Automatic detection of microaneurysms in colour fundus images for diabetic retinopathy screening. Issue 5 (July 2016)

• Record Type: Journal Article
• Authors: Rahim, Sarni; Jayne, Chrisina; Palade, Vasile; Shuttleworth, James
• Is Part Of: Neural computing & applications. Volume 27:Issue 5(2016)
• Page Start: 1149
• Page End: 1164
• Publication Date: 2016-07
• Subjects: Diabetic retinopathy -- Eye screening -- Colour fundus images -- Image processing -- Microaneurysms; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-1929-5
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract Regular eye screening is essential for the early detection and treatment of the diabetic retinopathy. This paper presents a novel automatic screening system for diabetic retinopathy that focuses on the detection of the earliest visible signs of retinopathy, which are microaneurysms. Microaneurysms are small dots on the retina, formed by ballooning out of a weak part of the capillary wall. The detection of the microaneurysms at an early stage is vital, and it is the first step in preventing the diabetic retinopathy. The paper first explores the existing systems and applications related to diabetic retinopathy screening, with a focus on the microaneurysm detection methods. The proposed decision support system consists of an automatic acquisition, screening and classification of diabetic retinopathy colour fundus images, which could assist in the detection and management of the diabetic retinopathy. Several feature extraction methods and the circular Hough transform have been employed in the proposed microaneurysm detection system, alongside the fuzzy histogram equalisation method. The latter method has been applied in the preprocessing stage of the diabetic retinopathy eye fundus images and provided improved results for detecting the microaneurysms.