The era of artificial intelligence (AI) has revolutionized our daily lives and AI has become a powerful force that is gradually transforming the field of medicine. Ophthalmology sits at the forefront of this transformation thanks to the effortless acquisition of an abundance of imaging modalities. There has been tremendous work in the field of AI for retinal diseases, with age-related macular degeneration being at the top of the most studied conditions. The purpose of the current systematic review was to identify and evaluate, in terms of strengths and limitations, the articles that apply AI to optical coherence tomography (OCT) images in order to predict the future evolution of age-related macular degeneration (AMD) during its natural history and after treatment in terms of OCT morphological structure and visual function. After a thorough search through seven databases up to 1 January 2022 using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 1800 records were identified. After screening, 48 articles were selected for full-text retrieval and 19 articles were finally included. From these 19 articles, 4 articles concentrated on predicting the anti-VEGF requirement in neovascular AMD (nAMD), 4 articles focused on predicting anti-VEGF efficacy in nAMD patients, 3 articles predicted the conversion from early or intermediate AMD (iAMD) to nAMD, 1 article predicted the conversion from iAMD to geographic atrophy (GA), 1 article predicted the conversion from iAMD to both nAMD and GA, 3 articles predicted the future growth of GA and 3 articles predicted the future outcome for visual acuity (VA) after anti-VEGF treatment in nAMD patients. Since using AI methods to predict future changes in AMD is only in its initial phase, a systematic review provides the opportunity of setting the context of previous work in this area and can present a starting point for future research.

Aim: The aim of this study was to evaluate the visual outcome of lens dislocation and subluxation managed by pars plana vitrectomy (PPV) and lensectomy in patients with open (OGIs) or closed globe injuries (CGIs). Methods: Medical records of 70 consecutive patients treated by PPV and lensectomy over a period of 11 years (1 January 2010–31 December 2020) were retrospectively reviewed. We collected demographic data, best corrected visual acuity (BCVA) using a Snellen Chart pre- and postoperatively, associated ocular injuries and treatment strategy. Visual outcome was evaluated according to the final BCVA which was defined as poor <0.1 or good ≥0.1. Results: The mean age was 57.9 ± 17.6 years. CGIs were present in 49 (70.0%) cases and open OGIs in 21 (30.0%) cases. The dislocation mechanism was zonular lysis in 59 cases (84.3%) and capsular rupture in 11 cases (15.7%). The intraocular lens implant (IOL) was sutured to the sclera in 51 (72.9%) cases or positioned in the capsular bag or in the sulcus in 3 (4.3%) cases and 1 (1.4%) case, respectively, whereas 15 (21.4%) patients remained aphakic. A good BCVA ≥ 0.1 was achieved in 45.71% of the eyes. The presence of retinal detachment (RD) (p = 0.014), iridodonesis (p = 0.011) and initial BCVA (p = 0.000) achieved statistical significance in predicting visual outcome. After treatment, 45.71% of patients achieved a final BCVA ≥ 0.1. Conclusion: RD, iridodonesis and initial BCVA were risk factors for poor visual outcome in our series.

Aim: To evaluate the visual outcome of traumatic endophthalmitis and describe the risk factors associated with poor visual acuity and retinal detachment (RD) development over an 11-year period. Methods: Medical records of 34 patients with traumatic endophthalmitis who underwent PPV over a period of 11 years (1 January 2010–31 December 2020) were reviewed. We extracted details regarding demographic data, initial and final best corrected visual acuity (BCVA) using a standard Snellen chart, wound and IOFB characteristics, ocular associated lesions, and treatment. The outcome was evaluated according to the final BCVA which was defined as poor < 0.1 or good ≥ 0.1 Results: Endophthalmitis rate was 29.8% in open globe injuries. The mean age was 43.6 ± 16.5 years and the majority of patients were males (32 out of 34, 94.1%). All patients received systemic (moxifloxacin) and intravitreal antibiotherapy. We performed pars plana vitrectomy (PPV) in all cases. Poor visual outcome was associated with wound size ≥ 3 mm (p = 0.02), the association of IOFB (p = 0.016), and the development of RD (p = 0.00). The presence of IOFB (p = 0.01) and wound size ≥ 3 mm (p = 0.01) were statistically associated with RD development. After treatment, 47.05% of patients achieved final BCVA ≥ 0.1. Conclusion: Wound size ≥ 3 mm, IOFB and RD were risk factors for poor visual outcomes in traumatic endophthalmitis.

Aim: To evaluate the visual outcome of penetrating ocular injuries with a retained intraocular foreign body (IOFB) managed by pars plana vitrectomy (PPV) and to describe the risk factors associated with poor visual acuity and retinal detachment (RD) development. Methods: Medical records of 56 patients with IOFB that were removed by PPV over a period of 11 years (1 January 2010–31 December 2020) were reviewed. We extracted the demographic data, initial and final best corrected visual acuity (BCVA) using standard Snellen chart, IOFB characteristics, complications and surgical details. Outcome was evaluated according to the final BCVA: poor <0.1, good 0.1–3mm in 30 (53.6%) cases. Preoperative RD was identified in 12 (21.4%) cases and endophthalmitis in 17 cases (30.4%). IOFBs larger than 3 mm and retinal location were associated with RD development. Poor visual outcome was associated with initial BCVA, retinal location, RD and endophthalmitis. Conclusion: Initial BCVA, retinal foreign body, RD and endophthalmitis were risk factors for poor visual outcome.

The purpose of this study is to analyze and compare pediatric normative data for the retinal nerve fiber layer of Romanian children using two different spectral domain optical coherence tomographs. Due to different scanning speeds and axial and transverse resolution, the results of the measurements of scans cannot be transposed. A total of 140 healthy children aged 4 to 18 were enrolled in the study. Overall, 140 eyes were scanned with a Spectralis SD-OCT (Heidelberg Technology), and the other 140 eyes were imaged with a Copernicus REVO SOCT (Optopol Technology (Zawiercie, Poland)). The mean global RNFL thickness and average RNFL thickness for the four quadrants were measured and compared. The average peripapillary RNFL thickness measured with the Spectralis was 104.03 ± 11.42 (range 81 to 126 µm), while the one measured with the Revo 80 was 127.05 ± 15.6 (range 111.43–158.28). The RNFL thickness measurements taken with the Spectralis in the superior, inferior, nasal, and temporal quadrants were 132 ±19.1, 133.5 ± 21.77, 74 ± 16.48, and 73 ± 11.95 µm, respectively, while those taken with the Revo 80 were 144.44 ± 9.25, 144.86 ±23.12, 96.49 ± 19.41, and 77 ± 11.4 µm, respectively. Multivariate analysis showed that the average RNFL thickness was not influenced by gender or eye laterality and was negatively correlated with age when we used the Spectralis device. This study provides normative data for SD-OCT peripapillary RNFL in healthy Romanian children for two different tomographs. These data help the clinician evaluate and interpret the results of optical coherence tomography for a child, taking into consideration all the technical and individual parameters.

Patient compliance in chronic illnesses is essential for disease management. This also applies to age-related macular degeneration (AMD), a chronic acquired retinal degeneration that needs constant monitoring and patient cooperation. Therefore, patients with AMD can benefit by being properly informed about their disease, regardless of the condition’s stage. Information is essential in keeping them compliant with lifestyle changes, regular monitoring, and treatment. Large language models have shown potential in numerous fields, including medicine, with remarkable use cases. In this paper, we wanted to assess the capacity of two large language models (LLMs), ChatGPT4 and PaLM2, to offer advice to questions frequently asked by patients with AMD. After searching on AMD-patient-dedicated websites for frequently asked questions, we curated and selected a number of 143 questions. The questions were then transformed into scenarios that were answered by ChatGPT4, PaLM2, and three ophthalmologists. Afterwards, the answers provided by the two LLMs to a set of 133 questions were evaluated by two ophthalmologists, who graded each answer on a five-point Likert scale. The models were evaluated based on six qualitative criteria: ($C_1$) reflects clinical and scientific consensus, ($C_2$) likelihood of possible harm, ($C_3$) evidence of correct reasoning, ($C_4$) evidence of correct comprehension, ($C_5$) evidence of correct retrieval, and ($C_6$) missing content. Out of 133 questions, ChatGPT4 received a score of five from both reviewers to 118 questions (88.72%) for $C_1$, to 130 (97.74%) for $C_2$, to 131 (98.50%) for $C_3$, to 133 (100%) for $C_4$, to 132 (99.25%) for $C_5$, and to 122 (91.73%) for $C_6$, while PaLM2 to 81 questions (60.90%) for $C_1$, to 114 (85.71%) for $C_2$, to 115 (86.47%) for $C_3$, to 124 (93.23%) for $C_4$, to 113 (84.97%) for $C_5$, and to 93 (69.92%) for $C_6$. Despite the overall high performance, there were answers that are incomplete or inaccurate, and the paper explores the type of errors produced by these LLMs. Our study reveals that ChatGPT4 and PaLM2 are valuable instruments for patient information and education; however, since there are still some limitations to these models, for proper information, they should be used in addition to the advice provided by the physicians.