Finding Ophthalmic Risk and Evaluating the Value of Eye exams and their predictive Reliability (FOREVER) - dataset

The Finding Ophthalmic Risk and Evaluating the Value of Eye exams and their predictive Reliability (FOREVER) dataset was developed in a collaborative effort with Synoptik A/S under the supervision of a consortium of international partners. It consists of extensive eye examinations carried out in optician stores, questionnaire responses, and for a subpopulation blood pressure measurements and saliva samples. In addition to the tabular examination data, e.g. intraocular pressure and visual acuity, the database consists of fundus images and optical coherence tomography (OCT) scans for a subset of the participants obtained through following data extraction methods: Fundus images are in DICOM format and can be analysed in a custom viewer or using the raw data. In addition, we provide vascular and optic cup/disc measurements extracted by AutoMorph. The raw OCT data are stored as Topcon .fda files and cannot be made available. However, the pixel arrays and segmentations can be extracted and provided in NumPy or DICOM format. Additionally, summary metrics are accessible in tabular format. The entire file can be viewed in Topcon IMAGEnet 6. All participant data is linked to the Danish National Health Registries with information about the participants' personal health history, including prescriptions and hospital diagnoses and procedure codes. To date, about 8000 saliva samples have been collected accompanied by 80,000 participants. An overall description of the data collection methods is provided on the website of Project FOREVER and in the corresponding method paper. As the dataset contains personal sensitive information, access to the dataset can only be granted after an application is submitted and approved by the steering group via this https://redcap.regionh.dk/surveys/?s=ELEFHYYN9KKT7LRK. After a positive decision, a data access contract will be signed between the two institutions. Authorized Users will then receive access to a secure compute environment. Please note that the data cannot be copied from the server, however, the access fee includes support for creating your personal subset of the dataset and assistance in moving necessary files onto the server. An overview of access fees is provided on the website. On the current platform it is possible to obtain access to an NVIDIA A100 GPU with a usage-based charge. The files attached to this Zenodo entry provide an overview of the variables included in the data. Please note that not all variables contain a value for each examination or participant. A complete overview of available entries is provided upon request. If the dataset is used in a research project the results should be published and articles should cite the dataset as well as the related articles mentioned under Related Works.