1 Classification and Calculation of Retinal Blood vessels Parameters Abstract - In this paper, we present an algorithm for the classification and calculation of retinal blood vessels parameters. Six parameters including Area, Length, thickness, Diameter, Mean Diameter and Tortuosity are calculated. The algorithm proceeds through three main steps 1. preprocessing operations on high resolution fundus images 2. For retinal vessel extraction, simple vessel segmentation techniques formulated in the language of 2D Median Filter 3. Segmentation for finding boundaries of the extracted blood vessels. Performance of this algorithm is tested using the fundus image database( 245 images) taken from Dr. Manoj Saswade, Dr.Neha Deshpande and online available databases diaretdb0, diaretdb1 and DRIVE. This algorithm achieves accuracy of 96% with 0.92 sensitivity and 0 specificityfor Saswade detabase , for diaretdb0 accuracy 95% with 0.95 sensitivity and 0 specificity, for diaretdb1 accuracy 96% with 0.96 sensitivity and 0 specificity, and for DRIVE database 98% accuracy with 0.98 sensitivity and 0 specificity and also used statistical techniques for result analysis, in this techniques used Mean, Standard Deviation, Variance, Covariance, Coefficient etc. Keywords - Blood Vessels, 2D Median Filter. 1. INTRODUCTION Proposed algorithm shows classification and calculation of retinal blood vessels parameters. Six parameters including Area, Length, thickness, Diameter, Mean Diameter and Tortuosity are calculated. In this algorithm we have used the Image Processing techniques for extraction of the retinal blood vessels and then classification and calculations of retinal blood vessels parameters. Firstly we have performed the preprocessing operation on high resolution fundus images. We have used 2D median filter for highlighting the blood vessel. For extraction of the blood vessels we have performed threshold function. Segmentation for detecting boundaries. For observing the result we have taken the images and formed a database from Dr. Manoj Saswade and Dr. Neha Despande (245 images), images from online databases diaretdb0, diaretdb1 and DRIVE. 2. METHODOLOGY Computer assisted diagnosis for various diseases are very common now a days and medical imaging is playing a vital role in such diagnosis. Image processing techniques can help in extractions of blood vessels and bifurcation points. The proposed algorithm has 3 stages, shown in the figure 1. In first stage preprocessing is done to remove the background noise from input fundus image. Blood vessels are highlighted and extracted in the second stage and in the third stage using segmentation technique boundaries are detected. 2.1 PREPROCESSING The Preprocessing is done to remove noise from the background and to enhance the image. We have taken out green channel, because green channel shows high intensity as compare to red and blue. shown in figure 2. Fundus Image Green channel image Manjiri B. Patwari Institute of Management Studies & Information Technology,Vivekana nd College Campus, Aurangabad MS (India) Manjiri.patwari@gma il.com Ramesh R. Manza Dept of CS and IT, Dr. B. A. M. University, Aurangabad MS (India) manzaramesh@g mail.com Yogesh M. Rajput Dept of CS and IT, Dr. B. A. M. University, Aurangabad MS (India) yogesh.rajput128@ gmail.com Deepali D. Rathod Dept of CS and IT, Dr. B. A. M. University, Aurangabad MS (India) rathoddeepali2@g mail.com Manoj Saswade Director, Saswade Eye Clinic” Aurangabad MS (India) nehad35@gmail. com Neha Deshpande Director, Guruprasad Netra Rugnalaya pvt. ltd, Aurangabad MS (India) prachims@yah oo.com