Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models
Continuous glucose monitoring (CGM) represents a significant advancement in diabetes management, playing an important role in glycemic control for patients with type 1 diabetes (T1D). Despite their benefits, their performance is affected by numerous factors such as the carbohydrate intake, alcohol c...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-10-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/21/9870 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850193059831087104 |
|---|---|
| author | David Chushig-Muzo Hugo Calero-Díaz Himar Fabelo Eirik Årsand Peter Ruben van Dijk Cristina Soguero-Ruiz |
| author_facet | David Chushig-Muzo Hugo Calero-Díaz Himar Fabelo Eirik Årsand Peter Ruben van Dijk Cristina Soguero-Ruiz |
| author_sort | David Chushig-Muzo |
| collection | DOAJ |
| description | Continuous glucose monitoring (CGM) represents a significant advancement in diabetes management, playing an important role in glycemic control for patients with type 1 diabetes (T1D). Despite their benefits, their performance is affected by numerous factors such as the carbohydrate intake, alcohol consumption, and physical activity (PA). Among these, PA could cause hypoglycemic episodes, which might happen after exercising. In this work, two main contributions are presented. First, we extend the performance evaluation of two glucose monitoring devices, Eversense and Free Style Libre (FSL), for measuring glucose concentrations during high-intensity PA and normal daily activity (NDA). The impact of PA is investigated considering (1) different glucose ranges (hypoglycemia, euglycemia, and hyperglycemia); and (2) four time periods throughout the day (morning, afternoon, evening, and night). Second, we evaluate the effectiveness of machine learning (ML) models, including logistic regression, K-nearest neighbors, and support vector machine, to automatically detect PA in T1D individuals using glucose measurements. The performance analysis showed significant differences between glucose levels obtained in the PA and NDA period for Eversense and FSL devices, specially in the hyperglycemic range and two time intervals (morning and afternoon). Both Eversense and FSL devices present measurements with large variability during strenuous PA, indicating that their users should be cautious. However, glucose recordings provided by monitoring devices are accurate for NDA, reaching similar values to capillary glucose device. Lastly, ML-based models yielded promising results to determine when an individual has performed PA, reaching an accuracy value of 0.93. The results can be used to develop an individualized data-driven classifier for each patient that categorizes glucose profiles based on the time interval during the day and according to if a patient performs PA. Our work contributes to the analysis of PA on the performance of CGM devices. |
| format | Article |
| id | doaj-art-b081238701254ade942a1cbcb5cb3ec2 |
| institution | OA Journals |
| issn | 2076-3417 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-b081238701254ade942a1cbcb5cb3ec22025-08-20T02:14:22ZengMDPI AGApplied Sciences2076-34172024-10-011421987010.3390/app14219870Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning ModelsDavid Chushig-Muzo0Hugo Calero-Díaz1Himar Fabelo2Eirik Årsand3Peter Ruben van Dijk4Cristina Soguero-Ruiz5Department of Signal Theory and Communications, Telematics and Computing Systems, Rey Juan Carlos University, Fuenlabrada, 28943 Madrid, SpainDepartment of Signal Theory and Communications, Telematics and Computing Systems, Rey Juan Carlos University, Fuenlabrada, 28943 Madrid, SpainFundación Canaria Instituto de Investigación Sanitaria de Canarias (FIISC), 35012 Las Palmas de Gran Canaria, SpainDepartment of Computer Science, Faculty of Science and Technology, UiT The Arctic University of Norway, 9019 Tromsø, NorwayDepartment of Internal Medicine, Divisions of Endocrinology, Isala, Diabetes Center, 8025 AB Zwolle, The NetherlandsDepartment of Signal Theory and Communications, Telematics and Computing Systems, Rey Juan Carlos University, Fuenlabrada, 28943 Madrid, SpainContinuous glucose monitoring (CGM) represents a significant advancement in diabetes management, playing an important role in glycemic control for patients with type 1 diabetes (T1D). Despite their benefits, their performance is affected by numerous factors such as the carbohydrate intake, alcohol consumption, and physical activity (PA). Among these, PA could cause hypoglycemic episodes, which might happen after exercising. In this work, two main contributions are presented. First, we extend the performance evaluation of two glucose monitoring devices, Eversense and Free Style Libre (FSL), for measuring glucose concentrations during high-intensity PA and normal daily activity (NDA). The impact of PA is investigated considering (1) different glucose ranges (hypoglycemia, euglycemia, and hyperglycemia); and (2) four time periods throughout the day (morning, afternoon, evening, and night). Second, we evaluate the effectiveness of machine learning (ML) models, including logistic regression, K-nearest neighbors, and support vector machine, to automatically detect PA in T1D individuals using glucose measurements. The performance analysis showed significant differences between glucose levels obtained in the PA and NDA period for Eversense and FSL devices, specially in the hyperglycemic range and two time intervals (morning and afternoon). Both Eversense and FSL devices present measurements with large variability during strenuous PA, indicating that their users should be cautious. However, glucose recordings provided by monitoring devices are accurate for NDA, reaching similar values to capillary glucose device. Lastly, ML-based models yielded promising results to determine when an individual has performed PA, reaching an accuracy value of 0.93. The results can be used to develop an individualized data-driven classifier for each patient that categorizes glucose profiles based on the time interval during the day and according to if a patient performs PA. Our work contributes to the analysis of PA on the performance of CGM devices.https://www.mdpi.com/2076-3417/14/21/9870continuous glucose monitoringtype 1 diabetesphysical activitymachine learningTabPFN |
| spellingShingle | David Chushig-Muzo Hugo Calero-Díaz Himar Fabelo Eirik Årsand Peter Ruben van Dijk Cristina Soguero-Ruiz Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models Applied Sciences continuous glucose monitoring type 1 diabetes physical activity machine learning TabPFN |
| title | Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models |
| title_full | Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models |
| title_fullStr | Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models |
| title_full_unstemmed | Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models |
| title_short | Characterizing the Impact of Physical Activity on Patients with Type 1 Diabetes Using Statistical and Machine Learning Models |
| title_sort | characterizing the impact of physical activity on patients with type 1 diabetes using statistical and machine learning models |
| topic | continuous glucose monitoring type 1 diabetes physical activity machine learning TabPFN |
| url | https://www.mdpi.com/2076-3417/14/21/9870 |
| work_keys_str_mv | AT davidchushigmuzo characterizingtheimpactofphysicalactivityonpatientswithtype1diabetesusingstatisticalandmachinelearningmodels AT hugocalerodiaz characterizingtheimpactofphysicalactivityonpatientswithtype1diabetesusingstatisticalandmachinelearningmodels AT himarfabelo characterizingtheimpactofphysicalactivityonpatientswithtype1diabetesusingstatisticalandmachinelearningmodels AT eirikarsand characterizingtheimpactofphysicalactivityonpatientswithtype1diabetesusingstatisticalandmachinelearningmodels AT peterrubenvandijk characterizingtheimpactofphysicalactivityonpatientswithtype1diabetesusingstatisticalandmachinelearningmodels AT cristinasogueroruiz characterizingtheimpactofphysicalactivityonpatientswithtype1diabetesusingstatisticalandmachinelearningmodels |