Updates in deep learning research in ophthalmology.

Ophthalmology has been one of the early adopters of artificial intelligence (AI) within the medical field. Deep learning (DL), in particular, has garnered significant attention due to the availability of large amounts of data and digitized ocular images. Currently, AI in Ophthalmology is mainly focused on improving disease classification and supporting decision-making when treating ophthalmic diseases such as diabetic retinopathy, age-related macular degeneration (AMD), glaucoma and retinopathy of prematurity (ROP). However, most of the DL systems (DLSs) developed thus far remain in the research stage and only a handful are able to achieve clinical translation. This phenomenon is due to a combination of factors including concerns over security and privacy, poor generalizability, trust and explainability issues, unfavorable end-user perceptions and uncertain economic value. Overcoming this challenge would require a combination approach. Firstly, emerging techniques such as federated learning (FL), generative adversarial networks (GANs), autonomous AI and blockchain will be playing an increasingly critical role to enhance privacy, collaboration and DLS performance. Next, compliance to reporting and regulatory guidelines, such as CONSORT-AI and STARD-AI, will be required to in order to improve transparency, minimize abuse and ensure reproducibility. Thirdly, frameworks will be required to obtain patient consent, perform ethical assessment and evaluate end-user perception. Lastly, proper health economic assessment (HEA) must be performed to provide financial visibility during the early phases of DLS development. This is necessary to manage resources prudently and guide the development of DLS.

Artificial intelligence for diabetic retinopathy screening, prediction and management.

PURPOSE OF REVIEW: Diabetic retinopathy is the most common specific complication of diabetes mellitus. Traditional care for patients with diabetes and diabetic retinopathy is fragmented, uncoordinated and delivered in a piecemeal nature, often in the most expensive and high-resource tertiary settings. Transformative new models incorporating digital technology are needed to address these gaps in clinical care. RECENT FINDINGS: Artificial intelligence and telehealth may improve access, financial sustainability and coverage of diabetic retinopathy screening programs. They enable risk stratifying patients based on individual risk of vision-threatening diabetic retinopathy including diabetic macular edema (DME), and predicting which patients with DME best respond to antivascular endothelial growth factor therapy. SUMMARY: Progress in artificial intelligence and tele-ophthalmology for diabetic retinopathy screening, including artificial intelligence applications in 'real-world settings' and cost-effectiveness studies are summarized. Furthermore, the initial research on the use of artificial intelligence models for diabetic retinopathy risk stratification and management of DME are outlined along with potential future directions. Finally, the need for artificial intelligence adoption within ophthalmology in response to coronavirus disease 2019 is discussed. Digital health solutions such as artificial intelligence and telehealth can facilitate the integration of community, primary and specialist eye care services, optimize the flow of patients within healthcare networks, and improve the efficiency of diabetic retinopathy management.

THE COLLABORATIVE OCULAR TUBERCULOSIS STUDY (COTS)-1: A Multinational Review of 251 Patients With Tubercular Retinal Vasculitis.

PURPOSE: Tubercular retinal vasculitis (TRV) is a heterogeneous disease that can be difficult to manage because of nonspecific presentation and limitations of confirmatory tests for tuberculosis. This is a big data analysis on phenotypes and treatment outcomes for TRV. METHODS: Multicentre retrospective study of patients with TRV between January 2004 and December 2014 and a minimum follow-up of 1 year. RESULTS: Two hundred and fifty-one patients with TRV with a mean age of 38.9 ± 14.4 years (range, 9-86 years) were included. The patients were predominantly males (n = 167/251; 66.5%) of Asian ethnicity (n = 174/246; 70.7%), and geographical origin (n = 137/251; 54.6%). Most patients had features of occlusive type of RV (n = 113/185; 61.1%) except Caucasians (n = 11; 28.2%). There was no significant difference in treatment failure whether patients received antitubercular therapy (ATT) (P = 0.29), although treatment failure was less frequent in patients who received ATT (13.6%; n = 31/228) compared with those who did not (21.7%, n = 5/23). Less treatment failures were observed in patients with occlusive type RV who received ATT; however, this was not significant on survival analysis (P = 0.09). Treatment with ATT was associated with higher failure rates in patients of Hispanic and African American race and those with TRV associated with panuveitis (compared with posterior uveitis). CONCLUSION: In this multinational study of TRV, there was no significant therapeutic effect of ATT. However, a definitive conclusion about the role of ATT could not be made because of a few patients who did not receive ATT. Because this is a retrospective study with a limited 1-year follow-up, the effect of ATT may have been overestimated (or underestimated) in the duration of follow-up.

PERIPHERAL RETINAL VASCULITIS: Analysis of 110 Consecutive Cases and a Contemporary Reappraisal of Tubercular Etiology.

PURPOSE: Describe the clinical features and outcomes of patients with peripheral retinal vasculitis (RV) and describe clinical characteristics of presumed tubercular RV in a nonendemic setting. METHODS: Retrospective cohort study of 110 consecutive patients with peripheral RV at a tertiary referral eye care center in the United Kingdom. Retinal vasculitis was defined as RV with vitritis associated with peripheral retinal ischemia. Patients who also had positive Quantiferon Gold in Tube test, positive tuberculin skin test, and/or other evidence of systemic tuberculosis such as biopsy were labeled with presumed tubercular RV. Treatment success was defined as resolution of inflammation, and successful tapering of oral corticosteroids to less than 10 mg/day or topical steroids to less than twice a day. RESULTS: Mean age of the study population was 42.69 ± 14.95 years. Patients were predominantly Asian (49.1%) and Male (67.0%). A total of 73.2% had bilateral involvement. Sixty-nine (62.72%) patients had presumed tubercular RV. A total of 52.8% patients received antitubercular therapy, 65.5% received oral corticosteroids, and 17.3% required steroid-sparing oral immunosuppressive agents. A total of 85.19% of patients with presumed tubercular RV achieved treatment success with concurrent antitubercular therapy as against 75.61% of patients with nontubercular RV. CONCLUSION: This is the largest study of the epidemiology, clinical features, and outcomes of both peripheral RV and presumed tubercular RV to date. Presumed tubercular RV commonly seems to affect young males of Asian descent and had vitreous hemorrhage as common clinical findings and also demonstrated a good treatment outcome with antitubercular therapy.

Aqueous humor immune factors and cytomegalovirus (CMV) levels in CMV retinitis through treatment - The CRIGSS study.

PURPOSE: This study aims to perform comprehensive longitudinal immune factor analysis of aqueous humor in relation to the aqueous CMV viral load and systemic CD4 counts during treatment of patients with co-infection of HIV and CMVR. METHODS: Aqueous humor samples were collected from 17 HIV-positive patients with CMVR scheduled to undergo weekly intravitreal ganciclovir therapy as part of the prospective CMV Retinitis Intravitreal Ganciclovir Singapore Study (CRIGSS) over the course of 1year. Full data across all the 4 time points was obtained and analyzed for CMV DNA viral load, 41 cytokine and chemokine factors using real-time PCR with the FlexMAP 3D (Luminex®) platform and assessed using the Milliplex Human Cytokine® kit. RESULTS: The following immune factors (Spearman correlation coefficient r value in parenthesis, p<0.05) showed strong correlation with CMV DNA load in the aqueous - MCP-1 (0.80, IFN-g (0.83), IP-10 (0.82), IL-8 (0.81), fractalkine (0.73), RANTES (0.68) - while the following showed moderate correlation - PDGF-AA (0.58), Flt-3L (0.59) and G-CSF (0.53). Only PDGF-AA revealed a statistically significant negative correlation with serum CD4 levels (r=-0.74). CONCLUSION: Immune factors that correlate with intraocular CMV DNA load are identified. They are indicative of a Th1 and monocyte-macrophage mediated response, and exhibit a decreasing trend longitudinally through the course of treatment. These factors may be an important new consideration in individualizing the treatment of patients with CMVR.

Clinical Features and CD4+ T Cells Count in AIDS Patients with CMV Retinitis: Correlation with Mortality.

PURPOSE: To explore the all-cause mortality in patients with acquired immune deficiency syndrome (AIDS) and Cytomegalovirus (CMV) retinitis. METHODS: A retrospective cohort study of patients with CMV retinitis (CMVR) presented to a tertiary referral center in Singapore from January 1, 2004, through December 31, 2015. RESULTS: A total of 144 patients were studied (87 survived, 11 lost to follow up, 46 died). Patients with bilateral CMVR and six-month follow up CD4 + T cell count < 50 cells/mm3 have shorter time to mortality, compared to patients with CD4 + T cell count > 50 cells/mm3 (p < .001) and unilateral disease (p = .043). Baseline CD4 + T cell count, size and zone of initial primary retinitis lesions, recurrences of retinitis, and timing of combined antiretroviral therapy (cART) are not significantly associated with mortality. CONCLUSION: Bilateral ocular involvement and lack of immune recovery in patients with AIDS and CMVR are associated with shorter survival time.

Acceptance and Perception of Artificial Intelligence Usability in Eye Care (APPRAISE) for Ophthalmologists: A Multinational Perspective.

Background: Many artificial intelligence (AI) studies have focused on development of AI models, novel techniques, and reporting guidelines. However, little is understood about clinicians' perspectives of AI applications in medical fields including ophthalmology, particularly in light of recent regulatory guidelines. The aim for this study was to evaluate the perspectives of ophthalmologists regarding AI in 4 major eye conditions: diabetic retinopathy (DR), glaucoma, age-related macular degeneration (AMD) and cataract. Methods: This was a multi-national survey of ophthalmologists between March 1st, 2020 to February 29th, 2021 disseminated via the major global ophthalmology societies. The survey was designed based on microsystem, mesosystem and macrosystem questions, and the software as a medical device (SaMD) regulatory framework chaired by the Food and Drug Administration (FDA). Factors associated with AI adoption for ophthalmology analyzed with multivariable logistic regression random forest machine learning. Results: One thousand one hundred seventy-six ophthalmologists from 70 countries participated with a response rate ranging from 78.8 to 85.8% per question. Ophthalmologists were more willing to use AI as clinical assistive tools (88.1%, n = 890/1,010) especially those with over 20 years' experience (OR 3.70, 95% CI: 1.10-12.5, p = 0.035), as compared to clinical decision support tools (78.8%, n = 796/1,010) or diagnostic tools (64.5%, n = 651). A majority of Ophthalmologists felt that AI is most relevant to DR (78.2%), followed by glaucoma (70.7%), AMD (66.8%), and cataract (51.4%) detection. Many participants were confident their roles will not be replaced (68.2%, n = 632/927), and felt COVID-19 catalyzed willingness to adopt AI (80.9%, n = 750/927). Common barriers to implementation include medical liability from errors (72.5%, n = 672/927) whereas enablers include improving access (94.5%, n = 876/927). Machine learning modeling predicted acceptance from participant demographics with moderate to high accuracy, and area under the receiver operating curves of 0.63-0.83. Conclusion: Ophthalmologists are receptive to adopting AI as assistive tools for DR, glaucoma, and AMD. Furthermore, ML is a useful method that can be applied to evaluate predictive factors on clinical qualitative questionnaires. This study outlines actionable insights for future research and facilitation interventions to drive adoption and operationalization of AI tools for Ophthalmology.

Digital health during COVID-19: lessons from operationalising new models of care in ophthalmology.

The COVID-19 pandemic has resulted in massive disruptions within health care, both directly as a result of the infectious disease outbreak, and indirectly because of public health measures to mitigate against transmission. This disruption has caused rapid dynamic fluctuations in demand, capacity, and even contextual aspects of health care. Therefore, the traditional face-to-face patient-physician care model has had to be re-examined in many countries, with digital technology and new models of care being rapidly deployed to meet the various challenges of the pandemic. This Viewpoint highlights new models in ophthalmology that have adapted to incorporate digital health solutions such as telehealth, artificial intelligence decision support for triaging and clinical care, and home monitoring. These models can be operationalised for different clinical applications based on the technology, clinical need, demand from patients, and manpower availability, ranging from out-of-hospital models including the hub-and-spoke pre-hospital model, to front-line models such as the inflow funnel model and monitoring models such as the so-called lighthouse model for provider-led monitoring. Lessons learnt from operationalising these models for ophthalmology in the context of COVID-19 are discussed, along with their relevance for other specialty domains.

Mobile technology: Usage and perspective of patients and caregivers presenting to a tertiary care emergency department.

BACKGROUND: Developments in information technology (IT) have driven a push in healthcare innovation in the emergency department (ED). Many of these applications rely on mobile technology (MT) such as smartphones but not everyone is comfortable with MT usage. Our study aims to characterize the technology usage behavior of users in the ED so as to guide the implementation of IT interventions in the ED. METHODS: A cross-sectional survey was conducted in the emergency department of a tertiary hospital. Patients and their caregivers aged 21 and above were recruited. The survey collected demographic information, technology usage patterns, and participant reported comfort level in the usage of MT. We performed descriptive statistics and multivariate logistic regression to identify factors differentially associated with comfort in usage of MT. RESULTS: A total of 498 participants were recruited, and 299 (60%) were patients. English was the most commonly written and read language (66.9%) and 64.2% reported a comfort level of 3/5 or more in using MT. Factors that were associated with being comfortable in using MT include having a tertiary education, being able to read and write English, as well as being a frequent user of IT. Caregivers were more likely to display these characteristics. CONCLUSION: A large proportion of ED patients are not comfortable in the usage of MT. Factors that predicted comfort level in the usage of MT were common amongst caregivers. Future interventions should take this into consideration in the design of MT interventions.

Health Economic and Safety Considerations for Artificial Intelligence Applications in Diabetic Retinopathy Screening.

Systematic screening for diabetic retinopathy (DR) has been widely recommended for early detection in patients with diabetes to address preventable vision loss. However, substantial manpower and financial resources are required to deploy opportunistic screening and transition to systematic DR screening programs. The advent of artificial intelligence (AI) technologies may improve access and reduce the financial burden for DR screening while maintaining comparable or enhanced clinical effectiveness. To deploy an AI-based DR screening program in a real-world setting, it is imperative that health economic assessment (HEA) and patient safety analyses are conducted to guide appropriate allocation of resources and design safe, reliable systems. Few studies published to date include these considerations when integrating AI-based solutions into DR screening programs. In this article, we provide an overview of the current state-of-the-art of AI technology (focusing on deep learning systems), followed by an appraisal of existing literature on the applications of AI in ophthalmology. We also discuss practical considerations that drive the development of a successful DR screening program, such as the implications of false-positive or false-negative results and image gradeability. Finally, we examine different plausible methods for HEA and safety analyses that can be used to assess concerns regarding AI-based screening.

Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: an economic analysis modelling study.

BACKGROUND: Deep learning is a novel machine learning technique that has been shown to be as effective as human graders in detecting diabetic retinopathy from fundus photographs. We used a cost-minimisation analysis to evaluate the potential savings of two deep learning approaches as compared with the current human assessment: a semi-automated deep learning model as a triage filter before secondary human assessment; and a fully automated deep learning model without human assessment. METHODS: In this economic analysis modelling study, using 39 006 consecutive patients with diabetes in a national diabetic retinopathy screening programme in Singapore in 2015, we used a decision tree model and TreeAge Pro to compare the actual cost of screening this cohort with human graders against the simulated cost for semi-automated and fully automated screening models. Model parameters included diabetic retinopathy prevalence rates, diabetic retinopathy screening costs under each screening model, cost of medical consultation, and diagnostic performance (ie, sensitivity and specificity). The primary outcome was total cost for each screening model. Deterministic sensitivity analyses were done to gauge the sensitivity of the results to key model assumptions. FINDINGS: From the health system perspective, the semi-automated screening model was the least expensive of the three models, at US$62 per patient per year. The fully automated model was $66 per patient per year, and the human assessment model was $77 per patient per year. The savings to the Singapore health system associated with switching to the semi-automated model are estimated to be $489 000, which is roughly 20% of the current annual screening cost. By 2050, Singapore is projected to have 1 million people with diabetes; at this time, the estimated annual savings would be $15 million. INTERPRETATION: This study provides a strong economic rationale for using deep learning systems as an assistive tool to screen for diabetic retinopathy. FUNDING: Ministry of Health, Singapore.

Dataset of aqueous humor cytokine profile in HIV patients with Cytomegalovirus (CMV) retinitis.

The data shows the aqueous humor cytokine profiling results acquired in a small cohort of 17 HIV patients clinically diagnosed with Cytomegalovirus retinitis using the FlexMAP 3D (Luminex®) platform using the Milliplex Human Cytokine® kit. Aqueous humor samples were collected from these patients at different time points (pre-treatment and at 4-weekly intervals through the 12-week course of intravitreal ganciclovir treatment) and 41 cytokine levels were analyzed at each time point. CMV DNA viral load was assessed in 8 patients at different time points throughout the course of ganciclovir treatment. The data described herein is related to the research article entitled "Aqueous humor immune factors and cytomegalovirus (CMV) levels in CMV retinitis through treatment - The CRIGSS study" (Iyer et al., 2016) [1]. Cytokine levels against the different time points which indicate the response to the given treatment and against the CMV viral load were analyzed.