Automated method for the detection and segmentation of drusen in colour fundus image for the diagnosis of age‐related macular degeneration. Issue 6 (1st June 2018)

• Record Type: Journal Article
• Authors: Mohaimin, Sultan Mohammad; Saha, Sajib Kumar; Khan, Alve Mahamud; Mohammad Arif, Abu Shamim; Kanagasingam, Yogesan
• Is Part Of: IET image processing. Volume 12:Issue 6(2018)
• Page Start: 919
• Page End: 927
• Publication Date: 2018-06-01
• Subjects: eye -- medical image processing -- image segmentation -- image colour analysis -- object detection; automated method -- drusen segmentation -- drusen detection -- colour fundus image -- age‐related macular degeneration diagnosis -- AMD -- quantitative mapping -- retinal abnormalities detection -- disease -- retina -- nonuniform illumination correction -- inter‐subject variability minimization -- colour normalisation method -- ARIA datasets -- STARE dataset; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2017.0685
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : Age‐related macular degeneration (AMD) is one of the main reasons for visual impairment worldwide. The assessment of risk for the development of AMD requires reliable detection and quantitative mapping of retinal abnormalities that are considered as precursors of the disease. Typical signs of the latter are the so‐called drusen that appear as yellowish spots in the retina. Automated detection and segmentation of drusen provide vital information about the severity of the disease. The authors propose a novel method for the detection and segmentation of drusen in colour fundus images. The method combines colour information of the object with its boundary information for the accurate detection and segmentation of drusen. To perform non‐uniform illumination correction and to minimise inter‐subject variability a novel colour normalisation method has been proposed. Experiments are conducted on publicly available STARE and ARIA datasets. The method achieves an overall accuracy of 96.62% which is about 4% higher than the state‐of‐the‐art method. The sensitivity and specificity of the proposed method are 95.96 and 97.64%, respectively.