A foundation model for generalizable disease detection from retinal images.

Medical artificial intelligence (AI) offers great potential for recognizing signs of health conditions in retinal images and expediting the diagnosis of eye diseases and systemic disorders1. However, the development of AI models requires substantial annotation and models are usually task-specific with limited generalizability to different clinical applications2. Here, we present RETFound, a foundation model for retinal images that learns generalizable representations from unlabelled retinal images and provides a basis for label-efficient model adaptation in several applications. Specifically, RETFound is trained on 1.6 million unlabelled retinal images by means of self-supervised learning and then adapted to disease detection tasks with explicit labels. We show that adapted RETFound consistently outperforms several comparison models in the diagnosis and prognosis of sight-threatening eye diseases, as well as incident prediction of complex systemic disorders such as heart failure and myocardial infarction with fewer labelled data. RETFound provides a generalizable solution to improve model performance and alleviate the annotation workload of experts to enable broad clinical AI applications from retinal imaging.

Developing a Continuous Severity Scale for Macular Telangiectasia Type 2 Using Deep Learning and Implications for Disease Grading.

PURPOSE: Deep learning (DL) models have achieved state-of-the-art medical diagnosis classification accuracy. Current models are limited by discrete diagnosis labels, but could yield more information with diagnosis in a continuous scale. We developed a novel continuous severity scaling system for macular telangiectasia (MacTel) type 2 by combining a DL classification model with uniform manifold approximation and projection (UMAP). DESIGN: We used a DL network to learn a feature representation of MacTel severity from discrete severity labels and applied UMAP to embed this feature representation into 2 dimensions, thereby creating a continuous MacTel severity scale. PARTICIPANTS: A total of 2003 OCT volumes were analyzed from 1089 MacTel Project participants. METHODS: We trained a multiview DL classifier using multiple B-scans from OCT volumes to learn a previously published discrete 7-step MacTel severity scale. The classifiers' last feature layer was extracted as input for UMAP, which embedded these features into a continuous 2-dimensional manifold. The DL classifier was assessed in terms of test accuracy. Rank correlation for the continuous UMAP scale against the previously published scale was calculated. Additionally, the UMAP scale was assessed in the κ agreement against 5 clinical experts on 100 pairs of patient volumes. For each pair of patient volumes, clinical experts were asked to select the volume with more severe MacTel disease and to compare them against the UMAP scale. MAIN OUTCOME MEASURES: Classification accuracy for the DL classifier and κ agreement versus clinical experts for UMAP. RESULTS: The multiview DL classifier achieved top 1 accuracy of 63.3% (186/294) on held-out test OCT volumes. The UMAP metric showed a clear continuous gradation of MacTel severity with a Spearman rank correlation of 0.84 with the previously published scale. Furthermore, the continuous UMAP metric achieved κ agreements of 0.56 to 0.63 with 5 clinical experts, which was comparable with interobserver κ values. CONCLUSIONS: Our UMAP embedding generated a continuous MacTel severity scale, without requiring continuous training labels. This technique can be applied to other diseases and may lead to more accurate diagnosis, improved understanding of disease progression, and key imaging features for pathologic characteristics. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Optic radiations representing different eccentricities age differently.

The neural pathways that carry information from the foveal, macular, and peripheral visual fields have distinct biological properties. The optic radiations (OR) carry foveal and peripheral information from the thalamus to the primary visual cortex (V1) through adjacent but separate pathways in the white matter. Here, we perform white matter tractometry using pyAFQ on a large sample of diffusion MRI (dMRI) data from subjects with healthy vision in the U.K. Biobank dataset (UKBB; N = 5382; age 45-81). We use pyAFQ to characterize white matter tissue properties in parts of the OR that transmit information about the foveal, macular, and peripheral visual fields, and to characterize the changes in these tissue properties with age. We find that (1) independent of age there is higher fractional anisotropy, lower mean diffusivity, and higher mean kurtosis in the foveal and macular OR than in peripheral OR, consistent with denser, more organized nerve fiber populations in foveal/parafoveal pathways, and (2) age is associated with increased diffusivity and decreased anisotropy and kurtosis, consistent with decreased density and tissue organization with aging. However, anisotropy in foveal OR decreases faster with age than in peripheral OR, while diffusivity increases faster in peripheral OR, suggesting foveal/peri-foveal OR and peripheral OR differ in how they age.

Race, Ethnicity, Insurance, and Population Density Associations with Pediatric Strabismus and Strabismic Amblyopia in the IRIS® Registry.

PURPOSE: To evaluate the associations of sociodemographic factors with pediatric strabismus diagnosis and outcomes. DESIGN: Retrospective cohort study. PARTICIPANTS: American Academy of Ophthalmology IRIS® Registry (Intelligent Research in Sight) patients with strabismus diagnosed before the age of 10 years. METHODS: Multivariable regression models evaluated the associations of race and ethnicity, insurance, population density, and ophthalmologist ratio with age at strabismus diagnosis, diagnosis of amblyopia, residual amblyopia, and strabismus surgery. Survival analysis evaluated the same predictors of interest with the outcome of time to strabismus surgery. MAIN OUTCOME MEASURES: Age at strabismus diagnosis, rate of amblyopia and residual amblyopia, and rate of and time to strabismus surgery. RESULTS: The median age at diagnosis was 5 years (interquartile range, 3-7) for 106 723 children with esotropia (ET) and 54 454 children with exotropia (XT). Amblyopia diagnosis was more likely with Medicaid insurance than commercial insurance (odds ratio [OR], 1.05 for ET; 1.25 for XT; P < 0.01), as was residual amblyopia (OR, 1.70 for ET; 1.53 for XT; P < 0.01). For XT, Black children were more likely to develop residual amblyopia than White children (OR, 1.34; P < 0.01). Children with Medicaid were more likely to undergo surgery and did so sooner after diagnosis (hazard ratio [HR], 1.23 for ET; 1.21 for XT; P < 0.01) than those with commercial insurance. Compared with White children, Black, Hispanic, and Asian children were less likely to undergo ET surgery and received surgery later (all HRs < 0.87; P < 0.01), and Hispanic and Asian children were less likely to undergo XT surgery and received surgery later (all HRs < 0.85; P < 0.01). Increasing population density and clinician ratio were associated with lower HR for ET surgery (P < 0.01). CONCLUSIONS: Children with strabismus covered by Medicaid insurance had increased odds of amblyopia and underwent strabismus surgery sooner after diagnosis compared with children covered by commercial insurance. After adjusting for insurance status, Black, Hispanic, and Asian children were less likely to receive strabismus surgery with a longer delay between diagnosis and surgery compared with White children. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Training Deep Learning Models to Work on Multiple Devices by Cross-Domain Learning with No Additional Annotations.

PURPOSE: To create an unsupervised cross-domain segmentation algorithm for segmenting intraretinal fluid and retinal layers on normal and pathologic macular OCT images from different manufacturers and camera devices. DESIGN: We sought to use generative adversarial networks (GANs) to generalize a segmentation model trained on one OCT device to segment B-scans obtained from a different OCT device manufacturer in a fully unsupervised approach without labeled data from the latter manufacturer. PARTICIPANTS: A total of 732 OCT B-scans from 4 different OCT devices (Heidelberg Spectralis, Topcon 1000, Maestro2, and Zeiss Plex Elite 9000). METHODS: We developed an unsupervised GAN model, GANSeg, to segment 7 retinal layers and intraretinal fluid in Topcon 1000 OCT images (domain B) that had access only to labeled data on Heidelberg Spectralis images (domain A). GANSeg was unsupervised because it had access only to 110 Heidelberg labeled OCTs and 556 raw and unlabeled Topcon 1000 OCTs. To validate GANSeg segmentations, 3 masked graders manually segmented 60 OCTs from an external Topcon 1000 test dataset independently. To test the limits of GANSeg, graders also manually segmented 3 OCTs from Zeiss Plex Elite 9000 and Topcon Maestro2. A U-Net was trained on the same labeled Heidelberg images as baseline. The GANSeg repository with labeled annotations is at https://github.com/uw-biomedical-ml/ganseg. MAIN OUTCOME MEASURES: Dice scores comparing segmentation results from GANSeg and the U-Net model with the manual segmented images. RESULTS: Although GANSeg and U-Net achieved comparable Dice scores performance as human experts on the labeled Heidelberg test dataset, only GANSeg achieved comparable Dice scores with the best performance for the ganglion cell layer plus inner plexiform layer (90%; 95% confidence interval [CI], 68%-96%) and the worst performance for intraretinal fluid (58%; 95% CI, 18%-89%), which was statistically similar to human graders (79%; 95% CI, 43%-94%). GANSeg significantly outperformed the U-Net model. Moreover, GANSeg generalized to both Zeiss and Topcon Maestro2 swept-source OCT domains, which it had never encountered before. CONCLUSIONS: GANSeg enables the transfer of supervised deep learning algorithms across OCT devices without labeled data, thereby greatly expanding the applicability of deep learning algorithms.

United States Population Disparities in Ophthalmic Care: Blindness and Visual Impairment in the IRIS® Registry (Intelligent Research in Sight).

PURPOSE: To evaluate associations of patient characteristics with United States eye care use and likelihood of blindness. DESIGN: Retrospective observational study. PARTICIPANTS: Patients (19 546 016) with 2018 visual acuity (VA) records in the American Academy of Ophthalmology's IRIS® Registry (Intelligent Research in Sight). METHODS: Legal blindness (20/200 or worse) and visual impairment (VI; worse than 20/40) were identified from corrected distance acuity in the better-seeing eye and stratified by patient characteristics. Multivariable logistic regressions evaluated associations with blindness and VI. Blindness was mapped by state and compared with population characteristics. Eye care use was analyzed by comparing population demographics with United States Census estimates and proportional demographic representation among blind patients versus a nationally representative US population sample (National Health and Nutritional Examination Survey [NHANES]). MAIN OUTCOME MEASURES: Prevalence and odds ratios for VI and blindness; proportional representation in the IRIS® Registry, Census, and NHANES by patient demographics. RESULTS: Visual impairment was present in 6.98% (n = 1 364 935) and blindness in 0.98% (n = 190 817) of IRIS patients. Adjusted odds of blindness were highest among patients ≥ 85 years old (odds ratio [OR], 11.85; 95% confidence interval [CI], 10.33-13.59 vs. those 0-17 years old). Blindness also was associated positively with rural location and Medicaid, Medicare, or no insurance vs. commercial insurance. Hispanic (OR, 1.59; 95% CI, 1.46-1.74) and Black (OR, 1.73; 95% CI, 1.63-1.84) patients showed a higher odds of blindness versus White non-Hispanic patients. Proportional representation in IRIS Registry relative to the Census was higher for White than Hispanic (2- to 4-fold) or Black (11%-85%) patients (P < 0.001). Blindness overall was less prevalent in NHANES than IRIS Registry; however, prevalence in adults aged 60+ was lowest among Black participants in the NHANES (0.54%) and second highest among comparable Black adults in IRIS (1.57%). CONCLUSIONS: Legal blindness from low VA was present in 0.98% of IRIS patients and associated with rural location, public or no insurance, and older age. Compared with US Census estimates, minorities may be underrepresented among ophthalmology patients, and compared with NHANES population estimates, Black individuals may be overrepresented among blind IRIS Registry patients. These findings provide a snapshot of US ophthalmic care and highlight the need for initiatives to address disparities in use and blindness. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Data Sources for Evaluating Health Disparities in Ophthalmology: Where We Are and Where We Need to Go.

Data provide an opportunity to discover disparities and inequities that may otherwise be unrecognized. Within the American Academy of Ophthalmology (AAO) Task Force on Disparities in Eye Care, the Leveraging Data Sub-task Force was charged with identifying data sources to study health disparities in eye care and to leverage data to advance health equity. We evaluated large data sources to determine their strengths, deficiencies, and relative accessibility in relation to the likelihood of identifying eye care disparities. We highlight the current challenges with these data sources and review key recommendations for improving future sources for studying health disparities in eye care.

Policy-Driven, Multimodal Deep Learning for Predicting Visual Fields from the Optic Disc and OCT Imaging.

PURPOSE: To develop and validate a deep learning (DL) system for predicting each point on visual fields (VFs) from disc and OCT imaging and derive a structure-function mapping. DESIGN: Retrospective, cross-sectional database study. PARTICIPANTS: A total of 6437 patients undergoing routine care for glaucoma in 3 clinical sites in the United Kingdom. METHODS: OCT and infrared reflectance (IR) optic disc imaging were paired with the closest VF within 7 days. EfficientNet B2 was used to train 2 single-modality DL models to predict each of the 52 sensitivity points on the 24-2 VF pattern. A policy DL model was designed and trained to fuse the 2 model predictions. MAIN OUTCOME MEASURES: Pointwise mean absolute error (PMAE). RESULTS: A total of 5078 imaging scans to VF pairs were used as a held-out test set to measure the final performance. The improvement in PMAE with the policy model was 0.485 (0.438, 0.533) decibels (dB) compared with the IR image of the disc alone and 0.060 (0.047, 0.073) dB with to the OCT alone. The improvement with the policy fusion model was statistically significant (P < 0.0001). Occlusion masking shows that the DL models learned the correct structure-function mapping in a data-driven, feature agnostic fashion. CONCLUSIONS: The multimodal, policy DL model performed the best; it provided explainable maps of its confidence in fusing data from single modalities and provides a pathway for probing the structure-function relationship in glaucoma.

From Data to Deployment: The Collaborative Community on Ophthalmic Imaging Roadmap for Artificial Intelligence in Age-Related Macular Degeneration.

OBJECTIVE: Health care systems worldwide are challenged to provide adequate care for the 200 million individuals with age-related macular degeneration (AMD). Artificial intelligence (AI) has the potential to make a significant, positive impact on the diagnosis and management of patients with AMD; however, the development of effective AI devices for clinical care faces numerous considerations and challenges, a fact evidenced by a current absence of Food and Drug Administration (FDA)-approved AI devices for AMD. PURPOSE: To delineate the state of AI for AMD, including current data, standards, achievements, and challenges. METHODS: Members of the Collaborative Community on Ophthalmic Imaging Working Group for AI in AMD attended an inaugural meeting on September 7, 2020, to discuss the topic. Subsequently, they undertook a comprehensive review of the medical literature relevant to the topic. Members engaged in meetings and discussion through December 2021 to synthesize the information and arrive at a consensus. RESULTS: Existing infrastructure for robust AI development for AMD includes several large, labeled data sets of color fundus photography and OCT images; however, image data often do not contain the metadata necessary for the development of reliable, valid, and generalizable models. Data sharing for AMD model development is made difficult by restrictions on data privacy and security, although potential solutions are under investigation. Computing resources may be adequate for current applications, but knowledge of machine learning development may be scarce in many clinical ophthalmology settings. Despite these challenges, researchers have produced promising AI models for AMD for screening, diagnosis, prediction, and monitoring. Future goals include defining benchmarks to facilitate regulatory authorization and subsequent clinical setting generalization. CONCLUSIONS: Delivering an FDA-authorized, AI-based device for clinical care in AMD involves numerous considerations, including the identification of an appropriate clinical application; acquisition and development of a large, high-quality data set; development of the AI architecture; training and validation of the model; and functional interactions between the model output and clinical end user. The research efforts undertaken to date represent starting points for the medical devices that eventually will benefit providers, health care systems, and patients.

Endophthalmitis Rate in Immediately Sequential versus Delayed Sequential Bilateral Cataract Surgery within the Intelligent Research in Sight (IRIS®) Registry Data.

PURPOSE: To compare the rate of postoperative endophthalmitis after immediately sequential bilateral cataract surgery (ISBCS) versus delayed sequential bilateral cataract surgery (DSBCS) using the American Academy of Ophthalmology Intelligent Research in Sight (IRIS®) Registry database. DESIGN: Retrospective cohort study. PARTICIPANTS: Patients in the IRIS Registry who underwent cataract surgery from 2013 through 2018. METHODS: Patients who underwent cataract surgery were divided into 2 groups: (1) ISBCS and (2) DSBCS (second-eye surgery ≥1 day after the first-eye surgery) or unilateral surgery. Postoperative endophthalmitis was defined as endophthalmitis occurring within 4 weeks of surgery by International Classification of Diseases (ICD) code and ICD code with additional clinical criteria. MAIN OUTCOME MEASURES: Rate of postoperative endophthalmitis. RESULTS: Of 5 573 639 IRIS Registry patients who underwent cataract extraction, 165 609 underwent ISBCS, and 5 408 030 underwent DSBCS or unilateral surgery (3 695 440 DSBCS, 1 712 590 unilateral surgery only). A total of 3102 participants (0.056%) met study criteria of postoperative endophthalmitis with supporting clinical findings. The rates of endophthalmitis in either surgery eye between the 2 surgery groups were similar (0.059% in the ISBCS group vs. 0.056% in the DSBCS or unilateral group; P = 0.53). Although the incidence of endophthalmitis was slightly higher in the ISBCS group compared with the DSBCS or unilateral group, the odds ratio did not reach statistical significance (1.08; 95% confidence interval, 0.87-1.31; P = 0.47) after adjusting for age, sex, race, insurance status, and comorbid eye disease. Seven cases of bilateral endophthalmitis with supporting clinical data in the DSBCS group and no cases in the ISBCS group were identified. CONCLUSIONS: Risk of postoperative endophthalmitis was not statistically significantly different between patients who underwent ISBCS and DSBCS or unilateral cataract surgery.

Artificial intelligence-based predictions in neovascular age-related macular degeneration.

PURPOSE OF REVIEW: Predicting treatment response and optimizing treatment regimen in patients with neovascular age-related macular degeneration (nAMD) remains challenging. Artificial intelligence-based tools have the potential to increase confidence in clinical development of new therapeutics, facilitate individual prognostic predictions, and ultimately inform treatment decisions in clinical practice. RECENT FINDINGS: To date, most advances in applying artificial intelligence to nAMD have focused on facilitating image analysis, particularly for automated segmentation, extraction, and quantification of imaging-based features from optical coherence tomography (OCT) images. No studies in our literature search evaluated whether artificial intelligence could predict the treatment regimen required for an optimal visual response for an individual patient. Challenges identified for developing artificial intelligence-based models for nAMD include the limited number of large datasets with high-quality OCT data, limiting the patient populations included in model development; lack of counterfactual data to inform how individual patients may have fared with an alternative treatment strategy; and absence of OCT data standards, impairing the development of models usable across devices. SUMMARY: Artificial intelligence has the potential to enable powerful prognostic tools for a complex nAMD treatment landscape; however, additional work remains before these tools are applicable to informing treatment decisions for nAMD in clinical practice.

Gaps in standards for integrating artificial intelligence technologies into ophthalmic practice.

PURPOSE OF REVIEW: The purpose of this review is to provide an overview of healthcare standards and their relevance to multiple ophthalmic workflows, with a specific emphasis on describing gaps in standards development needed for improved integration of artificial intelligence technologies into ophthalmic practice. RECENT FINDINGS: Healthcare standards are an essential component of data exchange and critical for clinical practice, research, and public health surveillance activities. Standards enable interoperability between clinical information systems, healthcare information exchange between institutions, and clinical decision support in a complex health information technology ecosystem. There are several gaps in standards in ophthalmology, including relatively low adoption of imaging standards, lack of use cases for integrating apps providing artificial intelligence -based decision support, lack of common data models to harmonize big data repositories, and no standards regarding interfaces and algorithmic outputs. SUMMARY: These gaps in standards represent opportunities for future work to develop improved data flow between various elements of the digital health ecosystem. This will enable more widespread adoption and integration of artificial intelligence-based tools into clinical practice. Engagement and support from the ophthalmology community for standards development will be important for advancing this work.

Using Deep Learning to Automate Goldmann Applanation Tonometry Readings.

PURPOSE: To develop an objective and automated method for measuring intraocular pressure using deep learning and fixed-force Goldmann applanation tonometry (GAT) techniques. DESIGN: Prospective cross-sectional study. PARTICIPANTS: Patients from an academic glaucoma practice. METHODS: Intraocular pressure was estimated by analyzing videos recorded using a standard slit-lamp microscope and fixed-force GAT. Video frames were labeled to identify the outline of the reference tonometer and the applanation mires. A deep learning model was trained to localize and segment the tonometer and mires. Intraocular pressure values were calculated from the deep learning-predicted tonometer and mire diameters using the Imbert-Fick formula. A separate test set was collected prospectively in which standard and automated GAT measurements were collected in random order by 2 independent masked observers to assess the deep learning model as well as interobserver variability. MAIN OUTCOME MEASURES: Intraocular pressure measurements between standard and automated methods were compared. RESULTS: Two hundred sixty-three eyes of 135 patients were included in the training and validation videos. For the test set, 50 eyes from 25 participants were included. Each eye was measured by 2 observers, resulting in 100 videos. Within the test set, the mean difference between automated and standard GAT results was -0.9 mmHg (95% limits of agreement [LoA], -5.4 to 3.6 mmHg). Mean difference between the 2 observers using standard GAT was 0.09 mmHg (LoA,-3.8 to 4.0 mmHg). Mean difference between the 2 observers using automated GAT videos was -0.3 mmHg (LoA, -4.1 to 3.5 mmHg). The coefficients of repeatability for automated and standard GAT were 3.8 and 3.9 mmHg, respectively. The bias for even-numbered measurements was reduced when using automated GAT. CONCLUSIONS: Preliminary measurements using deep learning to automate GAT demonstrate results comparable with those of standard GAT. Automated GAT has the potential to improve on our current GAT measurement standards significantly by reducing bias and improving repeatability. In addition, ocular pulse amplitudes could be observed using this technique.

Logistic Regression Classification of Primary Vitreoretinal Lymphoma versus Uveitis by Interleukin 6 and Interleukin 10 Levels.

PURPOSE: To assess the diagnostic performance and generalizability of logistic regression in classifying primary vitreoretinal lymphoma (PVRL) versus uveitis from intraocular cytokine levels in a single-center retrospective cohort, comparing a logistic regression model and previously published Interleukin Score for Intraocular Lymphoma Diagnosis (ISOLD) scores against the interleukin 10 (IL-10)-to-interleukin 6 (IL-6) ratio. DESIGN: Retrospective cohort study. PARTICIPANTS: Patient histories, pathology reports, and intraocular cytokine levels from 2339 patient entries in the National Eye Institute Histopathology Core database. METHODS: Patient diagnoses of PVRL versus uveitis and associated aqueous or vitreous IL-6 and IL-10 levels were collected retrospectively. From these data, cytokine levels were compared between diagnoses with the Mann-Whitney U test. A logistic regression model was trained to classify PVRL versus uveitis from aqueous and vitreous IL-6 and IL-10 samples and compared with ISOLD scores and IL-10-to-IL-6 ratios. MAIN OUTCOME MEASURES: Area under the receiver operating characteristic curve (AUC) for each classifier and sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) at the optimal cutoff (maximal Youden index) for each classifier. RESULTS: Seventy-seven lymphoma patients (10 aqueous samples, 67 vitreous samples) and 84 uveitis patients (19 aqueous samples, 65 vitreous samples) treated between October 5, 1999, and September 16, 2015, were included. Interleukin 6 levels were higher and IL-10 levels were lower in uveitis patients compared with lymphoma patients (P < 0.01). For vitreous samples, the logistic regression model, ISOLD score, and IL-10-to-IL-6 ratio achieved AUCs of 98.3%, 97.7%, and 96.3%, respectively. Sensitivity, specificity, PPV, and NPV at the optimal cutoffs for each classifier were 94.2%, 96.9%, 97%, and 94% for the logistic regression model; 92.7%, 100%, 100%, and 92.9% for the ISOLD score; and 94.2%, 95.3%, 95.6%, and 93.9% for the IL-10-to-IL-6 ratio. All models achieved complete separation between uveitis and lymphoma in the aqueous data set. CONCLUSIONS: The accuracy of the logistic regression model and generalizability of the ISOLD score to an independent patient cohort suggest that intraocular cytokine analysis by logistic regression may be a promising adjunct to cytopathologic analysis, the gold standard, for the early diagnosis of primary vitreoretinal lymphoma. Further validation studies are merited.

Big data requirements for artificial intelligence.

PURPOSE OF REVIEW: To summarize how big data and artificial intelligence technologies have evolved, their current state, and next steps to enable future generations of artificial intelligence for ophthalmology. RECENT FINDINGS: Big data in health care is ever increasing in volume and variety, enabled by the widespread adoption of electronic health records (EHRs) and standards for health data information exchange, such as Digital Imaging and Communications in Medicine and Fast Healthcare Interoperability Resources. Simultaneously, the development of powerful cloud-based storage and computing architectures supports a fertile environment for big data and artificial intelligence in health care. The high volume and velocity of imaging and structured data in ophthalmology and is one of the reasons why ophthalmology is at the forefront of artificial intelligence research. Still needed are consensus labeling conventions for performing supervised learning on big data, promotion of data sharing and reuse, standards for sharing artificial intelligence model architectures, and access to artificial intelligence models through open application program interfaces (APIs). SUMMARY: Future requirements for big data and artificial intelligence include fostering reproducible science, continuing open innovation, and supporting the clinical use of artificial intelligence by promoting standards for data labels, data sharing, artificial intelligence model architecture sharing, and accessible code and APIs.

Medicare Incentive Payments to United States Ophthalmologists for Use of Electronic Health Records: 2011-2016.

PURPOSE: To investigate ophthalmologists' rate of attestation to meaningful use (MU) of their electronic health record (EHR) systems in the Medicare EHR Incentive Program and their continuity and success in receiving payments in comparison with other specialties. DESIGN: Administrative database study. PARTICIPANTS: Eligible professionals participating in the Medicare EHR Incentive Program. METHODS: Based on publicly available data sources, subsets of payment and attestation data were created for ophthalmologists and for other specialties. The number of eligible professionals attesting was determined using the attestation data for each year and stage of the program. The proportion of attestations by EHR vendor was calculated using all attestations for each vendor. MAIN OUTCOME MEASURES: Numbers of ophthalmologists attesting by year and stage of the Medicare EHR Incentive Program, incentive payments, and number of attestations by EHR vendor. RESULTS: In the peak year of participation, 51.6% of ophthalmologists successfully attested to MU, compared with 37.1% of optometrists, 50.2% of dermatologists, 54.5% of otolaryngologists, and 64.4% of urologists. Across the 6 years of the program, ophthalmologists received an average of $17 942 in incentive payments compared with $11 105 for optometrists, $16 617 for dermatologists, $20 203 for otolaryngologists, and $23 821 for urologists. Epic and Nextgen were the most frequently used EHRs for attestation by ophthalmologists. CONCLUSIONS: Ophthalmology as a specialty performed better than optometry and dermatology, but worse than otolaryngology and urology, in terms of the proportion of eligible professionals attesting to MU of EHRs. Ophthalmologists were more likely to remain in the program after their initial year of attestation compared with all eligible providers. The top 4 EHR vendors accounted for 50% of attestations by ophthalmologists.

Associations between recent and established ophthalmic conditions and risk of Alzheimer's disease.

INTRODUCTION: Identifying ophthalmic diseases associated with increased risk of Alzheimer's disease (AD) may enable better screening and understanding of those at risk of AD. METHODS: Diagnoses of glaucoma, age-related macular degeneration (AMD), and diabetic retinopathy (DR) were based on International Classification of Diseases, 9th revision, codes for 3877 participants from the Adult Changes in Thought study. The adjusted hazard ratio for developing probable or possible AD for recent (within 5 years) and established (>5 years) diagnoses were assessed. RESULTS: Over 31,142 person-years of follow-up, 792 AD cases occurred. The recent and established hazard ratio were 1.46 (P = .01) and 0.87 (P = .19) for glaucoma, 1.20 (P = .12) and 1.50 (P < .001) for AMD, and 1.50 (P = .045) and 1.50 (P = .03) for DR. DISCUSSION: Increased AD risk was found for recent glaucoma diagnoses, established AMD diagnoses, and both recent and established DR. People with certain ophthalmic conditions may have increased AD risk.

Determinants of Outcomes of Adenoviral Keratoconjunctivitis.

PURPOSE: To determine host and pathogen factors predictive of outcomes in a large clinical cohort with keratoconjunctivitis. DESIGN: Retrospective analyses of the clinical and molecular data from a randomized, controlled, masked trial for auricloscene for keratoconjunctivitis (NVC-422 phase IIB, NovaBay; clinicaltrials.gov identifier, NCT01877694). PARTICIPANTS: Five hundred participants from United States, India, Brazil, and Sri Lanka with clinical diagnosis of keratoconjunctivitis and positive rapid test results for adenovirus. METHODS: Clinical signs and symptoms and bilateral conjunctival swabs were obtained on days 1, 3, 6, 11, and 18. Polymerase chain reaction (PCR) analysis was performed to detect and quantify adenovirus in all samples. Regression models were used to evaluate the association of various variables with keratoconjunctivitis outcomes. Time to resolution of each symptom or sign was assessed by adenoviral species with Cox regression. MAIN OUTCOME MEASURES: The difference in composite scores of clinical signs between days 1 and 18, mean visual acuity change between days 1 and 18, and time to resolution of each symptom or sign. RESULTS: Of 500 participants, 390 (78%) showed evidence of adenovirus by PCR. Among adenovirus-positive participants, adenovirus D species was most common (63% of total cases), but a total of 4 species and 21 different types of adenovirus were detected. Adenovirus D was associated with more severe signs and symptoms, a higher rate of subepithelial infiltrate development, and a slower decline in viral load compared with all other adenovirus species. The clinical courses of all patients with non-adenovirus D species infection and adenovirus-negative keratoconjunctivitis were similar. Mean change in visual acuity between days 1 and 18 was a gain of 1.9 letters; worse visual outcome was associated with older age. CONCLUSIONS: A substantial proportion of keratoconjunctivitis is not associated with a detectable adenovirus. The clinical course of those with adenovirus D keratoconjunctivitis is significantly more severe than those with non-adenovirus D species infections or adenovirus-negative keratoconjunctivitis; high viral load at presentation and non-United States origin of participants is associated with poorer clinical outcome.

VISUAL ACUITY IMPROVEMENT WHEN SWITCHING FROM RANIBIZUMAB TO AFLIBERCEPT IS NOT SUSTAINED.

PURPOSE: To assess whether visual benefits exist in switching to aflibercept in patients who have been chronically treated with ranibizumab for neovascular age-related macular degeneration. METHODS: A multicenter, national, electronic medical record database study was performed. Patients undergoing six continuous monthly ranibizumab injections and then switched to continuous aflibercept were matched to those on continuous ranibizumab therapy. Matching was performed in a 2:1 ratio and based on visual acuity 6 months before and at the time of the switch, and the number of previous ranibizumab injections. RESULTS: Patients who were switched to aflibercept demonstrated transiently significant improvement in visual acuity that peaked at an increase of 0.9 Early Treatment Diabetic Retinopathy Study letters 3 months after the switch, whereas control patients continued on ranibizumab treatment showed a steady decline in visual acuity. Visual acuity differences between the groups were significant (P < 0.05) at 2, 3, and 5 months after the switch. Beginning at 4 months after the switch, the switch group showed a visual acuity decline similar to the control group. CONCLUSION: Transient, nonsustained improvement in visual acuity occurs when switching between anti-vascular endothelial growth factor agents, which may have implications in treating patients on chronic maintenance therapy on one anti-vascular endothelial growth factor medication.