Reticular pseudodrusen (RPD) signify a critical phenotype driving vision loss in age-related macular degeneration (AMD). Their detection is paramount in the clinical management of those with AMD, yet they remain challenging to reliably identify. We thus developed a deep learning (DL) model to segment RPD from 9,800 optical coherence tomography B-scans, and this model produced RPD segmentations that had higher agreement with four retinal specialists (Dice similarity coefficient [DSC]=0.76 [95% confidence interval [CI] 0.71-0.81]) than the agreement amongst the specialists (DSC=0.68, 95% CI=0.63-0.73; p<0.001). In five external test datasets consisting of 1,017 eyes from 812 individuals, the DL model detected RPD with a similar level of performance as two retinal specialists (area-under-the-curve of 0.94 [95% CI=0.92-0.97], 0.95 [95% CI=0.92-0.97] and 0.96 [95% CI=0.94-0.98] respectively; p≥0.32). This DL model enables the automatic detection and quantification of RPD with expert-level performance, which we have made publicly available.

RPF reports personal fees from Alimera, Bayer, Biogen, Bohringer-Ingelheim, Caterna, Ellex, Novartis, ODOS, Ophtea, ProGenerika, and Roche/Genentech, and research funding from Biogen, CentreVue and Zeiss, outside of the submitted work. JHT reports research funding from Roche, Novartis, Bayer, Icare, Zeiss and Heidelberg Engineering, and personal fees from Novartis and Okko. PHG reports personal fees from Bayer, Horus, Zeiss, Novartis, Roche, Retinsight, Thea and Abbvie/Allergan outside the submitted work. LA reports personal fees from Horus, Thea, and Abbvie/Allergan outside the submitted work. PvW reports personal fees from Roche/Genentech, Bayer, Novartis, and Mylan outside the submitted work. RS reports employment with Apellis Pharmaceuticals outside the submitted work. AT reports personal fees from 4DMT, Adverum, Annexon, Apellis, Aviceda, Boehringer Ingleheim, Heidelberg Engineering, Iveric Bio, Janssen, Nanoscope, Novartis, OcuTerra, Ocular Therapeutix, Regenxbio, and Roche/Genentech outside the submitted work. AP reports personal fees from PYC Therapeutics and Cartherics outside the submitted work. RHG reports personal fees from Roche/Genentech, Bayer, Novartis and Apellis, Belite Bio, Ocular Therapeutix, Complement Therapeutics, Boehringer Ingelheim Pharmaceuticals, Character Bioscience, Janssen, AbbVie and Astellas outside the submitted work. All other authors report nothing to disclose.

This study was supported by the National Health & Medical Research Council of Australia (APP1027624 [RHG]; GNT1194667 [RHG], #2008382 [ZW]; GNT1181010 [RHG, ELF, MB, AP, BREA, ZW]), the Macular Disease Foundation Australia (research grant [ZW, RHG]), National Institutes of Health (R01AG060942 [CSL, YB], OT2OD32644 [CSL, AYL]), and Research to Prevent Blindness (an unrestricted grant [YB, CSL, AYL]). CERA receives operational infrastructure support from the Victorian Government. The sponsor or funding organization had no role in the design or conduct of this research.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

Human research ethics committee (HREC) or institutional review board (IRB) approval (and not a waiver) was obtained for each of the following studies as follows: Laser Intervention in the Early Stages of Age-Related Macular Degeneration (LEAD) Study: Adelaide Eye and Retina Centre: Bellberry Human Research Ethics Committee (2012-04-753-AA); Centre for Eye Research Australia: Human Research & Ethics Committee of the Royal Victorian Eye & Ear Hospital (11/1027H); Retinology Institute Victoria: Cabrini Human Research Ethics Committee recognises the approval of the Human Research & Ethics Committee of the Royal Victorian Eye & Ear Hospital (11/1027H); Lions Eye Institute: Sir Charles Gairdner Group Human Research Ethics Committee (2013-118); Marsden Eye Research: Bellberry Human Research Ethics Committee (2012-04-753); Belfast Health and Social Care Trust: Belfast HSC Trust (13068UC-SW). MACUSTAR: CS015: University Hospital Bonn Ethics Committee (384/17); CS003: Paris Ouest IV Ethics Committee (04/18_2); CS001: AIBILI Ethics Commitee (032/2017/AIBILI/CE); CS032: Comissao de Etica para a Saude do CHSJ (dated 12/01/2018); CS010: London Queen Square Research Ethics Committee (18/LO/0145); CS030: Center for Sundhed Glostrup Ethics Committee (H-18000126); CS034: Comitato Etico Milano (37910/2018); CS067: Il Comitato Etico dell Ospedale San Raffaele (dated 25/10/2018); CS017: Radboudumc Technology Center Ethics Committee (2017-3954); CS106: LUMC commissie medische ethiek (L18.055/SH/sh). The Northern Ireland Cohort for Longitudinal Study of Ageing (NICOLA): School of Medicine, Dentistry and Biomedical Sciences Ethics Committee, Queen's University Belfast (23/12) University of Washington: University of Washington Institutional Review Board (52049) University of Bonn: University Hospital Bonn Ethics Committee (045/18; 513/20) Montrachet: Kremlin Bicetre and Dijon Ethics Committee (2009-A00448-49)

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes

The deep learning model developed in this study has been made publicly available online here: https://github.com/uw-biomedical-ml/detectron2-rpd-yb. However, the datasets used to develop and externally evaluate the deep learning model in this study is not publicly available, as they either relate to real-world clinical data that cannot be shared publicly or require written research collaboration agreements.

https://github.com/uw-biomedical-ml/detectron2-rpd-yb