Diabetic retinopathy screening model in low and middle-income countries: a scoping review
Open accessThe DOI matches its Crossref record; 49 of the 49 references it cites were found in a registry or an index; 49 of the 90 sources its claims cite were checked for support.
Abstract
BACKGROUND: Diabetic retinopathy (DR), a leading cause of blindness in working-age adults, disproportionately affects low- and middle-income countries (LMICs). While preventable through early intervention, DR screening programs are often lacking in these resource-constrained settings. OBJECTIVE: This scoping review examines DR screening models implemented in LMICs, identifying evidence, research gaps, and potential improvement strategies. METHODS: A literature search across multiple databases identified studies on DR screening in LMICs, limited to peer-reviewed articles from the past 20 years focusing on DR screening model effectiveness or implementation. Key data (study design, screening techniques, outcomes) were extracted and synthesized narratively. RESULTS: This review synthesized 30 studies on DR screening in LMICs, mainly from India, South Africa, Pakistan, and Bangladesh. These studies explored diverse screening methods, from traditional techniques (ophthalmoscopy, funduscopy, slit lamp exams) to telemedicine and AI. Non-mydriatic fundus photography, often with AI-assisted grading, was common, as were remote grading and store-and-forward systems. Given the resource constraints in these settings, non-mydriatic methods were often preferred. Many studies optimized resources by training non-physicians for image acquisition, followed by specialist grading. The reviewed studies highlighted the effectiveness of community-based screening programs in expanding coverage and improving patient adherence. Furthermore, they demonstrated the cost-effectiveness of smartphone-based imaging devices and AI-driven systems. However, challenges remained, including limited infrastructure, inconsistent training, and follow-up difficulties. CONCLUSIONS: LMIC DR screening utilizes traditional and innovative technologies, with community-based approaches, telemedicine, and AI enhancing reach and accuracy. Transitioning from case-finding to population-based screening requires stronger diabetes surveillance and integration within primary care.
Overview
CoChat Summary
TL;DR
This scoping review synthesized 30 studies on diabetic retinopathy (DR) screening models in low- and middle-income countries (LMICs), covering traditional and innovative approaches such as community-based screening, telemedicine, and AI. It concludes that LMICs need stronger diabetes surveillance and integration of DR screening into primary care to move from opportunistic case-finding toward population-based screening.
Key findings
- The review included 30 studies published between 2003 and 2023, mainly from India (n = 12), South Africa (n = 2), Pakistan (n = 2), and Bangladesh (n = 2). Across studies, patients included 67,235 with Type 2 diabetes, 762 with Type 1 diabetes, and 150,370 with unspecified diabetes.
- Non-mydriatic fundus photography, often with AI-assisted grading, was common, as were remote grading and store-and-forward systems. Mydriasis was used in 17/30 studies; image acquisition was often by trained non-physician staff, while final grading was typically by ophthalmologists or retinal specialists in 22/30 studies.
- Screening models performed strongly overall: sensitivity 73.5–95.8%, NPV 93.5–98%, specificity 64.2–99.6%, PPV 32.1–89.7%, and AUC 0.878–0.99. AI software showed sensitivity up to 95.8% (95% CI 92.9–98.7) and AUC for referable DR up to 0.97.
- Challenges included limited infrastructure, inconsistent training, follow-up difficulties, and ungradable image rates from 0% to 37.8%. Only 5% of data came from low-income countries, with the rest from middle-income countries.
Methods
- Scoping review following PRISMA-ScR guidance. Databases searched on April 21, 2025 included EMBASE, Google Scholar, Global Health, MEDLINE, SCOPUS, PUBMED, CINAHL, and Web of Science. Inclusion criteria were peer-reviewed English articles from the past 20 years on adults with DR in LMICs and implementation of specified screening models; commentaries, editorials, non-empirical literature, pediatric populations, and other eye diseases were excluded.
- From 1,225 records, 298 duplicates were removed, 927 were screened, 119 full texts were assessed, and 30 studies were included. Three independent reviewers screened titles/abstracts, with a fourth reviewer resolving discrepancies.
- Data were extracted into an Excel standardized form and synthesized descriptively/narratively, covering settings, populations, screening techniques, mydriasis, personnel, image acquisition, remote grading, AI integration, and performance metrics. Methodological quality was not appraised in detail.
Limitations
- Stated by the authors: only 5% of data came from low-income countries, with the rest from middle-income countries, limiting global representation and generalizability; as a scoping review, it did not conduct detailed methodological quality appraisal; and low-income country infrastructure, resources, and demographics may differ significantly from middle-income settings.
- Stated by the authors: there is a lack of comprehensive data on diabetes prevalence and screening coverage, and many LMICs lack national diabetes registries, making it unclear what proportion of people with diabetes are reached for initial or repeat screening; many DR programs therefore function as opportunistic case-finding rather than systematic screening.
- Evident from the text: the search was confined to English-language sources and excluded grey literature, which may omit relevant local or non-English evidence; the absence of formal quality appraisal also limits confidence in the included studies’ findings.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
19 Valid · 1 Not valid · 5 Not checked · 25 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Various severity scales are used to classify diabetic retinopathy (DR), with the International Clinical Diabetic Retinopathy (ICDR) severity scale being the most commonly utilized in 15 out of 30 studies [9, 11–13, 15, 18, 21–23, 32, 34–36].”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Results > Study characteristics
What the sources say
- Not matchedcites [9] Mash B, Powell D, du Plessis F, van Vuuren U, Michalowska M, Levitt N. Screening for diabetic retinopathy in primary care with a mobile fundal camera–evaluation of a South African pilot project. S Afr Med J. 2007;97:1284–8.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Supportedcites [11] Rani PK, Raman R, Sharma V, Mahuli SV, Tarigopala A, Sudhir RR, et al. Analysis of a comprehensive diabetic retinopathy screening model for rural and urban diabetics in developing countries. Br J Ophthalmol. 2007;91:1425–9. https://doi.org/10.1136/bjo.2007.120659. source
The Methods section of Rani et al. (2007) states that 'Diabetic retinopathy was clinically graded following the norms of the international clinical diabetic retinopathy and diabetic macular oedema severity scales' (credited to Wilkinson et al. 2003, ref 10), and Table 3 reports grading into non-referable, severe NPDR, PDR and CSMO consistent with that scale — so this study is indeed one of the studies that used the ICDR severity scale, which is what the citation is offered for. The tally '15 out of 30 studies / most commonly utilised' is the citing review's own count across its reference list (the multiple citations appended), not a figure this single screening paper reports, and our copy contains no such pooled count. (LLM, conf=0.85)
- Not checked — no full text of the sourcecites [12] Khan T, Bertram MY, Jina R, Mash B, Levitt N, Hofman K. Preventing diabetes blindness: cost effectiveness of a screening programme using digital non-mydriatic fundus photography for diabetic retinopathy in a primary health care setting in South Africa. Diabetes Res Clin Pract. 2013;101:170–6. https://doi.org/10.1016/j.diabres.2013.05.006. source
- Supportedcites [13] Omar FJ, Sheeladevi S, Rani PK, Ning G, Kabona G. Evaluating the effectiveness of opportunistic eye screening model for people with diabetes attending diabetes clinic at Mnazi Mmoja hospital, Zanzibar. BMC Ophthalmol. 2014;14:81. https://doi.org/10.1186/1471-2415-14-81. source
The Methods section states that this study graded diabetic retinopathy using the International classification of Diabetic Retinopathy severity grading scale. This supports citing the paper as one study that used the scale; the 15-of-30 tally is the citing authors’ synthesis, not a finding attributed to this individual study. (LLM, conf=0.88)
- Partly supportedcites [15] Hansen MB, Abràmoff MD, Folk JC, Mathenge W, Bastawrous A, Peto T. Results of automated retinal image analysis for detection of diabetic retinopathy from the Nakuru study, Kenya. PLoS One. 2015;10:e0139148. https://doi.org/10.1371/journal.pone.0139148. sourceExperimental — accuracy not yet measured
The Methods section (“The grading protocol and ICDR reference standard”; page unavailable) and Table 1 show that this study used ICDR severity levels, so it supports counting this paper as one example. It does not assess 30 studies or establish that ICDR was the most commonly used scale in 15 of them. (LLM, conf=0.86)
- Not checked — no full text of the sourcecites [18] Matimba A, Woodward R, Tambo E, Ramsay M, Gwanzura L, Guramatunhu S. Tele-ophthalmology: opportunities for improving diabetes eye care in resource- and specialist-limited sub-Saharan African countries. J Telemed Telecare. 2016;22:311–6. https://doi.org/10.1177/1357633X15604083. source
- Supportedcites [21] Rajalakshmi R, Subashini R, Anjana RM, Mohan V. Automated diabetic retinopathy detection in smartphone-based fundus photography using artificial intelligence. Eye. 2018;32:1138–44. https://doi.org/10.1038/s41433-018-0064-9. source
The Methods state that this study graded retinopathy using the ICDR severity scale. In the citing sentence’s group of study citations, this source supports its inclusion as a study that used ICDR; the 15-of-30 count is the citing authors’ aggregate, not a result attributed to this individual study. (LLM, conf=0.92)
- Partly supportedcites [22] Wadhwani M, Vashist P, Singh SS, Gupta N, Malhotra S, Gupta A, et al. Diabetic retinopathy screening programme utilising non-mydriatic fundus imaging in slum populations of New Delhi, India. Trop Med Int Health. 2018;23:405–14. https://doi.org/10.1111/tmi.13039. sourceExperimental — accuracy not yet measured
Our copy confirms only the narrower part of the sentence that this source can carry: in 'Ocular examination' the optometrist 'graded the DR according to International Clinical Diabetic Retinopathy Scale (Table 1) [36]', and Table 1 reproduces the International Clinical Diabetic Retinopathy Severity Scale (Wilkinson et al. 2003). Wadhwani et al. 2018 is a single non-mydriatic screening programme in Delhi slums (8,432 participants), not a review of severity scales: no passage in our copy counts studies or states that the ICDR scale was the most commonly utilised in 15 of 30 studies (searched: 'International Clinical Diabetic Retinopathy Scale severity scale used', 'previous studies comparison of diabetic retinopathy grading scales most commonly used'). The 15-of-30 tally is therefore the citing authors' own synthesis, not this source's finding. (LLM, conf=0.70)
- Not checked — no full text of the sourcecites [23] Bellemo V, Lim ZW, Lim G, Nguyen QD, Xie Y, Yip MYT, et al. Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in africa: a clinical validation study. Lancet Digit Health. 2019;1:e35–44. https://doi.org/10.1016/S2589-7500(19)30004-4. source
- Partly supportedcites [32] Mathenge W, Whitestone N, Nkurikiye J, Patnaik JL, Piyasena P, Uwaliraye P, et al. Impact of artificial intelligence assessment of diabetic retinopathy on referral service uptake in a low-resource setting. Ophthalmol Sci. 2022;2:100168. https://doi.org/10.1016/j.xops.2022.100168. sourceExperimental — accuracy not yet measured
The Methods > AI Model section says this study used the International Clinical Disease Severity Scale to classify diabetic retinopathy, so it supports an example of the scale's use. It does not report a comparison across 30 studies or establish that 15 used the scale most commonly; searches for that count and for ICDR grading found no such passage. Page numbers were not available in our copy. (LLM, conf=0.87)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Remote grading of retinal images is widely utilized, with 18 out of 30 studies relying on this approach [10, 15, 16, 18, 21–27, 29–35, 37]”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Results > Study characteristics
What the sources say
- Not checked — no full text of the sourcecites [10] Raman R, Rani PK, Mahajan S, Paul P, Gnanamoorthy P, Krishna MS, et al. The tele-screening model for diabetic retinopathy: evaluating the influence of mydriasis on the gradability of a single-field 45° digital fundus image. Telemed e-Health. 2007;13:597–602. https://doi.org/10.1089/tmj.2006.0084. source
- Supportedcites [15] Hansen MB, Abràmoff MD, Folk JC, Mathenge W, Bastawrous A, Peto T. Results of automated retinal image analysis for detection of diabetic retinopathy from the Nakuru study, Kenya. PLoS One. 2015;10:e0139148. https://doi.org/10.1371/journal.pone.0139148. source
The Methods (Subject recruitment and imaging) state that retinal photographs from the Nakuru Eye Study in Kenya were graded by human graders at Moorfields Eye Hospitals Reading Centre in the UK. This supports citing the study as an example of remote grading; the 18-of-30 count describes the citing paper’s collection of studies, not a result attributed to this individual source. (LLM, conf=0.91)
- Partly supportedcites [16] Rajalakshmi R, Arulmalar S, Usha M, Prathiba V, Kareemuddin KS, Anjana RM, et al. Validation of smartphone based retinal photography for diabetic retinopathy screening. PLoS One. 2015;10:e0138285. https://doi.org/10.1371/journal.pone.0138285. sourceExperimental — accuracy not yet measured
The Methods describe retinal photographs graded by two specialists, and the Discussion says the smartphone system can transmit images for remote evaluation. Neither passage says grading in this study was performed remotely. As one of several citations attached to the sentence, this paper also does not establish the aggregate figure of 18 out of 30 studies. (LLM, conf=0.82)
- Not checked — no full text of the sourcecites [18] Matimba A, Woodward R, Tambo E, Ramsay M, Gwanzura L, Guramatunhu S. Tele-ophthalmology: opportunities for improving diabetes eye care in resource- and specialist-limited sub-Saharan African countries. J Telemed Telecare. 2016;22:311–6. https://doi.org/10.1177/1357633X15604083. source
- Supportedcites [21] Rajalakshmi R, Subashini R, Anjana RM, Mohan V. Automated diabetic retinopathy detection in smartphone-based fundus photography using artificial intelligence. Eye. 2018;32:1138–44. https://doi.org/10.1038/s41433-018-0064-9. source
The Methods (page not available) say retinal images were shared through the cloud with EyeNuk for automated analysis by its cloud-based EyeArt software. This supports citing the paper as an example of remote image grading; the 18-of-30 figure is the citing document’s tally across studies, not a result attributed to this individual paper. (LLM, conf=0.88)
- Could not verifycites [22] Wadhwani M, Vashist P, Singh SS, Gupta N, Malhotra S, Gupta A, et al. Diabetic retinopathy screening programme utilising non-mydriatic fundus imaging in slum populations of New Delhi, India. Trop Med Int Health. 2018;23:405–14. https://doi.org/10.1111/tmi.13039. sourceExperimental — accuracy not yet measured
Our copy of Wadhwani et al. (2018) is a single community screening programme report from the slums of Delhi, not a review pooling 30 studies: searches for 'remote grading', 'telemedicine/teleretinal', 'reading centre' and for any '18 of 30 studies' count returned no passage addressing the claim's point. The Introduction (p. 405) discusses only non-mydriatic imaging in clinical settings [26-30] and non-physician graders [34,35], and the Ocular examination section (p. 407) states the on-site optometrist viewed the transferred images on a laptop and graded DR himself, with regrading by an ophthalmologist for quality assurance (Training and quality assurance, p. 408) — so nothing in our copy supports the 'remote grading widely utilised in 18/30 studies' attribution, though absence is not contradiction. (LLM, conf=0.72)
- Not checked — no full text of the sourcecites [23] Bellemo V, Lim ZW, Lim G, Nguyen QD, Xie Y, Yip MYT, et al. Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in africa: a clinical validation study. Lancet Digit Health. 2019;1:e35–44. https://doi.org/10.1016/S2589-7500(19)30004-4. source
- Supportedcites [24] Joseph S, Kim R, Ravindran RD, Fletcher AE, Ravilla TD. Effectiveness of teleretinal imaging–based hospital referral compared with universal referral in identifying diabetic retinopathy: a cluster randomized clinical trial. JAMA Ophthalmol. 2019;137:786. https://doi.org/10.1001/jamaophthalmol.2019.1070. source
The Methods (Study Arms) describe retinal images transferred via the internet to a specialist who read and graded them remotely. This confirms that the cited trial is an example of the approach; the 18-of-30 tally is the citing paper’s count across studies, not a finding attributed to this individual trial. (LLM, conf=0.91)
- Supportedcites [25] Muqit MMK, Kourgialis N, Jackson-deGraffenried M, Talukder Z, Khetran ER, Rahman A, et al. Trends in diabetic retinopathy, visual acuity, and treatment outcomes for patients living with diabetes in a fundus photograph–based diabetic retinopathy screening program in Bangladesh. JAMA Netw Open. 2019;2:e1916285. https://doi.org/10.1001/jamanetworkopen.2019.16285. source
The Methods (Population) describe a reading center where retinal images from two screening locations were graded, supporting this paper as an example of remote image grading. The 18-of-30 figure is the citing document’s count across studies, not a result attributed to this individual paper. (LLM, conf=0.90)
- Partly supportedcites [26] Piyasena MMPN, Yip JLY, MacLeod D, Kim M, Gudlavalleti VSM. Diagnostic test accuracy of diabetic retinopathy screening by physician graders using a hand-held non-mydriatic retinal camera at a tertiary level medical clinic. BMC Ophthalmol. 2019;19:89. https://doi.org/10.1186/s12886-019-1092-3. sourceExperimental — accuracy not yet measured
The Methods (page not provided) describe physicians grading retinal images at a tertiary-level medical clinic. They do not establish that the grading was remote, and this single-study paper does not report the cited count of 18 out of 30 studies. (LLM, conf=0.82)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Every source the paper cites for this claim was checked and supports it.
In the paper
“A study conducted in India using the “Fundus on Phone” system also demonstrated lower costs compared to conventional fundus cameras [16].”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Results > Study characteristics
What the sources say
- Supportedcites [16] Rajalakshmi R, Arulmalar S, Usha M, Prathiba V, Kareemuddin KS, Anjana RM, et al. Validation of smartphone based retinal photography for diabetic retinopathy screening. PLoS One. 2015;10:e0138285. https://doi.org/10.1371/journal.pone.0138285. source
The Discussion states that the Fundus on Phone device is significantly cheaper than conventional fundus cameras and gives India-specific price comparisons: approximately one-fifth the price of a non-mydriatic camera and one-tenth to one-fifteenth that of a high-quality mydriatic camera. It also describes lower operating and maintenance costs. (LLM, conf=0.94)
boostedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Images obtained during DR screening can be non-gradable due to poor image quality, patient movement, improper equipment use, or ocular media opacities such as cataracts [42].”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Discussion
What the sources say
- Partly supportedcites [42] Ting DSW, Cheung CY, Lim G, et al. Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes. JAMA. 2017;318(22):2211–23. sourceExperimental — accuracy not yet measured
The Definition of Referable Diabetic Retinopathy section says a photograph was ungradable if more than one-third was obscured. The Discussion also describes poorer-quality images, including ungradable ones, in a diabetic retinopathy screening program. Neither passage identifies patient movement, improper equipment use, cataracts, or other ocular media opacities as causes; page numbers are unavailable in our copy. (LLM, conf=0.87)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Therefore, a hybrid approach—non-mydriatic screening with selective mydriasis for ungradable images—might optimizes accuracy and workflow [21, 43–45].”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Discussion
What the sources say
- Partly supportedcites [21] Rajalakshmi R, Subashini R, Anjana RM, Mohan V. Automated diabetic retinopathy detection in smartphone-based fundus photography using artificial intelligence. Eye. 2018;32:1138–44. https://doi.org/10.1038/s41433-018-0064-9. sourceExperimental — accuracy not yet measured
The Discussion links non-mydriatic imaging to possibly lower screening sensitivity, and the Limitations note that poor mydriasis can prevent image acquisition. But the Methods say pupils were dilated before photography for this study; the paper does not propose or test selective dilation of ungradable images, or show that a hybrid workflow improves accuracy or efficiency. (LLM, conf=0.81)
- Not matchedcites [43] Salongcay RP, Silva PS. The role of teleophthalmology in diabetic retinopathy screening. J Clin Med. 2020;9(7):2196.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not checked — no full text of the sourcecites [44] Ramasamy K, Raman R, Agarwal S, et al. Diabetic retinopathy screening using telemedicine in rural India. Telemed J E Health. 2014;20(8):748–54. https://doi.org/10.1111/j.1464-5491.2009.02797.x. source
- Not checked — no full text of the sourcecites [45] Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016;316(22):2402–10. source
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
10 more claims
Create a free account to see all 50 claims and to verify the 14 references still open.
- It may state two separate claims as one. 30
- Probably not a substantial claim. 10
- Borderline: one of the tests for a substantial claim is unclear. 1
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Methods > Eligibility criteria
“Articles published in peer-reviewed journals within the last 20 years.”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Results > Study characteristics
“The studies, published between 2003 and 2023, were conducted across Asia, Africa, and South America, with India contributing the largest number (n = 12).”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“The full texts of 119 articles were assessed, of which 90 were excluded due to incompatible study design (n = 41), non-LMIC settings (n = 28), irrelevance (n = 17), non-English language (n = 2), or availability of abstracts only (n = 2). A total of 30 studies were included in the final review[8–37].”
§ Diabetic retinopathy screening model in low and middle-income countries: a scoping review > Results > Search outcome
Reported 30 · recomputed 29
Citations
Cites
PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation
Tricco, A. C., Lillie, E., Zarin, W., O'Brien, K. K., Colquhoun, H., Levac, D., Moher, D., Peters, M. D. J., Horsley, T., Weeks, L., et al. (2018) · Annals of Internal Medicine · doi:10.7326/m18-0850 · Cited by 43,565 · cited 1×
“This scoping review was synthesized according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews guidelines [7].”
High-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450 · cited 1×
“…to protect patient information and ensure compliance with international standards Semi-automated models that combine AI with human graders have shown high accuracy at lower cost, while also reducing long-term healthcare expenditures through earlier detection and intervention [34, 46, 51, 52].”
Gulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., Venugopalan, S., Widner, K., Madams, T., Cuadros, J., et al. (2016) · JAMA · doi:10.1001/jama.2016.17216 · Cited by 7,697
Flaxman, S. R., Bourne, R. R. A., Resnikoff, S., Ackland, P., Braithwaite, T., Cicinelli, M. V., Das, A., Jonas, J. B., Keeffe, J., Kempen, J. H., et al. (2017) · The Lancet Global Health · doi:10.1016/s2214-109x(17)30393-5 · Cited by 3,477
Ting, D. S. W., Cheung, C. Y.-L., Lim, G., Tan, G. S. W., Quang, N. D., Gan, A., Hamzah, H., Garcia-Franco, R., San Yeo, I. Y., Lee, S. Y., et al. (2017) · JAMA · doi:10.1001/jama.2017.18152 · Cited by 2,402
Abràmoff, M. D., Lavin, P. T., Birch, M., Shah, N., & Folk, J. C. (2018) · npj Digital Medicine · doi:10.1038/s41746-018-0040-6 · Cited by 1,638
The Lancet Global Health Commission on Global Eye Health: vision beyond 2020
Burton, M. J., Ramke, J., Marques, A. P., Bourne, R. R. A., Congdon, N., Jones, I., Ah Tong, B. A. M., Arunga, S., Bachani, D., Bascaran, C., et al. (2021) · The Lancet Global Health · doi:10.1016/s2214-109x(20)30488-5 · Cited by 1,624
Artificial intelligence and deep learning in ophthalmology
Ting, D. S. W., Pasquale, L. R., Peng, L., Campbell, J. P., Lee, A. Y., Raman, R., Tan, G. S. W., Schmetterer, L., Keane, P. A., & Wong, T. Y. (2019) · British Journal of Ophthalmology · doi:10.1136/bjophthalmol-2018-313173 · Cited by 1,554
Artificial Intelligence and Deep Learning in Ophthalmology
Wang, Z., Keane, P. A., Chiang, M., Cheung, C. Y., Wong, T. Y., & Ting, D. S. W. (2022) · Artificial Intelligence in Medicine · doi:10.1007/978-3-030-64573-1_200 · Cited by 1,524
Guidelines on Diabetic Eye Care
Wong, T. Y., Sun, J., Kawasaki, R., Ruamviboonsuk, P., Gupta, N., Lansingh, V. C., Maia, M., Mathenge, W., Moreker, S., Muqit, M. M. K., et al. (2018) · Ophthalmology · doi:10.1016/j.ophtha.2018.04.007 · Cited by 839
Research works printed but not linked: 6
- International Diabetes Federation. IDF diabetes atlas. 10th ed. Brussels: International Diabetes Federation; 2021.
- Jani PK, Makwana HL, Patel P. Community-based screening for diabetic retinopathy using trained paramedical personnel. Indian J Ophthalmol. 2020;68(10):2200–4. https://doi.org/10.4103/0301-4738.191522. · doi:10.4103/0301-4738.191522
- Sharma A, Biswas P, Khokhar S, et al. Teleophthalmology as a model for detecting diabetic retinopathy in resource-poor settings. Indian J Ophthalmol. 2018;66(8):1109–12. https://doi.org/10.4103/0301-4738.232049. · doi:10.4103/0301-4738.232049
- Silva PS, Cavallerano JD, Aiello LM, Aiello LP. Telemedicine and diabetic retinopathy: moving forward into the new decade. Diabetes Technol Ther. 2011;13(10):1075–83.
- Salongcay RP, Silva PS. The role of teleophthalmology in diabetic retinopathy screening. J Clin Med. 2020;9(7):2196.
- Zimmer-Galler IE, Zeimer R, et al. The optos Panoramic200MA retinal imaging system: clinical effectiveness at detecting diabetic retinopathy. Eye (Lond). 2012;26(10):1427–35.
Cited by
Technology-driven diabetes care: innovation without equity?
Salomon, I., & Sam, S. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1783456 · Cited by 3
Comparação de arquiteturas CNNs e transformers na triagem automatizada de retinopatia diabética
Leite, D., Mendes, R., Custódio, A., Veloso, A. O. D. N., Ferraz, S., Ramalho, M., Câmara, J., & Moraes, R. D. (2026) · doi:10.5753/sbcas.2026.21542
Gupta, E., Aung, M. H., & Bowden, E. C. (2026) · Journal of Clinical Medicine · doi:10.3390/jcm15155827
Foroozanfar, Z., Ghadiri‐Anari, A., Azizi, R., & Mehrabbeik, A. (2026) · BMC Ophthalmology · doi:10.1186/s12886-026-05260-3 · Cited by 1
Metadata
- Authors
- Lestari, Y. D., Sitompul, R., Widyahening, I. S., Tahapary, D. L., Yusuf, P. A., Sasongko, M. B., Riono, P., Nursakina, Y., & Alexandra, G.
- Affiliations
- University of Indonesia (ID)
- Universitas Gadjah Mada (ID)
- Imperial College London (GB)
- Corresponding author
- Lestari, Yeni Dwi
- ORCID
- Venue
- BMC Public Health
- Published
- 2025-12-03
- Volume
- 25
- Issue
- 1
- Pages
- 4210-4210
- ISSN
- 1471-2458 (ISSN-L)
- PMC
- PMC12676764
- OpenAlex
- W4416949330
- Cited by
- 11
- References
- 45
- Citations by year
- 2026: 11
- FWCI
- 1.72
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by-nc-nd
- creativecommons.org/licenses/by-nc-nd/4.0 (tdm)
- creativecommons.org/licenses/by-nc-nd/4.0 (vor)
- Locations
- BMC Public Health (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Universitas Indonesia (NKB-294/UN2.RST/HKP.05.00/2024)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Retinal Imaging and Analysis
- Health Sciences › Medicine › Ophthalmology › Ophthalmology and Visual Health Research
- Health Sciences › Medicine › Ophthalmology › Retinal Diseases and Treatments
- Keywords
- diabetic retinopathy, diabetic retinopathy screening, low- and middle-income countries, telemedicine, artificial intelligence, non-mydriatic fundus photography, store-and-forward, community-based screening, smartphone-based imaging, cost-effectiveness, diabetes surveillance, primary care integration, blindness prevention
- MeSH terms
- Developing Countries, Diabetic Retinopathy, Humans, Mass Screening, Telemedicine
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Diabetic retinopathy screening model in low and middle-income countries: a scoping review. https://app.cochat.ai/papers/10.1186/s12889-025-25543-6
- The paper
- Lestari, Y. D., Sitompul, R., Widyahening, I. S., Tahapary, D. L., Yusuf, P. A., Sasongko, M. B., Riono, P., Nursakina, Y., & Alexandra, G. (2025). Diabetic retinopathy screening model in low and middle-income countries: a scoping review. BMC Public Health. https://doi.org/10.1186/s12889-025-25543-6
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.
