June 2020
Volume 61, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2020
Assessment of vessel parameters as a micro vascular biomarker using a Retinal Vessel Analysis System (VASP)
Author Affiliations & Notes
  • Rajiv Raman
    Sankara Nethralaya, Chennai, TN, India
  • Sajib Saha
    1. The Commonwealth Scientific and Industrial Research Organisation (CSIRO), Western Australia, Australia
  • Shaun Frost
    1. The Commonwealth Scientific and Industrial Research Organisation (CSIRO), Western Australia, Australia
  • Rehana Khan
    Sankara Nethralaya, Chennai, TN, India
  • Tarun Sharma
    Sankara Nethralaya, Chennai, TN, India
  • Yogesan Kanagasingam
    1. The Commonwealth Scientific and Industrial Research Organisation (CSIRO), Western Australia, Australia
  • Footnotes
    Commercial Relationships   Rajiv Raman, None; Sajib Saha, None; Shaun Frost, None; Rehana Khan, None; Tarun Sharma, None; Yogesan Kanagasingam, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 496. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Rajiv Raman, Sajib Saha, Shaun Frost, Rehana Khan, Tarun Sharma, Yogesan Kanagasingam; Assessment of vessel parameters as a micro vascular biomarker using a Retinal Vessel Analysis System (VASP). Invest. Ophthalmol. Vis. Sci. 2020;61(7):496.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To assess retinal vessel parameters in diabetic patients and to identify retinal biomarkers to predict diabetic retinopathy and correlating it with the systemic characteristics.

Methods : Vessel parameters were generated using a web based platform called VASP. It allows retinal images from various fundus cameras, pre-process and perform automated detection of optic disc and macula. It also generates the vessel network from the image and identify arteries and veins, bifurcations and cross- overs points. Generation of vascular parameters is a multi-stage process. In this study we have used 27 parameters that are applicable in diabetic retinopathy. 21 pairs of baseline and follow-up images of patients affected by diabetes were randomly chosen from Sankara Nethralaya dataset. Difference in thickness between the baseline and follow-up images were computed and normalized with the actual thickness of the baseline image. Prior to computing vascular parameters, baseline and follow-up images were registered based on the vessel centerline.

Results : Majority of these parameters shows 10~20% changes over time. Parameters like, Width ratio and tortuosity for vein in zone B, C & D in which the changes were consistent. Out of 27 parameters, few vessel parameters shows strong correlation with the systemic characteristics. Fractal dimension for vein in zone A, B & C and Number of trees with branch in zone B & C, were statistically increased from baseline to follow with (p Value – 0.014 and 0.043) respectively. It shows strong correlation with arterial pressure, diastolic blood pressure and lipoproteins level (r=0.559, r=0.582). Junctional exponent deviation for artery in zone B & C was statistically decreased from baseline to follow up, (p= 0.033). It shows strong correlation with triglycerides, lipoproteins, cholesterol and pulse pressure levels, (r=0.621).

Conclusions : Identifying early micro vascular changes in patients with diabetes mellitus will allow for the earlier intervention and treatment. Systemic characteristics like Blood pressure and lipid profile should be given more importance for the earlier intervention.

This is a 2020 ARVO Annual Meeting abstract.

 

Figure 1: a) Six major vessels classified into arteries and veins by VASP b) Baseline image, c) follow-up image, d) mosaic image after registration. Fundus area that was common to both of the images were only used for the analysis

Figure 1: a) Six major vessels classified into arteries and veins by VASP b) Baseline image, c) follow-up image, d) mosaic image after registration. Fundus area that was common to both of the images were only used for the analysis

 

Figure 2: Dividation of fundus image into zones

Figure 2: Dividation of fundus image into zones

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×