A deep-learning system predicts glaucoma incidence and progression using retinal photographs

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Background

Deep learning has been widely used for glaucoma diagnosis. However, there is no clinically validated algorithm for glaucoma incidence and progression prediction. This study aims to develop a clinically feasible deep-learning system for predicting and stratifying the risk of glaucoma onset and progression based on color fundus photographs (CFPs), with clinical validation of performance in external population cohorts.

Methods

We established data sets of CFPs and visual fields collected from longitudinal cohorts. The mean follow-up duration was 3 to 5 years across the data sets. Artificial intelligence (AI) models were developed to predict future glaucoma incidence and progression based on the CFPs of 17,497 eyes in 9346 patients. The area under the receiver operating characteristic (AUROC) curve, sensitivity, and specificity of the AI models were calculated with reference to the labels provided by experienced ophthalmologists. Incidence and progression of glaucoma were determined based on longitudinal CFP images or visual fields, respectively.

Results

The AI model to predict glaucoma incidence achieved an AUROC of 0.90 (0.81–0.99) in the validation set and demonstrated good generalizability, with AUROCs of 0.89 (0.83–0.95) and 0.88 (0.79–0.97) in external test sets 1 and 2, respectively. The AI model to predict glaucoma progression achieved an AUROC of 0.91 (0.88–0.94) in the validation set, and also demonstrated outstanding predictive performance with AUROCs of 0.87 (0.81–0.92) and 0.88 (0.83–0.94) in external test sets 1 and 2, respectively.

Conclusion

Our study demonstrates the feasibility of deep-learning algorithms in the early detection and prediction of glaucoma progression.

FUNDING

National Natural Science Foundation of China (NSFC); the High-level Hospital Construction Project, Zhongshan Ophthalmic Center, Sun Yat-sen University; the Science and Technology Program of Guangzhou, China (2021), the Science and Technology Development Fund (FDCT) of Macau, and FDCT-NSFC.

Abstract

Related collections

Most cited references 36

Record: found
Abstract: found
Article: not found

Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis.

Yih-Chung Tham, Xiang Li, Tien Wong … (2014)

Glaucoma is the leading cause of global irreversible blindness. Present estimates of global glaucoma prevalence are not up-to-date and focused mainly on European ancestry populations. We systematically examined the global prevalence of primary open-angle glaucoma (POAG) and primary angle-closure glaucoma (PACG), and projected the number of affected people in 2020 and 2040. Systematic review and meta-analysis. Data from 50 population-based studies (3770 POAG cases among 140,496 examined individuals and 786 PACG cases among 112 398 examined individuals). We searched PubMed, Medline, and Web of Science for population-based studies of glaucoma prevalence published up to March 25, 2013. Hierarchical Bayesian approach was used to estimate the pooled glaucoma prevalence of the population aged 40-80 years along with 95% credible intervals (CrIs). Projections of glaucoma were estimated based on the United Nations World Population Prospects. Bayesian meta-regression models were performed to assess the association between the prevalence of POAG and the relevant factors. Prevalence and projection numbers of glaucoma cases. The global prevalence of glaucoma for population aged 40-80 years is 3.54% (95% CrI, 2.09-5.82). The prevalence of POAG is highest in Africa (4.20%; 95% CrI, 2.08-7.35), and the prevalence of PACG is highest in Asia (1.09%; 95% CrI, 0.43-2.32). In 2013, the number of people (aged 40-80 years) with glaucoma worldwide was estimated to be 64.3 million, increasing to 76.0 million in 2020 and 111.8 million in 2040. In the Bayesian meta-regression model, men were more likely to have POAG than women (odds ratio [OR], 1.36; 95% CrI, 1.23-1.52), and after adjusting for age, gender, habitation type, response rate, and year of study, people of African ancestry were more likely to have POAG than people of European ancestry (OR, 2.80; 95% CrI, 1.83-4.06), and people living in urban areas were more likely to have POAG than those in rural areas (OR, 1.58; 95% CrI, 1.19-2.04). The number of people with glaucoma worldwide will increase to 111.8 million in 2040, disproportionally affecting people residing in Asia and Africa. These estimates are important in guiding the designs of glaucoma screening, treatment, and related public health strategies. Copyright © 2014 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.

0 comments Cited 1369 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

The pathophysiology and treatment of glaucoma: a review.

Robert N. Weinreb, Tin Aung, Felipe Medeiros (2014)

Glaucoma is a worldwide leading cause of irreversible vision loss. Because it may be asymptomatic until a relatively late stage, diagnosis is frequently delayed. A general understanding of the disease pathophysiology, diagnosis, and treatment may assist primary care physicians in referring high-risk patients for comprehensive ophthalmologic examination and in more actively participating in the care of patients affected by this condition. To describe current evidence regarding the pathophysiology and treatment of open-angle glaucoma and angle-closure glaucoma. A literature search was conducted using MEDLINE, the Cochrane Library, and manuscript references for studies published in English between January 2000 and September 2013 on the topics open-angle glaucoma and angle-closure glaucoma. From the 4334 abstracts screened, 210 articles were selected that contained information on pathophysiology and treatment with relevance to primary care physicians. The glaucomas are a group of progressive optic neuropathies characterized by degeneration of retinal ganglion cells and resulting changes in the optic nerve head. Loss of ganglion cells is related to the level of intraocular pressure, but other factors may also play a role. Reduction of intraocular pressure is the only proven method to treat the disease. Although treatment is usually initiated with ocular hypotensive drops, laser trabeculoplasty and surgery may also be used to slow disease progression. Primary care physicians can play an important role in the diagnosis of glaucoma by referring patients with positive family history or with suspicious optic nerve head findings for complete ophthalmologic examination. They can improve treatment outcomes by reinforcing the importance of medication adherence and persistence and by recognizing adverse reactions from glaucoma medications and surgeries.

0 comments Cited 786 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

The Ocular Hypertension Treatment Study: baseline factors that predict the onset of primary open-angle glaucoma.

Mae Gordon, Julia A Beiser, James Brandt … (2002)

The Ocular Hypertension Treatment Study (OHTS) has shown that topical ocular hypotensive medication is effective in delaying or preventing the onset of primary open-angle glaucoma (POAG) in individuals with elevated intraocular pressure (ocular hypertension) and no evidence of glaucomatous damage. To describe baseline demographic and clinical factors that predict which participants in the OHTS developed POAG. Baseline demographic and clinical data were collected prior to randomization except for corneal thickness measurements, which were performed during follow-up. Proportional hazards models were used to identify factors that predicted which participants in the OHTS developed POAG. In univariate analyses, baseline factors that predicted the development of POAG included older age, race (African American), sex (male), larger vertical cup-disc ratio, larger horizontal cup-disc ratio, higher intraocular pressure, greater Humphrey visual field pattern standard deviation, heart disease, and thinner central corneal measurement. In multivariate analyses, baseline factors that predicted the development of POAG included older age, larger vertical or horizontal cup-disc ratio, higher intraocular pressure, greater pattern standard deviation, and thinner central corneal measurement. Baseline age, vertical and horizontal cup-disc ratio, pattern standard deviation, and intraocular pressure were good predictors for the onset of POAG in the OHTS. Central corneal thickness was found to be a powerful predictor for the development of POAG.

0 comments Cited 417 times – based on 0 reviews      Review now

Bookmark

All references

Author and article information

Contributors

Fei Li

Yuandong Su:

ORCID: http://orcid.org/0000-0001-5030-4852

Fengbin Lin

Zhihuan Li

Yunhe Song

Sheng Nie

Linjiang Chen

Shiyan Chen:

ORCID: http://orcid.org/0000-0001-8596-0905

Hao Li

Kanmin Xue:

ORCID: http://orcid.org/0000-0003-0065-9131

Huixin Che

Zhengui Chen

Bin Yang

Huiying Zhang

Ming Ge

Weihui Zhong

Chunman Yang

Lina Chen

Fanyin Wang

Yunqin Jia

Wanlin Li

Yuqing Wu:

ORCID: http://orcid.org/0000-0003-2987-2497

Yingjie Li

Yuanxu Gao

Yong Zhou

Kang Zhang

Xiulan Zhang

Journal

Journal ID (nlm-ta): J Clin Invest

Journal ID (iso-abbrev): J Clin Invest

Journal ID (publisher-id): J Clin Invest

Title: The Journal of Clinical Investigation

Publisher: American Society for Clinical Investigation

ISSN (Print): 0021-9738

ISSN (Electronic): 1558-8238

Publication date (Electronic, pub): 1 June 2022

Publication date (Electronic, collection): 1 June 2022

Publication date (Print, pub): 1 June 2022

Publication date PMC-release: 1 June 2022

Volume: 132

Issue: 11

Electronic Location Identifier: e157968

Affiliations

[1 ]State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, China.

[2 ]State Key Laboratory of Biotherapy and Center for Translational Innovations, West China Hospital and Sichuan University, Chengdu, China.

[3 ]PKU-MUST Center for Future Technology, Faculty of Medicine, Macao University of Science and Technology, Macau, China.

[4 ]State Key Laboratory of Organ Failure Research, National Clinical Research Center for Kidney Disease and Nanfang Hospital, Southern Medical University, Guangzhou, China.

[5 ]Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Beijing Ophthalmology and Visual Science Key Lab, Beijing, China.

[6 ]Department of Ophthalmology, Nanfang Hospital, Southern Medical University, Guangzhou, China.

[7 ]Department of Ophthalmology, Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital, Chengdu, China.

[8 ]Department of Ophthalmology, Guizhou Provincial People’s Hospital, Guiyang, China.

[9 ]Nuffield Laboratory of Ophthalmology, Department of Clinical Neurosciences, University of Oxford and Oxford University Hospitals NHS Foundation Trust, Oxford, United Kingdom.

[10 ]He Eye Specialist Hospital, Shenyang, Liaoning Province, China.

[11 ]Jiangmen Xinhui Aier New Hope Eye Hospital, Jiangmen, Guangdong, China.

[12 ]Department of Ophthalmology, Zigong Third People’s Hospital, Zigong, China.

[13 ]Department of Ophthalmology, Fujian Provincial Hospital, Fuzhou, China.

[14 ]Department of Ophthalmology and Optometry, Guizhou Nursing Vocational College, Guiyang, China.

[15 ]Department of Ophthalmology, Guangzhou Development District Hospital, Guangzhou, China.

[16 ]Department of Ophthalmology, The Second Affiliated Hospital of Guizhou Medical University, Kaili, China.

[17 ]Department of Ophthalmology, The Third People’s Hospital of Dalian, Dalian, Liaoning Province, China.

[18 ]Department of Ophthalmology, Shenzhen Qianhai Shekou Free Trade Zone Hospital, Shenzhen, China.

[19 ]Department of Ophthalmology, Dali Bai Autonomous Prefecture People’s Hospital, Dali, China.

[20 ]Department of Ophthalmology, Wuwei People’s Hospital, Wuwei, Gansu Province, China.

[21 ]Department of Ophthalmology, Joint Shantou International Eye Center of Shantou University and the Chinese University of Hong Kong, Shantou, Guangdong, China.

[22 ]Department of Ophthalmology, The First Hospital of Nanchang City, Nanchang, China.

[23 ]State Key Laboratory of Lunar and Planetary Sciences, Macao University of Science and Technology, Taipa, Macau, China.

[24 ]Clinical Research Institute, Shanghai General Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China.

Author notes

Address correspondence to: Kang Zhang, Macau University of Science and Technology, Avenida Wai Long, Taipa, Macau 999078. Email: kang.zhang@ 123456gmail.com . Or to: Xiulan Zhang, Zhongshan Ophthalmic Center, No. 7 Jinsui Road, Guangzhou 510060, China. Email: zhangxl2@ 123456mail.sysu.edu.cn .

Authorship note: F Li, Y Su, F Lin, and ZL contributed equally to this work.

Author information

Yuandong Su http://orcid.org/0000-0001-5030-4852

Shiyan Chen http://orcid.org/0000-0001-8596-0905

Kanmin Xue http://orcid.org/0000-0003-0065-9131

Yuqing Wu http://orcid.org/0000-0003-2987-2497

Article

Publisher ID: 157968

DOI: 10.1172/JCI157968

PMC ID: 9151694

PubMed ID: 35642636

SO-VID: f2328850-bb8b-4674-98cb-8d730f82e8c0

License:

This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

History

Date received : 29 December 2021

Date accepted : 12 April 2022

Funding

Funded by: High-level Hospital Construction Project, Zhongshan Ophthalmic Center, Sun Yat-sen University

Award ID: 303020104

Funded by: National Natural Science Foundation of China

Award ID: 82101117,82070955

Funded by: FDCT

Award ID: 0070/2020/A2

Funded by: FDCT-NSFC

Award ID: 0007/2020/AFJ

Comments

Comment on this article

scite_

Cited by 17

See all cited by

Most referenced authors 519

See all reference authors

A deep-learning system predicts glaucoma incidence and progression using retinal photographs

Read this article at

Abstract

Background

Methods

Results

Conclusion

FUNDING

Abstract

Related collections

Karger: Ophthalmology

Most cited references 36

Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta-analysis.

The pathophysiology and treatment of glaucoma: a review.

The Ocular Hypertension Treatment Study: baseline factors that predict the onset of primary open-angle glaucoma.

Author and article information

Contributors

Journal

Affiliations

Author notes

Author information

Article

History

Funding

Categories

Comments

Comment on this article

Similar content 737

Cited by 17

Most referenced authors 519