Multicenter, Head-to-Head, Real-World Validation Study of Seven Automated Artificial Intelligence Diabetic Retinopathy Screening Systems
Open accessAbstract
OBJECTIVE: With rising global prevalence of diabetic retinopathy (DR), automated DR screening is needed for primary care settings. Two automated artificial intelligence (AI)–based DR screening algorithms have U.S. Food and Drug Administration (FDA) approval. Several others are under consideration while in clinical use in other countries, but their real-world performance has not been evaluated systematically. We compared the performance of seven automated AI-based DR screening algorithms (including one FDA-approved algorithm) against human graders when analyzing real-world retinal imaging data. RESEARCH DESIGN AND METHODS: This was a multicenter, noninterventional device validation study evaluating a total of 311,604 retinal images from 23,724 veterans who presented for teleretinal DR screening at the Veterans Affairs (VA) Puget Sound Health Care System (HCS) or Atlanta VA HCS from 2006 to 2018. Five companies provided seven algorithms, including one with FDA approval, that independently analyzed all scans, regardless of image quality. The sensitivity/specificity of each algorithm when classifying images as referable DR or not were compared with original VA teleretinal grades and a regraded arbitrated data set. Value per encounter was estimated. RESULTS: Although high negative predictive values (82.72–93.69%) were observed, sensitivities varied widely (50.98–85.90%). Most algorithms performed no better than humans against the arbitrated data set, but two achieved higher sensitivities, and one yielded comparable sensitivity (80.47%, P = 0.441) and specificity (81.28%, P = 0.195). Notably, one had lower sensitivity (74.42%) for proliferative DR (P = 9.77 × 10−4) than the VA teleretinal graders. Value per encounter varied at $15.14–$18.06 for ophthalmologists and $7.74–$9.24 for optometrists. CONCLUSIONS: The DR screening algorithms showed significant performance differences. These results argue for rigorous testing of all such algorithms on real-world data before clinical implementation.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
Autonomous point-of-care, centralized AI or teleophthalmology, and hybrid AI-plus-human grading models cannot be universally ranked because each offers a different balance of immediacy, specialist oversight, infrastructure requirements, and workflow complexity.
“Autonomous point-of-care, centralized AI or teleophthalmology, and hybrid AI-plus-human grading models should not be ranked universally, because each offers a different balance of immediacy, specialist oversight, infrastructure requirements, and workflow complexity.”
From From validation to public health impact: implementing equitable AI-enabled diabetic retinopathy screening (2026) · cites this paper as [10]
The Conclusions discuss AI as a first screening step with ophthalmologists reviewing flagged cases, and note integration, hardware, and cloud-service considerations for the VA teleretinal system (page not provided). They do not compare autonomous point-of-care, centralized, and hybrid models or conclude that those models should not be ranked universally on the stated criteria. (LLM, conf=0.83)
Experimental — accuracy not yet measuredAutonomous AI systems can provide same-visit results and reduce dependence on asynchronous grading; centralized AI/teleophthalmology models can support multisite quality assurance and specialist oversight; hybrid AI-plus-human models retain human review for selected or uncertain cases.
“Autonomous systems can provide same-visit results and reduce dependence on asynchronous grading, whereas centralized models may support multisite quality assurance and specialist oversight; hybrid models retain human review for selected or uncertain cases (8,10,11,14,17,21).”
From From validation to public health impact: implementing equitable AI-enabled diabetic retinopathy screening (2026) · cites this paper as [10]
Split into 3 atoms — 1 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.
Experimental — accuracy not yet measuredAI-enabled screening can reproduce or worsen inequity if devices are unavailable, internet connectivity is weak, staff are inadequately trained, costs are shifted to patients, or algorithms perform differently across populations and camera systems.
“However, the same technology can reproduce or worsen inequity if devices are unavailable, internet connectivity is weak, staff are inadequately trained, costs are shifted to patients, or algorithms perform differently across populations and camera systems (10,(31)(32)(33).”
From From validation to public health impact: implementing equitable AI-enabled diabetic retinopathy screening (2026) · cites this paper as [10]
The Conclusions report performance differences between sites and warn that algorithms may not generalize across populations; they also link image quality to technician training. The Methods identify camera-system specifications as part of the acquisition protocol. The paper does not establish that unavailable devices, weak internet connectivity, or costs shifted to patients worsen inequity; its systems were tested without an internet connection. (LLM, conf=0.87)
Experimental — accuracy not yet measuredIndependent head-to-head validation has shown that performance can vary materially across algorithms even when systems are applied to the same real-world data.
“Independent head-to-head validation has shown that performance can vary materially across algorithms even when systems are applied to the same real-world data (10).”
From From validation to public health impact: implementing equitable AI-enabled diabetic retinopathy screening (2026) · cites this paper as [10]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Abràmoff, M. D., Lou, Y., Erginay, A., Clarida, W., Amelon, R., Folk, J. C., & Niemeijer, M. (2016) · Investigative Opthalmology & Visual Science · doi:10.1167/iovs.16-19964 · Cited by 1,101 · see citing works
Cruz Rivera, S., Liu, X., Chan, A.-W., Denniston, A. K., Calvert, M. J., The SPIRIT-AI and CONSORT-AI Working Group, SPIRIT-AI and CONSORT-AI Steering Group, Darzi, A., Holmes, C., Yau, C., et al. (2020) · Nature Medicine · doi:10.1038/s41591-020-1037-7 · Cited by 748 · see citing works
Self-citationPrice, M. J., Ades, A., Soldan, K., Welton, N. J., Macleod, J., Simms, I., DeAngelis, D., Turner, K. M., & Horner, P. J. (2016) · Health Technology Assessment · doi:10.3310/hta20220 · Cited by 494 · see citing works
Bhaskaranand, M., Ramachandra, C., Bhat, S., Cuadros, J., Nittala, M. G., Sadda, S. R., & Solanki, K. (2019) · Diabetes Technology & Therapeutics · doi:10.1089/dia.2019.0164 · Cited by 211 · see citing works
Flagged first, then by citations
Research works printed but not linked: 26
- [1]Lee R, Wong TY, Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis (Lond) 2015; 2: 17 26605370 PMC4657234
- [2]Liew G, Michaelides M, Bunce C. A comparison of the causes of blindness certifications in England and Wales in working age adults (16-64 years), 1999-2000 with 2009-2010. BMJ Open 2014; 4: e004015 PMC3927710 24525390
- [3]Jampol LM, Glassman AR, Sun J. Evaluation and care of patients with diabetic retinopathy. N Engl J Med 2020; 382: 1629– 1637 32320570
- [4]Flaxel CJ, Adelman RA, Bailey ST,. Diabetic retinopathy preferred practice pattern®. Ophthalmology 2020; 127: 66– P145 31757498
- [5]American Diabetes Association. 11. Microvascular complications and foot care: Standards of Medical Care in Diabetes—2020. Diabetes Care 2020; 43(Suppl. 1): S135– S151 31862754
- [7]Resnikoff S, Felch W, Gauthier T-M, Spivey B. The number of ophthalmologists in practice and training worldwide: a growing gap despite more than 200,000 practitioners. Br J Ophthalmol 2012; 96: 783– 787 22452836
- [8]Kirkizlar E, Serban N, Sisson JA, Swann JL, Barnes CS, Williams MD. Evaluation of telemedicine for screening of diabetic retinopathy in the Veterans Health Administration. Ophthalmology 2013; 120: 2604– 2610 24084501
- [9]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– 792 31070699 PMC6512266
- [10]Tufail A, Kapetanakis VV, Salas-Vega S,. An observational study to assess if automated diabetic retinopathy image assessment software can replace one or more steps of manual imaging grading and to determine their cost-effectiveness. Health Technol Assess 2016; 20: 1– 72 PMC5204130 27981917
- [11]Tufail A, Rudisill C, Egan C,. Automated diabetic retinopathy image assessment software: diagnostic accuracy and cost-effectiveness compared with human graders. Ophthalmology 2017; 124: 343– 351 28024825
- [12]Ting DSW, Cheung CY-L, Lim G,. 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: 2211– 2223 29234807 PMC5820739
- [13]Abràmoff MD, Lavin PT, Birch M, Shah N, Folk JC. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digit Med 2018; 1: 39 31304320 PMC6550188
- [14]Keane PA, Topol EJ. With an eye to AI and autonomous diagnosis. NPJ Digit Med 2018; 1: 40 31304321 PMC6550235
- [16]Ogunyemi O, Kermah D. Machine learning approaches for detecting diabetic retinopathy from clinical and public health records. AMIA Annu Symp Proc 2015; 2015: 983– 990 26958235 PMC4765709
- [17]Kuzmak P, Demosthenes C, Maa A. Exporting diabetic retinopathy images from VA VISTA Imaging for research. J Digit Imaging 2019; 32: 832– 840 30511282 PMC6737199
- [18]Conlin PR, Fisch BM, Orcutt JC, Hetrick BJ, Darkins AW. Framework for a national teleretinal imaging program to screen for diabetic retinopathy in Veterans Health Administration patients. J Rehabil Res Dev 2006; 43: 741– 748 17310423
- [20]Xie Y, Nguyen QD, Hamzah H,. Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: an economic analysis modelling study. Lancet Digit Health 2020; 2: e240– e249 33328056
- [21]Wintergerst MWM, Brinkmann CK, Holz FG, Finger RP. Undilated versus dilated monoscopic smartphone-based fundus photography for optic nerve head evaluation. Sci Rep 2018; 8: 10228 29980724 PMC6035205
- [22]Silvar SD, Pollack RH. Racial differences in pigmentation of the fundus oculi. Psychon Sci 1967; 7: 159– 159
- [24]Gulshan V, Peng L, Coram M,. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 2016; 316: 2402– 2410 27898976
- [25]Gargeya R, Leng T. Automated identification of diabetic retinopathy using deep learning. Ophthalmology 2017; 124: 962– 969 28359545
- [26]Romero-Aroca P, Verges-Puig R, de la Torre J,. Validation of a deep learning algorithm for diabetic retinopathy. Telemed J E Health 2020; 26: 1001– 1009 31682189
- [28]Natarajan S, Jain A, Krishnan R, Rogye A, Sivaprasad S. Diagnostic accuracy of community-based diabetic retinopathy screening with an offline artificial intelligence system on a smartphone. JAMA Ophthalmol 2019; 137: 1182– 1188 31393538 PMC6692680
- [29]Xie Y, Gunasekeran DV, Balaskas K,. Health economic and safety considerations for artificial intelligence applications in diabetic retinopathy screening. Transl Vis Sci Technol 2020; 9: 22 PMC7396187 32818083
- [31]Liu X, Rivera SC, Moher D, Calvert MJ, Denniston AK; SPIRIT-AI and CONSORT-AI Working Group. Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI Extension. BMJ 2020; 370: m3164 32909959 PMC7490784
- [32]Ranganathan P, Aggarwal R. Common pitfalls in statistical analysis: understanding the properties of diagnostic tests - part 1. Perspect Clin Res 2018; 9: 40– 43 29430417 PMC5799952
Cited by
Sorted by citations
Ipp, E., Liljenquist, D., Bode, B., Shah, V. N., Silverstein, S., Regillo, C. D., Lim, J. I., Sadda, S., Domalpally, A., Gray, G., et al. (2021) · JAMA Network Open · doi:10.1001/jamanetworkopen.2021.34254 · Cited by 256 · see citing works
Artificial intelligence in ophthalmology: The path to the real-world clinic
Li, Z., Wang, L., Wu, X., Jiang, J., Qiang, W., Xie, H., Zhou, H., Wu, S., Shao, Y., & Chen, W. (2023) · Cell Reports Medicine · doi:10.1016/j.xcrm.2023.101095 · Cited by 207 · see citing works
Artificial Intelligence for Screening of Multiple Retinal and Optic Nerve Diseases
Dong, L., He, W., Zhang, R., Ge, Z., Wang, Y. X., Zhou, J., Xu, J., Shao, L., Wang, Q., Yan, Y., et al. (2022) · JAMA Network Open · doi:10.1001/jamanetworkopen.2022.9960 · Cited by 166 · see citing works
Rajesh, A. E., Davidson, O. Q., Lee, C. S., & Lee, A. (2023) · Diabetes Care · doi:10.2337/dci23-0032 · Cited by 166 · see citing works
Self-citationYim, D., Khuntia, J., Parameswaran, V., & Meyers, A. D. (2024) · JMIR Medical Informatics · doi:10.2196/52073 · Cited by 129 · see citing works
“Similarly, using GenAI in remote care helps improve glycemia and weight loss [95], yet challenges related to variable patient uptake and increased clinician participation necessitated by shared decision-making must be considered [96].”
González-Gonzalo, C., Thee, E. F., Klaver, C. C. W., Lee, A. Y., Schlingemann, R. O., Tufail, A., Verbraak, F., & Sánchez, C. I. (2022) · Progress in Retinal and Eye Research · doi:10.1016/j.preteyeres.2021.101034 · Cited by 103 · see citing works
Self-citationFoundational Considerations for Artificial Intelligence Using Ophthalmic Images
Abràmoff, M. D., Cunningham, B., Patel, B., Eydelman, M. B., Leng, T., Sakamoto, T., Blodi, B., Grenon, S. M., Wolf, R. M., Manrai, A. K., et al. (2022) · Ophthalmology · doi:10.1016/j.ophtha.2021.08.023 · Cited by 96 · see citing works
Scheetz, J., Koca, D., McGuinness, M., Holloway, E., Tan, Z., Zhu, Z., O’Day, R., Sandhu, S., MacIsaac, R. J., Gilfillan, C., et al. (2021) · Scientific Reports · doi:10.1038/s41598-021-94178-5 · Cited by 88 · see citing works
“…being marginally less accurate, to similar and high in accuracy when compared against validation using high quality retrospective datasets 12,14,16–18, It is promising that our system has exceeded sensitivity and specificity endpoints set by the United States Food and Drug Administration (FDA) (>…”
ResultCanayaz, M. (2022) · Applied Soft Computing · doi:10.1016/j.asoc.2022.109462 · Cited by 92 · see citing works
Artificial intelligence in retinal disease: clinical application, challenges, and future directions
Varela, M. D., Sen, S., de Guimarães, T. A. C., Kabiri, N., Pontikos, N., Balaskas, K., & Michaelides, M. (2023) · Graefe s Archive for Clinical and Experimental Ophthalmology · doi:10.1007/s00417-023-06052-x · Cited by 86 · see citing works
“Their real-world performance remains to be tested [127].”
BackgroundOng, J., Tavakkoli, A., Zaman, N., Kamran, S. A., Waisberg, E., Gautam, N., & Lee, A. G. (2022) · npj Microgravity · doi:10.1038/s41526-022-00222-7 · Cited by 75 · see citing works
Self-citationGrzybowski, A., & Brona, P. (2021) · Journal of Clinical Medicine · doi:10.3390/jcm10112352 · Cited by 64 · see citing works
Khokhar, P. B., Gravino, C., & Palomba, F. (2024) · Artif. Intell. Medicine · doi:10.1016/j.artmed.2025.103132 · Cited by 63 · see citing works
Zhang, J., Lin, S., Cheng, T., Xu, Y., Lu, L., He, J., Yu, T., Peng, Y., Zhang, Y., Zou, H., et al. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01109-5 · Cited by 51 · see citing works
“However, a signi fi cant reduction in diagnostic accuracy compared to the R&D stage was observed upon implementationinreal-worldcommunityeyediseasescreenings 9,10,17,18 .”
BackgroundGrzybowski, A. E., Singhanetr, P., Nanegrungsunk, O., & Ruamviboonsuk, P. (2023) · Ophthalmology and Therapy · doi:10.1007/s40123-023-00691-3 · Cited by 86 · 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
“Furthermore, the accuracy of the manual grading–based telemedicine screening was assumed to be 100%, which was in accordance with the DR diagnosis criteria [8,9,26,27].”
MethodKang, E. M., Ryu, I., Lee, G., Kim, J. K., Lee, I.-S., Jeon, G.-H., Song, H.-J., Kamiya, K., & Yoo, T. (2021) · Translational Vision Science & Technology · doi:10.1167/tvst.10.6.5 · Cited by 49 · see citing works
“31 To overcome this problem, we adopted the ensemble learning technique and validated the model using unseen datasets, including the prospectively designed dataset and completely independent external dataset.”
LLMEval-Med: A Real-world Clinical Benchmark for Medical LLMs with Physician ValidationPreprint
Zhang, M., Shen, Y.-J., Li, Z.-L., Sha, H., Hu, B., Wang, Y.-H., Huang, C.-H., Liu, S., Tong, J.-Q., Jiang, C.-H., et al. (2025) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2506.04078 · Cited by 48 · see citing works
“…treatments (Ross and Swetlitz, 2018); AI systems showed higher false-positive rates than radiologists in lung disease detection (Plesner et al., 2023); and diabetic retinopathy screening tools demonstrated inconsistent performance with sensitivities ranging from 51% to 86% (Lee et al., 2021).”
Artificial Intelligence for Predicting and Diagnosing Complications of Diabetes
Huang, J. K., Yeung, A. M., Armstrong, D. G., Battarbee, A. N., Cuadros, J. A., Espinoza, J., Kleinberg, S., Mathioudakis, N. N., Swerdlow, M. A., & Klonoff, D. C. (2022) · Journal of Diabetes Science and Technology · doi:10.1177/19322968221124583 · Cited by 59 · see citing works
“54 Researchers have also expressed reservations because of a lack of (1) detail in grading, 55 (2) grading of nondiabetic retinal conditions and glaucoma, and (3) an explanation of how the algorithm determines the retinal grade, which often incorporates multiple features from fundus imaging.”
BackgroundNovel artificial intelligence algorithms for diabetic retinopathy and diabetic macular edema
Yao, J., Lim, J., Lim, G. Y. S., Ong, J. C. L., Ke, Y., Tan, T. F., Tan, T., Vujosevic, S., & Ting, D. S. W. (2024) · Eye and Vision · doi:10.1186/s40662-024-00389-y · Cited by 44 · see citing works
“Real-world evaluations of several state-of-the-art AI DR screening systems have revealed inconsistent performance, with sensitivities ranging from 50.98% to 85.90%, and specificities from 60.42% to 83.69% [6].”
BackgroundInfluentialArtificial intelligence in retinal image analysis: Development, advances, and challenges
Oganov, A. C., Seddon, I. A., Jabbehdari, S., Uner, O. E., Fonoudi, H., Yazdanpanah, G., Outani, O., & Arévalo, J. F. (2023) · Survey of Ophthalmology · doi:10.1016/j.survophthal.2023.04.001 · Cited by 47 · see citing works
Jacoba, C. M., Doan, D., Salongccay, R. P., Aquino, L. C., Silva, J. P., Salva, C. M., Zhang, D., Alog, G. P., Zhang, K., Locaylocay, K. B., et al. (2023) · Ophthalmology Retina · doi:10.1016/j.oret.2023.03.003 · Cited by 41 · see citing works
Ong, A. Y., Taribagil, P., Sevgi, M., Kale, A. U., Dow, E. R., Macdonald, T., Kras, A., Maniatopoulos, G., Liu, X., Keane, P. A., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01726-8 · Cited by 39 · see citing works
“For example, in three researcher-led head-to-head comparison studies of multiple AIaMDs, several manufacturers either did not respond to enquiries or ultimately withdrew from participation, citing commercial or unspeci fi ed reasons 42 – 44 .”
Khan, A. Q., Sun, G.-M., Khalid, M., Imran, A., Bilal, A., Azam, M., & Sarwar, R. (2024) · PLoS ONE · doi:10.1371/journal.pone.0303094 · Cited by 36
Zhang, X., Yang, Y., Shen, Y., Zhang, K.-R., Jiang, Z., Ma, L., Ding, C., Wang, B.-Y., Meng, Y., & Liu, H. (2022) · European Radiology · doi:10.1007/s00330-022-08956-4 · Cited by 36 · see citing works
“Further, algorithm accuracy further depends on a variety of factors such as the type of study, i.e., multicenter or single-center study [21 – 23], the use of transfer learning [24 – 27] or data augmentation [28 – 30], whether the training dataset is well-balanced [31 – 33], adoption of DL or cML…”
Yin, S. (2025) · Biosensors · doi:10.3390/bios15100656 · Cited by 54 · see citing works
“Multicenter validation evaluates method stability, reproducibility, and generalizability across diverse institutions, populations, and environments, is among the first critical steps in translating research findings into real-world applications [128].”
Grzybowski, A., Rao, D. P., Brona, P., Negiloni, K., Krzywicki, T., & Savoy, F. M. (2023) · Ophthalmic Research · doi:10.1159/000534098 · Cited by 33 · see citing works
Owens, D., Gurudas, S., Sivaprasad, S., Zaidi, F., Tapp, R., Kazantzis, D., Evans, L., & Thomas, R. L. (2025) · Diabetes Research and Clinical Practice · doi:10.1016/j.diabres.2025.112346 · Cited by 33 · see citing works
Gonçalves, M. B., Nakayama, L., Ferraz, D., Faber, H., Korot, E., Malerbi, F., Regatieri, C., Maia, M., Celi, L., Keane, P., et al. (2023) · Eye · doi:10.1038/s41433-023-02717-3 · Cited by 32 · see citing works
“In ophthalmology models, using exclusively high-quality images can produce higher pre-deployment accuracy results; however, these algorithms fail when they are deployed in real-world settings [17, 18].”
Hayati, A., Homayuni, M. R. A., Sadeghi, R., Asadigandomani, H., Dashtkoohi, M., Eslami, S., & Soleimani, M. (2025) · Diagnostics · doi:10.3390/diagnostics15060737 · Cited by 31 · see citing works
“Additionally, ensuring the generalizability of AI models across diverse populations and clinical settings requires extensive validation through multicenter studies and large-scale datasets [14,15].”
Artificial Intelligence-Based Digital Biomarkers for Type 2 Diabetes: A Review.
Jabara, M., Kose, O., Perlman, G., Corcos, S., Pelletier, M., Possik, E., Tsoukas, M., & Sharma, A. (2024) · Canadian Journal of Cardiology · doi:10.1016/j.cjca.2024.07.028 · Cited by 31 · see citing works
“38 Although high negative predictive values (82.72%-93.69”
InfluentialVeritti, D., Rubinato, L., Sarao, V., Nardin, A. D., Foresti, G., & Lanzetta, P. (2023) · Graefe's Archive for Clinical and Experimental Ophthalmology · doi:10.1007/s00417-023-06245-4 · Cited by 30 · see citing works
“For instance, the performance of automated DR algorithms varies considerably due to limited training data, heterogeneity in disease presentations, and suboptimal image quality [9, 10].”
BackgroundUsing artificial intelligence for diabetic retinopathy screening: Policy implications
Raman, R., DasGupta, D., Ramasamy, K., George, R., Mohan, V., & Ting, D. (2021) · Indian Journal of Ophthalmology · doi:10.4103/ijo.ijo_1420_21 · Cited by 29 · see citing works
DF-DM: A foundational process model for multimodal data fusion in the artificial intelligence eraPreprint
Restrepo, D., Wu, C.-W., Vásquez-Venegas, C., Nakayama, L., Celi, L., & Lopez, D. M. (2024) · Research Square · doi:10.21203/rs.3.rs-4277992/v1 · Cited by 25 · see citing works
“Data bias can arise from imbalanced datasets or the underrepresentation of certain groups, leading to biased predictions [34, 65–67].”
ResultS.Rajarajeshwari, G. K. P., & Selvi, G. C. (2024) · IEEE Access · doi:10.1109/access.2024.3494840 · Cited by 57 · see citing works
“Lee et al. [119] conducted a non-interventional device validation study at multiple centers.”
Duggal, M., Chauhan, A., Gupta, V., Kankaria, A., Budhija, D., Verma, P., Miglani, V., Syal, P., Kaur, G., Kumar, L., et al. (2025) · JMIR Medical Informatics · doi:10.2196/67529 · Cited by 24 · see citing works
“%) in real-world performance [36].”
Rajalakshmi, R., Pramodkumar, T., Naziyagulnaaz, A. S., Anjana, R., Raman, R., Manikandan, S., & Mohan, V. (2024) · Seminars in Ophthalmology · doi:10.1080/08820538.2024.2432902 · Cited by 24 · see citing works
“39 The study showed significant variability in sensitivity, specificity and accuracy of AI devices across different data sets and secondly most of the algorithms did not outperform human graders.”
Doğan, M. E., Bilgin, A., Sarı, R., Bulut, M., Akar, Y., & Aydemir, M. (2024) · Eye · doi:10.1038/s41433-024-03000-9 · Cited by 24 · see citing works
Cui, T.-X., Lin, D.-R., Yu, S.-S., Zhao, X., Lin, Z.-Z., Zhao, L.-Q., Xu, F., Yun, D., Pang, J.-Y., Li, R.-Y., et al. (2023) · JAMA ophthalmology · doi:10.1001/jamaophthalmol.2023.4650 · Cited by 24 · see citing works
Global disparity bias in ophthalmology artificial intelligence applications
Nakayama, L. F., Kras, A., Ribeiro, L. Z., Malerbi, F. K., Mendonça, L. S., Celi, L. A., Regatieri, C. V. S., & Waheed, N. K. (2022) · BMJ Health & Care Informatics · doi:10.1136/bmjhci-2021-100470 · Cited by 23 · see citing works
Rajesh, A. E., Olvera-Barrios, A., Warwick, A., Wu, Y., Stuart, K., Biradar, M. I., Ung, C., Khawaja, A., Luben, R., Foster, P., et al. (2025) · Nature Communications · doi:10.1038/s41467-024-55198-7 · Cited by 23 · see citing works
Self-citationMayya, V., Kandala, R. N. V. P. S., Gurupur, V. P., King, C., Vu, G. T., & Wan, T. T. H. (2024) · Health Services Research and Managerial Epidemiology · doi:10.1177/23333928241275292 · Cited by 23 · see citing works
Artificial Intelligence in Ophthalmology – Status Quo and Future Perspectives
Matos, P. A. W., Reimer, R. P., Rokohl, A. C., Caldeira, L., Heindl, L. M., & Hokamp, N. G. (2022) · Seminars in Ophthalmology · doi:10.1080/08820538.2022.2139625 · Cited by 24 · see citing works
Big Data and Artificial Intelligence in Ophthalmology: Where Are We Now?
Lee, C. S., Brandt, J., & Lee, A. (2021) · Ophthalmology Science · doi:10.1016/j.xops.2021.100036 · Cited by 20 · see citing works
“Third, validation of both big data and AI study results in independent study cohorts are key.(17) As this field of research continues to advance, other critical issues remain.”
MethodSelf-citationWang, Y., Shi, D., Tan, Z., Niu, Y. Y., Jiang, Y., Xiong, R., Peng, G., & He, M. (2021) · Frontiers in Medicine · doi:10.3389/fmed.2021.740987 · Cited by 17 · see citing works
Review on diabetic retinopathy with deep learning methods
Shekar, S. C., Satpute, N., & Gupta, A. (2021) · Journal of Medical Imaging · doi:10.1117/1.jmi.8.6.060901 · Cited by 19 · see citing works
“Lee et al. 122 introduced the one-on-one, multicenter study of comparison of recent AI DR screening where 5 companies presented 7 algorithms among which one was FDA approved and a total of 23,724 patients were confronted.”
BackgroundMoannaei, M., Jadidian, F., Doustmohammadi, T., Kiapasha, A. M., Bayani, R., Rahmani, M., Jahanbazy, M. R., Sohrabivafa, F., Asadi Anar, M., Magsudy, A., et al. (2025) · BioMedical Engineering OnLine · doi:10.1186/s12938-025-01336-1 · Cited by 18 · see citing works
The biomedical applications of artificial intelligence: an overview of decades of research
Naskar, S., Sharma, S., Kuotsu, K., Halder, S., Pal, G., Saha, S., Mondal, S., Biswas, U. K., Jana, M., & Bhattacharjee, S. (2025) · Journal of drug targeting (Print) · doi:10.1080/1061186x.2024.2448711 · Cited by 18 · see citing works
“As Ai has [205] resnet and Mask r-Cnn Dr grading [206] opthai, airDoc, eyenuk, retina-ai Health, and retmarker identifying referable Dr [207] gaussian-sampling method built upon a multiple-instance learning framework identifying referable Dr [208] Ml-enabled software, eyeart v2.”
Real-world validation of artificial intelligence algorithms for ophthalmic imaging
Domalpally, A., & Channa, R. (2021) · The Lancet Digital Health · doi:10.1016/s2589-7500(21)00140-0 · Cited by 18 · see citing works
Penha, F. M., Priotto, B. M., Hennig, F., Przysiezny, B., Wiethorn, B. A., Orsi, J., Nagel, I. B. F., Wiggers, B., Stuchi, J. A., Lencione, D., et al. (2023) · International Journal of Retina and Vitreous · doi:10.1186/s40942-023-00477-6 · Cited by 18 · see citing works
“…heterogeneous distribution of PPVs among different datasets has been attributed to differences in disease prevalence, on the basis of Bayes theorem [11]: sites with higher prevalence rates had higher PPV; of note, our sample had the majority of patients with no signs of disease (68%), which may…”
BackgroundResultInfluentialMacdonald, T., Zhelev, Z., Liu, X., Hyde, C., Fajtl, J., Egan, C. A., Tufail, A., Rudnicka, A. R., Shinkins, B., Given-Wilson, R., et al. (2025) · The Lancet Digital Health · doi:10.1016/j.landig.2024.12.004 · Cited by 15 · see citing works
Alé-Chilet, A., Bernal-Morales, C., Barraso, M., Hernandez, T., Oliva, C., Vinagre, I., Ortega, E., Figueras-Roca, M., Sala-Puigdollers, A., Esquinas, C., et al. (2021) · Journal of Clinical Medicine · doi:10.3390/jcm11010197 · Cited by 16 · see citing works
Wang, Y., Han, X., Li, C., Luo, L., Yin, Q., Zhang, J., Peng, G., Shi, D., & He, M. (2024) · Journal of Medical Internet Research · doi:10.2196/52506 · Cited by 11 · see citing works
“For example, although most DL algorithms have achieved a sensitivity of over 90% in previous research, the reported sensitivity ranges from 55% to 85% in real-world tasks [4,7,10].”
InfluentialWroblewski, J. J., Sánchez-Buenfil, E., Inciarte, M., Berdia, J., Blake, L. R., Wroblewski, S., Patti, A., Suter, G., & Sanborn, G. E. (2023) · Journal of Diabetes Science and Technology · doi:10.1177/19322968231194644 · Cited by 18 · see citing works
“26 To our knowledge, our analysis is the first to use the Medios and EyeArt AI algorithms to analyze images from patients of Spanish-Mexican and Mayan descent in the Yucatan Peninsula, each having a distinct genetic phenotype.”
Chou, Y.-B., Kale, A. U., Lanzetta, P., Aslam, T. M., Barratt, J., Danese, C., Eldem, B. M., Eter, N., Gale, R. P., Korobelnik, J., et al. (2023) · Current Opinion in Ophthalmology · doi:10.1097/icu.0000000000000979 · Cited by 20 · see citing works
Chen, J. S., Marra, K., Robles-Holmes, H. K., Ly, K. B., Miller, J., Wei, G., Aguilar, E., Bucher, F., Ideguchi, Y., Coyner, A. S., et al. (2023) · Ophthalmology Science · doi:10.1016/j.xops.2023.100338 · Cited by 15 · see citing works
“36 The generalization performance of our model confers additional bene fi ts to OIR research, including the ability to potentially compare vascularity between images from different laboratories, which may promote data sharing among researchers.”
BackgroundArtificial Intelligence Algorithms in Diabetic Retinopathy Screening
Zafar, S., Mahjoub, H., Mehta, N., Domalpally, A., & Channa, R. (2022) · Current Diabetes Reports · doi:10.1007/s11892-022-01467-y · Cited by 16 · see citing works
Yuen, J., Pike, S., Khachikyan, S., & Nallasamy, S. (2022) · Exon Publications eBooks · doi:10.36255/exon-publications-digital-health-telehealth-ophthalmology · Cited by 15 · see citing works
Nakayama, L. F., de Matos, J. C. R. G., Quion, J., Novaes, F. D. C., Mitchell, W. G., Mwavu, R., Hung, C. J.-Y. J., Santiago, A. P. D., Phanphruk, W., Cardoso, J. S., et al. (2024) · PLOS Digital Health · doi:10.1371/journal.pdig.0000618 · Cited by 19 · see citing works
“Furthermore, commercial diabetic retinopathy screening algorithms have exhibited notable discrepancies in performance when applied across different populations within the same country, which can lead to delayed diagnosis and treatment in groups not represented in the training data set [33].”
BackgroundTrustworthy Evaluation of Clinical AI for Analysis of Medical Images in Diverse Populations.
Fajtl, J., Welikala, R., Barman, S., Chambers, R., Bolter, L., Anderson, J., Olvera-Barrios, A., Shakespeare, R., Egan, C., Owen, C. G., et al. (2024) · NEJM AI · doi:10.1056/aioa2400353 · Cited by 14 · see citing works
Li, Z., Yin, S., Wang, S., Wang, Y., Qiang, W., & Jiang, J. (2024) · Journal of Advanced Research · doi:10.1016/j.jare.2024.11.018 · Cited by 14 · see citing works
“Lee and colleagues (2021) [61] conducted a real-world validation study comparing seven automated AI-based DR screening systems.”
Variability of Grading DR Screening Images among Non-Trained Retina Specialists
Grzybowski, A., Brona, P., Krzywicki, T., gaca-Wysocka, M., Berlińska, A., & Święch, A. (2022) · Journal of Clinical Medicine · doi:10.3390/jcm11113125 · Cited by 13 · see citing works
Yuan, A., & Lee, A. (2022) · The Lancet Digital Health · doi:10.1016/s2589-7500(22)00027-9 · Cited by 17 · see citing works
Self-citationApplication of Deep Learning Methods in a Moroccan Ophthalmic Center: Analysis and Discussion
Farahat, Z., Zrira, N., Souissi, N., Benamar, S., Belmekki, M., Ngote, M. N., & Megdiche, K. (2023) · Diagnostics · doi:10.3390/diagnostics13101694 · Cited by 12 · see citing works
Li, Y.-T., Zhang, R., Dong, L., Shi, X.-H., Zhou, W.-D., Wu, H.-T., Li, H.-Y., Yu, C.-Y., & Wei, W.-G. (2024) · Eye · doi:10.1038/s41433-023-02914-0 · Cited by 12 · see citing works
“Secondly, the PROBAST contains several items of subjective judgment, and borderline results may impact overall interpretation.”
InfluentialTowards population-independent, multi-disease detection in fundus photographs
Matta, S., Lamard, M., Conze, P.-H., Guilcher, A. L., Lecat, C., Carette, R., Basset, F., Massin, P., Rottier, J.-B., Cochener, B., et al. (2023) · Scientific Reports · doi:10.1038/s41598-023-38610-y · Cited by 12 · see citing works
“ATrain 5935 (2856) 29 (13) 325 (158) 0 (0) 18 (10) 0 (0) 14 (7) 531 (220)”
Lin, S.-L., Li, L., Zou, H.-D., Xu, Y., & Lu, L.-N. (2022) · Journal of Medical Internet Research · doi:10.2196/40249 · Cited by 12 · see citing works
“Based on previously published literature [4-7, 16-18], 4 attributes were identified initially to describe the outline of the community-based eye disease screening, including the accuracy, screening result In Experiment 1, one municipal and 16 district-level eye disease control centers and over 250…”
InfluentialData Homogeneity Effect in Deep Learning-Based Prediction of Type 1 Diabetic Retinopathy
Lo, J.-E., Kang, E., Chen, Y.-N., Hsieh, Y.-T., Wang, N.-K., Chen, T.-C., Chen, K.-J., Wu, W.-C., Hwang, Y., Lo, F.-S., et al. (2021) · Journal of Diabetes Research · doi:10.1155/2021/2751695 · Cited by 12 · see citing works
“However, a large DR screening validation study found that most algorithms had significant performance differences and even obtained concerning results when evaluated through external validation, even though these algorithms were already in active use in realworld clinical settings [34].”
ResultRahmati, M., Smith, L., Piyasena, M. P., Bowen, M., Boyer, L., Fond, G., Kazemi, A., Yon, D. K., Lee, H., Sehmbi, T., et al. (2025) · Eye · doi:10.1038/s41433-025-03849-4 · Cited by 11 · see citing works
Qamar, W. U. R., Lee, M.-H., & Abibullaev, B. (2025) · Frontiers in Neuroscience · doi:10.3389/fnins.2025.1677898 · Cited by 11 · see citing works
Nakayama, L., Ribeiro, L. Z., Malerbi, F., & Regatieri, C. (2025) · International Journal of Retina and Vitreous · doi:10.1186/s40942-024-00596-8 · Cited by 11 · see citing works
“As an illustrative example, pneumonia screening and diabetic retinopathy algorithms exhibit high algorithmic accuracy but perform poorly when applied to data from external institutions, even when those institutions are within the same country [13, 14].”
Grzybowski, A. E., Peeters, F., Barão, R., Brona, P., Rommes, S., Krzywicki, T., Stalmans, I., & Jacob, J. (2024) · Acta ophthalmologica · doi:10.1111/aos.17428 · Cited by 11 · see citing works
“…sensitivity rates (comparable or superior to human graders), diagnostic performance has varied substantially and specificity is often lower than human graders, which increases referrals for human contact (Grzybowski et al., 2020; Gunasekeran et al., 2020; Lee et al., 2021; Rajesh et al., 2023).”
InfluentialA machine-learning-based algorithm for bone marrow cell differential counting
Yu, T., Yang, C.-K., Hsu, W.-H., Hsu, C.-A., Wang, H.-C., Hsiao, H.-J., Chao, H.-L., Hsieh, H.-P., Wu, J.-R., Tsai, Y., et al. (2024) · Int. J. Medical Informatics · doi:10.1016/j.ijmedinf.2024.105692 · Cited by 11
Zhu, A., Tailor, P., Verma, R., Zhang, I., Schott, B., Ye, C., Szirth, B., Habiel, M. M., & Khouri, A. S. (2023) · Journal of Telemedicine and Telecare · doi:10.1177/1357633x231158832 · Cited by 14 · see citing works
Peeters, F., Rommes, S., Elen, B., Gerrits, N., Stalmans, I., Jacob, J., & Boever, P. D. (2023) · Journal of Clinical Medicine · doi:10.3390/jcm12041408 · Cited by 11 · see citing works
Vaghefi, E., Yang, S., Xie, L. Y., Han, D., Yap, A., Schmeidel, O., Marshall, J., & Squirrell, D. (2022) · Eye · doi:10.1038/s41433-022-02217-w · Cited by 14 · see citing works
“While the accuracy of AI-based models for detecting DR have been demonstrated in many previous studies [5–9], most have failed to perform in the “real world” setting [16].”
MethodLee, A., Lee, C. S., Hunt, M. S., Yanagihara, R., Blazes, M., & Boyko, E. (2020) · Diabetes Care · doi:10.2337/dci21-0007 · Cited by 11
“Soliz for his expedient comments (1) regarding our work evaluating seven artificial intelligence (AI) algorithms in a head-to-head fashion (2).”
BackgroundSelf-citationGrzybowski, A. E., & Jin, K. (2026) · Ophthalmology and Therapy · doi:10.1007/s40123-026-01322-3 · Cited by 10 · see citing works
“…Affairs multicenter study involving over 23,000 patients reported sensitivities ranging from 50.9 to 85.9% across seven commercial algorithms [30], while the Finnish Study found mean sensitivity of 77.5%, specificity of 80.6%, and an ungradable-image rate of 1.9% among 21 CE-certified AI…”
InfluentialPinto, I., Olazarán, Á., Jurío, D., de la Osa, B., Sainz, M., Oscoz, A., Ballaz, J., Gorricho, J., Galar, M., & Andonegui, J. (2025) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1547045 · Cited by 14 · see citing works
Surya, J., Garima, Pandy, N., Rim, T. H., Lee, G., Priya, M. S., Subramanian, B., & Raman, R. (2023) · Indian Journal of Ophthalmology · doi:10.4103/ijo.ijo_11_23 · Cited by 10 · see citing works
Chawla, H., Uhr, J. H., Williams, J. S., Reinoso, M., & Weiss, J. (2023) · Ophthalmic Surgery Lasers and Imaging Retina · doi:10.3928/23258160-20230406-01 · Cited by 10 · see citing works
Eyes as the windows into cardiovascular disease in the era of big data
Chan, Y. K., Cheng, C.-Y., & Sabanayagam, C. (2023) · Taiwan Journal of Ophthalmology · doi:10.4103/tjo.tjo-d-23-00018 · Cited by 10 · see citing works
Application and prospect of artificial intellingence in diabetes care
Jia, W.-P., & Fisher, E. (2023) · Medical Review · doi:10.1515/mr-2022-0039 · Cited by 10 · see citing works
The march to harmonized imaging standards for retinal imaging
Gim, N., Ferguson, A. N., Blazes, M. S., Lee, C. S., & Lee, A. (2025) · Progress in Retinal and Eye Research · doi:10.1016/j.preteyeres.2025.101363 · Cited by 8 · see citing works
“However, despite these calls for action, adoption remains low, leaving a significant gap in the field (Lee et al., 2021a).”
InfluentialSelf-citationRajesh, A. E., Olvera-Barrios, A., Warwick, A., Wu, Y., Stuart, K., Biradar, M. I., Ung, C. Y., Khawaja, A., Luben, R., Foster, P., et al. (2023) · medRxiv · doi:10.1101/2023.06.28.23291873 · Cited by 9 · see citing works
Self-citationNissen, T. P. H., Nørgaard, T. L., Schielke, K., Vestergaard, P., Nikontovic, A., Dawidowicz, M., Grauslund, J., Vorum, H., & Aasbjerg, K. (2023) · Journal of Personalized Medicine · doi:10.3390/jpm13071128 · Cited by 9 · see citing works
The Role of Artificial Intelligence in the Diagnosis and Management of Diabetic Retinopathy
Ansari, A., Ansari, N. M., Khalid, U., Markov, D., Bechev, K. I., Aleksiev, V., Markov, G., & Poryazova, E. G. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14145150 · Cited by 8 · see citing works
“Although negative predictive values were high (82.72–93.69%), especially in the Atlanta subset, positive predictive values were generally modest, emphasizing the need for extensive external validation and performance benchmarking before clinical deployment of AI systems [16].”
InfluentialBondok, M., Selvakumar, R., Asdo, A., Naderi, B., Zhang, C.-H., Wong, C. N., & Felfeli, T. (2025) · Ophthalmology Retina · doi:10.1016/j.oret.2025.05.032 · Cited by 8 · see citing works
Deep learning detection of diabetic retinopathy in Scotland’s diabetic eye screening programme
Fleming, A. D., Mellor, J., McGurnaghan, S., Blackbourn, L., Goatman, K., Styles, C., Storkey, A., McKeigue, P., & Colhoun, H. (2023) · British Journal of Ophthalmology · doi:10.1136/bjo-2023-323395 · Cited by 8 · see citing works
“For example, systems have been trained, tested or both using images from screening programmes in Thailand [11], India [12], Singapore [13], Portugal [6], China [14], [15], several states of the USA [16], [7], Spain (Andalusia) [17] and England [18].”
BackgroundMethodNakayama, L., Ribeiro, L. Z., Malerbi, F., & Regatieri, C. (2022) · Frontiers in Ophthalmology · doi:10.3389/fopht.2022.898181 · Cited by 8 · see citing works
“Algorithms have been trained and tested with a few ethnic populations, mainly North American, European, and Asian patients (42).”
BackgroundAbellanas, M., Elena, M., Keane, P., Balaskas, K., Grewal, D., & Carreño, E. (2022) · Ocular immunology and inflammation · doi:10.1080/09273948.2022.2054433 · Cited by 8 · see citing works
Application Research of Artificial Intelligence Screening System for Diabetic Retinopathy
Wang, Y., Wang, L., Zhou, H., Liao, Y., & Yi, Q. (2022) · Journal of Healthcare Engineering · doi:10.1155/2022/2185547 · Cited by 8
“+rough regular screening of fundus lesions and early diagnosis, nearly 90% of visual impairment and blindness induced by diabetes can be completely avoided [6].”
BackgroundRetractedEvaluation of a New Neural Network Classifier for Diabetic Retinopathy
Katz, O., Presil, D., Cohen, L., Nachmani, R., Kirshner, N., Hoch, Y., Lev, T. H., Hadad, A., Hewitt, R. J., & Owens, D. (2021) · Journal of Diabetes Science and Technology · doi:10.1177/19322968211042665 · Cited by 8 · see citing works
“27,28 29-31 Recently, a small number have received approval for clinical use by the U.S. Food and Drug Administration 32 whilst others have already been incorporated into screening services elsewhere in the world as a filter (disease/no disease) or as a substitute for human graders.”
BackgroundKhan, A. Y. R., & Malik, M. B. (2025) · Cureus · doi:10.7759/cureus.96121 · Cited by 10 · see citing works
“A large-scale, multicenter, non-interventional validation study on seven different AI algorithms for the diagnosis of DR showed that only two could perform better than human examiners, thus raising serious concerns about the accuracy of such algorithms [129].”
Discriminative, generative artificial intelligence, and foundation models in retina imaging
Ruamviboonsuk, P., Arjkongharn, N., Vongsa, N., Pakaymaskul, P., & Kaothanthong, N. (2024) · Taiwan Journal of Ophthalmology · doi:10.4103/tjo.tjo-d-24-00064 · Cited by 9 · see citing works
Tadayoni, R., Massin, P., Bonnin, S., Magazzeni, S., Lay, B., Guilcher, A. L., Vicaut, E., Couturier, A., Quellec, G., & E.-N.T. Investigators (2024) · BMJ Open · doi:10.1136/bmjopen-2024-084574 · Cited by 7 · see citing works
Blair, J., Rodríguez, J., Vitar, R. L. L., Stadelmann, M., Abreu-González, R., Donate, J., Ciller, C., Apostolopoulos, S., Bermúdez, C., & Zanet, S. D. (2023) · Translational Vision Science & Technology · doi:10.1167/tvst.12.11.38 · Cited by 7 · see citing works
“13 It has been suggested that integration into a cloud-based system may facilitate an improved review process, resulting in higher real-world accuracy.”
BackgroundRan, A. R., Ding, J. L., Tang, Z., Lam, C., Nguyen, T. X., Zhou, J., Zhang, S., Fang, D., Yang, D., Ng, V., et al. (2026) · Diabetes Care · doi:10.2337/dc25-1493 · Cited by 7 · see citing works
Lin, S., Ma, Y., Li, L., Jiang, Y., Peng, Y., Yu, T., Qian, D., Xu, Y., Lu, L., Chen, Y., et al. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.109329 · Cited by 6 · see citing works
Artificial Intelligence-Based Screening System for Diabetic Retinopathy in Primary Care
Baget-Bernaldiz, M., Fontoba-Poveda, B., Romero-Aroca, P., Navarro-Gil, R., Hernando-Comerma, A., Bautista-Pérez, Á., Llagostera-Serra, M., Morente-Lorenzo, C., Vizcarro, M., & Mira-Puerto, A. (2024) · Diagnostics · doi:10.3390/diagnostics14171992 · Cited by 6 · see citing works
“Finally, a retrospective validation study compared seven different AIRSs for detecting RDR that had been previously validated [23].”
Method
- Authors
- Lee, A. Y., Yanagihara, R. T., Lee, C. S., Blazes, M., Jung, H. C., Chee, Y. E., Gencarella, M. D., Gee, H., Maa, A. Y., Cockerham, G. C., Lynch, M., & Boyko, E. J.
- Affiliations
- University of Washington (US)
- Office of Information and Technology, Clinical Imaging, Seattle, WA
- Emory University (US)
- Veterans Health Administration (US)
- United States Department of Veterans Affairs (US)
- Corresponding author
- Lee, Aaron Y.
- ORCID
- Venue
- Diabetes Care
- Publisher
- American Diabetes Association
- Published
- 2021-01-05
- Volume
- 44
- Issue
- 5
- Pages
- 1168-1175
- ISSN
- 0149-5992 (ISSN-L)
- 1935-5548
- PMC
- PMC8132324
- OpenAlex
- W3119783096
- Cited by
- 194
- References
- 32
- Citations by year
- 2026: 41 · 2025: 40 · 2024: 28 · 2023: 40 · 2022: 22 · 2021: 21 · 2020: 2
- FWCI
- 14.66
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Diabetes Care (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Eye Institute (K23-EY-029246, R01-AG-060942)
- Research to Prevent Blindness
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Retinal Imaging and Analysis
- Health Sciences › Medicine › Ophthalmology › Retinal Diseases and Treatments
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- diabetic retinopathy, automated screening, artificial intelligence, retinal image analysis, algorithm comparison, referable diabetic retinopathy, proliferative diabetic retinopathy, sensitivity and specificity, primary care screening, veterans
- MeSH terms
- Algorithms, Artificial Intelligence, Diabetes Mellitus, Diabetic Retinopathy, Humans, Mass Screening, Sensitivity and Specificity
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
