Implementación y evaluación de un sistema de detección mediante la captura de imágenes para la clasificación de estrabismo, utilizando redes neuronales convolucionales

thumbnail.default.alt
Share
Date
2021-11-24
Publisher
Universidad Antonio Nariño
Document type
COAR type
http://purl.org/coar/resource_type/c_7a1f
Citation
Bibliographic Managers
Document Viewer
Select a file to preview:
item.page.resume
Abstract
The diagnosis of strabismus, is very important to do in time during childhood, strabismus affects between 2% and 4% of the world population in children because this condition produces amblyopia, which consists of the loss of vision in the deviated eye, once developed the amblyopia, it cannot be treated, because the brain inhibits the signal from the deviated eye, resulting in the gradual and permanent loss of visual acuity in the affected eye.Convolutional neural networks were used for this study, In order to detect strabismus in patient images, the model used is DenseNet 201, an architecture designed for image classification tasks, trained by a set of own images, acquired by the authors, consisting of 332 images.
item.page.coverage.spatial
Hulia
item.page.coverage.temporal