1. Algorithm For Automatic Analysis Of Corneal Endothelial Images
- Author
-
Bling-Nan Hung, Joseph Jy-Haw Yu, Jenn-Lung Su, and Ming-Mu Hsieh
- Subjects
Corneal endothelium ,Microscope ,genetic structures ,Endothelium ,Computer science ,business.industry ,Matched filter ,media_common.quotation_subject ,Low-pass filter ,Image segmentation ,eye diseases ,law.invention ,Background noise ,medicine.anatomical_structure ,Optical microscope ,law ,medicine ,Contrast (vision) ,Computer vision ,sense organs ,Specular reflection ,Artificial intelligence ,business ,Algorithm ,media_common - Abstract
The images of corneal endothelium cells obtained from specular microscopes are blurred and obscure. We use adaptive local process and lowpass filter to enhance the contrast of the images and to reduce the background noise. A special matched filter is designed to detect the edges of cell borders. Then by applying thinning technique and line tracing rules, the borders of endothelium cells can be correctly identified. This algorithm provides the quantitative informations of cell size, density, and shape which are concerning the general status, viability, and functional reserve of this important cell layer.
- Published
- 2005
- Full Text
- View/download PDF