Quantifying the impact of physical activity on future glucose trends using machine learning

Tyler, Nichole S.; Mosquera-Lopez, Clara; Young, Gary P.; Youssef, Joseph El; Castle, Jessica R.; Jacobs, Peter G.

doi:10.1016/j.isci.2022.103888

Cited by 21 publications

(17 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have shown that when using the blood glucose data of real patients, we need additional features, such as the ones extracted from heart rate signals. The Logistic Regression, AdaBoost, Random Forest, and Multi-Layer Perceptron with the ReLU and Tanh activation function were found to be the best five models from our tests; they provide better or comparable results to those reported in similar studies, e.g., [ 61 , 62 , 63 , 64 ]. The other models require a post-processing of the predicted data to be more precise.…”

Section: Discussionsupporting

confidence: 68%

Detection of Physical Activity Using Machine Learning Methods Based on Continuous Blood Glucose Monitoring and Heart Rate Signals

Dénes-Fazakas

Siket

Szilágyi

et al. 2022

Sensors

View full text Add to dashboard Cite

Non-coordinated physical activity may lead to hypoglycemia, which is a dangerous condition for diabetic people. Decision support systems related to type 1 diabetes mellitus (T1DM) still lack the capability of automated therapy modification by recognizing and categorizing the physical activity. Further, this desired adaptive therapy should be achieved without increasing the administrative load, which is already high for the diabetic community. These requirements can be satisfied by using artificial intelligence-based solutions, signals collected by wearable devices, and relying on the already available data sources, such as continuous glucose monitoring systems. In this work, we focus on the detection of physical activity by using a continuous glucose monitoring system and a wearable sensor providing the heart rate—the latter is accessible even in the cheapest wearables. Our results show that the detection of physical activity is possible based on these data sources, even if only low-complexity artificial intelligence models are deployed. In general, our models achieved approximately 90% accuracy in the detection of physical activity.

show abstract

Section: Discussionsupporting

confidence: 68%

Detection of Physical Activity Using Machine Learning Methods Based on Continuous Blood Glucose Monitoring and Heart Rate Signals

Dénes-Fazakas

Siket

Szilágyi

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Although the mean decline in glucose differed among the three exercise types, and change in glucose was variable, we also report that TIR generally increased by ∼6% over the next 24 h after the 30-min structured exercise session, regardless of activity type, compared with a day with no reported exercise. The heterogeneity of each participant’s glucose change during exercise, as demonstrated by the low intraclass correlation coefficient of 0.12, has been observed in other studies that have also allowed variation in the time of day for exercise and the temporal relationship of exercise timing relative to bolus insulin dosing ( 13 ).…”

Section: Discussionmentioning

confidence: 84%

Examining the Acute Glycemic Effects of Different Types of Structured Exercise Sessions in Type 1 Diabetes in a Real-World Setting: The Type 1 Diabetes and Exercise Initiative (T1DEXI)

Riddell

Gal

et al. 2023

Diabetes Care

Self Cite

View full text Add to dashboard Cite

OBJECTIVE Maintenance of glycemic control during and after exercise remains a major challenge for individuals with type 1 diabetes. Glycemic responses to exercise may differ by exercise type (aerobic, interval, or resistance), and the effect of activity type on glycemic control after exercise remains unclear. RESEARCH DESIGN AND METHODS The Type 1 Diabetes Exercise Initiative (T1DEXI) was a real-world study of at-home exercise. Adult participants were randomly assigned to complete six structured aerobic, interval, or resistance exercise sessions over 4 weeks. Participants self-reported study and nonstudy exercise, food intake, and insulin dosing (multiple daily injection [MDI] users) using a custom smart phone application and provided pump (pump users), heart rate, and continuous glucose monitoring data. RESULTS A total of 497 adults with type 1 diabetes (mean age ± SD 37 ± 14 years; mean HbA1c ± SD 6.6 ± 0.8% [49 ± 8.7 mmol/mol]) assigned to structured aerobic (n = 162), interval (n = 165), or resistance (n = 170) exercise were analyzed. The mean (± SD) change in glucose during assigned exercise was −18 ± 39, −14 ± 32, and −9 ± 36 mg/dL for aerobic, interval, and resistance, respectively (P < 0.001), with similar results for closed-loop, standard pump, and MDI users. Time in range 70–180 mg/dL (3.9–10.0 mmol/L) was higher during the 24 h after study exercise when compared with days without exercise (mean ± SD 76 ± 20% vs. 70 ± 23%; P < 0.001). CONCLUSIONS Adults with type 1 diabetes experienced the largest drop in glucose level with aerobic exercise, followed by interval and resistance exercise, regardless of insulin delivery modality. Even in adults with well-controlled type 1 diabetes, days with structured exercise sessions contributed to clinically meaningful improvement in glucose time in range but may have slightly increased time below range.

show abstract

“…Subsequently, WGCNA analysis was performed to identify the modules associated with cell scorching, and ultimately 11 candidate genes were retrieved from the intersection, namely, SMIM1, FBLN2, ZFP2, B4GALT5, HCRT, SLC6A17, MUSK, SLC26A8, CRHR2, SEZ6L2, and KCNJ15. Machine learning is a powerful tool to perform complex algorithms to detect and diagnose clinical diseases [ 42 , 43 ]. In this research, the genes were filtered by the LASSO model, and ultimately the value of SMIM1 and SEZ6L2 in the diagnosis of IDD pyroptosis-related genes was clarified by the validation set.…”

Section: Discussionmentioning

confidence: 99%

Identification of SMIM1 and SEZ6L2 as Potential Biomarkers for Genes Associated with Intervertebral Disc Degeneration in Pyroptosis

et al. 2022

View full text Add to dashboard Cite

Background. Inflammatory reactions and pyroptosis play an important role in the pathology of intervertebral disc degeneration (IDD). The aim of the present study was to investigate pyroptosis in the nucleus pulposus cells (NPCs) of inflammatory induced IDD by bioinformatic methods and to search for possible diagnostic biomarkers. Methods. Gene expression profiles related to IDD were downloaded from the GEO database to identify differentially expressed genes (DEGs) between inflammation-induced IDD and non-inflammatory intervention samples. Pyroptosis genes were then searched for, and their expression in IDD was analyzed. Weighted gene co-expression network analysis (WGCNA) was then used to search for modules of IDD genes associated with pyroptosis and intersected with DEGs to discover candidate genes that would be diagnostically valuable. A LASSO model was developed to screen for genes that met the requirements, and ROC curves were created to clarify the diagnostic value of the genetic markers. Ultimately, the screened genes were further validated, and their diagnostic value assessed by selecting gene sets from the GEO database. RT-PCR was used to assess the mRNA expression of diagnostic markers in the nucleus pulposus (NP). Pan-cancer analysis was applied to demonstrate the expression and prognostic value of the screened genes in various tumors. Results. A total of 733 DEGs were identified in GSE41883 and GSE27494, which were mainly enriched in transmembrane receptor protein serine/threonine, kinase signaling pathway, response to lipopolysaccharide, and other biological processes, and they were mainly related to TGF beta signaling pathway, toll-like receptor signaling pathway, and TNF signaling pathway. A total of 81 genes related to pyroptosis were identified in the literature, and eight genes related to IDD were identified in the Veen diagram, namely, IL1A, IL1B, NOD2, GBP1, IL6, AK1, EEF2K, and PYCARD. Eleven candidate genes were obtained after locating the intersection of pyroptosis-related module genes and DEGs according to WGCNA analysis. A total of six valid genes were obtained after constructing a machine learning model, and five key genes were finally identified after correlation analysis. GSE23132 and GSE56081 validated the candidate genes, and the final IDD-related diagnostic markers were obtained as SMIM1 and SEZ6L2. RT-PCR results indicated that the mRNA expression of both was significantly elevated in IDD. The pan-cancer analysis demonstrated that SMIM1 and SEZ6L2 have important roles in the expression and prognosis of various tumors. Conclusion. In conclusion, this research identifies SMIM1 and SEZ6L2 as important biomarkers of IDD associated with pyroptosis, which will help to unravel the development and pathogenesis of IDD and determine potential therapeutic targets.

show abstract

Quantifying the impact of physical activity on future glucose trends using machine learning

Cited by 21 publications

References 48 publications

Detection of Physical Activity Using Machine Learning Methods Based on Continuous Blood Glucose Monitoring and Heart Rate Signals

Detection of Physical Activity Using Machine Learning Methods Based on Continuous Blood Glucose Monitoring and Heart Rate Signals

Examining the Acute Glycemic Effects of Different Types of Structured Exercise Sessions in Type 1 Diabetes in a Real-World Setting: The Type 1 Diabetes and Exercise Initiative (T1DEXI)

Identification of SMIM1 and SEZ6L2 as Potential Biomarkers for Genes Associated with Intervertebral Disc Degeneration in Pyroptosis

Contact Info

Product

Resources

About