Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Machine learning for glaucoma detection using fundus images

Machine learning for glaucoma detection using fundus images PurposeGlaucoma damages the optic nerve and causes permanent visual impairment. It cannot be recovered, so it is important to distinguish the disease over time. Fortunately, this is usually a state of progress, and if picked early, it is likely to be treated effectively. Early detection is the key to success preventing visual impairment. In this article, we will talk about how to diagnose glaucoma by means of two main steps: segmentation and classification using convolutional neural networks (CNN).MethodsFirst, we present different implementations of segmentation and classification on a database that contains a set of fundus data. This article proposes supervised and unsupervised methods for segmentation, detection, and classification of glaucoma in fundus images. A supervised method using U-net and Modified U-net (M-net) is first formed to segment and detect the optical disc and cup disc regions in fundus images. Then classing them in the next step. We are training proposed networks by using a graphics processor and public datasets. The unsupervised method based on K-means algorithm for optic and cup disc segmentation, then calculates CDR for classification to glaucoma or not.ResultsOur proposed M-net achieves an accuracy of 99.89% and unsupervised methods achieve an accuracy of 98.17%.ConclusionWith the highlights of recent advancements in deep learning-based approaches for glaucoma diagnosis. These techniques have shown promising improvement in detection performance.The results obtained have shown a significant improvement in the diagnostic performance of glaucoma abnormality. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Research on Biomedical Engineering Springer Journals

Machine learning for glaucoma detection using fundus images

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/machine-learning-for-glaucoma-detection-using-fundus-images-ZDPnqR7dmk

References (26)

Publisher
Springer Journals
Copyright
Copyright © The Author(s), under exclusive licence to The Brazilian Society of Biomedical Engineering 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
ISSN
2446-4732
eISSN
2446-4740
DOI
10.1007/s42600-023-00305-8
Publisher site
See Article on Publisher Site

Abstract

PurposeGlaucoma damages the optic nerve and causes permanent visual impairment. It cannot be recovered, so it is important to distinguish the disease over time. Fortunately, this is usually a state of progress, and if picked early, it is likely to be treated effectively. Early detection is the key to success preventing visual impairment. In this article, we will talk about how to diagnose glaucoma by means of two main steps: segmentation and classification using convolutional neural networks (CNN).MethodsFirst, we present different implementations of segmentation and classification on a database that contains a set of fundus data. This article proposes supervised and unsupervised methods for segmentation, detection, and classification of glaucoma in fundus images. A supervised method using U-net and Modified U-net (M-net) is first formed to segment and detect the optical disc and cup disc regions in fundus images. Then classing them in the next step. We are training proposed networks by using a graphics processor and public datasets. The unsupervised method based on K-means algorithm for optic and cup disc segmentation, then calculates CDR for classification to glaucoma or not.ResultsOur proposed M-net achieves an accuracy of 99.89% and unsupervised methods achieve an accuracy of 98.17%.ConclusionWith the highlights of recent advancements in deep learning-based approaches for glaucoma diagnosis. These techniques have shown promising improvement in detection performance.The results obtained have shown a significant improvement in the diagnostic performance of glaucoma abnormality.

Journal

Research on Biomedical Engineering – Springer Journals

Published: Sep 7, 2023

Keywords: Artificial intelligence; CAD system; Convolutional neural networks; Deep learning; Diabetic retinopathy; Funds images; Segmentation

There are no references for this article.