Benchmarking Machine Learning Algorithms on Blood Glucose Prediction for Type I Diabetes in Comparison With Classical Time-Series Models
Abstract
OBJECTIVE: This paper aims to compare the performance of several commonly known machine-learning (ML) models versus a classic Autoregression with Exogenous inputs (ARX) model in the prediction of blood glucose (BG) levels using time-series data of patients with Type 1 diabetes (T1D). METHODS: The ML algorithms include ML-based regression models and deep learning models such as a vanilla Long-Short-Term-Memory (LSTM) Network and a Temporal Convolution Network (TCN). Evaluations have been conducted with respect to different input features, regression model orders, as well as using the recursive method or direct method for multi-step prediction of BG levels. Prediction performance metrics include the average Root Mean Square Error (RMSE), temporal gain (TG) for early prediction, and the normalized energy of the second-order differences (ESOD) of the predicted time series to reflect risk of false alerts on hypo/hyper glycemia events. RESULTS: The ARX model achieved the lowest average RMSE for both recursive and direct methods, the second highest average TG under the direct method, but with a higher average normalized ESOD than some other models. CONCLUSION: There was no significant advantage observed from the ML models compared to the classic ARX model in predicting BG levels for T1D, except that TCN's performance was more robust with respect to BG trajectories with spurious oscillations, for which ARX tended to over-predict peak BG values and under-predict valley BG values. SIGNIFICANCE: Insight learned from this study could help researchers and clinical practitioners to select appropriate models for BG prediction.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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 Xie and Wang's comparison, the models showing best performance in terms of RMSE were ARX, Ridge regression, linear SVR, LSTM, and TCN, with RMSE between 19 and 20 mg/dL.
“The models showing best performance in terms RMSE were ARX, Ridge regression, linear SVR, LSTM, and TCN, with RMSE between 19 and 20 mg/dL.”
From Advanced Diabetes Management Using Artificial Intelligence and Continuous Glucose Monitoring Sensors (2020) · cites this paper as [58]
In Xie and Wang's comparison, the maximum time gain was achieved for Gradient boosting trees and LSTM, which predicted CGM data on average 7 min in advance.
“The maximum TG was achieved for Gradient boosting trees and LSTM, which made it possible to predict the CGM data, on average, 7 min in advance.”
From Advanced Diabetes Management Using Artificial Intelligence and Continuous Glucose Monitoring Sensors (2020) · cites this paper as [58]
In Xie and Wang's comparison on the OhioT1DM dataset, the best results were achieved by considering all available inputs in a 45-min time-window before the prediction time.
“The best results were achieved by considering all of the available inputs in a 45-min time-window before the prediction time.”
From Advanced Diabetes Management Using Artificial Intelligence and Continuous Glucose Monitoring Sensors (2020) · cites this paper as [58]
In Xie and Wang's comparison, considering both RMSE and time gain, linear SVR and TCN performed better, with RMSE around 20 mg/dL and a time gain of about 6 min.
“Considering both of the metrics, the models that performed better were linear SVR and TCN with RMSE around 20 mg/dL and a TG of about 6 min.”
From Advanced Diabetes Management Using Artificial Intelligence and Continuous Glucose Monitoring Sensors (2020) · cites this paper as [58]
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · see citing works
WaveNet: a generative model for raw audio
Oord, A. V. D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1609.03499 · Cited by 4,911 · see citing works
An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
Bai, S., Kolter, J. Z., & Koltun, V. (2018) · arXiv · doi:10.48550/arxiv.1803.01271 · Cited by 4,655 · see citing works
The Lack of A Priori Distinctions Between Learning Algorithms
Wolpert, D. H. (1996) · Neural Computation · doi:10.1162/neco.1996.8.7.1341 · Cited by 1,652 · see citing works
imputeTS: Time Series Missing Value Imputation in R
Möritz, S., & Bartz–Beielstein, T. (2017) · The R Journal · doi:10.32614/rj-2017-009 · Cited by 936 · see citing works
A Convergence Theory for Deep Learning via Over-ParameterizationPreprint
Allen-Zhu, Z., Li, Y., & Song, Z. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.03962 · Cited by 847 · see citing works
In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep LearningPreprint
Neyshabur, B., Tomioka, R., & Srebro, N. (2014) · International Conference on Learning Representations · doi:10.48550/arxiv.1412.6614 · Cited by 742 · see citing works
Learning Overparameterized Neural Networks via Stochastic Gradient Descent on Structured DataPreprint
Li, Y., & Liang, Y. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.01204 · Cited by 414 · see citing works
Partitioning glucose distribution/transport, disposal, and endogenous production during IVGTT
Hovorka, R., Shojaee-Moradie, F., Carroll, P. V., Chassin, L. J., Gowrie, I. J., Jackson, N. C., Tudor, R. S., Umpleby, A. M., & Jones, R. H. (2002) · American Journal of Physiology-Endocrinology and Metabolism · doi:10.1152/ajpendo.00304.2001 · Cited by 393 · see citing works
Sparacino, G., Zanderigo, F., Corazza, S., Maran, A., Facchinetti, A., & Cobelli, C. (2007) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2006.889774 · Cited by 339 · see citing works
Zecchin, C., Facchinetti, A., Sparacino, G., Nicolao, G. D., & Cobelli, C. (2012) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2012.2188893 · Cited by 181 · see citing works
Estimation of Future Glucose Concentrations with Subject-Specific Recursive Linear Models
Eren-Oruklu, M., Çinar, A., Quinn, L., & Smith, D. (2009) · Diabetes Technology & Therapeutics · doi:10.1089/dia.2008.0065 · Cited by 164 · see citing works
A Multi-Patient Data-Driven Approach to Blood Glucose Prediction
Aliberti, A., Pupillo, I., Terna, S., Macii, E., Cataldo, S. D., Patti, E., & Acquaviva, A. (2019) · IEEE Access · doi:10.1109/access.2019.2919184 · Cited by 130 · see citing works
Predicting Subcutaneous Glucose Concentration in Humans: Data-Driven Glucose Modeling
Gani, A., Gribok, A. V., Rajaraman, S., Ward, W. K., & Jaques Reifman, A. (2008) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2008.2005937 · Cited by 129 · see citing works
A Deep Learning Approach to Diabetic Blood Glucose Prediction
Mhaskar, H. N., Pereverzev, S. V., & van der Walt, M. D. (2017) · Frontiers in Applied Mathematics and Statistics · doi:10.3389/fams.2017.00014 · Cited by 117 · see citing works
A Machine Learning Approach to Predicting Blood Glucose Levels for Diabetes Management
Plis, K., Bunescu, R., Marling, C., et al. (2014) · Cited by 109 · see citing works
Lee, H., & Bequette, B. (2009) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2009.03.002 · Cited by 108 · see citing works
Using LSTMs to learn physiological models of blood glucose behavior
Mirshekarian, S., Bunescu, R., Marling, C., & Schwartz, F. (2017) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc.2017.8037460 · Cited by 100 · see citing works
The OhioT1DM Dataset For Blood Glucose Level Prediction.
Marling, C., & Bunescu, R. (2018) · International Joint Conference on Artificial Intelligence · Cited by 92 · see citing works
Wang, Q., Molenaar, P. C. M., Harsh, S., Freeman, K. A., Xie, J., Gold, C. H., Rovine, M., & Ulbrecht, J. S. (2014) · Journal of Diabetes Science and Technology · doi:10.1177/1932296814524080 · Cited by 90 · see citing works
Self-citationFox, I., Ang, L. P., Jaiswal, M., Pop‐Busui, R., & Wiens, J. (2018) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3219819.3220102 · Cited by 79 · see citing works
Statistical Software for State Space Methods
Commandeur, J. J. F., Koopman, S. J., & Ooms, M. (2011) · Journal of Statistical Software · doi:10.18637/jss.v041.i01 · Cited by 70 · see citing works
Zecchin, C., Facchinetti, A., Sparacino, G., & Cobelli, C. (2016) · Journal of Diabetes Science and Technology · doi:10.1177/1932296816654161 · Cited by 64 · see citing works
Percival, M. W., Bevier, W., Wang, Y.-Q., Dassau, E., Zisser, H., Jovanovic, L., & Doyle, F. (2010) · Journal of Diabetes Science and Technology · doi:10.1177/193229681000400522 · Cited by 61 · see citing works
Facchinetti, A., Sparacino, G., Trifoglio, E., & Cobelli, C. (2011) · Diabetes Technology & Therapeutics · doi:10.1089/dia.2010.0151 · Cited by 41 · see citing works
Practical Issues in the Identification of Empirical Models from Simulated Type 1 Diabetes Data
Finan, D. A., Zisser, H., Jovanovic, L. G., et al. (2007) · Diabetes Technology & Therapeutics · doi:10.1089/dia.2007.0202 · Cited by 37 · see citing works
Xie, J., & Wang, Q. (2018) · International Joint Conference on Artificial Intelligence · Cited by 30 · see citing works
Self-citationTurksoy, K., Bayrak, E. S., Quinn, L., Littlejohn, E., & Çınar, A. (2013) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.1109/acc.2013.6580275 · Cited by 23 · see citing works
Xie, J., & Wang, Q. (2019) · Smart Health · doi:10.1016/j.smhl.2019.100069 · Cited by 19 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Advanced Diabetes Management Using Artificial Intelligence and Continuous Glucose Monitoring Sensors
Vettoretti, M., Cappon, G., Facchinetti, A., & Sparacino, G. (2020) · Sensors · doi:10.3390/s20143870 · Cited by 176 · see citing works
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 · see citing works
A bibliometric analysis of the advance of artificial intelligence in medicine
Lin, M., Lin, L., Lin, L., Lin, Z., & Yan, X. (2025) · Frontiers in Medicine · doi:10.3389/fmed.2025.1504428 · Cited by 36 · see citing works
- Authors
- Xie, J., & Wang, Q.
- Affiliations
- Pennsylvania State University (US)
- ORCID
- Venue
- IEEE Transactions on Biomedical Engineering
- Published
- 2020-11
- Volume
- 67
- Issue
- 11
- Pages
- 3101-3124
- ISSN
- 0018-9294 (ISSN-L)
- 1558-2531
- OpenAlex
- W3008906415
- Cited by
- 158
- References
- 30
- Citations by year
- 2026: 15 · 2025: 39 · 2024: 28 · 2023: 30 · 2022: 26 · 2021: 16 · 2020: 4
- FWCI
- 14.37
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Biomedical Engineering (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Penn State University Social Science Research Institute (SSRI LEVEL 2)
- Topics
- Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Management and Research
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Keywords
- blood glucose prediction, type 1 diabetes, ARX model, LSTM, temporal convolutional network, multi-step forecasting, model comparison, regression models
- MeSH terms
- Machine Learning, Algorithms, Blood Glucose, Diabetes Mellitus, Type 1, Humans, Benchmarking
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
