A coarse-to-fine deep learning framework for optic disc segmentation in fundus images. (May 2019)

• Record Type: Journal Article
• Authors: Wang, Lei; Liu, Han; Lu, Yaling; Chen, Hang; Zhang, Jian; Pu, Jiantao
• Is Part Of: Biomedical signal processing and control. Volume 51(2019)
• Page Start: 82
• Page End: 89
• Publication Date: 2019-05
• Subjects: Image segmentation -- Optic disc -- Convolutional neural networks -- U-net model -- Color fundus images; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2019.01.022
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: A coarse-to-file deep learning framework developed on the basis of the U-net model. A novel vessel density map developed to highlight the location of the optic disc. Performance comparison with available U-net model and other optic disc segmentation methods. Abstract: Accurate segmentation of the optic disc (OD) depicted on color fundus images may aid in the early detection and quantitative diagnosis of retinal diseases, such as glaucoma and optic atrophy. In this study, we proposed a coarse-to-fine deep learning framework on the basis of a classical convolutional neural network (CNN), known as the U-net model, to accurately identify the optic disc. This network was trained separately on color fundus images and their grayscale vessel density maps, leading to two different segmentation results from the entire image. We combined the results using an overlap strategy to identify a local image patch (disc candidate region), which was then fed into the U-net model for further segmentation. Our experiments demonstrated that the developed framework achieved an average intersection over union (IoU) and a dice similarity coefficient (DSC) of 89.1% and 93.9%, respectively, based on 2978 test images from our collected dataset and six public datasets, as compared to 87.4% and 92.5% obtained by only using the sole U-net model. The comparison with available approaches demonstrated a reliable and relatively high performance of the proposed deep learning framework in automated OD segmentation.