Background: Large language models (LLMs) are becoming increasingly important as they are being used more frequently for providing medical information. Our aim is to evaluate the effectiveness of electronic artificial intelligence (AI) large language models (LLMs), such as ChatGPT-4, BingAI, and Gemini in responding to patient inquiries about retinopathy of prematurity (ROP). Methods: The answers of LLMs for fifty real-life patient inquiries were assessed using a 5-point Likert scale by three ophthalmologists. The models’ responses were also evaluated for reliability with the DISCERN instrument and the EQIP framework, and for readability using the Flesch Reading Ease (FRE), Flesch-Kincaid Grade Level (FKGL), and Coleman-Liau Index. Results: ChatGPT-4 outperformed BingAI and Gemini, scoring the highest with 5 points in 90% (45 out of 50) and achieving ratings of “agreed” or “strongly agreed” in 98% (49 out of 50) of responses. It led in accuracy and reliability with DISCERN and EQIP scores of 63 and 72.2, respectively. BingAI followed with scores of 53 and 61.1, while Gemini was noted for the best readability (FRE score of 39.1) but lower reliability scores. Statistically significant performance differences were observed particularly in the screening, diagnosis, and treatment categories. Conclusion: ChatGPT-4 excelled in providing detailed and reliable responses to ROP-related queries, although its texts were more complex. All models delivered generally accurate information as per DISCERN and EQIP assessments.

To evaluate the visual, refractive and corneal asphericity changes after intrastromal corneal ring segment (ICRS) implantation for visual rehabilitation of keratoconus. A total of 42 eyes of 32 patients were included. After creation of intrastromal tunnels of 5.01±0.03 mm inner diameter, 5.71±0.03 mm outer diameter and at 384.21±34.12 µm depth, 1 or 2 ICRS of 150-300 µm thickness and 90°-210° arc length were implanted. Changes in uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), refractive errors, mean simulated keratometry readings (SimK_avg), anterior and posterior corneal asphericity values (Q_ant and Q_post, respectively) measured with Scheimpflug topography were evaluated retrospectively. There was a significant improvement in UDVA and CDVA, along with a significant decrease in refractive spherical equivalent (SE), cylinder and SimK_avg postoperatively (P<0.001 for all). Mean Q_ant increased from -1.06±0.48 to -0.57±0.58 postoperatively (P0.05 for all); but were correlated with the preoperative SimK_avg and Q_ant values (P<0.001 for all). ICRS implantation seems to approximate the anterior corneal asphericity of "advanced prolate" shape to "optimal prolate" shape and an "ideal Q value" of -0.46; which may have a role in improved UDVA and CDVA postoperatively, besides decreased refractive cylinder values.

To evaluate the effectiveness of machine learning (ML) models in predicting the occurrence of retinopathy of prematurity (ROP) and treatment need. Four ML models were created using 49 parameters known within the first 24 h post-birth and obtained during the initial screening examination, encompassing demographic, maternal, clinical, and neonatal intensive care unit-related data. The models’ performances were assessed using five machine learning (ML) classifier algorithms: logistic regression (LR), decision tree (DT), support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost). Performance metrics were calculated, and the top ten parameters with the highest predictive value were identified. In the cohort of 355 preterm infants, Model I, predicting ROP development using birth data, achieved a balanced accuracy of 80%, with gestational age (GA), birth weight (BW) and mean corpuscular volume (MCV) as the top predictive parameters. Model II, predicting treatment-requiring ROP using birth data, exhibited a balanced accuracy of 81%. Key predictive parameters included low GA, BW, 1-minute and 5-minute APGAR scores, and low erythrocyte counts. For Model III, predicting ROP using the first screening examination data, and Model IV, predicting treatment-requiring ROP using the same data, the accuracy values were 80% and 66%, respectively, with BW, daily weight gain, total O2 support duration, and platelet/lymphocyte ratio emerged as the most significant predictive parameters in both models. This study demonstrates the potential of ML models to predict ROP development and treatment need. Incorporating clinical and intensive care-related parameters can enhance ROP screening and clinical decision-making.

Glaucoma is a chronic neurodegenerative disease of the optic nerve and a leading cause of irreversible blindness, worldwide. While the experimental research using animal models provides growing information about cellular and molecular processes, parallel analysis of the clinical presentation of glaucoma accelerates the translational progress towards improved understanding, treatment, and clinical testing of glaucoma. Optic nerve axon injury triggers early alterations of retinal ganglion cell (RGC) synapses with function deficits prior to manifest RGC loss in animal models of glaucoma. For testing the clinical relevance of experimental observations, this study analyzed the functional correlation of localized alterations in the inner plexiform layer (IPL), where RGCs establish synaptic connections with retinal bipolar and amacrine cells. Participants of the study included a retrospective cohort of 36 eyes with glaucoma and a control group of 18 non-glaucomatous subjects followed for two-years. The IPL was analyzed on consecutively collected macular SD-OCT scans, and functional correlations with corresponding 10–2 visual field scores were tested using generalized estimating equations (GEE) models. The GEE-estimated rate of decrease in IPL thickness (R = 0.36, P<0.001) and IPL density (R = 0.36, P0.05). These findings support early localized IPL alterations in correlation with progressing visual field defects in glaucomatous eyes. Considering the experimental data, glaucoma-related increase in IPL thickness/density might reflect dendritic remodeling, mitochondrial redistribution, and glial responses for synapse maintenance, but decreased IPL thickness/density might correspond to dendrite atrophy. The bridging of experimental data with clinical findings encourages further research along the translational path.