IRJET-A Review on Brain Tumor Detection using BFCFCM Algorithm

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 01 | Jan-2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 538
A REVIEW ON BRAIN TUMOR DETECTION USING BFCFCM ALGORITHM
Monika P Belekar1, Snehal S Thorat2
1MTech Student, Department of Electronics & Telecommunication, Government College of Engineering,
Amravati, India
2Assistant Professor, Department of Electronics & Telecommunication, Government College of Engineering,
Amravati, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Brain tumor is abnormal growth of cells within
the brain which may be cancerous or non-cancerous.Normally
the anatomy of the brain can be viewed by MRI scan or CT
scan. The MRI scan is more comfortable than any other scan
for diagnosis of Brain tumor. MRI does not practice any
radiation so it will not affect the human body. The current
work presents various segmentation techniques that are used
to detect brain tumor. The algorithm based on segmentation
using clustering techniques deals with the steps pre-
processing, skull masking, segmentation, feature extraction
and classification. After segmentationwhichisdonethroughc-
means clustering algorithm the Brain tumorisdetectedandits
exact location is identified. Also the patient’s stageisidentified
by this process whether it can be cured with medicines or not.
Key Words: MRI, Skull masking, SVM, ROI
1. Introduction
Brain is the most important part of central nervous system.
It has very complex structure. Brain is safely tightly
safeguarded inside skull that protectsitfromnormaldisease.
The Brain consists of white matter and gray matter.
Cerebrum, Cerebellum and the Brain stemarethethreemain
parts of Brain. Memory sensation and personalities are
affected when Brain gets damaged. Tumor is the abnormal
growth of tissues which causes damage to the functioning
cell. There are two type of tumor which is Benign and
Malignant tumor. Benign is non-cancerous and malignant is
cancerous tumor. Surgery, chemotherapy and radiation
therapy are widely used treatmentmethodsforthediagnosis
of Brain tumor. In last decades, radiologists perform the
diagnosis of Brain tumor manually on MRI images but it is
very time consuming process. With the advances of digital
image processing radiologists have a chance to improve
their performance with automatic methods like computer
aided detection (CAD) system and artificial neural network.
MRI gives high quality images and MR image can be
segmented into different tissue classessuch aswhitematter,
gray matter and cerebrospinal fluid. In order to generate or
display digital images MRI strongly depends on computer
technology. Detection of Brain tumor from MR images is a
very complex medical process. It cannot perform without
image processing technique. The segmentation and
clustering algorithm is used for the detection of brain tumor
with the study of physical and mental condition of the
person. In surgical and radiological operations it is used to
find the exact location and area of tumor.
2. Literature survey
Shweta Jain, Shubha Mishra proposed the artificial neural
network approach namely Back propagation network
(BPNs) and probabilistic neural network (PNN) to classify
brain cancer. It is used to classify the type of tumor in MRI
images of different patients with Astrocytoma type of brain
tumor.
V.P. Gladis Pushpa Rathi and Dr. S. Palani proposed a novel
method to classification of brain tumor using Linear
Discriminant Analysis which includes this steps, Image
collection, Normalization, Intensity, shape and Texture
feature extraction, feature selection and classification.Inthis
method the shape, Intensity and Texture features are
extracted and used for classification. Vital features are
selected using Linear DiscriminatAnalysis(LDA).Theresults
are compared with Principal Component Analysis (PCA)
dimension reduction techniques. The number of features
selected or features extracted by PCA and the classification
accuracy by The Support Vector Machine (SVM) is 98.87%.
then train the system by both continuous and without
continuous data to minimize the error rate as well as
increase the classification accuracy R. J. Deshmukh and R.S
Khule proposed Neuro-fuzzy systems use the combined
power of two methods: fuzzy logic and artificial neural
network (ANN) using to detect the brain tumor. The work
carried out involves processing of MRI images of brain
cancer affected patients for detection and Classification on
different types of brain tumors. A suitable Neuro Fuzzy
classifier is developed to recognize the different types of
brain tumors.
P.B.Nikam and V.D.Shinde proposed brain image
classification and detection using distance classifiermethod,
this theses presents a system for automatic classification of
healthy or affected person using Region growing
segmentation by watershed algorithm, Euclidean distance
classifier for fast computation, accompanied with pre-
processing and post processing method apply on database
consisting both normal and timorous samples of MR brain
images. This system had two main stages, first is pre-
processing of MRI images and then other post processing
operations, which includes operations like noise removal,
convert input image into gray scale image, High pass filter.
Segmentation process using Threshold segmentation; it is
the most common approach for detecting meaningful
discontinuities in gray level, second applied Morphological

operations and feature extracting process. Their work used
Watershed for segmentation and considers the gradient
magnitude of an image as a topographic surface and
Euclidean distance classifier; this classifier based on the
distance measure is direct and simple. The meanclassvalues
are used as class centers to calculate pixel-center distances
for use by the Euclidean distance rule. For major level
classification of a homogeneous area this scheme is better.
Its advantageous nature comes from the minimum time it
takes to classify Distance Measures are used to group or
cluster brightness values together. The result ensures that
the method is efficient , and satisfying for quick detection
whether person is healthy or unhealthy.
3. Proposed Method:
The proposed system hasfive modules:Pre-processing,skull
masking, segmentation, feature extraction andclassification.
Preprocessing is done by filtering, segmentation is done by
advanced fuzzy c-means algorithm, feature extraction is
done by thresholding and finally SVM (support vector
machine) classifier is used for classification.
A. Block Diagram:
Fig. Block Diagram for proposed method
Pre-processing:
The pre-processing convert the image according to the need
of next level. It performsfiltering of noise and other artifacts
in image. Image filtering is preprocessing stage used for
reducing image noise and highlighting important portions.
RGB to gray conversion and reshaping also takes place in
preprocessing.
Skull masking:
Detection of skull is used to control the boundaries of the
object. The edge information helps to find out the region of
interest (ROI) i.e. the portion of the image covering the
tumor. This work is done with the help of the calculating the
centroid in the image. Extraction of brain tissue from non-
brain tissues in MR images which is referred to as skull
stripping is an important step in many neuron imaging
studies. In this, we used automatic threshold value selector
to automatically choose threshold value.Then,mathematical
morphology operations on a binariesimageare appliedstage
by stage to achieve acceptable skull stripped brain images.
The proposed skull stripping method comprises four steps.
Initially image binarisation is completed using threshold
value and narrow connections are removed from binarised
image using morphologicalopening. Then,largestconnected
component from binarised image is selected by considering
the fact that brain is the largest connected structure inside
the head.
Segmentation:
Segmentation subdivides an image into its constituent
regions or objects and it should stop when the objects or
regions of interest in an application have been detected.
Segmentation is process of partitioning the image into
different parts having similar features. The pre-processing
stages needs to done on the image initially, and then
segmentation and feature extraction is applied for the
detection of the tumor which is the region of interest (ROI)
from the entire image. The features are intensity based, area
base, is the vital part of segmentation as the tumor must be
isolated from the brain image. For brain imagesegmentation
numerousimage processing techniqueshavebeenproposed,
for example- region growing, thresholding, classifiers and
clustering.
Feature Extraction
Features, the characteristics of the objects of interest, if
selected carefully are representative of the maximum
relevant information that the image has to offer for a
complete characterization of a lesion. Feature extraction
methodologies analyze objects and images to extract the
most prominent features that are representative of the
various classes of objects. Features are used as inputs to
classifiers that assign them to the class that they represent.
The purpose of feature extraction is to reduce the original
data by measuring certain properties, or features, that
distinguish one input pattern from another pattern. The
extracted feature should provide the characteristics of the

input type to the classifier by considering the description of
the relevant properties of the image into feature vectors. In
this proposed method we extract the following features.
 Shape Features - circularity, irregularity, Area,
Perimeter, Shape Index
 Intensity features – Mean, Variance, Standard
Variance, Median Intensity,Skewness,andKurtosis.
 Texture features –Contrast, Correlation, Entropy,
Energy, Homogeneity, cluster shade, sum of square
variance.
Accordingly, 3 kinds of features are extracted, which
describe the structure information of intensity, shape, and
texture. These features certainly have some redundancy,but
the purpose of this step is to find the potential by useful
features. In the next step the feature selection will be
performed to reduce the redundancy.Featureselectionisthe
technique of selecting a subset of relevant features for
building robust learning modelsbyremovingmostirrelevant
and redundant features from the data, feature selection
helps improve the performance of learning models by:
 Alleviating the effect of the curse of dimensionality.
 Enhancing generalization capability
Classification:
The SVM is a supervised learning method. It isagoodtoolfor
data analysis and classification. SVM classifier has a fast
learning speed even in large data. SVM is used for two or
more class classification problems. Support Vector Machine
is based on the conception of decision planes. A decision
plane is one that separates between a set of items having
dissimilar class memberships. The Classification and
detection of brain tumor was done by using the Support
Vector Machine technique. Classification is done to identify
the tumor class present in the image. The use of SVM
involves two basic steps of training and testing.
Clustering algorithm:
K-Means is the one of the unsupervised learning algorithm
for clusters. In k-means algorithm initially requiredtodefine
the number of clusters k. Then k-cluster center are chosen
randomly. The distance between the each pixel to each
cluster centers are calculated. The distance may be ofsimple
Euclidean function. Single pixel is compared to all cluster
centers using the distance formula. The pixel is moved to
particular cluster which has shortest distance among all.
Then the centroid is re-estimated. Again each pixel is
compared to all centroid. The processis continuousuntilthe
center coverage. K-means clustering algorithm clustersdata
by iteratively computing a mean intensity for each class and
segmenting the image by classifying each pixel in the class
with the closest mean.
Algorithm for K means Clustering:
Step 1: Compute the intensity distribution
Step 2: Initialize the centers with k random values
Step 3: Cluster the pixels based on distance of their
intensities from the center
Step 4: Compute the new center for each of the clusters
Step 5: Repeat the following steps until the cluster center of
the image does not change anymore
Fuzzy C-Mean clustering algorithm introduced by Bezdek is
an improvement of earlier clustering methods. It is basedon
minimizing an objective function, with respect to fuzzy
membership, and set of cluster centroid. The FCM algorithm
iteratively optimizes with the continuous update of fuzzy
membership and set of cluster centroid. The drawback of
FCM for image segmentation is the objective functionofFCM
does not take into consideration any spatial dependence
among.
Step 1: Randomly select c cluster centers.
Step 2: Calculate the fuzzy membership function and center
for each cluster
Step 3: Compute the new membership value and update
fussy membership degree
Step 4: Repeat previous Step until the membership value is
less than or equal to previous one.
4. CONCLUSIONS
There are different types of tumors available. They may be
massin the brain or malignant over the brain. Suppose if itis
a mass, then K- means algorithm is enough to extract it from
the brain cells. If there is any noise present in the MR image
it is removed before the K-means process. The noise free
image is given as input to the k-means and tumors are
extracted from the MRI image. The performance of brain
tumor segmentation is evaluated based on K-means
clustering. Segmentation is done by advanced K-means
algorithm and fuzzy c means algorithm.

REFERENCES
[1] Shweta A. Ingle, Snehal M. Gajbhiye; “A review on Brain
tumor detection technique” , International Journal of
Science and Research,2017
[2] Priyanka Shah, Manila Jeshnani ,SagarKukreja,Priyanka
Ailani; “Survey on algorithm for Brain tumor detection”
International Journal of Computer Science and
Information Technologies, Vol. 8 (1) , 2017, 56-58
[3] Amitava Haider, Chandan Giri, Amiya Haider; “Brain
Tumor Detection using Segmentation based Object
Labeling Algorithm”, in Proceedings in IEEE-
International Conference On Advances In Engineering,
Science And Management,2016
[4] Jinal A. Shah, S. R. Suralkar; “A review on Brain tumor
detection for MRI images” International Conference on
Global Trends in Engineering, Technology and
Management (ICGTETM-2016)
[5] Arvan Aulia Rachman , Zuherman Rustam ; “Cancer
Classification using Fuzzy C-Means with Feature
Selection” 2016 12th International Conference on
Mathematics, Statistics, and Their Applications(ICMSA),
Banda Aceh, Indonesia
[6] Zeynel Cebeci, Figen Yildiz ; “Comparison of K-Means
and Fuzzy C-Means Algorithms on Different
Cluster Structures” Journal of Agricultural Informatics
(ISSN 2061-862X) 2015 Vol. 6, No. 3:13-23

IRJET-A Review on Brain Tumor Detection using BFCFCM Algorithm

Recommended

More Related Content

What's hot (20)

Similar to IRJET-A Review on Brain Tumor Detection using BFCFCM Algorithm (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET-A Review on Brain Tumor Detection using BFCFCM Algorithm