June 2015
Volume 56, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2015
Interactive Retinal Blood Flow Analysis of Macular Region using the Retinal Function Imager
Author Affiliations & Notes
  • Jing Tian
    Bascom Palmer Eye Institute, University of Miami, Miami, FL
  • Gabor Mark Somfai
    Bascom Palmer Eye Institute, University of Miami, Miami, FL
  • Delia DeBuc
    Bascom Palmer Eye Institute, University of Miami, Miami, FL
  • Footnotes
    Commercial Relationships Jing Tian, None; Gabor Somfai, None; Delia DeBuc, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2015, Vol.56, 1466. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jing Tian, Gabor Mark Somfai, Delia DeBuc; Interactive Retinal Blood Flow Analysis of Macular Region using the Retinal Function Imager. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):1466.

      Download citation file:


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

      ×
  • Supplements
Abstract
 
Purpose
 

Retinal Function Imager (RFI; Optical Imaging, Rehovot, Israel) is a non-invasive and non-contact functional imaging modality that directly measures hemodynamic parameters, which can facilitate the diagnosis of a variety of ocular and systematic diseases. In this study, we aim to develop a fast and reliable tool to calculate the blood flow and blood velocity using the RFI images.

 
Methods
 

The proposed method is based on the interactive segmentation of blood vessels and optimization of the vessel center intensity profile’s cross-correlation. An RFI blood flow scanning session consists of 8 images taken at the time interval of 17.5 ms. The current built-in software of RFI only measures the blood flow velocity using a time-consuming semi-automatic segmentation of the vessel on the fundus images. In our method, the center line of the vessel is traced automatically by using the shortest-path graph search once the user clicks on the two ends points of the vessels. The vessel walls can be detected automatically on the vessel map and the vessel diameters are obtained easily as illustrated in Fig.1(a). The blood flow velocity is calculated by maximizing the cross-correlation of intensity on the ratio images. The registration of five RFI blood flow scanning sessions is performed using I2K Retina Software and blood velocity is calculated simultaneously as shown in Fig.1 (b).

 
Results
 

Our developed software was tested on 30 vessels from 3 subjects (one healthy, one with severe non-proliferative diabetic retinopathy (NPDR) and one with proliferative diabetic retinopathy (PDR)). Each study subject had 5 serial RFI’s blood flow scanning sessions. The mean of coefficients of variation of blood flow between the five sessions in the three subjects were 10.07%, 17.62% and 12.12%, respectively. The mean blood flow velocities were 3.95, 2.44 and 1.42 mm/s and the mean blood flow were 0.64, 0.51 and 0.12 µl/s for healthy, NPDR and PDR subjects, respectively.

 
Conclusions
 

In this preliminary study, we implemented a faster tool to reliably measure the retinal blood flow and blood velocity. We successfully tested our software in three subjects and a decrease in the blood velocity was observed in the pathological eyes. Further work is needed to evaluate the performance of our methodology.  

 
(a)The automatic segmentation of a vessel from a RFI image, and (b) the observed motion contrast both obtained by our software.
 
(a)The automatic segmentation of a vessel from a RFI image, and (b) the observed motion contrast both obtained by our software.

 
×
×

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.

×