Canadian journal of ophthalmology. Journal canadien d'ophtalmologie最新文献

Objective: To assess the effect of prompt engineering on the accuracy, comprehensiveness, readability, and empathy of large language model (LLM)-generated responses to patient questions regarding retinal disease.

Design: Prospective qualitative study.

Participants: Retina specialists, ChatGPT3.5, and GPT4.

Methods: Twenty common patient questions regarding 5 retinal conditions were inputted to ChatGPT3.5 and GPT4 as a stand-alone question or preceded by an optimized prompt (prompt A) or preceded by prompt A with specified limits to length and grade reading level (prompt B). Accuracy and comprehensiveness were graded by 3 retina specialists on a Likert scale from 1 to 5 (1: very poor to 5: very good). Readability of responses was assessed using Readable.com, an online readability tool.

Results: There were no significant differences between ChatGPT3.5 and GPT4 across any of the metrics tested. Median accuracy of responses to a stand-alone question, prompt A, and prompt B questions were 5.0, 5.0, and 4.0, respectively. Median comprehensiveness of responses to a stand-alone question, prompt A, and prompt B questions were 5.0, 5.0, and 4.0, respectively. The use of prompt B was associated with a lower accuracy and comprehensiveness than responses to stand-alone question or prompt A questions (p < 0.001). Average-grade reading level of responses across both LLMs were 13.45, 11.5, and 10.3 for a stand-alone question, prompt A, and prompt B questions, respectively (p < 0.001).

Conclusions: Prompt engineering can significantly improve readability of LLM-generated responses, although at the cost of reducing accuracy and comprehensiveness. Further study is needed to understand the utility and bioethical implications of LLMs as a patient educational resource.

Objective: To quantify post-cataract surgery health care utilization caused by dry eye symptoms (DES) and identify preoperative risk factors and mediators of postoperative DES.

Setting: An outpatient surgical centre in Toronto, Canada.

Design: Retrospective cohort study.

Participants: Included patients had cataract surgery between April 2019 and January 2020, completed a preoperative Dry Eye Questionnaire 5 (DEQ5), and were over age 18.

Methods: Data collected included DES risk factors, prophylaxis, and intraoperative and postoperative details. DES health care utilization and prevalence were reported as percentages. Risk of DES follow-up with DES prophylaxis use, and within each DEQ5 severity group, were analyzed with χ²-square test of independence and odds ratios. Binomial logistic regression assessed for significance of multiple preoperative and intraoperative risk factors, controlling for one another.

Results: Of 1074 patients (46% male, mean age: 71, mean DEQ5: 5.12), 18.1% had at least one nonroutine postoperative appointment due to DES. Patients with moderate/severe DEQ5 scores were 1.843 times likelier to have postoperative DES concerns than those with no/mild scores (CI = 1.307-2.599). Preoperative DES prophylaxis did not reduce DES risk in any severity group. Higher DEQ5 score, female sex, bilateral sequential surgeries, and femtosecond laser-assisted cataract surgery had higher likelihood of postoperative DES (B = 0.054; p < 0.001; B = -0.351; p = 0.037; B = 0.695; p = 0.003; B = 0.491; p = 0.003, respectively).

Conclusion: Nearly 1 in 5 patients had DES postcataract surgery, with 1 in 12 requiring one or more nonroutine follow-ups for DES, suggesting substantial health care burden. Current standard preoperative DES treatment may not reduce postoperative DES, and further studies are needed to elucidate why.

Objective: To examine the association between quantitative vascular parameters extracted from intravenous fluorescein angiography (IVFA) and baseline clinical characteristics of patients with retinal vein occlusion (RVO).

Methods: Our prospective single-centre study in Toronto, Canada, recruited patients with a diagnosis of macular edema secondary to RVO presenting with a central macular thickness (CMT) ≥310 μm from 2017 to 2023. IVFA images were captured using an ultra-widefield scanning laser ophthalmoscope and processed using the artificial intelligence-based RETICAD system to extract quantitative measurements of blood flow, perfusion, and blood-retinal barrier (BRB) permeability. Univariable and multivariable regression models were used to investigate associations between quantitative IVFA parameters and baseline best-corrected visual acuity (BCVA), CMT, and macular volume.

Results: The study included 41 eyes from 41 RVO patients. In the multivariable analysis, BRB permeability was significantly associated with both CMT (p < 0.001) and macular volume (p = 0.005). Subgroup analyses revealed that in central RVO patients, central BRB permeability remained significantly associated with CMT (p = 0.022) and macular volume (p = 0.010); however, there was no association with BCVA (p = 0.921). In branch RVO patients, central BRB permeability was significantly associated with BCVA (p = 0.006) and CMT (p = 0.009), but not with macular volume (p = 0.723). Additionally, both central and peripheral BRB permeability were significantly higher in patients with RVO compared to healthy controls (p < 0.001).

Conclusions: Our investigation reveals novel associations between baseline clinical characteristics and quantitative IVFA parameters in RVO patients, which may serve as clinically relevant biomarkers. Future studies should explore these associations in diverse RVO patient populations with extended follow-up.

Objective: To investigate the localization, distribution, and type of central microaneurysms (MAs) and their relationship with retinal vascular alterations in patients with retinal vein occlusion (RVO).

Methods: In this cross-sectional study, ultra-widefield color fundus photography (UWF-CF), standard and single-capture 65° widefield (WF) optical coherence tomography angiography (OCTA) were performed in consecutive patients with RVO treated at the Department of Ophthalmology and Optometry, Medical University of Vienna. UWF-CF, en face and B-Scans in 6 mm × 6 mm OCTA were examined for detection of MAs. Nonperfusion areas (NPA) and collateral vessels (CV) were evaluated on WF-OCTA, ghost vessels (GV), and tortuous vessels (TV) on UWF-CF.

Results: One-hundred-and-twelve patients were included in the study, and data from 59 eyes of 59 patients with disease duration longer than 3 months, good image quality, and without relevant ocular comorbidities were eligible for statistical analysis. Fifty-six of 59 (94.9%) patients were previously treated with anti-vascular endothelial growth factor agents for macular edema, 31 of 59 (52.5%) patients presented with MAs in the central 6 mm and 60 MAs were found in total using multimodal imaging. There was no statistically significant difference in the greatest diameter of fluid-associated versus non-fluid-associated MAs (p = 0.53). Eyes with MAs were associated with CV, TV, and GV (χ²-test; p < 0.001, p = 0.0498, and p = 0.001). Median NPA was 27.3 mm² (quartiles 1.3-62.8 mm²) in eyes with MAs and 0 mm² (quartiles 0-36.2 mm²) in eyes without MAs (Mann-Whitney-U-test; p = 0.018).

Conclusion: MAs were associated with extensive NPA, the presence of CV, GV, and TV. There was no correlation between the diameter of the MA and the adjacent intraretinal fluid in our predominantly pretreated RVO study patients.

Objective: To develop an automated, new framework based on machine learning to diagnose malignant eyelid skin tumors.

Methods: This study used eyelid lesion images from Sheba Medical Center, a large tertiary center in Israel. Before model training, we pretrained our models on the International Skin Imaging Collaboration (ISIC) 2019 dataset consisting of 25,332 images. The proprietary eyelid data set was then used for fine-tuning. The data set contained multiple images per patient, aiming to classify malignant lesions in comparison to benign counterparts.

Results: The analyzed data set consisted of images representing both benign and malignant eyelid lesions. For the benign category, a total of 373 images were sourced. By comparison, for the malignant category, 186 images were sourced. For the final model, at sensitivity of 93.8% (95% CI 80.0-100.0%), the model has a corresponding specificity of 73.7% (95% CI 60.0-87.1%). To further understand the decision-making process of our model, we employed heatmap visualization techniques, specifically gradient-weighted Class Activation Mapping.

Discussion: This study introduces a dependable model-aided diagnostic technology for assessing eyelid skin lesions. The model demonstrated accuracy comparable to human evaluation, effectively determining whether a lesion raises a high suspicion of malignancy or is benign. Such a model has the potential to alleviate the burden on the health care system, particularly benefiting rural areas, and enhancing the efficiency of clinicians and overall health care.

Background: Whether optical computed tomography (OCT) and magnetic resonance imaging (MRI) findings are associated with final visual acuity in children with myelin oligodendrocyte glycoprotein antibody disease (MOGAD) optic neuritis is unclear.

Methods: We retrospectively reviewed the charts of pediatric patients with MOGAD optic neuritis seen at St. Louis Children's Hospital/Barnes Jewish Hospital since 2016.

Results: In the 12 patients in this study, presenting visual acuity was worse in the optic neuritis-affected eyes but significantly improved from presentation to follow-up, such that, at last follow-up, there was no longer a statistical difference between the affected and unaffected eyes. The number of affected eyes with nerve enhancement and the amount of optic nerve affected, as well as thickness of the retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), and macula, decreased from presentation to follow-up. Ultimately, none of these variables were associated with final visual acuity.

Conclusion: In this cohort, pediatric MOGAD optic neuritis patients had positive visual outcomes despite significant RNFL thinning and involvement of the optic nerve on MRI, leading to a lack of correlation between follow-up visual acuity and OCT and MRI measures of disease severity, respectively.

Purpose: Over 6 million people in the United States participate in martial arts, and the focus on striking the face poses a significant risk of injury to the eye. The objective of this study is to characterize the nature and incidence of martial arts-related eye injuries in the United States.

Methods: A retrospective analysis of the National Electronic Injury Surveillance System (NEISS) database served as the basis of the study. The NEISS database was queried for eye injuries related to martial arts from 2012 to 2021. The entries that resulted were coded for demographic information, mechanism of injury, martial arts discipline, and diagnosis. Data were analyzed using SPSS version 28 through descriptive and χ² analyses.

Results: There were 162 reported cases of martial arts-related eye injuries in 4- to 62-years-old patients with most injuries occurring in males (84%). The most represented martial arts discipline was boxing (30.5%), and the most frequently reported mechanism of injury was punching (30.2%). The most frequently reported diagnosis resulting from eye injury was corneal abrasion (40.5%), while sight-threatening diagnoses were reported, including retinal detachment (n = 4), vitreous hemorrhage (n = 2), and open globe (n = 2).

Conclusions: Despite the significant risk of ocular injury from martial arts, corneal abrasion was the most commonly diagnosed with excellent prognosis. Sight-threatening diagnoses were also reported, indicating that martial arts athletes should be educated about the risks of eye injury and the importance of regular eye examination when practicing combat sports.

Objective: To report the surgical outcomes of a modified trabeculectomy technique involving implanting the Tenon's layer under the scleral flap.

Design: Prospective, interventional case series.

Participants: 51 eyes with medically uncontrolled glaucoma were enrolled for this study. A new trabeculectomy technique, the Tenon's filtration technique for trabeculectomy (TFT-LEC) was used in 26 eyes, while a conventional procedure, normal trabeculectomy (N-LEC), was used for 25 eyes.

Methods: Intraocular pressure (IOP) control, the number of glaucoma medications, the need for additional interventions, and postoperative complications were assessed.

Results: Twelve months postoperatively, the mean IOP was 13.5 ± 0.5 mmHg in the TFT-LEC group and 15.4 ± 0.5 mmHg in the N-LEC group (p = 0.13). The TFT-LEC group required an average of 1.3 ± 1.0 additional glaucoma medications (21 cases required only ripasudil) postoperatively, with no cases of bleb needling revision or reoperation. The N-LEC group required an average of 1.7 ± 1.5 glaucoma medications (p = 0.43) compared to TFT-LEC group, eight cases (32%) required bleb needling revision (p = 0.002), and one case (4%) of reoperation (p = 0.49). The complications in the TFT-LEC group included shallow anterior chamber in six (23 %) cases (p = 1.00) compared to N-LEC group, choroidal detachment in two (8%) cases (p = 0.42), and anterior chamber hemorrhage in seven (27%) cases (p = 0.29). None of these complications affected visual function.

Conclusions: This new technique involving implanting the Tenon's layer under the scleral flap may improve the postoperative outcomes of trabeculectomy.