Evaluating the Relationship between Neighborhood-Level Social Vulnerability and Patient Adherence to Ophthalmology Appointments.

PURPOSE: To examine the association between neighborhood-level social vulnerability and adherence to scheduled ophthalmology appointments. METHODS: In this retrospective cohort study, records of all patients ≥18 years scheduled for an ophthalmology appointment between September 12, 2020, and February 8, 2021, were reviewed. Primary exposure is neighborhood-level Social Vulnerability Index (SVI) based on the patient's residential location. SVI is a rank score of 15 social factors into four themes (socioeconomic status, household composition/disability, minority status/language, and housing type/transportation), ranging from 0 to 1.0, with higher ranks indicating greater social vulnerability. The overall SVI score and each theme were analyzed separately as the primary exposure of interest in multivariable logistic regression models that controlled for age, sex, appointment status (new or established), race, and distance from clinic. The primary outcome, non-adherence, was defined as missing more than 25% of scheduled appointments. RESULTS: Of 8,322 patients (41% non-Hispanic Black, 24% Hispanic, 22% non-Hispanic White) with scheduled appointments, 28% were non-adherent. Non-adherence was associated with greater social vulnerability (adjusted odds ratio [aOR] per 0.01 increase in overall SVI = 2.46 [95% confidence interval, 1.99, 3.06]) and each SVI theme (socioeconomic status: aOR = 2.38 [1.94, 2.91]; household composition/disability: aOR = = 1.51 [1.26, 1.81]; minority status/language: aOR = 2.03 [1.55, 2.68]; housing type/transportation: aOR = 1.41 [1.16, 1.73]). CONCLUSION: Neighborhood-level social vulnerability is associated with greater risk of non-adherence to scheduled ophthalmology appointments, controlling for individual characteristics. Multi-level intervention strategies that incorporate neighborhood-level vulnerabilities are needed to reduce disparities in access to ophthalmology care.

Evaluation of the Impact of Meibomian Gland Dysfunction Using a Novel Patient-Reported Outcome Instrument.

Purpose: This study was intended to characterize the impact of meibomian gland dysfunction (MGD) on patients' quality of life. Methods: In this prospective, multicenter, noninterventional clinical study (NCT01979887), eligible individuals (age ≥40 years; absence of uncontrolled ocular/systemic disease) were categorized, based on composite grading of ocular symptoms, Schirmer score, and meibum quality, into (1) non-MGD, (2) mild/moderate MGD, or (3) severe MGD cohorts. The MGD Impact Questionnaire (MGD IQ), a 10-item patient-reported outcome measure, was self-administered at clinic visit on day 1, and readministered on day 22 to assess intervisit agreement regarding MGD IQ responses. Results: In total, 75 subjects were assigned to the study cohorts (25 per cohort). Across cohorts, MGD IQ item scores rose incrementally with increasing MGD severity. The severe MGD cohort experienced greater difficulty with reading and performance of leisure activities, greater time on eye care, and greater bother with eye care and eye appearance than the mild/moderate MGD cohort (all P < 0.05). Compared with the non-MGD cohort, the mild/moderate MGD cohort had greater difficulty working on computer, whereas the severe MGD cohort had greater difficulty reading, driving, and performing leisure activities, more frequent difficulty with outdoor activities, more time on eye care, and greater bother with eye care (all P < 0.05). Intervisit agreement between MGD IQ responses was fair to moderate (weighted kappa statistic 0.33‒0.58). Conclusions: Vision-related activities are negatively impacted by increasing severity of MGD. The MGD IQ instrument can help characterize disease severity and amplify the patient's voice in patient-centric clinical research. ClinicalTrials.gov NCT01979887.

Real-world treatment patterns and economic burden of post-cataract macular edema.

BACKGROUND: Post-cataract macular edema (PCME) is a condition that can occur in patients following cataract surgery without risk factors and complications. Although 80% of patients experience spontaneous resolution after 3 to 12 months, in persistent cases, it can lead to permanent vision loss if left untreated. There are currently no standardized treatment guidelines for PCME, and there have been limited studies showing the impact of PCME on annual Medicare spending and ophthalmology-related outpatient visits per case compared to those without the complication. This study aims to evaluate real-world treatment patterns and the economic burden of patients with PCME. METHODS: This retrospective claims analysis identified patients from the IBM® MarketScan® Commercial and Medicare Supplemental databases. Patients with (n = 2430) and without (n = 7290) PCME 1 year post cataract surgery were propensity score matched 1:3 based on age, geographic region, diabetes presence, cataract surgery type, and Charlson Comorbidity Index. Treatment pattern analysis for each PCME patient summarized the distribution of medications across lines of therapy. Economic burden analysis compared the mean number and costs of eye-related outpatient visits, optical coherence tomography imaging scans, and ophthalmic medications between the 2 groups using linear regression models. RESULTS: Treatment pattern analysis found 27 different treatment combinations across 6 treatment lines. The most common first-line treatments were topical steroid drops (372 [30%]), topical nonsteroidal anti-inflammatory drug drops (321 [27%]), and intraocular or periocular injectable steroids (189 [15%]). Compared to match controls, PCME patients averaged 6 additional eye-related outpatient office visits (95% CI: 5.7-6.2) resulting in an additional $3,897 (95% CI: $3,475 - $4,319) in total costs. Patients filled 3 more ophthalmology-related outpatient prescription medications (95% CI: 2.8-3.2), adding $371 in total cost (95% CI: $332 - $410). CONCLUSIONS: PCME treatment patterns showed wide clinical variability in treatments and time, specifically regarding injectable treatments and combination therapy. Additionally, significantly higher healthcare resource use and economic burden were found for both patients and payers when comparing PCME patients to non-PMCE controls. These results highlight the need for treatment standardization and demonstrate that interventions targeted at preventing PCME may be valuable.

Contrastive learning-based pretraining improves representation and transferability of diabetic retinopathy classification models.

Diabetic retinopathy (DR) is a major cause of vision impairment in diabetic patients worldwide. Due to its prevalence, early clinical diagnosis is essential to improve treatment management of DR patients. Despite recent demonstration of successful machine learning (ML) models for automated DR detection, there is a significant clinical need for robust models that can be trained with smaller cohorts of dataset and still perform with high diagnostic accuracy in independent clinical datasets (i.e., high model generalizability). Towards this need, we have developed a self-supervised contrastive learning (CL) based pipeline for classification of referable vs non-referable DR. Self-supervised CL based pretraining allows enhanced data representation, therefore, the development of robust and generalized deep learning (DL) models, even with small, labeled datasets. We have integrated a neural style transfer (NST) augmentation in the CL pipeline to produce models with better representations and initializations for the detection of DR in color fundus images. We compare our CL pretrained model performance with two state of the art baseline models pretrained with Imagenet weights. We further investigate the model performance with reduced labeled training data (down to 10 percent) to test the robustness of the model when trained with small, labeled datasets. The model is trained and validated on the EyePACS dataset and tested independently on clinical datasets from the University of Illinois, Chicago (UIC). Compared to baseline models, our CL pretrained FundusNet model had higher area under the receiver operating characteristics (ROC) curve (AUC) (CI) values (0.91 (0.898 to 0.930) vs 0.80 (0.783 to 0.820) and 0.83 (0.801 to 0.853) on UIC data). At 10 percent labeled training data, the FundusNet AUC was 0.81 (0.78 to 0.84) vs 0.58 (0.56 to 0.64) and 0.63 (0.60 to 0.66) in baseline models, when tested on the UIC dataset. CL based pretraining with NST significantly improves DL classification performance, helps the model generalize well (transferable from EyePACS to UIC data), and allows training with small, annotated datasets, therefore reducing ground truth annotation burden of the clinicians.

An Association Between Large Optic Cupping and Total and Regional Brain Volume: The Women's Health Initiative.

PURPOSE: To investigate the relationships between optic nerve cupping and total and regional brain volumes. DESIGN: Secondary analysis of randomized clinical trial data. METHODS: Women 65 to 79 years of age without glaucoma with cup-to-disc ratio (CDR) measurements from the Women's Health Initiative (WHI) Sight Examination study and magnetic resonance imaging (MRI)-based total and regional brain volumes from the WHI Memory Study MRI-1 were included. Large CDR was defined as 0.6 or greater in either eye. Generalized estimating equation models were used to account for intra-brain correlations between the right and left sides. The final analysis was adjusted for demographic and clinical characteristics and for total brain volume (for regional analyses). RESULTS: Final analyses included 471 women, with the mean age ± SD was 69.2 ± 3.6 years; 92.8% of the subjects were white. Of 471 women, 34 (7.2%) had large CDR. Controlling for total brain volume and for demographic and clinical characteristics, lateral ventricle volume was 3.01 cc larger for subjects with large CDR compared to those without large CDR (95% CI = 0.02 to 5.99; P = .048). Furthermore, frontal lobe volume was 4.78 cc lower for subjects with large CDR compared to those without (95% CI = -8.71, -0.84; P = 0.02), and occipital lobe volume was 1.86 cc lower for those with large CDR compared to those without (95% CI = -3.39, -0.3; P =.02). CONCLUSIONS: Our analysis suggests that in women aged 65 years or more, large CDR is associated with lower relative total brain volume and absolute regional volume in the frontal and occipital lobes. Enlarged CDR in individuals without glaucoma may represent a sign of optic nerve and brain aging, although more longitudinal data are needed.

Characterization of Errors in Retinopathy of Prematurity Diagnosis by Ophthalmologists-in-Training in the United States and Canada.

PURPOSE: To identify the prominent factors that lead to misdiagnosis of retinopathy of prematurity (ROP) by ophthalmologists-in-training in the United States and Canada. METHODS: This prospective cohort study included 32 ophthalmologists-in-training at six ophthalmology training programs in the United States and Canada. Twenty web-based cases of ROP using wide-field retinal images were presented, and ophthalmologists-in-training were asked to diagnose plus disease, zone, stage, and category for each eye. Responses were compared to a consensus reference standard diagnosis for accuracy, which was established by combining the clinical diagnosis and the image-based diagnosis by multiple experts. The types of diagnostic errors that occurred were analyzed with descriptive and chi-squared analysis. Main outcome measures were frequency of types (category, zone, stage, plus disease) of diagnostic errors; association of errors in zone, stage, and plus disease diagnosis with incorrectly identified category; and performance of ophthalmologists-in-training across postgraduate years. RESULTS: Category of ROP was misdiagnosed at a rate of 48%. Errors in classification of plus disease were most commonly associated with misdiagnosis of treatment-requiring (plus error rate = 16% when treatment-requiring was correctly diagnosed vs 81% when underdiagnosed as type 2 or pre-plus; mean difference: 64.3; 95% CI: 51.9 to 76.7; P < .001) and type 2 or pre-plus (plus error rate = 35% when type 2 or pre-plus was correctly diagnosed vs 76% when overdiagnosed as treatment-requiring; mean difference: 41.0; 95% CI: 28.4 to 53.5; P < .001) disease. The diagnostic error rate of postgraduate year (PGY)-2 trainees was significantly higher than PGY-3 trainees (PGY-2 category error rate = 61% vs PGY-3 = 35%; mean difference, 25.4; 95% CI: 17.7 to 33.0; P < .001). CONCLUSIONS: Ophthalmologists-in-training in the United States and Canada misdiagnosed ROP nearly half of the time, with incorrect identification of plus disease as a leading cause. Integration of structured learning for ROP in residency education may improve diagnostic competency. [J Pediatr Ophthalmol Strabismus. 2023;60(5):337-343.].

Automated Region of Interest Selection Improves Deep Learning-Based Segmentation of Hyper-Reflective Foci in Optical Coherence Tomography Images.

Hyperreflective foci (HRF) have been associated with retinal disease progression and demonstrated as a negative prognostic biomarker for visual function. Automated segmentation of HRF in retinal optical coherence tomography (OCT) scans can be beneficial to identify the formation and movement of the HRF biomarker as a retinal disease progresses and can serve as the first step in understanding the nature and severity of the disease. In this paper, we propose a fully automated deep neural network based HRF segmentation model in OCT images. We enhance the model's performance by using a patch-based strategy that increases the model's compute on the HRF pixels. The patch-based strategy is evaluated against state of the art HRF segmentation pipelines on clinical retinal image data. Our results shows that the patch-based approach demonstrates a high precision score and intersection over union (IOU) using a ResNet34 segmentation model with Binary Cross Entropy loss function. The HRF segmentation pipeline can be used for analyzing HRF biomarkers for different retinopathies.

Association between myopia progression and quantity of laser treatment for retinopathy of prematurity.

BACKGROUND: Previous studies found that infants with retinopathy of prematurity (ROP) who were treated for more posterior disease with a greater number of laser spots developed higher myopia. These studies included multiple physicians with variations in laser density. In treatments by a single physician, laser spot count is a better surrogate for area of avascular retina and anterior-posterior location of disease, so that the relationship with myopia can be better assessed. METHODS: Our retrospective study included infants treated with laser for ROP by a single surgeon at a single center. Exclusion criteria were irregularities during laser and additional treatment for ROP. We assessed correlation between laser spot count and change in refractive error over time using a linear mixed effects model. RESULTS: We studied 153 eyes from 78 subjects treated with laser for ROP. The average gestational age at birth was 25.3±1.8 weeks, birth weight 737±248 grams, laser spot count 1793±728, and post-treatment follow up 37±29 months. Between corrected ages 0-1 years, the mean spherical equivalent was +0.4±2.3 diopters; between ages 1-2, it was -1.3±3.2D; and ages 2-3 was -0.8±3.1D. Eyes that received more laser spots had significantly greater change in refractive error over time (0.30D more myopia per year per 1000 spots). None of the eyes with hyperopia before 18 months developed myopia during the follow-up period. CONCLUSIONS: Greater myopia developed over time in infants with ROP treated by laser to a larger area of avascular retina.

Evaluation of an Artificial Intelligence System for Retinopathy of Prematurity Screening in Nepal and Mongolia.

Purpose: To evaluate the performance of a deep learning (DL) algorithm for retinopathy of prematurity (ROP) screening in Nepal and Mongolia. Design: Retrospective analysis of prospectively collected clinical data. Participants: Clinical information and fundus images were obtained from infants in 2 ROP screening programs in Nepal and Mongolia. Methods: Fundus images were obtained using the Forus 3nethra neo (Forus Health) in Nepal and the RetCam Portable (Natus Medical, Inc.) in Mongolia. The overall severity of ROP was determined from the medical record using the International Classification of ROP (ICROP). The presence of plus disease was determined independently in each image using a reference standard diagnosis. The Imaging and Informatics for ROP (i-ROP) DL algorithm was trained on images from the RetCam to classify plus disease and to assign a vascular severity score (VSS) from 1 through 9. Main Outcome Measures: Area under the receiver operating characteristic curve and area under the precision-recall curve for the presence of plus disease or type 1 ROP and association between VSS and ICROP disease category. Results: The prevalence of type 1 ROP was found to be higher in Mongolia (14.0%) than in Nepal (2.2%; P < 0.001) in these data sets. In Mongolia (RetCam images), the area under the receiver operating characteristic curve for examination-level plus disease detection was 0.968, and the area under the precision-recall curve was 0.823. In Nepal (Forus images), these values were 0.999 and 0.993, respectively. The ROP VSS was associated with ICROP classification in both datasets (P < 0.001). At the population level, the median VSS was found to be higher in Mongolia (2.7; interquartile range [IQR], 1.3-5.4]) as compared with Nepal (1.9; IQR, 1.2-3.4; P < 0.001). Conclusions: These data provide preliminary evidence of the effectiveness of the i-ROP DL algorithm for ROP screening in neonatal populations in Nepal and Mongolia using multiple camera systems and are useful for consideration in future clinical implementation of artificial intelligence-based ROP screening in low- and middle-income countries.

Patient Adherence to Immunosuppressive Therapy in Treatment of Chronic Inflammatory Eye Disease.

TITLE: Patient Adherence to Immunosuppressive Therapy for Chronic Inflammatory Eye Disease. PURPOSE: To investigate adherence rates to immunosuppressive therapy (IMT) for treatment of noninfectious inflammatory eye disease (IED), adherence and disease control, and factors associated with nonadherence. METHOD: Retrospective review of medical charts from 2015 to 2020 was conducted on patients with IED at 6 months, 1 and 2 years after initiation of IMT. RESULTS: Of 183 patients, adherence rates at 6 months and 1 year were 70% and 58% by 2 years. Eighty-two percent, 78%, and 65% of patients with disease quiescence were adherent at 6 months, 1 and 2 years, respectively. Adherent patients have 1.86 (95% CI 1.09, 3.20) times greater likelihood for disease control compared to nonadherent. Primary reason for nonadherence was patient self-discontinuation. No specific factors were associated with nonadherence. CONCLUSION: Patients on IMT for IED had steady adherence rates up to 1 year, with decreased adherence at 2 years. Adherence to IMT significantly correlates with disease quiescence.

Association between cognitive function and large optic nerve cupping, accounting for cup-disc-ratio genetic risk score.

PURPOSE: To investigate if accounting for a cup-to-disc ratio (CDR) genetic risk score (GRS) modified the association between large CDR and cognitive function among women. DESIGN: This was a retrospective study using data from the Women's Health Initiative. METHODS: Patients with glaucoma or ocular hypertension were excluded. Large CDR was defined as ≥ 0.6 in either eye. Cognitive function was measured by the Modified Mini-Mental State Examination (3MSE). We used the combined effects from 13 single nucleotide polymorphisms (SNPs) to formulate the GRS for CDR. We used logistic regression to investigate associations between weighted GRS and large CDR, then a linear regression to assess the association between weighted GRS and 3MSE scores, and between weighted GRS, CDR, and 3MSE scores, adjusted for demographic and clinical characteristics. RESULTS: Final analyses included 1,196 White women with mean age of 69.60 ± 3.62 years and 7.27% with large CDR. Mean GRS in women with and without large CDR was 1.51 ± 0.31 vs. 1.41 ± 0.36, respectively (p = 0.004). The odds of large CDR for a one unit increase in GRS was 2.30 (95% CI: (1.22, 4.36), p = 0.011). Adding the CDR GRS in the model with CDR and 3MSE, women with large CDR still had statistically significantly lower 3MSE scores than those without large CDR, yielding a predicted mean difference in 3MSE scores of 0.84 (p = 0.007). CONCLUSIONS: Independent of the CDR GRS, women with large CDR had a lower cognitive function.

Optometrist's perspectives of Artificial Intelligence in eye care.

PURPOSE: The application of artificial intelligence (AI) in diagnosing and managing ocular disease has gained popularity as research highlights the utilization of AI to improve personalized medicine and healthcare outcomes. The objective of this study is to describe current optometric perspectives of AI in eye care. METHODS: Members of the American Academy of Optometry were sent an electronic invitation to complete a 17-item survey. Survey items assessed perceived advantages and concerns regarding AI using a 5-point Likert scale ranging from "strongly agree" to "strongly disagree." RESULTS: A total of 400 optometrists completed the survey. The mean number of years since optometry school completion was 25 ± 15.1. Most respondents reported familiarity with AI (66.8%). Though half of optometrists had concerns about the diagnostic accuracy of AI (53.0%), most believed it would improve the practice of optometry (72.0%). Optometrists reported their willingness to incorporate AI into practice increased from 53.3% before the COVID-19 pandemic to 65.5% after onset of the pandemic (p<0.001). CONCLUSION: In this study, optometrists are optimistic about the use of AI in eye care, and willingness to incorporate AI in clinical practice also increased after the onset of the COVID-19 pandemic.

External Validation of a Retinopathy of Prematurity Screening Model Using Artificial Intelligence in 3 Low- and Middle-Income Populations.

Importance: Retinopathy of prematurity (ROP) is a leading cause of preventable blindness that disproportionately affects children born in low- and middle-income countries (LMICs). In-person and telemedical screening examinations can reduce this risk but are challenging to implement in LMICs owing to the multitude of at-risk infants and lack of trained ophthalmologists. Objective: To implement an ROP risk model using retinal images from a single baseline examination to identify infants who will develop treatment-requiring (TR)-ROP in LMIC telemedicine programs. Design, Setting, and Participants: In this diagnostic study conducted from February 1, 2019, to June 30, 2021, retinal fundus images were collected from infants as part of an Indian ROP telemedicine screening program. An artificial intelligence (AI)-derived vascular severity score (VSS) was obtained from images from the first examination after 30 weeks' postmenstrual age. Using 5-fold cross-validation, logistic regression models were trained on 2 variables (gestational age and VSS) for prediction of TR-ROP. The model was externally validated on test data sets from India, Nepal, and Mongolia. Data were analyzed from October 20, 2021, to April 20, 2022. Main Outcomes and Measures: Primary outcome measures included sensitivity, specificity, positive predictive value, and negative predictive value for predictions of future occurrences of TR-ROP; the number of weeks before clinical diagnosis when a prediction was made; and the potential reduction in number of examinations required. Results: A total of 3760 infants (median [IQR] postmenstrual age, 37 [5] weeks; 1950 male infants [51.9%]) were included in the study. The diagnostic model had a sensitivity and specificity, respectively, for each of the data sets as follows: India, 100.0% (95% CI, 87.2%-100.0%) and 63.3% (95% CI, 59.7%-66.8%); Nepal, 100.0% (95% CI, 54.1%-100.0%) and 77.8% (95% CI, 72.9%-82.2%); and Mongolia, 100.0% (95% CI, 93.3%-100.0%) and 45.8% (95% CI, 39.7%-52.1%). With the AI model, infants with TR-ROP were identified a median (IQR) of 2.0 (0-11) weeks before TR-ROP diagnosis in India, 0.5 (0-2.0) weeks before TR-ROP diagnosis in Nepal, and 0 (0-5.0) weeks before TR-ROP diagnosis in Mongolia. If low-risk infants were never screened again, the population could be effectively screened with 45.0% (India, 664/1476), 38.4% (Nepal, 151/393), and 51.3% (Mongolia, 266/519) fewer examinations required. Conclusions and Relevance: Results of this diagnostic study suggest that there were 2 advantages to implementation of this risk model: (1) the number of examinations for low-risk infants could be reduced without missing cases of TR-ROP, and (2) high-risk infants could be identified and closely monitored before development of TR-ROP.

Assessment of a novel ophthalmology tele-triage system during the COVID-19 pandemic.

BACKGROUND: In response to the COVID-19 pandemic, a web-based tele-triage system was created to prioritize in-person clinic visits and ensure safety at the University of Illinois at Chicago Department of Ophthalmology and Visual Sciences during a statewide shelter-in-place order. The aim of this study is to evaluate the impact of the tele-triage system on urgent visit volume and explore the characteristics of acute visit requests at a tertiary referral eye center. METHODS: This retrospective study analyzed acute visit requests between April 6, 2020 and June 6, 2020. Descriptive statistics, chi-square tests, ANOVA, and bivariate logistic regression were used to compare variables with a p-value of 0.05. RESULTS: Three hundred fifty-eight surveys were completed. Mean age was 49.7 ± 18.8 years (range 2-91). The majority of requests were determined as urgent (63.0%) or emergent (0.8%). Forty-nine patients had recent eye trauma (13.7%), and the most common reported symptoms were new onset eye pain (25.7%) and photophobia (22.9%). Most patients were self-referred (63.7%), though provider referral was more common in patients with symptoms of new onset lid swelling (p < 0.01), diplopia (p < 0.01), flashing lights (p = 0.02), or droopy eyelid (p < 0.01). Patients presenting with symptom onset within 48 h tended to be younger (45.8 years) versus those with symptom duration of 48 h to 1 week (49.6 years), or more than 1 week (52.6 years; p < 0.01). CONCLUSION: This novel tele-triage system screened out one-third of acute visit requests as non-urgent, which limited in-person visits during the initial shelter-in-place period of the pandemic. Tele-triage systems should be implemented in eye care practices for future emergency preparedness.

Imaging and artificial intelligence for progression of age-related macular degeneration.

Age-related macular degeneration (AMD) is a leading cause of severe vision loss. With our aging population, it may affect 288 million people globally by the year 2040. AMD progresses from an early and intermediate dry form to an advanced one, which manifests as choroidal neovascularization and geographic atrophy. Conversion to AMD-related exudation is known as progression to neovascular AMD, and presence of geographic atrophy is known as progression to advanced dry AMD. AMD progression predictions could enable timely monitoring, earlier detection and treatment, improving vision outcomes. Machine learning approaches, a subset of artificial intelligence applications, applied on imaging data are showing promising results in predicting progression. Extracted biomarkers, specifically from optical coherence tomography scans, are informative in predicting progression events. The purpose of this mini review is to provide an overview about current machine learning applications in artificial intelligence for predicting AMD progression, and describe the various methods, data-input types, and imaging modalities used to identify high-risk patients. With advances in computational capabilities, artificial intelligence applications are likely to transform patient care and management in AMD. External validation studies that improve generalizability to populations and devices, as well as evaluating systems in real-world clinical settings are needed to improve the clinical translations of artificial intelligence AMD applications.

Probabilistic Forecasting of Anti-VEGF Treatment Frequency in Neovascular Age-Related Macular Degeneration.

Purpose: To probabilistically forecast needed anti-vascular endothelial growth factor (anti-VEGF) treatment frequency using volumetric spectral domain-optical coherence tomography (SD-OCT) biomarkers in neovascular age-related macular degeneration from real-world settings. Methods: SD-OCT volume scans were segmented with a custom deep-learning-based analysis pipeline. Retinal thickness and reflectivity values were extracted for the central and the four inner Early Treatment Diabetic Retinopathy Study (ETDRS) subfields for six retinal layers (inner retina, outer nuclear layer, inner segments [IS], outer segments [OS], retinal pigment epithelium-drusen complex [RPEDC] and the choroid). Machine-learning models were probed to predict the anti-VEGF treatment frequency within the next 12 months. Probabilistic forecasting was performed using natural gradient boosting (NGBoost), which outputs a full probability distribution. The mean absolute error (MAE) between the predicted versus actual anti-VEGF treatment frequency was the primary outcome measure. Results: In a total of 138 visits of 99 eyes with neovascular AMD (96 patients) from two clinical centers, the prediction of future anti-VEGF treatment frequency was observed with an accuracy (MAE [95% confidence interval]) of 2.60 injections/year [2.25-2.96] (R2 = 0.390) using random forest regression and 2.66 injections/year [2.31-3.01] (R2 = 0.094) using NGBoost, respectively. Prediction intervals were well calibrated and reflected the true uncertainty of NGBoost-based predictions. Standard deviation of RPEDC-thickness in the central ETDRS-subfield constituted an important predictor across models. Conclusions: The proposed, fully automated pipeline enables probabilistic forecasting of future anti-VEGF treatment frequency in real-world settings. Translational Relevance: Prediction of a probability distribution allows the physician to inspect the underlying uncertainty. Predictive uncertainty estimates are essential to highlight cases where human-inspection and/or reversion to a fallback alternative is warranted.

Evaluation of pediatric ophthalmologists' perspectives of artificial intelligence in ophthalmology.

PURPOSE: To survey pediatric ophthalmologists on their perspectives of artificial intelligence (AI) in ophthalmology. METHODS: This is a subgroup analysis of a study previously reported. In March 2019, members of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) were recruited via the online AAPOS discussion board to voluntarily complete a Web-based survey consisting of 15 items. Survey items assessed the extent participants "agreed" or "disagreed" with statements on the perceived benefits and concerns of AI in ophthalmology. Responses were analyzed using descriptive statistics. RESULTS: A total of 80 pediatric ophthalmologists who are members of AAPOS completed the survey. The mean number of years since graduating residency was 21 years (range, 0-46). Overall, 91% (73/80) reported understanding the concept of AI, 70% (56/80) believed AI will improve the practice of ophthalmology, 68% (54/80) reported willingness to incorporate AI into their clinical practice, 65% (52/80) did not believe AI will replace physicians, and 71% (57/80) believed AI should be incorporated into medical school and residency curricula. However, 15% (12/80) were concerned that AI will replace physicians, 26% (21/80) believed AI will harm the patient-physician relationship, and 46% (37/80) reported concern over the diagnostic accuracy of AI. CONCLUSIONS: Most pediatric ophthalmologists in this survey viewed the role of AI in ophthalmology positively.