Abstract  Brain tumor is one of the foremost causes for the i ncrease in mortality among children and adults. Com puter visions are being used by doctors to analysis and diagnose the medical problems. Magnetic Resonance Imaging (M RI) is a medical imaging technique, which is used to visualize inter nal structures of MRI brain images for analyzing no rmal and abnormal prototypes of brain while diagnosing. It is a non%i nvasive method to take picture of brain and the sur rounding images. Image processing techniques are used to extract meaningfu l information from medical images for the purpose o f diagnosis and prognosis. Raw MRI brain images are not suitable fo r processing and analysis since noise and low contrast affect the quality of the MRI images. The classification of MRI brain images is emphasized in this paper for cancer diagn osis. It can consist of four steps : Pre%processing, identification of Region of Intere st (ROI), feature extraction and classification. For improving quality of the image, partial differential equation s method is proposed and its result is compared wit h other methods such as block analysis method, opening by reconstruction me thod and histogram equalization method using statistical parameters such as carrier signal to ratio, peak signal%to%ratio, s tructural similarity index measure, figure of merit , mean square error. The enhanced image is converted into bi%level image, wh ich is utilized for sharpening the regions and filling the gaps in the binarized image using morphological operators ROI i s identified by applying region growing method for extorting the five features. The classification is performed based on the extracted image feature to determine whether th e brain image is normal or abnormal and it is also, introduced hybridizatio n of Neural Network (NN) with bee colony optimizati on for the classification and estimation of cancer affect on g iven MRI image. The performance of the proposed cla ssifier is compared with traditional NN classifier using statistical me asures such as sensitivity, specificity and accurac y. The experiment is conducted over 100 MRI brain images.

References

[1] Cline H., Lorensen E., Kikinis R., and Jolesz F., Three Dimensional Segmentation of MR Images of the Head using, Journal of Computer Assisted Tomography , vol. 14, no. 6, pp. 1037*1045, 1990.

[2] Cocosco A., Zijdenbos P., and Evans C., A Fully Automatic and Robust Brain MRI Tissue Classification Method, Medical Image Analysis, vol. 7, no. 4, pp. 513*527, 2003.

[3] Coulibaly P. and Evora D., Comparison of Neural Network Methods for Infilling Missing Daily Weather Records, Journal of Hydrology, vol. 341, no. 2, pp. 27*41, 2007 .

[4] Gonzalez C., Woods E., and Masters R., Digital Image Processing, Journal of Biomedical Optics , vol. 14, no. 2, pp. 029901*029901, 2009.

[5] Karaboga N., Kalinli A., and Karaboga D., Designing Digital IIR Filters using Ant Colony Optimisation Algorithm, Engineering Applications of Artificial Intelligence , vol. 17, no. 3, pp. 301*309, 2004 .

[6] Kelleg oz T., Toklu B., and Wilson J., Elite Guided Steady*State Genetic Algorithm for Minimizing Total Tardiness in Flowshops, Computers and Industrial Engineering, vol. 58, no. 2, pp. 300*306, 2010.

[7] Kiranyaz S., Ince T., Yildirim A., and Gabbouj M., Evolutionary Artificial Neural Networks by Multi*Dimensional Particle Swarm Optimization, Neural Networks, vol. 22, no. 10, pp. 1448*1462, 2009.

[8] Mahsa C. and Woo C., Medical Image Segmentation using a Multi*Agent System Approach, The International Arab Journal of Information Technology, vol. 10, no. 3, pp. 222* 229, 2013.

[9] Martin H., Howard D., Mark B., Orlando D., Neural Network Design, Vikas Publishing House, New Delhi, India, 1996.

[10] Perona P. and Malik, J., Scale Space and Edge Detection using Anisotropic Diffusion, IEEE Transaction on Pattern Analysis and Machine Intelligence , vol. 12, no. 7, pp. 629*639, 1990.

[11] Primary Brain Tumors in the United States Statistical Report 2002*2004. available at: http://www.cbtrus.org/reports/2007*2008/ 2007 report.pdf, last visited 2007,

[12] Robotham H., Bosch P., Guti rrez*Estrada J., Castillo J., Pulido*Calvo I., Acoustic Identification of Small Pelagic Fish Species in Chile using Support Vector Machines and Neural Networks, Fisheries Research, vol. 102, no. 2, pp. 115*122, 2010.

[13] Sathya S. and Manavalan R., Analysis of Background Detection and Contrast Enhancement of MRI Images, International Journal of Computer Applications , vol. 36, no. 12, 2011.

[14] Shujun F., Qiuqi R., Chengpo M., and Wenqia W., Edge Detection for Image Measurement based on Nonlinear Diffusion Filterin, in Optics and Precision Engineering , vol. 15, no. 2, pp. 289*293, 2007

[15] Zhang S., Gong S., Guan Y., Zhang P., and Gong Q., A Novel Iga*Edspso Hybrid Algorithm for the Synthesis of Sparse Arrays, Progress in Electromagnetics Research , vol. 89, pp. 121*134, 2009.

[16] Zulaikha B. and Mohamed S., A Robust Segmentation Approach for Noisy Medical Images using Fuzzy Clustering With Spatial Probability, the International Arab Journal of Information Technology , vol. 9, no. 1, pp. 74*83, 2012 . Sathya Subramaniam Obtained MSc, computer science from Vivekananda College of Arts and Science, Periyar University, India, in 2008, MPhil in computer science from KSR College of Arts and Science, Periyar University, India in 2011. She worked as Asst Prof, Department of Computer Science and Applications, Kandaswami kandar s College, India. 124 The International Arab Journal of Information Tech nology Manavalan Radhakrishnan obtained MSc, computer science from St.Joseph s College of Bharathidasan University, India in 1999 and MPhil in computer science from Manonmaniam Sundaranar University, India in 2002. He works as Asst Prof and Head, Department of Computer Science and Applications, KSR College of Arts and Science, India. He pursues PhD in Medical Image Processing. His areas of interest are medical image processing and analysis, soft computing, pattern recognition and theory of computation.