April 2010
Volume 51, Issue 13
Free
ARVO Annual Meeting Abstract  |   April 2010
Improvements in High Resolution Imaging With Adaptive Optics Scanning Laser Ophthalmoscopy
Author Affiliations & Notes
  • N. M. Putnam
    School of Optometry and Vision Science Graduate Group,
    University of California, Berkeley, Berkeley, California
  • K. Y. Li
    School of Optometry and Vision Science Graduate Group,
    University of California, Berkeley, Berkeley, California
  • G. Kumar
    College of Optometry, University of Houston, Houston, Texas
  • S. B. Stevenson
    College of Optometry, University of Houston, Houston, Texas
  • F. Marchis
    Dept of Astronomy,
    University of California, Berkeley, Berkeley, California
  • A. Roorda
    School of Optometry and Vision Science Graduate Group,
    University of California, Berkeley, Berkeley, California
  • Footnotes
    Commercial Relationships  N.M. Putnam, None; K.Y. Li, None; G. Kumar, None; S.B. Stevenson, None; F. Marchis, None; A. Roorda, US Patent # 7,118,216, P.
  • Footnotes
    Support  NIH EY014375, NSF AST9876783, AOF Ezell Fellowship
Investigative Ophthalmology & Visual Science April 2010, Vol.51, 2310. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      N. M. Putnam, K. Y. Li, G. Kumar, S. B. Stevenson, F. Marchis, A. Roorda; Improvements in High Resolution Imaging With Adaptive Optics Scanning Laser Ophthalmoscopy. Invest. Ophthalmol. Vis. Sci. 2010;51(13):2310.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: : The use of adaptive optics in a scanning laser ophthalmoscope (AOSLO) offers many benefits, but the presence of noise, coherence artifacts and motion artifacts affect the radiometric precision and the highest achievable resolution of the system. Here we present a series of efforts to overcome these limits.

Methods: : The improvements are applied in three areas: image collection, image processing and post-processing. Image collection: We use statistically weighted wavefront reconstruction, which penalizes improbable wavefront modes (ie local waffling) and nonuniform subaperture illumination providing higher fidelity and more robust operation. We set the pixel sampling density to be appropriate for the structures that we are trying to resolve. Finally, we employ schemes that use optical stimulation to scramble the phase relationships between retinal cells, so that average images are effectively incoherent. Image processing: We employ FFT-based correction of intraframe image distortions that are caused by continuous eye motion. Post-processing: High fidelity images generated in the first two steps lend themselves to deconvolution. We employ AIDA (adaptive image deconvolution algorithm, http://msg.ucsf.edu/AIDA/) as the final step. The algorithm works with single or multiple frames and employs a myopic deconvolution scheme (ie aided by good estimates of the point spread function and the residual error of the motion correction algorithms). We implemented this sequence of improvements in an effort to resolve foveolar cones.

Results: : More robust wavefront control permits imaging for longer periods of time at high Strehl ratios, even without dilation and cycloplegia. The quality of the correction is higher as measured by a relative increase in the amount of light collected through the confocal aperture. The improvements in quality and uniformity permit video stabilization to sub-cellular levels over long video sequences. Finally, the deconvolution reveals more structure in the image, while preserving its radiometric quantities. These combined improvements have allowed us to resolve the highest density foveolar cones in healthy, normal eyes.

Conclusions: : Implementation of improvements in every step of the AOSLO imaging process yields measurable improvements in image quality. Routine imaging of the smallest retinal features, such as foveal cones, and detecting features with intrinsically low contrast will be enabled through these steps.

Keywords: imaging/image analysis: non-clinical • image processing • retina 
×
×

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.

×