Transformer-Based Deep Learning Prediction of 10-Degree Humphrey Visual Field Tests From 24-Degree Data.

Purpose: To predict 10-2 Humphrey visual fields (VFs) from 24-2 VFs and associated non-total deviation features using deep learning. Methods: We included 5189 reliable 24-2 and 10-2 VF pairs from 2236 patients, and 28,409 reliable pairs of macular OCT scans and 24-2 VF from 19,527 eyes of 11,560 patients. We developed a transformer-based deep learning model using 52 total deviation values and nine VF test features to predict 68 10-2 total deviation values. The mean absolute error, root mean square error, and the R2 were evaluation metrics. We further evaluated whether the predicted 10-2 VFs can improve the structure-function relationship between macular thinning and paracentral VF loss in glaucoma. Results: The average mean absolute error and R2 for 68 10-2 VF test points were 3.30 ± 0.52 dB and 0.70 ± 0.11, respectively. The accuracy was lower in the inferior temporal region. The model placed greater emphasis on 24-2 VF points near the central fixation point when predicting the 10-2 VFs. The inclusion of nine VF test features improved the mean absolute error and R2 up to 0.17 ± 0.06 dB and 0.01 ± 0.01, respectively. Age was the most important 24-2 VF test parameter for 10-2 VF prediction. The predicted 10-2 VFs achieved an improved structure-function relationship between macular thinning and paracentral VF loss, with the R2 at the central 4, 12, and 16 locations of 24-2 VFs increased by 0.04, 0.05 and 0.05, respectively (P < 0.001). Conclusions: The 10-2 VFs may be predicted from 24-2 data. Translational Relevance: The predicted 10-2 VF has the potential to improve glaucoma diagnosis.

Technology and Methodology in Glaucoma Case Detection.

Even the most powerful handheld screening tool for glaucoma will not find cases cost-effectively because the disease prevalence is low. However, a tier 1 biomarker tool that enhances disease prevalence followed by the use of a screening tool could serve as a future glaucoma screening strategy.

Implementation and Consensus.

■■■.

The clinical utility of a glaucoma polygenic risk score in four population-based European-ancestry cohorts.

OBJECTIVE: We used a polygenic risk score (PRS) to identify high-risk groups for primary open-angle glaucoma (POAG) within population-based cohorts. DESIGN: Secondary analysis of four prospective population-based studies. PARTICIPANTS: We included four European-ancestry cohorts: the United States (US) based Nurses' Health Study (NHS), NHS2, and the Health Professionals Follow-up Study, and the Rotterdam Study (RS) in the Netherlands. The US cohorts included female nurses and male health professionals aged 55+ years. The RS included residents aged 45 years or older living in Rotterdam, the Netherlands. METHODS: PRS weights were estimated by applying the Lassosum method on imputed genotype and phenotype data from the UK-Biobank. This resulted in 144,020 variants, single nucleotide polymorphism (SNPs) and indels, with non-zero betas that we used to calculate a PRS in the target populations. Using multivariable Cox proportional hazard models, we estimated the relationship between the standardized PRS and relative risk for POAG. Additionally, POAG prediction was tested by calculating these models' concordance (Harrell's C-statistic). Finally, we assessed the association between PRS tertiles and glaucoma-related traits. MAIN OUTCOME MEASURES: The relative risk for POAG and Harrell's C-statistic (the equivalent of an area-under-the-curve for longitudinal models). RESULTS: Among 1,046 cases and 38,809‬ controls, the relative risk (95% confidence interval) for POAG for participants in the highest PRS quintile was 3.99 (3.08, 5.18) in the US cohorts, and 4.89 (2.93, 8.17) in the Rotterdam Study, compared with participants with median genetic risk (3rd quintile). In restricted cubic spline analyses, the relation between continuous PRS and POAG risk increased exponentially in the US and Rotterdam cohorts (Pspline<0.05). Combining age, sex, intraocular pressure >25 mmHg, and family history resulted in a meta-analyzed concordance of 0.75 (0.73, 0.75). Adding the PRS to this model improved the concordance to 0.82 (0.80, 0.84). In a meta-analysis of all cohorts, cases in the highest tertile had a larger cup-disc ratio at diagnosis, by 0.11 (0.07, 0.15), and a 2.07-fold increased risk of requiring glaucoma surgery (1.19, 3.60). CONCLUSIONS: Incorporating a PRS into a POAG predictive model improves identification concordance from 0.75 up to 0.82, supporting its potential for guiding more cost-effective screening strategies.

Impact of Demographics on Regional Visual Field Loss and Deterioration in Glaucoma.

Purpose: To elucidate the impact of demographics, including gender, race, ethnicity, and preferred language, on regional visual field (VF) loss and progression in glaucoma. Methods: Multivariable linear mixed regressions were performed to determine the impact of race, ethnicity, and preferred language on regional VF loss with adjustment for age and gender. Regional VF loss was defined by pointwise total deviation values and VF loss patterns quantified by an unsupervised machine learning method termed archetypal analysis. All cross-sectional and longitudinal analyses were performed both without and with adjustment for VF mean deviation, which represented overall VF loss severity. P values were corrected for multiple comparisons. Results: All results mentioned had corrected P values less than 0.05. Asian and Black patients showed worse pointwise VF loss than White patients with superior hemifield more affected. Patients with a preferred language other than English demonstrated worse pointwise VF loss than patients with English as their preferred language. Longitudinal analyses revealed Black patients showed worse VF loss/year compared to White patients. Patients with a preferred language other than English demonstrated worse VF loss/year compared to patients preferring English. Conclusions: Blacks and non-English speakers have more severe VF loss, with superior hemifield being more affected and faster VF worsening. Translational Relevance: This study furthered our understanding of racial, ethnic, and socioeconomic disparities in glaucoma outcomes. Understanding the VF loss burden in different racial, ethnic, and socioeconomic groups may guide more effective glaucoma screening and community outreach efforts. This research could help reduce vision loss and improve quality of life in disproportionately affected populations by guiding public health efforts to promote glaucoma awareness and access to care.

Diagnostic Capability of OCTA-Derived Macular Biomarkers for Early to Moderate Primary Open Angle Glaucoma.

Background/Objectives: To investigate macular vascular biomarkers for the detection of primary open-angle glaucoma (POAG). Methods: A total of 56 POAG patients and 94 non-glaucomatous controls underwent optical coherence tomography angiography (OCTA) assessment of macular vessel density (VD) in the superficial (SCP), and deep (DCP) capillary plexus, foveal avascular zone (FAZ) area, perimeter, VD, choriocapillaris and outer retina flow area. POAG patients were classified for severity based on the Glaucoma Staging System 2 of Brusini. ANCOVA comparisons adjusted for age, sex, race, hypertension, diabetes, and areas under the receiver operating characteristic curves (AUCs) for POAG/control differentiation were compared using the DeLong method. Results: Global, hemispheric, and quadrant SCP VD was significantly lower in POAG patients in the whole image, parafovea, and perifovea (p < 0.001). No significant differences were found between POAG and controls for DCP VD, FAZ parameters, and the retinal and choriocapillaris flow area (p > 0.05). SCP VD in the whole image and perifovea were significantly lower in POAG patients in stage 2 than stage 0 (p < 0.001). The AUCs of SCP VD in the whole image (0.86) and perifovea (0.84) were significantly higher than the AUCs of all DCP VD (p < 0.05), FAZ parameters (p < 0.001), and retinal (p < 0.001) and choriocapillaris flow areas (p < 0.05). Whole image SCP VD was similar to the AUC of the global retinal nerve fiber layer (RNFL) (AUC = 0.89, p = 0.53) and ganglion cell complex (GCC) thickness (AUC = 0.83, p = 0.42). Conclusions: SCP VD is lower with increasing functional damage in POAG patients. The AUC for SCP VD was similar to RNFL and GCC using clinical diagnosis as the reference standard.

Classifying and quantifying changes in papilloedema using machine learning.

Background: Machine learning (ML) can differentiate papilloedema from normal optic discs using fundus photos. Currently, papilloedema severity is assessed using the descriptive, ordinal Frisén scale. We hypothesise that ML can quantify papilloedema and detect a treatment effect on papilloedema due to idiopathic intracranial hypertension. Methods: We trained a convolutional neural network to assign a Frisén grade to fundus photos taken from the Idiopathic Intracranial Hypertension Treatment Trial (IIHTT). We applied modified subject-based fivefold cross-validation to grade 2979 longitudinal images from 158 participants' study eyes (ie, the eye with the worst mean deviation) in the IIHTT. Compared with the human expert-determined grades, we hypothesise that ML-estimated grades can also demonstrate differential changes over time in the IIHTT study eyes between the treatment (acetazolamide (ACZ) plus diet) and placebo (diet only) groups. Findings: The average ML-determined grade correlated strongly with the reference standard (r=0.76, p<0.001; mean absolute error=0.54). At the presentation, treatment groups had similar expert-determined and ML-determined Frisén grades. The average ML-determined grade for the ACZ group (1.7, 95% CI 1.5 to 1.8) was significantly lower (p=0.0003) than for the placebo group (2.3, 95% CI 2.0 to 2.5) at the 6-month trial outcome. Interpretation: Supervised ML of fundus photos quantified the degree of papilloedema and changes over time reflecting the effects of ACZ. Given the increasing availability of fundus photography, neurologists will be able to use ML to quantify papilloedema on a continuous scale that incorporates the features of the Frisén grade to monitor interventions.

A Health-Related Quality of Life Measure for Patients Who Undergo Minimally Invasive Glaucoma Surgery.

PURPOSE: To develop a patient-reported outcome measure to assess the impact of glaucoma and treatment, including minimally invasive glaucoma surgery (MIGS). DESIGN: Observational study before and after concomitant cataract and Food and Drug Administration-approved implantable MIGS device surgery. SETTING: Survey administration was on a computer, iPad, or similar device. PATIENT POPULATION: 184 adults completed the baseline survey, 124 a survey 3 months after surgery, and 106 the 1-month test-retest reliability survey. The age range was 37 to 89 (average age = 72). Most were female (57%), non-Hispanic White (81%), and had a college degree (56%). MAIN OUTCOME MEASURES: The Glaucoma Outcomes Survey (GOS) assesses functional limitations (27 items), vision-related symptoms (7 items), psychosocial issues (7 items), and satisfaction with microinvasive glaucoma surgery (1 item). These multiple-item scales were scored on a 0 to 100 range, with a higher score indicating worse health. RESULTS: Internal consistency reliability estimates ranged from 0.75 to 0.93, and 1-month test-retest intraclass correlations ranged from 0.83 to 0.92 for the GOS scales. Product-moment correlations among the scales ranged from 0.56 to 0.60. Improvement in visual acuity in the study eye from baseline to the 3-month follow-up was significantly related to improvements in GOS functional limitations (r = 0.18, P = .0485), vision-related symptoms (r = 0.19, P = .0386), and psychosocial concerns (r = 0.18, P = .0503). Responders to treatment ranged from 17% for vision-related symptoms to 48% for functional limitations. CONCLUSIONS: This study supports using the GOS for ophthalmic procedures such as MIGS. Further evaluation of the GOS in different patient subgroups and clinical settings is needed.

Use of Diagnostic Codes for Primary Open-Angle Glaucoma Polygenic Risk Score Construction in Electronic Health Record-Linked Biobanks.

PURPOSE: Polygenic risk scores (PRSs) likely predict risk and prognosis of glaucoma. We compared the PRS performance for primary open-angle glaucoma (POAG), defined using International Classification of Diseases (ICD) codes vs manual medical record review. DESIGN: Retrospective cohort study. METHODS: We identified POAG cases in the Mount Sinai BioMe and Mass General Brigham (MGB) biobanks using ICD codes. We confirmed POAG based on optical coherence tomograms and visual fields. In a separate 5% sample, the absence of POAG was confirmed with intraocular pressure and cup-disc ratio criteria. We used genotype data and either self-reported glaucoma diagnoses or ICD-10 codes for glaucoma diagnoses from the UK Biobank and the lassosum method to compute a genome-wide POAG PRS. We compared the area under the curve (AUC) for POAG prediction based on ICD codes vs medical records. RESULTS: We reviewed 804 of 996 BioMe and 367 of 1006 MGB ICD-identified cases. In BioMe and MGB, respectively, positive predictive value was 53% and 55%; negative predictive value was 96% and 97%; sensitivity was 97% and 97%; and specificity was 44% and 53%. Adjusted PRS AUCs for POAG using ICD codes vs manual record review in BioMe were not statistically different (P ≥.21) by ancestry: 0.77 vs 0.75 for African, 0.80 vs 0.80 for Hispanic, and 0.81 vs 0.81 for European. Results were similar in MGB (P ≥.18): 0.72 vs 0.80 for African, 0.83 vs 0.86 for Hispanic, and 0.74 vs 0.73 for European. CONCLUSIONS: A POAG PRS performed similarly using either manual review or ICD codes in 2 electronic health record-linked biobanks; manual assessment of glaucoma status might not be necessary for some PRS studies. However, caution should be exercised when using ICD codes for glaucoma diagnosis given their low specificity (44%-53%) for manually confirmed cases of glaucoma.

The Evolving Surgical Paradigm of Scleral Allograft Bio-Tissue Use in Ophthalmic Surgery: Techniques and Clinical Indications for Ab-Externo and Ab-Interno Scleral Reinforcement.

To review the latest surgical advances and evolving clinical use of scleral bio-tissue for reinforcement in the eye and review the published literature on novel surgical applications of scleral allograft bio-tissue. Conventional surgical procedures for scleral reinforcement using homologous scleral allograft have been traditionally ab-externo interventions comprising of anterior or posterior reinforcement of the sclera for clinical indications such as trauma, scleromalacia, glaucoma drainage device coverage, scleral perforation, buckle repair as well as posterior reinforcement for pathologic myopia and staphyloma. There have been a few novel ab-interno uses of scleral bio-tissue for reinforcement in both retina and glaucoma. Over the last decade, there has been an increase in peer-reviewed publications on scleral reinforcement, reflecting more interest in its clinical applications. With favorable biological and biomechanical properties, scleral allograft may be an ideal substrate for an array of new applications and surgical uses.

The Association of Urinary Sodium Excretion with Glaucoma and Related Traits in a Large United Kingdom Population.

PURPOSE: Excessive dietary sodium intake has known adverse effects on intravascular fluid volume and systemic blood pressure, which may influence intraocular pressure (IOP) and glaucoma risk. This study aimed to assess the association of urinary sodium excretion, a biomarker of dietary intake, with glaucoma and related traits, and determine whether this relationship is modified by genetic susceptibility to disease. DESIGN: Cross-sectional observational and gene-environment interaction analyses in the population-based UK Biobank study. PARTICIPANTS: Up to 103 634 individuals (mean age: 57 years; 51% women) with complete urinary, ocular, and covariable data. METHODS: Urine sodium:creatinine ratio (UNa:Cr; mmol:mmol) was calculated from a midstream urine sample. Ocular parameters were measured as part of a comprehensive eye examination, and glaucoma case ascertainment was through a combination of self-report and linked national hospital records. Genetic susceptibility to glaucoma was calculated based on a glaucoma polygenic risk score comprising 2673 common genetic variants. Multivariable linear and logistic regression, adjusted for key sociodemographic, medical, anthropometric, and lifestyle factors, were used to model associations and gene-environment interactions. MAIN OUTCOME MEASURES: Corneal-compensated IOP, OCT derived macular retinal nerve fiber layer and ganglion cell-inner plexiform layer (GCIPL) thickness, and prevalent glaucoma. RESULTS: In maximally adjusted regression models, a 1 standard deviation increase in UNa:Cr was associated with higher IOP (0.14 mmHg; 95% confidence interval [CI], 0.12-0.17; P < 0.001) and greater prevalence of glaucoma (odds ratio, 1.11; 95% CI, 1.07-1.14; P < 0.001) but not macular retinal nerve fiber layer or ganglion cell-inner plexiform layer thickness. Compared with those with UNa:Cr in the lowest quintile, those in the highest quintile had significantly higher IOP (0.45 mmHg; 95% CI, 0.36-0.53, P < 0.001) and prevalence of glaucoma (odds ratio, 1.30; 95% CI, 1.17-1.45; P < 0.001). Stronger associations with glaucoma (P interaction = 0.001) were noted in participants with a higher glaucoma polygenic risk score. CONCLUSIONS: Urinary sodium excretion, a biomarker of dietary intake, may represent an important modifiable risk factor for glaucoma, especially in individuals at high underlying genetic risk. These findings warrant further investigation because they may have important clinical and public health implications. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Genome-first evaluation with exome sequence and clinical data uncovers underdiagnosed genetic disorders in a large healthcare system.

Population-based genomic screening may help diagnose individuals with disease-risk variants. Here, we perform a genome-first evaluation for nine disorders in 29,039 participants with linked exome sequences and electronic health records (EHRs). We identify 614 individuals with 303 pathogenic/likely pathogenic or predicted loss-of-function (P/LP/LoF) variants, yielding 644 observations; 487 observations (76%) lack a corresponding clinical diagnosis in the EHR. Upon further investigation, 75 clinically undiagnosed observations (15%) have evidence of symptomatic untreated disease, including familial hypercholesterolemia (3 of 6 [50%] undiagnosed observations with disease evidence) and breast cancer (23 of 106 [22%]). These genetic findings enable targeted phenotyping that reveals new diagnoses in previously undiagnosed individuals. Disease yield is greater with variants in penetrant genes for which disease is observed in carriers in an independent cohort. The prevalence of P/LP/LoF variants exceeds that of clinical diagnoses, and some clinically undiagnosed carriers are discovered to have disease. These results highlight the potential of population-based genomic screening.

Harvard Glaucoma Fairness: A Retinal Nerve Disease Dataset for Fairness Learning and Fair Identity Normalization.

Fairness (also known as equity interchangeably) in machine learning is important for societal well-being, but limited public datasets hinder its progress. Currently, no dedicated public medical datasets with imaging data for fairness learning are available, though underrepresented groups suffer from more health issues. To address this gap, we introduce Harvard Glaucoma Fairness (Harvard-GF), a retinal nerve disease dataset including 3,300 subjects with both 2D and 3D imaging data and balanced racial groups for glaucoma detection. Glaucoma is the leading cause of irreversible blindness globally with Blacks having doubled glaucoma prevalence than other races. We also propose a fair identity normalization (FIN) approach to equalize the feature importance between different identity groups. Our FIN approach is compared with various state-of-the-art fairness learning methods with superior performance in the racial, gender, and ethnicity fairness tasks with 2D and 3D imaging data, demonstrating the utilities of our dataset Harvard-GF for fairness learning. To facilitate fairness comparisons between different models, we propose an equity-scaled performance measure, which can be flexibly used to compare all kinds of performance metrics in the context of fairness. The dataset and code are publicly accessible via https://ophai.hms.harvard.edu/datasets/harvard-gf3300/.

Polygenic Risk Scores for Glaucoma Onset in the Ocular Hypertension Treatment Study.

Importance: Primary open-angle glaucoma (POAG) is a highly heritable disease, with 127 identified risk loci to date. Polygenic risk score (PRS) may provide a clinically useful measure of aggregate genetic burden and improve patient risk stratification. Objective: To assess whether a PRS improves prediction of POAG onset in patients with ocular hypertension. Design, Setting, and Participants: This was a post hoc analysis of the Ocular Hypertension Treatment Study. Data were collected from 22 US sites with a mean (SD) follow-up of 14.0 (6.9) years. A total of 1636 participants were followed up from February 1994 to December 2008; 1077 participants were enrolled in an ancillary genetics study, of which 1009 met criteria for this analysis. PRS was calculated using summary statistics from the largest cross-ancestry POAG meta-analysis, with weights trained using 8 813 496 variants from 449 186 cross-ancestry participants in the UK Biobank. Data were analyzed from July 2022 to December 2023. Exposures: From February 1994 to June 2002, participants were randomized to either topical intraocular pressure-lowering medication or close observation. After June 2002, both groups received medication. Main Outcomes and Measures: Outcome measures were hazard ratios for POAG onset. Concordance index and time-dependent areas under the receiver operating characteristic curve were used to compare the predictive performance of multivariable Cox proportional hazards models. Results: Of 1009 included participants, 562 (55.7%) were female, and the mean (SD) age was 55.9 (9.3) years. The mean (SD) PRS was significantly higher for 350 POAG converters (0.24 [0.95]) compared with 659 nonconverters (-0.12 [1.00]) (P < .001). POAG risk increased 1.36% (95% CI, 1.08-1.64) with each higher PRS decile, with conversion ranging from 9.52% (95% CI, 7.09-11.95) in the lowest PRS decile to 21.81% (95% CI, 19.37-24.25) in the highest decile. Comparison of low-risk and high-risk PRS tertiles showed a 2.0-fold increase in 20-year POAG risk for participants of European and African ancestries. In the subgroup randomized to delayed treatment, each increase in PRS decile was associated with a 0.52-year (95% CI, 0.01-1.03) decrease in age at diagnosis (P = .047). No significant linear association between PRS and age at POAG diagnosis was present in the early treatment group. Prediction models significantly improved with the addition of PRS as a covariate (C index = 0.77) compared with the Ocular Hypertension Treatment Study baseline model (C index = 0.75) (P < .001). Each 1-SD higher PRS conferred a mean hazard ratio of 1.25 (95% CI, 1.13-1.44) for POAG onset. Conclusions and Relevance: Higher PRS was associated with increased risk for POAG in patients with ocular hypertension. The inclusion of a PRS improved the prediction of POAG onset. Trial Registration: ClinicalTrials.gov Identifier: NCT00000125.

Relationship Between Oral Health and Glaucoma Traits in the United Kingdom.

PRCIS: In this cross-sectional analysis of UK Biobank participants, we find no adverse association between self-reported oral health conditions and either glaucoma or elevated intraocular pressures. PURPOSE: Poor oral health may cause inflammation, which accelerates the progression of neurodegenerative diseases. We investigated the relationship between oral health and glaucoma. PATIENTS: United Kingdom Biobank participants. METHODS: This is a cross-sectional analysis of participants categorized by self-reported oral health status. Multivariable linear and logistic regression models were used. Primary analysis examined the association with glaucoma prevalence. Secondary analyses examined associations with IOP, macular retinal nerve fiber layer (mRNFL), and ganglion cell inner plexiform layer (mGCIPL) thicknesses, and interaction terms with multitrait glaucoma polygenic risk scores (MTAG PRS) or intraocular pressure (IOP) PRS. RESULTS: A total of 170,815 participants (34.3%) reported current oral health problems, including painful or bleeding gums, toothache, loose teeth, and/or denture wear. A In all, 33,059, 33,004, 14,652, and 14,613 participants were available for analysis of glaucoma, IOP, mRNFL, and mGCIPL, respectively. No association between oral health and glaucoma was identified [odds ratio (OR): 1.04, 95% CI: 0.95-1.14]. IOPs were slightly lower among those with oral disease (-0.08 mm Hg, 95% CI: -0.15, -0.009); specifically, among those with loose teeth ( P =0.03) and denture-wearers ( P <0.0001). mRNFL measurements were lower among those with oral health conditions (-0.14 µm, 95% CI: -0.27, -0.0009), but mGCIPL measurements ( P =0.96) were not significantly different. A PRS for IOP or glaucoma did not modify relations between oral health and IOP or glaucoma ( P for interactions ≥​​​​0.17). CONCLUSIONS: Self-reported oral health was not associated with elevated IOP or an increased risk of glaucoma. Future studies should confirm the null association between clinically diagnosed oral health conditions and glaucoma.

Assessment of a Large Language Model's Responses to Questions and Cases About Glaucoma and Retina Management.

Importance: Large language models (LLMs) are revolutionizing medical diagnosis and treatment, offering unprecedented accuracy and ease surpassing conventional search engines. Their integration into medical assistance programs will become pivotal for ophthalmologists as an adjunct for practicing evidence-based medicine. Therefore, the diagnostic and treatment accuracy of LLM-generated responses compared with fellowship-trained ophthalmologists can help assess their accuracy and validate their potential utility in ophthalmic subspecialties. Objective: To compare the diagnostic accuracy and comprehensiveness of responses from an LLM chatbot with those of fellowship-trained glaucoma and retina specialists on ophthalmological questions and real patient case management. Design, Setting, and Participants: This comparative cross-sectional study recruited 15 participants aged 31 to 67 years, including 12 attending physicians and 3 senior trainees, from eye clinics affiliated with the Department of Ophthalmology at Icahn School of Medicine at Mount Sinai, New York, New York. Glaucoma and retina questions (10 of each type) were randomly selected from the American Academy of Ophthalmology's commonly asked questions Ask an Ophthalmologist. Deidentified glaucoma and retinal cases (10 of each type) were randomly selected from ophthalmology patients seen at Icahn School of Medicine at Mount Sinai-affiliated clinics. The LLM used was GPT-4 (version dated May 12, 2023). Data were collected from June to August 2023. Main Outcomes and Measures: Responses were assessed via a Likert scale for medical accuracy and completeness. Statistical analysis involved the Mann-Whitney U test and the Kruskal-Wallis test, followed by pairwise comparison. Results: The combined question-case mean rank for accuracy was 506.2 for the LLM chatbot and 403.4 for glaucoma specialists (n = 831; Mann-Whitney U = 27976.5; P < .001), and the mean rank for completeness was 528.3 and 398.7, respectively (n = 828; Mann-Whitney U = 25218.5; P < .001). The mean rank for accuracy was 235.3 for the LLM chatbot and 216.1 for retina specialists (n = 440; Mann-Whitney U = 15518.0; P = .17), and the mean rank for completeness was 258.3 and 208.7, respectively (n = 439; Mann-Whitney U = 13123.5; P = .005). The Dunn test revealed a significant difference between all pairwise comparisons, except specialist vs trainee in rating chatbot completeness. The overall pairwise comparisons showed that both trainees and specialists rated the chatbot's accuracy and completeness more favorably than those of their specialist counterparts, with specialists noting a significant difference in the chatbot's accuracy (z = 3.23; P = .007) and completeness (z = 5.86; P < .001). Conclusions and Relevance: This study accentuates the comparative proficiency of LLM chatbots in diagnostic accuracy and completeness compared with fellowship-trained ophthalmologists in various clinical scenarios. The LLM chatbot outperformed glaucoma specialists and matched retina specialists in diagnostic and treatment accuracy, substantiating its role as a promising diagnostic adjunct in ophthalmology.

Integrating genetic regulation and single-cell expression with GWAS prioritizes causal genes and cell types for glaucoma.

Primary open-angle glaucoma (POAG), characterized by retinal ganglion cell death, is a leading cause of irreversible blindness worldwide. However, its molecular and cellular causes are not well understood. Elevated intraocular pressure (IOP) is a major risk factor, but many patients have normal IOP. Colocalization and Mendelian randomization analysis of >240 POAG and IOP genome-wide association study (GWAS) loci and overlapping expression and splicing quantitative trait loci (e/sQTLs) in 49 GTEx tissues and retina prioritizes causal genes for 60% of loci. These genes are enriched in pathways implicated in extracellular matrix organization, cell adhesion, and vascular development. Analysis of single-nucleus RNA-seq of glaucoma-relevant eye tissues reveals that the POAG and IOP colocalizing genes and genome-wide associations are enriched in specific cell types in the aqueous outflow pathways, retina, optic nerve head, peripapillary sclera, and choroid. This study nominates IOP-dependent and independent regulatory mechanisms, genes, and cell types that may contribute to POAG pathogenesis.