Detection of Brain Tumor by Particle Swarm Optimization using Image Segmentation

Background and Objective: Image processing is a technique or set of operations to get meaningful information from an image for the usefulness and effectiveness of images. Image segmentation is an efficient technique in extracting and separating some of the features in the images. Methods: The main objective of this research work is to find the best fit of FCM algorithm over finding the axial and coronal plane of MRI brain imagesvia its accuracy and computational time.In the preprocessing, brain images of MRI have been converted from the DICOM format into standard image. Preprocessing is carried out by Gaussian filter technique to remove the noises in the images. The Fuzzy C Means (FCM) algorithm is implemented to segment the tumor affected region in the MR images. Results: By comparing the histogram values of the images (before and after segmentation) with the cluster center values by the FCM algorithm, the efficiency and accuracy of the algorithm is evaluated. Conclusion: The best fit of FCM algorithm into the axial and coronal plane is identified based on the computational time in this work.


Introduction
Image Processing (IP) is an analysis of a picture using techniques which can identify shapes, colors and relationships between them which are not perceived by the human eye.IP manipulates the data using an equation, or series of equations and then stores the result of the computation for each pixel (picture element )and generate a new image.IP is a technique in which the data from an image are digitized and various mathematical operations are applied to the image data to create an enhanced image.Image processing widely exists in many fields, such as education, remote sensing, military, film industry, video, medical, and so on 1 .
Segmentation is the process of dividing an image into regions with similar properties such as gray level, color, texture, brightness, and contrast.Analyzing the image is to extract the image part from the original image and then using method for image segmentation, which is quite important.e main aim for segmentation is to partition an image into mutually exclusive regions such that each region with respect to the pixel intensity is homogenous to a predefined criterion.It divides the image into a number of regions with specific and unique properties 1 .

Detection of Brain Tumor by Particle Swarm Optimization using Image Segmentation
e role of segmentation is to sub divide the objects in an image; in case of medical image segmentation the aim is to study anatomical structure and identify the region of interest i.e. locate tumor, lesion and other abnormalities.Measuring the tissue volume to measure the growth of tumor (also decrease in size of tumor with treatment) and helping in treatment planning prior to radiation therapy and calculate the drug.
e existing image segmentation methods can be mainly divides into four categories: first one is based on the threshold segmentation method; and second is based on the region and the third one is based on edges and final is based on the theory 1 .
MRI is an invasive diagnostic technique that produces the computerized digital images of tissues of the internal body based on the nuclear magnetic resonance of atoms of the body.
e application of radio waves in MRI radiates the magnetic field and pulses of energies of radio waves to make the images of organs, tissues and structures inside the body.An MRI can reveal minor damage to tendons, ligaments, and muscles.MRI is the brain for tumors, an aneurysm, bleeding in the brain, nerve injury, and other problems, such as damage caused by a stroke.e advantage of MRI with other diagnostic imaging modalities is its higher spatial resolution and the discrimination of so tissue 2 .e three primary imaging planes that are utilized in neuroimaging are axial plane: Transverse images represent "slices" of the brain.Sagittal plane: Images taken perpendicular to the axial plane which separate the le and right sides (lateral view).Coronal plane: Images taken perpendicular to the sagittal plane which separate the front from the back (frontal view), all the three types are shown in Figure 1.Optimum approach of detecting abnormalities in MRI brain images using texture feature analysis is a research work done by Abhinav Das and Nitin Jain 3 .In this research work the brain cancer images are achieved by three basic stages to attain more quality and accuracy.Image segmentation stage uses the threshold segmentation mechanism done by Fuzzy C Means threshold algorithm.e final stage is helped for making a comparison between normal and abnormal images.V. Sheejakumari and B. Sankara Gomathi carried their research work titled as MRI brain in the aid of improved particle swarm optimization for the classification of healthy and pathological tissues 2 .ey proposed a new method Improved Particle Swarm Optimization (IPSO) to categorize the MRI images of healthy and pathological tissues.It implements the results and analyzed based on the various statistical performance measures.
3-D image segmentation of hybridizing of the Particle Swarm Optimization (PSO) and Biogeography-Based Optimization (BBO) algorithm is implemented in a research work 4 .
e experimental result revealed that the proposed hybridization algorithm is performed well for 2D and 3D images.Another research work proposes an improvement method for image segmentation using the Fuzzy C-Means Clustering (FCM) algorithm.is algorithm is widely applied and achieved the best result in the field of image segmentation.ey suggested three different levels for further improvement of these results.
e first is related to Fuzzy C-Means algorithm and uses the Metaheuristic optimization.e second level is anxious with the use of Mahalanobis distance to reduce the geometrical shape in different classes with the spatial gray-level information of the image in the segmentation process.e final level corresponds in correcting the error of clustering by reallocating the potentially misclassified pixels to refining the segmentation results 5 .
Data clustering using Particle Swarm Optimization is discussed in a research work by DWvander Merwe 6  and AP Engelbrecht.ey propose two new approaches using PSO to cluster data.
is work explores PSO the user specified number of clusters to find the centroids.e second algorithm uses PSO to refine the clusters formed by k-means 6 .Image segmentation for amultilevelthresholding method for image segmentation is based on the hybridization of modified PSO and Otsu's method by Hamdaoui and et al 7 .e proposed method is compared with multi level segmentation methods and basic PSO method to expose its efficiency.It improves the objective function value to update its velocity and position of the particle.Akhilesh et al. 8 carried out a research work titled as momentum component adaptive PSO algorithm for image segmentation.In this work, a proposed variant of PSO is described on optimal multilevel thresholding algorithm.It also proposes iterative scheme that is more suitable for practically obtained initial values 8 .
Mahamed G.H. Omran et al. 9 have carried out a research work as application of unsupervised image classification using Particle Swarm Optimization in Dynamic Clustering 9 .
ey proposed the Dynamic Clustering algorithm based on PSO (DCPSO) approach which automatically determines the optimum number of clusters in the data set and simultaneously minimal the user interface.It uses a validity index to measure the quality of the resultant clustering.It has been applied on MRI images, satellite images and also nature images.e goal of this research is to distinguish the MRI scanning brain images from abnormal and normal images.
ree different hybridization methods of PSO and ABC, they proposed three new variants consisting IABAPFNN, ABC-SPSO-FNN, and HPA-FNN of Feed-Forward Neural Network (FNN) 10 .
Sahil J Prajapati and Kalpesh R Jadhav 11 have carried their research work in the introduction of nonnegative matrix factorization detecting the brain tumor by various image segmentation techniques.ey have utilized the nonnegative matrix factorization as one of the most promising technique to reduce the dimensionality of the data.In this, the goal of Nonnegative Matrix Factorization (NMF) is to find two nonnegative matrices and all matrices to contain only nonnegative elements and the results shows no cancellations, linear super position and considerable sparsity 11 .Parag Puranik et al. 12 did a research work in comprehensive learning Particle Swarm Optimization for human perception-based color image segmentation.e main objective is to generate a fuzzy system for image segmentation and color classification with the minimum error rate and least number of rules.Fuzzy sets are defined on the Hue (H), Saturation (S) and Lightness (L) components of the HSL color 12 .
is research work is organized as follows.e section 2 describes the methodology and the PSO algorithm and its applications.
e experimental results and the comparison of different criteria are presented in section 3. Section 4 concludes this research work.

Methodology
e detection of blocks in brain (tumor) carries various methods in the last decade.In this, the PSO algorithm is well been suited for the MRI brain data analysis.Brain tumor is a collection of growth of abnormal cells in the brain and the area closer to it.Many different types of brain tumors exist.Brain tumors of type benign, which is an early stage cannot be spread to the other regions.But the cancer tumor of type malignant spreads to other regions and other parts of the body.In general brain tumors that cannot cause cancer are benign and tumors that spread the cancer cells to the other region of the body are malignant.is work is to analyze the brain tumor and its affected region.

PSO Algorithm
PSO was initially developed by Kennedy and Eberhart in 1995 2 .e researchers adopted due to its optimization accuracy to solve variety of engineering optimization problems.In this decade PSO based approaches are widely efficient in image segmentation application.PSO is a heuristic global optimization method and also an optimization algorithm, based on swarm intelligence.
e concept of PSO is originated from the behavior of particles of swarm and the social interaction between particles.While searching for the food, the birds get scattered or they move together to find the food is the nature of behavior.e birds search for the food from one place to another, the bird which is nearer to food can smell the food.e basic algorithm of PSO consists of n swarm particle, and the position of each of the particle stands for the potential solution.e swarm particle changes its position according to the three principles.
Keep its inertia Update the condition with respect to its optimal position Update the condition with respect to the most optimal position of swarm.
Detection of Brain Tumor by Particle Swarm Optimization using Image Segmentation In these two equations, k id v and k id x denotes the speed of the particle of i and its k times and the d-dimension quantity of its position, k id pbest represents the d-dimension quantity of the individual i at its most optimist position at its k times.k id gbest is the optimist position of the d-dimension quantity of the swarm.e speeding figures are indicated by c 1 and c 2 it regulates the length when moving to the most particle of the whole swarm and to the most optimist individual particle.e random fiction is represented by r 1 and r 2 , in which the random range is denoted by the interval 0- e optimization of the particle and the most optimist particle can transmit information onto the other particles and the speed is very high and efficient through the new generations.e PSO accepts the real number code, the calculation is very simple, and it easily depicts the direct solution.e number of dimension is equal to the constant of the solution 13 .Biomedical application includes human tremor analysis, cancer classification, survival prediction, gene clustering, protein structure prediction and docking, drug design and in many pharmaceutical applications.Control-it is the largest sharing around 7.6% of application area of PSO in the IEEE Xplore database.e area includes designing of controllers, automatic generation of control tuning, ultrasonic motor control, power plants, and traffic control flow and system control 14 .
e application areas of PSO are system design, multiobjective optimization, classification, pattern recognition, biological system modeling, scheduling, signal processing, games, robotics applications, decision making, simulation and identification 15 .e MRI brain image data is taken as input to detect the tumor by image segmentation.
e research work uses the fast border mark algorithm based on horizontal and vertical lines of MR slices, to detect the abnormal region is carried out by K. Bharathi and S. Karthikeyan titled as A Novel Implementation of Image Segmentation for Extracting Abnormal Images in Medical Image Applications 16 .

Experimental Results
e detection and identification of brain diseases through the PSO algorithm and the extraction of infected area are carried out in this research work. is work takes the MRI brain image as input data.
e experiments were carried out on the platform of IBM running Windows 7 operating system.e algorithm was developed via Matrix Laboratory (MATLAB) (R2008a).In this work, two different planes, axial and the coronal plane of the MRI brain image data are given as input.Here, the PSO algorithm uses the MRI brain image as input, in thoseparticles are the pixel values and velocity is replaced by intensity of the image.
Converting the Digital Imaging Communication in Medicine (DICOM) file format into image file is the first stage.On applying the algorithm the following resultant image is obtained in the second stage.e MRI brain images of different person is taken and the algorithm is applied to detect the block in the given dataset.With starting the segmentation level n=1 the PSO algorithm shows the null value, so the default value is taken n=2.When n=2 the algorithm work for the MRI brain image data shows the result as in the Figure 2      e best resultant image from the different segmentation level is achieved with the equation 3. e Table 1 shows the elapsed time of axial and coronal plane and the average of the two planes for the comparison and Figure 4. depicts the chart of the elapsed time and segmentation level of the axial and coronal plane.On the basis of elapsed time and the Figure 4 shows that the segmentation level n=4 gives the accuracy result.e final stage involves the removal of noise from the image and it also extracts the affected region i.e. tumorportion from the best resultant image.Figures 5 and 6 results the abnormal growth of cells observed in the brain separately with the help of filtering techniques in image processing.From Table 1, it is easy to identify that the PSO algorithm is best suitable for the coronal plane compared with axial plane.On taking the average elapsed time of the axial and coronal planes, the results shows that the execution time for axial plane has 17.44990 seconds and for the coronal plane is 15.87230 seconds.
erefore, the result of coronal plane has higher accuracy and the axial plane is less than the coronal plane.From the Figures 5 and 6, the segmentation process shows that the axial plane has still some noise and result is not as good as the coronal plane.But, the coronal plane has clearly segmented the affected region separately and the elapsed time is relatively low.Hence, the PSO algorithm is best suitable and efficient for the coronal plane.

Conclusion
Manual brain segmentation probably is more accurate than fully automated segmentation.However, the major drawbacks of manual image segmentation are time consuming and subjectivity of human segmentation.It is significant to develop a reliable automated segmentation to overcome the drawbacks of manual segmentation.e challenges for automatic segmentation of the MRI brain images have given rise to many different approaches.e methodology carries some of the metrics of automation.
e axial and coronal plane of the image is detected; still sagittal plane automation is not yet finished in this work.Hence, it is concluded that the automation of MRI brain images to detect the brain tumor is based on the criteria of elapsed time and the segmentation level.e process of segmenting the MRI brain image by PSO algorithm is better for the coronal plane than the axial plane.which converts image into a digital form and perform some operation on it in order to get an enhanced image or to extract some meaningful and useful information from it.Image processing is basically of two aspects: one is improving the visual appearance of images to a human viewer and the secondmeasuring some of the features and structures present in the images.The digital image can be optimized for the application by segmenting or altering the appearance of structures within it based on the body part, diagnostic task, viewing preferences etc. Analyzing the images in the computer will help the radiologists and to detect the important or suspicious region in the image to diagnose the disease.
An exciting source of images is the medical field.Imaging plays a unique importance practically in the medical informatics.The field of biomedical analysis is rapidly evolved with the last two decades.Over the years crossed, the major technical challenge in the images is increase in size and its dimensionality.In medical scenario, we have so many algorithms which automatically or semi-automatically detect lesion, diseases and tumors and enhance some of the locations in the images.The research process of medical image analysis has two dimensions one with the patient-centered medical world and the second one is computer-centered medical world.There different imaging modalities such as CT-Computed Tomography, PET-Positron Emission Tomography, MRI-Magnetic Resonance Imaging, Ultrasound and X-Ray [15].The efforts of the research in the medical field is mainly devoted on processing and analyzing medical image to segment the meaningful data such as shape, to detect abnormalities, quantifying the changes in the position and growth of the suspected regions, motion of organs and its volume.
In the medical domain the MRI evolution creates a huge revolution.In 1971, Prof. Raymond Damadianinitials the first concept of MRI.From the day of invention, researchers and doctors keep on improvising and using the MRI scans not only to assist in medical diagnose but also in the research area.An MRI scanner is an important tool to analyze patients and produces the MRI images.The scanner is developed with two powerful magnets, the first magnet aligns the water molecules in one direction either north or south and second one is to generate the quick pulses series in turn on and off.The magnets should be cooled constantly in the temperature of absolute zero (-459.67°F).By traditionally liquid helium is act as cooling agents for the magnets.The MRI scan images are useful to predict uterine abnormalities in women for the treatment of infertility, some types of heart problems, injuries and abnormal behaviors of bone joints, detect the lumps in the liver and abdominal organs, abnormalities in the brain and spinal cord, pelvic pain for women and etc.
Image segmentation is the process of partitioning an input image into set of connected pixels.Technically, image segmentation is refers to the decomposition of a scene into different meaningful regions.Scientifically, the image segmentation is a vision task of hypothetical middle-level.The aim of the image segmentation is to identify the region of interest i.e., to locate tumor and radiation dose calculation in radiation therapy in automatic segmentation of medical images [8].The segmentation is based on the measurements taken from the images based on the texture, color, depth, motion or intensity values of an image.In general, image segmentation is a step by step process to study all the image regions in depth.One of the major applications of image segmentation is to identify the objects in the scene and measuring its shape and size.In this research work, the MRI brain images are analyzed with the FCM algorithm and the results are verified with the histograms values of the images to evaluate the accuracy of the results.This article is organized as follows.Section II deals with the literature review of the variousrelated works, section III discusses about the materials and methods applied in this research work.The results and its discussion are explained in section IV and finally section V concludes the research work.
2. Literature Survey There are many researchers performing research to find the efficiency of tumor affected region to help the radiologists.In particularly, MRI brain image the performance of the FCM is highly important clustering algorithm for its efficiency and effectiveness.Some of the research works for different persons are discussed here from which gives a different perspective of FCM.3D brain tumor segmentation in MRI using fuzzy classification, symmetry analysis and spatially constrained deformable models discussed about the segmentation of brain tumor.Fuzzy classification and approximate brain asymmetry plane based on these two different approaches the tumor detection is carried out and its effective for all different types of tumor [10].M.N.Ahmed etal.are carried out in their research work in bias field estimation and adaptive segmentation of MRI data using a modified fuzzy c-means algorithm.They proposes a novel algorithm to segment the MRI data and evaluate the inhomogeneities intensity of fuzzy logic [1].Current Methods in the Automatic Tissue Segmentation of 3D Magnetic Resonance Brain Images they provide the current methods in the tissue segmentation with the detailed study of each method with the mathematical representation and advantages and disadvantages of the methods.They provide the conventional fuzzy c-means with the two ideas intensity nonuniformity INU and spatial context with the pixel values in clustering process [12].
Automatic Tumor Segmentation Using Knowledge-Based Techniques is research work done by Matthew C. Clarkand etal.The suspected tumor is identified by multispectral histogram analysis and region analysis is used for the intracranial region.They generated a system that automatically identifies tumor segments and labels of glioblastoma-multiforme tumors in the human brain with the help of magnetic resonance images [6].Discrete dynamic contour model with adjacent vertices consists of vertices and edges is the main evaluation criteria to segment thalamus from MRI brain images which is an important neuro-anatomic structure in brain is the research work discussed by LadanAmini and et.al [4].Intensity space map(ISM) is combined with the fuzzy c-means clustering algorithm to segment the color MRI images for tumor detection work.Though the manual segmentation of MRI data is possible, this algorithm segments the muscle regions and its time consuming [19].
Keh-Shih Chuang, Hong-Long Tzeng, Sharon Chen, Jay Wu and Tzong-Jer Chen carried their research work in Fuzzy c-means clustering with spatial information for image segmentation.The spatial information incorporates with the fuzzy c-means algorithm for clustering into the membership function.Each pixel is taken consideration with the summation of spatial function with the neighborhood membership function [5].Another research work proposes a new automatic clustering approach with a hybrid algorithm in the combination of Artificial Bee colony with fuzzy c-means to determine the tumor region.The hybridization of (FCMAB), to segment the MRI brain image extracts abnormal cell growth with the cluster centers [3].The log bias field is stack of spline surfaces, that reduces the spline coefficients of 3-bias field reduces in finding the spline coefficients is effective two-stage algorithm.They proposed algorithm inorder to account the spatial continuity constraints between image volume element.The MR imaging signal is formulated by a multiplicative bias field with INU artifact [11].A Modified Fuzzy C-Means Algorithm for Bias Field Estimation and Segmentation of MRI Data is a research work carried by Mohamed N. Ahmed.The objectivefunctions of the standard fuzzy cmeans algorithm for inhomogeneities and allow voxel the volume element of an image is influences the adjacent neighborhood.Theregularize and biases of the neighborhood pixels leads towards the homogeneous labeling in the piecewise information [2].
Brain Tumor Detection Using MRI Images is done by the research authors of PranitaBalaji Kanade1 and P.P Gumaste.They proposed algorithm which consists of six stage process which includes test images,preprocessing, denoising and SWT, segmentation, feature extraction and SVM/PNN.The algorithm is have higher accuracy and low error rates.The image segmentation algorithm will have following features accuracy, reliability, repeatability, robustness and least dependency [8].Ming Zhao and etal.carriedtheir research work titled as Automatic Threshold Level Set Model Applied on MRI Image Segmentation of Brain Tissue [20].A mathematical representation and proof for Chan and Vese model for the different mean and variance for automatic threshold level set for the image segmentation.The extraction of MR images was developed by threshold level set without edges for the tissues in brain.Dzung L. Pham and Jerry L. Prince evaluated a novel algorithm with the multiplicative intensity inhomogeneities for the obtaining the fuzzy segmentation of images.An iterative algorithm is developed to minimize the objective function of fuzzy c-means [13].

3.Materials and Methods
Dunn was the first person to introduce the Fuzzy C-Means clustering algorithm and it's extended by Bezdek.Duun Clustering is a widely used technique to classify images, that pixels of same group are belongs to one group and differentiated pixels are belongs to some other groups [15].The MRI scans has been taken for the patients for many reasons.To detect and identify bleeding, injury, blood vessels, tumors in brain area.MRI also diagnoses more problem than on X-ray, ultrasound scan or CT scan.Brain is the main central nervous unit which connects all the nerves of the body.It's very important for the patients to diagnose the various problems in the brain.So the detection and prediction of the brain tumor affected region is must and it should be precise for the radiologists to produce accurate results to the physicians.Fuzzy c-means (FCM) clustering algorithm is a commonly used in image segmentation, to assigns pixels of the images into distinct classes based on the features of the image.One of the important domains of Fuzzy C Means is medical image analysis.This paper analysis and detect brain tumor affected region separately with the help of the FCM in the coordination of histogram values of the image.The MRI scanner scans the brain and produces the resultant image in three different planes for the detailed description.The magnetic resonance scanner provides various scanning images of the brain from different directions.Figure 1 shows the different planes of the MRI brain images.One is axial plane which slices the brain and provides the information, the second one is the coronal plane shows the information from the back with the spinal cord and the last one is the sagittal plane.The sagittal plane describes the brain image from the left and right side of the brain anatomy shows in the Fig. 1.The FCM clustering algorithm is an iterative process to produce the cluster centers partition by minimizing the weighted group of squared error objective functions summation.The MRI brain image is segmented by FCM algorithm is carried by many researchers.In this paper, the segmented region pixel values is verified with the histogram of the resultant images.The histogram values of the resultant image is been analyzed.The research methodology of this research work includes four modules.The first module is converting the DICOM images in to some real world images which can be easily handled by any software.The second module is preprocessing the input image by removing noise.The third module is applying the traditional FCM algorithm to the preprocessed images and finding the cluster values to the resultant images and the last module is generation of histogram values and cross examining with the results from the third module to find the accuracy with the coordination of pixels intensity.The resultant images produce the segmented region of the input MRI brain images based on the evaluation of the intensity values of the images.By segmenting the image the cluster center values are been recorded with the help of FCM algorithm.

Gaussian filter
The MRI input brain imagenoises areremoved with the Gaussian filter in the preprocessing stage.The Gaussian noise is an form of white noise.It is caused due to the signals random fluctuations.The white noise is a random signal with spectral densities flat power and it is also called as additive noise [9].An simple and important filtering techniques for images is the Gaussian filtering which is a subtitle of bilateral filtering.
) , ( Where (‫‬ǡ ݂ሻ = ‫(‬ െ ݂ሻ=ԡ‫‬ െ ݂ԡis a suitable measure of distance between the intensity values of ‫‬ and f.The geometric spread ߪ d is the domain is chosen based on the desired amount of low-pass filtering.The Gaussian range filter is insensitive to the overall filters and additives changes are subjected to image intensity.

The Fuzzy C-Means Algorithm
Fuzzy c-means is clustering algorithm to classify the pixels into two or more groups.FCM is mainly used to segment the images.Though, several approaches are exists in the real world for MRI brain image the FCM produces more efficient and effective results than the others.The FCM clustering algorithm works on the image based on some features like intensity values, texture, and pixels regions etc.In this approach, we cluster the image by taking the intensity of the pixels.The merely same intensity values of the pixels will belongs to one cluster and other pixels which may have the same intensity values form the next cluster.FCM clustering algorithm is based on minimizing the weighted square mean error of the objective function The above equation is the weighting function of the FCM clustering algorithm.Centroids cluster of I is in C i and the u value is the range of between 0 and 1; Euclidean distance between the i th centroids and j th data joint is represented by d ij and weighting function is m£(1,) [7].

Objective function of FCM
The procedure convergesthe saddle point of Y m or minimizes the local minima.The fuzzy segmentation of image data is done by the iterative process by optimizing the objective function is carried as follows 1. C and q values are to be set.2. Initialize the matrix M=M(M ij ) 3. Loop counter b=0.

Center vectors calculation is done by
is the termination condition This iteration process is stop when it meets the following condition.The termination criterion with the range between 0 and 1, and k is the number of iteration steps.In the above equations m>1 and x i is data measured in d-dimensional, M ij is the X membership degree in the j cluster and center of the cluster is in the d dimension of R j [7].Theԡ‫כ‬ԡ is expressing the similarity between center and the measured data.The optimization of the objective function is carried out in the segmentation of fuzzy, with the update of the cluster centers and objective functions of the equation(1) [17].The above equations 2 and 3 are the FCM to create clusters of the MRI brain image to segment and separated the tumor affected region based on the intensity values of the image.In the, FCM the C represents the clustering which means creating group of objects which are similar belongs to one group and dissimilar are belongs to different group.

Application Areas of FCM
The applications of clustering techniques are document categorization, customer/market segmentation, scientific data analysis, city planning, land use and in earthquake studies [16].Fuzzy, hard, remote sensing, satellite signal receiving, are the broad classification area of clustering algorithms.The clustering is widely used in many research areas such as data mining, artificial intelligence, fuzzy systems, pattern recognition, machine learning etc.The application of cluster analysis is widely used in chemistry, the systematize chemical and physical properties are analyzed and give detailed report in the field of analytical chemistry.The FCM is been applied in the data mining domain widely.The comparative analysis of FCM with k Medoids for the statistical data points are analysis in the research work and the results are discussed.The experimental results are discussed and effectiveness of the FCM is shown clearly for the distributed data points [18].

Histograms
Histogram is a graphical representation of an image based on the intensity distribution of an image and quantifies the each intensity values based on the number of pixels considered.In this work the detection of affected region is identified based on the intensity level of images using FCM [10].If f be a image it is been represented by m r and m c matrix of pixel integer intensities ranging from 0 to L-1.The set of possible intensity value is 0 to 255.pixels) of number n)/(total intensity with the pixels of (number     The table I shows the computation time values of AXI_IMG_01, AXI_IMG_02 and AXI_IMG_03.The computation time is the CPU execution time of FCM algorithm which not includes the preprocessing time.The computation time is calculated in seconds.Fig. 4(a) shows before segmentation with the full intensity value is peak at the somewhere between 120 and 130.The value is equal to cluster center c1=128.559.After the segmentation the histogram value is peak at the point 10.Fig. 4(b) segmented region intensity value is equal to the cluster center value c2=10.330 of FCM algorithm which shows the accuracy of the segmented region.
The cluster centers of c1 and c2 of the sample data are shown in the table.The table shows the cluster center values of the image before and after segmentation of images.This value is approximately equal to the histogram peak point values for the corresponding images to maintain the accuracy and to achieve adequate results.Diagnosing tumor plays an important role in patients to produce accurate results is a highly an important criteria.The coronal plane is another plane in the MRI brain image which shows some tumor affected region with the back plane of the human.In the axial plane we can detect some types of tumor.The coronal plane sample MRI data set is shown in the        Fig. 16shows the comparative computational time of the axial and coronal plane for the three image data set.The computational time is calculated to discuss about the more efficiency of the FCM clustering algorithm in the real world over the axial and coronal plane.The computational time may differ based on the system specification.The time calculated is depends on the CPU evaluation time of only the fuzzy c means clustering algorithm.The axial plane has taken more time than the coronal plane.The elapsed time is calculated in seconds.In the real world, on diagnosis tumor to the large amount of patient the time consumption is an important criterion.On using the fuzzy c means for coronal plane it require only adequate time, so the FCM is well suitable for the coronal plane than the axial plane.

Conclusion
The prediction of brain tumor is a critical problem in the medical field.The tumor affected region is separated from the MRI brain images by comparing the computational time of the axial and coronal plane and also finding the best fit of the FCM algorithm over the two planes.The structured approach discusses so far will help the physicians to detect the tumor affected region very easily.The preprocessing of images is carried out byGaussian filter method to remove the noisesfrom the images.After preprocessing, the traditional FCM is applied to segment the tumor affected region.The affected region is cross examined with the cluster center values of FCM.A physician can detect the tumor affected region very straightforwardly with the help of result of FCM algorithm.But, the FCM algorithm alone cannot diagnose the tumor for some type brain images.It helps the doctors or radiologists in finding the affected region perfectly to detect the tumor.The proposed approach is more suitable and robust for the coronal plane rather than the axial plane based on the computational time when comparing with the axial plane.The future work is to calculate the area of the tumor affected region by means of intensity based pixel values.

INTRODUCTION
Images are one of the effective ways to reveal information to the world.Immeasurable techniques are used to process the images and used for decision making.In medical ¿ eld, radiological images are processed by Computer Aided Diagnosis (CAD) tool to detect the abnormalities occur in the image and for the automated detection and quantitative analysis.Segmentation is one of the techniques which divulge some information to the user.It is the process of partitioning an image into regions that shares related features and characteristics or grouped to interpret the images.Magnetic resonance imaging (MRI) of brain is examined by performing with several coil types, depending on the MRI unit design and the information provided by it.The imaging protocols of the MRI brain should meet few important criteria, the clinical questions should be answered and they must be completely provides all the necessary information.The MRI much be has to be short as possible to reduce the time the patient has to spend in the magnet and must be reproducible.The imaging protocols should be standard one to maintain its continuity over a period of time.Numerous changes in the MRI protocols should be avoided, the operator of the equipment may confuse.MRI of the whole brain is generally of 3.0 T imaging slices, this thinner slices is superior in accurate detection of brain structure.In general, MRI studies of the brain should be including two weightings and two imaging planes.Some standard sequences of the MRI brain are shown in Table 1.1FLAIR, FLAIR + Gd, DWI/ADC, SWI, T2, T1 r Gd.

FLAIR
x Lesion detection.
x Less sensitive in posterior fossa.
x Usually applied in axial and coronal imaging planes.

FLAIR + Gd
x Indicated for the detection of leptomeningeal disease PD/T2 x Proton density (¿ rst echo) can be used as an alternative to FLAIR x more sensitive for the detection of posterior fossa lesions x T2-WI second echo for detection of long T2 lesions.

T2
x Echo sequence provides information about hemoglobin breakdown and calci¿ cations x Sensitivity effects is proportional to TE and ¿ eld strength.
x Excellent differentiation of gray and white matter to detect the tumor and cell density.
Brain tumors are the formation of new blood vessels to increase the capacity of brain which is also known as angiogenesis.When brain tumor starts growing it outgrows its blood supply, the blood vessels which are developed in this way is irregular, disorganized, abnormal, and tortuous.The tumor region is enhanced after the injection of a gadolinium-chelate, because the contrast agent leaks out the abnormal blood vessels.These abnormalities in tumor blood vessels can be used in MR perfusion imaging.MR perfusion imaging is helpful in identifying high-grade tumor components and glial tumor.The hyper intense areas on FLAIR or T2-WI always correspond to the true tumor margins and the MR imaging also assist in radiation therapy and surgical planning by outlining the tumor boundaries.Image segmentation has many application in all the ¿ elds such as medicine, agriculture, ERP development, toll plaza, industry etc. Methods of processing this images requires segmentation, extraction [9].The medical imaging applications such as diagnosis, study of anatomical structure, treatment planning, quanti¿ cation of tissue volumes, computer integrated surgery, functional imaging data, identi¿ cation of tumor in liver, brain etc.
In this paper, a segmentation method called level set method based on EMO algorithm is introduced.The algorithm takes random dataset of MR brain imaging planes.The datasets are preprocessed with linear high resolution weiner ¿ lter and improve the contrast of the images.The quality of each pixel is evaluated by the objective function of the EM algorithm using the attraction and repulsion operators.The images generated through EMO is again evolved through level set method, contours and surfaces are represented by zero level set usually called as level set function to segment the tumor affected region.Experimental results shows that the performance evidence of the implementation of this novel approach to the image segmentation process.The rest of the paper is organized as follows.In section 2, the related literature review is explained.Section 3 gives a simple description of the standard EMO algorithm and level set method functions are introduced.Section 4 discusses the experimental results and bias ¿ eld, histogram of the MR images.Finally the work is concluded in section 5.

REVIEW OF LITERATURE
In the recent years, so many researchers have developed a new optimization algorithm and level set methods to solve numerous problem of the real world.Some of these new approaches to images are discussed here.A new Hybrid Electromagnetism-like Algorithm for capacitated vehicle routing problems is carried out by the researchers AlkÕn Yurtkuran and Erdal Emel.A modi¿ ed objective function value is utilitzing the random key procedure for EM algorithm to solve the known capacitated vehicle routing problem.The proposed algorithms computational times are high because of the sorting algorithm which uses RKP [17].Ana Maria A. C. Rocha and Edite M.G.P. Fernandes carried their research work to propose a new local search procedure based on a population shrinking strategy with the pattern search method to improve ef¿ ciency.This new process explores in the search space with smaller number of functions [13].Hamid A. Jalab carried his research work to present a novel Emag algorithm for content based image retrieval based on electromagnetism optimization technique.The results shows an average precision value and a signi¿ cant improvement over the traditional CBIR technique [3].A hybrid electromagnetism-like algorithm for single machine scheduling problem research work discuss about the work in population based meta heuristic to solve continuous problems effectively.The convergence and diversity effects are achieved iteratively to solve the problem.The genetic algorithm is the key random key concept for this hybrid algorithm for single machine problems [1].
The multilevel thresholding with the electromagnetic optimization like algorithm forms a new proposed approach MTHEMO.This study explores the study of the two versions of MTHEMO.One is using the otsu function and the second with the kapur criteria.The ef¿ ciency is evaluated by using the STD and PSNR values [12].Chujian et al. discuss the research work titled as An improved electromagnetism-like mechanism algorithm for constrained optimization [18].The study proposes an improving EM algorithm with FAD rules and corresponding charge formula are used as constraint handling techniques.The modi¿ ed algorithm is helpful not only for constrained optimization but also for optimization problems.Ching Hung Lee and Fu-Kai Chaung carried out their research work titled as Fractional-order PID controller optimization via improved Electromagnetism-like algorithm.They propose a new evolutionary algorithm with the improved EM algorithm along with the genetic algorithm technique (IEMGA).It is an evolutionary method and avoids the use of calculus and capable of decreasing the computational complexity [5].
The multiphase level set formulation is generated to avoid the problems of overlap and vacuum; in the set functions of n phase in the piecewise constant case.The proposed model is validating by numerical results in image denoising, segmentation and in Sethian level set method [16].Level set approaches to image segmentation with intensity inhomogeneity discuss a novel level set method.The inhomogeneous objects are modeled by Gaussian distribution with different variances and means with sliding window map to original image.This new methodology is been directly applied to 3 and 7T MR images with the bias corrections [19].Bing Nan Li et al.Discuss about the computed tomography images in the liver tumor segmentation [6].The liver tumor is estimated by fuzzy clustering, and the probabilistic distribution is enhanced by object function and regulates region competition.The new uni¿ ed level set model is an good measurement for the liver tumor segmentation of computed tomography images.The tumor in MRI brain image and the results of K-Means algorithm for the quality of cluster algorithm is discussed in the Identi¿ cation of Calci¿ cation in MRI Brain Images by k-Means Algorithm [11].The author describes the role of clustering and its various applications and techniques in his research work.
Yunjie Chen, Jianwei Zhang and Jim Macione carried their research work titled as An improved level set method for brain MR images segmentation and bias correction.They propose a new region based on active contour model for level set formulation with the bias correction of images.They ¿ rst de¿ ne a localized k means clustering objective function for image intensities.The cluster centers in the objective function have multiplicative factor which estimates the bias of the neighborhood intensity of the images [2].Sukassini and Velmurugan carried their research work in segmentation techniques for the mammogram images [15].The authors discussed various preprocessing techniques for images including morphological operation which gives very quality and effective images as results.They discuss various results of the different research papers by different authors which have greater impact on our novel approach.Distance Regularized Level Set Evolution and Its Application to Image Segmentation is been discussed by Chunming Li and et al.A new approach for level set evolution called distance regularized level set evolution (DRLSE).It avoids the induced numerical errors and thus eliminates the need of initialization.Since, it's a numerical implementation; it requires large time steps to reduce the number of iterations.It's an edge base active contour model for image segmentation and it greatly reduces the computational cost [7].

METHODOLOGY
This paper proposes a hybrid framework that combines EM-like algorithm and Level Set Methods (ELSM) for solving MR imaging problems of brain.To reduce the time complexity and to obtain ef¿ cient quality segmentation the EM algorithm with combined with the level set methods form the new hybrid algorithm ELSM.The purpose of this hybrid framework is to take the advantage of EM with a high diversity population, and level set methods; the combined algorithm works faster.The new combinational algorithm proposed in this paper follows the collective attraction-repulsion mechanism by combining with the level set methods.This new algorithm is composed of two phases: electromagnetism optimization technique and level set methods.The clinical and research applications of MR imaging is relying on segmentation of the affected region in each image to diagnose it.In general this novelistic approach towards the EM like algorithm with the level set methods is to detect the tumor affected region of the brain and it can be applied to other imaging problems.The proposed hybrid ELSM algorithm for the MR brain images is evaluated as follows: Step 1: Converting the general MRI DICOM(Digital Imaging and Communication in Medicine) in to standard image ¿ le formats.
Step 2: Preprocessing the images with adequate linear ¿ lters to remove high signal noise and white noise for the images to achieve effective resolution.
Step 3: Implementing the modi¿ ed Electromagnetic optimization algorithm for the images to highlighting white and contrast of the MR brain image with varying intensities for further detection and segmentation of tumor based on the value of n (2,3,4,5).
Step 4: Initializing the level set methods with the step 3 results and starts segmenting the tumor affected region through the iterative process with the varying levels denoted by the value of k (1,2,….) to detect the tumor affected boundary region separately.
Step 5 : The max and min concepts are used to select the best resultant images and time step of hybrid ELSM algorithm is marked.

A. Electromagnetism optimization algorithm
Electromagnetism-like algorithm (EMO) is a global optimization algorithm which undergoes the concept of electromagnetism law of physics, it's a population based methods.EMO is relatively new meta-heuristic algorithm introduced ¿ rst by Birbil and Fang in the year 2003.The members of the population are guided by the objective function values which follows attraction and repulsion mechanism.In this paper, the second module of ELSM algorithm is implementing the EMO algorithm to segment the images.The idea behind the EMO is to move a particle to the space by the force of the rest of the particles in the population.In the brain tumor segmentation is detection of white matter i.e. particles from the image is the main goal of this research work.In this research work EMO is used highlight the different regions of particles in the image.EMO is an nonlinear optimization problem that involves multiple objective function, in mathematical terms it can be formulated as maximize f(x), x = (x 1 ,……., where f: R n ĺ R is a nonlinear function [12].The solution of EM algorithm can be viewed as a charged particle in search space and its charges are relates with the objective function values.The electromagnetic force exists between the two particles works as with the force of the one particle with more charge will attract the other and the other one will repulse the former.The charges of the particle is based on the objective function values which determines the attraction or repulsion.There are four phases in the EMO algorithm: initialization of the algorithm, calculation of total force, movement along the force direction and the last one is neighborhood search in ¿ nding the local minima [18]. The initialization part, the population is randomly generated in the search space as in the other optimization algorithm.Each particle obeys a uniform distribution in the upper bound and as in lower bound.After calculating with the function value of each particle, the point with the best function value is stored in a best particle.The EMO algorithm needs a local search randomly to gather neighborhood information, we need two parameters to achieve this one is LSITER representing the number of iterations and the other one is G denotes the multiplier of the neighborhood search.The total force determines the detection and step length of movement of each particle.The charge of each point I, which identi¿ es the power of attraction and repulsion of particles, q i is calculated by the formula (2).According to the EM mechanism the particle with the better objective function values attract each other otherwise repel each other the total force of this particles if formulated in the equation (3).
The last stage in the EM algorithm is the movement of particles after evaluating the total force of the particles.According to the equation (4) the movement is calculated and the O is the interval of 0 and 1. Range is a vector which denotes the movement of particle in upper and lower bound which the corresponding direction.The implementation details and the step by procedure of all the four stages are evaluated ef¿ ciently.In the an improved electromagnetism-like algorithm for global optimization research paper discuss the criteria and recommended steps of the lsiter, total force and move function [4].
The main objective of the ELSM algorithm is segmenting the images with the values of n and the resultant image is leveled by the boundary detection of the tumor affected region through the number of level based on the k value.The new EM algorithm is been introduced in order to segment the images based on the intensity value of pixels in the image to detect the tumor.The modi¿ ed EM algorithm is been implemented based on the equation (7) in this the value of q i is the charge of each point (pixel) in the image and the power of n denotes the cluster value of segmentation.By calculating the neighborhood pixel value the x best and x last values are obtained.The equation ( 8) F k represents the total force calculation of the modi¿ ed ELSM algorithm.The q i and q j represents the x and y values of the pixels and x i represents the intensity values of the particular pixels.By using the equation ( 7) and ( 8) with the attraction and repulsion mechanism the little contrast intensity values of the images are obtained by the using the cluster values of n the best resultant image is evaluated for the enhancement of image.
The level set methods is been modi¿ ed and the tumor affected region is separated by modi¿ ed the original.The modi¿ ed level set method overcomes the disadvantages and solves the tumor detection problem in combining with the modi¿ ed EM algorithm.On ¿ nding the number of level segmentation in detecting the boundary region the value of k are being processed with modi¿ ed level set methods.The changes in the level set methods are described in the equation (9) with the level values k.The equation 5 has been modi¿ ed with initial at t = 0, it would be possible to know at any time t with motion equation with the applying chain rule give us the equation (9).Ongoing through number derivations the value of x t will be denoted as in the equation (10) where k denotes the level value.
( ( ), ) x t t t wI w = 0 (9) where In general the boundary of the pixels are identi¿ ed by active contours, its computer generated curves that move within the image to ¿ nd the object boundaries.The intensity I of the tumor tissue should take a speci¿ c value with the physical property.In our hybrid algorithm the modi¿ ed EM algorithm is partially segments the tumor with the various n values.The segmented region of tumor is estimated with the active contour is slowly varying across the image domain with the various levels of k.The image intensities are approximately same within each class of tissue.We propose a method which divides this small region pixel intensity separately [2].The level set functions are evaluated and active contours are applied to detect the affected region separately.The general active contour functions are implemented as a last phase of the third module of the ELSM algorithm.

EXPERIMENTAL RESULTS AND DISCUSSION
To evaluate the effectiveness of the proposed algorithm ELSM, experiments are carried out on a subset of abnormal MR brain images from the real world.The image dataset is collected from Swami Vivekananda Diagnostic Center scan center in chennai.Different patient's data from different age and places are taken and the hybrid algorithm is evaluated.The ELSM algorithm main aim is to detect the tumor affected region separately from the normal tissue of the brain to support the radiologists and to assists the physicians.The algorithm works on by computing the coef¿ cient of joint variation of tissue between grey and white     The result of the intensity highlighted image is drives through the third phase of ELSM algorithm.In this last phase, the tumor affected region of the segmented images is evaluated through the level set function with the active contours.By carrying the inner and outer layer of the boundary region in the images, iteratively we will reach the tumor affected region separately.The boundary region calculation of the image in the ELSM algorithm is been devised through with the active contours as described in section 3.By using, the active contours with the level set function the affected region of the MR image is been identi¿ ed in the ELSM algorithm.The various levels of LSM are denoted by the value of k.By taking the value of k from 1,2 and 3, the n value results from the second phase is evaluated.Since the range of third step of the ELSM algorithm is basically starts with the value starts from 1.The resultant of image of each n values is been evaluated based on the each k value is carried out and the results are shows in the ¿ gure 5.This ¿ gure 5 shows the top down approach of our ELSM algorithm based on n and k values.The ¿ rst stage of the ¿ gure shows the preprocessed image and the second stage shows the modi¿ ed EM algorithm results based on the cluster value of n.The last stage describes the various levels of the boundary detection for the values of k.By manually observing the best ¿ t value for n and k are analyzed.By changing the values of k the resultant images are observed.Based on the observation from the table 2 the resultant images are analyzed manually.On minimizing both the value of n and k the detected region is small in size.On maximizing both the values of k and n the detected region include the artifacts detail of image.Minimizing the value of n and maximizing the value of k results feasible solution for the image.By minimizing the value of n and maximizing the values of k the best resultant image is obtained by observing the images.Figure 6 shows the tumor affected region when cluster value of n = 2 with the changing level value of k 5.The ¿ gure 6(a) shows the images partially and 6(b) represents the region with missing some regions and 6(c) detects the entire tumor affected region with the minimum of the n value and the maximum of the k value.The best resultant image is obtained from the value of n = 2 and k = 3.Table 3 shows the elapsed time for the changing n and k values, the value of [2,1] will be small but the best resultant image as in table 2 is [2,3].The difference between this two min and max value of elapsed time is small, so its negotiable.Its worth noting that our algorithm allows À exible initialization of the level set functions.The initial contour of the image can be inside, outside, or cross object boundaries.The proposed ELSM hybrid algorithm performance tests were implemented using Matlab R2008a on i5 processor with 2.2 GHZ, 4GB DDR3 memory, 64-bit, and Windows 7. The ¿ gure 7 shows the elapsed time in seconds during the implementation of ELSM algorithm with the number of cluster and level values taken for the dataset.From the ¿ gure 7 the x axis represents the cluster values n and the k value number of levels and the space occupies and y axis represents the time taken.From the ¿ gure it is observed that the when x = 3 the y value is minimized.Figure 8 shows that the space occupied by the resultant images the n and k values are minimized; meantime the space is also reduced.By analysis in the all the criteria the best resultant images is obtained when n = 2 and k = 3.
In the medical ¿ eld, to segmentation of tumor or it's the detection of white matter of brain from the given MR images is an important task even in the modern era.By the proposed hybrid ELSM algorithm, both the objectives taken for analysis are achieved in ef¿ cient manner.The time complexity and space complexity are also minimized by this approach.The resultant image is validated with the medical expert and thus provides 90% accuracy.The ELSM algorithm allows the use of relatively large time steps to signi¿ cantly reduce iteration number in the last phase of the algorithm.The ef¿ ciency and accuracy of the ELSM algorithm will ¿ nd its utility in more application area of the image segmentation.

Figure 1 .
Figure 1.Different planes of MRI brain image.
Feedforward optimized by hybridization of PSO and Artificial Bee Colony (ABC) abnormal brain detection research work carried by Shuihual Wang et al. e work starts with an automatic classification system based on Artificial Bee Colony (ABC) and Particle Swarm Optimization (PSO).
and 3.It represents the change in the value of n(segmentation level) starts from 2, 3, 4, 5, 6, 7 and 8 on the axial plane and coronal plane respectively.

Figure 2 .
Figure 2. PSO algorithm in the axial plane.

Figure 3 .
Figure 3. PSO algorithm in the coronal plane.

e
third stage is by taking the resultant images of the axial and coronal plane,with the help of the elapsed time.e best resultant image is calculated by the elapsed time of the plane either axial or coronal, n t represents the total number of segmentation level and n denotes the segmentation level.
In the Gaussian filtering function both the closeness function and similarity function are the Gaussian function ,661 3ULQW ,661 2QOLQH 79HOPXUXJDQ HW DO ,QWHUQDWLRQDO -RXUQDO RI (QJLQHHULQJ DQG 7HFKQRORJ\ ,-(7 '2, LMHWYL 9RO 1R 'HF -DQarguments of the Euclidean distance.The closeness function is described in the equation 5 and the similarity function is illustrated in the equation 6[14].where d(ߦ,x) = d(ߦ-x) = ԡߦ െ ‫ݔ‬ԡ is the Euclidean distance between ߦ and x.

Fig. 2 .
Fig.2.Input abnormal images of the axial plane.The second step is preprocessing starts with 3 MRI image for three different persons AXI_IMG_01 by removing the noise.The noises in the images are removed with the gaussian filtering techniques.Some of the pixels could be affected some form of white or some other noise so the removal of noise by gaussian technique will produce more standard resultant image.Since, the MRI images are black and white the gaussian filterin gtechnique might be correct choice inorder to get a clartity of the image.The results of the preprocessing module is shown in the Fig. 3.The third step is implementing the fuzzy c-means algorithm with the steps are evaluated as in the objective function of FCM.The equations of the FCM are implemented IBM machine with intel®core @ Duo processor and 8 GB RAM, running Windows7 operating system.The algorithm was developed via Matlab (R2008a).In this work, the segmentation of images is based on the intensity values of pixels in the imaging plane.The FCM is a clustering algorithm which gives the resultant images or values based on the cluster center of the pixels.The default segmentation level of the FCM algorithm set as n=2 are shown in the Fig. 3(b) and 3(c).The images are been segmented based on the intensity values of pixels; however the cluster center produces the highest point of intensity values of the image.The last level, separate the brain tumour affected region shown in the Fig. 3(d).

Fig. 3 .
Fig.3.Preprocessing and FCM result of AXI_IMG_01.The last step in this research work is cross-examing the cluster centers of FCM algorithm with the intesnity values of the resultant image with the histogram graph.The diagnoziation of brain tumor is highly important.We are segmenting the MRI brain images based on the pixels intensity values, so on reexaming the resultant values with the histogram graph give more adequate results.The resultant image of the FCM is compared with the histogram values of the image before and after segmentation.Since, the segmentation is based on the intensity values, the cluster values gives the intensity range of the image.With this, the histogram graph are been compared to achieve the correctness of the image.the validation of the resultant image is been

Fig. 4 .
Fig.4.Histograms of FCM forAXI_IMG_01The axial plane image AXI_IMG_02 sample data is implemented as same as the AXI_IMG_01.The FCM clustering algorithm produces the same result as with the coordination of histogram.The cluster center values of AXI_IMG_02 is c1=109.0320and c2=10.330.Fig.5shows the preprocessing and various steps of AXI_IMG_02 and separated brain tumor affected region separately.Cluster center values are compared with the histogram graph intensity of the AXI_IMG_02 before and after segmentation are shown in the Fig.6.The process is implemented for AXI_IMG_03 with the cluster centers c1=128.5591and c2=13.4650and the results are shown in the Fig.7.andFig.8.

Fig. 9 .
COR_IMG_01 data is also preprocessed with the Gaussian filter technique.The cluster values of COR_IMG_02 is c1=40.0859and c2=182.6475and the results are shown in the Fig. 10.and histogram graph of the image are shown in the Fig. 11.

Fig. 13 .
Fig.13.Histograms of COR_IMG_02 In the coronal plane nearly 20 images are taken as input among that 3 image with their results are discussed here.The COR_IMG_02 is done with the preprocessing module and the traditional fuzzy c-means clustering algorithm is applied and the results are compared with the histogram values which are shown in the figures Fig. 12and Fig. 13 with the two cluster center values as c1= 26.7629 and c2=138.3620.The COR_IMG_03 has undergone the same step of process as in the structured approach and produces the results as shown in the Fig. 14. and Fig. 15.With the cluster center values in the coordination of the histogram.The computation time for the three image of the coronal plane with their cluster center values obtained from the proposed approach is shown in the Table II. .

Figure 1 :
Figure 1: Shows the abnormal MR Data

Figure 2 :
Figure 2: Results of traditional EMO and level set methods Figure 4 shows the second phase of ELSM algorithm based on the values of n.By increasing the value of n the images are changing as per its intensity values.

Figure 4 :
Figure 4: Shows the second phase of ELSM algorithm based on the values of n

Figure 5 :
Figure 5: Shows the top down approach ELSM algorithm

Figure 6 :
Figure 6: Tumor affected region when n = 2 with the changing k value (a) shows the results of the image when n = 2 and k = 1.(b) shows the results of the image when n = 2 and k = 2 (c) shows the results of the image when n = 2 and k = 3 Number of clusters (Value of ) n Level Values (Value of ) kElapsed time in seconds

Figure 7 :Figure 8 :
Figure 7: Elapsed time of the dataset 1. e methodology implements the PSO algorithm with some image segmentation techniques.It includes four stages to obtain the affected brain tumor region.e main aim of this paper is to extract the brain tumor separately. is work detects the MRI image, whether it is benign or malignant.e research paper method includes four stages.e first stage is to convert the Digital Imaging and Communication in Medicine (DICOM) file format into image file format.e second stage is implementing the PSO algorithm with the change in the value of n taking n=2 as default value.e third stage is based on the elapsed time selecting the best resultant image from the segmented images.

Table 1 .
Elapsed time of axial and coronal plane

Table I :
Computational time of the axial plane.
MRI IMAGE DATA SET CLUSTER CENTER -C1 CLUSTER CENTER -C2 RUN TIME (sec)

Table II :
Computational time of the coronal plane