July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Fractal Dimension Analysis of OCTA Images in Normal and Diabetic Eyes using the Circular Mass-radius Method
Author Affiliations & Notes
  • Jennifer Lopez
    Ophthalmology, New York University, New York, New York, United States
    NYU School of Medicine, San Jose, California, United States
  • Bing Chiu
    Ophthalmology, New York University, New York, New York, United States
  • Harrison Chiu
    Ophthalmology, New York University, New York, New York, United States
  • Preethi Kumar
    Ophthalmology, New York University, New York, New York, United States
  • Shariq Hashmi
    Ophthalmology, New York University, New York, New York, United States
    NYU School of Medicine, San Jose, California, United States
  • Akash Gupta
    Ophthalmology, New York University, New York, New York, United States
  • Soshian Sarrafpour
    Ophthalmology, New York University, New York, New York, United States
  • Joshua A Young
    Ophthalmology, New York University, New York, New York, United States
  • Footnotes
    Commercial Relationships   Jennifer Lopez, None; Bing Chiu, None; Harrison Chiu, None; Preethi Kumar, None; Shariq Hashmi, None; Akash Gupta, None; Soshian Sarrafpour, None; Joshua Young, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 5345. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jennifer Lopez, Bing Chiu, Harrison Chiu, Preethi Kumar, Shariq Hashmi, Akash Gupta, Soshian Sarrafpour, Joshua A Young; Fractal Dimension Analysis of OCTA Images in Normal and Diabetic Eyes using the Circular Mass-radius Method. Invest. Ophthalmol. Vis. Sci. 2019;60(9):5345.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Fractal analysis has been used to investigate lung and brain tumors as well as the retinal vasculature. We analyzed fractal dimensions (FD) of macular optical coherence tomography angiography (OCTA) images via circular mass-radius method to evaluate their role in differentiating microvascular differences between normal and diabetic eyes.

Methods : A retrospective study was conducted using OCTA images of 3x3mm and 6x6mm scans for superficial and deep capillary plexuses from 49 control eyes and 58 eyes with diabetic retinopathy. ImageJ was used to standardize and binarize grayscale OCTA images with default thresholding of background pixel values. Fractalyse was used to determine fractal dimension and correlation coefficient via circular mass-radius method centered at the barycentre. One-way analysis of variance (ANOVA) and post hoct-tests with Bonferroni correction were used to compare control eyes to those with proliferative diabetic retinopathy (PDR) and each stage of non-proliferative diabetic retinopathy (NPDR).

Results : FD of control and diabetic eyes were determined at default threshold values for 3x3mm and 6x6mm OCTA images using the mass-radius method (Table 1). ANOVA showed a significant difference between the FD of control and diabetic eyes in both the 3x3mm and 6x6mm deep plexus scans, and the 3x3mm superficial plexus scan (p <0.05). In the 3x3mm superficial plexus, the FD of severe NPDR and PDR were significantly lower compared to control. The scans of the deep plexus showed only moderate NPDR was significantly reduced in the 6x6mm scan, while moderate NPDR, severe NPDR and PDR were significantly lower in the 3x3mm scan (Table 2).

Conclusions : The mass-radius FD were significantly lower in eyes with diabetic retinopathy for the 3x3mm scans of the superficial and deep plexus layer; however, were only significantly lower in the 6x6mm scan of the deep plexus layer. The study suggests using 3x3mm scans for fractal analysis may be more effective as a measure of microvascular dropout compared to the 6x6mm scan. Furthermore, the data suggests diabetic changes to the deep plexus layer may be more substantial than changes to the superficial plexus.

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

 

Table 1. Mean FD and Standard Deviation

Table 1. Mean FD and Standard Deviation

 

Table 2. Post hoct-tests with Bonferroni correction

Table 2. Post hoct-tests with Bonferroni correction

×
×

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.

×