Abstract
The localization and segmentation of optic disc (OD) in fundus images is a crucial step in the pipeline for detecting the early onset of retinal diseases, such as macular degeneration, diabetic retinopathy, glaucoma, etc. In this paper, we are proposing a novel convolutional neural network architecture for the precise segmentation of the OD in fundus images. We modify the basic architectures of DeepLab v3+ and U-Net models by integrating a novel attention module between the encoder and decoder to attain the finest accuracy. We also use fully-connected conditional random fields to further boost the performance of these architectures. We compare the results of our best proposed architecture against other established architectures for optic disc segmentation on our private dataset, as well as on publicly available datasets, namely, DRIONS-DB, RIM-ONE v.3, and DRISHTI-GS. The results obtained with the proposed method outperforms the existing methods in the literature.
Highlights
Recent advancements in deep learning and computer vision have shown the effectiveness of Convolutional Neural Networks (CNNs) in solving challenging tasks such as image classification, image segmentation, image captioning, object detection and tracking, surpassing traditional algorithms, and achieving state-of-the-art results
We use our private dataset to check how these architectures perform on a wide range of fundus images with differing qualities
In this paper, we show that attention mechanisms and conditional random fields (CRFs) can be used to boost the performance of deep convolutional neural network based models for optic disc (OD) segmentation
Summary
Recent advancements in deep learning and computer vision have shown the effectiveness of Convolutional Neural Networks (CNNs) in solving challenging tasks such as image classification, image segmentation, image captioning, object detection and tracking, surpassing traditional algorithms, and achieving state-of-the-art results. Segmentation is a significant medical imaging task, as the automatic delineation of biological structures of importance is required for the automatic detection of disease, computer-. The precise segmentation of OD in fundus images is a challenging problem primarily owing to the retinal diseases bring in the pathological changes in the anatomy of the retina. In many cases, the quality of fundus images is not good enough to detect the OD precisely.The reasons like image distortions, noise introduced, and lack of technical expertise of the technician are the major causes of degraded fundus image quality. Figure [something] shows three fundus images having different levels of visibility of the OD with respect to its appearance
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
