Browse by author
Lookup NU author(s): Professor David SteelORCiD
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
© The Author(s) 2024. 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.
Author(s): Rajesh AE, Olvera-Barrios A, Warwick AN, Wu Y, Stuart KV, Biradar MI, Ung CY, Khawaja AP, Luben R, Foster PJ, Cleland CR, Makupa WU, Denniston AK, Burton MJ, Bastawrous A, Keane PA, Chia MA, Turner AW, Lee CS, Tufail A, Lee AY, Egan C, Zheng Y, Yates M, Woodside J, Williams C, Williams K, Weedon M, Vitart V, Viswanathan A, Trucco E, Thomas D, Tapp R, Sun Z, Sudlow C, Strouthidis N, Stratton I, Steel D, Sivaprasad S, Sergouniotis P, Self J, Sattar N, Rudnicka A, Rahi J, Pontikos N, Petzold A, Peto T, Paterson E, Patel P, Owen C, Oram R, O'Sullivan E, Morgan J, Moore T, McKibbin M, McKay G, McGuinness B, Madhusudhan S, Mackie S, MacGillivray T, Luthert P, Lotery A, Littlejohns T, Lascaratos G, Hysi P, Hogg R, Harding S, Hardcastle A, Hammond C, Guggenheim J, Gibson J, Garway-Heath DT, Gallacher J, Ennis S, Doney A, Dick A, Dhillon B, Desai P, Day A, Chua S, Chan M, Chakravarthy U, Carare R, Braithwaite T, Black G, Bishop P, Barrett J, Barman S, Balaskas K, Atan D, Aslam T, Allen N
Publication type: Article
Publication status: Published
Journal: Nature Communications
Year: 2025
Volume: 16
Online publication date: 01/02/2025
Acceptance date: 05/12/2024
Date deposited: 13/01/2025
ISSN (electronic): 2041-1723
Publisher: Springer Nature
URL: https://doi.org/10.1038/s41467-024-55198-7
DOI: 10.1038/s41467-024-55198-7
Data Access Statement: Access to the UK Biobank is restricted to safeguard the privacy of the participants and requires an application. The restrictions depend on the level of access granted. One can apply for access on their website. You cannot share the UK Biobank data with researchers who are not registered with the UK Biobank. Registrations are reviewed within 10 working days of submission. The length of access depends on the access granted. Access to the EPIC-Norfolk Eye study is restricted and requires an application because of the desire to safeguard the privacy of participants. One can request access via the EPIC-Norfolk Management Committee. The data is available to researchers with relevant scientific and ethics approvals for their research, including those in other countries and in commercial companies who are looking for new treatments or laboratory tests. Applications are generally reviewed within 1 month.
Altmetrics provided by Altmetric
