Simple, Mobile-based Artificial Intelligence Algo<b>r</b>ithm in the detection of Diabetic Retinopathy (SMART) study

Sosale, Bhavana; Aravind, S.R.; Murthy, Hemanth; Narayana, Srikanth; Sharma, Usha; Gowda, Sahana G V; Naveenam, Muralidhar

doi:10.1136/bmjdrc-2019-000892

Cited by 60 publications

(46 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…At individual level, we report high sensitivity for ARIA in detecting any DR (91.4%). This is similar to that reported by Rajalakshmi et al (2018) where they used an alternate AI system, (EyeArt), at a tertiary care diabetes centre in India, capturing three retinal images using a smartphone‐based device and reported a sensitivity of 95.8% for detection of any DR. Another study, using the Medios AI system, captured two retinal images and reported similar sensitivity and specificity values 26 . The high sensitivity produced for any DR in this study suggests that the ARIA used is a useful primary‐care tool for triaging to human graders and ultimately reducing the workload, time, cost and variability of human DR grading.…”

Section: Discussionsupporting

confidence: 54%

Facilitating diabetic retinopathy screening using automated retinal image analysis in underresourced settings

et al. 2021

View full text Add to dashboard Cite

Aim To evaluate an automated retinal image analysis (ARIA) of indigenous retinal fundus images against a human grading comparator for the classification of diabetic retinopathy (DR) status. Methods Indigenous Australian adults with type 2 diabetes (n = 410) from three remote and very remote primary‐care services in the Northern Territory, Australia, underwent teleretinal DR screening. A single, central retinal fundus photograph (opportunistic mydriasis) for each eye was later regraded using a single ARIA and a UK human grader and national DR classification system. The sensitivity and specificity of ARIA were assessed relative to the comparator. Proportionate agreement and a Kappa statistic were also computed. Results Retinal images from 391 and 393 participants were gradable for ‘Any DR’ by the human grader and ARIA grader, respectively. ‘Any DR’ was detected by the human grader in 185 (47.3%) participants and by ARIA in 202 (48.6%) participants (agreement =88.0%, Kappa = 0.76,), whereas proliferative DR was detected in 31 (7.9%) and 37 (9.4%) participants (agreement = 98.2%, Kappa = 0.89,), respectively. The ARIA software had 91.4 (95% CI, 86.3–95.0) sensitivity and 85.0 (95% CI, 79.3–89.5) specificity for detecting ‘Any DR’ and 96.8 (95% CI, 83.3–99.9) sensitivity and 98.3 (95% CI, 96.4–99.4) specificity for detecting proliferative DR. Conclusions This ARIA software has high sensitivity for detecting ‘Any DR’, hence could be used as a triage tool for human graders. High sensitivity was also found for detection of proliferative DR by ARIA. Future versions of this ARIA should include maculopathy and referable DR (CSME and/or PDR). Such ARIA software may benefit diabetes care in less‐resourced regions.

show abstract

Section: Discussionsupporting

confidence: 54%

Facilitating diabetic retinopathy screening using automated retinal image analysis in underresourced settings

et al. 2021

View full text Add to dashboard Cite

show abstract

“…10 Finally, artificial intelligence coupled with telemedicine and mHealth could help to reach populations with poor health system access, either because of geographical isolation or scarcity of human resources for eye health. 280,284 There is little information on implementing artificial intelligence in eye health delivery and even less on whether it could improve care or outcomes. Before deployment into routine clinical practice, artificial intelligence applications require external validation on data that were not previously used in algorithm development and should undergo rigorous safety and efficacy testing in prospective clinical studies.…”

Section: Artificial Intelligencementioning

confidence: 99%

Section: Section 6: Beyond 2020—delivering High-quality Universal Eyementioning

confidence: 99%

The Lancet Global Health Commission on Global Eye Health: vision beyond 2020

Burton

Ramke

Marques

et al. 2021

The Lancet Global Health

818

572

View full text Add to dashboard Cite

Eye health is essential to achieve the Sustainable Development Goals; vision needs to be reframed as a development issue There is extensive evidence showing that improving eye health contributes directly and indirectly to achieving many Sustainable Development Goals, including reducing poverty and improving work productivity, general and mental health, and education and equity. Improving eye health is a practical and cost-effective way of unlocking human potential. Eye health needs to be reframed as an enabling, cross-cutting issue within the sustainable development framework. Almost everyone will experience impaired vision or an eye condition during their lifetime and require eye care services; urgent action is necessary to meet the rapidly growing eye health need In 2020, 1•1 billion people had distance vision impairment or uncorrected presbyopia. By 2050, this figure is expected to rise to 1•8 billion. Most affected people live in low-income and middle-income countries (LMICs) with avoidable causes of vision impairment. During the life course, most people will experience vision impairment, even if just the need for reading glasses. Because of unmet needs and an ageing global population, eye health is a major public health and sustainable development concern which warrants urgent political action. Eye health is an essential component of universal health coverage; it must be included in planning, resourcing, and delivery of health care Universal health coverage is not universal without affordable, high quality, equitable eye care. In line with the WHO World report on vision, we urge countries to consider eye care as an essential service within universal health coverage. To deliver comprehensive services including promotion, prevention, treatment, and rehabilitation, eye care needs to be included in national strategic health plans and development policies, health financing structures, and health workforce planning. Coordinated intersectoral action is needed to systematically improve population eye health, also within healthy ageing initiatives, schools, and the workplace. Integration of eye health services with multiple relevant components of health service delivery and at all levels of the health system is of central importance.

show abstract

“…Bias could have been introduced by poor reporting of patient characteristics of the included studies. Finally, except for Abramoff et al’s trial [ 15 ] and 5 other prospectively conducted studies [ 48 , 52 , 55 , 60 , 66 ], all other studies on ML-based DR diagnosis were validated by retrospective data. Due to spectrum bias, an overestimation of ML’s performance in a real-world setting is possible and should be considered.…”

Section: Discussionmentioning

confidence: 99%

Performance and Limitation of Machine Learning Algorithms for Diabetic Retinopathy Screening: Meta-analysis

Wu¹,

Liu²,

Hsu³

et al. 2021

J Med Internet Res

View full text Add to dashboard Cite

Background Diabetic retinopathy (DR), whose standard diagnosis is performed by human experts, has high prevalence and requires a more efficient screening method. Although machine learning (ML)–based automated DR diagnosis has gained attention due to recent approval of IDx-DR, performance of this tool has not been examined systematically, and the best ML technique for use in a real-world setting has not been discussed. Objective The aim of this study was to systematically examine the overall diagnostic accuracy of ML in diagnosing DR of different categories based on color fundus photographs and to determine the state-of-the-art ML approach. Methods Published studies in PubMed and EMBASE were searched from inception to June 2020. Studies were screened for relevant outcomes, publication types, and data sufficiency, and a total of 60 out of 2128 (2.82%) studies were retrieved after study selection. Extraction of data was performed by 2 authors according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), and the quality assessment was performed according to the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Meta-analysis of diagnostic accuracy was pooled using a bivariate random effects model. The main outcomes included diagnostic accuracy, sensitivity, and specificity of ML in diagnosing DR based on color fundus photographs, as well as the performances of different major types of ML algorithms. Results The primary meta-analysis included 60 color fundus photograph studies (445,175 interpretations). Overall, ML demonstrated high accuracy in diagnosing DR of various categories, with a pooled area under the receiver operating characteristic (AUROC) ranging from 0.97 (95% CI 0.96-0.99) to 0.99 (95% CI 0.98-1.00). The performance of ML in detecting more-than-mild DR was robust (sensitivity 0.95; AUROC 0.97), and by subgroup analyses, we observed that robust performance of ML was not limited to benchmark data sets (sensitivity 0.92; AUROC 0.96) but could be generalized to images collected in clinical practice (sensitivity 0.97; AUROC 0.97). Neural network was the most widely used method, and the subgroup analysis revealed a pooled AUROC of 0.98 (95% CI 0.96-0.99) for studies that used neural networks to diagnose more-than-mild DR. Conclusions This meta-analysis demonstrated high diagnostic accuracy of ML algorithms in detecting DR on color fundus photographs, suggesting that state-of-the-art, ML-based DR screening algorithms are likely ready for clinical applications. However, a significant portion of the earlier published studies had methodology flaws, such as the lack of external validation and presence of spectrum bias. The results of these studies should be interpreted with caution.

show abstract

Simple, Mobile-based Artificial Intelligence Algorithm in the detection of Diabetic Retinopathy (SMART) study

Cited by 60 publications

References 18 publications

Facilitating diabetic retinopathy screening using automated retinal image analysis in underresourced settings

Facilitating diabetic retinopathy screening using automated retinal image analysis in underresourced settings

The Lancet Global Health Commission on Global Eye Health: vision beyond 2020

Performance and Limitation of Machine Learning Algorithms for Diabetic Retinopathy Screening: Meta-analysis

Contact Info

Product

Resources

About