Lesions seem to be smaller than normal superficial epithelial cells (which are approximately 25 × 50 μm) and might correspond to the staining of dying shrunken cells, according to recent investigations. These new quantitative data will help in developing automated recognition algorithms to obtain reliable objective classification of corneal staining.
Purpose:
Lissamine green (LG) is often used in addition to fluorescein to assess the severity of conjunctival damage in dry eye syndrome, which is graded manually. Our purpose was to describe an algorithm designed for image analysis of LG conjunctival staining.
Methods:
Twenty pictures of patients suffering from dry eye with visible LG conjunctival staining were selected. The images were taken by 2 different digital slit lamps with a white light source and a red filter transmitting over the wavelengths absorbed by LG. Conjunctival staining appeared in black on a red background. The red channel was extracted from the original image. Stained areas were then detected using a Laplacian of Gaussian filter and applying a threshold whose value was determined manually on a subset of images. The same algorithm parameters remained constant thereafter. LG-stained areas were also drawn manually by 2 experts as a reference.
Results:
The delineation obtained by the algorithm closely matched the actual contours of the punctate dots. In 19 cases of 20 (95%), the algorithm found the same Oxford grade as the experts, even for confluent staining that was detected as a multitude of dots by the algorithm but not by the experts, resulting in a high overestimation of the total number of dots (without mismatching the Oxford grade estimated by the experts). The results were similar for the 2 slit-lamp imaging systems.
Conclusions:
This efficient new image-analysis algorithm yields results consistent with subjective grading and may offer advantages of automation and scalability in clinical trials.
Purpose Measurement of the endothelial cell density (ECD) is the main criterion to validate the quality of corneal grafts in eye banks. Because of corneal curvature and of deep storage‐induced endothelial folds, parallax errors may result in ECD overestimation. We designed a software with 3D reconstruction and optimized segmentation methods to improve cell count reliability
Methods After acquisition of endothelial Z stacks by conventional motorized bright field microscopy, a depth map was obtained by researching maximized measurements of the focus in images (shape‐from‐focus method). Texture of the surface was built by summing image parts presenting the right focus leadind to an all‐in‐focus image. By calculating first derivative in all directions of depth map, precise estimation of cells in folds surfaces could be found. In order to improve and facilitate the cell count, a segmentation method based on watershed technique was applied on the texture image. Taking account of the new 3D area, a precise corrected ECD (cECD) could be calculated. We validated the cECD by comparison with ECD obtained manually by an expert. Two types of images were used: the Keratotest (cells boundaries engraved on a quartz wafer) and real endothelial images. Two parameters were compared: quality of cells borders recognition (correct shapes) and ECD.
Results Comparison results showed the relevance of these algorithms for both criteria. Automation of processing allowed gain of time compared to manual estimation
Conclusion 3D endothelial count can be useful to improve corneas selection in eye banks. Grant: ANR2012 CORIMMO3D
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.