MATLAB Based Brain Tumour Extraction Using Artificial Neural Network
International Journal on Recent and Innovation Trends in Computing and Communication
Volume: 4 Issue: 3
ISSN: 2321-8169
360 - 364
_______________________________________________________________________________________
MATLAB Based Brain Tumour Extraction Using Artificial Neural Network
Kalpana U. Rathod
Prof. Y. D. Kapse
Dept. of E&TC
Government college of Engg.
Jalgaon, Maharashtra, India
kalpanarathod6@gmail.com
Dept. of E&TC
Government college of Engg.
Jalgaon, Maharashtra, India
yogitakapse2013@gmail.com
Abstract- Brain tumour is the major cause of mortality among children and adults. The chance of survival can be greater when the tumour is detected
correctly at its early stage. This paper presents a neural network techniques for the classification of magnetic resonance brain image. The proposed
technique consists of three stages, features extraction from gray scale MR Image using gray level co-occurrence matrix, MR image segmentation
using k-mean clustering method and classification of MR Image into normal and abnormal (tumourous) image using feed-forward neural network.
This technique have been developed on MATLAB version 7.5.0 platform.
Keywords- Brain Tumor; MRI; Segmentation; Feature Extraction; GLCM; Artificial Neural Network; MATLAB.
___________________________________________________*****________________________________________________
3.
I. INTRODUCTION
Brain cancer can be counted among the most deadly and
intractable diseases. Tumours may be embedded in regions of
the brain that are critical to orchestrating the bodys vital
functions, while they shed cells to invade other parts of the
brain, forming more tumours too small to detect using
conventional imaging techniques. Brain cancers location and
ability to spread quickly makes treatment with surgery or
radiation like fighting an enemy hiding out among minefields
and caves.
Malignant Tumour
Malignancy is the type of tumour, which grows worst with
the passage of time and ultimately results in the death of a
person.
The proposed system is an efficient system for detection of
tumour and classification for given MRI images in normal and
tumorous image. The method of detection and classification
work is carried out during the process is explained. This method
is developed in MATLAB environment in order to check for
applicability of proposed method.
In recent years, the occurrence of brain tumors has been on
the rise. Unfortunately, many of these tumors will be detected
too late, after symptoms appear. It is much easier and safer to
remove a small tumor than a large one. About 60 percent of
glioblastomas start out as a lower-grade tumor. But small tumors
become big tumors. Low-grade gliomas become high-grade
gliomas. Once symptoms appear, it is generally too late to treat
the tumor. Computer-assisted surgical planning and advanced
image-guided technology have become increasingly used in
Neuro surgery [1].
II. PROPOSED METHODOLOGY
The proposed algorithm starts by reading the input brain MR
image and converting it into grey scale image. There are three
major steps in the proposed approach. The first step is image
pre-processing, the second step is feature extraction using
GLCM [5], the third step is classification using feed forward
neural network. Fig. 1 shows the steps included in proposed
system for brain tumour detection using artificial neural
network.
1.
Types of Tumour
There are three general types of Tumour: 1. Benign 2. Premalignant 3. Malignant
1.
Benign Tumour
A benign tumor is a tumour which does not expand in an
abrupt way; it doesnt affect its neighbouring healthy tissues and
also does not expand to non-adjacent tissues.
2. Pre-Malignant Tumour
Premalignant Tumour is a precancerous stage. It is considered as
a disease, if not properly treated it may lead to cancer.
Image Acquisition
Images are obtained using MRI scan and these scanned
images are displayed in a two dimensional matrices having
pixels as its elements. These matrices are dependent on matrix
size and its field of view. Images are stored in MATLAB and
displayed as a gray scale image of size 256*256. The entries of a
gray scale image are ranging from 0 to 255, where 0 shows total
black color and 255 shows pure white color. Entries between this
range vary in intensity from black to white [6].
For the implementation of this application we need to have
the images of different patients in our database in order to
360
IJRITCC | March 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication
Volume: 4 Issue: 3
ISSN: 2321-8169
360 - 364
_______________________________________________________________________________________
identify their condition. The MRI image is stored along with our
main file from various sources.
Image Acquisition
Image Pre-processing
Segmentation Using KMeans Clustering
Feature Extraction
g) Re-estimate centre.
h) Repeat the process until the centre doesn't move.
4. Feature Extraction
Texture features or more precisely, Gray Level Cooccurrence Matrix (GLCM) features are used to distinguish
between normal and abnormal brain tumors. Five co-occurrence
matrices are constructed in four spatial orientations horizontal,
right diagonal, vertical and left diagonal (0, 45, 90 and 135.)
A fifth matrix is constructed as the mean of the preceding four
matrices, From the mean co-occurrence matrix obtained, we
have to extract the 8 different statistical features and these
extracted features is applied at input of neural network. This
features are as follow.
(i) CONTRAST
Neural Network Classifier
Contrast is defined as the separation between the darkest and
brightest area.
Suspicious Regions
Figure 1
. Steps for the brain tumor extraction using artificial neural
network
2.
Image Pre-processing
In this phase image is enhanced in the way that finer details
are improved and noise is removed from the image. The first
step is to get the MRI image and application of pre-processing
steps. There are various methods which come under this step,
we will be dealing with only grey scale and filters. Basically
preprocessing is done to remove noise and blurring as well as
ringing effect in order to get the enhanced and much clear image
for our purpose. The filter which we have used is median filter
but as we are working on image samples that are required for the
medical purpose. The median filter has to be passed with mask
for better image.
3.
Image segmentation using k-means clustering
In our project we are using k-means clustering for to segment
the brain tumour purpose. In this method, we are grouping the
data and to select the mid value, then we are cluster the data in
k-means cluster the grouping data are also present in another
group of data, but fuzzy and c-means method it is not possible to
group same data present in another cluster so we are avoiding
fuzzy c-means and c-means clustering algorithm, instead we
used k-means clustering method [4].
STEPS FOR K-MEANS:
a) Give the no of cluster value as k.
b) Randomly choose k cluster centres.
c) Calculate mean or centre of the cluster.
d) Calculate the distance between each pixel to each cluster
centre.
e) If the distance is near to the centre then move to that cluster.
f) Otherwise move to next cluster.
(ii) CORRELATION
Correlation is computed into what is known as the correlation
coefficient, which ranges between -1 and +1.
(iii) HOMOGENITY
Homogeneity is defined as the quality or state of being
homogeneous.
(iv) ENTROPY
Entropy is a measure of the uncertainty in a random variable.
(v) ENERGY
It provides the sum of squared elements in the GLCM .Also
known as the uniformity or the angular second moment.
(vi) VARIANCE
The variance defines the variation of intensity around the
N 1
mean.
P i , j . ( i - )2
Varience =
i , j 0
Where,
361
IJRITCC | March 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication
Volume: 4 Issue: 3
ISSN: 2321-8169
360 - 364
_______________________________________________________________________________________
a)
The Sigmoid Function
(vii) MAXIMUM PROBABILITY
This is simply the largest entry in the matrix, and corresponds
to the strongest response. This could be the maximum in any of
the matrices or the maximum overall.
b) Hyperbolic Tangent Activation Function
Maximum Probability = max i, j p(i, j)
(viii)
DISSIMILARITY
c)
A Linear Function
N 1
Dissimilarity =
Pi, j * | i j |
i, j 0
6. Neural network Classifier
A Neural network classifier (Feed forward neural network.)
is used to detect candidate-circumscribed tumour. Generally, the
input layer consists of eight neurons corresponding to the eight
features. The output layer consists of one neuron indicating
whether the MRI is a candidate circumscribed tumour or not,
and the hidden layer changes according to the number of rules
that give best recognition rate for each group of features.
The Structure of a Feed forward Neural Network
Fig. 2 illustrates a typical feed forward neural network with a
single hidden layer.
Figure 2
III. EXPERIMENTAL RESULTS
In this paper, the pre-processing stage performs image
enhancement. The median filter is used for image enhancement,
it is used to remove the noise in an image. It is better than mean
filter, Weiner filter, Gaussian filter. In image segmentation kmeans clustering method is used which gives more accurate
result, using the K-means algorithm, it has an advantage of less
computing time. In other words, the partitioned clustering is
faster than the hierarchical clustering.
Fig. 3 shows the basic GUI layout of the proposed system.
. Feed forward neural network (single hidden layer).
Most neural networks passes the output of their layers
through activation functions. These activation functions scale the
output of the neural network into proper ranges.
There are total three activation function provided that can be
used are
Figure 3 . The basic layout of the proposed system.
In the proposed system the image is first loaded as shown in
the fig. 4.
362
IJRITCC | March 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication
Volume: 4 Issue: 3
ISSN: 2321-8169
360 - 364
_______________________________________________________________________________________
Figure 7 . Decision of Feed forward Classifier
Figure 4
. An image is loaded.
Fig. 5 shows the GUI with cluster image, extracted features
value, tumor part and graph of normalized histogram, sum of
distribution and distribution of intensity among the pixels of
input MR brain image.
IV. CONCLUSION
Image processing plays vital role in todays world. It has
important application in biomedical field. The proposed system
is developed for the diagnosing of tumour from magnetic
resonance imaging pictures of the brain. This method makes the
diagnosing in many phases. In the pre-processing stage filtering
is performed on brain MR images. In image segmentation stage
K-mean clustering method used to segment an MR image. After
that texture features are extracted from gray scale MR image
using GLCM method, these extracted features are given as input
to artificial neural network for classifying MR image into normal
and abnormal (tumorous) image.
ACKNOWLEDGMENT
Kalpana U. Rathod would like to thanks Prof. Y. D. Kapse
for their immense support and help.
REFERENCES
Figure 5 . Brain tumor segmentation
Figure 6
. Training of feed forward network
[1] H. E. Cline, E. Lorensen, R. Kikinis and F. Jolesz, Threedimensional segmentation of MR images of the head using
probability and connectivity, J. Comput Assist Tomography,
14:10371045, 1990.
[2] P. Gibbs, D. L. Buckley, S.J. Blackband and A. Horsman, Tumor
volume determination from MR images by morphological
segmentation, Phys. Med Biol, 41:24372446, 1996.
[3] D. L. Collins, T. M. Peters, W. Dai and A. C. Evans. Model based
segmentation of individual brain structures from MRI data, SPIE
Vis Biomed Comput, 1808:1023, 1992.
[4] S. M. Ali, Loay Kadom Abood, and Rabab Saadoon Abdoon,
Brain Tumor Extraction in MRI images using Clustering and
Morphological Operations Techniques, International Journal of
geographical Information System Application and Remote
Sensing, volume 4, No.1, ,ISSN 2077-771x, ICSRS publication,
June 2013.
[5] Shweta Jain, Brain Cancer Classification Using GLCM Based
Feature Extraction in Artificial Neural Network, International
Journal of Computer Science & Engineering Technology, Vol. 4,
No. 07, Jul 2013.
[6] Roopali R. .Laddha, S. A. Ladhake A Review on Brain Tumor
Detection Using Segmentation And Threshold Operations,
International Journal of Computer Science & Information
Technologies, Vol. 5, 2014.
[7] Swapnali Sawakare and Dimple Chaudhari, Classification of
Brain Tumor using Discrete Wavelet Transform, Principle
363
IJRITCC | March 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
International Journal on Recent and Innovation Trends in Computing and Communication
Volume: 4 Issue: 3
ISSN: 2321-8169
360 - 364
_______________________________________________________________________________________
[8]
[9]
[10]
[11]
[12]
Component Analysis & Probabilistic Neural Network,
International Journal For Research In Emerging Science And
Technology, Volume-1, Issue-6, November-2014.
Anam Mustaqeem, Ali Javed and Tehseen Fatima An Efficient
Brain Tumor Detection Algorithm Using Watershed &
Thresholding Based Segmentation, Image Graphics & Signal
Processing, 2012.
S. K. Warfield, J. Dengler and J. Zaers, Automatic identification
of gray matter structures from MRI to improve the segmentation of
white matter lesions, J Image Guid Surg, 1:326338, 1995.
Rohini Paul Joseph, C. Senthil Singh and M. Manikandan, Brain
Tumor MRI Image Segmentation & Detection In Image
Processing, International Journal Of Research in Engineering &
Technology, Vol.3, Issue 01, 2014.
N. J. Bonnie, M. B. Fukui and C. C. Meltzer, tumor volume
measurement: comparison of manual and semiautomated
methods,;212:811816, Radiology 1999.
H. Zhu, H.Y. Francis, F. K. Lam, PWF Poon, Deformable region
model for locating the boundary of brain tumors, In: Proceedings
of the IEEE 17th Annual Conference on Engineering in Medicine
and Biology 1995, Montreal, Quebec, Canada: IEEE,1995; 411.
364
IJRITCC | March 2016, Available @ http://www.ijritcc.org
_______________________________________________________________________________________
