The DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Optimized glycemic control of type 2 diabetes with reinforcement learning: a proof-of-concept trial
Open accessAbstract
The personalized titration and optimization of insulin regimens for treatment of type 2 diabetes (T2D) are resource-demanding healthcare tasks. Here we propose a model-based reinforcement learning (RL) framework (called RL-DITR), which learns the optimal insulin regimen by analyzing glycemic state rewards through patient model interactions. When evaluated during the development phase for managing hospitalized patients with T2D, RL-DITR achieved superior insulin titration optimization (mean absolute error (MAE) of 1.10 ± 0.03 U) compared to other deep learning models and standard clinical methods. We performed a stepwise clinical validation of the artificial intelligence system from simulation to deployment, demonstrating better performance in glycemic control in inpatients compared to junior and intermediate-level physicians through quantitative (MAE of 1.18 ± 0.09 U) and qualitative metrics from a blinded review. Additionally, we conducted a single-arm, patient-blinded, proof-of-concept feasibility trial in 16 patients with T2D. The primary outcome was difference in mean daily capillary blood glucose during the trial, which decreased from 11.1 (±3.6) to 8.6 (±2.4) mmol L −1 ( P < 0.01), meeting the pre-specified endpoint. No episodes of severe hypoglycemia or hyperglycemia with ketosis occurred. These preliminary results warrant further investigation in larger, more diverse clinical studies. ClinicalTrials.gov registration: NCT05409391 .
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures






![Extended Data Fig. 1 The development of the AI system. a, The sequential decision process with patient model and policy model in AI system. Given a trajectory, for the initial step, the representation function fR receives as input the past observations O1:t from the trajectory. The model is subsequently unrolled recurrently for K steps. At each step κ∈[1,K], the policy model π receives the hidden state st+k−1 and generates an action at+k−1. The dynamics function fT of patient model subsequently receives as input the hidden state st+k−1 from the previous step and the action at+k−1 and produces the hidden state of the next step st+k, and the prediction function of patient model predicts diabetes status yt+k−1. The hidden states and actions recurrently update. b, The AI system training pipeline. Left, the patient model learning for patient tracking. Given a hidden state st and an actual action at, the patient generates the predicted status yˆt to estimate the current status yt, and produces the next state st+1. The estimated reward rˆt compared to the actual reward rt was calculated from yˆt. The patient model is joint optimized by the objective of consistency loss of state transition LT and supervised loss of status prediction Lp. Right, the policy update for dynamic regimen with combined supervised learning and reinforcement learning. The policy model π receives the hidden state st and then generates an action aˆt. Subsequently, the patient model receives as input the hidden statest from the previous step and the action at, then produces the hidden state of the next step st+1 and returns a reward rˆt. The policy model π is joint optimized by the objective of combined reinforcement learning LRL1, LRL2 and supervised learning LSL.](https://figures.cochat.ai/paper-figures/1606a2f37b324ceebdb03463fa31422f/optimized-glycemic-control-of-type-2-diabetes-with-reinforcement-learning-a-figure-7.jpg)






Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“AI-driven real-time patient monitoring systems are revolutionizing clinical care by enabling continuous, data-driven tracking of patient health.90,104,105”
From A foundational architecture for AI agents in healthcare (2025) · cites this paper as [105]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362
Sun, H., Saeedi, P., Karuranga, S., Pinkepank, M., Ogurtsova, K., Duncan, B. B., Stein, C., Basit, A., Chan, J. C. N., Mbanya, J. C., et al. (2022) · Diabetes Research and Clinical Practice · doi:10.1016/j.diabres.2021.109119 · Cited by 10,123
Stratton, I. M. (2000) · BMJ · doi:10.1136/bmj.321.7258.405 · Cited by 9,059
Reinforcement Learning: A Survey
Kaelbling, L. P., Littman, M. L., & Moore, A. W. (1996) · Journal of Artificial Intelligence Research · doi:10.1613/jair.301 · Cited by 8,972 · cited 1×
“Reinforcement learning (RL) has been proposed as a subfield of machine learning, enabling an agent to learn effective strategies through trial-and-error interactions with a dynamic environment 13 .”
BackgroundReinforcement Learning: An Introduction
Johnson, J. D., Li, J., & Chen, Z. (2000) · Neurocomputing · doi:10.1016/s0925-2312(00)00324-6 · Cited by 8,414
Self-citation10-year follow-up of intensive glucose control in type 2 diabetes
Holman, R. R., Paul, S. K., Bethel, M. A., Matthews, D. R., & Neil, H. A. W. (2008) · New England Journal of Medicine · doi:10.1056/nejmoa0806470 · Cited by 7,751
Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning
Kermany, D. S., Goldbaum, M., Cai, W., Valentim, C. C. S., Liang, H., Baxter, S. L., McKeown, A., Yang, G., Wu, X., Yan, F., et al. (2018) · Cell · doi:10.1016/j.cell.2018.02.010 · Cited by 4,842
Self-citationArtificial intelligence in cancer imaging: Clinical challenges and applications
Bi, W. L., Hosny, A., Schabath, M. B., Giger, M. L., Birkbak, N. J., Mehrtash, A., Allison, T., Arnaout, O., Abbosh, C., Dunn, I. F., et al. (2019) · CA: A Cancer Journal for Clinicians · doi:10.3322/caac.21552 · Cited by 1,846 · cited 2×
“Artificial intelligence (AI) approaches have emerged as potentially powerful tools to aid in disease diagnosis and management 8–10 .”
BackgroundMoghissi, E. S., Korytkowski, M. T., DiNardo, M., Einhorn, D., Hellman, R., Hirsch, I. B., Inzucchi, S. E., Ismail-Beigi, F., Kirkman, M. S., & Umpierrez, G. E. (2009) · Diabetes Care · doi:10.2337/dc09-9029 · Cited by 1,444 · cited 1×
“The traditional clinical methods of insulin dosage titration were used as the standard clinical methods for comparison, consisting of guidelines 42 and consensus formulas 43,44 for premixed insulin regimen, basal regimen and basal-bolus regimen. https://doi.org/10.1038/s41591-023-02552-9 The…”
MethodInfluentialMastering Atari, Go, chess and shogi by planning with a learned model
Schrittwieser, J., Antonoglou, I., Hubert, T., Simonyan, K., Sifre, L., Schmitt, S., Guez, A., Lockhart, E., Hassabis, D., Graepel, T., et al. (2020) · RePEc: Research Papers in Economics · Cited by 1,310
Research works printed but not linked: 5
- Thomas, M. M., Joost, B., Aske, P. & Catholijn, M. J. Model-based reinforcement learning: a survey. Found. Trends Mach. Learn 16, 1–118 (2020).
- Farajtabar, M., Chow, Y. & Ghavamzadeh, M. More robust doubly robust off-policy evaluation. In Proc. of the 35th International Conference on Machine Learning, Vol. 80, 1447–1456 (PMLR, 2018).
- Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014).
- Fox, I. & Wiens, J. Reinforcement learning for blood glucose control: challenges and opportunities. https://openreview.net/forum?id=ByexVzSAs4 (2019).
- Zhang, H., Xu, W. & Yu, H. Generative planning for temporally coordinated exploration in reinforcement learning. In 10th International Conference on Learning Representations (ICLR 2022) (IEEE, 2022).
Cited by
A Primer on Reinforcement Learning in Medicine for Clinicians
Jayaraman, P., Desman, J. M., Sabounchi, M., Nadkarni, G. N., & Sakhuja, A. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01316-0 · Cited by 191
“Despite their complexity, studies have demonstrated the suitability of relatively simple methods like FQE for policy 77,78 79 andaproof-of-concepttrialofinsulincontrolfortype2diabetes 80 .”
Artificial intelligence for diabetes care: current and future prospects
Sheng, B., Pushpanathan, K., Guan, Z., Lim, Q. H., Lim, Z. W., Yew, S. M. E., Goh, J. H. L., Bee, Y. M., Sabanayagam, C., Sevdalis, N., et al. (2024) · The Lancet Diabetes & Endocrinology · doi:10.1016/s2213-8587(24)00154-2 · Cited by 106
Large language models for diabetes care: Potentials and prospects
Sheng, B., Guan, Z., Lim, L., Jiang, Z., Mathioudakis, N. N., Li, J., Liu, R., Bao, Y., Bee, Y. M., Wang, Y. X., et al. (2024) · 中国科学通报:英文版 · doi:10.1016/j.scib.2024.01.004 · Cited by 72
Nagayach, A., Bhaskar, R., Ghosh, S., Singh, D. K. K., Han, S. S., & Sinha, J. K. (2024) · Ageing Research Reviews · doi:10.1016/j.arr.2024.102450 · Cited by 58
A foundational architecture for AI agents in healthcare
Liu, F., Niu, Y., Zhang, Q., Wang, K., Dong, Z., Wong, I. N., Cheng, L., Li, T., Duan, L., Li, K., et al. (2025) · Cell Reports Medicine · doi:10.1016/j.xcrm.2025.102374 · Cited by 41
“90,104,105 Through integration of wearable devices, biosensors, and remote monitoring tools, AI agents can collect and analyze a range of physiological data, including heart rate, blood pressure, and blood glucose levels.”
Metadata
- Authors
- Wang, G., Liu, X., Ying, Z., Yang, G., Chen, Z., Liu, Z., Zhang, M., Yan, H., Lu, Y., Gao, Y., Xue, K., Li, X., & Chen, Y.
- Affiliations
- Shanghai Medical College of Fudan University (CN)
- Beijing University of Posts and Telecommunications (CN)
- Fudan University (CN)
- Shanghai Xuhui Central Hospital (CN)
- Peking University (CN)
- University of Oxford (GB)
- Corresponding author
- Wang, Guangyu
- Li, Xiaoying
- Chen, Ying
- ORCID
- Venue
- Nature Medicine
- Published
- 2023-09-14
- Volume
- 29
- Issue
- 10
- Pages
- 2633-2642
- ISSN
- 1078-8956 (ISSN-L)
- 1546-170X
- PMC
- PMC10579102
- OpenAlex
- W4386741020
- Cited by
- 156
- References
- 43
- Citations by year
- 2026: 49 · 2025: 75 · 2024: 26 · 2023: 6
- FWCI
- 31.77
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- Nature Medicine (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Funders
- National Natural Science Foundation of China (62272055, 81820108008, 31830041, 82000822)
- the Tencent Foundation through the XPLORER PRIZE, and Young Elite Scientists Sponsorship Program by cs
- Topics
- Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Management and Research
- Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Hyperglycemia and glycemic control in critically ill and hospitalized patients
- Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Treatment and Management
- Keywords
- type 2 diabetes, insulin dose adjustment, glycemic control, reinforcement learning, hospitalized patients, hypoglycemia, hyperglycemia, capillary blood glucose
- MeSH terms
- Glycemic Control, Artificial Intelligence, Blood Glucose, Diabetes Mellitus, Type 2, Humans, Hypoglycemic Agents, Insulin
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Optimized glycemic control of type 2 diabetes with reinforcement learning: a proof-of-concept trial. https://app.cochat.ai/papers/10.1038/s41591-023-02552-9
- The paper
- Wang, G., Liu, X., Ying, Z., Yang, G., Chen, Z., Liu, Z., Zhang, M., Yan, H., Lu, Y., Gao, Y., Xue, K., Li, X., & Chen, Y. (2023). Optimized glycemic control of type 2 diabetes with reinforcement learning: a proof-of-concept trial. Nature Medicine. https://doi.org/10.1038/s41591-023-02552-9
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.