Abstract
Purpose: :
To develop and demonstrate the application of forward ray tracing in corneal topography, both in simulation and surface reconstruction. Previous techniques rely on backward ray tracing. Forward ray tracing will add the ability to determine expected images when reflection sources and surface types are known. This is important because recent developments in customized refractive surgery demand more accurate instrument performance of corneal topographers(Curr Opin Ophthalmol 2007; 18:325-333).
Methods: :
A forward ray tracing model is presented to determine the exact image projection in a general corneal topography system. This is implemented for several surface types (Rand-like surfaces, OVS 1997; 74(11):926-30) taking note that post refractive surgery corneas have high trefoil and quadrafoil features: sphere + trefoil feature , sphere + quadrafoil feature. Consequently, the skew ray error in Placido based topography can be demonstrated by looking into appropriate Zernike coefficients (evaluated at 7 mm corneal zone) corresponding to the surfaces. RMS of Zernike coefficients (until 8th order) derived from theoretical calculations, forward ray tracing (FRT) surface reconstruction and meridional ray tracing surface reconstruction are compared.
Results: :
The output of the FRT reconstruction and the theoretical calculation are practically the same with their minute difference dependent on the finite precision of the numerical processor. In this case the accuracy of FRT surface reconstruction is better than 10-10 µm in specifying Zernike coefficients. The error of the meridional ray tracing reconstruction algorithm in determining trefoil and quadrafoil aberration (RMS up to 8th order) is on average 61% and 77% respectively.
Conclusions: :
Forward ray tracing is an accurate technique that is useful for simulation and surface reconstruction of the anterior corneal surface. It is able to clearly demonstrate the effect of skew ray error in Placido based corneal topography.
Keywords: cornea: basic science • computational modeling