Machine learning derived retinal pigment score from ophthalmic imaging shows ethnicity is not biology
Anand E. Rajesh, Abraham Olvera‐Barrios, Alasdair Warwick, Yue Wu, Kelsey V. Stuart, Mahantesh I. Biradar, Chuin Ying Ung, Anthony P. Khawaja, Robert Luben, Paul J. Foster, Charles R. Cleland, William Makupa, Alastair K. Denniston, Matthew J. Burton, Andrew Bastawrous, Pearse A. Keane, Mark A. Chia, Angus Turner, Cecilia S. Lee, Adnan Tufail, Aaron Lee, Catherine Egan, Naomi E. Allen, Tariq Aslam, Denize Atan, Konstantinos Balaskas, Sarah Barman, Jenny Barrett, Paul Bishop, Graeme C. Black, Tasanee Braithwaite, Roxana O. Carare, Usha Chakravarthy, Michelle Chan, Sharon Chua, Alexander Day, Parul Desai, Baljean Dhillon, Andrew D. Dick, Alex S. F. Doney, Sarah Ennis, John Gallacher, David F. Garway‐Heath, Jane Whitney Gibson, Jeremy A. Guggenheim, Christopher J. Hammond, Alison J. Hardcastle, Simon Harding, Ruth Hogg, Pirro G. Hysi, Gerassimos Lascaratos, Thomas J. Littlejohns, Andrew Lotery, Philip J. Luthert, Tom MacGillivray, Sarah Mackie, Savita Madhusudhan, Bernadette McGuinness, Gareth J. McKay, Martin McKibbin, Tony Moore, James P. Morgan, Eoin O’Sullivan, Richard A. Oram, Christopher G. Owen, Praveen J. Patel, Euan Paterson, Tünde Pető, Axel Petzold, Nikolas Pontikos, Jugnoo S. Rahi, Alicja R. Rudnicka, Naveed Sattar, Jay Self, Panagiotis I. Sergouniotis, Sobha Sivaprasad, David Steel, Irene Stratton, Nicholas G. Strouthidis, Cathie Sudlow, Zihan Sun, Robyn J. Tapp, Dhanes Thomas, Emanuele Trucco, Ananth C. Viswanathan, Véronique Vitart, Mike Weedon, Katie Williams, Cathy Williams, Jayne V. Woodside, Max Yates, Yalin Zheng
Abstract
Few metrics exist to describe phenotypic diversity within ophthalmic imaging datasets, with researchers often using ethnicity as a surrogate marker for biological variability. We derived a continuous, measured metric, the retinal pigment score (RPS), that quantifies the degree of pigmentation from a colour fundus photograph of the eye. RPS was validated using two large epidemiological studies with demographic and genetic data (UK Biobank and EPIC-Norfolk Study) and reproduced in a Tanzanian, an Australian, and a Chinese dataset. A genome-wide association study (GWAS) of RPS from UK Biobank identified 20 loci with known associations with skin, iris and hair pigmentation, of which eight were replicated in the EPIC-Norfolk cohort. There was a strong association between RPS and ethnicity, however, there was substantial overlap between each ethnicity and the respective distributions of RPS scores. RPS decouples traditional demographic variables from clinical imaging characteristics. RPS may serve as a useful metric to quantify the diversity of the training, validation, and testing datasets used in the development of AI algorithms to ensure adequate inclusion and explainability of the model performance, critical in evaluating all currently deployed AI models. The code to derive RPS is publicly available at: https://github.com/uw-biomedical-ml/retinal-pigmentation-score .