Development of Non-Invasive Continuous Glucose Prediction Models Using Multi-Modal Wearable Sensors in Free-Living Conditions
Sensors · 20.05.2025 · 25 (10) · 3207
DOI: 10.3390/s25103207 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 20.05.2025. Verfasst von Thilini S Karunarathna und 1 weiteren Autor:innen. Beteiligte Einrichtung: Ubiquitous and Personal Computing Laboratory, Kyoto University of Advanced Science (KUAS), Kyoto 615-8577, Japan. Hauptthemen laut MEDLINE: Wearable Electronic Devices, Blood Glucose, Blood Glucose Self-Monitoring, Biosensing Techniques. Die Arbeit wurde bisher 25-mal zitiert (Crossref) und nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: KUAS Advanced Research Grant.
Abstract (Originalsprache)
Continuous monitoring of glucose levels is important for diabetes management and prevention. While traditional glucose monitoring methods are often invasive and expensive, recent approaches using machine learning (ML) models have explored non-invasive alternatives—but many still depend on manually logged food intake and activity, which is burdensome and impractical for everyday use. In this study, we propose a novel approach that eliminates the need for manual input by utilizing only passively collected, automatically recorded multi-modal data from non-invasive wearable sensors. This enables practical and continuous glucose prediction in real-world, free-living environments. We used the BIG IDEAs Lab Glycemic Variability and Wearable Device Data (BIGIDEAs) dataset, which includes approximately 26,000 CGM readings, simultaneous ly collected wearable data, and demographic information. A total of 236 features encompassing physiological, behavioral, circadian, and demographic factors were constructed. Feature selection was conducted using random-forest-based importance analysis to select the most relevant features for model training. We evaluated the effectiveness of various ML regression techniques, including linear regression, ridge regression, random forest regression, and XGBoost regression, in terms of prediction and clinical accuracy. Biological sex, circadian rhythm, behavioral features, and tonic features of electrodermal activity (EDA) emerged as key predictors of glucose levels. Tree-based models outperformed linear models in both prediction and clinical accuracy. The XGBoost (XR) model performed best, achieving an R-squared of 0.73, an RMSE of 11.9 mg/dL, an NRMSE of 0.52 mg/dL, a MARD of 7.1%, and 99.4% of predictions falling within Zones A and B of the Clarke Error Grid. This study demonstrates the potential of combining feature engineering and tree-based ML regression techniques for continuous glucose monitoring using wearable sensors.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Ubiquitous and Personal Computing Laboratory, Kyoto University of Advanced Science (KUAS), Kyoto 615-8577, Japan.
- Institute of Industrial Science, The University of Tokyo, Tokyo 153-8505, Japan.
ORCID-Profile: Thilini S Karunarathna, Zilu Liang
Förderung
- KUAS Advanced Research Grant
Zitieren
APA
Karunarathna, T. S., & Liang, Z. (2025). Development of Non-Invasive Continuous Glucose Prediction Models Using Multi-Modal Wearable Sensors in Free-Living Conditions. Sensors, 25(10), 3207. https://doi.org/10.3390/s25103207
Vancouver
Karunarathna TS, Liang Z. Development of Non-Invasive Continuous Glucose Prediction Models Using Multi-Modal Wearable Sensors in Free-Living Conditions. Sensors. 2025;25(10):3207. doi:10.3390/s25103207
BibTeX
@article{Karunarathna2025,
author = {Karunarathna, Thilini S and Liang, Zilu},
title = {Development of Non-Invasive Continuous Glucose Prediction Models Using Multi-Modal Wearable Sensors in Free-Living Conditions},
journal = {Sensors},
year = {2025},
volume = {25},
number = {10},
pages = {3207},
doi = {10.3390/s25103207}
}
EUVIMED