Artificial Intelligence for Retinal and Optic-Nerve Imaging Biomarkers: A Biomarker-Centred Systematic Review and Diagnostic Meta-Analysis
Protocol, extraction worksheets, analysis scripts and verification files of a systematic review and diagnostic meta-analysis of 97 studies (92,388 patients; 60,183 eyes) of machine-learning and deep-learning imaging biomarkers in retinal and optic-nerve disease. Databases were searched from inception to 22 July 2026. The review was not entered in a prospective register; the protocol was specified before screening and is deposited here in full, together with the material needed to recompute every reported estimate. This deposit is a time-stamped archive of those materials and is not presented as a prospective registration. Reporting follows PRISMA 2020 and PRISMA-DTA. See README.md for the folder structure and GLOSSARY_EN.md for a bilingual key to the worksheets.
Authors
- Lucas Antônio Avelar Carvalho (ORCID: https://orcid.org/0009-0005-5785-4589)
- Eric Pinheiro de Andrade (ORCID: https://orcid.org/0000-0002-3331-786X)
- Kevin Waquim Pessoa Carvalho (ORCID: https://orcid.org/0000-0002-3923-0503)
- Francisco Murta
- Carolina Oliveira de Ávila (ORCID: https://orcid.org/0000-0001-5864-5221)
- Rubens Belfort Jr.
- Renato Passos
- Flavio Galhardo
Publication Details
- Journal
- Open Science Framework
- Published
- 2026-10-28
- DOI
- https://doi.org/10.17605/osf.io/t4n3z
- Primary Topic
- Retinal Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00