You are here

Article Text

Article menu
Download PDFPDF
Retina
Deep learning detection of diabetic retinopathy in Scotland’s diabetic eye screening programme
  1. Alan D Fleming1,
  2. Joseph Mellor2,
  3. Stuart J McGurnaghan1,
  4. Luke A K Blackbourn1,
  5. Keith A Goatman3,
  6. Caroline Styles4,
  7. Amos J Storkey5,
  8. Paul M McKeigue2,
  9. Helen M Colhoun1
  1. 1 The Institute of Genetics and Cancer, University of Edinburgh Western General Hospital, Edinburgh, UK
  2. 2 Usher Institute, The University of Edinburgh, Edinburgh, UK
  3. 3 King’s College, Aberdeen, UK
  4. 4 Queen Margaret Hospital, NHS Fife, Dunfermline, Fife, UK
  5. 5 School of Informatics, The University of Edinburgh, Edinburgh, UK
  1. Correspondence to Dr Joseph Mellor, The University of Edinburgh, Edinburgh, Edinburgh, UK; joe.mellor{at}ed.ac.uk

Abstract

Background/Aims Support vector machine-based automated grading (known as iGradingM) has been shown to be safe, cost-effective and robust in the diabetic retinopathy (DR) screening (DES) programme in Scotland. It triages screening episodes as gradable with no DR versus manual grading required. The study aim was to develop a deep learning-based autograder using images and gradings from DES and to compare its performance with that of iGradingM.

Methods Retinal images, quality assurance (QA) data and routine DR grades were obtained from national datasets in 179 944 patients for years 2006–2016. QA grades were available for 744 images. We developed a deep learning-based algorithm to detect whether either eye contained ungradable images or any DR. The sensitivity and specificity were evaluated against consensus QA grades and routine grades.

Results Images used in QA which were ungradable or with DR were detected by deep learning with better specificity compared with manual graders (p<0.001) and with iGradingM (p<0.001) at the same sensitivities. Any DR according to the DES final grade was detected with 89.19% (270 392/303 154) sensitivity and 77.41% (500 945/647 158) specificity. Observable disease and referable disease were detected with sensitivities of 96.58% (16 613/17 201) and 98.48% (22 600/22 948), respectively. Overall, 43.84% of screening episodes would require manual grading.

Conclusion A deep learning-based system for DR grading was evaluated in QA data and images from 11 years in 50% of people attending a national DR screening programme. The system could reduce the manual grading workload at the same sensitivity compared with the current automated grading system.

  • Retina
  • Public health
  • Imaging
  • Telemedicine

Data availability statement

Data may be obtained from a third party and are not publicly available.

https://doi.org/10.1136/bjo-2023-323395

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Data availability statement

    Data may be obtained from a third party and are not publicly available.

    View Full Text

    Footnotes

    • ADF and JM contributed equally.

    • Contributors ADF and JM contributed equally to this paper. ADF and JM conceived and designed the study. HMC, PMM and AJS made important contributions to study design. LAKB, KAG and SJM were involved in data cleaning, harmonisation, quality control and databasing of the data. JM, ADF and KAG coded and performed the data analysis methods. ADF drafted the initial manuscript. All authors made critically important contributions to manuscript revision. All authors approved the final manuscript. HMC is guarantor of the overall content.

    • Funding This work was funded by Juvenile Diabetes Research Foundation UK (JDRF UK; grant number 2-SRA-2019-857-S-B).

    • Competing interests HMC is the principal investigator on the JDRF UK grant. The employment of ADF and JM was with this funding. HMC and PMM have declared stock options in Bayer AG and Roche Pharmaceuticals. HMC has received grants from AstraZeneca LP, Regeneron, Pfizer, Novo Nordisk and Eli Lilly and Company, and is on the advisory panels or boards of Novo Nordisk, Eli Lilly and Company, Regeneron, Novartis Pharmaceuticals, Bayer AG and Sanofi-Aventis. HMC has received payments for speakers bureaus and honoraria from Eli Lilly and Company, Regeneron and Novartis Pharmaceuticals.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

    Linked Articles

    • Highlights from this issue
      At a glance
      Frank Larkin
      British Journal of Ophthalmology 2024; 108 i-ii Published Online First: 20 Jun 2024. doi: 10.1136/bjo-2024-325908