Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene

Nikolas Pontikos; William A Woof; Siying Lin; Biraja Ghoshal; Bernardo S Mendes; Advaith Veturi; Quang Nguyen; Behnam Javanmardi; Michalis Georgiou; Alexander Hustinx; Miguel A Ibarra-Arellano; Ismail Moghul; Yichen Liu; Kristina Pfau; Maximilian Pfau; Mital Shah; Jing Yu; Saoud Al-Khuzaei; Siegfried K Wagner; Malena Daich Varela; Thales Antonio Cabral de Guimarães; Sagnik Sen; Gunjan Naik; Dayyanah Sumodhee; Dun Jack Fu; Nathaniel Kabiri; Jennifer Furman; Bart Liefers; Aaron Y Lee; Samantha R De Silva; Caio Marques; Fabiana Motta; Yu Fujinami-Yokokawa; Alison J Hardcastle; Gavin Arno; Birgit Lorenz; Philipp Herrmann; Kaoru Fujinami; Juliana Sallum; Savita Madhusudhan; Susan M Downes; Frank G Holz; Konstantinos Balaskas; Andrew R Webster; Omar A Mahroo; Peter M Krawitz; Michel Michaelides

doi:10.1038/s42256-025-01040-8

Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene

Nat Mach Intell. 2025;7(6):967-978. doi: 10.1038/s42256-025-01040-8. Epub 2025 Jun 18.

Authors

Nikolas Pontikos^{1

2}, William A Woof^{1

2}, Siying Lin^{1

2}, Biraja Ghoshal^{1

2}, Bernardo S Mendes^{1

2}, Advaith Veturi¹, Quang Nguyen¹, Behnam Javanmardi³, Michalis Georgiou^{1

2}, Alexander Hustinx³, Miguel A Ibarra-Arellano³, Ismail Moghul^{1

2}, Yichen Liu^{1

2}, Kristina Pfau^{4

5}, Maximilian Pfau^{4

5}, Mital Shah^{1

2

6}, Jing Yu⁷, Saoud Al-Khuzaei^{6

7}, Siegfried K Wagner^{1

2}, Malena Daich Varela^{1

2}, Thales Antonio Cabral de Guimarães^{1

2

8}, Sagnik Sen^{1

2}, Gunjan Naik^{1

2}, Dayyanah Sumodhee^{1

2}, Dun Jack Fu^{1

2}, Nathaniel Kabiri¹, Jennifer Furman⁹, Bart Liefers², Aaron Y Lee^{10

11}, Samantha R De Silva^{6

7}, Caio Marques¹², Fabiana Motta¹², Yu Fujinami-Yokokawa¹³, Alison J Hardcastle^{1

2}, Gavin Arno^{1

2}, Birgit Lorenz⁴, Philipp Herrmann⁴, Kaoru Fujinami¹³, Juliana Sallum¹², Savita Madhusudhan¹⁴, Susan M Downes^{6

7}, Frank G Holz⁴, Konstantinos Balaskas^{1

2}, Andrew R Webster^{1

2}, Omar A Mahroo^{1

2}, Peter M Krawitz³, Michel Michaelides^{1

2}

Affiliations

¹ University College London Institute of Ophthalmology, University College London, London, UK.
² Moorfields Eye Hospital, London, UK.
³ Institute for Genomic Statistic and Bioinformatics, University Hospital Bonn, Rheinische-Friedrich-Wilhelms University, Bonn, Germany.
⁴ Department of Ophthalmology, University Hospital Bonn, Rheinische-Friedrich-Wilhelms Universität Bonn, Bonn, Germany.
⁵ Department of Ophthalmology, University Hospital Basel, Basel, Switzerland.
⁶ Oxford Eye Hospital, John Radcliffe Hospital, Oxford, UK.
⁷ Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, Oxford, UK.
⁸ Department of Ophthalmology, Faculdade São Leopoldo Mandic, Campinas, São Paulo, Brazil.
⁹ UCL Translational Research Office, UCL Maple House, London, UK.
¹⁰ Department of Ophthalmology, University of Washington School of Medicine, Seattle, WA USA.
¹¹ Roger and Angie Karalis Johnson Retina Center, University of Washington, Seattle, WA USA.
¹² Department of Ophthalmology and Visual Sciences, Escola Paulista de Medicina, Federal University of Sao Paulo, São Paulo, Brazil.
¹³ Laboratory of Visual Physiology, Division of Vision Research, National Institute of Sensory Organs, National Hospital Organization Tokyo Medical Center, Tokyo, Japan.
¹⁴ St Paul's Eye Unit, The Royal Liverpool and Broadgreen University Hospitals, Liverpool, UK.

Abstract

Rare eye diseases such as inherited retinal diseases (IRDs) are challenging to diagnose genetically. IRDs are typically monogenic disorders and represent a leading cause of blindness in children and working-age adults worldwide. A growing number are now being targeted in clinical trials, with approved treatments increasingly available. However, access requires a genetic diagnosis to be established sufficiently early. Critically, the timely identification of a genetic cause remains challenging. We demonstrate that a deep learning algorithm, Eye2Gene, trained on a large multimodal imaging dataset of individuals with IRDs (n = 2,451) and externally validated on data provided by five different clinical centres, provides better-than-expert-level top-five accuracy of 83.9% for supporting genetic diagnosis for the 63 most common genetic causes. We demonstrate that Eye2Gene's next-generation phenotyping can increase diagnostic yield by improving screening for IRDs, phenotype-driven variant prioritization and automatic similarity matching in phenotypic space to identify new genes. Eye2Gene is accessible online (app.eye2gene.com) for research purposes.

Keywords: Genetic testing; Image processing; Molecular medicine; Retinal diseases.

Abstract

Grants and funding