eprintid: 64271 rev_number: 26 eprint_status: archive userid: 1903 dir: disk0/00/06/42/71 datestamp: 2020-10-07 14:37:22 lastmod: 2020-10-07 14:37:22 status_changed: 2020-10-07 14:37:22 type: article metadata_visibility: show creators_name: Gómez Valverde, Juan José creators_name: Antón López, Alfonso creators_name: Fatti, Gianluca creators_name: Liefers, Bart creators_name: Herranz Cabarcos, Alejandra creators_name: Santos Lleo, Andres de creators_name: Sánchez, Clara, I creators_name: Ledesma Carbayo, María Jesús creators_id: andres@die.upm.es creators_id: mledesma@die.upm.es title: Automatic glaucoma classification using color fundus images based on convolutional neural networks and transfer learning publisher: Optical Society of America rights: by-nc-nd ispublished: pub subjects: electronica subjects: medicina subjects: telecomunicaciones full_text_status: public abstract: 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. date_type: published date: 2019-01-25 publication: Biomedical Optics Express volume: 10 number: 2 pagerange: 892-913 id_number: 10.1364/BOE.10.000892 institution: Telecomunicacion department: Ingenieria_Electronica refereed: TRUE issn: 2156-7085 official_url: https://www.osapublishing.org/boe/abstract.cfm?uri=boe-10-2-892 comprojects_type: MINECO comprojects_code: TEC2015-66978-R comprojects_title: Tecnología óptica para elastografía del tejido citation: Gómez Valverde, Juan José and Antón López, Alfonso and Fatti, Gianluca and Liefers, Bart and Herranz Cabarcos, Alejandra and Santos Lleo, Andres de and Sánchez, Clara, I and Ledesma Carbayo, María Jesús (2019). Automatic glaucoma classification using color fundus images based on convolutional neural networks and transfer learning. "Biomedical Optics Express", v. 10 (n. 2); pp. 892-913. ISSN 2156-7085. https://doi.org/10.1364/BOE.10.000892 . document_url: http://oa.upm.es/64271/1/INVE_MEM_2019_325141.pdf