Considering 3D topography of endothelial folds to improve cell count of organ cultured corneas.

The posterior side of the cornea is covered by the endothelial monolayer, which governs corneal transparency but cannot proliferate. Determination of endothelial cell density (ECD) is therefore the minimal and mandatory quality control in all eye banks. It avoids primary graft failures caused by endothelial insufficiency, and allows allocation of corneas to surgical techniques requiring different numbers of endothelial cells (ECs). Corneas stored in organ culture (17% of grafts worldwide), are characterized by heavy stromal swelling and numerous deep endothelial folds, up to 200 µm high. During microscopic en face observation, flat surfaces are thus exceptional and EC counting is biased by parallax errors, resulting in overestimated eye bank ECD (ebECD). We used a motorized transmitted light microscope to acquire Z-stacks of images every 10 µm, and processed them to reconstruct the 3D surface of the folded endothelium. This method (3D-ECD) takes into account the local point-by-point slope in order to correct ECD. On a set of 30 corneas, we compared 3D-ECD and ebECD determined on five identical zones at the center of the cornea. 3D reconstruction allowed us to visualize twice as many cells, and ebECD was 8.1 ± 4.5% (95%CI 6.4-9.7) higher than 3D-ECD, with 1744 ± 488 versus 1606 ± 473 cells/mm2. 3D counting makes it possible to increase cell sampling and to correct overestimation by the conventional en face counting still routinely performed in eye banks.

Comparison and supervised learning of segmentation methods dedicated to specular microscope images of corneal endothelium.

The cornea is the front of the eye. Its inner cell layer, called the endothelium, is important because it is closely related to the light transparency of the cornea. An in vivo observation of this layer is performed by using specular microscopy to evaluate the health of the cells: a high spatial density will result in a good transparency. Thus, the main criterion required by ophthalmologists is the cell density of the cornea endothelium, mainly obtained by an image segmentation process. Different methods can perform the image segmentation of these cells, and the three most performing methods are studied here. The question for the ophthalmologists is how to choose the best algorithm and to obtain the best possible results with it. This paper presents a methodology to compare these algorithms together. Moreover, by the way of geometric dissimilarity criteria, the algorithms are tuned up, and the best parameter values are thus proposed to the expert ophthalmologists.

Inherent errors of the fixed-frame counting method for corneal endothelial cell density in eye banks.

The aim of this work was to analyze the magnitude of inherent errors associated with the fixed-frame counting method for corneal endothelial cell density (ECD) measurements. This technique is common among most eye banks worldwide. Three types of mosaics were used: regular and irregular tessellated mosaics (eight increasing densities ranging from 800 to 3,600 cells/mm(2) by steps of 400 cells/mm(2)) generated by a computer, and real mosaics (four specimens) obtained from human corneal endothelium flat mounted and stained with Alizarin red. On the three mosaics, the fixed-frame counting method was applied using a computer program. The ECD was calculated for 3,000 successive random positions from calibrated grids which area ranged from 50 × 50 to 300 × 300 µm(2) (incremental steps of 25 µm). For each grid, the ECD was expressed either as a single count, a mean of five or a mean of 10 measures. The fixed-frame count was constantly associated with an inherent variability but repeatability increased with larger grid size and ECD. The mean calculated out of 10 measures was the most reliable, but still, we noted ±5 % of residual variability from the real ECD. The 100 × 100 µm(2) grid manual counts, performed in many eye banks, should be abandoned and upgraded to at least 200 × 200 µm(2) grid counts. Digital image analysis with a variable frame counting method would be the best alternative.

Robust 3-D reconstruction of surfaces from image focus by local cross-sectional multivariate statistical analyses: application to human ex vivo corneal endotheliums.

The considered problem of 3-D reconstruction consists in computationally and passively recovering both topography and texture of a scene surface observed by optical sectioning with a limited depth-of-field imaging system (typically a conventional optical microscope). Throughout a sequence of registered 2-D images, the concepts of shape-from-focus and extended-depth-of-field involve recovering both topography (depth map) and texture image of the surface by researching in-focus information, respectively. Toward that aim, traditional approaches generally follow a 2-D sectional way and thereby fail to deal with noisy and disturbed acquisitions, quite frequent in transmitted light observations and of interest in this paper. Such examples are the acquisitions of human ex vivo corneal endotheliums from the medical issue addressed in this paper, which are mainly damaged by cellular fragments in the sample immersion medium and by emphasized contrast reversals. To achieve with such noisy and disturbed acquisitions, a new focus analysis is introduced that originally adopts a 3-D strategy throughout the image sequence. This method exploits simultaneously all available cross-sectional cues that effectively strengthens the robustness. More precisely, it locally performs multivariate statistical analyses over cross-sectional spatial windows so as to find sectional in-focus positions. Comparisons to state-of-the-art methods on both synthetic data and real acquisitions from the deal-with medical issue demonstrate the efficiency and the robustness of the proposed approach.

Endothelial morphometry by image analysis of corneas organ cultured at 31 degrees C.

PURPOSE: To determine the factors influencing endothelial morphometry by using image analysis of corneas stored in organ culture to determine the coefficient of variation (CV) in cell area and percentage of hexagonal cells. METHODS: The endothelia of 505 of the 559 corneas consecutively stored at the eye bank were routinely analyzed with Sambacornea image-analysis software (ver. 1.2.10; Tribvn, Châtillon, France) on three large-field images of 750 x 1000 microm, obtained after osmotic dilation of the intercellular spaces with 0.9% sodium chloride. Analysis was performed on at least 300 cells. The quality of the three-image set was graded poor, average, or good by an independent observer. The studied parameters were donor age and sex, lens status, storage time, and intrinsic quality of captured images. Statistics were analyzed by nonparametric tests. RESULTS: Image analysis was possible for 504 of the 505 assessed corneas. Donor age correlated significantly with endothelial cell density (ECD; r = -0.343), CV (r = 0.221), and hexagonality (r = -0.314; P < 0.001 for the three). Image quality significantly influenced these three parameters. ECD and hexagonality decreased parallel to image quality, whereas the CV increased. In the 258 corneas assessed twice (on average, at day [D] +4, then D +14) ECD, CV, and hexagonality decreased during storage. CONCLUSIONS: Despite the sometimes mediocre quality of the transmitted light microscopy images, endothelial parameters supplied by the analyzer were clinically reliable, since variations similar to those long known in specular microscopy were found. Endothelial morphometry (CV and hexagonality) is likely to provide further information on the endothelial function of the graft tissue, perhaps particularly for grafts of borderline ECD, close to the discard threshold.

Reproducibility of endothelial assessment during corneal organ culture: comparison of a computer-assisted analyzer with manual methods.

PURPOSE: To investigate the reproducibility of endothelial assessment of organ-cultured corneas with the computer-assisted Sambacornea analyzer in comparison with manual METHODS: methods. Seven observers of two eye banks determined the endothelial cell density (ECD) of 30 corneas through a grid overlay placed on endothelial photographs using two manual modes, unaided (naked-eye) and pointing (point-out). ECD was measured with the analyzer, first in automated mode, where analysis was completely machine determined, and then in touched-up mode, where the observer selected the analysis zone and corrected poorly drawn cell borders. Interobserver variability of ECD for the different methods was compared. Reproducibility of morphometry parameters was determined for the touched-up mode. RESULTS: Interobserver variability was +/-19.2% (95% confidence interval [CI], 13.0-25.4) and +/-17.6% (95% CI, 11.9-23.3) for the naked-eye and point-out mode, respectively, whereas the touched-up mode gave the least variability of +/-9.6% (95% CI, 6.5-12.7), confirmed by the highest intraclass correlation coefficient of 0.95 (95% CI, 0.91-0.97). Interobserver variability increased with worsening image quality. Manual modes underestimated ECD (naked-eye by a mean 10.7% [SD, 2.9%]; point-out by a mean 6.9% [SD, 2.3%]), whereas the automated mode overestimated ECD by a mean 14.7% (SD, 24.3%). Reproducibility of morphometric parameters by the touched-up mode was acceptable but was influenced by endothelial pleomorphism. CONCLUSIONS: Manual counting shows systematic underestimation of ECD with high interobserver variability. The analyzer in automated mode overestimates ECD and is absolutely unreliable. Detection of cell contours by the specific algorithm, combined with manual correction by a skilled technician, appears to be the most reliable method of ECD and morphometry determination.

Screening for diabetic retinopathy by ophthalmologists and endocrinologists with pupillary dilation and a nonmydriatic digital camera.

PURPOSE: To assess the effects of pupillary dilation on image quality and certitude of screening diabetic retinopathy by trained endocrinologists using a digital nonmydriatic camera. DESIGN: Prospective, comparative, observational case series. METHODS: One hundred fifty patients with diabetes attending a hospital-based photographic screening clinic had five-field (45 degree) digital retinal imaging and mosaic construction, first through dark-adapted pupils, then after a single application of tropicamide 1%. The 600 sets of images (150 patients, 300 eyes, before and after dilation) were graded independently and in a blinded manner by two endocrinologists and two ophthalmologists, with the consensual opinion on dilated images of the latter serving as the gold standard. The criteria of evaluation were assessment of image quality and certitude of detecting and grading retinopathy. RESULTS: Of 300 eyes, pharmacological mydriasis improved image quality, with an increase in the number of eyes with five good images from seven to 160 and good-quality mosaics from 99 to 233. Dilation allowed better identification of maculopathy in the second eye because there was a decrease in ungradeable central images from 127 to 15 eyes. Dilation increased the number of eyes having retinopathy detected with certainty from 153 to 252 and graded with certainty from 173 to 277. No adverse effects were recorded. CONCLUSIONS: Single application of tropicamide 1% significantly improves image quality and certitude of screening diabetic retinopathy with nonmydriatic camera by endocrinologists.