August 2019
Volume 60, Issue 11
Open Access
ARVO Imaging in the Eye Conference Abstract  |   August 2019
Method for artificially degrading the signal from Optical Coherence Tomography cube to assess variability in retinal pigment epithelial detachment volumetric measurement
Author Affiliations & Notes
  • Sophie Kubach
    Carl Zeiss Meditec, Inc, Dublin, California, United States
  • Luis De Sisternes
    Carl Zeiss Meditec, Inc, Dublin, California, United States
  • Warren Lewis
    Bayside Photonics, Inc, Yellow Springs, Ohio, United States
  • Giovanni Gregori
    Bascom Palmer Eye Institute, Miami, Florida, United States
  • Footnotes
    Commercial Relationships   Sophie Kubach, Carl Zeiss Meditec, Inc (E); Luis De Sisternes, Carl Zeiss Meditec, Inc (E); Warren Lewis, Carl Zeiss Meditec, Inc (C); Giovanni Gregori, Carl Zeiss Meditec, Inc (C)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science August 2019, Vol.60, PB0175. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Sophie Kubach, Luis De Sisternes, Warren Lewis, Giovanni Gregori; Method for artificially degrading the signal from Optical Coherence Tomography cube to assess variability in retinal pigment epithelial detachment volumetric measurement. Invest. Ophthalmol. Vis. Sci. 2019;60(11):PB0175.

      Download citation file:

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

  • Supplements

Purpose : Measurement of retinal pigment epithelium detachment volume (RPED) is an important characteristic to be able to measure when designing the needed signal in an OCT system. In this study we assess the variability in the RPED volume in dry AMD eyes with signal-to-noise reduction by artificially degrading the OCT signal

Methods : Repeated Angio 6x6mm scans from 5 dry AMD eyes imaged on the PLEX® Elite 9000 swept-source OCT (ZEISS, Dublin, CA) were artificially corrupted to simulate different reduction in signal. The method to reduce the OCT signal is the following: Data from an 8-bit OCT cube are unlogged after determining the pixels value above and below the noise level. The noise is then subtracted from the signal and the signal is scaled by a factor 10(deltadB/10) where deltadB is the signal reduction factor. The noise is then added back to the signal and the log is taken to produce an 8-bit OCT cube. After segmentation of the Retinal pigment epithelial (RPE), RPED volume corresponding to the volume between the elevated RPE and RPE fit in the central 5 mm circle was computed on the original cubes and on the cubes with different reduction in signal.

Results : Table 1 shows for each eye, the mean and standard deviation for RPED volume within a 5mm radius circle when considering the original dataset (i.e no signal reduction) and the same dataset degraded by 1.5 dB, 3 dB and 4.5 dB.
The sensitivity in RPED volume with OCT signal degradation varies from eye to eye. The change in RPED volume with signal degradation is small, less than half the standard deviation for all eyes except for eye #2 where the change is statistically significant, with a volume reduction of 5 sigma at 4.5 dB from baseline.
Figure 1 shows for each eye the RPED volume and error bar as a function of OCT function of signal. Eye 2 (blue plot) shows a statistical change in RPED volume. If we take a closer look at the RPE elevation map for all 5 eyes (figure 2), there is a clear difference in the RPE elevation profile, with a more pronounced RPE elevation for eye #1 and #3 and smaller elevations more spread out for eye #2.

Conclusions : The described method for artificially degrading the OCT signal can be used to assess the sensitivity of the RPED volume independently from any system-to-system variations.

This abstract was presented at the 2019 ARVO Imaging in the Eye Conference, held in Vancouver, Canada, April 26-27, 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.