April 2011
Volume 52, Issue 14
Free
ARVO Annual Meeting Abstract  |   April 2011
Keratoconus Screening Based On Broader Applications Of CLMI Algorithm
Author Affiliations & Notes
  • Jennifer R. Lewis
    Ophthalmology, Biomed Engineering, The Ohio State University, Columbus, Ohio
  • Beatrice E. Frueh
    Ophthalmology, Univ of Bern Inselspital, Bern, Switzerland
  • Christoph Tappeiner
    Ophthalmology, Univ of Bern Inselspital, Bern, Switzerland
  • Ashraf M. Mahmoud
    Ophthalmology, Biomed Engineering, The Ohio State University, Columbus, Ohio
  • Cynthia J. Roberts
    Ophthalmology, Biomed Engineering, The Ohio State University, Columbus, Ohio
  • Footnotes
    Commercial Relationships  Jennifer R. Lewis, Ziemer (E); Beatrice E. Frueh, None; Christoph Tappeiner, None; Ashraf M. Mahmoud, Ziemer (P); Cynthia J. Roberts, Ziemer (C)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science April 2011, Vol.52, 5167. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jennifer R. Lewis, Beatrice E. Frueh, Christoph Tappeiner, Ashraf M. Mahmoud, Cynthia J. Roberts; Keratoconus Screening Based On Broader Applications Of CLMI Algorithm. Invest. Ophthalmol. Vis. Sci. 2011;52(14):5167.

      Download citation file:


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

      ×
  • Supplements
Abstract
 
Purpose:
 

To test the ability of the CLMI algorithm (Cone Location and Magnitude Index) applied to other maps, in order to separate keratoconic and normal subject data based on data from dual Scheimpflug tomographic data.

 
Methods:
 

The CLMI algorithm was applied to anterior and posterior curvature and elevation maps to find the steepest and highest values, respectively, as well as a modified version (FLMI) on the pachymetry map to find the thinnest value. All data were acquired from a dual Scheimpflug-Placido based tomographer (GALILEI, Ziemer Group) in a retrospective analysis. Data from two distinct databases containing the right eye data of normal (N=67) and keratoconic (N=41) subjects were exported for analysis using custom software to apply the CLMI algorithm to multiple map types, as described. The keratoconic data were identified prior to imaging by slit-lamp clinical diagnosis. Stepwise logistic regressions (SAS) were used to determine the strongest predictor(s) of separation between the two groups. Additionally, t-test analysis was used to determine which parameters distinguished the two groups with a significance of P<0.05.

 
Results:
 

Complete separation of the two groups was obtained using the CLMI algorithm on the posterior axial map. Multiple other parameters were significantly different (p<0.0001) between the two groups, but with some overlap in the distributions. These include CLMI on anterior axial and both anterior and posterior tangential, as well as the average magnitude of the data within the circle located by the routine for anterior and posterior axial and tangential. Finally, FLMI on the pachymetry map, as well as the magnitude of the thinnest spot were also significantly different between groups (p<0.0001).

 
Conclusions:
 

The CLMI algorithm based on posterior curvature dual Scheimpflug-Placido data has been shown to provide complete separation between normal and keratoconic corneas with 100% sensitivity and 100% specificity. It is likely that posterior axial screening will be more robust than other posterior elevation approaches, since elevation can change with reference sphere and data coverage.

 
Keywords: refractive surgery: corneal topography • keratoconus • imaging/image analysis: 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.

×