Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: an economic analysis modelling study
Open accessAbstract
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…
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Xie et al. and Huang et al. showed that performance differences between semi-automated and fully automated AI models directly impacted cost per patient and ICERs.
“the performance differences between semi-automated and fully automated AI models—as shown by Xie et al.23 and Huang et al.36 —directly impacted cost per patient and ICERs”
From Systematic review of cost effectiveness and budget impact of artificial intelligence in healthcare (2025) · cites this paper as Xie et al.23
Xie et al. and Huang et al. found that AI-driven models for diabetic retinopathy screening reduced per-patient screening costs by 14-19.5% and achieved ICERs as low as $1107.63 per QALY.
“both Xie et al.23 and Huang et al.36 found that AI-driven models reduced per-patient screening costs by 14–19.5% and achieved ICERs as low as $1107.63 per QALY.”
From Systematic review of cost effectiveness and budget impact of artificial intelligence in healthcare (2025) · cites this paper as Xie et al.23
AI technology can increase the productivity of existing DR tele-ophthalmology screening programs and lower unit economic costs within the community.
“For example, with the growing burden of DR, AI technology can increase the productivity of existing DR tele-ophthalmology screening programs and lower unit economic costs within the community.123”
From Artificial intelligence in diabetes management: Advancements, opportunities, and challenges (2023) · cites this paper as [123]
Without a screening program, all patients diagnosed with type 1 diabetic mellitus and type 2 diabetic mellitus are referred for an annual eye examination.
“Without a screening program, all patients diagnosed with type 1 diabetic mellitus and type 2 diabetic mellitus are referred for an annual eye examination;46”
From Autonomous Artificial Intelligence in Diabetic Retinopathy Testing—Lessons Learned on Successful Health System Adoption (2026) · cites this paper as [46]
In Singapore, autonomous testing costs $11-15 less per patient per year, leading to an annual national savings of $15 million by 2050.
“In Singapore, it is estimated that autonomous testing costs $11-15 less per patient per year, leading to an annual national savings of $15 million by 2050.63”
From Autonomous Artificial Intelligence in Diabetic Retinopathy Testing—Lessons Learned on Successful Health System Adoption (2026) · cites this paper as [63]
Artificial intelligence algorithms have demonstrated a high sensitivity of 95% to 100% in clinical implementation, whereas specificity is relatively lower and with greater variation at 60% to 90%.
“Artificial intelligence algorithms have demonstrated a high sensitivity of 95% to 100% in clinical implementation, whereas specificity is relatively lower and with greater variation at 60% to 90%.38,40,41,46”
From Autonomous Artificial Intelligence in Diabetic Retinopathy Testing—Lessons Learned on Successful Health System Adoption (2026) · cites this paper as [46]
At Stanford, a hybrid workflow routing AI-ungradable nonmydriatic images to retina specialists at the Stanford Reading Center increases overall gradability of nonmydriatic images to 95.6% of encounters.
“With this process, the overall gradability of nonmydriatic images increases to 95.6% of encounters, as the human expert graders in this system are experienced retina specialists and have greater capacity for interpreting lower quality nonmydriatic images than the AI is currently programmed to accept.46”
From Autonomous Artificial Intelligence in Diabetic Retinopathy Testing—Lessons Learned on Successful Health System Adoption (2026) · cites this paper as [46]
Using autonomous AI for diabetic retinopathy screening has been shown to be lower cost in other countries.
“using autonomous AI for DR screening has been shown to be lower cost in other countries.13–16”
From Screening for diabetic retinopathy with artificial intelligence in a primary care setting: a comparative cost analysis (2026) · cites this paper as [13]
Increasing AI sensitivity increases detection of high-risk patients but may raise medical costs.
“Improving AI’s sensitivity finds more high-risk patients but may raise medical costs while increasing specificity reduces unnecessary referrals but may weaken detection capability.”
From Economic evaluation for medical artificial intelligence: accuracy vs. cost-effectiveness in a diabetic retinopathy screening case (2024) · cites this paper as [13]
Increasing AI specificity reduces unnecessary referrals and associated costs but may compromise detection capability.
“Improving AI’s sensitivity finds more high-risk patients but may raise medical costs while increasing specificity reduces unnecessary referrals but may weaken detection capability.”
From Economic evaluation for medical artificial intelligence: accuracy vs. cost-effectiveness in a diabetic retinopathy screening case (2024) · cites this paper as [13]
Cites
Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement
Husereau, D., Drummond, M., Petrou, S., Carswell, C., Moher, D., Greenberg, D., Augustovski, F., Briggs, A. H., Mauskopf, J., Loder, E., et al. (2013) · Cost Effectiveness and Resource Allocation · doi:10.1186/1478-7547-11-6 · Cited by 381 · see citing works
Prevalence of diabetic retinopathy in type 2 diabetes in developing and developed countries
Ruta, L. M., Magliano, D. J., LeMesurier, R., Taylor, H. R., Zimmet, P. Z., & Shaw, J. E. (2013) · Diabetic Medicine · doi:10.1111/dme.12119 · Cited by 311 · see citing works
Gulshan, V., Rajan, R. P., Widner, K., Wu, D., Wubbels, P., Rhodes, T., Whitehouse, K., Coram, M., Corrado, G., Ramasamy, K., et al. (2019) · JAMA Ophthalmology · doi:10.1001/jamaophthalmol.2019.2004 · Cited by 303 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Rajpurkar, P., Chen, E., Banerjee, O., & Topol, E. J. (2022) · Nature Medicine · doi:10.1038/s41591-021-01614-0 · Cited by 3,212 · see citing works
Artificial intelligence in healthcare: transforming the practice of medicine
Bajwa, J., Munir, U., Nori, A., & Williams, B. (2021) · Future Healthcare Journal · doi:10.7861/fhj.2021-0095 · Cited by 1,856 · see citing works
Digital technology, tele-medicine and artificial intelligence in ophthalmology: A global perspective
Li, J.-P. O., Liu, H., Ting, D. S. J., Jeon, S., Chan, R. V. P., Kim, J. E., Sim, D. A., Thomas, P. B. M., Lin, H., Chen, Y., et al. (2021) · Progress in Retinal and Eye Research · doi:10.1016/j.preteyeres.2020.100900 · Cited by 626 · see citing works
Self-citationArtificial intelligence in diabetes management: Advancements, opportunities, and challenges
Guan, Z., Li, H., Liu, R., Cai, C., Liu, Y., Li, J., Wang, X., Huang, S., Wu, L., Liu, D., et al. (2023) · Cell Reports Medicine · doi:10.1016/j.xcrm.2023.101213 · Cited by 287 · see citing works
“123 This will enable providers to project expertise to geographically remote locations, improve the productivity of existing resources, and optimize the flow of patients within entire health systems.”
BackgroundRuamviboonsuk, P., Tiwari, R., Sayres, R., Nganthavee, V., Hemarat, K., Kongprayoon, A., Raman, R., Levinstein, B., Liu, Y., Schaekermann, M., et al. (2022) · The Lancet Digital Health · doi:10.1016/s2589-7500(22)00017-6 · Cited by 250 · see citing works
Gomez Rossi, J., Rojas-Perilla, N., Krois, J., & Schwendicke, F. (2022) · JAMA Network Open · doi:10.1001/jamanetworkopen.2022.0269 · Cited by 141 · see citing works
Systematic review of cost effectiveness and budget impact of artificial intelligence in healthcare
El Arab, R. A., & Al Moosa, O. A. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01722-y · Cited by 149 · see citing works
“Similarly, regional cost differences are highlightedinstudiesonDRscreening,whereSingapore 23 andruralChina 35 reported how optimized integration and regional economic context directly in fl uenced the generalizability of AI bene fi ts, reinforcing the thematic understanding of contextual variability in AI economic evalu…”
InfluentialOng, J. C. L., Seng, B. J. J., Law, J. Z. F., Low, L. L., Kwa, A. L. H., Giacomini, K. M., & Ting, D. S. W. (2024) · Cell Reports Medicine · doi:10.1016/j.xcrm.2023.101356 · Cited by 111 · see citing works
Self-citationCost-effectiveness of artificial intelligence screening for diabetic retinopathy in rural China
Huang, X.-M., Yang, B.-F., Zheng, W.-L., Liu, Q., Xiao, F., Ouyang, P.-W., Li, M.-J., Li, X.-Y., Meng, J., Zhang, T.-T., et al. (2022) · BMC Health Services Research · doi:10.1186/s12913-022-07655-6 · Cited by 89 · see citing works
Wang, Y., Liu, C., Hu, W., Luo, L., Shi, D., Zhang, J., Yin, Q., Zhang, L., Han, X., & He, M. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01032-9 · Cited by 76 · see citing works
Wolf, R. M., Liu, T. Y. A., Thomas, C., Prichett, L., Zimmer-Galler, I., Smith, K., Abramoff, M. D., & Channa, R. (2021) · Diabetes Care · doi:10.2337/dc20-1671 · Cited by 62 · see citing works
Lin, S., Ma, Y., Xu, Y., Lu, L., He, J., Zhu, J., Peng, Y., Yu, T., Congdon, N., & Zou, H. (2023) · JMIR Public Health and Surveillance · doi:10.2196/41624 · Cited by 52 · see citing works
Patel, D., Ananthakrishnan, A., Lin, T., Channa, R., Liu, T. Y. A., & Wolf, R. M. (2022) · Journal of Clinical Medicine · doi:10.3390/jcm11237120 · Cited by 43 · see citing works
Dow, E. R., Khan, N. C., Chen, K. M., Mishra, K., Perera, C., Narala, R., Basina, M., Dang, J., Kim, M., Levine, M., et al. (2023) · Ophthalmology Science · doi:10.1016/j.xops.2023.100330 · Cited by 40 · see citing works
Yang, Q., Bee, Y. M., Lim, C. C., Sabanayagam, C., Yim-Lui Cheung, C., Wong, T. Y., Ting, D. S. W., Lim, L.-L., Li, H., He, M., et al. (2025) · eClinicalMedicine · doi:10.1016/j.eclinm.2025.103089 · Cited by 38 · see citing works
Self-citationIncorporating Artificial Intelligence into Healthcare Workflows: Models and Insights
Dai, T., & Abràmoff, M. D. (2023) · Tutorials in Operations Research: Advancing the Frontiers of OR/MS: From Methodologies to Applications · doi:10.1287/educ.2023.0257 · Cited by 29 · see citing works
Li, H., Li, G., Li, N., Liu, C., Yuan, Z., Gao, Q., Hao, S., Fan, S., & Yang, J. (2023) · PLOS ONE · doi:10.1371/journal.pone.0291390 · Cited by 27 · see citing works
Teng, C. W., Patel, S. D., Barkmeier, A. J., Liu, T. Y. A., Myung, D., Henderer, J., Liu, J., Hansen, E., & Al-Aswad, L. A. (2026) · Ophthalmology Science · doi:10.1016/j.xops.2025.100935 · Cited by 23 · see citing works
Wewetzer, L., Held, L. A., Goetz, K., & Steinhäuser, J. (2023) · DIGITAL HEALTH · doi:10.1177/20552076231176644 · Cited by 23 · see citing works
Wolf, R. M., Channa, R., Lehmann, H. P., Abramoff, M. D., & Liu, T. Y. A. (2024) · Clinical Diabetes · doi:10.2337/cd23-0019 · Cited by 12 · see citing works
Di Matteo, L., Whitestone, N., & Bhambhwani, V. (2026) · BMJ Open Ophthalmology · doi:10.1136/bmjophth-2026-002904
- Authors
- Xie, Y., Nguyen, Q. D., Hamzah, H., Lim, G., Bellemo, V., Gunasekeran, D. V., Yip, M. Y. T., Qi Lee, X., Hsu, W., Li Lee, M., Tan, C. S., Tym Wong, H., Lamoureux, E. L., Tan, G. S. W., Wong, T. Y., Finkelstein, E. A., & Ting, D. S. W.
- Affiliations
- Singapore National Eye Center (SG)
- National University of Singapore (SG)
- Duke-NUS Medical School (SG)
- Tan Tock Seng Hospital (SG)
- SingHealth (SG)
- Corresponding author
- Ting, Daniel S W
- ORCID
- Venue
- The Lancet Digital Health
- Publisher
- Elsevier BV
- Published
- 2020-05
- Volume
- 2
- Issue
- 5
- Pages
- e240-e249
- ISSN
- 2589-7500 (ISSN-L)
- OpenAlex
- W3020619771
- Cited by
- 288
- References
- 3
- Citations by year
- 2026: 31 · 2025: 52 · 2024: 57 · 2023: 47 · 2022: 48 · 2021: 42 · 2020: 9
- FWCI
- 17.78
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- The Lancet Digital Health (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- National University of Singapore (submittedVersion, cc-by-nc-nd)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Retinal Imaging and Analysis
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Health Systems, Economic Evaluations, Quality of Life
- Health Sciences › Medicine › Ophthalmology › Ophthalmology and Visual Health Research
- Keywords
- diabetic retinopathy screening, teleophthalmology, deep learning, artificial intelligence, cost-minimization analysis, semi-automated screening, automated screening, fundus images, triage, diabetes mellitus, health system costs
- MeSH terms
- Machine Learning, Adult, Aged, Artificial Intelligence, Cost-Benefit Analysis, Decision Trees, Diabetes Mellitus, Diabetic Retinopathy, Diagnostic Techniques, Ophthalmological, Humans, Image Processing, Computer-Assisted, Mass Screening, Middle Aged, Models, Biological, Ophthalmology, Photography, Physical Examination, Retina, Sensitivity and Specificity, Singapore, Health Care Costs, 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.
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
