Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/118526
Title: Diagnóstico de miopía patológica en imágenes de fondo de ojo mediante aprendizaje profundo
Author: López Robles, Daniel
Tutor: Nuñez Do Rio, Joan Manuel
Others: Ventura, Carles  
Abstract: Myopia is a global public health problem. Its most severe form, called pathological myopia, can lead to irreversible loss of vision. The complexity of the diagnosis is aggravated by the emerging issues of population ageing around the world. The rate of people over 60 is growing twice as high as that of ophthalmology professionals. Therefore, it is essential to promote the use of new technologies for an early diagnosis of pathological myopia. Artificial Intelligence provides promising tools to facilitate clinical diagnosis through Deep Learning. Convolutional Neural Networks, specialized in image processing, allow to address complex problems such as medical image classification. This paper explores convolutional neural network-based solutions for detecting pathological myopia in fundus images, including the use of transfer learning with the VGGNet, ResNet and GoogleNet models. The results achieved demonstrate the potential of Deep Learning to contribute to the early diagnosis of pathological myopia.
Keywords: eye fundus image
deep learning
pathologic myopia
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: 19-Jun-2020
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.