Cover Image

A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis

Diabetic Retinopathy (DR) is the most common complication of Diabetes Mellitus and can lead to blindness if not detected early. Since DR is often asymptomatic in its early stage, timely diagnosis is crucial. Artificial Intelligence (AI) has the potential to facilitate early disease detection and tre...

Full description

Saved in:
Bibliographic Details
Main Authors: Ng, Jun Chen, Yeoh, Pauline Shan Qing, Bing, Li, Wu, Xiang, Hasikin, Khairunnisa, Lai, Khin Wee
Format: Article
Published: Institute of Electrical and Electronics Engineers 2024
Subjects:
R Medicine (General)
T Technology (General)
Online Access:http://eprints.um.edu.my/47078/
https://doi.org/10.1109/ACCESS.2024.3469537
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.um.eprints.47078
record_format eprints
spelling my.um.eprints.470782025-01-03T08:44:00Z http://eprints.um.edu.my/47078/ A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis Ng, Jun Chen Yeoh, Pauline Shan Qing Bing, Li Wu, Xiang Hasikin, Khairunnisa Lai, Khin Wee R Medicine (General) T Technology (General) Diabetic Retinopathy (DR) is the most common complication of Diabetes Mellitus and can lead to blindness if not detected early. Since DR is often asymptomatic in its early stage, timely diagnosis is crucial. Artificial Intelligence (AI) has the potential to facilitate early disease detection and treatment, but its implementation in the medical field raises significant privacy concerns. The sensitive nature of healthcare data, which includes personal information and medical history, makes data privacy a critical issue. This paper explores the implementation of AI models to predict DR risks while incorporating common defense algorithms to enhance data privacy. An unstructured dataset, specifically the DDR dataset, was used to train Deep Learning (DL) models. Two families of DL models, ResNets and DenseNets, were trained and evaluated based on the performance metrics. ResNet 50 and DenseNet 169 demonstrated superior performance and were selected for further privacy enhancement using encryption. The results indicated that privacy-preserving methods, particularly encryption, did not significantly impact the model performance. In summary, this paper highlights the potential of privacy-preserving AI in predicting the risks of DR. Institute of Electrical and Electronics Engineers 2024 Article PeerReviewed Ng, Jun Chen and Yeoh, Pauline Shan Qing and Bing, Li and Wu, Xiang and Hasikin, Khairunnisa and Lai, Khin Wee (2024) A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis. IEEE Access, 12. pp. 145159-145173. ISSN 2169-3536, DOI https://doi.org/10.1109/ACCESS.2024.3469537 <https://doi.org/10.1109/ACCESS.2024.3469537>. https://doi.org/10.1109/ACCESS.2024.3469537 10.1109/ACCESS.2024.3469537
institution Universiti Malaya
building UM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaya
content_source UM Research Repository
url_provider http://eprints.um.edu.my/
topic R Medicine (General)
T Technology (General)
spellingShingle R Medicine (General)
T Technology (General)
Ng, Jun Chen
Yeoh, Pauline Shan Qing
Bing, Li
Wu, Xiang
Hasikin, Khairunnisa
Lai, Khin Wee
A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis
description Diabetic Retinopathy (DR) is the most common complication of Diabetes Mellitus and can lead to blindness if not detected early. Since DR is often asymptomatic in its early stage, timely diagnosis is crucial. Artificial Intelligence (AI) has the potential to facilitate early disease detection and treatment, but its implementation in the medical field raises significant privacy concerns. The sensitive nature of healthcare data, which includes personal information and medical history, makes data privacy a critical issue. This paper explores the implementation of AI models to predict DR risks while incorporating common defense algorithms to enhance data privacy. An unstructured dataset, specifically the DDR dataset, was used to train Deep Learning (DL) models. Two families of DL models, ResNets and DenseNets, were trained and evaluated based on the performance metrics. ResNet 50 and DenseNet 169 demonstrated superior performance and were selected for further privacy enhancement using encryption. The results indicated that privacy-preserving methods, particularly encryption, did not significantly impact the model performance. In summary, this paper highlights the potential of privacy-preserving AI in predicting the risks of DR.
format Article
author Ng, Jun Chen
Yeoh, Pauline Shan Qing
Bing, Li
Wu, Xiang
Hasikin, Khairunnisa
Lai, Khin Wee
author_facet Ng, Jun Chen
Yeoh, Pauline Shan Qing
Bing, Li
Wu, Xiang
Hasikin, Khairunnisa
Lai, Khin Wee
author_sort Ng, Jun Chen
title A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis
title_short A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis
title_full A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis
title_fullStr A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis
title_full_unstemmed A Privacy-Preserving Approach Using Deep Learning Models for Diabetic Retinopathy Diagnosis
title_sort privacy-preserving approach using deep learning models for diabetic retinopathy diagnosis
publisher Institute of Electrical and Electronics Engineers
publishDate 2024
url http://eprints.um.edu.my/47078/
https://doi.org/10.1109/ACCESS.2024.3469537
_version_ 1821001885420617728
score 13.23648

Similar Items