الفهرس Only 14 pages are availabe for public view
 |  | from | 132 |  |  |
| | | from | 132 | | |
Abstract
Diabetic Retinopathy (DR) is one of the leading causes of blindness. The risk of vision loss due to this disease could be avoided and reduced by timely diagnosis. Early treatment of DR (Diabetic Retinopathy) through screening and controlling the duration and degree of diabetes are an effective steps in fighting and preventing the disease progress thus minimizing the danger of vision loss. The early detection of DR can reduce the risk of vision loss by 50 percent. Medical imagining is one of the most important tools among the health care commu- nity. Not only for the record storing and visual documentation, but also for information extraction of many diseases. We can extract great valuable information and make useful history which helps us in treatment and healing by tracking the disease progress of the patient. Diabetic retinopathy is among these diseases that can be early discovered by analyzing the retinal images using one of these tools before involving in critical stages that may be lead to blindness. Analysis of the retinal blood vessels structure is a main step for this tools as it can lead us to ensure the presence of disease or not. Manually performing such a task is considered a major obstacle as it has many disadvantages like prone to human error and waste time because of the vast amount of images and vessel structure complication. Soan accurate automated retinal blood vessel segmentation technique is needed to detect the retinal blood vessels perimeter and area or determine the optic disc shape to help as a preprocess phase before the disease detection step. In recent years, several researches have been proposed for vessel segmentation using differ- ent techniques such as Support Vector Machine(SVM), K-nearst neighbor, Na{u00A8}ıve Bayes, Fuzzy C Means (FCM) and others technique. However there are still some problems that need to be handled such as analyzing of thin vessels or dealing with abnormal images with their symptoms that complicate the segmentation process and effect on the results quality