watershed is constructed at the maximum level between them. The
merge tree is traversed such that each region contains either include points or exclude
points but not both. This interaction style is called marker-based watershed
segmentation. There are many variants of the watershed transform. For example,
merging may consider also gradient information or other criteria for homogeneity.
A frequently used variant is to merge regions where the difference of the mean gray
value is below a threshold. This process can be carried out iteratively and results also
in a hierarchical merging tree.

25
 CHAPTER 5

SYSTEM IMPLEMENTATION

5.1Data Analysis and Visualization

MATLAB provides tools to acquire, analyze, and visualize data, enabling you
to gain insight into your data in a fraction of the time it would take using spreadsheets
or traditional programming languages. You can also document and share your results
through plots and reports or as published MATLAB code

5.2Acquiring Data

MATLAB lets you access data from files, other applications, databases, and
external devices. You can read data from popular file formats such as Microsoft
Excel; text or binary files; image, sound, and video files; and scientific files such as
net CDF and HDF. File I/O functions let you work with data files in any format.

Using MATLAB with add-on products, you can acquire data from hardware devices,
such as your computer’s serial port or sound card, as well as stream live, measured
data directly into MATLAB for analysis and visualization. You can also
communicate with instruments such as oscilloscopes, function generators, and signal
analyzers.

5.3Analyzing Data

MATLAB lets you manage, filter, and preprocess your data. You can perform
exploratory data analysis to uncover trends, test assumptions, and build descriptive

26
models. MATLAB provides functions for filtering and smoothing, interpolation,
convolution, and fast Fourier transforms (FFTs). Add-on products
providecapabilities for curve and surface fitting, multivariate statistics, spectral
analysis,image analysis, system identification, and other analysis tasks.

5.4Visualizing Data

MATLAB provides built-in 2-D and 3-D plotting functions, as well as volume
visualization functions. You can use these functions to visualize and understand data
and communicate results. Plots can be customized either interactively or
programmatically. The MATLAB plot gallery provides examples of many ways to
display data graphically in MATLAB. For each example, you can view and
download source code to use in your MATLAB application.

5.5Programming and Algorithm Development

MATLAB provides a high-level language and development tools that let you
quickly develop and analyze algorithms and applications.

5.5.1 The MATLAB Language

The MATLAB language provides native support for the vector and matrix
operations that are fundamental to solving engineering and scientific problems,
enabling fast development and execution. With the MATLAB language, you can
write programs and develop algorithms faster than with traditional languages
because you do not need to perform low-level administrative tasks such as declaring
variables, specifying data types, and allocating memory. In many cases, the support
for vector and matrix operations eliminates the need for for-loops. As a result, one
line of MATLAB code can often replace several lines of C or C++ code. MATLAB
27
provides features of traditional programming languages, including f low control,
error handling, and object-oriented programming (OOP). You can use fundamental
data types or advanced data structures, or you can define custom data types. You
canproduce immediate results by interactively executing commands one at a time.
This approach lets you quickly explore multiple options and iterate to an optimal
solution. You can capture interactive steps as scripts and functions to reuse and
automate your work. MATLAB add-on products provide built-in algorithms for
signal processing and communications, image and video processing, control
systems, and many other domains. By combining these algorithms with your own,
you can build complex programs and applications.

28
 5.6Development Tools

MATLAB includes a variety of tools for efficient algorithm development,

including:

5.6.1 Command Window –Lets you interactively enter data, execute commands
and programs, and display results
5.6.2 MATLAB Editor–Provides editing and debugging features, such as setting
break points and stepping through individual lines of code. Code Analyzer–
Automatically checks code for problems and recommendsmodifications to
maximize performance and maintainability

5.6.3 MATLAB Profiler–Measures performance of MATLAB programs and

identifies areas of code to modify for improvement

29
 CHAPTER 6
SOFTWARE DESCRIPTION

6.1 MATLAB:

MATLAB is a programming language developed by MathWorks. It started

out as a matrix programming language where linear algebra programming was
simple. It can be run both under interactive sessions and as a batch job. MATLAB
(matrix laboratory) is a fourth generation high-level programming language and
interactive environment for numerical computation, visualization and
programming. MATLAB is developed by Math Works. It allows matrix
manipulations; plotting of functions and data; implementation of algorithms;
creation of user interfaces; interfacing with programs written in other languages,
including C, C++, Java, and FORTRAN; analyze data; develop algorithms; and
create models and applications. It has numerous built-in commands and math
functions that help you in mathematical calculations, generating plots and
performing numerical methods.

6.2MATLAB'S POWER OF COMPUTATIONAL MATHEMATICS:

MATLAB is used in every facet of computational mathematics. Following are some

commonly used mathematical calculations where it is used most commonly:

• Dealing with Matrices and Arrays

• Plotting and graphics

30
 • Linear Algebra

• Algebraic Equations

• Non-linear Functions

• Statistics

• Data Analysis

• Calculus and Differential Equations

• Numerical Calculations

• Integration

• Transforms

• Curve Fitting

6.3FEATURES OF MATLAB:

It is a high-level language for numerical computation, visualization and

application development.It also provides an interactive environment for iterative
exploration, design and problem solving.It provides vast library of mathematical
functions for linear algebra, statistics, Fourier analysis, filtering, optimization,
numerical integration and solving ordinary differential equations.It provides built-
in graphics for visualizing data and tools for creating custom plots.MATLAB's
programming interface gives development tools for improving
codequality,maintainabilityand maximizing performance.It provides tools for
building applications with custom graphical interfaces.It provides functions for
integrating MATLAB based algorithms with externalapplications and languages
such as C, Java, .NET and Microsoft Excel.
31
6.4USES OF MATLAB:

MATLAB is widely used as a computational tool in science and

engineering encompassing the fields of physics, chemistry, math and all engineering
streams.
It is used in a range of applications including:

6.4.1 signal processing and Communications

6.4.2 image and video Processing

6.4.3 control systems

6.4.4 test and measurement

6.4.5 computational finance

6.4.6 computational biology

6.5DIGITAL IMAGE PROCESSING:

Digital image processing deals with manipulation of digital images through a

digital computer. DIP focuses on developing a computer system that is able to
perform processing on an image. The input of that system is a digital image and the
system process that image using efficient algorithms, and gives an image as an
output. The most common example is Adobe Photoshop. It is one of the widely used
application for processing digital images.

32
6.5.1 WORKING:

Figure 6.5.1 WORKFLOW OF DIP

Courtesy: https://images.app.goo.gl/pAxVqMtb2r4f2DHLA

In the above figure, an image has been captured by a camera and has been
sent to a digital system to remove all the other details, and just focus on the water
drop by zooming it in such a way that the quality of the image remains the same.

6.6SIGNAL PROCESSING:

Signal processing is a discipline in electrical engineering and in mathematics

that deals with analysis and processing of analog and digital signals , and deals with
storing , filtering , and other operations on signals. These signals include
transmission signals, sound or voice signals , image signals , and other signals e.t.c.

Out of all these signals, the field that deals with the type of signals for which
the input is an image and the output is also an image is done in image processing.
As it name suggests, it deals with the processing on images. It can be further
dividedinto analog image processing and digital image processing.

33
6.7 ANALOG IMAGE PROCESSING:

Analog image processing is done on analog signals. It includes processing on

two dimensional analog signals. In this type of processing, the images are
manipulated by electrical means by varying the electrical signal. The common
example include is the television image.

Digital image processing has dominated over analog image processing with
the passage of time due its wider range of applications.
6.8DIGITAL IMAGE PROCESSING

The digital image processing deals with developing a digital system that
performs operations on an digital image.

6.9IMAGE:

An image is nothing more than a two dimensional signal. It is defined by the

mathematical function f(x,y) where x and y are the two co-ordinates horizontally and
vertically.

The value of f(x,y) at any point is gives the pixel value at that point of an
image.The above figure is an example of digital image that you are now viewing on
your computer screen. But actually , this image is nothing but a two dimensional
array of numbers ranging between 0 and 255.

34
6.10 HUMAN VOICE

Human voice is an example of analog signals. When you speak, the voice that is
produced travel through air in the form of pressure waves and thus belongs to a
mathematical function, having independent variables of space and time and a value
corresponding to air pressure.

Another example is of sin wave which is shown in the figure below.

Y = sin(x) where x is independent

35
 CHAPTER-7
RESULT AND DISSCUSSION

7.1 OUTPUT

7.1.1 IMAGE ACQUISITION

Fig .7.1.1:Image Acquisition

In Fig 7.1.1 The image is acquired by clicking the image acquisition button in
the image.

36
 7.1.2 IMAGE PREPROCESSING

Fig .7.1.2:Image Preprocessing

In fig 7.1.2 the acquired image undergoes preprocessing step where original
image is converted into gray scale and it involves removal of noise

7.1.3 HISTOGRAM EQUILIZATION

Fig 7.1.3 Histogram Equilization

In Fig 7.1.3 the contrast of the image is enhanced using histogram
equalization method
37
7.1.4 ADAPTIVE EQUILIZATION

Fig 7.1.4 Adaptive Equalization

7.1.5 EDGE DETECTION

Fig 7.1.5 Edge Detection

In Fig 7.1.5 The edges will be detected using Sobel edge detection method

38
 7.1.6 OTSU’S THRESHOLD

Fig 7.1.6 Otsu’s Threshold

In Fig 7.1.6 The otsu’s algorithm is performed in which the algorithm returns
the single intensity threshold that separates pixels into 2 classes foreground and
background

7.1.7 WATERSHED THRESHOLD

Fig 7.1.7 Watershed Method

39
 In fig 7.1.7 after extracting sure background and foreground images from
otsu’s method this watershed algorithm uses markers which will make
watershed run and detect the exact boundaries.This algorithm helps in detecting
touching and overlapping objects in the image

7.2 OUTPUT IMAGES WITH FEATURE VALUES:

CT IMAGES STAGES FEATURE

EXTRACTION
VALUES
Initial Stage 0.90

Mid Stage 0.91

End Stage 0.95

Table 1.1 stages and feature extraction values of the CT images

40
 CHAPTER-8
CONCLUSION AND FUTURE SCOPE

Different CT Images were acquired from the internet, and the images
undergo basic Otsu preprocessing technique for threshold verification to classify the
images as foreground and background. Watershed Segmentation is used for
segmenting the image. After processing the image, it provides information about
different stages of cancer like early, medium and mature stage to provide perfect
indication to doctor , so that they could give a treatment according the stage.. It
was observed that for a few images, segmentation was done correctly. So the future
works includes creating a GLCM feature extraction and giving the stage of the
cancer with accuracy. Increasing the accuracy of the images provide a clear
explanation to the doctors and technicians. It also helps to diagnose the cancer
quickly and provide timely treatment to the patients.

41
 REFERENCE

1.N. P. Asimakis, I. S. Karanasiou, P. K. Gkonis and N. K. Uzunoglu, "Theoretical

analysis of a passive acoustic brain monitoring system", Progress In
Electromagnetics Research B, vol. 23, pp. 165-180, 2010.
2.C. M. Chaturvedi, V. P. Singh, P. Singh, P. Basu, M. Singaravel, R. K. Shukla, et
al., "2.45GHz (CW) microwave irradiation alters circadian organization spatial
memory DNA structure in the brain cells and blood cell counts of male mice mus
musculus", Progress In Electromagnetics Research B, vol. 29, pp. 23-42, 2011.

3.M. Emin Tagluk, M. Akin and N. Sezgin, "Classification of sleep apnea by using
wavelet transform and artificial neural networks", Expert Systems with Applications,
vol. 37, no. 2, pp. 1600-1607, 2010.

4.L. Golestanirad, A. P. Izquierdo, S. J. Graham, J. R. Mosig and C. Pollo, "Effect

of realistic modeling of deep brain stimulation on the prediction of volume of
activated tissue", Progress In Electromagnetics Research, vol. 126, pp. 1-16, 2012.

5.S. A. Mohsin, "Concentration of the specific absorption rate around deep brain
stimulation electrodes during MRI", Progress In Electromagnetics Research, vol.
121, pp. 469-484, 2011.

6.S. A. Mohsin, N. M. Sheikh and U. Saeed, "MRI induced heating of deep brain
stimulation leads: Effect of the air-tissue interface", Progress In Electromagnetics
Research, vol. 83, pp. 81-91, 2008.

7.A. Oikonomou, I. S. Karanasiou and N. K. Uzunoglu, "Phasedarray near field

radiometry for brain intracranial applications", Progress In Electromagnetics
Research, vol. 109, pp. 345-360, 2010.

42
8.R. Scapaticci, L. Di Donato, I. Catapano and L. Crocco, "A feasibility study on
microwave imaging for brain stroke monitoring", Progress In Electromagnetics
Research B, vol. 40, pp. 305-324, 2012.

9. Zhang, L. Wu and S. Wang, "Magnetic resonance brain image classification by

an improved artificial bee colony algorithm", Progress In Electromagnetics
Research, vol. 116, pp. 65-79, 2011.

43