May 2008
Volume 49, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2008
Wavelet Based Assessment of Photoreceptor Response to Light Flicker With Functional Spectral Optical Coherence Tomography
Author Affiliations & Notes
  • T. Schmoll
    Center of Biomed. Eng. and Physics, Medical University of Vienna, Vienna, Austria
  • D. Van de Ville
    Radiology Department, University Hospital Geneva, Geneva, Austria
  • R. Michaely
    Laboratoire d'Optique Biomedicale, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland
  • R. A. Leitgeb
    Center of Biomed. Eng. and Physics, Medical University of Vienna, Vienna, Austria
    Laboratoire d'Optique Biomedicale, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland
  • Footnotes
    Commercial Relationships  T. Schmoll, None; D. Van de Ville, None; R. Michaely, None; R.A. Leitgeb, None.
  • Footnotes
    Support  FNS Grant 205321-109704/1; FP7-HEALTH-A Grant 201880 FUNOCT
Investigative Ophthalmology & Visual Science May 2008, Vol.49, 3198. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      T. Schmoll, D. Van de Ville, R. Michaely, R. A. Leitgeb; Wavelet Based Assessment of Photoreceptor Response to Light Flicker With Functional Spectral Optical Coherence Tomography. Invest. Ophthalmol. Vis. Sci. 2008;49(13):3198.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: : To extract human photoreceptor reflectance response signatures in-vivo using intensity modulated illumination from high-speed spectral optical coherence tomography (SOCT) B-scans applying advanced wavelet based signal-processing routines known from functional MRI.

Methods: : The retina of a healthy dilated volunteer is flickered para-foveally with a square pattern at 5Hz. The volunteer is dark adapted for 20min prior to measurement. SOCT B-scans at a rate of 16 per second are recorded across the square pattern at a fixed position on the retina. Such time series consist of 150 tomograms. The LED (630nm) illumination power at the cornea is 10µW. The SOCT source is centered at 850nm supporting an axial resolution of 8µm. Illumination and measurement light are separated via dichroic beam splitter. A wavelet based signal fit at 5Hz modulation was applied to each pixel of the B-scan time series. Such algorithms are well known from fMRI for their high sensitivity to reveal signal changes below 1%. A functional tomogram is obtained showing color-coded correlation coefficients between the measured signal and the expected 5Hz response. In addition, the wavelet analysis yields phase information that gives access to relative time processes within the retinal layers.

Results: : For proof-of-principle analysis, only parts of the B-scan time series are used that show least motion artifacts. The wavelet-based analysis shows highest correlation of the outer and inner photoreceptor segment boundary signal with the 5Hz excitation. The signature was only present within the stimulated area at the B-scan center. For negative control we took a tomogram series without stimulation at the same position. This time there is no significant correlation visible.

Conclusions: : We applied fMRI signal processing methods to in-vivo high speed SOCT B-scan time series to reveal reflectivity changes within the retina caused by photoreceptor response to modulated excitation. We found a significant change of signal within the photoreceptor layers, in particular along the inner-outer photoreceptor segment boundary. Due to the modulated excitation at fixed frequency we can offset the sensitivity of the method to suppress heart beat artifacts as well as motion artifacts such as backing of the volunteer from flash excitation. In addition we gain temporal phase sensitivity to follow signaling pathways.

Keywords: retina • photoreceptors • image processing 
×
×

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.

×