557 by ides editor

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

Segmentation of Brain MRI for tumor Detection Using Ant Colony Optimization Anjum Sheikh1, R.K. Krishna2 1

M Tech Student, EMS, RCERT, Chandrapur,India anjum_nazir@rediffmail.com 2 Asst. Prof.,Department of Electronics Engg. RCERT, Chandrapur,India choosing the path at previous times and with the strength of the pheromone concentration laid on it [10, 11]. The designed algorithm has been simulated in matlab.In this paper Ant Colony Optimization algorithm is used to extract the tumor from the brain magnetic resonance (MR) images.

Abstract- Tumor segmentation from MRI data is an important but time consuming task performed manually by medical expertsImage analysis is still performed manually which is often a difficult and time-consuming task. As a result, there is an increasing need for computerized image analysis to facilitate image based diagnosis. In this paper we propose an Ant colony Optimizaton algorithm for tumor detection. Ant colony optimization (ACO) is a nature-inspired optimization algorithm motivated by the natural phenomenon that ants deposit pheromone on the ground in order to mark some favorable path that should be followed by other members of the colony.

II. REVIEW OF EXISTING WORKS Several methods have been proposed to segment the tumor from brain MR images. An automatic segmentation method separates non-enhancing brain tumors from healthy tissues in MR images to aid in the task of tracking tumor size over time [18]. The MR feature images used for the segmentation consist of three weighted images (T1, T2 and proton density (PD)) for each axial slice through the head. An initial segmentation is computed using an unsupervised fuzzy clustering algorithm. Then, integrated domain knowledge and image processing techniques contribute to the final tumor segmentation. In [10] an efficient method for automatic brain tumor segmentation for the extraction of tumor tissues from MR images has been described. Perona and Malik anisotropic diffusion model for image enhancement and Kmeans clustering technique for grouping tissues belonging to a specific group have been combined for extraction of tumor tissues from MR images [16]. The proposed method uses T1, T2 and PD weighted gray level intensity images. SHE incorporates the spatial interpolation accuracy of low-resolution images into the optimization procedure of the Hidden Markov Random Field (HMRF) to segment tumor using multi-channel MR images with different resolutions, e.g., high-resolution T1-weighted and low resolution T2-weighted images [19]. In experiments, this algorithm was evaluated using a set of simulated multichannel brain MR images with known ground-truth tissue segmentation and also applied it to a dataset of MR images obtained during clinical trials of brain tumor chemotherapy. Neuro-fuzzy segmentation process of the MRI data is used to detect various tissues like white matter, gray matter, csf and tumor[14]. The advantage of hierarchical self organizing map and fuzzy c means algorithms are used to classify the image layer by layer. The lowest level weight vector is achieved by the abstraction level. A higher value of tumor pixels has been achieved by this neuro-fuzzy approach. Watershed segmentation based algorithm has been used for detection of tumor in 2D and in 3D.[15] .Compared to traditional metaheuristic segmentation methods, the proposed ACO has advantages that it can effectively segment the fine details. The suggested image segmentation strategy is tested

Index Terms: ACO, M RI image, tumor, segmentation, pheremone

I. INTRODUCTION The computer-aided diagnoses is to use computer to process the medical images to extract the useful information so that the doctor can make a diagnoses decision easier and quicker. The process of segmenting tumors in MRI images as opposed to natural scenes is particularly challenging [4]. The tumors vary greatly in size and position, have a variety of shape and appearance properties, have intensities overlapping with normal brain tissue, and often an expanding tumor can deflect and deform nearby structures in the brain giving an abnormal geometry also for healthy tissue.[5] But 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 image processing technologies plays a pivotal role in processing and analyzing the images and also in forming the images. Image segmentation is a complex visual computation problem, which refers to the process of distinguishing objects from background. Ant colony optimization (ACO) is a cooperative search algorithm inspired by the behavior of real ants. [7] Ant Colony Optimization (ACO) is a population-based approach first designed by Marco Dorigo and coworkers, inspired by the foraging behavior of ant colonies [9]. Individuals ants are simple insects with limited memory and capable of performing simple actions. However, the collective behavior of ants provides intelligent solutions to problems such as finding the shortest paths from the nest to a food source. Ants foraging for food lay down quantities of a volatile chemical substance named pheromone, marking their path that it follows. Ants smell pheromone and decide to follow the path with a high quantity of pheromone. The probability that an ant chooses a path increases with the number of ants ÂŠ 2011 ACEEE DOI: 02.CEMC.2011.01. 557

Poster Paper Proc. of Int. Colloquiums on Computer Electronics Electrical Mechanical and Civil 2011 ants produce identical solutions during one iteration. Additionally, because of the local pheromone update, the minimum values of the pheromone are limited. Thus the pheromones are updated by adding equation 1 and 2. This procedure is repeated for all the ants and for all the iterations. In this experiment the number of iterations is taken as 30. To further enhance the value, this entire procedure can be repeated for any number of times.

on a set of synthetic MR Brain images. The improved accuracy rate according to the experimental results is due to better characterization of natural brain structure. III. ACO FOR SEGMENTATION OF BRAIN TUMOR A. Image Acquisition Images of a patient obtained by MRI scan is displayed as an array of pixels and stored in Matlab7.1. All the test images were converted to gray scale and resized for 2 the size 56X256. In image processing, it is necessary to smoothen an image while preserving its edge. The assumption is that noise is captured by the high frequency coefficients, thus by filtering these coefficients, the unwanted noise is removed. But the edges are also high frequency components hence its necessary to preserve these while removal of noise. A gray scale image can be specified by giving a large matrix whose entries are between 0 to 255, 0 corresponds to black while 255 corresponds to white pixels. All the test images were converted to gray scale and resized for 2 the size 56X256.

IV. RESULTS AND DISCUSSIONS The algorithm was tested for synthetic images. In this implementation 30 numbers of iterations have been used and 5 levels of image display were used to display the results. The number of ants and the number of steps to be taken by the ants are equal to the number of rows and the number of columns of the image respectively. The accuracy of the results can be improved by having more levels of image display and increasing the number of iterations. The segmentation results indicate that the segmentation of T2 weighted images give better results as compared to the T1 images. Fig 1,2 & 3 indicate the outputs for 3 different brain tumor images.

B. Segmentation Using ACO Segmentation of images holds an important position in the area of image processing. Image segmentation in medical imaging is a tool to delineate anatomical structure and other regions of interest whose a priori knowledge is generally available.[17] It becomes more important while typically dealing with medical images where pre-surgery and post surgery decisions are required for the purpose of initiating and speeding up the recovery process[10].The segmentation of tumor in a brain will make the surgeon able to see the tumor and then ease the treatment. The algorithm consists of the following steps:

Fig 1.(a) Original Grayscale Image (b) Segmented Image

1.Pheromone Initialization For each ant assign the initial pheromone value T0. And for each ant select a random pixel from the image which has not been selected previously. To find out the pixels is been selected or not, a flag value is assigned for each pixel. Initially the flag value is assigned as 0, once the pixel is selected the flag is changed to 1. This procedure is followed for all the ants. For each ant a separate column for pheromone and flag values are allocated in the solution matrix.

Fig 2(a) Original Grayscale Image (b) Segmented Image

2. Local Pheromone Update The local pheromone update is performed by all ants after each construction step. Each ant applies it only to the last edge traversed: Tnew=[(1-ρ)*Told+ρ*ΔT0ld] ΔT0ld (1) Tnew2=[(1-ρ)*Told] |1- ΔT0ld | (2) where Told and Tnew are the old and new pheromone values, and ñ is rate of pheromone evaporation parameter in local update.The main goal of the local update is to diversify the search performed by subsequent ants during one iteration. In fact, decreasing the pheromone concentration on the edges as they are traversed during one iteration encourages subsequent ants to choose other edges and hence to produce different solutions. This makes less likely that several © 2011 ACEEE DOI: 02.CEMC.2011.01. 557

Fig 3. (a) Original Grayscale Image (b) Segmented Image

Poster Paper Proc. of Int. Colloquiums on Computer Electronics Electrical Mechanical and Civil 2011 V. CONCLUSION In this paper a metaheuristic based image segmentation method was approached. The results show that Ant Colony Optimization (ACO) method can successively segment a tumor provided the parameters are chosen properly. Tumor identification and the investigation are carried out for the potential use of MRI data for improving the tumor shape and 2D visualization of the surgical planning. The aim of segmentation is to maintain quality of the segmentation to be more actual than manual segmentation and will speed up segmentation in operative imaging. Implemented algorithm deals with only 2D images but the algorithm can be improved to process the 3D images as the MRI data is 3D in nature.

[9] M. Dorigo, V. Maniezzo and A. Colorni,” Positive feed back as a search strategy”, Technical Report (16) Politecnico di Milano, Italy, 1991. [10] Y. Ge, Q. C. Meng, C. J. Yan and J. Xu. A Hybrid Ant Colony “Algorithm for Global Optimization of Continuous Multi-Extreme Functions”,. Proceedings of the Third International Conference on Machine Learning and Cybernetics, Shanghai, 2427-2432, 2004. [11] R. Beckers, J. L. Deneubourg, and S. Goss. “Trails and Uturns in the selection of the shortest path by the ant”, Lasius. Niger. J. Theor. Bio. 159:397–415, 1992. [12]R. Laptik, D. Navakauskas, “Application of Ant Colony Optimization for Image Segmentation”, ISSN 1392 – 1215, Electronics And Electrical Engineering, 2007. No. 8(80) [13] M. Dorigo and G. Di Caro, “The Ant Colony\ Optimization meta-heuristic”, in New Ideas in Optimization”, D. Corne et al., Eds., McGraw Hill, London, UK, pp. 11-32, 1999. [14] 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 [15] 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 [16] M. Masroor Ahmed, Dzulkifli Bin Mohamad, “Segmentation of Brain MR Images for Tumor Extraction by Combining Kmeans Clustering and Perona-Malik Anisotropic Diffusion Model” [17] V. Dey , Y. Zhang , M. Zhong , “A Review On Image Segmentation Techniques With Remote Sensing Perspective,” In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium – 100 Years ISPRS, Vienna, Austria, July 5– 7, 2010, IAPRS, Vol. XXXVIII, Part 7A [18] Lynn M Fletcher-Heath, Lawrence O Hall, Dmitry B Goldgof, F.Reed Murtagh, “Automatic segmentation of non-enhancing brain tumors in magnetic resonance images”, Artificial Intelligence in Medicine, Volume 21, Issues 1-3, January-March 2001, Pages 4363 [19]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” [20] http://www.linac.com/products.html

REFRENCES [1] M. Dorigo and S. Thomas, “Ant Colony Optimization”. Cambridge: MIT Press, 2004. [2] M. Dorigo, M. Birattari, and T. Stutzle, “Ant colony optimization,” IEEE Computational Intelligence Magazine, vol. 1, pp. 28–39, Nov. 2006. [3] Matei Mancas, Bernard Gosselin, “Towards an automatic tumor segmentation using iterative watersheds”, Signal Processing & Circuit Theory Lab, Faculté Polytechnique de Mons Bâtiment Multitel, Parc Initialis, avenue Copernic 1, 7000, Mons, Belgium [4] A. Amali Asha, S.P. Victor, A. Lourdusamy,” Feature Extraction in Medical Image using Ant Colony Optimization : A Study”, International Journal on Computer Science and Engineering (IJCSE) ISSN : [5]Dana Cobzas, Neil Birkbeck, Mark Schmidt, Martin Jagersand, Albert Murtha, “3D Variational Brain Tumor Segmentation using a High Dimensional Feature Set” [6] Saif D. Salman & Ahmed A. Bahrani, “Segmentation of tumor tissue in gray medical images using watershed transformation method”, International Journal of Advancements in Computing Technology Volume 2, Number 4, October 2010 [7] Wei Gao, “Study on Immunized Ant Colony Optimization”, Third International Conference on Natural Computation (ICNC 2007) 0-7695-2875-9/07 $25.00 © 2007,IEEE [8] Yuanjing Feng and Zhejin Wang, “Ant Colony Optimization for Image Segmentation”, Zhejiang University of Technology. China