EyeWeS: Weakly supervised pre-trained convolutional neural networks for diabetic retinopathy detection

  • Pedro Costa
  • , Teresa Araujo
  • , Guilherme Aresta
  • , Adrian Galdran
  • , Ana Maria Mendonca
  • , Asim Smailagic
  • , Aurelio Campilho

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

27 Citations (Scopus)

Abstract

Diabetic Retinopathy (DR) is one of the leading causes of preventable blindness in the developed world. With the increasing number of diabetic patients there is a growing need of an automated system for DR detection. We propose Eye WeS, a method that not only detects DR in eye fundus images but also pinpoints the regions of the image that contain lesions, while being trained with image labels only. We show that it is possible to convert any pre-trained convolutional neural network into a weakly-supervised model while increasing their performance and efficiency. EyeWeS improved the results of Inception V3 from 94.9% Area Under the Receiver Operating Curve (AUC) to 95.8% AUC while maintaining only approximately 5% of the Inception V3's number of parameters. The same model is able to achieve 97.1% AUC in a cross-dataset experiment.

Original languageEnglish
Title of host publicationProceedings of the 16th International Conference on Machine Vision Applications, MVA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9784901122184
DOIs
Publication statusPublished - May 2019
Externally publishedYes
Event16th International Conference on Machine Vision Applications, MVA 2019 - Tokyo, Japan
Duration: 27 May 201931 May 2019

Publication series

NameProceedings of the 16th International Conference on Machine Vision Applications, MVA 2019

Conference

Conference16th International Conference on Machine Vision Applications, MVA 2019
Country/TerritoryJapan
CityTokyo
Period27/05/1931/05/19

Fingerprint

Dive into the research topics of 'EyeWeS: Weakly supervised pre-trained convolutional neural networks for diabetic retinopathy detection'. Together they form a unique fingerprint.

Cite this