Dataset for Hypertensive Retinopathy

Hypertensive Retinopathy (HR) is a retinal disease that is caused due to consistent high blood pressure (hypertension) and leads to vision loss. The millions of people in the world are suffering from HR disease due to hypertension. The abnormalities occur on the retina due to hypertension such as tortuosity in retinal vessels, hard exudates, cotton wool spots, hemorrhages and swelling of optic disc (OD). The ophthalmologists can examine these retinopathy signs due to HR by capturing the digital images through fundus camera. This disease does not have any early signs and in many cases, HR is diagnosed at later stages when the disease leads to the blindness or vision loss. Therefore, it is necessary for hypertensive patients to have routine examination of their eyes.

Hypertensive Retinopathy Database:

AVRDB (Annotated Dataset for Vessel Segmentation and Calculation of Arteriovenous Ratio) is a newly developed HR database which will be publicly available at in future for research community. It consists of 100 fundus retinal images which are captured through TOPCON TRC-NW8 and annotated with the help of expert ophthalmologists from Armed Forces Institute of Ophthalmology (AFIO), Pakistan. The vascular network is categorized into arteriolar and venular pattern. These 100 images having dimensions of 1504 x 1000 contain retinal arteries, veins, AVR and whole vascular structure for ground truths. It also has annotation at image level for HR and PE.


Please cite our articles if you use this database for evaluation

  • Shahzad Akbar, Muhammad Usman Akram, Muhammad Sharif, Anam Tariq, and Ubaid ullah Yasin. “Decision Support System for Detection of Papilledema through Fundus Retinal Images.” Journal of Medical Systems 41, no. 4 (2017): 66.
  • Shahzad Akbar, Taimur Hassan, M. Usman Akram, Ubaid Ullah Yasin, Imran Basit, “AVRDB: Annotated Dataset for Vessel Segmentation and Calculation of Arteriovenous Ratio”, 21st International Conference on Image Processing, Computer Vision & Pattern Recognition, July 17-20, 2017, Las Vegas, USA.

Request for Dataset: