A system for the automatic estimation of morphometric parameters of corneal endothelium in alizarine red-stained images

Background/aims A computer program for the automatic estimation of endothelium morphometric parameters (cell density, pleomorphism, polymegethism) in alizarine red-stained images is presented and evaluated. Methods Images of corneal endothelium from 30 porcine eyes stained with alizarine red were acquired with an optical microscope and saved as grey-level digital images. Each image was first pre-processed for luminosity correction and contrast enhancement. An artificial neural network was used to classify all pixels as cell contour or cell body pixels. The segmented cell contours were then used to obtain estimates of morphometric parameters. The central area was assessed and the mean area per cornea was 0.54+/-0.07 mm(2). The whole system was implemented as a computer program using the Matlab language. Estimated parameters were compared with the corresponding values derived from manual contour detection on the same images used for the automatic estimation. Results For the 30 images in our dataset, the mean differences for automatic versus manual parameters were -12+/-52 (range -103 to +145) cells/mm(2) for density, 0.5+/-2.6% (range -5.6 to +5.6%) for pleomorphism and -0.7+/-1.9% (range -4.1 to +2.8%) for polymegethism. Conclusion The evaluation of the automatic system on 30 images from porcine eyes confirmed its ability to estimate reliably morphometric parameters with respect to parameter values derived by manual analysis.