EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Validation of a deep learning system for the detection of diabetic retinopathy in Indigenous Australians

Mark A Chia iD, Fred Hersch, Rory Sayres, Pinal Bavishi, Richa Tiwari, Pearse A Keane iD, Angus W Turner iD

British Journal of Ophthalmology · 2024

DOI: 10.1136/bjo-2022-322237 ↗

Kurz gefasst

Fachartikel in der Zeitschrift British Journal of Ophthalmology, 2024. Verfasst von Mark A Chia und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Ophthalmology, University College London, London, UK mark.a.chia@outlook.com. Hauptthemen laut MEDLINE: Australasian People, Deep Learning, Diabetes Mellitus, Diabetic Retinopathy. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Medical Research Council, UK Research and Innovation, Moorfields Eye Charity.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/aims Deep learning systems (DLSs) for diabetic retinopathy (DR) detection show promising results but can underperform in racial and ethnic minority groups, therefore external validation within these populations is critical for health equity. This study evaluates the performance of a DLS for DR detection among Indigenous Australians, an understudied ethnic group who suffer disproportionately from DR-related blindness.

Methods

We performed a retrospective external validation study comparing the performance of a DLS against a retinal specialist for the detection of more-than-mild DR (mtmDR), vision-threatening DR (vtDR) and all-cause referable DR. The validation set consisted of 1682 consecutive, single-field, macula-centred retinal photographs from 864 patients with diabetes (mean age 54.9 years, 52.4% women) at an Indigenous primary care service in Perth, Australia. Three-person adjudication by a panel of specialists served as the reference standard.

Results

For mtmDR detection, sensitivity of the DLS was superior to the retina specialist (98.0% (95% CI, 96.5 to 99.4) vs 87.1% (95% CI, 83.6 to 90.6), McNemar’s test p

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Australasian PeopleDeep LearningDiabetes MellitusDiabetic RetinopathyFemaleHumansMaleMiddle AgedAustraliaEthnicityMinority GroupsRetrospective StudiesAustralian Aboriginal and Torres Strait Islander Peoples

Stichwörter der Autor:innen

Diagnostic tests/InvestigationImagingRetina

Autor:innen und Institutionen

  • Institute of Ophthalmology, University College London, London, UK mark.a.chia@outlook.com.
  • Moorfields Eye Hospital NHS Foundation Trust, London, UK.
  • Lions Outback Vision, Lions Eye Institute, Nedlands, Western Australia, Australia.
  • Google Health, Palo Alto, California, USA.
  • Institute of Ophthalmology, University College London, London, UK.
  • Centre for Ophthalmology and Visual Science, The University of Western Australia, Nedlands, Western Australia, Australia.

ORCID-Profile: Mark A Chia, Pearse A Keane, Angus W Turner

Förderung

  • Medical Research Council
  • UK Research and Innovation
  • Moorfields Eye Charity
  • General Sir John Monash Foundation
  • Google LLC

Zitieren

APA

Chia, M. A., Hersch, F., Sayres, R., Bavishi, P., Tiwari, R., Keane, P. A., & Turner, A. W. (2024). Validation of a deep learning system for the detection of diabetic retinopathy in Indigenous Australians. British Journal of Ophthalmology. https://doi.org/10.1136/bjo-2022-322237

Vancouver

Chia MA, Hersch F, Sayres R, Bavishi P, Tiwari R, Keane PA, et al. Validation of a deep learning system for the detection of diabetic retinopathy in Indigenous Australians. British Journal of Ophthalmology. 2024. doi:10.1136/bjo-2022-322237

BibTeX
@article{Chia2024,
  author = {Chia, Mark A and Hersch, Fred and Sayres, Rory and Bavishi, Pinal and Tiwari, Richa and Keane, Pearse A and Turner, Angus W},
  title = {Validation of a deep learning system for the detection of diabetic retinopathy in Indigenous Australians},
  journal = {British Journal of Ophthalmology},
  year = {2024},
  doi = {10.1136/bjo-2022-322237}
}
Kostenlos starten →