Artificial intelligence (AI) has seen significant progress in medical diagnostics, particularly in image and video analysis. This review focuses on the application of AI in analyzing in vivo confocal microscopy (IVCM) images for corneal diseases. The cornea, as an exposed and delicate part of the body, necessitates the precise diagnoses of various conditions. Convolutional neural networks (CNNs), a key component of deep learning, are a powerful tool for image data analysis. This review highlights AI applications in diagnosing keratitis, dry eye disease, and diabetic corneal neuropathy. It discusses the potential of AI in detecting infectious agents, analyzing corneal nerve morphology, and identifying the subtle changes in nerve fiber characteristics in diabetic corneal neuropathy. However, challenges still remain, including limited datasets, overfitting, low-quality images, and unrepresentative training datasets. This review explores augmentation techniques and the importance of feature engineering to address these challenges. Despite the progress made, challenges are still present, such as the “black-box” nature of AI models and the need for explainable AI (XAI). Expanding datasets, fostering collaborative efforts, and developing user-friendly AI tools are crucial for enhancing the acceptance and integration of AI into clinical practice.

Background & Objective: Corneal endothelial cells (ECD) are characterized by limited regenerative potential, which is additionally impaired in patients with diabetes. This retrospective study included 27 patients aged 58.1±13.6, 16 female and 11 males, who underwent 23-gauge vitrectomy in combination with cataract surgery (phacovitrectomy) and further Ex-PRESS shunt implantation throughout 2013-2017 at St. Barbara Hospital in Sosnowiec, Poland.Methods: In our study, we distinguished 4 periods: initial period; post phacovitrectomy and removal of oil tamponade; and 3 and 12 months post implantation of the Ex-PRESS shunt. Statistical analysis was performed at the level of statistical significance of p<0.05. It included an analysis of variance (ANOVA) and Tukey’s post-hoc test in order to determine the differences in the density of ECD cells/mm2 between the periods of observation. The paired-samples t-Student test was also performed to determine whether the differences in visual acuity values before and after PPV and before and after Ex-PRESS shunt were statistically significant.Results: The initial count of ECD cells was 2381.1±249, which decreased to 1872.8±350.7 cell/mm2 and finally to 1677.9±327 at the endpoint. Differences in the density of ECD cells/mm2 were observed to be statistically significant between the periods: after PPV vs. initial number of ECD (p = 0.000138); before 3 months after Ex-PRESS shunt vs. initial number of ECD (p = 0.000138); 12 months after Ex- PRESS shunt vs. initial number of ECD (p = 0000138). Analyzing the changes in visual acuity, we observed a deterioration both before and 3 months after Ex-PRESS shunt (p = 0.007944) and before and after PPV (p = 0.060334). In turn, correlation analysis indicated that there is a statistically significant, moderate, positive relationship. The relationship between visual acuity after Ex-PRESS shunt and ECD cells/mm2 density turned out to be statistically significant (r = +0.521381; p < 0.05).Conclusion: Regardless of the period of observation and the choice of ophthalmic treatment of diabetic complications, we observed a decrease in the number of ECD cells and a deterioration in visual acuity. It is, therefore, reasonable to provide the patient with complete information about the proposed procedures and to consider the risk-benefit balance.

Background/Objectives: Keratoconus is a progressive corneal disorder that significantly impairs visual quality and daily functioning. This study aimed to translate, culturally adapt, and validate the Polish version of the Keratoconus Outcomes Research Questionnaire (KORQ), a tool designed to assess keratoconus-related vision problems and their impact on quality of life. Methods: This study involved three phases: translation, validation, and optimization. The translation followed cross-cultural adaptation guidelines, including forward translation, back translation, and pretesting. In the validation phase, 100 patients with keratoconus completed the Polish KORQ. Psychometric properties were assessed using Rasch analysis to evaluate item fit, reliability, unidimensionality, and targeting. Exploratory and confirmatory factor analyses (EFA and CFA) were conducted to examine the latent structure of the questionnaire. Regression analysis examined the demographic and clinical factors influencing keratoconus severity and vision-related quality of life, with the total KORQ score as the dependent variable. Results: The Polish KORQ demonstrated excellent psychometric properties. Internal consistency was high (α = 0.95 for activity limitation, α = 0.87 for symptoms). Rasch analysis confirmed good item fit and measurement reliability. EFA identified a two-factor structure consistent with the original questionnaire, explaining 53% of the total variance. CFA supported this model, with good fit indices (CFI = 0.981, TLI = 0.980), despite slightly elevated RMSEA (0.104) and SRMR (0.109). The two latent factors were moderately correlated (r = 0.729). Regression analysis showed that uncorrected visual acuity and disease severity significantly predicted lower quality of life. Conclusions: The Polish KORQ is a reliable and valid tool for assessing the impact of keratoconus in clinical and research settings.

Background: Managing post-keratoplasty astigmatism remains challenging. Even though graft viability is the main concern in keratoplasty, astigmatism might hinder vision recovery following a successful corneal transplant. Photorefractive keratectomy (PRK) and topography-guided PRK may be options for correcting refractive errors in patients who underwent keratoplasty. The aim of the study was to compare the results of PRK and topography-guided PRK in patients who had undergone corneal keratoplasty. Methods: This study was conducted at the Chair and Ophthalmology Clinic of the Medical University of Silesia, at the Railway Hospital Katowice, from 2023 to 2024. Patients who underwent penetrating keratoplasty due to keratoconus or corneal scar (post-traumatic and post-inflammatory) with a residual spherical myopic or mixed myopic astigmatic refractive defect were included in this study. The studied patients were divided into two groups: 15 patients underwent PRK and 15 patients underwent topography-guided PRK. Each participant underwent a preoperative examination, including uncorrected visual acuity (UCVA) and best spectacle-corrected visual acuity (BSCVA) measured on the Snellen chart (LCD panel for visual acuity testing Frey CP-400, Optotech Medical, Niepołomnice, Niepołomice, Poland), cycloplegic refraction, corneal pachymetry and topography (Schwind Sirius+, Pentacam), wavefront aberrations (Schwind Peramis, Pentacam), applanation tonometry, and anterior and posterior segment examinations, conducted at baseline and 1, 3, 6, and 12 months. Results: Keratoconus was the most common reason for keratoplasty (80% vs. 60%). Following PRK, the mean KI in group (1) decreased significantly to 43.88 ± 3.64 (p < 0.001), and in the group (2), the mean diopters decreased significantly after the intervention to 46.46 ± 2.80 (p < 0.001). The mean spherical refractive error in group (2) changed significantly after the intervention, reaching −2.72 ± 1.28 D (p < 0.001). The mean cylinder in group (1) changed significantly after PRK to have a mean of −2.75 ± 1.44 D (p < 0.001). Also, in group (2), it changed significantly after the intervention to have a mean of −2.95 ± 1.99 D (p < 0.001). There was a significant increase in both uncorrected visual acuity (UCVA) and best-corrected visual acuity (BCVA) after topography-guided PRK at all the follow-up points of 1, 3, 6, and 12 months (p < 0.001). The mean higher-order corneal aberrations (HOAs) after topography-guided PRK were 1.33 ± 0.50, ranging from 0.22 to 2.34 (p < 0.001). Conclusions: Topography-guided PRK is safer and more effective in correcting aberrations and refractive errors after corneal keratoplasty than regular PRK. Additionally, topography-guided PRK reduces high-order aberrations by utilizing advanced topographic mapping of the cornea, enabling precise customization of the treatment to address individual corneal surface irregularities.

Background: In vivo confocal microscopy (IVCM) provides high-resolution corneal imaging that may enhance preoperative and postoperative assessment in refractive surgery. This pilot study aimed to evaluate the diagnostic utility of IVCM in identifying subclinical corneal abnormalities that could influence surgical qualification and outcomes. Methods: A total of 7 patients (3 males, 4 females; mean age 48.8 ± 14.5 years) undergoing qualification or follow-up for refractive surgery were prospectively examined between May 2021 and March 2025. Each participant underwent a comprehensive ophthalmic evaluation, including slit-lamp biomicroscopy, corneal topography, anterior segment optical coherence tomography (AS-OCT), and IVCM using the Heidelberg Retina Tomograph II with Rostock Cornea Module. Patients with prior ocular surgery, active infection, or systemic corneal disease were excluded. Results: IVCM revealed subtle epithelial, stromal, and endothelial abnormalities undetectable by conventional methods. Findings such as Thygeson’s keratitis, pre-Descemet’s dystrophy, and subclinical herpes simplex keratitis led to modifications of surgical plans or disqualification in selected cases. The technique also aided postoperative evaluation of epithelial–stromal interface disorders. Conclusions: IVCM proved to be a valuable adjunct in detecting subclinical corneal pathology, refining patient selection, and improving safety in refractive surgery. Larger multicenter studies are warranted to validate its clinical role and define standardized indications for preoperative screening.

Background/Objectives: This study assessed outcomes between penetrating keratoplasty (PK) and Descemet’s stripping automated endothelial keratoplasty (DSAEK) when combined with cataract surgery as part of the triple procedure. Methods: Retrospective analysis of 727 triple procedures (525 PK and 202 DSAEK) from 2007–2023. Graft survival, visual acuity, and refractive outcomes were analyzed. Kaplan–Meier and Cox regression were used for survival and prognostic analysis. Results: No statistically significant difference in survival was found (PK—42 months; DSAEK—47 months). DSAEK had better visual acuity improvement and refractive stability. PK had higher astigmatism and variability in refractive error. Conclusions: While graft survival was comparable, DSAEK offers superior visual rehabilitation, supporting its use when refractive predictability is important.