Detection of Diabetic Retinopathy in Fundus Images using Extreme Learning Machines
- February 11, 2019
- Posted by: RSIS
- Category: Electrical and Electronics Engineering
International Journal of Research and Scientific Innovation (IJRSI) | Volume VI, Issue I, January 2019 | ISSN 2321–2705
Lisa Maria Macedo, Prof. Amita Dessai
Dept. of Electronics and Telecommunication Engineering, Electronics Communication and Instrumentation, Goa College of Engineering, Farmagudi, Goa, India
Abstract— Diabetic Retinopathy (DR) is the term used to describe the retinal damage due to diabetes.
Initially, diabetic retinopathy may cause none to mild symptoms but sight loss at an advanced stage.
Hence detecting lesions automatically in retinal images can assist in diagnosis and screening of DR at an early stage.
The detection of the different lesions in fundus images is therefore of interest. This project proposes thepre-processing of the image using a Median Filter and Contrast Limited Adaptive Histogram Equalization(CLAHE), optic disc detection using Hough Transform, feature extraction using Gray-Level Co-Occurrence Matrix (GLCM)and Extreme Learning Machines (ELM) for classification.
Keywords— Diabetic Retinopathy, Median Filter, Contrast Limited Adaptive Histogram Equalization (CLAHE), Gray Level Co-Occurrence Matrix (GLCM), Extreme Learning Machines (ELM).
When one has diabetes, the body’s ability to produce or respond to the hormone insulin is impaired. Diabetic retinopathy is a diabetes complication that affects eyes. It damages the blood vessels of the retina. At first, diabetic retinopathy may cause none or mild vision problems. However, eventually causing blindness. Imaging of fundus helps in identifying and further classify the DR. The spatial distribution of exudates and microaneurysms and hemorrhages, can be used to determine the severity of the DR. Color retinal images are studied closely by ophthalmologists.