560 by ides editor

Poster Paper Proc. of Int. Colloquiums on Computer Electronics Electrical Mechanical and Civil 2011

Segmentation of Medical Images Using Ant Colony Optimization Anjum Qureshi1, Shilpa Kamdi2 Dept. of Electronics Engg., Rajiv Gandhi College Of Engg. Research & Technology,Chandrapur. 1 anjum_nazir@rediffmail.com 2. shilpa_kamdi@rediffmail.com understand the image feature. Since edges often occur at image locations representing object boundaries, edge detection is extensively used in image segmentation. Medical image edge detection is very much useful for object recognition of the human organs such as heart and lungs. Idea to do image processing by Ant Colony Optimization algorithm is not new, but fresh [1] and promising. Ant Colony Optimization [2] belongs to evolutionary computation methods, based on the swarm intelligence approach. The main advantage of swarm intelligence approach [2] is that system of simple communicating agents is capable of solving complex problems. Ant Colony Optimization (ACO) being a branch of swarm intelligence is here considered and its use for important biomedical image processing application is investigated..Though a variety of edge detection algorithms are available a new technique is adopted in this paper based on ACO. For segmentation, metaheuristic based Parallel Ant colony Optimization (PACO) approach has been implemented. The system has been simulated in the Mat lab for the parallel processing, using the master slave approach and information exchange. Here parallelism is inherent in program loops, which focused on performing searching operation in parallel.

Abstract--Medical image segmentation and detection at the early stage played vital roles for many health-related applications such as medical diagnostics, drug evaluation, medical research, training and teaching. Medical images are at the core of medical science and an enormous source of information that need to be utilized. Image processing techniques with regards to biomedical images are generally either used for the retrieval of images or for analysis and modification of images. The aim of this study is to develop ACO system for medical image segmentation applications due to the rapid execution for obtaining and extracting the Region of Interest from the images for diagnostic purposes in medical field. For segmenta tion, metaheuristic based Parallel Ant colony Optimization (PACO) approach has been implemented. Keywords—Segmentation, Pheromone, ACO, PACO, Medical Images

I. INTRODUCTION At the core of medical science are biomedical images – images of the human body that help in the understanding of the nature of human biological systems. These images may be at the molecular level or images of complete organs, organ systems, and body parts. These images are enormous sources of information and like any other source of information need to be tapped and analyzed to pave the way for better understanding[14]. In understanding and gathering information from these images, the technique of image processing is of utmost importance. Image Processing is the process of modifying or interpreting existing pictures,such as photographs. Commonly it uses image processing operators to prepare image for further analysis. When compared to ordinary images the medical images, consists of so many information, in which the feature extraction is very difficult. Medical images, such as CT, MRI, show the information inside the patient body by noninvasive method, so that it is much helpful for doctor’s diagnoses and less painful for patients. However the raw data can only give the material to doctor, the doctor has to decide by himself the priority of data need. It is very difficult to locate the problems in medical images if it contains noise or the image is not in a proper format due to irregular structure of human body. Applying [13] image processing technologies plays a pivotal role in processing and analyzing the images and also in forming the images. It is an essential pre-processing step in medical image segmentation. Detection of edges in an image helps us to © 2011 ACEEE DOI: 02.CEMC.2011.01.560

II. EXISTING WORK A clustering based approach using a Self Organizing Map (SOM) algorithm for medical image segmentation [18] consists of two phases. In the first phase, the MRI brain image is acquired from patient database. In that film artifact and noise are removed. In the second phase (MR) image segmentation is to accurately identify the principal tissue structures in these image volumes. A fuzzy based segmentation process to detect brain tumor was implemented. In that performance of the MRI image in terms of weight vector, execution time and tumor pixels detected. Watershed segmentation [19] based algorithm has been used for detection of tumor in 2D and in 3D. This method can segment a tumor provided that the desired parameters are set properly. An automatic approach to segment Cardiac Magnetic Resonance (CMR) images includes a preprocessing step that consists in filtering the image using connected operators (area opening and closing filters) which is applied in order to homogenize the cavity and solve the problems due to the papillary muscles. Thereby the GVF snake algorithm is applied with one point clicked in the cavity as initialization and an optimized tuning of parameters for the endocardial contour extraction. The epicardial border is then obtained using the endocardium as initialization. A high agreement between manual and automatic 68

Poster Paper Proc. of Int. Colloquiums on Computer Electronics Electrical Mechanical and Civil 2011 region is segmented using PACO and performance evaluations are evaluated. The fundamental principle of PACO is to divide K ants into M sub colonies, so that the number of ants per each sub ant colony is the total number of ants divided by the number of sub colonies. The fundamental principle of PACO is to divide K ants into M sub colonies, so that the number of ants per each sub ant colony is the total number of ants divided by the number of sub colonies. In the algorithm designing, each colony is treated as an independent processor and then the ant colony can search the best solution independently. In order to avoid the local optimization in some colonies when the ant is doing the job, the other sub colonies should carry out the information exchange with each other in the chosen fixed time interval condition so that the execution time is reduced very much[5] In the proposed PACO approach, totally M colonies are considered in which M-1 colonies are treated as slaves and one colony is assigned for master. Each colonies visit all the pixels without revisit. Initially, the pheromone value for all the colonies is initialized and the posterior energy values are computed. Finally each slave colony yields global optimum value and the master colony system also yields global optimum value. Therefore M-1 slave colonies produce M-1 optimum values. . These values are compared and the highest global optimum value from slave colonies is computed and compared with the master global value. If the values of the slave colonies are less this optimum value is treated as adaptive threshold value. The entire procedure is repeated for number of times to obtain the more accurate value.

contours was obtained with correlation scores of 0.96 for the endocardium and 0.90 for the epicardium. [24] The use of anti-geometric diffusion as a means of segmenting human knee cartilage performs satisfactorily for most of the knee images. It performs well when the femoral and tibial cartilages are clearly distinguishable and not in contact. . Automatic segmentation of knee cartilage is a challenging problem and it seems unlikely that anti-geometric diffusion alone will solve the problem. [23] III. PACO FOR IMAGE SEGMENTATION A. Image acquisition and pre-processing The synthetic images obtained were stored in the JPEG format in matlab 7.Image preprocessing [5] indicates that the same tissue type may have a different scale of signal intensities for different images. It depends on the modality and corrects the system irregularities such as differential light detection efficiency, dead pixels or dark noise. A potentially good image is selected , which is filtered by convolving with a 3 × 3 Gaussian low pass filter with variance σ = 3. To eliminate the noise in the image, two successive stages of morphological filtration, i.e., dilation and erosion have been applied. B. Image Enhancement The pre-processed images may contain a high intensity salt and pepper noise which appears due to the presence of gray scale variations in the image which is removed by applying suitable filters and performing normalization. Hence the objective of enhancement is de-noising the high frequency components [17].Weighted Median (WM) filters has been used for removing noise. WM filtering is an enhancement technique for removing noise without significantly reducing the sharpness of the image. WM filter reduces noise in an image by preserving useful details. These [5] filters have the robustness and edge preserving capability of the classical median filter. A WM value W(x, y) is calculated using Eq. 1: W(x, y) =median {w 1 × x 1………w n × x n} (1) wh ere, x 1…....x n are the intensity values inside a sliding window centered at (x,y) and w × n denotes replication of x, w times.

IV. RESULTS AND DISCUSSIONS Set N =11, M = 100, K =10 (10 ants), a = 1, ζ max = 50 , r2 = 3 .The designed algorithm has been tested on three types of MR images including cardiac, brain tumor and arthritis. Results indicate that the segmentation is very clear for the images where the region of interest is enhanced. Fig 1, 2 and 3 illustrate the segmentation results for the brain tumor, cardiac and knee bone MRI respectively. Figure 2 shows Short axis MRI image of the heart ventricles in end-diastolic phase. The circular-section left ventricle is shown on the right side, and the meniscus shaped cross section of the right ventricle is shown on the left side of the figure. Figure 3 shows Magnetic resonance imaging (MRI) examination of the knee in a child with juvenile rheumatoid arthritis. Coronal gradient echo (MPGR) MRI shows high signal joint effusion, meniscal hypoplasia, and widening of the intercondylar notch..

C. Parallel Ant Colony Optimization The Segmentation of an image entails the division or separation of the image into regions of similar attribute. The ultimate aim in a large number of image processing applications is to extract important features from the image data, from which a d e sc r i p t i o n , interpretation, or understanding of the scene can be provided by the machine[7]. In this paper single image segmentation through PACO has been proposed. PACO [17] is a parallel implementation of ACO where ants do their study simultaneously on different processing units. This intuitively provides improved performance a nd speeds up t h e searching process by exchanging information about the solutions they found. In this proposed PACO, master-slave and information exchange approaches are combined to improve the result of image segmentation. The suspicious © 2011 ACEEE DOI: 02.CEMC.2011.01. 560

(a) Original Image

(b) Segmented image

Fig 1 MRI of brain tumor

Poster Paper Proc. of Int. Colloquiums on Computer Electronics Electrical Mechanical and Civil 2011

[6]Marco Dorigo ,”Swarm Intelligence, Ant Algorithms and Ant Colony Optimization:” IRIDIA Université Libre de Bruxelles ,Belgio” [7] Teerapun Saeh eaw, Nivit Charoenchai, and Wichai Chattinnawat:” Application of Ant colony optimization for Multiobjective Production Problems”,World Academy of Science, Engineering and Technology 602009 [8]Marco Dorigo, Mauro Birattari, and Thomas St¨utzle,Universit ´e Libre de Bruxelles, Belgium,”Ant colony Optimization,Artificial ants as a computational intelligence technique” [9] Omolbani Mohamad Rezapour-Amirahmad dehghani,,”A review of antcolony optimization for suspend ed sed iment estimation”,Australian Journal of Basic and Applied Sciences, 4(7): 2099-2108, 2010 ISSN 1991-8178 ©2010, insinet Publication [10].Anirudh Shekhawat ,Pratik Poddar, Dinesh Bowal,”Ant colony optimization algorithms : introduction and beyond,”Artificial Intellligence Seminar 2009 [11] R. B. Dubey, M. Hanmandlu, S. K. Gupta, “Semiautomatic Segmentation of MRI Brain Tumor”, ICGST-GVIP Journal, ISSN: 1687- 398X, Volume 9, Issue 4, August 2009 [12] Yuanjing Feng and Zhejin Wang, “Ant Colony Optimization for Image Segmentation” [13] H.P. Ng, S.H. Ong, K.W.C. Foong, ,P.S. Goh5, W.L. Nowinski, Medical Image Segmentation Using K-Mean s Clustering And Improved Watershed Algorithm [14] K.M.M. Rao, V.D.P. Rao, Medical Image Processing [15] Marco Dorigo, Vittorio Maniezz, Alberto Colorni, “The Ant System: Optimization by a colony of cooperating agents,” IEEE Transactions on Systems, Man, and Cybernetics–Part B, Vol.26, No.1, 1996, pp.1-13 [16] P.Vasuda, S.Satheesh, “Improved Fuzzy C-Means Algorithm for MR Brain Image Segmentation “,P. Vasuda et. al. / (IJCSE) International Journal on Computer Science and En neering, [17] J.Jaya and K.Thanushkodi, “Segmentation of MR Brain tumor using Parallel ACO”, (IJCNS) International Journal of Computer and Network Security [18] T.Logeswari and M.Karnan, “An Improved Implementation of Brain Tumor Detection Using Segmentation Based on Hierarchical Self Organizing Map”, International Journal of Computer Theory and Engineering, Vol. 2, No. 4, August, 2010 [19] Rajeev Ratan , Sanjay Sharma , S. K. Sharma, “Brain Tumor Detection based on Multi-parameter MRI Image Analysis”, ICGSTGVIP Journal, ISSN 1687-398X, Volume (9), Issue (III), June 2009 [20] Jingxin Nie, Zhong Xue, Tianming Liu, Geoffrey S. Young, Kian Setayesh, Lei Guo and Stephen T.C. Wong, “Automated brain tumor segmentation using spatial accuracy-weighted hidden Markov Random Field” [21] S. Murugavalli and V. Rajamani, “An Improved Implementation of Brain Tumor Detection Using Segmentation Based on Neuro Fuzzy Technique”, Journal of Computer Science 3 (11): 841-846, 2007 ISSN 1549-3636 © 2007 Science Publications [22] M. Masroor Ahmed, Dzulkifli Bin Mohamad, Segmentation of Brain MR Images for Tumor Extraction by Combining Kmeans Clustering and Perona-Malik model [23] James Cheong and David Suter, “A Study on Anti-Geometric Diffusion for the Segmentation of Human Knee Cartilage”MECSE15-2004 [24] R. El Berbari, I. Bloch, A. Redheuil, E. Angelini, E. Mousseaux, F. Frouin and A. Herment,”An automated myocardial segmentation in cardiac MRI”, Proceedings of the 29th Annual International Conference of the IEEE EMBS Cité Internationale, Lyon, France August 23-26, 2007

(a) Original image (b) Segmented image Figure 2: Short axis MRI image of the heart

(a) Original image (b) segmented image Figure 3 :MRI examination of the knee in a child with juvenile rheumatoid arthritis.

IV. CONCLUSION An image segmentation method based on the parallel ant colony optimization algorithm has been discussed. This work has provided results for the application of segmentation algorithm for three types of MRI images. Implementing segmentation algorithms for the medical images can help in the proper detection of the region of interest .Further it can be helpful for the surgical planning. But the limitation of this algorithm is that it is applicable only for 2D images. The algorithm should be improved for processing 3D images as the MRI data is 3D in nature. REFRENCES [1] R. Laptik, D. Navakauskas,”Application of Ant Colony Optimization for Image Segmentation” [2] R. Laptik, D. Navakauskas “,MAX-MIN Ant System in Image Preprocessing” [3]Christian Blum “Ant colony optimization: Introduction and recent trends” [4]Amali Asha, S.P. Victor, A. Lourdusamy, A. Amali Asha “Feature Extraction in Med ical Image using Ant Colony Optimization : A Study”A. et al. / International Journal on Computer Science and Engineering (IJCSE) ISSN : 0975-3397 Vol. 3 No. 2 Feb 2011 [5] Jayabalan Jaya and Keppanagowder Than ushkodi “Implementation of Computer Aided Diagnosis System Based on Parallel Approach of Ant Based Medical Image Segmentation,”, Journal of Computer Science 7 (2):291-297, 2011 ISSN 1549-3636 © 2011 Science Publications