Predicting 1, 2 and 3 year emergent referable diabetic retinopathy and maculopathy using deep learning.


Journal

Communications medicine
ISSN: 2730-664X
Titre abrégé: Commun Med (Lond)
Pays: England
ID NLM: 9918250414506676

Informations de publication

Date de publication:
21 Aug 2024
Historique:
received: 22 02 2024
accepted: 07 08 2024
medline: 22 8 2024
pubmed: 22 8 2024
entrez: 21 8 2024
Statut: epublish

Résumé

Predicting diabetic retinopathy (DR) progression could enable individualised screening with prompt referral for high-risk individuals for sight-saving treatment, whilst reducing screening burden for low-risk individuals. We developed and validated deep learning systems (DLS) that predict 1, 2 and 3 year emergent referable DR and maculopathy using risk factor characteristics (tabular DLS), colour fundal photographs (image DLS) or both (multimodal DLS). From 162,339 development-set eyes from south-east London (UK) diabetic eye screening programme (DESP), 110,837 had eligible longitudinal data, with the remaining 51,502 used for pretraining. Internal and external (Birmingham DESP, UK) test datasets included 27,996, and 6928 eyes respectively. Internal multimodal DLS emergent referable DR, maculopathy or either area-under-the receiver operating characteristic (AUROC) were 0.95 (95% CI: 0.92-0.98), 0.84 (0.82-0.86), 0.85 (0.83-0.87) for 1 year, 0.92 (0.87-0.96), 0.84 (0.82-0.87), 0.85 (0.82-0.87) for 2 years, and 0.85 (0.80-0.90), 0.79 (0.76-0.82), 0.79 (0.76-0.82) for 3 years. External multimodal DLS emergent referable DR, maculopathy or either AUROC were 0.93 (0.88-0.97), 0.85 (0.80-0.89), 0.85 (0.76-0.85) for 1 year, 0.93 (0.89-0.97), 0.79 (0.74-0.84), 0.80 (0.76-0.85) for 2 years, and 0.91 (0.84-0.98), 0.79 (0.74-0.83), 0.79 (0.74-0.84) for 3 years. Multimodal and image DLS performance is significantly better than tabular DLS at all intervals. DLS accurately predict 1, 2 and 3 year emergent referable DR and referable maculopathy using colour fundal photographs, with additional risk factor characteristics conferring improvements in prognostic performance. Proposed DLS are a step towards individualised risk-based screening, whereby AI-assistance allows high-risk individuals to be closely monitored while reducing screening burden for low-risk individuals. Diabetic retinopathy (DR) is a disease where the light-sensing layer at the back of the eye (retina) becomes damaged by raised blood sugar levels. It affects around one in three of the 463 million people with diabetes worldwide and is a leading cause of acquired vision loss in working-age adults. In this study, we developed computer-based models to predict when DR would reach a stage where vision could be threatened up to 3-years in the future. Our study shows that this system can accurately predict sight-threatening DR in patients with diabetes. This could mean fewer unnecessary visits for individuals at low-risk of DR progression, but closer monitoring and potentially earlier treatment for individuals at high-risk of DR progression, which could reduce the risk of vision loss.

Sections du résumé

BACKGROUND BACKGROUND
Predicting diabetic retinopathy (DR) progression could enable individualised screening with prompt referral for high-risk individuals for sight-saving treatment, whilst reducing screening burden for low-risk individuals. We developed and validated deep learning systems (DLS) that predict 1, 2 and 3 year emergent referable DR and maculopathy using risk factor characteristics (tabular DLS), colour fundal photographs (image DLS) or both (multimodal DLS).
METHODS METHODS
From 162,339 development-set eyes from south-east London (UK) diabetic eye screening programme (DESP), 110,837 had eligible longitudinal data, with the remaining 51,502 used for pretraining. Internal and external (Birmingham DESP, UK) test datasets included 27,996, and 6928 eyes respectively.
RESULTS RESULTS
Internal multimodal DLS emergent referable DR, maculopathy or either area-under-the receiver operating characteristic (AUROC) were 0.95 (95% CI: 0.92-0.98), 0.84 (0.82-0.86), 0.85 (0.83-0.87) for 1 year, 0.92 (0.87-0.96), 0.84 (0.82-0.87), 0.85 (0.82-0.87) for 2 years, and 0.85 (0.80-0.90), 0.79 (0.76-0.82), 0.79 (0.76-0.82) for 3 years. External multimodal DLS emergent referable DR, maculopathy or either AUROC were 0.93 (0.88-0.97), 0.85 (0.80-0.89), 0.85 (0.76-0.85) for 1 year, 0.93 (0.89-0.97), 0.79 (0.74-0.84), 0.80 (0.76-0.85) for 2 years, and 0.91 (0.84-0.98), 0.79 (0.74-0.83), 0.79 (0.74-0.84) for 3 years.
CONCLUSIONS CONCLUSIONS
Multimodal and image DLS performance is significantly better than tabular DLS at all intervals. DLS accurately predict 1, 2 and 3 year emergent referable DR and referable maculopathy using colour fundal photographs, with additional risk factor characteristics conferring improvements in prognostic performance. Proposed DLS are a step towards individualised risk-based screening, whereby AI-assistance allows high-risk individuals to be closely monitored while reducing screening burden for low-risk individuals.
Diabetic retinopathy (DR) is a disease where the light-sensing layer at the back of the eye (retina) becomes damaged by raised blood sugar levels. It affects around one in three of the 463 million people with diabetes worldwide and is a leading cause of acquired vision loss in working-age adults. In this study, we developed computer-based models to predict when DR would reach a stage where vision could be threatened up to 3-years in the future. Our study shows that this system can accurately predict sight-threatening DR in patients with diabetes. This could mean fewer unnecessary visits for individuals at low-risk of DR progression, but closer monitoring and potentially earlier treatment for individuals at high-risk of DR progression, which could reduce the risk of vision loss.

Autres résumés

Type: plain-language-summary (eng)
Diabetic retinopathy (DR) is a disease where the light-sensing layer at the back of the eye (retina) becomes damaged by raised blood sugar levels. It affects around one in three of the 463 million people with diabetes worldwide and is a leading cause of acquired vision loss in working-age adults. In this study, we developed computer-based models to predict when DR would reach a stage where vision could be threatened up to 3-years in the future. Our study shows that this system can accurately predict sight-threatening DR in patients with diabetes. This could mean fewer unnecessary visits for individuals at low-risk of DR progression, but closer monitoring and potentially earlier treatment for individuals at high-risk of DR progression, which could reduce the risk of vision loss.

Identifiants

pubmed: 39169209
doi: 10.1038/s43856-024-00590-z
pii: 10.1038/s43856-024-00590-z
doi:

Types de publication

Journal Article

Langues

eng

Pagination

167

Subventions

Organisme : Diabetes UK
ID : 20/0006144
Pays : United Kingdom

Informations de copyright

© 2024. The Author(s).

Références

Saeedi, P. et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the international diabetes federation diabetes Atlas, 9(th) edition. Diabetes Res. Clin. Pract. 157, 107843 (2019).
pubmed: 31518657
Ting, D. S., Cheung, G. C. & Wong, T. Y. Diabetic retinopathy: global prevalence, major risk factors, screening practices and public health challenges: a review. Clin. Exp. Ophthalmol. 44, 260–277 (2015).
Steinmetz, J. D. et al. Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global Burden of Disease Study. Lancet Glob. Health 9, e144–e160 (2021).
Scanlon, P. H. The english national screening programme for diabetic retinopathy 2003-2016. Acta Diabetol. 54, 515–525 (2017).
pubmed: 28224275 pmcid: 5429356
Heydon, P. et al. Prospective evaluation of an artificial intelligence-enabled algorithm for automated diabetic retinopathy screening of 30,000 patients. Br. J. Ophthalmol. 105, 723–728 (2021).
pubmed: 32606081
Thomas, R. L. et al. Retrospective analysis of newly recorded certifications of visual impairment due to diabetic retinopathy in Wales during 2007–2015. BMJ Open 7, e015024 (2017).
pubmed: 28720613 pmcid: 5541630
Scanlon, P. H. The contribution of the English NHS diabetic eye screening programme to reductions in diabetes-related blindness, comparisons within Europe, and future challenges. Acta Diabetol. 58, 521–530 (2021).
pubmed: 33830332 pmcid: 8053650
Broadbent, D. M. et al. Safety and cost-effectiveness of individualised screening for diabetic retinopathy: the ISDR open-label, equivalence RCT. Diabetologia 64, 56–69 (2021).
pubmed: 33146763
Byrne, P. et al. Personalising screening of sight-threatening diabetic retinopathy - qualitative evidence to inform effective implementation. BMC Public Health 20, 881 (2020).
pubmed: 32513143 pmcid: 7278114
Sharif, A., Jendle, J. & Hellgren, K. J. Screening for diabetic retinopathy with extended intervals, safe and without compromising adherence: a retrospective cohort study. Diabetes Ther. 12, 223–234 (2021).
pubmed: 33165837
Taylor-Phillips, S. et al. Extending the diabetic retinopathy screening interval beyond 1 year: systematic review. Br. J. Ophthalmol. 100, 105–114 (2016).
pubmed: 25586713
Eleuteri, A. et al. Individualised variable-interval risk-based screening for sight-threatening diabetic retinopathy: the liverpool risk calculation engine. Diabetologia 60, 2174–2182 (2017).
pubmed: 28840258 pmcid: 6448900
Haider, S., Sadiq, S. N., Moore, D., Price, M. J. & Nirantharakumar, K. Prognostic prediction models for diabetic retinopathy progression: a systematic review. Eye 33, 702–713 (2019).
pubmed: 30651592 pmcid: 6707154
García-Fiñana, M. et al. Personalized risk-based screening for diabetic retinopathy: a multivariate approach versus the use of stratification rules. Diabetes Obes. Metab. 21, 560–568 (2019).
pubmed: 30284381
Stratton, I. M., Aldington, S. J., Taylor, D. J., Adler, A. I. & Scanlon, P. H. A simple risk stratification for time to development of sight-threatening diabetic retinopathy. Diabetes Care 36, 580–585 (2013).
pubmed: 23150285 pmcid: 3579348
Leese, G. P. et al. Progression of diabetes retinal status within community screening programs and potential implications for screening intervals. Diabetes Care 38, 488–494 (2015).
pubmed: 25524948
Lund, S. H. et al. Individualised risk assessment for diabetic retinopathy and optimisation of screening intervals: a scientific approach to reducing healthcare costs. Br. J. Ophthalmol. 100, 683–687 (2016).
pubmed: 26377413
Aspelund, T. et al. Individual risk assessment and information technology to optimise screening frequency for diabetic retinopathy. Diabetologia 54, 2525–2532 (2011).
pubmed: 21792613
Kashim, R. M., Newton, P. & Ojo, O. Diabetic retinopathy screening: a systematic review on patients’ non-attendance. Int. J. Environ. Res. Public Health 15, 157 (2018).
pubmed: 29351207 pmcid: 5800256
Olvera-Barrios, A. et al. Two-year recall for people with no diabetic retinopathy: a multi-ethnic population-based retrospective cohort study using real-world data to quantify the effect. Br. J. Ophthalmol. 107, 1839–1845 (2023).
pubmed: 37875374
Arcadu, F. et al. Author correction: deep learning algorithm predicts diabetic retinopathy progression in individual patients. npj Digit. Med. 3, 160 (2020).
pubmed: 33293570 pmcid: 7723990
Bora, A. et al. Predicting the risk of developing diabetic retinopathy using deep learning. Lancet Digit. Health 3, e10–e19 (2021).
pubmed: 33735063
Rom, Y., Aviv, R., Ianchulev, T. & Dvey-Aharon, Z. Predicting the future development of diabetic retinopathy using a deep learning algorithm for the analysis of non-invasive retinal imaging. BMJ Open Ophthalmol. 7, e001140 (2022).
pmcid: 9809299
Dai, L. et al. A deep learning system for predicting time to progression of diabetic retinopathy. Nat. Med. 30, 584–594 (2024).
pubmed: 38177850 pmcid: 10878973
PHE. NHS Diabetic Eye Screening Programme: Grading definitions for referable disease, https://www.gov.uk/government/publications/diabetic-eye-screening-retinal-image-grading-criteria/nhs-diabetic-eye-screening-programme-grading-definitions-for-referable-disease (2021).
McLennan, D. et al. The English Indices of Deprivation 2019. (Ministry of Housing, Communities and Local Government, 2019).
Nderitu, P. et al. Automated image curation in diabetic retinopathy screening using deep learning. Sci. Rep. 12, 11196 (2022).
pubmed: 35778615 pmcid: 9249740
Tan, M. & Le, Q. V. EfficientNetV2: Smaller models and faster training. arXiv:2104.00298. https://ui.adsabs.harvard.edu/abs/2021arXiv210400298T (2021).
Arik, S. O. & Pfister, T. TabNet: attentive interpretable tabular learning. arXiv:1908.07442. https://ui.adsabs.harvard.edu/abs/2019arXiv190807442A (2019).
Haider, S. et al. Predictors for diabetic retinopathy progression-findings from nominal group technique and Evidence review. BMJ Open Ophthalmol. 5, e000579 (2020).
pubmed: 33083555 pmcid: 7549478
Photocoagulation Treatment of Proliferative Diabetic Retinopathy. Clinical application of Diabetic Retinopathy Study (DRS) findings, DRS Report Number 8. Ophthalmology 88, 583–600 (1981).
Klein, R. et al. The relation of retinal vessel caliber to the incidence and progression of diabetic retinopathy: XIX: the wisconsin epidemiologic study of diabetic retinopathy. Arch Ophthalmol. 122, 76–83 (2004).
pubmed: 14718299
Zhang, K. et al. Deep-learning models for the detection and incidence prediction of chronic kidney disease and type 2 diabetes from retinal fundus images. Nat. Biomed. Eng. 5, 533–545 (2021).
pubmed: 34131321
Saputro, S. A., Pattanaprateep, O., Pattanateepapon, A., Karmacharya, S. & Thakkinstian, A. Prognostic models of diabetic microvascular complications: a systematic review and meta-analysis. System. Rev. 10, 288 (2021).
Oke, J. L., Stratton, I. M., Aldington, S. J., Stevens, R. J. & Scanlon, P. H. The use of statistical methodology to determine the accuracy of grading within a diabetic retinopathy screening programme. Diab. Med. 33, 896–903 (2016).
Zhou, Y. et al. A foundation model for generalizable disease detection from retinal images. Nature 622, 156–163 (2023).
pubmed: 37704728 pmcid: 10550819
Nderitu, P. Github Code Respository: Automated Image Curation in Diabetic Retinopathy Screening using Deep Learning, https://github.com/pnderitu/DUK_Automated_Curation (2022).

Auteurs

Paul Nderitu (P)

Section of Ophthalmology, Faculty of Life Sciences and Medicine, King's College London, London, UK. p.nderitu@doctors.org.uk.
Department of Ophthalmology, King's Ophthalmology Research Unit (KORU), King's College Hospital, London, UK. p.nderitu@doctors.org.uk.

Joan M Nunez do Rio (JM)

Department of Ophthalmology, King's Ophthalmology Research Unit (KORU), King's College Hospital, London, UK.
School of Biomedical Engineering & Imaging Sciences, King's College London, London, UK.

Laura Webster (L)

Department of Ophthalmology, South East London Diabetic Eye Screening Service, St Thomas' Hospital, London, UK.

Samantha Mann (S)

Department of Ophthalmology, South East London Diabetic Eye Screening Service, St Thomas' Hospital, London, UK.

M Jorge Cardoso (MJ)

School of Biomedical Engineering & Imaging Sciences, King's College London, London, UK.

Marc Modat (M)

School of Biomedical Engineering & Imaging Sciences, King's College London, London, UK.

David Hopkins (D)

Institute of Diabetes, Endocrinology and Obesity, King's Health Partners, London, UK.

Christos Bergeles (C)

School of Biomedical Engineering & Imaging Sciences, King's College London, London, UK.

Timothy L Jackson (TL)

Section of Ophthalmology, Faculty of Life Sciences and Medicine, King's College London, London, UK.
Department of Ophthalmology, King's Ophthalmology Research Unit (KORU), King's College Hospital, London, UK.

Classifications MeSH