Cargando…

Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis

Retrospective studies on artificial intelligence (AI) in screening for diabetic retinopathy (DR) have shown promising results in addressing the mismatch between the capacity to implement DR screening and increasing DR incidence. This review sought to evaluate the diagnostic test accuracy (DTA) of AI...

Descripción completa

Detalles Bibliográficos
Autores principales: Uy, Holijah, Fielding, Christopher, Hohlfeld, Ameer, Ochodo, Eleanor, Opare, Abraham, Mukonda, Elton, Minnies, Deon, Engel, Mark E.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10511145/
https://www.ncbi.nlm.nih.gov/pubmed/37729122
http://dx.doi.org/10.1371/journal.pgph.0002160
_version_ 1785108084987789312
author Uy, Holijah
Fielding, Christopher
Hohlfeld, Ameer
Ochodo, Eleanor
Opare, Abraham
Mukonda, Elton
Minnies, Deon
Engel, Mark E.
author_facet Uy, Holijah
Fielding, Christopher
Hohlfeld, Ameer
Ochodo, Eleanor
Opare, Abraham
Mukonda, Elton
Minnies, Deon
Engel, Mark E.
author_sort Uy, Holijah
collection PubMed
description Retrospective studies on artificial intelligence (AI) in screening for diabetic retinopathy (DR) have shown promising results in addressing the mismatch between the capacity to implement DR screening and increasing DR incidence. This review sought to evaluate the diagnostic test accuracy (DTA) of AI in screening for referable diabetic retinopathy (RDR) in real-world settings. We searched CENTRAL, PubMed, CINAHL, Scopus, and Web of Science on 9 February 2023. We included prospective DTA studies assessing AI against trained human graders (HGs) in screening for RDR in patients with diabetes. Two reviewers independently extracted data and assessed methodological quality against QUADAS-2 criteria. We used the hierarchical summary receiver operating characteristics (HSROC) model to pool estimates of sensitivity and specificity and, forest plots and SROC plots to visually examine heterogeneity in accuracy estimates. From our initial search results of 3899 studies, we included 15 studies comprising 17 datasets. Meta-analyses revealed a sensitivity of 95.33% (95%CI: 90.60–100%) and specificity of 92.01% (95%CI: 87.61–96.42%) for patient-level analysis (10 datasets, N = 45,785) while, for the eye-level analysis, sensitivity was 91.24% (95%CI: 79.15–100%) and specificity, 93.90% (95%CI: 90.63–97.16%) (7 datasets, N = 15,390). Subgroup analyses did not provide variations in the diagnostic accuracy of country classification and DR classification criteria. However, a moderate increase was observed in diagnostic accuracy in the primary-level healthcare settings: sensitivity of 99.35% (95%CI: 96.85–100%), specificity of 93.72% (95%CI: 88.83–98.61%) and, a minimal decrease in the tertiary-level healthcare settings: sensitivity of 94.71% (95%CI: 89.00–100%), specificity of 90.88% (95%CI: 83.22–98.53%). Sensitivity analyses did not show any variations in studies that included diabetic macular edema in the RDR definition, nor studies with ≥3 HGs. This review provides evidence, for the first time from prospective studies, for the effectiveness of AI in screening for RDR in real-world settings. The results may serve to strengthen existing guidelines to improve current practices.
format Online
Article
Text
id pubmed-10511145
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-105111452023-09-21 Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis Uy, Holijah Fielding, Christopher Hohlfeld, Ameer Ochodo, Eleanor Opare, Abraham Mukonda, Elton Minnies, Deon Engel, Mark E. PLOS Glob Public Health Research Article Retrospective studies on artificial intelligence (AI) in screening for diabetic retinopathy (DR) have shown promising results in addressing the mismatch between the capacity to implement DR screening and increasing DR incidence. This review sought to evaluate the diagnostic test accuracy (DTA) of AI in screening for referable diabetic retinopathy (RDR) in real-world settings. We searched CENTRAL, PubMed, CINAHL, Scopus, and Web of Science on 9 February 2023. We included prospective DTA studies assessing AI against trained human graders (HGs) in screening for RDR in patients with diabetes. Two reviewers independently extracted data and assessed methodological quality against QUADAS-2 criteria. We used the hierarchical summary receiver operating characteristics (HSROC) model to pool estimates of sensitivity and specificity and, forest plots and SROC plots to visually examine heterogeneity in accuracy estimates. From our initial search results of 3899 studies, we included 15 studies comprising 17 datasets. Meta-analyses revealed a sensitivity of 95.33% (95%CI: 90.60–100%) and specificity of 92.01% (95%CI: 87.61–96.42%) for patient-level analysis (10 datasets, N = 45,785) while, for the eye-level analysis, sensitivity was 91.24% (95%CI: 79.15–100%) and specificity, 93.90% (95%CI: 90.63–97.16%) (7 datasets, N = 15,390). Subgroup analyses did not provide variations in the diagnostic accuracy of country classification and DR classification criteria. However, a moderate increase was observed in diagnostic accuracy in the primary-level healthcare settings: sensitivity of 99.35% (95%CI: 96.85–100%), specificity of 93.72% (95%CI: 88.83–98.61%) and, a minimal decrease in the tertiary-level healthcare settings: sensitivity of 94.71% (95%CI: 89.00–100%), specificity of 90.88% (95%CI: 83.22–98.53%). Sensitivity analyses did not show any variations in studies that included diabetic macular edema in the RDR definition, nor studies with ≥3 HGs. This review provides evidence, for the first time from prospective studies, for the effectiveness of AI in screening for RDR in real-world settings. The results may serve to strengthen existing guidelines to improve current practices. Public Library of Science 2023-09-20 /pmc/articles/PMC10511145/ /pubmed/37729122 http://dx.doi.org/10.1371/journal.pgph.0002160 Text en © 2023 Uy et al https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Research Article
Uy, Holijah
Fielding, Christopher
Hohlfeld, Ameer
Ochodo, Eleanor
Opare, Abraham
Mukonda, Elton
Minnies, Deon
Engel, Mark E.
Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis
title Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis
title_full Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis
title_fullStr Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis
title_full_unstemmed Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis
title_short Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis
title_sort diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: a systematic review and meta-analysis
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10511145/
https://www.ncbi.nlm.nih.gov/pubmed/37729122
http://dx.doi.org/10.1371/journal.pgph.0002160
work_keys_str_mv AT uyholijah diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT fieldingchristopher diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT hohlfeldameer diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT ochodoeleanor diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT opareabraham diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT mukondaelton diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT minniesdeon diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis
AT engelmarke diagnostictestaccuracyofartificialintelligenceinscreeningforreferablediabeticretinopathyinrealworldsettingsasystematicreviewandmetaanalysis