A multimodal deep learning architecture for predicting interstitial glucose for effective type 2 diabetes management
The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
The accurate prediction of blood glucose is critical for the effective management of diabetes. Modern continuous glucose monitoring (CGM) technology enables real-time acquisition of interstitial glucose concentrations, which can be calibrated against blood glucose measurements. However, a key challenge in the effective management of type 2 diabetes lies in forecasting critical events driven by glucose variability. While recent advances in deep learning enable modeling of temporal patterns in glucose fluctuations, most of the existing methods rely on unimodal inputs and fail to account for individual physiological differences that influence interstitial glucose dynamics. These limitations highlight the need for multimodal approaches that integrate additional personalized physiological information. One of the primary reasons for multimodal approaches not being widely studied in this field is the bottleneck associated with the availability of subjects’ health records. In this paper, we propose a multimodal approach trained on sequences of CGM values and enriched with physiological context derived from health records of 40 individuals with type 2 diabetes. The CGM time series were processed using a stacked Convolutional Neural Network (CNN) and a Bidirectional Long Short-Term Memory (BiLSTM) network followed by an attention mechanism. The BiLSTM learned long-term temporal dependencies, while the CNN captured local sequential features. Physiological heterogeneity was incorporated through a separate pipeline of neural networks that processed baseline health records and was later fused with the CGM modeling stream. To validate our model, we utilized CGM values of 30 min sampled with a moving window of 5 min to predict the CGM values with a prediction horizon of (a) 15 min, (b) 30 min, and (c) 60 min. We achieved the multimodal architecture prediction results with Mean Absolute Point Error (MAPE) between 14 and 24 mg/dL, 19–22 mg/dL, 25–26 mg/dL in case of Menarini sensor and 6–11 mg/dL, 9–14 mg/dL, 12–18 mg/dL in case of Abbot sensor for 15, 30 and 60 min prediction horizon respectively. The results suggested that the proposed multimodal model achieved higher prediction accuracy compared to unimodal approaches; with upto 96.7% prediction accuracy; supporting its potential as a generalizable solution for interstitial glucose prediction and personalized management in the type 2 diabetes population.
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
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?
In the citing paper
“Emerging multimodal deep learning models that fuse real‑time CGM and clinical data can now forecast interstitial glucose 15–60 min ahead with clinically acceptable accuracy, creating a window for algorithms to anticipate glycemic excursions rather than merely record them [23].”
From Digital Management of Early-Onset Type 2 Diabetes: Empowerment, Challenges, and Future Outlook (2026) · cites this paper as [23]
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
Pathophysiology of type 2 diabetes mellitus
Galicia-Garcia, U., Benito-Vicente, A., Jebari, S., Larrea-Sebal, A., Siddiqi, H., Uribe, K. B., Ostolaza, H., & Martín, C. (2020) · International Journal of Molecular Sciences · doi:10.3390/ijms21176275 · Cited by 3,360
The Performance and Usability of a Factory-Calibrated Flash Glucose Monitoring System
Bailey, T., Bode, B. W., Christiansen, M. P., Klaff, L. J., & Alva, S. (2015) · Diabetes Technology & Therapeutics · doi:10.1089/dia.2014.0378 · Cited by 685 · cited 1×
“In contrast, Continuous Glucose Monitoring (CGM) measures the concentration of glucose in the interstitial fluid at regular intervals 10 .”
Type 2 diabetes mellitus in older adults: clinical considerations and management
Bellary, S., Kyrou, I., Brown, J. E., & Bailey, C. J. (2021) · Nature Reviews Endocrinology · doi:10.1038/s41574-021-00512-2 · Cited by 639
Deng, Y., Lu, L., Aponte, L., Angelidi, A. M., Novak, V., Karniadakis, G. E., & Mantzoros, C. S. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00480-x · Cited by 152 · cited 1×
“Regarding T2D, Deng et al. 35 employed deep transfer learning with data augmentation to improve robustness under limited data conditions.”
Technical Aspects of the Parkes Error Grid
Pfützner, A., Klonoff, D. C., Pardo, S., & Parkes, J. L. (2013) · Journal of Diabetes Science and Technology · doi:10.1177/193229681300700517 · Cited by 122
Research works printed but not linked: 32
- Zimmet, P., Alberti, K. G., Magliano, D. J. & Bennett, P. H. Diabetes mellitus statistics on prevalence and mortality: Facts and fallacies. Nat. Rev. Endocrinol. 12(10), 616–622 (2016). · DOI · Paper page
- Sun, H. et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res. Clin. Pract. 183, 109119 (2022). · DOI · Paper page
- Parker, E. D. et al. Economic costs of diabetes in the US in 2022. Diabetes Care 47(1), 26–43 (2024). · doi:10.2337/dci23-0085
- Kaufman, J. M., Thommandram, A. & Fossat, Y. Acoustic analysis and prediction of type 2 diabetes mellitus using smartphone-recorded voice segments. Mayo Clinic Proceedings: Digital Health 1(4), 534–544 (2023).
- N. S. Padhye, J. Wang, Pattern of active and inactive sequences of diabetes self-monitoring in mobile phone and paper diary users, in 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2015: IEEE, pp. 7630–7633. · DOI · Paper page
- Ernawati, U., Wihastuti, T. A. & Utami, Y. W. Effectiveness of diabetes self-management education (DSME) in type 2 diabetes mellitus (T2DM) patients: Systematic literature review. J. Public Health Res. 10(2), jphr-2021 (2021). · DOI · Paper page
- Wagner, J., Malchoff, C. & Abbott, G. Invasiveness as a barrier to self-monitoring of blood glucose in diabetes. Diabetes Technol. Ther. 7(4), 612–619 (2005). · DOI · Paper page
- A. D. Association. The American Diabetes Association Releases Standards of Care in Diabetes—2025. https://www.diabesrelief.online/newsroom/press-releases/american-diabetes-association-releases-standards-care-diabetes-2025 (accessed 02–05, 2025).
- Kebede, M. M., Zeeb, H., Peters, M., Heise, T. L. & Pischke, C. R. Effectiveness of digital interventions for improving glycemic control in persons with poorly controlled type 2 diabetes: A systematic review, meta-analysis, and meta-regression analysis. Diabetes Technol. Ther. 20(11), 767–782 (2018). · DOI · Paper page
- Sobel, S. I., Chomentowski, P. J., Vyas, N., Andre, D. & Toledo, F. G. Accuracy of a novel noninvasive multisensor technology to estimate glucose in diabetic subjects during dynamic conditions. J. Diabetes Sci. Technol. 8(1), 54–63 (2014). · DOI
- Andellini, M. et al. Artificial intelligence for non-invasive glycaemic-events detection via ECG in a paediatric population: Study protocol. Heal. Technol. 13(1), 145–154 (2023). · DOI
- Cisuelo, O. et al. Development of an artificial intelligence system to identify hypoglycaemia via ECG in adults with type 1 diabetes: Protocol for data collection under controlled and free-living conditions. BMJ Open 13(4), e067899 (2023). · DOI
- Fokkert, M. et al. Performance of the FreeStyle Libre Flash glucose monitoring system in patients with type 1 and 2 diabetes mellitus. BMJ Open Diabetes Res. Care 5(1), e000320 (2017). · DOI · Paper page
- van Doorn, W. P. et al. Machine learning-based glucose prediction with use of continuous glucose and physical activity monitoring data: The Maastricht Study. PLoS ONE 16(6), e0253125 (2021). · DOI · Paper page
- Liu, K. et al. Machine learning models for blood glucose level prediction in patients with diabetes mellitus: Systematic review and network meta-analysis. JMIR Med. Inform. 11(1), e47833 (2023). · DOI · Paper page
- Service, F. J. Glucose variability. Diabetes 62(5), 1398–1404 (2013). · DOI · Paper page
- Kim, S. & Kim, H. A new metric of absolute percentage error for intermittent demand forecasts. Int. J. Forecast. 32(3), 669–679. https://doi.org/10.1016/j.ijforecast.2015.12.003 (2016). · DOI · Paper page
- Kim, T. K. T test as a parametric statistic. Korean J. Anesthesiol. 68(6), 540–546 (2015). · DOI · Paper page
- de Batlle, J. et al. GATEKEEPER’s strategy for the multinational large-scale piloting of an eHealth platform: tutorial on how to identify relevant settings and use cases. J. Med. Internet Res. 25, e42187. https://doi.org/10.2196/42187 (2023). · DOI
- LeRoith, D. et al. Treatment of diabetes in older adults: An endocrine society* clinical practice guideline. J. Clin. Endocrinol. Metab. 104(5), 1520–1574. https://doi.org/10.1210/jc.2019-00198 (2019). · DOI · Paper page
- Portability, I. & Act, A. Guidance regarding methods for de-identification of protected health information in accordance with the health insurance portability and accountability act (HIPAA) privacy rule (Human Health Services, 2012).
- Felizardo, V., Garcia, N. M., Pombo, N. & Megdiche, I. Data-based algorithms and models using diabetics real data for blood glucose and hypoglycaemia prediction–a systematic literature review. Artif. Intell. Med. 118, 102120 (2021). · DOI
- Martinsson, J., Schliep, A., Eliasson, B. & Mogren, O. Blood glucose prediction with variance estimation using recurrent neural networks. J. Healthc. Inform. Res. 4, 1–18 (2020). · DOI · Paper page
- C. Marling, R. Bunescu, The OhioT1DM dataset for blood glucose level prediction: Update 2020, in CEUR workshop proceedings, 2020, vol. 2675: NIH Public Access, p. 71.
- Koca, Ö. A., Kabak, H. Ö. & Kılıç, V. Attention-based multilayer GRU decoder for on-site glucose prediction on smartphone. J. Supercomput. 80(17), 25616–25639 (2024). · DOI
- Seo, W., Park, S.-W., Kim, N., Jin, S.-M. & Park, S.-M. A personalized blood glucose level prediction model with a fine-tuning strategy: A proof-of-concept study. Comput. Methods Programs Biomed. 211, 106424 (2021). · DOI
- Zhu, T., Li, K., Herrero, P. & Georgiou, P. Personalized blood glucose prediction for type 1 diabetes using evidential deep learning and meta-learning. IEEE Trans. Biomed. Eng. 70(1), 193–204 (2022). · DOI · Paper page
- Daniels, J., Herrero, P. & Georgiou, P. A multitask learning approach to personalized blood glucose prediction. IEEE J. Biomed. Health Inform. 26(1), 436–445 (2021). · DOI
- Y. Sun, P. Kosmas, Integrating bayesian approaches and expert knowledge for forecasting continuous glucose monitoring values in type 2 diabetes mellitus, IEEE Journal of Biomedical and Health Informatics 2024. · DOI
- Yang, T., Yu, X., Tao, R., Li, H. & Zhou, J. Blood glucose prediction for type 2 diabetes using clustering-based domain adaptation. Biomed. Signal Process. Control 105, 107629 (2025). · DOI
- Montaser, E., Díez, J.-L. & Bondia, J. Glucose prediction under variable-length time-stamped daily events: A seasonal stochastic local modeling framework. Sensors 21(9), 3188 (2021). · DOI
- Piao, C. et al. GARNN: an interpretable graph attentive recurrent neural network for predicting blood glucose levels via multivariate time series. Neural Netw. 185, 107229 (2025). · DOI
Cited by
Digital Management of Early-Onset Type 2 Diabetes: Empowerment, Challenges, and Future Outlook
Guo, L.-J., Low, K.-S., Tae, S.-K., & Lim, L.-L. (2026) · Current Diabetes Reports · doi:10.1007/s11892-026-01633-6
Metadata
- Authors
- Haleem, M. S., Katsarou, D., Georga, E. I., Dafoulas, G. E., Bargiota, A., Lopez-Perez, L., Rujas, M., Fico, G., Pecchia, L., Fotiadis, D., Gatekeeper Consortium, Caimi, C., Tamporale, C., Manea, M., Bonferini, C., Gaeta, E., Sánchez, G. C., Drympeta, I., Votis, K., Folkvord, F., & de Battle, J.
- Venue
- Scientific Reports
- Published
- 2025-07-29
- Volume
- 15
- Issue
- 1
- ISSN
- 2045-2322
- Cited by
- 14
- References
- 5
- Language
- en
- Update notices
- correction (2025-08-20)
- Licence
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Funders
- GATEKEEPER EU Horizon 2020 (857223)
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: A multimodal deep learning architecture for predicting interstitial glucose for effective type 2 diabetes management. https://app.cochat.ai/papers/10.1038/s41598-025-07272-3
- The paper
- Haleem, M. S., Katsarou, D., Georga, E. I., Dafoulas, G. E., Bargiota, A., Lopez-Perez, L., Rujas, M., Fico, G., Pecchia, L., Fotiadis, D., Gatekeeper Consortium, Caimi, C., Tamporale, C., Manea, M., Bonferini, C., Gaeta, E., Sánchez, G. C., Drympeta, I., Votis, K., . . . de Battle, J. (2025). A multimodal deep learning architecture for predicting interstitial glucose for effective type 2 diabetes management. Scientific Reports. https://doi.org/10.1038/s41598-025-07272-3
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.