To develop a comprehensive dataset for anterior segment eye disease to enhance research into conditions affecting the front of the eye.
Approach:
Dataset Development: The CADMUS dataset includes 945,243 images from 22,482 patients, collected during routine clinical care at Moorfields Eye Hospital from December 2019 to September 2024.
Data Composition: The dataset encompasses various anterior segment conditions, including keratoconus, Fuchs’ corneal dystrophy, corneal scarring, and cataracts, and includes over 40,000 surgical records from more than 12,000 patients.
Data Integration: CADMUS combines imaging data, quantitative measurements, and linked electronic health records.
Key Findings:
96% of patients have multiple follow-up visits, allowing for the study of disease progression.
The dataset supports research into keratoconus progression and deep learning models.
Routine clinical images contain biological signals that can predict patient age and biological sex.
Interpretation:
The CADMUS dataset is a significant resource for advancing research in anterior segment diseases.
Limitations:
The dataset is limited to patients treated at Moorfields Eye Hospital, which may affect its generalizability to other populations.
Conclusion:
The CADMUS dataset is a significant resource for advancing research in anterior segment eye diseases.
