Purchase this article with an account.
J. Einighammer, T. Oltrup, T. Bende, B. Jean; Exact Raytracing Based on Geometry Data of Real Pseudophakic Eyes . Invest. Ophthalmol. Vis. Sci. 2006;47(13):305.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Exact raytracing based on snell’s law was used to predict the refraction of pseudophakic eyes. Therefore measured geometry of real eyes together with the known intraocular lens geometry were used. The outcome was compared to the postoperative refraction obtained by autorefractometry.
38 pseudophakic eyes were evaluated. The anteriour corneal surface was measured with Technomed C–Scan ®, the posteriour surface was approximated from the anteriour (both spline interpolated). The anteriour chamber depth (ACD) was not measurable with Zeiss IOLMaster ® and therefore estimated from axial length by a linear scaling approach. The geometry of the lens (Alcon AcrySof SA60AT ®) was used according to the manufacturer’s specification (anteriour and posteriour radius of curvature, thickness, refractive index). Axial length was measured with the IOLMaster. Refractive indices were assumed according to Gullstrand. Raytracing was performed using 2500 rays entering a 3.5 mm pupil. The predicted refraction (SPH, CYL, AX) is the outcome of an iterative multidimensional minimizing process of the Stiles–Crawford weighted RMS spot size as a function of the geometry of correcting spectacles. The predicted refraction together with the spherical equivalent refractive error (SEQ) was compared to the measured refraction obtained by a Canon RF–10 ® autorefractor.
The mean difference between predicted and measured refraction: SEQ 0.25 dpt ± 0.74 (SD); SPH 0.17 ± 0.76; CYL 0.17 ± 0.57; AX 26.95 ± 18.96. Pearson’s correlation coefficient: SEQ r=0.81, P<.01; SPH r=0.81, P<.01; CYL r=0.74, P<.01; AX r=0.89, P<.01. (AX was only compared for 22 eyes having a CYL>=0.5 dpt.) All variables show linear dependence and significant correlation. The predicted refraction shows a slight mean hyperopic shift in sphere and cylinder and noticeable standard deviation. Main error sources are measurement errors (cornea, axial length) and the estimated lens position.
It has been proofed possible to simulate the geometrical optical properties of the individual human eye by raytracing according to snell’s law. The exact geometry (total eye morphometry) is needed, including high resolution topographic data. The method should be further developed, using data from advanced measurement technics (e.g. including a correct IOL position). The method could also be useful in IOL calculations beyond paraxial optics, e.g. in handling post refractive surgery corneas, calculating aspheric, multifocal IOLs or considering pupil sizes in order to create an optimal individual IOL.
This PDF is available to Subscribers Only