A pretrained transformer model for decoding individual glucose dynamics from continuous glucose monitoring data
National Science Review · 04.04.2025 · 12 (5)
DOI: 10.1093/nsr/nwaf039 ↗Kurz gefasst
Fachartikel in der Zeitschrift National Science Review, online erschienen am 08.02.2025. Verfasst von Yurun Lu und 13 weiteren Autor:innen. Beteiligte Einrichtung: Center for Excellence in Mathematical Sciences, National Center for Mathematics and Interdisciplinary Sciences, Hua Loo-Keng Center for Mathematical …. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 91 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, CAS Project for Young Scientists in Basic Research, National Natural Science Foundation of China.
Abstract (Originalsprache)
Continuous glucose monitoring (CGM) technology has grown rapidly to track real-time blood glucose levels and trends with improved sensor accuracy. The ease of use and wide availability of CGM will facilitate safe and effective decision making for diabetes management. Here, we developed an attention-based deep learning model, CGMformer, pretrained on a well-controlled and diverse corpus of CGM data to represent individual's intrinsic metabolic state and enable clinical applications. During pretraining, CGMformer encodes glucose dynamics including glucose level, fluctuation, hyperglycemia, and hypoglycemia into latent space with self-supervised learning. It shows generalizability in imputing glucose value across five external datasets with different populations and metabolic states (MAE = 3.7 mg/dL). We then fine-tuned CGMformer towards a diverse panel of downstream tasks in the screening of diabetes and its complications using task-specific data, which demonstrated a consistently boosted predictive accuracy over direct fine-tuning on a single task (AUROC = 0.914 for type 2 diabetes (T2D) screening and 0.741 for complication screening). By learning an intrinsic representation of an individual's glucose dynamics, CGMformer classifies non-diabetic individuals into six clusters with elevated T2D risks, and identifies a specific cluster with lean body-shape but high risk of glucose metabolism disorders, which is overlooked by traditional glucose measurements. Furthermore, CGMformer achieves high accuracy in predicting an individual's postprandial glucose response with dietary modelling (Pearson correlation coefficient = 0.763) and helps personalized dietary recommendations. Overall, CGMformer pretrains a transformer neural network architecture to learn an intrinsic representation by borrowing information from a large amount of daily glucose profiles, and demonstrates predictive capabilities fine-tuned towards a broad range of downstream applications, holding promise for the early warning of T2D and recommendations for lifestyle modification in diabetes management.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Center for Excellence in Mathematical Sciences, National Center for Mathematics and Interdisciplinary Sciences, Hua Loo-Keng Center for Mathematical Sciences, Key Laboratory of Management, Decision and Information Syste…
- School of Mathematics, University of Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing 100049, China.
- Department of Endocrinology and Metabolism, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai Diabetes Institute, Shanghai Clinical Center for Diabetes, Shanghai K…
- Key Laboratory of Systems Health Science of Zhejiang Province, School of Life Science, Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China.
- BGI-Research, Hangzhou 310030, China.
- School of Mathematics, South China University of Technology, Guangzhou 510640, China.
ORCID-Profile: Yurun Lu, Rui Liu, Pei Chen, Zhanying Feng, Weiping Jia, Yong Wang
Förderung
- National Key Research and Development Program of China
- CAS Project for Young Scientists in Basic Research
- National Natural Science Foundation of China
- Zhejiang Province Vanguard Goose-Leading Initiative
- JST Moonshot Research and Development Program
Zitieren
APA
Lu, Y., Liu, D., Liang, Z., Liu, R., Chen, P., Liu, Y., Li, J., Feng, Z., Li, L. M., Sheng, B., Jia, W., Chen, L., Li, H., & Wang, Y. (2025). A pretrained transformer model for decoding individual glucose dynamics from continuous glucose monitoring data. National Science Review, 12(5). https://doi.org/10.1093/nsr/nwaf039
Vancouver
Lu Y, Liu D, Liang Z, Liu R, Chen P, Liu Y, et al. A pretrained transformer model for decoding individual glucose dynamics from continuous glucose monitoring data. National Science Review. 2025;12(5). doi:10.1093/nsr/nwaf039
BibTeX
@article{Lu2025,
author = {Lu, Yurun and Liu, Dan and Liang, Zhongming and Liu, Rui and Chen, Pei and Liu, Yitong and Li, Jiachen and Feng, Zhanying and Li, Lei M and Sheng, Bin and Jia, Weiping and Chen, Luonan and Li, Huating and Wang, Yong},
title = {A pretrained transformer model for decoding individual glucose dynamics from continuous glucose monitoring data},
journal = {National Science Review},
year = {2025},
volume = {12},
number = {5},
doi = {10.1093/nsr/nwaf039}
}
EUVIMED