Aims/hypothesis: Corneal confocal microscopy is a rapid non-invasive ophthalmic imaging technique that identifies peripheral and central neurodegenerative disease. Quantification of corneal sub-basal nerve plexus morphology, however, requires either time-consuming manual annotation or a less-sensitive automated image analysis approach. We aimed to develop and validate an artificial intelligence-based, deep learning algorithm for the quantification of nerve fibre properties relevant to the diagnosis of diabetic neuropathy and to compare it with a validated automated analysis program, ACCMetrics.Methods: Our deep learning algorithm, which employs a convolutional neural network with data augmentation, was developed for the automated quantification of the corneal sub-basal nerve plexus for the diagnosis of diabetic neuropathy. The algorithm was trained using a high-end graphics processor unit on 1698 corneal confocal microscopy images; for external validation, it was further tested on 2137 images. The algorithm was developed to identify total nerve fibre length, branch points, tail points, number and length of nerve segments, and fractal numbers. Sensitivity analyses were undertaken to determine the AUC for ACCMetrics and our algorithm for the diagnosis of diabetic neuropathy.Results: The intraclass correlation coefficients for our algorithm were superior to those for ACCMetrics for total corneal nerve fibre length (0.933 vs 0.825), mean length per segment (0.656 vs 0.325), number of branch points (0.891 vs 0.570), number of tail points (0.623 vs 0.257), number of nerve segments (0.878 vs 0.504) and fractals (0.927 vs 0.758). In addition, our proposed algorithm achieved an AUC of 0.83, specificity of 0.87 and sensitivity of 0.68 for the classification of participants without (n= 90) and with (n= 132) neuropathy (defined by the Toronto criteria).Conclusions/interpretation: These results demonstrated that our deep learning algorithm provides rapid and excellent localisation performance for the quantification of corneal nerve biomarkers. This model has potential for adoption into clinical screening programmes for diabetic neuropathy.Data availability: The publicly shared cornea nerve dataset (dataset 1) is available athttp://bioimlab.dei.unipd.it/Corneal%20Nerve%20Tortuosity%20Data%20Set.htmandhttp://bioimlab.dei.unipd.it/Corneal%20Nerve%20Data%20Set.htm.

During infectious keratitis, the production of collagenolytic and inflammatory substances, along with increased corneal matrix metalloproteinase (MMP) activity, induces the degradation of corneal collagen and may cause postkeratitis complications, such as opacity, thinning, and corneal perforation. MMPs, especially MMP-2 and MMP-9, are overexpressed in infectious keratitis and sustained over time by inflammatory and nonmicrobial mechanisms. The high MMP levels are correlated with excessive corneal destruction in bacterial, herpetic, fungal, and acanthamoeba infections. Nonspecific treatments, such as tetracyclines, particularly doxycycline, or corticosteroids, are used as adjuvants to antimicrobials to alleviate the disproportionate degradation and inflammation of the corneal layers caused by corneal MMPs and decrease the recruitment and infiltration of inflammatory cells. Treatments showing inhibition of specific MMPs (Galardin, ZHAWOC7726), interfering with pro-MMP activation (EDTA, ascorbic acid), or showing anticytokine effect (epigallocatechin-2-gallate, TRAM-34) have been reported. Other treatments show a direct action over corneal collagen structure such as corneal cross-linking or have been associated with reduction of MMP levels such as amniotic membrane grafting. Although the use of these drugs has been shown in studies to be effective in controlling inflammation, especially in experimental ones, robust studies are still needed based on randomized and randomized clinical trials to demonstrate their potential effect as adjuvants in the management of infectious keratitis.

Less than 1% of all microorganisms of the available environmental microbiota can be cultured with the currently available techniques. Metagenomics is a new methodology of high-throughput DNA sequencing, able to provide taxonomic and functional profiles of microbial communities without the necessity to culture microbes in the laboratory. Metagenomics opens to a ‘hypothesis-free’ approach, giving important details for future research and treatment of ocular diseases in ophthalmology, such as ocular infection and ocular surface diseases.

Objective To evaluate the accuracy of convolutional neural networks technique (CNN) in detecting keratoconus using colour-coded corneal maps obtained by a Scheimpflug camera. Design Multicentre retrospective study. Methods and analysis We included the images of keratoconic and healthy volunteers’ eyes provided by three centres: Royal Liverpool University Hospital (Liverpool, UK), Sedaghat Eye Clinic (Mashhad, Iran) and The New Zealand National Eye Center (New Zealand). Corneal tomography scans were used to train and test CNN models, which included healthy controls. Keratoconic scans were classified according to the Amsler-Krumeich classification. Keratoconic scans from Iran were used as an independent testing set. Four maps were considered for each scan: axial map, anterior and posterior elevation map, and pachymetry map. Results A CNN model detected keratoconus versus health eyes with an accuracy of 0.9785 on the testing set, considering all four maps concatenated. Considering each map independently, the accuracy was 0.9283 for axial map, 0.9642 for thickness map, 0.9642 for the front elevation map and 0.9749 for the back elevation map. The accuracy of models in recognising between healthy controls and stage 1 was 0.90, between stages 1 and 2 was 0.9032, and between stages 2 and 3 was 0.8537 using the concatenated map. Conclusion CNN provides excellent detection performance for keratoconus and accurately grades different severities of disease using the colour-coded maps obtained by the Scheimpflug camera. CNN has the potential to be further developed, validated and adopted for screening and management of keratoconus.

Purpose: To share the experience and provide a standardized protocol for Descemet membrane endothelial keratoplasty (DMEK) graft preparation. Methods: A retrospective study based on 527 prestripped DMEK tissues that were prepared between 2014 and 2017. The experience of using different instruments and techniques has been described, and a standardized technique for preparing DMEK grafts has been identified. The tissues in general were prepared by superficially tapping the endothelial side with a Moria trephine (9.5 mm diameter). The plane of cleavage was identified using a cleavage hook, and the DMEK graft was deadhered from the trephined site throughout the circumference for ease of excising the graft. The DMEK graft was peeled using either one or multiple quadrant methods depending on the challenges faced during excision. The graft was finally marked with the letter “F” to identify the orientation during surgery. Data on endothelial cell loss (ECL) and challenging cases were observed, monitored, and recorded during this period. Results: Less than 1 percent trypan blue-positive cells with tissue wastage of <6% was observed during the study period. Our standardized stripping technique has resulted in an overall ECL of 4.6%. Marking Descemet membrane showed 0.5% cell mortality. Conclusions: Standardizing DMEK technique using specific tools and simple techniques would help new surgeons to decide the instruments and improve their tissue preparation skills also in challenging cases such as previous cataract incisions or horseshoe-shaped tears, further reducing ECL or tissue wastage.