Machine learning derived retinal pigment score from ophthalmic imaging shows ethnicity is not biology.

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.

Association of Ambient Air Pollution Exposure With Incident Glaucoma: 12-Year Evidence From the UK Biobank Cohort.

Purpose: Glaucoma is the leading cause of irreversible blindness worldwide. Despite growing concerns about air quality and its impact on ocular health, there remains a knowledge gap regarding the long-term association between air pollution and glaucoma risk. This study investigates the relationship between exposure to ambient air pollution and incidence of glaucoma.

How many future dementia cases would be missed by a high-risk screening program? A retrospective cohort study in a population-based cohort.

Introduction: Risk prediction models aim to identify those at high risk to receive targeted interventions. We aimed to identify the proportion of future dementia cases that would be missed by a high-risk screening program.

Methods: We identified validated dementia risk prediction models from systematic reviews.

Unsupervised representation learning on high-dimensional clinical data improves genomic discovery and prediction.

Although high-dimensional clinical data (HDCD) are increasingly available in biobank-scale datasets, their use for genetic discovery remains challenging. Here we introduce an unsupervised deep learning model, Representation Learning for Genetic Discovery on Low-Dimensional Embeddings (REGLE), for discovering associations between genetic variants and HDCD. REGLE leverages variational autoencoders to compute nonlinear disentangled embeddings of HDCD, which become the inputs to genome-wide association studies (GWAS).