Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis

<b>Purpose:</b> To evaluate the performance of deep learning (DL) in diagnosing glaucoma and predicting its progression using fundus photography and retinal optical coherence tomography (OCT) images. <b>Materials and Methods:</b> Relevant studies published up to 30 October 20...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiao Chun Ling, Henry Shen-Lih Chen, Po-Han Yeh, Yu-Chun Cheng, Chu-Yen Huang, Su-Chin Shen, Yung-Sung Lee
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Biomedicines
Subjects:
Online Access:https://www.mdpi.com/2227-9059/13/2/420
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849718837981741056
author Xiao Chun Ling
Henry Shen-Lih Chen
Po-Han Yeh
Yu-Chun Cheng
Chu-Yen Huang
Su-Chin Shen
Yung-Sung Lee
author_facet Xiao Chun Ling
Henry Shen-Lih Chen
Po-Han Yeh
Yu-Chun Cheng
Chu-Yen Huang
Su-Chin Shen
Yung-Sung Lee
author_sort Xiao Chun Ling
collection DOAJ
description <b>Purpose:</b> To evaluate the performance of deep learning (DL) in diagnosing glaucoma and predicting its progression using fundus photography and retinal optical coherence tomography (OCT) images. <b>Materials and Methods:</b> Relevant studies published up to 30 October 2024 were retrieved from PubMed, Medline, EMBASE, Cochrane Library, Web of Science, and ClinicalKey. A bivariate random-effects model was employed to calculate pooled sensitivity, specificity, positive and negative likelihood ratios, and area under the receiver operating characteristic curve (AUROC). <b>Results:</b> A total of 48 studies were included in the meta-analysis. DL algorithms demonstrated high diagnostic performance in glaucoma detection using fundus photography and OCT images. For fundus photography, the pooled sensitivity and specificity were 0.92 (95% CI: 0.89–0.94) and 0.93 (95% CI: 0.90–0.95), respectively, with an AUROC of 0.90 (95% CI: 0.88–0.92). For the OCT imaging, the pooled sensitivity and specificity were 0.90 (95% CI: 0.84–0.94) and 0.87 (95% CI: 0.81–0.91), respectively, with an AUROC of 0.86 (95% CI: 0.83–0.90). In predicting glaucoma progression, DL models generally showed less robust performance, with pooled sensitivities and specificities ranging lower than in diagnostic tasks. Internal validation datasets showed higher accuracy than external validation datasets. <b>Conclusions:</b> DL algorithms achieve excellent performance in diagnosing glaucoma using fundus photography and OCT imaging. To enhance the prediction of glaucoma progression, future DL models should integrate multimodal data, including functional assessments, such as visual field measurements, and undergo extensive validation in real-world clinical settings.
format Article
id doaj-art-eb77eac9738b4d21891b086725daed2a
institution DOAJ
issn 2227-9059
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Biomedicines
spelling doaj-art-eb77eac9738b4d21891b086725daed2a2025-08-20T03:12:16ZengMDPI AGBiomedicines2227-90592025-02-0113242010.3390/biomedicines13020420Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-AnalysisXiao Chun Ling0Henry Shen-Lih Chen1Po-Han Yeh2Yu-Chun Cheng3Chu-Yen Huang4Su-Chin Shen5Yung-Sung Lee6Department of Ophthalmology, Chang Gung Memorial Hospital, Linkou, Taoyuan 333, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou, Taoyuan 333, TaiwanDepartment of Ophthalmology, New Taipei Municipal Tucheng Hospital, New Taipei 236, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou, Taoyuan 333, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou, Taoyuan 333, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou, Taoyuan 333, TaiwanDepartment of Ophthalmology, Chang Gung Memorial Hospital, Linkou, Taoyuan 333, Taiwan<b>Purpose:</b> To evaluate the performance of deep learning (DL) in diagnosing glaucoma and predicting its progression using fundus photography and retinal optical coherence tomography (OCT) images. <b>Materials and Methods:</b> Relevant studies published up to 30 October 2024 were retrieved from PubMed, Medline, EMBASE, Cochrane Library, Web of Science, and ClinicalKey. A bivariate random-effects model was employed to calculate pooled sensitivity, specificity, positive and negative likelihood ratios, and area under the receiver operating characteristic curve (AUROC). <b>Results:</b> A total of 48 studies were included in the meta-analysis. DL algorithms demonstrated high diagnostic performance in glaucoma detection using fundus photography and OCT images. For fundus photography, the pooled sensitivity and specificity were 0.92 (95% CI: 0.89–0.94) and 0.93 (95% CI: 0.90–0.95), respectively, with an AUROC of 0.90 (95% CI: 0.88–0.92). For the OCT imaging, the pooled sensitivity and specificity were 0.90 (95% CI: 0.84–0.94) and 0.87 (95% CI: 0.81–0.91), respectively, with an AUROC of 0.86 (95% CI: 0.83–0.90). In predicting glaucoma progression, DL models generally showed less robust performance, with pooled sensitivities and specificities ranging lower than in diagnostic tasks. Internal validation datasets showed higher accuracy than external validation datasets. <b>Conclusions:</b> DL algorithms achieve excellent performance in diagnosing glaucoma using fundus photography and OCT imaging. To enhance the prediction of glaucoma progression, future DL models should integrate multimodal data, including functional assessments, such as visual field measurements, and undergo extensive validation in real-world clinical settings.https://www.mdpi.com/2227-9059/13/2/420glaucomadeep learningartificial intelligencefundus photographyoptical coherence tomographydiagnosis
spellingShingle Xiao Chun Ling
Henry Shen-Lih Chen
Po-Han Yeh
Yu-Chun Cheng
Chu-Yen Huang
Su-Chin Shen
Yung-Sung Lee
Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis
Biomedicines
glaucoma
deep learning
artificial intelligence
fundus photography
optical coherence tomography
diagnosis
title Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis
title_full Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis
title_fullStr Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis
title_full_unstemmed Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis
title_short Deep Learning in Glaucoma Detection and Progression Prediction: A Systematic Review and Meta-Analysis
title_sort deep learning in glaucoma detection and progression prediction a systematic review and meta analysis
topic glaucoma
deep learning
artificial intelligence
fundus photography
optical coherence tomography
diagnosis
url https://www.mdpi.com/2227-9059/13/2/420
work_keys_str_mv AT xiaochunling deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis
AT henryshenlihchen deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis
AT pohanyeh deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis
AT yuchuncheng deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis
AT chuyenhuang deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis
AT suchinshen deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis
AT yungsunglee deeplearninginglaucomadetectionandprogressionpredictionasystematicreviewandmetaanalysis