May 2007
Volume 48, Issue 13
ARVO Annual Meeting Abstract  |   May 2007
Computational Methods for OCT Corneal Strain Mapping
Author Affiliations & Notes
  • M. R. Ford
    Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio
  • W. J. Dupps, Jr.
    Cole Eye Institute, Cleveland Clinic, Cleveland, Ohio
  • A. M. Rollins
    Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio
  • Footnotes
    Commercial Relationships M.R. Ford, patent application, P; W.J. Dupps, patent application, P; A.M. Rollins, patent application, P.
  • Footnotes
    Support Research to Prevent Blindness, CWRU VSTP NIH Grant: T32 EY07157
Investigative Ophthalmology & Visual Science May 2007, Vol.48, 3874. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      M. R. Ford, W. J. Dupps, Jr., A. M. Rollins; Computational Methods for OCT Corneal Strain Mapping. Invest. Ophthalmol. Vis. Sci. 2007;48(13):3874.

      Download citation file:

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

  • Supplements

The viscoelastic properties of the cornea are important determinants of the corneal response to surgery and disease. The purpose of this work is to evaluate various techniques of measuring displacement with regards to accuracy, speed, and directional information acquired.


Displacement fields were measured in a human donor cornea that was mounted on an artificial anterior chamber and then subjected to controlled pressure increments. Mounted specimens were imaged by a high-speed FDOCT system and imaged in 3D under various IOP conditions. Algorithms were written to perform cross-correlation and phase measurements to track displacement. These routines were validated against phantoms, known displacements, and displacements with added noise. Thresholding was performed to remove noise and poorly correlated data from the final results.


All methods are able to accurately measure the displacement in phantoms and in tissue experiments. Increasing the amount of data scanned dimensionally increases the maximum measurable displacement. Increasing dimensionality incurs a non-linear increase in the data processing time. 1D and 2D algorithms require anywhere from several seconds to a few minutes to process depending on the window size chosen, whereas 3-D algorithms require a few hours to several days. Phase tracking algorithms function much more rapidly than 2D and 3D algorithms, but generate much noisier results with only axial information.


Software analysis of OCT feature flow in the corneal stroma can provide high-resolution information about the corneal response to an applied stress. The various techniques applied provide differing levels of detail regarding the corneal response but are limited by speed. Phase tracking algorithms have promise but are currently limited by noise problems limiting their current utility. 2D tracking algorithms provide the best tradeoff between accuracy, speed, and directional information acquired.Figure(1): A) OCT image of the cornea. B) A corneal displacement map in microns. C) A strain map of the cornea in percent.  

Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • anterior segment • cornea: clinical science 

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.