Related Project:

Retina: vessel analysis


Retinal Vessel Tortuosity Data Set






Retinal Vessel Tortuosity [.zip] [size: 16.2 MB]

Details:

The dataset contains 60 images of retinal vessels from normal and hypertensive patients and information about their manually estimated tortuosity.
  • 30 acquired images of retinal arteries of similar length and caliber (Arteries_with_Markers.zip), and

  • 30 images of the same arteries preprocessed by a normalization algorithm (Reduced_Arteries_Iso.zip) [1].

  • 30 acquired images of retinal veins of similar length and caliber (Veins_with_Markers.zip), and

  • 30 images of the same veins preprocessed by a normalization algorithm (Reduced_Veins_Iso.zip) [1]

  • Matlab data structures (ManualData.mat) containing for each record:
    - the name of the image
    - the x and y samples of the manually drawn vessel center line
    - the ordinal position of the vessel in the manual ordering for increasing tortuosity.

  • List of image names ordered by increasing tortuosity (Clinical Ordering.xls). Vessels were manually ordered by a retinal specialist, Dr. S. Piermarocchi, Department of Ophthalmology, University of Padova.

[1] M. Foracchia, E. Grisan, and A. Ruggeri: ‘Luminosity and Contrast Normalization in Retinal Images’, Medical Image Analysis, 9 (3):179-190, 2005

The acquired images were acquired with a 50° fundus camera (TRC 50, Topcon, Japan) and digitized with a scanner at 1100x1300 pixels.

Image Format. Acquired images: JPG, compressed, color. Preprocessed images: TIF non compressed, color.

Source: Department of Ophthalmology, University of Padova, Padova, Italy.

We request any researcher reporting results using this database to acknowledge the database by referencing the publication:

E. Grisan, M. Foracchia, A. Ruggeri: ‘A novel method for the automatic grading of retinal vessel tortuosity’, IEEE Transactions on Medical Imaging, 2008, 27(3), 310-319

and by acknowledging the contribution of Dr. Piermarocchi:

The authors wish to thank Dr. S. Piermarocchi, from the Department of Ophthalmology, University of Padova, Italy, for having kindly provided the fundus images and the manual tortuosity grading.
 



Download:

All the images and manual data in the RET-TORT database are compressed in one zip file of approximately 16.2 MB.

In order to verify the download, we ask you to fill in the form below, indicating a valid e-mail address.
An e-mail message containing a link to download the database will be sent to this address in a few seconds. Your e-mail adddress will be used only for verification purposes.


Name Surname
Istitution e-mail

return to top