A lightweight deep learning model for mobile eye fundus image quality assessment

Andrés D. Pérez, Oscar Julian Perdomo Charry, Fabio A. González

Resultado de la investigación: Capítulo en Libro/Reporte/ConferenciaContribución a la conferencia

3 Citas (Scopus)

Resumen

Image acquisition and automatic quality analysis are fundamental stages and tasks to support an accurate ocular diagnosis. In particular, when eye fundus image quality is not appropriate, it can hinder the diagnosis task performed by experts. Portable, smart-phone-based eye fundus image acquisition devices have the advantage of their low cost and easy deployment, however, their main disadvantage is the sacrifice of image quality. This paper presents a deep-learning-based model to assess the eye fundus image quality which is small enough to be deployed in a smart phone. The model was evaluated in a public eye fundus dataset with two sets of annotations. The proposed method obtained an accuracy of 0.911 and 0.856, in the binary classification task and the three-classes classification task respectively. Besides, the presented method has a small number of parameters compared to other state-of-the-art models, being an alternative for a mobile-based eye fundus quality classification system.

Idioma originalInglés estadounidense
Título de la publicación alojada15th International Symposium on Medical Information Processing and Analysis
EditoresEduardo Romero, Natasha Lepore, Jorge Brieva
EditorialSPIE
ISBN (versión digital)9781510634275
ISBN (versión impresa)9781510634275, 9781510634282
DOI
EstadoPublicada - ene 3 2020
Evento15th International Symposium on Medical Information Processing and Analysis, SIPAIM 2019 - Medellin, Colombia
Duración: nov 6 2019nov 8 2019

Serie de la publicación

NombreProceedings of SPIE - The International Society for Optical Engineering
Volumen11330
ISSN (versión impresa)0277-786X
ISSN (versión digital)1996-756X

Conferencia

Conferencia15th International Symposium on Medical Information Processing and Analysis, SIPAIM 2019
PaísColombia
CiudadMedellin
Período11/6/1911/8/19

All Science Journal Classification (ASJC) codes

  • Materiales electrónicos, ópticos y magnéticos
  • Física de la materia condensada
  • Informática aplicada
  • Matemáticas aplicadas
  • Ingeniería eléctrica y electrónica

Huella Profundice en los temas de investigación de 'A lightweight deep learning model for mobile eye fundus image quality assessment'. En conjunto forman una huella única.

Citar esto