Purchase this article with an account.
M.D. Karon, S.D. Klyce, M.K. Smolek; Device–Independent Statistical Indexes in Corneal Topography . Invest. Ophthalmol. Vis. Sci. 2004;45(13):2874.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Comparison of exams obtained from different corneal topographers is difficult owing to variations in measurement and calculations. In order to overcome this limitation, we have developed a tool that converts topography data to a common format and common spatial resolution. Methods: Using a 2–D Fast Fourier Transform (FFT), a low pass Butterworth filter was developed and tested with corneal topography exams from the NIDEK OPDScan, the NIDEK Magellan, the Tomey TMS, and the Humphrey Atlas corneal topographers. The bandwidth cutoff was determined through correlations of Wavefront RMS error of paired exams from the fine mire Magellan and wide mire OPDScan, which represent the extremes of target spatial resolution among current corneal topographers. The resulting filtered data were mapped onto a standard, uniform ring structure. Eighteen corneal topography indexes similar to those previously published were calculated for each exam and linear correlation analysis was used for evaluation. For consistency with previous published studies, the statistics were scaled where necessary to those previously established with the Tomey TMS topographer. Results: The figure displays the average RMS of these exams at various filter cutoff frequencies and shows that 0–5 rad/pixel (1 pixel = 58.6 µm) would be the optimal frequency that produces a common spatial resolution between topographers. Linear correlations of topography indexes between devices had an average correlation coefficient of 0.84. Conclusions: A method has been developed that will produce a set of corneal topography statistical indexes that are quantitatively similar among corneal topographers including both wide mire (OPDScan) and fine mire (Atlas) large Placido disk corneal topographers, as well as fine mire, small Placido cone corneal topographers (TMS and Magellan). This approach paves the way for the development of device–independent automatic classification of corneal topography to assist in clinical interpretation. View OriginalDownload SlideView OriginalDownload Slide
This PDF is available to Subscribers Only