Global RETFound: A Protocol for the World’s First Global Medical Foundation Model

@article{paulnderitu2026global,
  title = {{Global RETFound: A Protocol for the World’s First Global Medical Foundation Model}},
  author = {Paul Nderitu, FRCOphth, PhD and Yukun Zhou, PhD and An Ran Ran, MMed, PhD and Jocelyn Hui Lin Goh, BEng and Hongyang Jiang, PhD and Sahana Srinivasan, BEng and Gatera Fiston Kitema, MPHEC, PhD and Polly Rawlinson, BA and Lucy Arnold, MSc and Ka Wai Yuen, MSc and Ke Zou, PhD and Gabriel Dawei Yang, MMed, PhD and Carol Y. Cheung, PhD and Pearse A. Keane, MD, PhD and Yih Chung Tham, PhD and Hatem M. Abdelakher and Eberechukwu Achigbu and Feyi Grace Adepoju and Leonard Agbahounga and Rupesh Agrawal and Referano Agustiawan and Islam Sh. Ahmed and Kwadwo Owusu Akuffo and Shehab Ahmad Al-Abed and Esraa Khalaf Al-Ajaleen and Minhaj Nur Alam and Sura Yousef Alaroud and Abdulrahman F. Albloushi and Neilya Aldasheva and Tsedeke Asaminew Alemu and Nchifor Alice and Faisal Alkathery and Mutaz Al-Nawaflh and Otim Ambrose and Kwesi N. Amissah-Arthur and Damara Andalia and Fares Antaki and Nomo Arlette Francine and Lucy Arnold and Louis Arnould and Simon Arunga and Tariq Aslam and Aamir Asrar and Olufemi Emmanuel Babalola and Vilma Jurate Balciuniene and Olivia Baldivieso and Ana Banc and Covadonga Bascaran and Christos Bergeles and Ciara Bergin and Carolina Bernal-Morales and Ansgar Beuse and Mukharram M. Bikbov and Guzel M. Bikbova and Iryna Blagun and Damir Bosnar and Rupert Bourne and Tasanee Braithwaite and Caroline Brandl and Andres Bravo},
  journal = {Ophthalmology Science},
  year = {2026},
  month = aug,
  doi = {10.1016/j.xops.2026.101384},
  url = {https://doi.org/10.1016/j.xops.2026.101384},
  note = {CC BY-NC},
}

TY  - JOUR
TI  - Global RETFound: A Protocol for the World’s First Global Medical Foundation Model
AU  - Paul Nderitu, FRCOphth, PhD
AU  - Yukun Zhou, PhD
AU  - An Ran Ran, MMed, PhD
AU  - Jocelyn Hui Lin Goh, BEng
AU  - Hongyang Jiang, PhD
AU  - Sahana Srinivasan, BEng
AU  - Gatera Fiston Kitema, MPHEC, PhD
AU  - Polly Rawlinson, BA
AU  - Lucy Arnold, MSc
AU  - Ka Wai Yuen, MSc
AU  - Ke Zou, PhD
AU  - Gabriel Dawei Yang, MMed, PhD
AU  - Carol Y. Cheung, PhD
AU  - Pearse A. Keane, MD, PhD
AU  - Yih Chung Tham, PhD
AU  - Hatem M. Abdelakher
AU  - Eberechukwu Achigbu
AU  - Feyi Grace Adepoju
AU  - Leonard Agbahounga
AU  - Rupesh Agrawal
AU  - Referano Agustiawan
AU  - Islam Sh. Ahmed
AU  - Kwadwo Owusu Akuffo
AU  - Shehab Ahmad Al-Abed
AU  - Esraa Khalaf Al-Ajaleen
AU  - Minhaj Nur Alam
AU  - Sura Yousef Alaroud
AU  - Abdulrahman F. Albloushi
AU  - Neilya Aldasheva
AU  - Tsedeke Asaminew Alemu
AU  - Nchifor Alice
AU  - Faisal Alkathery
AU  - Mutaz Al-Nawaflh
AU  - Otim Ambrose
AU  - Kwesi N. Amissah-Arthur
AU  - Damara Andalia
AU  - Fares Antaki
AU  - Nomo Arlette Francine
AU  - Lucy Arnold
AU  - Louis Arnould
AU  - Simon Arunga
AU  - Tariq Aslam
AU  - Aamir Asrar
AU  - Olufemi Emmanuel Babalola
AU  - Vilma Jurate Balciuniene
AU  - Olivia Baldivieso
AU  - Ana Banc
AU  - Covadonga Bascaran
AU  - Christos Bergeles
AU  - Ciara Bergin
AU  - Carolina Bernal-Morales
AU  - Ansgar Beuse
AU  - Mukharram M. Bikbov
AU  - Guzel M. Bikbova
AU  - Iryna Blagun
AU  - Damir Bosnar
AU  - Rupert Bourne
AU  - Tasanee Braithwaite
AU  - Caroline Brandl
AU  - Andres Bravo
PY  - 2026
DA  - 2026/08/26
JO  - Ophthalmology Science
DO  - 10.1016/j.xops.2026.101384
UR  - https://doi.org/10.1016/j.xops.2026.101384
AB  - Purpose: Foundation models (FMs) are pretrained on large-scale unlabeled data sets and are adaptable to a wide spectrum of downstream diagnostic and prognostic tasks. However, limited representation in existing ophthalmic artificial intelligence (AI) data sets, particularly from Latin America, Africa, the Middle East, and Central Asia, could exacerbate health disparities. We describe an international collaborative protocol to develop and validate Global RETFound, an open-access FM for ophthalmic and oculomics research, pretrained on synthetic and anonymized real color fundus photographs (CFPs). Design: International multicenter study. Subjects: Data from a global consortium of institutions. Methods: International collaborators are supported with Docker-based tools, model/data sharing agreements, open-access licenses, tailored protocols, and standardized guides. Under a flexible 2-tier contribution framework, collaborators share fine-tuned generative AI (GenAI) weights or anonymized real CFPs. Main Outcome Measures: Prespecified outcomes include successful development of a globally representative, open-access FM and downstream performance evaluation across ocular and systemic diseases, assessed using discrimination metrics (area under the receiver operating characteristic curve and area under the precision–recall curve), sensitivity, specificity, F1-score, calibration, and task-specific regression and segmentation measures. Results: The Global RETFound consortium has established an international collaborative study with broad geographic representation. As of the 10th of July 2026, 232 collaborators from 78 countries have been onboarded, and 88 study assets (in the form of GenAI weights or anonymized real CFPs) from 39 countries have been shared. Conclusions: The Global RETFound protocol is a novel, privacy-preserving, inclusive strategy for developing a large-scale FM using synthetic and real CFPs. By enabling global participation from institutions with varying technical resources, and focusing on underrepresented populations, the initiative aims to advance equitable and generalizable FMs for ophthalmic and systemic disease. This initiative is positioned as a blueprint for future global FM collaborations in other medical specialties. Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
SN  - 2666-9145
ER  -