Quality Control in the Corneal Bank with Artificial Intelligence: Comparison of a New Deep Learning-based Approach with Conventional Endothelial Cell Counting by the "Rhine-Tec Endothelial Analysis System". / Qualitätskontrolle in der Hornhautbank mit künstlicher Intelligenz: Vergleich eines neuen Deep-Learning-basierten Ansatzes mit der konventionellen Endothelzelldichtenbestimmung durch das "Rhine-Tec Endothelial Analysis System".

Endothelial cell density (ECD) is a crucial parameter for the release of corneal grafts for transplantation. The Lions Eye Bank of Baden-Württemberg uses the "Rhine-Tec Endothelial Analysis System" for ECD quantification, which is based on a fixed counting frame method considering only a small sample of 15 to 40 endothelial cells. The measurement result therefore depends on the frame placement and manual correction of the cells counted within the frame. To increase the sample size and create higher objectivity, we developed a new method based on "deep learning" that automatically detects all visible endothelial cells in the image. This study aims to compare this new method with the conventional Rhine-Tec system. 9375 archived phase-contrast microscopic images of consecutive grafts from the Lions Eye Bank were evaluated with the deep learning method and compared with the corresponding archived analyses of the Rhine-Tec system. Means, Bland-Altman and correlation analyses were compared. Comparable results were obtained for both methods. The mean difference between the Rhine-Tec system and the deep learning method was only - 23 cells/mm2 (95% confidence interval - 29 to - 17). There was a statistically significant positive correlation between the two methods, with a correlation coefficient of 0.748. What was striking in the Bland-Altman analysis were clustered deviations in the cell density range between 2000 and 2500 cells/mm2 - with higher values in the Rhine-Tec system. The comparable results for cell density measurement values underline the validity of the deep learning-based method. The deviations around the formal threshold for graft release of 2000 cells/mm2 are most likely explained by the higher objectivity of the deep learning method and the fact that measurement frames and manual corrections were specifically selected to reach the formal threshold of 2000 cells/mm2 when the full area endothelial quality was good. This full area assessment of the graft endothelium cannot currently be replaced by deep learning methods and remains the most important basis for graft release for keratoplasty.

Influence of Organ Culture on the Characteristics of the Human Limbal Stem Cell Niche.

Organ culture storage techniques for corneoscleral limbal (CSL) tissue have improved the quality of corneas for transplantation and allow for longer storage times. Cultured limbal tissue has been used for stem cell transplantation to treat limbal stem cell deficiency (LSCD) as well as for research purposes to assess homeostasis mechanisms in the limbal stem cell niche. However, the effects of organ culture storage conditions on the quality of limbal niche components are less well described. Therefore, in this study, the morphological and immunohistochemical characteristics of organ-cultured limbal tissue are investigated and compared to fresh limbal tissues by means of light and electron microscopy. Organ-cultured limbal tissues showed signs of deterioration, such as edema, less pronounced basement membranes, and loss of the most superficial layers of the epithelium. In comparison to the fresh limbal epithelium, organ-cultured limbal epithelium showed signs of ongoing proliferative activity (more Ki-67+ cells) and exhibited an altered limbal epithelial phenotype with a loss of N-cadherin and desmoglein expression as well as a lack of precise staining patterns for cytokeratin ((CK)14, CK17/19, CK15). The analyzed extracellular matrix composition was mainly intact (collagen IV, fibronectin, laminin chains) except for Tenascin-C, whose expression was increased in organ-cultured limbal tissue. Nonetheless, the expression patterns of cell-matrix adhesion proteins varied in organ-cultured limbal tissue compared to fresh limbal tissue. A decrease in the number of melanocytes (Melan-A+ cells) and Langerhans cells (HLA-DR+, CD1a+, CD18+) was observed in the organ-cultured limbal tissue. The organ culture-induced alterations of the limbal epithelial stem cell niche might hamper its use in the treatment of LSCD as well as in research studies. In contrast, reduced numbers of donor-derived Langerhans cells seem associated with better clinical outcomes. However, there is a need to consider the preferential use of fresh CSL for limbal transplants and to look at ways of improving the limbal stem cell properties of stored CSL tissue.

[Conjunctival tumors in children : Histopathological diagnoses in 262 cases]. / Bindehauttumoren bei Kindern : Histopathologische Diagnosen in 262 Fällen.

BACKGROUND: The spectrum of conjunctival tumors in children and adolescents significantly differs from that of adults. OBJECTIVE: The aim of this study was to show the distribution of histopathological diagnoses of pediatric conjunctival tumors. MATERIAL AND METHODS: Retrospective case series including all conjunctival tumors of children and adolescents under 18 years of age examined histologically over a period of 25 years (1990-2015) at the Eye Center in Freiburg, Germany. RESULTS: Among the 262 conjunctival specimens, the most common diagnoses were conjunctival nevi (148, [56%]), most often with junctional activity (106/148, [72%]), pyogenic granulomas (26, [10%]), and dermolipomas (18, [7%]). In one case a malignant tumor (conjunctival melanoma) was detected (1, < 1%). CONCLUSION: The results suggest that pediatric conjunctival tumors are benign in most cases. Most of them were of melanocytic origin, with a junctional conjunctival nevus being the most common diagnosis.

Histological Diagnosis of Ocular and Periocular Tuberculosis 1945 - 2020. / Histologisch diagnostizierte okuläre und periokuläre Tuberkulose von 1945 bis 2020.

BACKGROUND: Ocular tuberculosis is a rare but important differential diagnosis for inflammatory conditions of all eye tissues, including the ocular surface and adnexa. Tissue diagnostics may prove challenging as some ocular tissues are difficult to biopsy and the detection of pathogens may be insensitive. We were interested in how many cases in the archive of the ophthalmopathological laboratory had been diagnosed with (peri)ocular tuberculosis since 1945. MATERIALS AND METHODS: Retrospective analysis of historical records and specimens of the ophthalmopathology laboratory of the eye department at Freiburg university hospital. Systematic re-evaluation of available slides for presence of granuloma, necrosis, giant cells, acid fast bacteria, and chronic as well as acute inflammation, plus comparison of current and historic evaluations. In addition, we describe a recent case with tuberculoma of the iris. RESULTS: There were 50,418 records archived since 1945, of which 23 specimens taken from 22 patients had been diagnosed as (peri)ocular tuberculosis. Of these, 22 (96%) were archived and available for re-interpretation. Four specimens (17%) had been excised from children. The most common tissues were enucleated eye globes (10/23, 44%), followed by the lacrimal sac (5/23, 22%) and conjunctiva (2/23, 9%). The most frequent histopathological findings were granulomas (23/23, 100%), chronic inflammation (22/23, 96%), giant cells (21/23, 91%), and necrosis (14/23, 61%). An acute inflammatory response was found in 4/23 specimens (17%). Ziehl-Neelsen stains for acid-fast bacteria had been performed in five cases, of which three were positive (60%). The greatest discrepancy between current and historical findings related to the presence of necrosis (59% consensus). In other findings, the consensus was high (78 - 96%). In a recent case of a patient with wasting syndrome attributed to lymphoma, histopathological workup of an iris tumour led to the diagnosis of tuberculosis. CONCLUSION: Ocular tuberculosis is a rare but important histopathological differential diagnosis. In the available specimens, the classic finding of necrotizing inflammation was rarest and showed least consensus on histological re-evaluation. Other typical findings, such as giant cells and a predominantly lymphocytic infiltrate, are sometimes not found even with proven presence of Mycobacterium tuberculosis. They should not be considered essential in cases where there is strong clinical suspicion.