@article{,
title= {Longitudinal diabetic retinopathy screening data},
keywords= {},
author= {},
abstract= {This data set contains
repeated 4-field color fundus photos (1120 in total) of 70 patients in the diabetic retinopathy screening program of the Rotterdam Eye Hospital (Rotterdam, The Netherlands);
the result of intra- and inter-visit registration by two methods (i2kRetina and WeVaR); and
the grading and ranking of these results by two graders.

### Inclusion criteria:
All patients with diabetes who were screened for diabetic retinopathy during one week in June 2013.

### Exclusion criteria:
First-time patients and
Patients who were not examined in the year before.

Fundus image of both eyes of each patient were acquired using a non-mydriatic digital fundus camera (Topcon TRC-NW65) with a 45 degrees field of view after pupil dilation. Images of each visit were registered by two methods: WeVaR (see Adal et al, A Hierarchical Coarse-to-Fine Approach for Image Registration) and i2k Retina (DualAlign LLC; see here). Image mosaic movies were created from the registered, normalized fundus images and were presented to two graders. In addition, the graders compared the image mosaics side-by-side and ranked them. Images of consecutive visits were similarly registered and the image mosaic movies were again presented to two graders.

Included data:
Patient's gender and age
All color fundus images
All normalized fundus images
Intra-visit image mosaic movies for both registration methods
Intra-visit Image mosaics for both registration methods
Inter-visit image mosaic movies for both registration methods
Scores of two graders (not every movie/image was graded by both graders)

### Publications
This data set, or a part thereof, was used by us in the following paper(s). Please cite one or more of these papers in any of your publication(s) that uses (parts of) this data set.

K.M. Adal, P.G. van Etten, J.P. Martinez, L.J. van Vliet, K.A. Vermeer. Accuracy Assessment of Intra and Inter-Visit Fundus Image Registration for Diabetic Retinopathy Screening. Invest Ophthalmol Vis Sci. 2015. Accepted for publication.

### Contributors
The following people did all the hard work on assembling and releasing this data set:
Kedir Adal, Peter van Etten, Jose Martinez and Koen Vermeer

https://i.imgur.com/suusenq.png
},
terms= {The Rotterdam Ophthalmic Institute allows you to use this data free of charges, provided that
you do not use the data for commercial activities;
you do not hold us liable for any errors in the data set;
you do not redistribute the data;
any incidental findings will be reported back to us; and
you will cite one or more of our relevant publications in all publications that make use of this data.},
license= {},
superseded= {},
url= {http://www.rodrep.com/longitudinal-diabetic-retinopathy-screening---description.html}
}