April 2011
Volume 52, Issue 14
Free
ARVO Annual Meeting Abstract  |   April 2011
Rapid Cone AOSLO Imaging And Analysis
Author Affiliations & Notes
  • Stephen A. Burns
    School of Optometry, Indiana University, Bloomington, Indiana
  • Weiyao Zou
    School of Optometry, Indiana University, Bloomington, Indiana
  • Xiaofeng Qi
    School of Optometry, Indiana University, Bloomington, Indiana
  • Zhangyi Zhong
    School of Optometry, Indiana Univ Bloomington, Bloomington, Indiana
  • Gang Huang
    School of Optometry, Indiana University, Bloomington, Indiana
  • Footnotes
    Commercial Relationships  Stephen A. Burns, None; Weiyao Zou, None; Xiaofeng Qi, None; Zhangyi Zhong, None; Gang Huang, None
  • Footnotes
    Support  NIH Grants EY04395, EY14375, P30EY019008.
Investigative Ophthalmology & Visual Science April 2011, Vol.52, 3195. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Stephen A. Burns, Weiyao Zou, Xiaofeng Qi, Zhangyi Zhong, Gang Huang; Rapid Cone AOSLO Imaging And Analysis. Invest. Ophthalmol. Vis. Sci. 2011;52(14):3195.

      Download citation file:


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

      ×
  • Supplements
Abstract
 
Purpose:
 

To develop AOSLO techniques for obtaining and analyzing photorecptor images. In the current work we test a new algorithm which while less specific than complete counting rapidly estimates cone spacing properties and also generates a lower bound estimate of AO control accuracy.

 
Methods:
 

Images are acquired with the Indiana dual-DM AOSLO which includes computer controlled steering mirrors that can place the AO within a 30 degree field of view. AOSLO imaging is initiated while the subject is fixating, and approximately 2 seconds of data is collected at 30 frames per second. The AO imaging field is then displaced according to a pre-programmed pattern, and another image acquisition is initiated. Locations and system parameters are saved to a database. Off-line, images for each location are averaged and stitched. The composite image is then analyzed using a sweeping window of 200 microns. Within each window, the brightest pixels are automatically selected. These pixels typically represent bright cones. A 25-50 micron region around each bright region is then extracted, and the subregions averaged using a shift-add approach, providing an image of an average bright cone.

 
Results:
 

Cone image mosaics along all four principal meridia from the fovea to 10 degrees within a one hour session. Image analysis produces excellent estimates of the hexagonal structure of the cone array. Figure 1 shows results from 2, 1.4, 0.8 and 0 degrees (averaged within 120 micron wide regions). The width of the central island represents a convolution of the point spread function of the AO system with the average cone within that region. At the fovea the cones are small placing an upper bound on the width of the psf. In the example the half-width was 2.6 microns. The nearest neighbors and second nearest

 
Conclusions:
 

The automatic estimation of average cone neighborhoods provides a quick estimates of the normality of cone packing in subjects. While, more detailed analysis techniques will be needed to examine local pathological changes the new approach allows rapid evaluation of the retina and AO control.  

 
Keywords: imaging/image analysis: non-clinical • 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.

×