Fourier transform method to determine human corneal endothelial morphology

Authors:

Abstract
The statistical evaluation of the size, shape, density and regularity of the cells in the human corneal endothelium is an important diagnostic technique. A method based on the Fourier transform of the cell boundaries was developed which can yield these statistical properties. The development of a hybrid optical/digital technique to obtain these statistical perimeters is our goal. The input images were tracings of human endothelial cell patterns. The optical Fourier transform of each image was obtained, and the radial projection and the angular correlation function were plotted versus distance and angle respectively. The average size of the cells was obtained from the first peak of the radial projection. The width of this peak is related to the coefficient of variation of the average cell size. The separation of the peaks in the normalized angular correlation plot is related to cell shape. This method is suitable for rapid analysis of large numbers of endothelial cell images. This technique may have potential for diagnostic ophthalmology.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.