Detection of Diabetic Retinopathy from Ultra-Widefield Scanning Laser Ophthalmoscope Images: A Multicenter Deep Learning Analysis

Published:January 31, 2021DOI:


      To develop a deep learning (DL) system that can detect referable diabetic retinopathy (RDR) and vision-threatening diabetic retinopathy (VTDR) from images obtained on ultra-widefield scanning laser ophthalmoscope (UWF-SLO).


      Observational, cross-sectional study.


      A total of 9392 UWF-SLO images of 1903 eyes from 1022 subjects with diabetes from Hong Kong, the United Kingdom, India, and Argentina.


      All images were labeled according to the presence or absence of RDR and the presence or absence of VTDR. Labeling was performed by retina specialists from fundus examination, according to the International Clinical Diabetic Retinopathy Disease Severity Scale. Three convolutional neural networks (ResNet50) were trained with a transfer-learning procedure for assessing gradability and identifying VTDR and RDR. External validation was performed on 4 datasets spanning different geographical regions.

      Main Outcome Measures

      Area under the receiver operating characteristic curve (AUROC); area under the precision-recall curve (AUPRC); sensitivity, specificity, and accuracy of the DL system in gradability assessment; and detection of RDR and VTDR.


      For gradability assessment, the system achieved an AUROC of 0.923 (95% confidence interval [CI], 0.892–0.947), sensitivity of 86.5% (95% CI, 77.6–92.8), and specificity of 82.1% (95% CI, 77.3–86.2) for the primary validation dataset, and >0.82 AUROCs, >79.6% sensitivity, and >70.4% specificity for the geographical external validation datasets. For detecting RDR and VTDR, the AUROCs were 0.981 (95% CI, 0.977–0.984) and 0.966 (95% CI, 0.961–0.971), with sensitivities of 94.9% (95% CI, 92.3–97.9) and 87.2% (95% CI, 81.5–91.6), specificities of 95.1% (95% CI, 90.6–97.9) and 95.8% (95% CI, 93.3–97.6), and positive predictive values (PPVs) of 98.0% (95% CI, 96.1–99.0) and 91.1% (95% CI, 86.3–94.3) for the primary validation dataset, respectively. The AUROCs and accuracies for detecting both RDR and VTDR were >0.9% and >80%, respectively, for the geographical external validation datasets. The AUPRCs were >0.9, and sensitivities, specificities, and PPVs were >80% for the geographical external validation datasets for RDR and VTDR detection.


      The excellent performance achieved with this DL system for image quality assessment and detection of RDR and VTDR in UWF-SLO images highlights its potential as an efficient and effective diabetic retinopathy screening tool.


      Abbreviations and Acronyms:

      AI (artificial intelligence), AUPRC (area under the precision-recall curve), AUROC (area under the receiver operating characteristic curve), CI (confidence interval), CNN (convolutional neural network), DL (deep learning), DM (diabetes mellitus), DME (diabetic macular edema), DR (diabetic retinopathy), ETDRS (Early Treatment Diabetic Retinopathy Study), NPDR (nonproliferative DR), PDR (proliferative diabetic retinopathy), PPV (positive predictive value), RDR (referable diabetic retinopathy), UWF (ultra-widefield), UWF-SLO (ultra-widefield scanning laser ophthalmoscope), VTDR (vision-threatening diabetic retinopathy)
      To read this article in full you will need to make a payment
      Subscribe to Ophthalmology Retina
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect


        • Wong T.Y.
        • Cheung C.M.
        • Larsen M.
        • et al.
        Diabetic retinopathy.
        Nat Rev Dis Primers. 2016; 2: 16012
        • Garvican L.
        • Clowes J.
        • Gillow T.
        Preservation of sight in diabetes: developing a national risk reduction programme.
        Diabet Med. 2000; 17: 627-634
        • Vujosevic S.
        • Aldington S.J.
        • Silva P.
        • et al.
        Screening for diabetic retinopathy: new perspectives and challenges.
        Lancet Diabetes Endocrinol. 2020; 8: 337-347
        • Wong T.Y.
        • Sun J.
        • Kawasaki R.
        • et al.
        Guidelines on Diabetic Eye Care: The International Council of Ophthalmology Recommendations for Screening, Follow-up, Referral, and Treatment Based on Resource Settings.
        Ophthalmology. 2018; 125: 1608-1622
        • Solomon S.D.
        • Chew E.
        • Duh E.J.
        • et al.
        Diabetic retinopathy: a position statement by the American Diabetes Association.
        Diabetes Care. 2017; 40: 412-418
        • Lawrenson J.G.
        • Bourmpaki E.
        • Bunce C.
        • et al.
        Trends in diabetic retinopathy screening attendance and associations with vision impairment attributable to diabetes in a large nationwide cohort.
        Diabet Med. 2020; : e14425
        • Kernt M.
        • Hadi I.
        • Pinter F.
        • et al.
        Assessment of diabetic retinopathy using nonmydriatic ultra-widefield scanning laser ophthalmoscopy (Optomap) compared with ETDRS 7-field stereo photography.
        Diabetes Care. 2012; 35: 2459-2463
        • Silva P.S.
        • Cavallerano J.D.
        • Sun J.K.
        • et al.
        Nonmydriatic ultrawide field retinal imaging compared with dilated standard 7-field 35-mm photography and retinal specialist examination for evaluation of diabetic retinopathy.
        Am J Ophthalmol. 2012; 154: 549-559.e2
        • Silva P.S.
        • Cavallerano J.D.
        • Tolson A.M.
        • et al.
        Real-time ultrawide field image evaluation of retinopathy in a diabetes telemedicine program.
        Diabetes Care. 2015; 38: 1643-1649
        • Hirano T.
        • Imai A.
        • Kasamatsu H.
        • et al.
        Assessment of diabetic retinopathy using two ultra-wide-field fundus imaging systems, the Clarus(R) and Optos systems.
        BMC Ophthalmol. 2018; 18: 332
        • Silva P.S.
        • Cavallerano J.D.
        • Sun J.K.
        • et al.
        Peripheral lesions identified by mydriatic ultrawide field imaging: distribution and potential impact on diabetic retinopathy severity.
        Ophthalmology. 2013; 120: 2587-2595
        • Sun J.K.
        • Aiello L.P.
        The future of ultrawide field imaging for diabetic retinopathy: pondering the retinal periphery.
        JAMA Ophthalmol. 2016; 134: 247-248
        • Silva P.S.
        • Horton M.B.
        • Clary D.
        • et al.
        Identification of diabetic retinopathy and ungradable image rate with ultrawide field imaging in a national teleophthalmology program.
        Ophthalmology. 2016; 123: 1360-1367
        • Arcadu F.
        • Benmansour F.
        • Maunz A.
        • et al.
        Deep learning algorithm predicts diabetic retinopathy progression in individual patients.
        NPJ Digit Med. 2019; 2: 92
        • Silva P.S.
        • Cavallerano J.D.
        • Haddad N.M.
        • et al.
        Peripheral lesions identified on ultrawide field imaging predict increased risk of diabetic retinopathy progression over 4 years.
        Ophthalmology. 2015; 122: 949-956
        • Ting D.S.W.
        • Cheung C.Y.
        • Lim G.
        • et al.
        Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes.
        JAMA. 2017; 318: 2211-2223
        • Li Z.
        • Keel S.
        • Liu C.
        • et al.
        An automated grading system for detection of vision-threatening referable diabetic retinopathy on the basis of color fundus photographs.
        Diabetes Care. 2018; 41: 2509-2516
        • Gulshan V.
        • Peng L.
        • Coram M.
        • et al.
        Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs.
        JAMA. 2016; 316: 2402-2410
        • Abramoff M.D.
        • Lou Y.
        • Erginay A.
        • et al.
        Improved automated detection of diabetic retinopathy on a publicly available dataset through integration of deep learning.
        Invest Ophthalmol Vis Sci. 2016; 57: 5200-5206
        • Gargeya R.
        • Leng T.
        Automated identification of diabetic retinopathy using deep learning.
        Ophthalmology. 2017; 124: 962-969
        • Cheung C.Y.
        • Tang F.
        • Ting D.S.W.
        • et al.
        Artificial intelligence in diabetic eye disease screening.
        Asia Pac J Ophthalmol (Phila). 2019 Apr 24; ([Epub ahead of print])
        • Bellemo V.
        • Lim G.
        • Rim T.H.
        • et al.
        Artificial intelligence screening for diabetic retinopathy: the real-world emerging application.
        Curr Diab Rep. 2019; 19: 72
        • Xie Y.
        • Nguyen Q.D.
        • Hamzah H.
        • et al.
        Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: an economic analysis modelling study.
        Lancet Digit Health. 2020; 2: e240-e249
        • Li Z.
        • Guo C.
        • Nie D.
        • et al.
        Deep learning from "passive feeding" to "selective eating" of real-world data.
        NPJ Digit Med. 2020; 3: 143
        • Tang F.Y.
        • Ng D.S.
        • Lam A.
        • et al.
        Determinants of quantitative optical coherence tomography angiography metrics in patients with diabetes.
        Sci Rep. 2017; 7: 2575
        • Sivaprasad S.
        • Prevost A.T.
        • Vasconcelos J.C.
        • et al.
        Clinical efficacy of intravitreal aflibercept versus panretinal photocoagulation for best corrected visual acuity in patients with proliferative diabetic retinopathy at 52 weeks (CLARITY): a multicentre, single-blinded, randomised, controlled, phase 2b, non-inferiority trial.
        Lancet. 2017; 389: 2193-2203
        • Wilkinson C.P.
        • Ferris 3rd, F.L.
        • Klein R.E.
        • et al.
        Proposed international clinical diabetic retinopathy and diabetic macular edema disease severity scales.
        Ophthalmology. 2003; 110: 1677-1682
        • Kermany D.S.
        • Goldbaum M.
        • Cai W.
        • et al.
        Identifying medical diagnoses and treatable diseases by image-based deep learning.
        Cell. 2018; 172: 1122-1131.e9
        • Zhou B.
        • Khosla A.
        • Lapedriza A.
        • et al.
        Learning deep features for discriminative localization.
        2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016; : 2921-2929
        • Chalakkal R.J.
        • Abdulla W.H.
        • Thulaseedharan S.S.
        Quality and content analysis of fundus images using deep learning.
        Comput Biol Med. 2019; 108: 317-331
        • Zago G.T.
        • Andreao R.V.
        • Dorizzi B.
        • Teatini Salles E.O.
        Retinal image quality assessment using deep learning.
        Comput Biol Med. 2018; 103: 64-70
        • Abramoff M.D.
        • Lavin P.T.
        • Birch M.
        • et al.
        Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices.
        NPJ Digit Med. 2018; 1: 39
        • Beede E.
        • Baylor E.
        • Hersch F.
        • et al.
        A Human-Centered Evaluation of a Deep Learning System Deployed in Clinics for the Detection of Diabetic Retinopathy.
        Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. Association for Computing Machinery, Honolulu, HI2020
        • Li Z.
        • Guo C.
        • Nie D.
        • et al.
        Deep learning for detecting retinal detachment and discerning macular status using ultra-widefield fundus images.
        Commun Biol. 2020; 3: 15
        • Solanki K.
        • Ramachandra C.
        • Bhat S.
        • et al.
        EyeArt: automated, high-throughput, image analysis for diabetic retinopathy screening.
        Invest Ophthalmol Vis Sci. 2015; 56: 1429
        • Olvera-Barrios A.
        • Heeren T.F.
        • Balaskas K.
        • et al.
        Diagnostic accuracy of diabetic retinopathy grading by an artificial intelligence-enabled algorithm compared with a human standard for wide-field true-colour confocal scanning and standard digital retinal images.
        Br J Ophthalmol. 2021; 105: 265-270
        • Araújo T.
        • Aresta G.
        • Mendonça L.
        • et al.
        DR|GRADUATE: uncertainty-aware deep learning-based diabetic retinopathy grading in eye fundus images.
        Med Image Anal. 2020; 63: 101715
        • Nagasawa T.
        • Tabuchi H.
        • Masumoto H.
        • et al.
        Accuracy of ultrawide-field fundus ophthalmoscopy-assisted deep learning for detecting treatment-naive proliferative diabetic retinopathy.
        Int Ophthalmol. 2019; 39: 2153-2159
        • Seider M.I.
        • Lee R.Y.
        • Wang D.
        • et al.
        Optic disk size variability between African, Asian, white, Hispanic, and Filipino Americans using Heidelberg retinal tomography.
        J Glaucoma. 2009; 18: 595-600
        • Ctori I.
        • Huntjens B.
        The association between foveal morphology and macular pigment spatial distribution: an ethnicity study.
        PLoS One. 2017; 12e0169520
        • Wong T.Y.
        • Islam F.M.
        • Klein R.
        • et al.
        Retinal vascular caliber, cardiovascular risk factors, and inflammation: the multi-ethnic study of atherosclerosis (MESA).
        Invest Ophthalmol Vis Sci. 2006; 47: 2341-2350
        • Burnett S.
        • Hurwitz B.
        • Davey C.
        • et al.
        The implementation of prompted retinal screening for diabetic eye disease by accredited optometrists in an inner-city district of North London: a quality of care study.
        Diabet Med. 1998; 15: S38-S43