A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning
Among various eye diseases, glaucoma is one of the leading causes of blindness. Glaucoma is also one of the most common eye diseases in Taiwan. Glaucoma screenings can use optical coherence tomography (OCT) to locate areas in which the retinal nerve fiber layer is thinning. However, because OCT equi...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-02-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/13/4/2195 |
_version_ | 1797622632238022656 |
---|---|
author | Chuan-Wang Chang Chuan-Yu Chang Yuan-Yi Lin Wei-Wen Su Henry Shen-Lih Chen |
author_facet | Chuan-Wang Chang Chuan-Yu Chang Yuan-Yi Lin Wei-Wen Su Henry Shen-Lih Chen |
author_sort | Chuan-Wang Chang |
collection | DOAJ |
description | Among various eye diseases, glaucoma is one of the leading causes of blindness. Glaucoma is also one of the most common eye diseases in Taiwan. Glaucoma screenings can use optical coherence tomography (OCT) to locate areas in which the retinal nerve fiber layer is thinning. However, because OCT equipment is costly, only large hospitals with well-equipped facilities will have OCT, and regular eye clinics cannot afford such expensive equipment. This has caused many glaucoma patients to worsen because they cannot get an early diagnosis in regular eye clinics in time. This paper proposes a method of using a generative adversarial network (GAN) to generate corresponding OCT images from fundus images to assist family doctors in judging whether further examination is needed based on the generated OCT images to achieve early detection and treatment of glaucoma. In addition, in order to improve the classification accuracy of the system deployed in different hospitals or clinics, this paper also proposes to use the incremental training method to fine-tune the model. The model can be quickly applied by adding a small number of images from a specific clinic or hospital. Experimental results show that the cosine similarity between the generated OCT image and the real OCT image is 97.8%. Combined with the proposed transfer learning method, the classification accuracy of the classification model reaches 83.17%. As well as the use of the incremental method, the accuracy of identifying glaucoma is approximately 78.94%, which is 8.77% higher than the 70.17% accuracy of the initial model. Experimental results show the effectiveness and feasibility of our proposed method. |
first_indexed | 2024-03-11T09:13:06Z |
format | Article |
id | doaj.art-789a14fd717e4ccb8c62c3375741aa70 |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-11T09:13:06Z |
publishDate | 2023-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-789a14fd717e4ccb8c62c3375741aa702023-11-16T18:52:15ZengMDPI AGApplied Sciences2076-34172023-02-01134219510.3390/app13042195A Glaucoma Detection System Based on Generative Adversarial Network and Incremental LearningChuan-Wang Chang0Chuan-Yu Chang1Yuan-Yi Lin2Wei-Wen Su3Henry Shen-Lih Chen4Department of CSIE, National Chin-Yi University of Technology, Taichung City 411030, TaiwanDepartment of CSIE, National Yunlin University of Science and Technology, Douliu City 640301, TaiwanDepartment of CSIE, National Yunlin University of Science and Technology, Douliu City 640301, TaiwanGlaucoma Division, Chang Gung Memorial Hospital, Taoyuan City 333423, TaiwanGlaucoma Division, Chang Gung Memorial Hospital, Taoyuan City 333423, TaiwanAmong various eye diseases, glaucoma is one of the leading causes of blindness. Glaucoma is also one of the most common eye diseases in Taiwan. Glaucoma screenings can use optical coherence tomography (OCT) to locate areas in which the retinal nerve fiber layer is thinning. However, because OCT equipment is costly, only large hospitals with well-equipped facilities will have OCT, and regular eye clinics cannot afford such expensive equipment. This has caused many glaucoma patients to worsen because they cannot get an early diagnosis in regular eye clinics in time. This paper proposes a method of using a generative adversarial network (GAN) to generate corresponding OCT images from fundus images to assist family doctors in judging whether further examination is needed based on the generated OCT images to achieve early detection and treatment of glaucoma. In addition, in order to improve the classification accuracy of the system deployed in different hospitals or clinics, this paper also proposes to use the incremental training method to fine-tune the model. The model can be quickly applied by adding a small number of images from a specific clinic or hospital. Experimental results show that the cosine similarity between the generated OCT image and the real OCT image is 97.8%. Combined with the proposed transfer learning method, the classification accuracy of the classification model reaches 83.17%. As well as the use of the incremental method, the accuracy of identifying glaucoma is approximately 78.94%, which is 8.77% higher than the 70.17% accuracy of the initial model. Experimental results show the effectiveness and feasibility of our proposed method.https://www.mdpi.com/2076-3417/13/4/2195deep learningunsupervised learninggenerative adversarial networkincremental learning |
spellingShingle | Chuan-Wang Chang Chuan-Yu Chang Yuan-Yi Lin Wei-Wen Su Henry Shen-Lih Chen A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning Applied Sciences deep learning unsupervised learning generative adversarial network incremental learning |
title | A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning |
title_full | A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning |
title_fullStr | A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning |
title_full_unstemmed | A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning |
title_short | A Glaucoma Detection System Based on Generative Adversarial Network and Incremental Learning |
title_sort | glaucoma detection system based on generative adversarial network and incremental learning |
topic | deep learning unsupervised learning generative adversarial network incremental learning |
url | https://www.mdpi.com/2076-3417/13/4/2195 |
work_keys_str_mv | AT chuanwangchang aglaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT chuanyuchang aglaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT yuanyilin aglaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT weiwensu aglaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT henryshenlihchen aglaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT chuanwangchang glaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT chuanyuchang glaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT yuanyilin glaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT weiwensu glaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning AT henryshenlihchen glaucomadetectionsystembasedongenerativeadversarialnetworkandincrementallearning |