International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847

Detection of Abnormal Tissue Growth in MR Imaging using Biogeography Based Optimization

Gaganpreet Kaur1 ,Dr. Dheerendra Singh2 ,Harpreet Kaur3
Assistant Professor, Sri Guru Granth Sahib World University, Fatehgarh Sahib 2 Professor & Head, SUSCET, Tangori 3 Student Masters Of Technology, Sri Guru Granth Sahib World University, Fatehgarh Sahib
1

Abstract
The Medical image segmentation is an essential tool in visual surveying and analyzing magnetic resonance images and solving the problems in medical imaging. Many algorithm have been elaborated for medical image segmentation. In this paper we have used a biogeography based optimization. Biogeography is the study of the distribution of animal and plants over time. We have applied a BBO on the MRI to detect the abnormal tissue growth. We have detect cancer in this work, for cancer detection images that are used in our work is Aorta, Coronal, Transverse. This technique has main characteristic it does not involve reproduction of the generation. The total time taken to detect abnormal tissue growth in our work is from 3.89 to 4.08 in seconds.

Keywords: Magnetic resonance imaging (MRI), image segmentation, Biogeography Based Optimization.

1. INTRODUCTION
The Segmentation involves portioning an image into set of homogenous and meaningful regions, such that the pixel in each partitioned region possesses an identical set of properties. Segmentation can be used for object recognition, image compression, image editing [1]. The quality of the segmentation depends upon the digital image. Segmentation is the process of dividing an image into distinct regions with property that each region is characterized by unique feature such as intensity, color etc .Further it refers to the process of dividing a digital image into multiple segments such as a sets of pixels, also known as super pixels [2]. The main objective of segmentation is to transform the representation of an image into meaningful image that is more understandable and easier to analyze [2]. Segmentation is basically a collection of methods that allowing Image segmentation is a technique which uniquely identifies pixels or close parts of image which share certain visual characteristics. Medical image is the technique and process used to create images of the human body. In medical image processing segmentation has been used for various purposes like: surgical planning, heart image extraction from cardiac cine angiograms, detection of tumors, measuring tumor volume, detection of micro classification on mammograms, detection of the coronary border in angiograms, its response to therapy, automated classification of blood cells, etc. Image segmentation may be defined as a process of assigning pixels to homogenous and disjoint regions which form a partition of the image that share certain visual characteristics[2]. Image segmentation is used to locate and find objects and boundaries (lines, curves, etc.) in images[2]. It basically aims at dividing an image into subparts based on certain feature. Features could be based on certain boundaries, contour, color, intensity or texture pattern, geometric shape or any other pattern [2]. It provides an easier way to analyze and represent an image. Practical application of image segmentation range from filtering of noisy images, medical applications (Locate tumors and other pathologies, Measure tissue volumes, Computer guided surgery, Diagnosis, Treatment planning, study of anatomical structure), Locate objects in satellite images (roads, forests, etc.), Face Recognition, Finger print Recognition, etc. 1.1 Requirements that good image segmentation meets 1. Every pixel in the image belongs to a region 2. A region is connected: any two pixels in a particular region can be connected by a line that doesnt leave the region 3. Each region is homogeneous with respect to a chosen characteristic. The characteristic could be syntactic (for example, colour, intensity or texture) or based on semantic interpretation 4. Adjacent regions cant be merged into a single homogeneous region 5. No regions overlap

2. MAGNETIC RESONANCE IMAGING

Medical image is the technique and process used to create images of the human body. In medical image processing segmentation has been used for various purposes like: surgical planning, heart image extraction from cardiac cine angiograms, detection of tumors, measuring tumor volume, detection of micro classification on mammograms, detection of the coronary border in angiograms, its response to therapy, automated classification of blood cells, etc. The advances in image technology, diagnostic imaging has become an indispensable tool in medicine today. X-ray angiography (XRA), magnetic resonance angiography(MRA), magnetic resonance imaging(MRI), computed tomography(CT), and other

Volume 2, Issue 8, August 2013

Page 74

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847
imaging modalities are heavily used in clinical practice[3]. Such images provide complementary information about the patient[3]. We have applied a Biogeography Based Optimization technique on MR images. The magnetic resonance imaging of the body to performed to get structural detail of the brain, liver, chest, abdomen and pelvis which help in diagnosis or monitoring the treatment. The different types of Magnetic resonance imaging are: (1) Brain Magnetic resonance imaging: MR is generally more sensitive in detecting brain abnormalities during the early stages of disease, and is excellent in early detection of cases of cerebral infarction, brain tumors, or infections. MR is particularly useful in detecting white matter disease. (2) MR Liver imaging: MR provides outstanding intrinsic soft contrast that can enhance subtle differences between normal and pathologic tissues and tissues of different histologic subtypes. (3) Chest Magnetic resonance imaging: The chest MRI is used to detect following disorders: thymus tumor, lung masses, esophageal mass, other masses (aggregations of cells) or tumors of the chest, abnormal lymph nodes, swollen glands and enlarged lymph nodes in any location of the chest, staging of tumors including invasion of blood vessels, alveolar bullae (COPD), bronchial abnormalities, bronchiectasis, cystic lung lesions, pleural abnormalities, including thickening or pleural effusion, abnormal pulmonary vessels, aortic stenosis, etc. (4) Abdominal Magnetic resonance imaging: The Abdominal MRI may reveal many medical conditions, including: abscess, acute tubular necrosis, adrenal masses, cancer, enlarged spleen or liver, gallbladder or bile duct problems, gallstones, bile duct stones, hemangiomas, kidney infection, kidney damage, lymphadenopathy, obstructed venacava, pancreatic cancer, tumor of the gallbladder, abdominal aortic aneurysm, ovarian cancer, etc. 2.1 Advantages and Disadvantages of MRI: Advantages are: 1. It has an excellent capability for soft tissue imaging 2. It has very high resolution of the order of 1mm cubic voxels 3. It has high signal to noise ratio 4. Multi channel images with variable contrast can be achieved by using different pulse sequences; this can be further utilized for segmenting and classifying different structures. Disadvantages are: 1. MR acquisition takes considerably longer time as compared to CT and 2. In case of MR it is more difficult to obtain uniform image quality.

3. OPTIMIZATION M ETHODS
(a) Ant colony optimization: Ant colony optimization based on the pheromone deposition of ants. ACO is a Meta heuristic that can be used to refine method applicable to a wide set of problems with few modifications. The ant based clustering algorithms are based upon the brood sorting behavior of ants. (b) Particle Swarm Optimization: Particle swarm optimization is based on the swarming behavior of birds, fish, etc. In this each particle learns from its neighbors and adjusts itself accordingly. PSO has no crossover and mutation operator. In PSO the solutions are known as particle which fly through the problem space.The particles are also called swarm. In each iteration, the particles are updated using two values one is pbest and other is gbest. (c) Genetic algorithm: The genetic algorithm based on the natural selection in biological evolution. The GA reproduce children by crossover, namely their solution disappear at the end of each generation. (d) Evolutionary strategy: The evolutionary strategy based on multiple parents contributing to offspring. The most fit offspring become the next generation of parents. The ES reproduce children by crossover, their solution disappear at the end of each generation. (e) Differential Evolution: The DE is based on the simple difference-based modification of solutions. The DE its solution directly, but changes in a particular DE solution are based on difference between other DE solutions. (f) Population Based Incremental Learning: PBIL is an optimization algorithm and an estimation of distribution algorithm, This is a type of genetic algorithm where the genotype of an entire population is evolved rather than individual member. (g) Biogeography Based Optimization: BBO is population based optimization algorithm it does not involve reproduction or the generation.

4.BIOGEOGRAPHY BASED OPTIMIZATION

Biogeography Based Optimization is a population based evolutionary algorithm(EA) motivated by migration mechanism ecosystem. It is based on the mathematics of biogeography. In BBO, problem solutions are represented as islands, and the sharing of features between solution is represented as emigration and immigration.(BBO satellite). Biogeography Based optimization was first presented in December 2008 by Dan Simon[4]. The mathematical model of biogeography describes how species migrate from one island to another, and how species arise. The geographical areas that are well suited for the residence of biological species are said to have high suitability

Volume 2, Issue 8, August 2013

Page 75

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847
index(HIS) and the variables that characterize habitability are called suitability index variable(SIV).(medical image quantization). SIVs can be considered the independent variables of the habitat, and HSI can be considered the dependent variable[4]. Habitats with a high HSI tend to have a large number of species, while those with a low HSI have a small number of species. Habitats with a high HSI have many species that emigrate to nearby habitats, simply by virtue of the large number of species that they host. Habitats with a high HSI have a low species immigration rate because they are already nearly saturated with species. Therefore, high HSI habitats are more static in their species distribution than low HSI habitats. By the same token, high HSI habitats have a high emigration rate; the large number of species on high HSI islands have many opportunities to emigrate to neighboring habitats. Habitats with a low HSI have a high species immigration rate because of their sparse populations. This immigration of new species to low HSI habitats may raise the HSI of the habitat, because the suitability of a habitat is proportional to its biological diversity. However if a habitats HSI remains low, then the species that reside there will tend to go extinct, which will further open the way for additional immigration. Due to this, low HSI habitats are more dynamic in their species distribution than high HSI habitats. A good solution is analogous to an island with a high HSI, and a poor solution represents an island with a low HSI. High HSI solutions resist change more than low HSI solutions. By the same token, high HSI solutions tend to share their features with low HSI solutions. Poor solutions accept a lot of new features from good solutions. This addition of new features to low HIS solutions may raise the quality of those solutions. We call this new approach to problem solving biogeographybased optimization (BBO)[4]. 4.1 BBO basically depends upon the following terms: Migration: The BBO migration strategy in which we decide whether to migrate from one region to other or not. The migration rates of each solution are used to probabilistically share features between solution. BBO migration is used to change existing habitat. Migration in BBO is adaptive process. It is used to modify existing islands. Migration stages arises when LSI occurs. When species are less compatible with their habitat they migrate. Mutation: Mutation is a probabilistic operator the randomly modifies a solution feature. The purpose of mutation is to increase habitat among the population. For low value solution, mutation gives them a chance of enchancing the quality of the solution, and for high fitness value solution, mutation try to improve the

Figure 1. Species Model value as compared to the previous value. The implemented mutation mechanism is problem dependent. In which a new regions are created by hybrid the other the region. Certain Features of Biogeography Based optimization:1. BBO has a way of sharing information between solutions. 2. BBO solutions survive forever (although their characteristics change as the optimization process progresses) 3. BBO solutions do not necessarily have any built-in tendency to cluster. 4. It does not require a priori knowledge of the number of partitions in the image.

5. PROPOSED BBO BASED SCHEME FOR IMAGE SEGMENTATION

BBO can be better to detect abnormal growth of tissues as compare to other optimization method. BBO is a population based optimization algorithm it does not involve reproduction or the generations. BBO is uniquely a biology technique that we used for image segmentation. BBO is very fast algorithm. BBO is an application of biogeography to EAs. It is modelled after the immigration and emigration of species between habitats to achieve information sharing. For each generation, BBO uses the fitness of each solution to determine its immigration and emigration rate. BBO has a way of sharing information between solutions. Its solution survives forever. It does not require a priori knowledge of the number of partitions in the image. It yields regions, more homogeneous than the existing method even in presence of noise. So, BBO based segmentation is used in this work with medical images to help in object detection.

Proposed algorithm of Biogeography Based Optimization

Volume 2, Issue 8, August 2013

Page 76

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847
Step 1: Take MRI image Step 2: Convert it into grayscale image Step 3: Initialize the BBO parameters Step 4: Initialize a random set of habitat, each habitat corresponding to a potential solution Step 5: Define scale, Smax, immigration rate() and emigration rate(), HSI Step 6: Pi(n+1)= Pi(n)=r1(PHSIi-Xi)+(r2(Hhsi)-Xi) (use for habitat(image in our case)outward, if habitat HIS is high then the pixels are migrate and if low HIS then pixels are immigrate) Step 7: Xi(n+1)=Xi(n)+Pi(n+1) Step 8: After 3, According to random migration & immigration value new matrix is obtained. Step 9: To detect ROI, segment the image on basis of immigration and migrate value Step 10: After 4, segment image and must compare the ROI part of the image Step 11: With different scale and different value new circle or Hough is obtained Step 12: end Based on step (6) (7) the pixels can migrate to habitat having high HIS, otherwise the pixel values immigrate. The Flowchart of the proposed algorithm is shown in Figure 2. By following this algorithm we reached the target efficiently.

Figure 2: Flow chart of proposed scheme The following criterion is follow for image segmentation using Biogeography Based Optimization: The proposed algorithm is implemented with Matlab 7.5.0. In this work first take MRI image. Edge detected using canny edge method. The canny method finds edges by looking for local maxima of the gradient of image. The method uses two thresholds, to detect strong and weak edges, and includes the weak edges in the output only if they are connected to strong edges. This method is therefore less likely than the others to be fooled by noise, and more likely to detect true weak edges. Hough circle used to detect circle, The Hough Circle used to detect features of the image. Function of Hough Circle used in this work like this [y0detect, x0detect, Accumulator]= Hough circle(Imbinary, r, thresh, region) Y0detect= row coordinate of detected circle X0detect= column coordinate of detected circle Accumulator= the accumulator array in Hough space Imbinary= binary image, r= radius, thresh= threshold value, region= region to search for circle centers with in (x, y, w, h). In this work we have take values of these parameters is r=45, thresh= 4. The Region of Interest used to filter out the portion of the image. Apply Biogeography Based Optimization algorithm and initialize all parameters of BBO. Segment the object that we want to detect in the MR Images.

Volume 2, Issue 8, August 2013

Page 77

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847 6. IMPLEMENTATION
In this work we have implemented Biogeography Based Optimization to segment the MR imaging. We have detect a cancer in medical images. Different parameters are implement of BBO, the population size= 50, Generation limit=100, Mutation probability=0( for small number of population). We have checked the performance of different function ( Rastrigin, Schwefel 2.26, Sphere, Step, Griewank ) at different population size, for mean min cost and mean generation count at different samples by BBO algorithm mean improvement segment generation and mean improvement amounts checked it gives better results. So we have applied the proposed BBO algorithm at MR imaging for segmentation. The implementation of Biogeography Based Optimization done in MATLAB 7.5.0.

Figure 3: Implementation of Biogeography Based Optimization

7. RESULTS
The proposed algorithm is used for segmentation of MR images using Biogeography Based Optimization .Different images are used in the following results to detect cancer in the MR images, the images are Transverse, Aorta, Coronal. The Transverse images are muscles images. The aorta images are that supplies all of the systemic circulation, which means that the body except for the respiratory zone of the lung gets it blood from the aorta. It supplies the blood to neck, heart, arm. The coronal images are those that are divided the body of into vertical and dorsal(belly and back) section. The Table 1 shows the time for segmentation using proposed BBO algorithm, the time taken by this algorithm very less to detect cancer at proper position in MR imaging. Table 1: TIME TAKEN BY BBO BASED MR IMAGING SEGMENTATION IMAGES Name TIME IN SEC Transverse .jpg Coronal .jpg Aorta .jpg 4.04 4.08 3.89

Figure 4. (a)Transverse original image b) Edge detection using canny method c) Hough circle d) Segmented image e) 3D view of segmented image

Volume 2, Issue 8, August 2013

Page 78

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847

Figure 5. a)Coronal original image b) Edge detection using canny method c) Hough circle d) Segmented image e) 3D view of segmented image

Figure 6. a)Aorta original image b) Edge detection by canny method c) Hough circle d) Segmented image e) 3D view of segmented image In earlier papers, work has been done on color images but in this work we used MRI images instead of color images and checked the abnormal tissue in correct position and detect in very less time by Biogeography Based Optimization. By this technique easily detect the proper area of cancer. This proposed algorithm gives better results as compare to other optimization methods. The output of the above algorithm can be seen in above figures.

7. CONCLUSION
Segmentation is a collection of methods allowing interpreting close parts of the image as objects. BBO is a biogeography technique used for image segmentation which provide more accurate segmented images as compared to other optimization method. In this paper, We have used new approach using BBO based segmentation technique on medical images to improve the quality of detection of object which help to detect abnormal tissue growth. In this work the MRI database is used for testing, the major objective of this algorithm is to detect cancer from different parts of the human body that images shows in results. Similarly, this proposed approach can achieve better results than previous approach that uses image segmentation techniques like, ACO, GA, DE, PSO, ES.

ACKNOWLEDGMENT
This thesis work is based on Biogeography Based optimization technique to segment the MR images by using MATLAB. The Main objective of this thesis is to study and implement the Biogeography Based Optimization for MR images segmentation, the images[4] [5] [6]. The basic software required are Window 7, MATLAB version 7.5.0, Image processing toolbox. I would like to thank my guide Mrs. Gaganpreet Kaur, SRI GURU GRANTH SAHIB WORLD UNIVERSITY, FATEHGARH SAHIB for her valuable advice and healthy criticism throughout my thesis which helped me immensely to complete my work successfully. I would like to thanks for the authors of all those books and papers which I have consulted during my thesis work as well as for preparing the report.

REFERENCES
[1] Bhavna Sharma, Prof k. Venugopalan, Performance Comparison of Standard Segmentation Methods for Identification of Hematoma in Brain CT Scan images, International Journal of Computer Engineering & Technology(IJCET) Volume 3, Issue 1, January- june(2012), pp.126-134.

Volume 2, Issue 8, August 2013

Page 79

International Journal of Application or Innovation in Engineering & Management (IJAIEM)

Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com Volume 2, Issue 8, August 2013 ISSN 2319 - 4847
[2] Surbhi Gupta, Parvinder S. Sandhu, Implementing Color Image Segmentation Using Biogeography Based Optimization, International Conference on Software and Computer Applications IPCSIT Vol.9(2011) IACSIT press, singapore. [3] Mrs. P. vijayalakshmi, Segmentation of Brain MRI Using K Mean Clustering Algorithm, International journal of engineering trends and technology-May to June Issue 2011. [4] Dan Simon, Senior Member, IEEE, Biogeography Based Optimization, IEEE Transaction On Evolutionary Computation, Vol. 12, No. 6, DECEMBER 2008. [5] Sushil Kumar, Tarun Kumar Sharma, Millie Pant, A.K.Ray, Adapative Artificial Bee Colony for Segmentation of CT lung images, International Confrence on Recent Advances and Future Trends in Information Technology, Proceeding published in International Journal of Computer Application(IJCA). [6] Aniruddha Bhattacharya, Member, IEEE, and Pranab Kumar Chattopadhyay, Hybrid Differential Evolution with Biogeography Based Optimization for Solution of Load Dispatch, IEEE Transaction on Power system, Vol. 25, No.4, November 2010. [7] Urvinder Singh, Tara Singh Kamal, Design Of yagi-Uda Antenna using Biogeography Based Optimization, IEEE Transaction on Antennas Propagation, Vol.58, No.10, October 2010. [8] V.K.Panchal, et.al, Biogeography Based Satellite Image Classification, International journal of Computer Science and Information Security, Vol .6, No.2, 2009. [9] Dan Simon, Mehmet Ergezer, Population Distribution in Biogeography Based Optimization Algorithms with Elitism, Proceeding of The 2009 IEEE International Conference on Systems, Man and Cybernetics San Antonio, TX, USA October 2009. [10] Monica sood, Mandeep Kaur, Shortest path Finding in Country using Hybrid Approach of BBO and BCO, international journal of computer application (0975-8887), Volume 40- No.6, February2012. [11] S. Shenbaga Ezhil, Dr. C. Vijayalakshmi, Prediction of Colon Rectum Cancer Survivability Using Artificial Neural Network International journal of computer engineering and technology, Volume 3,issue 1, January-june(2012). [12] J. Umamaheswari, G. Radhamani, A Hybrid Approach for DICOM Image segmentation using Fuzzy Technique, International journal of fuzzy logic system, Vol.2, No.3, July 2012. [13] R Venkateswaran, S Muthukumar, Genetic Approach on Medical Image Segmentation by Generalized Spatial Fuzzy C- mean Algorithm, 2010 IEEE International conference on computational intelligence and computing research. [14] A. Umaamaheshvari, K. Prabhakaran, Biogeography for Watermarking Medical images, IOSR journal of engineering(IOSRJEN), Vol.3, Issue 1(Jan.2013), V4, PP31-35. [15] Arthanari Rajendran, Raghavan Dhanasekaran, Brain Tumor Segmentation on MRI Brain Images with fuzzy clustering and GVF snake Model, INT J COMPUT COMMUN, ISSN 1841-9836, Vol.7(2012), No.3, September, pp.530-539. [16] Dr. H. B. Kekre, Detection of Tumor In MRI Using Vector Quantization Segmentation, International Journal of Engineering Science and technology, Vol.2(8), 2010, 3753-3757. [17] Dr. H. B. Kekre, Dr. Tanuja Sarode, Image segmentation of MRI Image using KMCG and KFCG Algorithms, 2nd International conference and workshop on emerging trends and technology (ICWET) 2011, Proceeding published by international journal of computer application(IJCA). [18] Gaganpreet Kaur, Dr. Dheerendra Singh, Pollination Based Optimization For Color Image Segmentation, International Journal of computer Engineering and & technology, Volume.3, Issue 2, July-September(2012), pp 407414. [19] Er. Krishma Bhuchar, Er.Rekha Rani, Performance Evaluation of biogeography based image segmentation, UACEE international journal of computer and its application- Volume 2: Issue 1[ISSN 2250- 3765]. [20] Rajeshwar dass, Priyanka, Swapna Devi, Image Segmentation Technique, IJECT Vol.3, Issue 1, Jan-March 2012. [21] Surbhi Gupta, Gurpreet singh Sandhu, Implementing Color Image Segmentation Using Biogeography Based Optimization, International conference on computer and communication Technologies(ICCCT2012) May 26-27, 2012 Phuket. [22] Aniruddha Bhattacharya, Member, IEEE, and Pranab Kumar Chattopadhyay, Biogeography Based Optimization for Different Economic Load Dispatch Problems, IEEE Transaction on power systems, Vol. 25, NO. 2, MAY 2010 [23] Marsil De A. Costa e Silva, Leandro dos S. Coelho, Multiobjective Biogeography Based Optimization on predator based approach, IEEE Transaction on Magnetics, Vol.48, No.2, February 2012.

Volume 2, Issue 8, August 2013

Page 80