این سایت در حال حاضر پشتیبانی نمی شود و امکان دارد داده های نشریات بروز نباشند
International Journal of Nonlinear Analysis and Applications، جلد ۱۳، شماره ۲، صفحات ۳۲۰۳-۳۲۱۲
عنوان فارسی
چکیده فارسی مقاله
کلیدواژه‌های فارسی مقاله
عنوان انگلیسی Diabetic retinopathy detection and classification based on deep learning: A review
چکیده انگلیسی مقاله Diabetic retinopathy can be defined as an eye disease that occurs specifically in diabetic patients and results in damaging the small blood vessels of the retina because of high and low blood sugar. Delayed detection and treatment often lead to blindness, so one of the most significant issues is early detection of this disease, which is necessary for successful treatment. Many deep learning methods have been suggested for diabetic retinopathy detection and classification. Manual inspection of the fundus images to check diabetic retinopathy is highly tedious and time-consuming work. Thus, an automatic method for early diabetic retinopathy diagnosis utilizing the fundus images is very useful tool that helps experts. In this review paper, several deep learning models that are widely used in literature investigated diabetic retinopathy classification will be presented and discussed. In addition, a comparative analysis for classification performance accuracy of diabetic retinopathy using these deep learning models will be reviewed comprehensively. Thus, an automatic approach for early diabetic retinopathy diagnosis utilizing the fundus images is a very useful tool that helps the experts.
کلیدواژه‌های انگلیسی مقاله Diabetic Retinopathy, Deep learning, Fundus images, CNN, Accuracy metrics
نویسندگان مقاله Abeer Ahmed Ali |
Computer Science Department, Collage of Science, University of Baghdad, Baghdad, Iraq

Faten Abd Ali Dawood |
Computer Science Department, Collage of Science, University of Baghdad, Baghdad, Iraq

نشانی اینترنتی https://ijnaa.semnan.ac.ir/article_6805_f3dfc8f6451f39db4bf981ff42dfeb90.pdf
فایل مقاله فایلی برای مقاله ذخیره نشده است
کد مقاله (doi)
زبان مقاله منتشر شده en
موضوعات مقاله منتشر شده
نوع مقاله منتشر شده
برگشت به: صفحه اول پایگاه   |   نسخه مرتبط   |   نشریه مرتبط   |   فهرست نشریات