An Ameliorate Technique for Brain Lumps Detection Using Fuzzy C-Means Clustering

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 01 | Jan -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 335
An Ameliorate Technique for Brain Lumps Detection Using Fuzzy C-
Means Clustering
Pritam R. Dungarwal1, Prof. Dinesh D. Patil
1Research Scholar Student, M.E (CSE), SSGBCOET,Bhusawal,Maharashtra,India
2HOD, Professor, Dept. of CSE, SSGBCOET, Bhusawal,Maharashtra,India
---------------------------------------------------------------------***--------------------------------------------------------------------
ABSTRACT-Clustering is commonly used in biomedical
applications particularly for brain lumps detection in
Abnormal magnetic resonance images (MRI). In terms of
segmentation efficiency Fuzzy clustering using fuzzy local
information C-means algorithm proved to be greater over the
other clustering methodologies. Most Research in developed
countries show that the number of people who have brain
cancer were died due to the fact of inaccurate detection and
not in time. Generally, CT scan or MRI that is directed into
intracranial cavity produces a complete image of brain. This
image is visually examined by the physician for detection &
diagnosis of brain cancer. However this method of detection
resists the accurate determination of size of lump. In addition,
it also reduces the time for analysis. At the end of the process
the lump is extracted from the MR image and itsexactposition
and the shape also determined. The graphbasedonpixel value
is drawn taking the various points from the swelledcellsliesin
the original position from the affected region. Here the
affected region is considered as ellipsoid shape and the
volumes have been calculated from it. A fuzzy level set
algorithm is proposed in this paper to facilitatemedical image
segmentation & on this performance of evaluation of the
proposed algorithm was carried.
Index Terms— introduction, clustering, Fuzzy clustering,
image segmentation, Fuzzy C-Mean, result
1 INTRODUCTION
Now a day’s brain abnormality mainly brain lumps are
one of the most common brain diseases, so detection
and quantification of lumps in MRI are important in
medical diagnosis. Through past many researchers have
prepared important research in the field of brain
abnormality segmentation but still now it is very
important research fields due to the large number of
variation of MRI of brain. The accurate segmentation
of internal structures of the brain is of great interest
for the study and very helpful for the treatment of lumps. It
aims at reducing the mortality and improving the surgical or
radio therapeutic management of lumps. The most
important aim of medical image analysis in general and
brain MRI analysis in particular, is to extract clinical
information that would improve diagnosis and treatment of
disease. The aim is to provide information associated to
anatomical structures as well as potential abnormal
tissues necessary to treatment planning and patient follow-
up. There are different brain lumps detection and
segmentation methods to detect and segment a brain
lumps from MRI images. The measurement of brain
lumps volume can assist lumps staging for brain lumps
volume measurements is developed which overcome
the problem of inter-operator variance, besides partial
volume effects and shows satisfactory performance
for segmentation, thus segmentation is very important for
volume calculation. Different segmentation techniques
such as thresholding based segmentation methodology,
Region Growing based segmentation, K-nearest
neighbours (KNN), Bayesian approach, Markov Random
Field Models , Expectation maximization (EM),
Support vector machine (SVM), Fuzzy c-means
algorithms, K-means algorithms, Morphology-based
segmentation, Atlas-guided based segmentation,
Knowledge based segmentation, Texture-based
segmentation, Artificial neural networks (ANNs), Fusion-
based, Fuzzy connectedness, Watershed Methods, Level
set based segmentation, Hybrid Self Organizing Map
(SOM), SOM, Graph Cut based segmentation, Fractal-
based segmentation, Parametric deformable models
(snakes), Boundary based methods, Geometric
deformable model, The Combination of Watershed and
Level Set segmentation, Spatio-Temporal Model, Hidden
Markov Model, Genetic algorithms based segmentation,
Kohon-en Self Organizing Map (SOM) with a common
phase pre-processing and segmentation and its steps are
shown below. Thus accurate segmentation over full field
of view is another very much problem but during the
segmentation procedure verification of results is
another source of difficulty. Statistical classification may
not allow differentiation between non-enhancinglumps and
normal tissue due to overlapping intensity
distributions of healthy tissue with lumps and surrounding
edema. Manually segmenting brain lumps from MR imaging
is generally time consuming and difficult.
Fig: 1 Segmentation process
Brain MRI Images
Pre-processing
Detection Phase
Post-Processing Segmentation
Volume Calculation
Diagnosis

An automate segmentation method is desirable because it
reduces the load on the operator and generates satisfactory
results.
The region growing segmentation is used to segment the
brain lumps due to its wide range of applications and
automatic features. After taking the image of the lumps
brain there is a need to process it. The image clearly shows
the place of the lumps portion of the brain. The image
does not give the information about the numerical
parameters such as area and volume of the lumps portion
of the brain. After segmentation the desired lumps area is
selected from the segmented image. This selected region
is used to calculate the area and volumeofthelumpspresent
in the MR image.
2. LITERATURE SURVEY
The previous methods for brain lumps segmentation are
thresholding, region growing & clustering. Thresholding is
the simplest method of image segmentation. From a
grayscale image, thresholding can be used to create binary
images. During the thresholding process, individual
pixels in an image are marked as "object" pixels if
their value is greater than some threshold value
(assuming an object to be brighter than the background)
and as "background" pixels otherwise. This convention is
known as threshold above. Variants include threshold
below, which is opposite of threshold above; threshold
inside, where a pixel is labelled "object" if its value is
between two thresholds; and threshold outside, which is
the opposite of threshold inside. Typically, an object pixel
is given a value of “1” while a background pixel is given
a value of “0.” Finally, a binary image is created by
colouring each pixel white or black, depending on a
pixel's labels.
The major drawback to threshold-based approaches is that
they often lack the sensitivity and specificity needed for
accurate classification. Brain cancer cells have high
proteinaceous fluid which has very high density and hence
very high intensity, therefore watershedsegmentationisthe
best tool to classify cancers and high intensity tissues of
brain [6]. Watershed segmentation can classify the
intensities with very small difference also, which is not
possible with snake and level set method. A similar method
for tumor detection is proposed by Malhotra [2], but multi-
parameter extraction was not used. Hossam and P Vasuda
[10] have proposed a method for brain tumor detection and
segmentation using histogram Thresholding detects the
tumor but the result shown crops excessive area ofbrain. An
efficient and improved brain tumor detectionalgorithm was
developed by Rajeev Ratan, Sanjay Sharma and S. K. Sharma
[8] which makes use of multi-parameter MRI analysis and
the lumps cannot be segmented in 3-D unless and until we
have 3-D MRI image data set. So, a relatively simple method
for detection of brain lumps is presentedwhichmakesuseof
marker basedwatershedsegmentationwithimprovementto
avoid over & under segmentation. The Segmentation of an
image entails the division or separation of the image into
regions of similar attribute. The ultimate aim in a large
number of image processing applications is to extract
important features from the image data, from which a
description, interpretation, or understanding of the scene
can be provided by the machine. The segmentation of brain
lumps from magnetic resonance images is an important but
time consuming task performed by medical experts.[5] The
digital image processing community has developed several
segmentation methods, many of them ad hoc. Suchendra et
al. (1997) suggested a multi scale image segmentation using
a hierarchical self-organizing map; a high speed parall fuzzy
c-mean algorithm for brain lumps segmentation; an
improved implementation of brain lumps detection using
segmentation based on neuron fuzzy technique designed a
method on 3D variational segmentation forprocessesdueto
the high diversity in appearance of lumps tissue from
various patients. When experts work on lumps images then
they use three different types of algorithms. Some of the
techniques based on pixel based, some based on texture of
images and some of them based on structure of images.
3. AN AMELIORATE TECHNIQUE FORBRAINLUMPS
DETECTION
Content based image retrieval methods areoneofthem with
widely used in retrieving medical imagesfromMRIdata sets.
In this research paper we are presenting An Ameliorate
Technique, based on modified classifiers&feedback method
for lumps image retrieval from MRI images. This technique
uses new concepts of texture and shape properties of the
images by using fuzzy c-means clustering and modified SVM
classifier. It also uses subjective feedback method for
efficient and accurate results. This method uses following
phases –
Fig2: Phases of Brain Lumps Detection
Algorithm for an Ameliorate Technique for Brain Lumps
Detection
Input- Set of MRI images
Output- Detect Brain Lumps accurately
1- Select input image from image data set
2- Apply pre-processing on image
o Noise removal- by using Median filter

 Select a noisy MRI image and separate each plane and
each scalar component is treated in dependently.
 Generate zero arrays around an image based on image
mask size using pad array command.
 Select 3 * 3 masks from an image and the mask should
be odd sized
 For each component of each point under the mask a
single median component is determined.
 Contrast enhancement
3- Feature Extraction
 Boundary detection
 The Intensity Gradient and Texture map are calculated
 Extra features extraction such as texture and shape
properties
 Measure SVM classifiers
4- Modified Subjective feedback phase
 Apply Fussy C-Means clustering
4. CLUSTERING TECHNIQUES
Clustering is the processes of collection of objects which are
similar between them and are dissimilar objects belong to
the other clusters. The clustering is suitable in biomedical
image segmentation when the number of cluster is known
for particular clustering of human anatomy. Clustering
(pixel) is belong to the one cluster then it couldnotbelong to
another cluster. K-mean is the example of exclusive
clustering algorithm. In overlapping clustering, one data
(pixel) is belonging to the two or more clusters. Fuzzy C-
mean is example of the overlapping clustering algorithm.
[10]
K-MEANS CLUSTERING
K-mean clustering is unsupervised algorithms that solve
clustering problem. The procedure for k mean clustering
algorithm is the simple and easy way to segment the image
using basic knowledge of the clustering value. In k-mean
clustering initially randomly define k centroids. Selection of
this k centroid is placed in thecunningwaybecausedifferent
location makes the different clustering. So that, the better is
to place centroid value will be as much as far away from the
each other. Secondly calculate the distance between each
pixel to selected cluster centroid. Each pixel compares with
the k clusters centroids and finding the distance using the
distance formula. If pixel has theshortestdistanceamong all,
than it is move to particular cluster. Repeatthisprocessuntil
all pixel compare to the cluster centroids. The process
continues until some convergence criteria are the met [10]
FUZZY C-MEANS CLUSTERING
Fuzzy C-means clustering is the overlapping clustering
technique. One pixel value depending upon the two or more
clusters centers. It is also known as soft clustering method.
Most widely used the fuzzy clustering algorithm is theFuzzy
C-means (FCM) algorithm (Bezdek 1981). The FCM
algorithm is the partition of the n element X={x1,...,xn} intoa
collection of the c fuzzy clustering with respect to the below
given criteria [9][10][4] It is based on the minimization of
the following objective function:
Where, m = level of the fuzziness and real number
greater than 1. uij= degree of the membership of xi in the
cluster cj x = data value Fuzzy C-means is the popular
method for medical image segmentation but only consider
the image intensity thereby producingunsatisfactoryresults
in noisy images. A bunch of the algorithms are proposed to
make the FCM robust against noise and in homogeneity but
it’s still not perfect. In 2012, J. Selvakumar, A,LakshmiandT.
Arivoli [7] proposed a technique for brain lumps
segmentation using the k-means and the fuzzy c-means
algorithm. Its use the pre-processing steps for filtering the
noise and other artefacts in image and apply the K-means
and fuzzy c-mean algorithm. This purposed algorithm,fuzzy
c-mean is slower than the K-means in efficiencybutgivesthe
accurate prediction of lumps cellswhicharenotpredicted by
the K-means algorithm.
5. EXPERIMENTAL RESULT
The proposed algorithm and Fuzzy C-Means algorithm is
implemented using MATLAB software and tested on the
brain MRI images to explore the segmentation accuracy of
the proposed approach. The comparison is made between
the Fuzzy C-Means and proposed algorithm. The quality of
the segmentation of the proposed algorithm can be
calculated by segmentation accuracy which is given as.
The input image and the corresponding segmented image is
shown below.
Fig: 3 Experimental Results on different brain images
Following table & graph shows result comparison of k-
means and FCM algorithm
Table 1. Results
Parameters In %
Existing K-Means Implemented FCM
Precision 91 94.73
Recall 90.54 92.52
Specificity 88.70 96.77

Accuracy 89.90 92.51
Fig 4. Graphical comparison of Results
6. CONCLUSION
Recently, the applications of the image processing can be
found in the areas like electronics, remote sensing, natural
scene detection, biometrics, bio-medical etc. If we
concentration the bio-medical applications, image
processing is commonly used for the diagnosis of different
tissues purpose. By use of suitable image segmentation
method and the use of accurate input image we get accurate
results.
REFERENCES
[1] B.Jyothia, Y.MadhaveeLathab, P.G.Krishna Mohanc
,V.S.K.Reddy,” Integrated MultipleFeaturesforTumorImage
Retrieval Using Classifier and Feedback Methods”,
International Conference on Computational Modeling and
Security (CMS 2016), Science direct, 2016, PP 141-148.
[2] L.A. Khoo, P. Taylor, and R.M. Given-Wilson, “Computer-
Aided Detection in the United Kingdom National Breast
Screening Programme: Prospective Study,” Radiology, vol.
237, pp. 444-449, 2005.
[3] Pritam Dungarwal and Prof. Dinesh Patil, “Literature
Survey on Detection of lumps in brain” International Journal
of Modern Trends in Engineering and Research (IJMTER)
Volume 03, Issue 12, December – 2016,pp 312-316
[4] Krit Somkantha, Nipon Theera-Umpon, “Boundary
Detection in Medical ImagesUsingEdgeFollowingAlgorithm
Based on Intensity Gradient and TextureGradientFeatures,”
in Proc. IEEE transactionson biomedical engineering,vol.58,
no. 3, march 2011,pp. 567–573.
[5] B.Jyothi, Y.MadhaveeLatha, P.G.Krishna Mohan,”
Multidimentional Feature Vector Space for an Effective
Content Based Medical Image Retrieval 5th IEEE
International Advance Conputing Conference (IACC-
2015),BMS College of engineering Bangalore,June 12 to 13
,2015.
[6] Nitin Jain & Dr. S. S. Salankar,” Color & Texture Feature
Extraction for Content Based Image Retrieval”, IOSR Journal
of Electrical andElectronicsEngineering(IOSR-JEEE)e-ISSN:
2278-1676, p-ISSN: 2320-3331 PP 53-58.
[7] J. Selvakumar, A. Lakshmi, T. Arivoli, “Brain tumor
segmentation and its Area calculation in Brain MR images
using K-Mean clustering and Fuzzy C-Mean Algorithm”,
IEEE-International conference on advances in engineering
science and management, March-2012.
[8] Cs Pillai, “A Survey of Shape Descriptors forDigital Image
Processing”, IRACST - International Journal of Computer
Science and Information Technology & Security (IJCSITS),
ISSN: 2249-9555 Vol. 3, No.1, February 2013
[9] Prashant Aher and Umesh Lilhore, “An improved cbmir
architecture, based on modified classifiers & feedback
method for tumor image retrieval from mri images”
International Journal of Modern Trends in Engineering and
Research (IJMTER) Volume 03, Issue 12,Dec – 2016,pp-157-
160
[10] Ramakrishna Reddy, Eamani and G.V Hari Prasad,”
Content-Based Image Retrieval Using Support Vector
Machine in digital image processing techniques”,
International Journal of Engineering and Technology
(IJEST),IIN:0977-5462 vol.04 No04 April 2012

An Ameliorate Technique for Brain Lumps Detection Using Fuzzy C-Means Clustering

More Related Content

What's hot (20)

Similar to An Ameliorate Technique for Brain Lumps Detection Using Fuzzy C-Means Clustering (20)

More from IRJET Journal (20)

Recently uploaded (20)

An Ameliorate Technique for Brain Lumps Detection Using Fuzzy C-Means Clustering