Feature extraction from retinal images for diagnosis of diabetic retinopathy

Abstract

ABSTRA CT There is an ever—increasing interest in the development of automatic medical diagnosis systems due to the advancement in computing technology and also to improve the service by medical community. The knowledge about health and disease is required for reliable and accurate medical diagnosis. Digital information is acquired at different scales, quickly and efficiently by means of image processing techniques. So the algorithms can be developed for computer- aided medical diagnosis based on image processing technology. In this project; a design and algorithms of a digital diabetic retinopathy system for the screening of diabetic retinopathy is proposed. The project proposes algorithms for retinal image preprocessing, blood vessel enhancement and segmentation, and optic disk localization and detection. The developed methods are tested on three different publicly available databases i. e. DRIVE, STARE, Diaretdb. Four- parameters are used to check the validity of proposed algorithms i.e. visual inspection, accuracy, area under the receiver operating curves (‘ROC) and calculation time. The proposed method for preprocessing achieves an average accuracy of 98.04% whereas proposed method for vessel segmentation achieves an average accuracy of 94.85% and an average area under the receiver operating characteristic curve of 0.9669. The algorithm for optic disk achieves an average accuracy of 96.7% for localization. The proposed methods are compared with recently published methods amid experimental results show that proposed methods outperform all others.