The significance of using different methods for analysing photokeratoscopic data

With the photokeratoscope of Wesley-Jessen, 45 pictures were taken from the right eye of 18 normal persons and analysed by 2 different methods. The first method is based on the assumption that radius of the corneal curvature is constant between 2 points of reflection, the second method is without such a preliminary assumption on the corneal contour. The parameters estimated: radius of the central curvature (K) and the coefficient of radius variation (RV) are compared. A difference in K at 1.5% corresponding in case to 0.6 diopters and a difference in RV at 46% are demonstrated. The differences are correlated to the error introduced by the assumption on constant radius. Further it is established that K measured by the keratometer of Javal-Schiøtz is about 0.14 mm larger (in case 0.7 diopters) than K estimated by the method of photokeratoscopic data analysis without preliminary assumption.