Automatic detection of non-proliferative diabetic retinopathy in retinal fundus images using convolution neural network
- PDF / 1,130,497 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 22 Downloads / 236 Views
ORIGINAL RESEARCH
Automatic detection of non‑proliferative diabetic retinopathy in retinal fundus images using convolution neural network P. Saranya1 · S. Prabakaran1 Received: 25 June 2020 / Accepted: 4 September 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Diabetic retinopathy (DR) is one of the complications of diabetes and a leading cause of blindness in the world. The tiny blood vessels inside the retina are damaged due to diabetes and result in various vision-related problems and it may lead to complete vision loss without early detection and treatment. Diabetic retinopathy may not cause any symptoms during its earlier stage of the disease and many physical tests such as visual acuity tests, pupil dilation, etc., are required to detect diabetic retinopathy disease. So, early detection of diabetic retinopathy disease is required to avoid vision loss. This work aims to automate the detection and grading of non-proliferative Diabetic Retinopathy from retinal fundus images using Convolution Neural Networks. The model was tested on two popular datasets such as MESSIDOR and IDRiD. Before applying the Convolution Neural Network (CNN) layers, the images were pre-processed and resolution was adjusted (256 × 256). The maximum accuracy achieved is 90.89% using MESSIDOR images. The research can be carried forward by applying various preprocessing techniques before putting them through different computational layers. Keywords Diabetic retinopathy · Retinal fundus images · CNN
1 Introduction Diabetic Retinopathy causes retinal damage and might lead to loss of vision. Therefore early detection of the severity of DR plays an important role in the treatment. DR affects 80% of the people who have diabetes. It does not cause any symptom during its initial stage and some people may notice minute changes in their vision. According to the WHO, there are 31.7 million people affected by Diabetes in India and it is expected to rise to 79.4 million by 2030 (Gadkari et al. 2018). National Diabetes and diabetic Retinopathy survey-2019, says that the prevalence of diabetes is 11.8% in the last four years from 2015 to 2019. The major cause of DR is poor control over their blood sugar level and the majority of diabetic patients will never go for a fundamental evaluation of DR since it is a time-consuming procedure. Diagnosis of diabetic retinopathy is time-consuming and it requires several clinical tests such as visual acuity tests, * P. Saranya [email protected] 1
Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai 603203, India
pupil dilation, and optical coherence tomography to detect it. Scientists and experts all around the globe are working hard to develop techniques to detect the symptoms faster and with precision so that no one has to lose their vision because of this horrible disease. Automated detection with leading technologies available in the market will lead to a large amount of savings of time and effort. Two main types of diabetic ret
Data Loading...
