See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/236002493

Detection and Quantification of Brain Tumor from MRI of Brain and it’s
Symmetric Analysis

Article · June 2012

CITATIONS READS

106 6,953

2 authors:

Sudipta Roy Samir Bandyopadhyay

Jio Institute University of Calcutta
92 PUBLICATIONS 983 CITATIONS 1,043 PUBLICATIONS 9,658 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Machine learning Technique for automatic detection and recognition View project

Establishments of Medical Exposure Characteristics of the Moroccan population and the National Dose References Levels in Diagnostic and Interventional Imaging.
View project

All content following this page was uploaded by Sudipta Roy on 11 May 2014.

The user has requested enhancement of the downloaded file.

 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

Detection and Quantification of Brain Tumor from MRI of Brain and it’s
Symmetric Analysis
Sudipta Roy, Samir K. Bandyopadhyay
Department of Computer Science and Engineering,
University of Calcutta, 92 A.P.C. Road, Kolkata-700009, India.

ABSTRACT
In this work a fully automatic algorithm to detect brain tumors by using symmetry analysis is proposed. Here we detect the tumor,
segment the tumor and calculate the area of the tumor. The quantitative analysis of MRI brain tumor allows obtaining useful key
indicators of disease progression. The complex problem of segmenting tumor in MRI can be successfully addressed by considering
modular and multi-step approaches mimicking the human visual inspection process. The tumor detection is often an essential
preliminary phase to solve the segmentation problem successfully. The experiments showed good results also in complex situations.
Segmentation of images embraces a significant position in the region of image processing. It becomes more and more significant
while normally dealing with medical images; magnetic resonance (MR) imaging suggest more perfect information for medical
examination than that of other medical images such as ultrasonic , CT images and X-ray. Tumor segmentation and area calculation
from MRI data is an essential but fatigue, boring and time unbearable task when it completed manually by medical professional
when evaluate with present day’s high speed computing machines which facilitate us to visual study the area and position of
unnecessary tissues.

Keywords: MRI image, Segmentation, Tumor detection, Morphological analysis, Symmetry analysis.

the parts contained in the human body.MRI imaging is often

1. INTRODUCTION used when treating brain tumors. These high resolution
images are used to examine human brain development and
The principle of our task is to recognize a tumor and discover abnormalities. Nowadays there are several
its quantifications from a particular MRI scan of a brain methodologies for classifying MR images. Among all
image using digital image processing techniques and compute medical image processing, image segmentation is initial and
the area of the tumor by fully automated process and its important work, for example, quantification of specified area
symmetry analysis. In recent years a great effort of the must based on accurate segmentation.
research in field of medical imaging was focused on brain A tumor is a mass of tissue that grows out of control
tumors segmentation. The automatic segmentation has great of the normal forces that regulates growth. The multifaceted
potential in clinical medicine by freeing physicians from the brain tumors can be split into two common categories
burden of manual labelling; whereas only a quantitative depending on the tumors beginning, their enlargement
measurement allows to track and modelling precisely the prototype and malignancy. Primary brain tumors are tumors
disease. Despite the undisputed usefulness of automatic tumor that take place commencing cells in the brain or commencing
segmentation, this is not yet a widespread clinical practice, the wrapper of the brain. An inferior or metastatic brain tumor
therefore the automatic brain tumor segmentation is still a takes place when cancer cells extend to the brain from a
widely studied research topic. The main difficulties in field of primary cancer in a different component of the body. The
automatic tumor segmentation are related to the fact that the majority of investigations in developed countries demonstrate
brain tumors are very heterogeneous in terms of shape, color, that the amount of people who develop brain tumors and
texture and position and they often deform other nearby depart this life from them has greater than before maybe as
anatomical structures. An healthy brain has a strong sagittal much as 300 over past three decades. The computationally
symmetry, that is weakened by the presence of tumor. The efficient method runs orders of magnitude faster than current
comparison between the healthy and ill hemisphere, state of the art techniques giving comparable or improved
considering that tumors are generally not symmetrically results. Our quantitative results indicate the benefit of
placed in both hemispheres, was used to detect the anomaly. incorporating model aware affinities into the segmentation
One of the motivations to make the substandard segmentation process for the difficult case of brain tumor. This paper
of good organization is the occurrence of artefact in the MR expresses a well-organized technique for automatic brain
images. One such artefact is the additional cranial tissues tumor segmentation for the removal of tumor tissues from
(skull). These additional cranial tissues repeatedly hamper MR images. A well acknowledged segmentation trouble
with the ordinary tissues throughout segmentation that within MRI is the task of category voxels according to their
accounts for the substandard segmentation efficiency. tissue type which take account of White Matter (WM), Grey
Magnetic Resonance Imaging (MRI) is an advanced medical Matter (GM) , Cerebrospinal Fluid (CSF) and occasionally
imaging technique used to produce high resolution images of pathological tissues like tumor etc. A brain tumor is an

477
 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

intracranial mass produced by an uncontrolled growth of cells edema or swelling in the nearby brain, As all tumor do not
either normally found in the brain such as neurons, lymphatic have a clear boundary between active and necrotic parts there
tissue, glial cells, blood vessels, pituitary and pineal gland, is need to define a clear boundary between edema and brain
skull, or spread from cancers primarily located in other tissues. Hassan Khotanlou et all [4] recommend a common
organs [1]. Brain tumors are classified based on the type of automatic scheme for segmenting brain tumors in 3D MRI.
tissue involved, the location of the tumor, whether it is benign Our scheme is valid in dissimilar types of tumors with MRI
or malignant, and other considerations. Primary (true) brain images. Its effect represent the initialization of a segmentation
tumors are the tumors that originated in the brain and are technique based on a mixture of a deformable model and
named for the cell types from which they originated. They spatial associations, principal to a particular segmentation of
can be benign (non cancerous), meaning that they do not the tumors. P.Narendran, V.K. Narendira Kumar, K.
increase in a different place or attack neighbouring tissues. Somasundaram [5] proposed a new method for segmentation
They can also be malignant and invasive (spreading to of pathological brain structures. This method combines prior
neighbouring area). Secondary or metastasis brain tumors information of structures and image information (region and
take their origin from tumor cells which increase to the brain edge) for segmentation. The automated brain tumor
from a different position in the body. Most frequently cancers segmentation method that we have developed consists of two
that increase to the brain to reason secondary brain tumors main components: pre-processing and segmentation. The
begin in the lumy, breast, and kidney or from melanomas in inputs of this system are two different modalities of MR
the skin. The first aim of this work is to develop a framework images: CE-T1w and FLAIR that we believe are sufficient for
for a robust and accurate segmentation of a large class of brain tumor segmentation [6]. The Graph Cut [7] method
brain tumors in MR images. Most existing methods are attempts to solve the min cut/max flow problem. Snakes and
region-based. They have several advantages, but line and Level Sets are active contour methods that evolve a curve
edge information in computer vision systems are also based upon geometric and image constraints. For the problem
important. The proposed method tries to combine region and of brain tumor segmentation, Lefohn et al. [8] implemented a
edge information, thus taking advantage of both approaches level set solver on the GPU. Quantitative results of this level
while cancelling their drawbacks. We first segment the brain set formulation compare well with hand contouring results.
to remove non-brain data. However, in pathological cases, Kaus et al. [9] used an atlas and statistical information to
standard segmentation methods fail, in particular when the segment brain tumors. Edward Kim et al. [10] method utilizes
tumor is located very close to the brain surface. Therefore we statistical seed distributions to overcome the local bias seen in
propose an improved segmentation method, relying on the the traditional cellular automata framework. Our results show
approximate symmetry plane. improved accuracy, robustness, and competitive usability.
Further, with a GPU implementation, the method produces
2. RELATED WORK results at interactive rates.

Image segmentation represents a method of 3. MATERIALS AND METHODS

separation a portion of image into separate area. A great
assortment of dissimilar segmentation approaches for images We have used these basic concepts to detect tumor in
have been developed. The Segmentation of an image entails our paper, the component of the image hold the tumor
the division or separation of the image into regions of similar generally has extra concentration then the other segment and
attribute. The ultimate aim in a large number of image we can guess the area, shape and radius of the tumor in the
processing applications is to extract important features from image. We calculate the area in pixel. Noise existing in the
the image data, from which a description, interpretation, or image can decrease the capability of region growing filter to
understanding of the scene can be provided by the machine. grow large regions or may result as a fault edges. When faced
Among them, the clustering technique have been with noisy images, it is generally convenient to pre-process
comprehensively explore and used in T.Logeswari and the image by using median filter. Median filters have the
M.Karnan [2], a clustering support come close to using a self robustness and edge preserving capability of the classical
organizing map (SOM) algorithm is projected for medical median filter. In pre-processing some fundamental image
image segmentation. This paper illustrate segmentation enhancement and noise lessening procedure are applied.
scheme consists of two stages. In the opening stages, the MRI Apart from that dissimilar traditions to identify edges and
brain image is obtained from patient database. In that film doing segmentations have also been used. The intention of
artefact and noise are disconnected. In the subsequent stages these steps is fundamentally to recover the image and the
(MR) image segmentation is to precisely recognize the major image superiority to get more guarantee and ease in identify
tissue arrangement in these image areas. In R. Rajeswari et al. the tumor. The noise is reducing by the conversion of gray
[3] proposed a Spectral leakage has the effect of the scale image. Then this gray scale image pass in to the filter.
frequency analysis of finite-length signals or finite-length We use here a high pass filter imfilter function in matlab to
segments of infinite signals. In brain the tumor itself, filter an image, replaces each pixel of the image with a
comprising a necrotic (dead) part and an active part, the weighted average of the surrounding pixels. The weights are

478
 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

determined by the values of the filter, and the number of of the integer type is shortened, and fractional values are
surrounding pixels is determined by the size of the filter used. rounded.
Then the gray image and filtered image are merged together Step5:- Add step2, step4 image and a integer value 45 and
to enhanced the image quality. Here we use Median filtering pass it in to a median filter to get the resultant enhanced
which is a nonlinear operation often used in image processing image.
to reduce "salt and pepper" noise. A median filter is more Step6:- Computes a global threshold that can be used to
effective than convolution when the goal is to simultaneously convert an intensity image (Step5) to a binary image with a
reduce noise and preserve edges. We use here matlab normalized intensity value which lies in between range 0 and
command medfilt2. Then we convert the filtered image into 1.
binary image by the thresholding method which computes a Step7:- Compute watershed segmentation by matlab
global threshold that can be used to convert an intensity command watershed (step6 image).
image to a binary image with normalized intensity value Step8:-Compute the morphological operation by two matlab
between 0 and 1.The uses Otsu's method [19], which chooses command imerode and imdilate and strel with arbitrary shape.
the threshold to minimize the intraclass variance of the black Step9:- Store the size of the step 8 image into var1 and var2
and white pixels. Then segment the threshold image by i.e no. Of rows and column in pixels by
watershed segmentation because It is the best method to [var1 var2]=size(step8 image)
segment an image to separate a tumor but it suffers from over Step10:- For i=1:1:var1 do
and under segmentation, due to which we have used it as a Step11:- For j=1:1:var2 do
check to our output. It not give the better result after that Step12:- If step8 image (i,j) == 1 do
some morphological operations are applied on the image after Step13:- step2 image (i,j) = 255
converting it into binary form. The basic purpose of the Step14:- Else do
operations is to show only that part of the image which has Step15:- step2 image (i,j) = step2 image (i,j) * 0.3
the tumor that is the part of the image having more intensity Step16:- End If
and more area then that specified in the strel command. The Step17:- End For
basic commands used in this step are strel, imerode and Step18:- End For
imdilate, Imerode: It is used to erode an image. Imdilate: It is Step19:- Convert in to binary image and traces the exterior
used to dilate an image. Marge these morphological outputs boundaries of objects, as well as boundaries of holes inside
with grayscale image by the step9 to step19 and we get these objects, in the binary image and into an RGB color
resultant output in which tumor detect sharply. Then we make image for the purpose of visualizing labeled regions.
the resultant image with sharp location of tumor by Step20:- Show only tumor portion of the image by remove
morphological output image and gray image from step9 to the small object area.
step18. Traces the exterior boundaries of objects, as well as Step21:- Compute edge detection using sobel edge detection
boundaries of holes inside these image, in the binary image, it technique.
also descends into the outermost objects (parents) and traces
their children (objects completely enclosed by the parents). It 3.2 Algorithm for Area Calculation
must be a binary image where nonzero pixels belong to an
object and 0 pixels constitute the background. Output is Input: Tumor portion of the image.
shown only in the color portion of the image with tumor.
Output: Area of the tumor.
Then tumor area is calculated from 2nd algorithms. From this
area we can assume the dangerousness of tumor.
Step1:- Read the input color or grayscale image.
3.1 Algorithm for Detecting Brain Tumor Step2:- Converts input colour image in to grayscale image
which is done by forming a weighted sum of each three
Input: MRI of brain image. (RGB) component, eliminating the saturation and hue
Output: Tumor portion of the image. information while retaining the luminance and the image
returns a grayscale colour map and store it into variable I.
Step1:- Read the input color or grayscale image. Step3:- Compute numbers of rows and column in pixels by
Step2:- Converts input colour image in to grayscale image [r2 c2] = size (I)
which is done by forming a weighted sum of each three Step4:- Initialize a variable a=0
(RGB) component, eliminating the saturation and hue Step4:- For i=1:1:r2 do
information while retaining the luminance and the image Step4:- For j=1:1:c2 do
returns a grayscale colour map. Step4:- If I (i,j)==255 do
Step3:- Resize this image in to 200 × 200 image matrix. Step4:- a=a+0
Step4:- Filters the multidimensional array with the Step4:- Else do
multidimensional filter. Each element of the output an integer Step4:- a=a+1
or in array, then output elements that exceed the certain range Step4:- EndIF

479
 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

Step4:- EndFor
Step4:- End For
Step4:- Display the area a.

Results are shown below with image name BT1.

Figure 1 is the original MRI scan image; Figure 2 shows
grayscale conversion of the image; Figure 3 is the output of
the after grayscale image pass into the high pass filtered
image and then Figure2 and figure3 image are superpose
with a median filter image and get the resultant enhanced Figure 7: Morphol Figure 8: Output
image of Figure 4; Figure 5 and Figure 6 shows the threshold -ogical Output with Tumor Location
segmentation with threshold value 0.35 and watershed
segmentation to localize the tumor portion of the image.
Morphological operations with arbitrary shape are applied in
Figure 7; Location of the tumor with input image is shown in
figure 8; Figure 9 is the edge detection with sobel parameter.
Figure 10 and Figure 11 are the colour output tumor with
noise and tumor without noise.

Figure 9: Sobel Figure 10: Tumor

Edge Detection with Noise

Figure 1: Original Figure 2: Grayscale

Input Image Image
Figure 11: Final output
with Tumor Portion only

Table 1: Contains image size with tumor size in

pixels i.e tumor area in pixels with different images.

Image name Image size Tumor size

BT1 200×200 6040

Figure 3: High Figure 4: Enhanced
pass filter Image Image
BT2 200×200 5080

BT3 200×200 4913

BT4 200×200 2144

BT5 200×200 2778

BT6 200×200 8080

Figure 5: Thresh- Figure 6: Watersh-

hold segmentation ade segmentation

480
 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

3.3 Some other Results are shown below

A1 A2 A3 A4

BT2

BT3

BT4

BT5

BT6

Figure 12: Shows the output image with different input MRI image where BT2,BT3,BT4,BT5,BT6 are
the different input image name and A1,A2,A3,A4 are the input image , morphological output, tumor
location with image, colored output with only tumor portion of the MRI image.

481
 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

3.4 Flow Chart of our Proposed Method 4. CONCLUSIONS & FUTURE WORKS
We proposed an interactive segmentation method
that enables users to quickly and efficiently segment tumors
in MRI of brain. We proposed a new method that in addition
to area of the region and edge information uses a type of prior
information also its symmetry analysis which is more
consistent in pathological cases. Since tumor is a rather
general concept in medicine, limitations of the proposed
approach might become apparent as soon as unforeseen
pathologic tissue types that could not adequately be captured
by the discriminative model appear in previously unseen
patients. Especially secondary tumors might be composed of
an enormous variety of tissue types depending on the primary
tumor site. Its application to several datasets with different
tumors sizes, intensities and locations shows that it can
automatically detect and segment very different types of brain
tumors with a good quality.
For our future work, we plan to work with a greater
number of brain structures and explore incorporating
additional information to guide our proposal. We would also
like to explore higher dimensional data and improve our user
interface and investigate possibilities to handle this issue. The
goal is to detect, to segment, and to identify most types of
pathological tissue that occur within pediatric brain tumors.

REFERENCES
[1] Dou, W., Ruan, S., Chen, Y., Bloyet, D., and Constans, J.
M. (2007), “A framework of fuzzy information fusion for
segmentation of brain tumor tissues on MR images”,
Image and Vision Computing, 25:164–171.

[2] T.Logeswari and M.Karnan, “An Improved

Implementation of Brain Tumor Detection Using
Segmentation Based on Hierarchical Self Organizing
Map”, International Journal of Computer Theory and
Engineering, Vol. 2, No. 4, August, 2010,pp.1793-8201.

[3] R. Rajeswari and P. Anandhakumar, “Segmentation and

Identification of Brain Tumor MRI Image with Radix4
FFT Techniques”, European Journal of Scientific
Research, Vol.52 No.1 (2011), pp.100-109.

[4] Hassan Khotanlou, Olivier Colliot and Isabelle Bloch,

“Automatic brain tumor segmentation using symmetry
analysis and deformable models”, GET-Ecole Nationale
Superieure des Telecommunications, France.

[5] P.Narendran, V.K. Narendira Kumar, K. Somasundaram,

“3D Brain Tumors and Internal Brain Structures
Segmentation in MR Images”, I.J. Image, Graphics and
Signal Processing, 2012, 1, 35-43.

[6] Dou, W., Ruan, S., Chen, Y., Bloyet, D., and Constans, J.
M. (2007), “ A framework of fuzzy information fusion

482
 Volume 2 No. 6, June 2012 ISSN 2223-4985

International Journal of Information and Communication Technology Research

©2012 ICT Journal. All rights reserved

http://www.esjournals.org

for segmentation of brain tumor tissues on MR images”, Image Segmentation,” World Academy of Science,
Image and Vision Computing, 25:164–171. Engineering and Technology 4, 2005.

[7] Y. Boykov and V. Kolmogorov. An experimental [17] K. Pyun, J. Lim, C. S. Won, and R. M. Gray, “Image
comparison of min-cut/max-flow algorithms for energy segmentation using hidden Markov gauss mixture
minimization in vision. IEEE Transactions on Pattern models”, IEEE Trans. Image Process., vol. 16, no. 7,
Analysis and Machine Intelligence, pages 1124–1137, pp.1902–1911,Jul. 2007.
2004.
[18] Sudipta Roy, Atanu Saha and Prof. Samir K.
[8] A. Lefohn, J. Cates, and R. Whitaker. Interactive, gpu- Bandyopadhyay “ Brain tumor segmentation and
based level sets for 3d segmentation. MICCAI, pages quantification from mri of brain”, Journal of Global
564–572, 2003. Research in Computer Science, Volume 2, No. 4, April
2011.
[9] M. Kaus, S. K. Warfield, A. Nabavi, P. M. Black, F. A.
Jolesz, and R. Kikinis. “Automated Segmentation of MRI [19] Otsu, N. “A threshold selection method from gray-level
of Brain Tumors”, Radiology, 218(2)(586-91), 2001 Feb. histogram”, IEEE Trans. Syst. Man Cybern. 9, 62–66
(1979).
[10] Edward Kim, Tian Shen, Xiaolei Huang, “A Parallel
Cellular Automata with Label Priors for Interactive Brain
Tumor Segmentation”, Lehigh University, Department of ABOUT AUTHORS
Computer Science and Engineering, Bethlehem, PA,
USA, 2010.
Sudipta Roy, He is pursuing M.Tech
[11] Sudipta Roy and Prof. Samir K. Bandyopadhyay in the Dept. Of Computer Science &
“Contour Detection of Human Knee”, IJCSET Engineering , University of Calcutta,
,September 2011 , Vol 1, Issue 8,pp. 484-487. India. He received B.Sc(Phys Hons)
from Burdwan University and B.Tech
[12] T. Logeswari and M. Karnan , “An improved from Calcutta University. He is
implementation of brain tumor detection using Author of more than five publications
segmentation based on soft computing” Journal of in National and International Journal. Field of interest is
Cancer Research and Experimental Oncology Vol. 2(1) Biomedical Image Analysis, Image Processing,
pp. 006-014, March, 2010. Steganography, Database Management System , Data
Structure, Artificial Intelligence, Programming Languages
[13] Prof. Samir K. Bandyopadhyay and Sudipta Roy etc.
“Detection of Sharp Contour of the element of the WBC
and Segmentation of two leading elements like Nucleus
and Cytoplasm”, International Journal of Engineering Samir K Bandyopadhyay, He is
Research and Applications (IJERA), Vol. 2, Issue 1,Jan- Professor of Dept. Of Computer
Feb 2012, pp.545-551. Science & Engi-neering, University of
Calcutta, Kolkata, India. Chairman,
[14] J. J. Corso, E. Sharon, and A. Yuille, “Multilevel Science & Engineering Research
Segmentation and Integrated Bayesian Model Support Society(SERSC, Indian Part),
Classification with an Application to Brain Tumor Fellow of Computer Society of India,
Segmentation,” in Medical Image Computing and Sectional President of ICT of Indian
Computer Assisted Intervention, vol. 2, 2006, pp. 790– Science Congress Association, 2008-2009,Senior Member of
798. IEEE, Member of ACM, Fellow of Institution of Engineers
(India), Fellow of Institution of Electronics & Tele
[15] Sudipta Roy and Prof. Samir K. Bandyopadhyay “Visual Communication Engineering, India, Reviewer of International
Image Based Hand Recognitions”, Asian Journal Of Journals IEEE Trans on Neural Networks, ACM, Springer
Computer Science And Information Technology1:4 Publications. Field of Specialization Bio-medical Engg,
(2011), pp.106 – 110. Mobile Computing, Pattern Recognition, Graph Theory,
Image Processing, etc. Published Books like Datastructure
[16] Terrence Chen, and Thomas S. Huang, “Region Based Using C, Addison Wesley, 2003, C Language, Pear-son
Hidden Markov Random Field Model for Brain MR Publication, 2010. Author of more than hundred publications
in National and International Journal and Conference.

483

View publication stats