{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T15:42:27Z","timestamp":1773157347964,"version":"3.50.1"},"reference-count":36,"publisher":"MDPI AG","issue":"19","license":[{"start":{"date-parts":[[2023,10,6]],"date-time":"2023-10-06T00:00:00Z","timestamp":1696550400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"State Research Programme project in biomedical, medical technologies and pharmaceuticals","award":["VPP-EM-BIOMEDIC\u012aNA-2022\/1-0001"],"award-info":[{"award-number":["VPP-EM-BIOMEDIC\u012aNA-2022\/1-0001"]}]},{"name":"State Research Programme project in biomedical, medical technologies and pharmaceuticals","award":["KLIF-1076 to HS"],"award-info":[{"award-number":["KLIF-1076 to HS"]}]},{"name":"Austrian Science Fund (FWF)","award":["VPP-EM-BIOMEDIC\u012aNA-2022\/1-0001"],"award-info":[{"award-number":["VPP-EM-BIOMEDIC\u012aNA-2022\/1-0001"]}]},{"name":"Austrian Science Fund (FWF)","award":["KLIF-1076 to HS"],"award-info":[{"award-number":["KLIF-1076 to HS"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>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 (&lt;70 mg\/dL), euglycemia (70\u2013180 mg\/dL), and hyperglycemia (&gt;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.<\/jats:p>","DOI":"10.3390\/s23198269","type":"journal-article","created":{"date-parts":[[2023,10,6]],"date-time":"2023-10-06T08:29:12Z","timestamp":1696580952000},"page":"8269","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Comparative Analysis of Predictive Interstitial Glucose Level Classification Models"],"prefix":"10.3390","volume":"23","author":[{"given":"Svjatoslavs","family":"Kistkins","sequence":"first","affiliation":[{"name":"Research Institute of Pauls Stradins Clinical University Hospital, LV-1002 Riga, Latvia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Timurs","family":"Mihailovs","sequence":"additional","affiliation":[{"name":"Institute of Smart Computing Technologies, Riga Technical University, LV-1048 Riga, Latvia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-9896-0817","authenticated-orcid":false,"given":"Sergejs","family":"Lobanovs","sequence":"additional","affiliation":[{"name":"Research Institute of Pauls Stradins Clinical University Hospital, LV-1002 Riga, Latvia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Valdis","family":"P\u012br\u0101gs","sequence":"additional","affiliation":[{"name":"Research Institute of Pauls Stradins Clinical University Hospital, LV-1002 Riga, Latvia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3510-9594","authenticated-orcid":false,"given":"Harald","family":"Sourij","sequence":"additional","affiliation":[{"name":"Interdisciplinary Metabolic Medicine Trials Unit, Division of Endocrinology and Diabetology, Medical University of Graz, 8010 Graz, Austria"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Othmar","family":"Moser","sequence":"additional","affiliation":[{"name":"Division of Exercise Physiology and Metabolism, Institute of Sport Science, University of Bayreuth, 95447 Bayreuth, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4252-9220","authenticated-orcid":false,"given":"Dmitrijs","family":"B\u013ciz\u0146uks","sequence":"additional","affiliation":[{"name":"Institute of Smart Computing Technologies, Riga Technical University, LV-1048 Riga, Latvia"},{"name":"SIA \u201cR4U\u201d, LV-1016 Riga, Latvia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2023,10,6]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"313","DOI":"10.1089\/dia.2018.0364","article-title":"Lag Time Remains with Newer Real-Time Continuous Glucose Monitoring Technology During Aerobic Exercise in Adults Living with Type 1 Diabetes","volume":"21","author":"Zaharieva","year":"2019","journal-title":"Diabetes Technol. Ther."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"S3","DOI":"10.1089\/dia.2015.0417","article-title":"Continuous Glucose Monitoring: A Review of Successes, Challenges, and Opportunities","volume":"18","author":"Rodbard","year":"2016","journal-title":"Diabetes Technol. Ther."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"3158","DOI":"10.1007\/s10439-016-1679-2","article-title":"A Review of Safety and Design Requirements of the Artificial Pancreas","volume":"44","author":"Blauw","year":"2016","journal-title":"Ann. Biomed. Eng."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s41666-019-00059-y","article-title":"Blood Glucose Prediction with Variance Estimation Using Recurrent Neural Networks","volume":"4","author":"Martinsson","year":"2019","journal-title":"J. Healthc. Inform. Res."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1007\/s41666-020-00068-2","article-title":"Dilated Recurrent Neural Networks for Glucose Forecasting in Type 1 Diabetes","volume":"4","author":"Zhu","year":"2020","journal-title":"J. Healthc. Inform. Res."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1109\/JBHI.2021.3100558","article-title":"A Multitask Learning Approach to Personalized Blood Glucose Prediction","volume":"26","author":"Daniels","year":"2022","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Khadem, H., Nemat, H., Elliott, J., and Benaissa, M. (2023). Blood Glucose Level Time Series Forecasting: Nested Deep Ensemble Learning Lag Fusion. Bioengineering, 10.","DOI":"10.3390\/bioengineering10040487"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Rodr\u00edguez-Rodr\u00edguez, I., Rodr\u00edguez, J.-V., Molina-Garc\u00eda-Pardo, J.-M., Zamora-Izquierdo, M.-\u00c1., and Mart\u00ednez-Ingl\u00e9s, M.-T.M.-I.I. (2020). A Comparison of Different Models of Glycemia Dynamics for Improved Type 1 Diabetes Mellitus Management with Advanced Intelligent Analysis in an Internet of Things Context. Appl. Sci., 10.","DOI":"10.3390\/app10124381"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Rabby, M.F., Tu, Y., Hossen, M.I., Lee, I., Maida, A.S., and Hei, X. (2021). Stacked LSTM based deep recurrent neural network with kalman smoothing for blood glucose prediction. BMC Med. Inform. Decis. Mak., 21.","DOI":"10.1186\/s12911-021-01462-5"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"112673","DOI":"10.1016\/j.chaos.2022.112673","article-title":"A novel blood glucose time series prediction framework based on a novel signal decomposition method","volume":"164","author":"Zhu","year":"2022","journal-title":"Chaos Solitons Fractals"},{"key":"ref_11","unstructured":"(2023, August 03). Available online: https:\/\/github.com\/jxx123\/simglucose."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"849","DOI":"10.1111\/dom.14643","article-title":"Humoral immune response to COVID-19 vaccination in diabetes is age-dependent but independent of type of diabetes and glycaemic control: The prospective COVAC-DM cohort study","volume":"24","author":"Sourij","year":"2022","journal-title":"Diabetes Obes. Metab."},{"key":"ref_13","unstructured":"Sadegh, M., Hui, S., Razvan, B., and Cindy, M. (2019, January 23\u201327). Lstms and neural attention models for blood glucose prediction: Comparative experiments on real and synthetic data. Proceedings of the 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), IEEE, Berlin, Germany."},{"key":"ref_14","unstructured":"Ali, M., Alexander, J.R., Nicklas, H., Peter, C.E., Jens, T.M., Morten, J.L., Henrik, B., and Morten, M. (2020). Short term blood glucose prediction based on continuous glucose monitoring data. arXiv."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"842","DOI":"10.1177\/1932296820922622","article-title":"Feature-Based Machine Learning Model for Real-Time Hypoglycemia Prediction","volume":"15","author":"Dave","year":"2021","journal-title":"J. Diabetes Sci. Technol."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Prendin, F., Del Favero, S., Vettoretti, M., Sparacino, G., and Facchinetti, A. (2021). Forecasting of Glucose Levels and Hypoglycemic Events: Head-to-Head Comparison of Linear and Nonlinear Data-Driven Algorithms Based on Continuous Glucose Monitoring Data Only. Sensors, 21.","DOI":"10.3390\/s21051647"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Mujahid, O., Contreras, I., and Vehi, J. (2021). Machine Learning Techniques for Hypoglycemia Prediction: Trends and Challenges. Sensors, 21.","DOI":"10.3390\/s21020546"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e0253125","DOI":"10.1371\/journal.pone.0253125","article-title":"Brouwers, Machine learning-based glucose prediction with use of continuous glucose and physical activity monitoring data: The Maastricht Study","volume":"16","author":"Foreman","year":"2021","journal-title":"PLoS ONE"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1152\/physrev.00045.2011","article-title":"Mechanisms of diabetic complications","volume":"93","author":"Forbes","year":"2013","journal-title":"Physiol. Rev."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"S61","DOI":"10.2337\/dc19-S006","article-title":"6. Glycemic Targets: Standards of Medical Care in Diabetes-2019","volume":"42","year":"2019","journal-title":"Diabetes Care"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"385","DOI":"10.1016\/S2213-8587(18)30315-2","article-title":"Hypoglycaemia, cardiovascular disease, and mortality in diabetes: Epidemiology, pathogenesis, and management","volume":"7","year":"2019","journal-title":"Lancet Diabetes Endocrinol."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1186\/1479-5868-5-56","article-title":"A comparison of direct versus self-report measures for assessing physical activity in adults: A systematic review","volume":"5","author":"Prince","year":"2008","journal-title":"Int. J. Behav. Nutr. Phys. Act."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2501","DOI":"10.1007\/s00125-020-05263-9","article-title":"Glucose management for exercise using continuous glucose monitoring (CGM) and intermittently scanned CGM (isCGM) systems in type 1 diabetes: Position statement of the European Association for the Study of Diabetes (EASD) and of the International Society for Pediatric and Adolescent Diabetes (ISPAD) endorsed by JDRF and supported by the American Diabetes Association (ADA)","volume":"63","author":"Moser","year":"2020","journal-title":"Diabetologia"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"527","DOI":"10.1007\/s00592-019-01454-y","article-title":"Frequent physical activity is associated with reduced risk of severe diabetic retinopathy in type 1 diabetes","volume":"57","author":"Forsblom","year":"2020","journal-title":"Acta Diabetol."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"1727","DOI":"10.2337\/dc17-0615","article-title":"Physical Activity Reduces Risk of Premature Mortality in Patients With Type 1 Diabetes With and Without Kidney Disease","volume":"40","author":"Forsblom","year":"2017","journal-title":"Diabetes Care"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"e0136489","DOI":"10.1371\/journal.pone.0136489","article-title":"Effects of High-Intensity Interval Exercise versus Moderate Continuous Exercise on Glucose Homeostasis and Hormone Response in Patients with Type 1 Diabetes Mellitus Using Novel Ultra-Long-Acting Insulin","volume":"10","author":"Moser","year":"2015","journal-title":"PLoS ONE"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"164","DOI":"10.1016\/j.diabet.2011.10.005","article-title":"The barriers to physical activity in Type 1 diabetes (BAPAD-1) scale: Predictive validity and reliability","volume":"38","author":"Brazeau","year":"2012","journal-title":"Diab. Metabol."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"2243","DOI":"10.1111\/dom.15102","article-title":"Impact of severe hypoglycaemia requiring hospitalization on mortality in people with type 1 diabetes: A national retrospective observational cohort study","volume":"25","author":"Moser","year":"2023","journal-title":"Diabetes Obes. Metab."},{"key":"ref_29","unstructured":"(2023, August 03). Standards of Medical Care in Diabetes. Available online: https:\/\/diabetesjournals.org\/care\/issue\/45\/Supplement_1."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"574","DOI":"10.1007\/s00125-016-4189-8","article-title":"Frequent and intensive physical activity reduces risk of cardiovascular events in type 1 diabetes","volume":"60","author":"Forsblom","year":"2017","journal-title":"Diabetologia"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"125","DOI":"10.1016\/j.pedn.2022.05.022","article-title":"Related factors associated with fear of hypoglycemia in parents of children and adolescents with type 1 diabetes\u2014A systematic review","volume":"66","author":"Zhang","year":"2022","journal-title":"J. Pediatr. Nurs."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1639","DOI":"10.1007\/s11845-021-02741-w","article-title":"Adherence to physical activity recommendations and barriers to physical activity participation among adults with type 1 diabetes","volume":"191","author":"Finn","year":"2022","journal-title":"Ir. J. Med. Sci."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1373","DOI":"10.1007\/s00592-021-01718-6","article-title":"Symptoms of depression are associated with reduced leisure-time physical activity in adult individuals with type 1 diabetes","volume":"58","author":"Ahola","year":"2021","journal-title":"Acta Diabetol."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"109","DOI":"10.1016\/j.artmed.2019.07.007","article-title":"Data-driven modeling and prediction of blood glucose dynamics: Machine learning applications in type 1 diabetes","volume":"98","author":"Woldaregay","year":"2019","journal-title":"Artif. Intell. Med."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"8605206","DOI":"10.1155\/2019\/8605206","article-title":"Blood Glucose Level Prediction for Diabetics Based on Nutrition and Insulin Administration Logs Using Personalized Mathematical Models","volume":"2019","author":"Gyuk","year":"2019","journal-title":"J. Healthc. Eng."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"e000499","DOI":"10.1136\/bmjdrc-2017-000499","article-title":"Development and validation of a prediction model for insulin-associated hypoglycemia in non-critically ill hospitalized adults","volume":"6","author":"Mathioudakis","year":"2018","journal-title":"BMJ Open Diabetes Res. Care"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/19\/8269\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T21:01:54Z","timestamp":1760130114000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/23\/19\/8269"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,6]]},"references-count":36,"journal-issue":{"issue":"19","published-online":{"date-parts":[[2023,10]]}},"alternative-id":["s23198269"],"URL":"https:\/\/doi.org\/10.3390\/s23198269","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,10,6]]}}}