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

doi:10.1109/ITT59889.2023.10184244

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 proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

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 language	English
Title of host publication	2023 9th International Conference on Information Technology Trends
Subtitle of host publication	ITT 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	145-150
Number of pages	6
ISBN (Electronic)	9798350327502
ISBN (Print)	9798350327519
DOIs	https://doi.org/10.1109/ITT59889.2023.10184244
Publication status	Published - 24 Jul 2023
Event	9th 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 2023 → 25 May 2023 Conference number: 9 https://ieee.ae/event/the-9th-international-conference-on-information-technology-trends-itt-2023/ https://hct.ac.ae/en/events/information-technology-trends-itt-2023/

Conference

Conference	9th International IEEE Conference in Information Technology Trends 2023
Abbreviated title	ITT 2023
Country/Territory	United Arab Emirates
City	Dubai
Period	24/05/23 → 25/05/23
Internet address	https://ieee.ae/event/the-9th-international-conference-on-information-technology-trends-itt-2023/ https://hct.ac.ae/en/events/information-technology-trends-itt-2023/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ITT59889.2023.10184244Licence: Unspecified

Cite this

Aydin, B. A., Hussain, M., Hill, R., & Al-Aqrabi, H. (2023). Domain Modelling For A Lightweight Convolutional Network Focused On Automated Exudate Detection in Retinal Fundus Images. In 2023 9th International Conference on Information Technology Trends: ITT 2023 (pp. 145-150). [10184244] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ITT59889.2023.10184244

@inproceedings{35d08eb9d7bc4f4d8be5f519ddb06c7f,

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

abstract = "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%.",

keywords = "Computer Vision, Convolutional Networks, Data Augmentations, Exudates, Lightweight",

author = "Aydin, {Burcu Ataer} and Muhammad Hussain and Richard Hill and Hussain Al-Aqrabi",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 9th International IEEE Conference in Information Technology Trends 2023 : The Application of AI in Sustainable Computing, ITT 2023 ; Conference date: 24-05-2023 Through 25-05-2023",

year = "2023",

month = jul,

day = "24",

doi = "10.1109/ITT59889.2023.10184244",

language = "English",

isbn = "9798350327519",

pages = "145--150",

booktitle = "2023 9th International Conference on Information Technology Trends",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

url = "https://ieee.ae/event/the-9th-international-conference-on-information-technology-trends-itt-2023/, https://hct.ac.ae/en/events/information-technology-trends-itt-2023/",

}

Aydin, BA, Hussain, M , Hill, R & Al-Aqrabi, H 2023, Domain Modelling For A Lightweight Convolutional Network Focused On Automated Exudate Detection in Retinal Fundus Images. in 2023 9th International Conference on Information Technology Trends: ITT 2023., 10184244, Institute of Electrical and Electronics Engineers Inc., pp. 145-150, 9th International IEEE Conference in Information Technology Trends 2023, Dubai, United Arab Emirates, 24/05/23. https://doi.org/10.1109/ITT59889.2023.10184244

Domain Modelling For A Lightweight Convolutional Network Focused On Automated Exudate Detection in Retinal Fundus Images. / Aydin, Burcu Ataer; Hussain, Muhammad ; Hill, Richard et al.

2023 9th International Conference on Information Technology Trends: ITT 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 145-150 10184244.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Domain Modelling For A Lightweight Convolutional Network Focused On Automated Exudate Detection in Retinal Fundus Images

AU - Aydin, Burcu Ataer

AU - Hussain, Muhammad

AU - Hill, Richard

AU - Al-Aqrabi, Hussain

N1 - Conference code: 9

PY - 2023/7/24

Y1 - 2023/7/24

N2 - 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%.

AB - 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%.

KW - Computer Vision

KW - Convolutional Networks

KW - Data Augmentations

KW - Exudates

KW - Lightweight

UR - http://www.scopus.com/inward/record.url?scp=85168561374&partnerID=8YFLogxK

U2 - 10.1109/ITT59889.2023.10184244

DO - 10.1109/ITT59889.2023.10184244

M3 - Conference contribution

AN - SCOPUS:85168561374

SN - 9798350327519

SP - 145

EP - 150

BT - 2023 9th International Conference on Information Technology Trends

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 9th International IEEE Conference in Information Technology Trends 2023

Y2 - 24 May 2023 through 25 May 2023

ER -

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

Abstract

Conference

UN SDGs

Access to Document

Other files and links

Cite this