Researchers at Moorfields Eye Hospital NHS Foundation Trust and the UCL Institute of Ophthalmology, London, UK, have developed what they describe as the world’s most comprehensive anterior segment eye disease dataset, designed to accelerate research into conditions affecting the front of the eye.

The CADMUS dataset contains 945,243 images from 22,482 patients, collected during routine clinical care at Moorfields Eye Hospital between December 2019 and September 2024. Approximately 96 percent of patients have multiple follow-up visits, enabling researchers to study disease progression and long-term outcomes.

The dataset includes a broad range of anterior segment conditions, including keratoconus, Fuchs’ corneal dystrophy, corneal scarring, and cataracts. It also contains more than 40,000 surgical records from over 12,000 patients, covering cataract surgery, corneal cross-linking, corneal transplant, keratectomy, and other corneal procedures.

CADMUS combines raw imaging from the MS-39 anterior segment OCT tomographer, derived quantitative measurements such as keratometry and pachymetry, and linked electronic health record data including demographics, diagnoses, visual acuity, refraction, and surgical history.

The datasheet’s lead author, Shafi Balal, an ophthalmic surgeon at Moorfields and NIHR doctoral fellow at UCL, said early work using CADMUS has already supported research into keratoconus progression measurement and deep learning models based on anterior segment scans. “One model can predict patient age and biological sex from anterior segment scans, demonstrating that routine clinical images carry rich biological signals invisible to the human eye,” he added.

The dataset was developed through INSIGHT, the Eye and Oculomics Health Data Research Hub at Moorfields, and has been published as a datasheet in Ophthalmology Science. Researchers will be able to apply for access through INSIGHT’s Data Use Application process.