Pulse Brain · Growing Health Evidence Index
Tier 3 — Observational / field trialPeer-reviewed

AlzEye: longitudinal record-level linkage of ophthalmic imaging and hospital admissions of 353 157 patients in London, UK

Siegfried K. Wagner, Fintan Hughes, Mario Cortina‐Borja, Nikolas Pontikos, Robbert Struyven, Xiaoxuan Liu, Hugh Montgomery, Daniel C. Alexander, Eric J. Topol, Steffen E. Petersen, Konstantinos Balaskas, Jack Hindley, Axel Petzold, Jugnoo S. Rahi, Alastair K. Denniston, Pearse A. Keane

BMJ Open · 2022

Read source ↗ All evidence

Summary

AlzEye represents a large-scale resource linking longitudinal multimodal retinal imaging from routine NHS ophthalmic care with nationally recorded hospital admission data for over 353,000 patients in London. This cohort, enriched with subpopulations affected by systemic disease, provides a rare large-labelled dataset to support development of deep-learning models for oculomics—the detection of systemic disease signatures from retinal imaging. The study exemplifies a privacy-by-design approach to record-level linkage of routine clinical data across health service boundaries.

UK applicability

This study is directly applicable to UK healthcare, being conducted within the NHS at Moorfields Eye Hospital and utilising nationally collected Hospital Episode Statistics. The methodology and findings could inform development of retinal imaging-based screening or prognostication tools for systemic diseases within NHS ophthalmology and primary care settings.

Key measures

Number of participants with retinal imaging (154,830); total retinal images acquired (6,261,931); hospital admissions (1,337,711 episodes); prevalence of systemic diagnoses (myocardial infarction n=12,022; stroke n=11,735; dementia n=13,363); distribution of retinal image modalities

Outcomes reported

The study linked longitudinal multimodal retinal imaging from NHS routine ophthalmic care with hospital admission data for systemic diseases including myocardial infarction, stroke, and dementia in a cohort of over 353,000 patients aged 40+ years. The dataset comprised over 6.2 million retinal images of seven modalities acquired from over 154,000 patients, with hospital episode statistics capturing 1.3 million admission episodes.

Theme
Measurement & metrics
Subject
Measurement methods & nutrient profiling
Study type
Research
Study design
Observational cohort
Source type
Peer-reviewed study
Status
Published
Geography
United Kingdom
System type
Human clinical
DOI
10.1136/bmjopen-2021-058552
Catalogue ID
BFmor3g48f-n9ks6y

Topic tags

Pulse AI · ask about this record

Dig deeper with Pulse AI.

Pulse AI has read the whole catalogue. Ask about this record, its theme, or how the findings apply to UK farming and policy — every answer cites the underlying studies.