Automatic glaucoma classification using color fundus images based on convolutional neural networks and transfer learning
Gómez-Valverde, Juan J.
Antón López, Alfonso
Sánchez, Clara I.
Ledesma-Carbayo, María J.
Glaucoma detection in color fundus images is a challenging task that requires expertise and years of practice. In this study we exploited the application of different Convolutional Neural Networks (CNN) schemes to show the influence in the performance of relevant factors like the data set size, the architecture and the use of transfer learning vs newly defined architectures. We also compared the performance of the CNN based system with respect to human evaluators and explored the influence of the integration of images and data collected from the clinical history of the patients. We accomplished the best performance using a transfer learning scheme with VGG19 achieving an AUC of 0.94 with sensitivity and specificity ratios similar to the expert evaluators of the study. The experimental results using three different data sets with 2313 images indicate that this solution can be a valuable option for the design of a computer aid system for the detection of glaucoma in large-scale screening programs.
61 - Medicina
617 - Cirurgia. Ortopèdia. Oftalmologia
Optical Society of America
Is part of
Biomedical Optics Express
Gómez-Valverde, Juan J.; Antón López, Alfonso; Fatti, Gianluca; Liefers, Bart; Herranz, Alejandra; Santos, Andrés; Sánchez, Clara I.; Ledesma-Carbayo, María J. «Automatic glaucoma classification using color fundus images based on convolutional neural networks and transfer learning». Biomedical Optics Express, 2019, vol. 10, núm. 2, p. 892-913. Disponible en: <https://www.osapublishing.org/boe/abstract.cfm?uri=boe-10-2-892>. Fecha de acceso: 26 sept. 2019. DOI: https://doi.org/10.1364/BOE.10.000892
This item appears in the following Collection(s)
The following license files are associated with this item:
Except where otherwise noted, this item's license is described as http://creativecommons.org/licenses/by-nc-nd/4.0/