April 2009
Volume 50, Issue 13
ARVO Annual Meeting Abstract  |   April 2009
Improve Adaptive Optics Wavefront Control Accuracy for Retinal Imaging With Dual-influence-matrix Method
Author Affiliations & Notes
  • W. Zou
    School of Optometry, Indiana University, Bloomington, Indiana
  • S. A. Burns
    School of Optometry, Indiana University, Bloomington, Indiana
  • Footnotes
    Commercial Relationships  W. Zou, None; S.A. Burns, None.
  • Footnotes
    Support  NIH EYO4395
Investigative Ophthalmology & Visual Science April 2009, Vol.50, 1052. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      W. Zou, S. A. Burns; Improve Adaptive Optics Wavefront Control Accuracy for Retinal Imaging With Dual-influence-matrix Method. Invest. Ophthalmol. Vis. Sci. 2009;50(13):1052.

      Download citation file:

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

  • Supplements

Purpose: : We present a technique for improving wavefront control of adaptive optics (AO) systems: the imaging aberrations are corrected first with a generic influence matrix and then with an influence matrix recalibrated in the presence of overall static errors.

Methods: : We have shown that the design of a control system for an Adaptive Optics Scanning Laser Ophthalmoscope (AOSLO) with a Woofer-Tweeter deformable mirrors (DMs) can produce both a wide range of corrections and a high precision of correction by implementing a novel control algorithm that simultaneously optimizes both mirrors (for our design, a Boston Micromachines DM, and an Imagine Optic Mirao DM). Even with the Woofer-Tweeter approach, however, the AO wavefront control is still limited by the nonlinearity of the DM influence functions and shows fluctuations even after convergence. The fluctuations arise because there is a nonlinear relation between the mirror deflection and the actuator voltage. Typically the influence function is calibrated when the DM deflection is zero. Thus, when the AO correction is working and there is large amplitude of wavefront aberrations, the wavefront control can oscillate slightly due to the inaccurate influence matrix. Using the typical approach, we find that the residual wavefront RMS error can be controlled around 0.08-0.15 µm, which is similar to other reported results. In adaptive optics, the accuracy of wavefront control depends in part on how accurate we can calibrate the actuator influence function and thereby the DM influence matrix. We therefore implemented a two step process. Step One, the DM influence matrix is calibrated with the DM deflection close to zero. With this zero-deflection influence matrix, wavefront aberrations are corrected by the AO control process, and the actuator stroke values will be recoded when the residual wavefront error reach its minimum; Step Two, the DM influence matrix is recalibrated with the DM deflection at the recorded actuator stroke values obtained in Step One, the wavefront aberration will be corrected with the new influence matrix.

Results: : Calibration of the Woofer DM influence matrix showed a systematic, linear relation to the defocus being corrected. By adjusting the influence matrix for the applied aberrations, residual wavefront RMS error decreased to around 0.026-0.051µm and the amplitude of fluctuations after convergence decreased remarkably.

Conclusions: : A variable-influence-matrix methodology for improving wavefront control for AO retinal imaging is proposed and proved to be effective. This approach can also be generalized to any AO system where there is a large static contribution to the wavefront error.

Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • aberrations • imaging/image analysis: non-clinical 

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.