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.

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.

The longitudinal psychological, physical activity, and financial impact of a COVID-19 lockdown on older adults in Singapore: The PIONEER-COVID population-based study.

BACKGROUND: Several countries have implemented 'lockdown' measures to curb the spread of the coronavirus disease 2019 (COVID-19). AIMS: To examine the psychological, physical activity (PA), and financial impact of a 2-month COVID-19 lockdown on older adults aged ≥60 years in Singapore, and to identify factors associated with adverse lockdown-related outcomes. METHOD: We interviewed 496 community-dwelling adults (mean age [standard deviation]: 73.8 [7.6] years; 54.8% female) during the lockdown who had previously participated in a population-based epidemiological study. Validated questionnaires were utilised to assess loneliness and depressive symptoms at both timepoints, while inhouse questionnaires were used to assess PA and financial difficulty during lockdown. Multivariable regression models determined the lockdown-related change in loneliness and depression scores, and the factors associated with adverse outcomes. RESULTS: Loneliness increased significantly during the lockdown period (p < 0.001) while depressive symptoms decreased (p = 0.022). Decreased PA, greater financial problems, male gender, Indian ethnicity, living alone, having a greater body mass index and perceived susceptibility to COVID-19 were all associated with worsening loneliness scores. A total of 36.9% and 19.6% participants reported decreased PA and had financial problems during the lockdown, respectively. Unemployment was associated with decreased PA, while self-employed individuals, cleaners, retail workers and smokers had greater odds of experiencing financial difficulty. CONCLUSION: Despite a decrease in depressive symptoms, our population of older Asians reported a significant increase in loneliness and decreased PA, with one-fifth experiencing financial problems during lockdown. Our data suggest that more targeted public health efforts are needed to reduce repercussions of future lockdowns.

Knowledge, attitudes, and practice towards COVID-19 among multi-ethnic elderly Asian residents in Singapore: a mixed methods study.

INTRODUCTION: We investigated knowledge, attitudes, and practice (KAP) about COVID-19 and related preventive measures in Singaporeans aged ≥ 60 years. METHODS: This was a population-based, cross-sectional, mixed-methods study (13 May 2020-9 June 2020) of participants aged ≥60 years. Self-reported KAP about ten COVID-19 symptoms and six government-endorsed preventive measures were evaluated. Multivariable regression models identified sociodemographic and health-related factors associated with knowledge, attitudes and practices in our sample. Associations between knowledge/attitude scores, and practice categories were determined using logistic regression. 78 participants were interviewed qualitatively about practice of additional preventive measures and data were analysed thematically. RESULTS: Mean awareness score of the symptoms was 7.2/10. Fever (93.0%) and diarrhoea (33.5%) were the most- and least-known symptoms, respectively. Most knew all six preventive measures (90.4%), perceived them as effective (78.7%), and practiced 'wear a mask' (97.2%). Indians, Malays, and those in smaller housing had poorer mean knowledge of COVID-19 symptoms scores. Older participants had poorer attitudes towards preventive measures. Compared to Chinese, Indians had lower odds of practicing 3/6 recommendations. A one-point increase in knowledge of and attitudes towards preventive measures score had higher odds of always practicing 3/6 and 2/6 measures, respectively. Qualitative interviews revealed use of other preventive measures, e.g. maintaining a healthy lifestyle. CONCLUSION: Elderly Singaporeans displayed high levels of KAP about COVID-19 and related preventive measures, with a positive association between levels of knowledge/attitude, and practice. However, important ethnic and socioeconomic disparities were evident, suggesting key vulnerabilities remain, requiring immediate attention.

A Peer-to-Peer Live-Streaming Intervention for Children During COVID-19 Homeschooling to Promote Physical Activity and Reduce Anxiety and Eye Strain: Cluster Randomized Controlled Trial.

BACKGROUND: The COVID-19 pandemic has led to worldwide school closures, with millions of children confined to online learning at home. As a result, children may be susceptible to anxiety and digital eye strain, highlighting a need for population interventions. OBJECTIVE: The objective of our study was to investigate whether a digital behavior change intervention aimed at promoting physical activity could reduce children's anxiety and digital eye strain while undergoing prolonged homeschooling during the COVID-19 pandemic. METHODS: In this cluster randomized controlled trial, homeschooled grade 7 students at 12 middle schools in southern China were recruited through local schools and randomly assigned by the school to receive (1:1 allocation): (1) health education information promoting exercise and ocular relaxation, and access to a digital behavior change intervention, with live streaming and peer sharing of promoted activities (intervention), or (2) health education information only (control). The primary outcome was change in self-reported anxiety score. Secondary outcomes included change in self-reported eye strain and sleep quality. RESULTS: On March 16, 2020, 1009 children were evaluated, and 954 (94.5%) eligible children of consenting families were included in the intention-to-treat analysis. Children in the intervention (n=485, 6 schools) and control (n=469, 6 schools) groups were aged 13.5 (SD 0.5) years, and 52.3% (n=499) were male. The assigned interventions were completed by 896 children (intervention: n=467, 96.3%; control: n=429, 91.5%). The 2-week change in square-root-transformed self-reported anxiety scores was greater in the intervention (-0.23, 95% CI -0.27 to -0.20) vs control group (0.12, 95% CI 0.09-0.16; unadjusted difference -0.36, 95% CI -0.63 to -0.08; P=.02). There was a significant reduction in square-root-transformed eye strain in the intervention group (-0.08, 95% CI -0.10 to 0.06) compared to controls (0.07, 95% CI 0.05-0.09; difference -0.15, 95% CI -0.26 to -0.03; P=.02). Change in sleep quality was similar between the two groups. CONCLUSIONS: This digital behavior change intervention reduced children's anxiety and eye strain during COVID-19-associated online schooling. TRIAL REGISTRATION: ClinicalTrials.gov NCT04309097; http://clinicaltrials.gov/ct2/show/NCT04309097.