Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Oct 6;23(19):8269.
doi: 10.3390/s23198269.

Comparative Analysis of Predictive Interstitial Glucose Level Classification Models

Svjatoslavs Kistkins  1 , Timurs Mihailovs  2 , Sergejs Lobanovs  1 , Valdis Pīrāgs  1 , Harald Sourij  3 , Othmar Moser  4 , Dmitrijs Bļizņuks  2   5
Affiliations

Comparative Analysis of Predictive Interstitial Glucose Level Classification Models

Svjatoslavs Kistkins et al. Sensors (Basel). .

Abstract

Background: New methods of continuous glucose monitoring (CGM) provide real-time alerts for hypoglycemia, hyperglycemia, and rapid fluctuations of glucose levels, thereby improving glycemic control, which is especially crucial during meals and physical activity. However, complex CGM systems pose challenges for individuals with diabetes and healthcare professionals, particularly when interpreting rapid glucose level changes, dealing with sensor delays (approximately a 10 min difference between interstitial and plasma glucose readings), and addressing potential malfunctions. The development of advanced predictive glucose level classification models becomes imperative for optimizing insulin dosing and managing daily activities.

Methods: The aim of this study was to investigate the efficacy of three different predictive models for the glucose level classification: (1) an autoregressive integrated moving average model (ARIMA), (2) logistic regression, and (3) long short-term memory networks (LSTM). The performance of these models was evaluated in predicting hypoglycemia (<70 mg/dL), euglycemia (70-180 mg/dL), and hyperglycemia (>180 mg/dL) classes 15 min and 1 h ahead. More specifically, the confusion matrices were obtained and metrics such as precision, recall, and accuracy were computed for each model at each predictive horizon.

Results: As expected, ARIMA underperformed the other models in predicting hyper- and hypoglycemia classes for both the 15 min and 1 h horizons. For the 15 min forecast horizon, the performance of logistic regression was the highest of all the models for all glycemia classes, with recall rates of 96% for hyper, 91% for norm, and 98% for hypoglycemia. For the 1 h forecast horizon, the LSTM model turned out to be the best for hyper- and hypoglycemia classes, achieving recall values of 85% and 87% respectively.

Conclusions: Our findings suggest that different models may have varying strengths and weaknesses in predicting glucose level classes, and the choice of model should be carefully considered based on the specific requirements and context of the clinical application. The logistic regression model proved to be more accurate for the next 15 min, particularly in predicting hypoglycemia. However, the LSTM model outperformed logistic regression in predicting glucose level class for the next hour. Future research could explore hybrid models or ensemble approaches that combine the strengths of multiple models to further enhance the accuracy and reliability of glucose predictions.

Keywords: ARIMA; CGM; LSTM; diabetes; hyperglycemia; hypoglycemia; logistic regression; prediction.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflicts of interest.

Figures

Figure 1
Figure 1
CGM data excerpts for real (left) and virtual patients (right).
Figure 2
Figure 2
(a) Engineered features, (b) engineered features—a snippet for 54–62 h range.
Figure 2
Figure 2
(a) Engineered features, (b) engineered features—a snippet for 54–62 h range.
Figure 3
Figure 3
Confusion matrices for ARIMA model.
Figure 4
Figure 4
Confusion matrices for logistic regression model.
Figure 5
Figure 5
Confusion matrices for LSTM network model.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Zaharieva D.P., Turksoy K., McGaugh S.M., Pooni R., Vienneau T., Ly T., Riddell M.C. Lag Time Remains with Newer Real-Time Continuous Glucose Monitoring Technology During Aerobic Exercise in Adults Living with Type 1 Diabetes. Diabetes Technol. Ther. 2019;21:313–321. doi: 10.1089/dia.2018.0364. - DOI - PMC - PubMed
    1. Rodbard D. Continuous Glucose Monitoring: A Review of Successes, Challenges, and Opportunities. Diabetes Technol. Ther. 2016;18((Suppl. S2)):S3–S13. doi: 10.1089/dia.2015.0417. - DOI - PMC - PubMed
    1. Blauw H., Keith-Hynes P., Koops R., DeVries J.H. A Review of Safety and Design Requirements of the Artificial Pancreas. Ann. Biomed. Eng. 2016;44:3158–3172. doi: 10.1007/s10439-016-1679-2. - DOI - PMC - PubMed
    1. Martinsson J., Schliep A., Eliasson B., Mogren O. Blood Glucose Prediction with Variance Estimation Using Recurrent Neural Networks. J. Healthc. Inform. Res. 2019;4:1–18. doi: 10.1007/s41666-019-00059-y. - DOI - PMC - PubMed
    1. Zhu T., Li K., Chen J., Herrero P., Georgiou P. Dilated Recurrent Neural Networks for Glucose Forecasting in Type 1 Diabetes. J. Healthc. Inform. Res. 2020;4:308–324. doi: 10.1007/s41666-020-00068-2. - DOI - PMC - PubMed