July 2019
Volume 60, Issue 9
ARVO Annual Meeting Abstract  |   July 2019
Automated Image Alignment for Comparing Vascular Changes in Fundus Fluorescein Angiography and Optical Coherence Tomography Angiography in the Macula of Patients with Diabetic Retinopathy
Author Affiliations & Notes
  • David J Ramsey
    Ophthalmology, Tufts University School of Medicine, Boston, Massachusetts, United States
    Ophthalmology, Lahey Hospital and Medical Center, Peabody, Massachusetts, United States
  • Ayman G Elnahry
    Department of Ophthalmology, Cairo University, Cairo, Egypt
  • Footnotes
    Commercial Relationships   David Ramsey, None; Ayman Elnahry, None
  • Footnotes
    Support  None.
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 3035. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      David J Ramsey, Ayman G Elnahry; Automated Image Alignment for Comparing Vascular Changes in Fundus Fluorescein Angiography and Optical Coherence Tomography Angiography in the Macula of Patients with Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2019;60(9):3035.

      Download citation file:

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

  • Supplements

Purpose : To compare retinal vascular changes in the macula of diabetic patients using fundus fluorescein angiography (FFA) and optical coherence tomography angiography (OCTA) images aligned using a commercially available alignment software.

Methods : Patients with diabetic retinopathy were recruited from Cairo University retina clinic. FFA was performed with either a TRC 50DX (Topcon, Tokyo, Japan) or a Spectralis HRA2 (Heidelberg Engineering, Heidelberg, Germany) camera. OCTA was performed by an Optovue RTvue XR Avanti machine (Optovue, Fremont, CA) using the 6x6 mm macular scan. All FFA images were automatically cropped and aligned with their respective OCTA image as the target using the i2k Align Retina software (Dual-Align, Clifton Park, NY). Each image was then randomized and evaluated after removing identifiers. The foveal avascular zone (FAZ) was manually traced and microaneurysms (MAs) marked using ImageJ (NIH, Bethesda, MD).The fractal dimension (FD), a global measurement of the density and complexity of the vasculature, was calculated using the FracLac plugin of ImageJ.

Results : Eighteen eyes of 13 patients were successfully aligned; 6 eyes could not be registered by the software and were excluded, 1 eye was excluded due to movement artifact, and 1 eye did not have a macular scan available. The FAZ area was 0.461±0.153 mm2 for the FFA images versus 0.413±0.166 mm2 in the OCTA images (NS). The FD was 1.664±0.048 (range 1.562 to 1.751) for FFA images compared to 1.717±0.023 (range 1.688 to 1.767) for OCTA images (p<0.001). Significantly more MAs were identified in the FFA images (55.50±32.93) compared to the OCTA images (25.28±10.56, p<0.0001). The number of MAs identified in the OCTA image as a percentage of total in the FFA image declined as the total number of MAs increased (R2=0.516, p<0.001).

Conclusions : OCTA is a non-invasive tool capable of depth resolving the retinal vasculature of the macula in higher detail when compared to conventional FFA. Although both modalities were capable of resolving the FAZ, the retinal vascular fractal analysis demonstrates that OCTA is capable of imaging the central macula of diabetic patients with greater resolution of detail compared to FFA. OCTA, however, detects far fewer MAs compared to FFA, a limitation illustrated by overlaying images following alignment.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.


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.