Domain Modelling For A Lightweight Convolutional Network Focused On Automated Exudate Detection in Retinal Fundus Images

Burcu Ataer Aydin, Muhammad Hussain, Richard Hill, Hussain Al-Aqrabi

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

2 Citations (Scopus)


Retinal Fundus images contain important signs implying various stages of diabetic retinopathy. One of the early signs of diabetic retinopathy is exudates. The timely detection of this sign from fundus images can prevent or at least suppress the advancement of this condition. To this effect, this research presents a CNN-based automated exudate detection architecture for the timely detection of this early sign. In order to overcome the challenges of a limited dataset several domain specific augmentations are proposed for improving the generalization capacity of the developed architecture. The lightweight architecture consists of only 6.42 million parameters, compared to Resnet-18 (11.69 million) whilst achieving an overall F1 score of 89%.
Original languageEnglish
Title of host publication2023 9th International Conference on Information Technology Trends
Subtitle of host publicationITT 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages6
ISBN (Electronic)9798350327502
ISBN (Print)9798350327519
Publication statusPublished - 24 Jul 2023
Event9th International IEEE Conference in Information Technology Trends 2023: The Application of AI in Sustainable Computing - HCT-Dubai Men’s campus, Dubai, United Arab Emirates
Duration: 24 May 202325 May 2023
Conference number: 9


Conference9th International IEEE Conference in Information Technology Trends 2023
Abbreviated titleITT 2023
Country/TerritoryUnited Arab Emirates
Internet address

Cite this