Deep Learning Applied to Blood Glucose Prediction from Flash Glucose Monitoring and Fitbit Data

2020 ◽  
pp. 59-63
Author(s):  
Pietro Bosoni ◽  
Marco Meccariello ◽  
Valeria Calcaterra ◽  
Cristiana Larizza ◽  
Lucia Sacchi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Glucose Monitoring ◽  
Flash Glucose Monitoring ◽  
Glucose Prediction

Applying Deep Learning to Track Food Consumption and Human Activity for Non-intrusive Blood Glucose Monitoring

2021 ◽  
Author(s):  
Mohamed Amr Samir ◽  
Zeinab A. Mohamed ◽  
Mona Abdelmotaleb A. Hussein ◽  
Ayman Atia
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Food Consumption ◽  
Human Activity ◽  
Glucose Monitoring ◽  
Blood Glucose Monitoring

scholarly journals Improved HbA1c and reduced glycaemic variability after 1-year intermittent use of flash glucose monitoring

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Wenhui Zhang ◽  
Yu Liu ◽  
Baosheng Sun ◽  
Yanjun Shen ◽  
Ming Li ◽  
...  
Keyword(s):  
Blood Glucose ◽  
Insulin Dose ◽  
Glucose Monitoring ◽  
Final Visit ◽  
Glycaemic Variability ◽  
Self Monitoring ◽  
Flash Glucose Monitoring ◽  
Significant Difference ◽  
Mean Blood Glucose ◽  
Hba1c Measurement

AbstractFlash glucose monitoring (FGM) was introduced in China in 2016, and it might improve HbA1c measurements and reduce glycaemic variability during T1DM therapy. A total of 146 patients were recruited from October 2018 to September 2019 in Liaocheng. The patients were randomly divided into the FGM group or self-monitoring blood glucose (SMBG) group. Both groups wore the FGM device for multiple 2-week periods, beginning with the 1st, 24th, and 48th weeks for gathering data, while blood samples were also collected for HbA1c measurement. Dietary guidance and insulin dose adjustments were provided to the FGM group patients according to their Ambulatory Glucose Profile (AGP) and to the SMBG group patients according to their SMBG measurements taken 3–4 times daily. All of the participants underwent SMBG measurements on the days when not wearing the FGM device. At the final visit, HbA1c, time in range (TIR), duration of hypoglycaemia and the number of diabetic ketoacidosis (DKA) events were taken as the main endpoints. There were no significant difference in the baseline characteristics of the two groups. At 24 weeks, the HbA1c level of the FGM group was 8.16 ± 1.03%, which was much lower than that of the SMBG group (8.68 ± 1.01%) (p = 0.003). The interquartile range (IQR), mean blood glucose (MBG), and the duration of hypoglycaemia in the FGM group also showed significant declines, compared with the SMBG group (p < 0.05), while the TIR increased in the FGM group [(49.39 ± 17.54)% vs (42.44 ± 15.49)%] (p = 0.012). At 48 weeks, the differences were more pronounced (p < 0.01). There were no observed changes in the number of episodes of DKA by the end of the study [(0.25 ± 0.50) vs (0.28 ± 0.51), p = 0.75]. Intermittent use of FGM by T1DM patients can improve their HbA1c and glycaemic control without increasing the hypoglycaemic exposure in insulin-treated individuals with type 1 diabetes in an developing country.

scholarly journals Flash glucose monitoring helps achieve better glycemic control than conventional self-monitoring of blood glucose in non-insulin-treated type 2 diabetes: a randomized controlled trial

2020 ◽  
Vol 8 (1) ◽  
pp. e001115 ◽  
Author(s):  
Eri Wada ◽  
Takeshi Onoue ◽  
Tomoko Kobayashi ◽  
Tomoko Handa ◽  
Ayaka Hayase ◽  
...  
Keyword(s):  
Type 2 Diabetes ◽  
Blood Glucose ◽  
Glycemic Control ◽  
Glucose Monitoring ◽  
Analysis Of Covariance ◽  
Self Monitoring ◽  
Flash Glucose Monitoring ◽  
Baseline Values ◽  
The Mean

IntroductionThe present study aimed to evaluate the effects of flash glucose monitoring (FGM) and conventional self-monitoring of blood glucose (SMBG) on glycemic control in patients with non-insulin-treated type 2 diabetes.Research design and methodsIn this 24-week, multicenter, open-label, randomized (1:1), parallel-group study, patients with non-insulin-treated type 2 diabetes at five hospitals in Japan were randomly assigned to the FGM (n=49) or SMBG (n=51) groups and were provided each device for 12 weeks. The primary outcome was change in glycated hemoglobin (HbA1c) level, and was compared using analysis of covariance model that included baseline values and group as covariates.ResultsForty-eight participants in the FGM group and 45 in the SMBG group completed the study. The mean HbA1c levels were 7.83% (62.1 mmol/mol) in the FGM group and 7.84% (62.2 mmol/mol) in the SMBG group at baseline, and the values were reduced in both FGM (−0.43% (−4.7 mmol/mol), p<0.001) and SMBG groups (−0.30% (−3.3 mmol/mol), p=0.001) at 12 weeks. On the other hand, HbA1c was significantly decreased from baseline values in the FGM group, but not in the SMBG group at 24 weeks (FGM: −0.46% (−5.0 mmol/mol), p<0.001; SMBG: −0.17% (−1.8 mmol/mol), p=0.124); a significant between-group difference was also observed (difference −0.29% (−3.2 mmol/mol), p=0.022). Diabetes Treatment Satisfaction Questionnaire score was significantly improved, and the mean glucose levels, SD of glucose, mean amplitude of glycemic excursions and time in hyperglycemia were significantly decreased in the FGM group compared with the SMBG group.ConclusionsGlycemic control was better with FGM than with SMBG after cessation of glucose monitoring in patients with non-insulin-treated type 2 diabetes.Trial registration numberUMIN000026452, jRCTs041180082.

scholarly journals Flash Glucose Monitoring Technology Impact on Diabetes Self-Care Behavior

2019 ◽  
Vol 14 (2) ◽  
pp. 130-132 ◽  
Author(s):  
Nicole D. White ◽  
Emily Knezevich
Keyword(s):  
Blood Glucose ◽  
Glucose Monitoring ◽  
Self Care ◽  
Innovative Technology ◽  
Future Research ◽  
Self Monitoring ◽  
Flash Glucose Monitoring ◽  
Glucose Levels ◽  
Blood Glucose Levels ◽  
Care Behavior

Individuals with diabetes play a significant role in the control of their condition by participating in their own care. Self-monitoring of blood glucose is of particular importance in maintaining adequate glycemic control but when obtained using traditional fingerstick methods, is often limited with by cost, fear of needles or pain and inconvenience. Flash glucose monitoring is an innovative technology available to address these barriers and help people with diabetes better manage their blood glucose levels. Data demonstrating increased frequency in glucose monitoring, patient perspectives related to self-care behaviors, and implications for practice and future research are described.

scholarly journals Deep transfer learning: a novel glucose prediction framework for new subjects with type 2 diabetes

2021 ◽  
Author(s):  
Xia Yu ◽  
Tao Yang ◽  
Jingyi Lu ◽  
Yun Shen ◽  
Wei Lu ◽  
...  
Keyword(s):  
Type 2 Diabetes ◽  
Deep Learning ◽  
Transfer Learning ◽  
Prediction Models ◽  
Glucose Monitoring ◽  
Learning Networks ◽  
Deep Learning Network ◽  
Dynamic Time ◽  
Glucose Prediction

AbstractBlood glucose (BG) prediction is an effective approach to avoid hyper- and hypoglycemia, and achieve intelligent glucose management for patients with type 1 or serious type 2 diabetes. Recent studies have tended to adopt deep learning networks to obtain improved prediction models and more accurate prediction results, which have often required significant quantities of historical continuous glucose-monitoring (CGM) data. However, for new patients with limited historical dataset, it becomes difficult to establish an acceptable deep learning network for glucose prediction. Consequently, the goal of this study was to design a novel prediction framework with instance-based and network-based deep transfer learning for cross-subject glucose prediction based on segmented CGM time series. Taking the effects of biodiversity into consideration, dynamic time warping (DTW) was applied to determine the proper source domain dataset that shared the greatest degree of similarity for new subjects. After that, a network-based deep transfer learning method was designed with cross-domain dataset to obtain a personalized model combined with improved generalization capability. In a case study, the clinical dataset demonstrated that, with additional segmented dataset from other subjects, the proposed deep transfer learning framework achieved more accurate glucose predictions for new subjects with type 2 diabetes.

scholarly journals Benefits of the addition of continuous or flash glucose monitoring versus standard practice using self-monitored blood glucose and haemoglobin A1c in the primary care of diabetes mellitus: a systematic review protocol

BMJ Open ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. e050027
Author(s):  
Alexander Kieu ◽  
Romona Devi Govender ◽  
Linda Östlundh ◽  
Jeffrey King
Keyword(s):  
Diabetes Mellitus ◽  
Systematic Review ◽  
Primary Care ◽  
Blood Glucose ◽  
Glycaemic Control ◽  
Glucose Monitoring ◽  
Physician Satisfaction ◽  
Haemoglobin A1c ◽  
Flash Glucose Monitoring ◽  
Patients With Diabetes

IntroductionStudies demonstrate that optimal glycaemic control reduces morbidity from diabetes mellitus but remains elusive in a significant portion of patients. Although research shows that continuous glucose monitoring (CGM) and flash glucose monitoring (FGM) improves glycaemic control in selected subsets of patients with diabetes in specialty practices, we found no systematic reviews evaluating the use of CGM/FGM in primary care, where the majority of patients with diabetes are cared for.This systematic review aims to answer the questions: ‘compared with usual care of self-monitoring blood glucose and haemoglobin A1c (HbA1c), does the addition of CGM/FGM use in the primary care of patients with diabetes improve glycaemic control, decrease rates of hypoglycaemia, and improve patient and physician satisfaction?’ and if so, ‘what subgroups of primary care patients with diabetes are most likely to benefit?’.Methods and analysisAligning with the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols guidelines, a search will be conducted in PubMed, EMBASE, Scopus, CINAHL, Cochrane Central Register of Controlled Trials and Web of Science. We will include studies investigating CGM/FGM use and reporting the primary outcome measure of HbA1c and secondary outcome measures of hypoglycaemia, time in range, time below range, time above range and patient/staff satisfaction. We will examine which patient populations appear to benefit from CGM/FGM. Three independent researchers will use the Covidence systematic review software for blinded screening and study selection. The National Heart, Lung, and Blood Institute quality assessment tool and Grading of Recommendations Assessment, Development and Evaluation will be used to assess the risk of bias and quality of evidence.Ethics and disseminationThe systematic review methodology does not require ethics approval due to the nature of the study design. Study findings will be publicly available to a wide readership across disciplines and will be published in a peer-reviewed journal.PROSPERO registration numberCRD42021229416.

scholarly journals Useful Measurement of Glucose Variability by Flash Glucose Monitoring (FGM) with the Efficacy of Sodium-Glucose Cotransporter 2 (SGLT2) Inhibitor

2020 ◽  
Vol 2 (S1) ◽  
pp. 1-8 ◽  
Author(s):  
Ebe K ◽  
Bando H ◽  
Muneta T ◽  
Bando M ◽  
Yonei Y
Keyword(s):  
Blood Glucose ◽  
Sglt2 Inhibitor ◽  
Glucose Monitoring ◽  
Glucose Variability ◽  
Postprandial Blood Glucose ◽  
Diabetic Patients ◽  
Flash Glucose Monitoring ◽  
Low Carbohydrate Diet ◽  
Sodium Glucose Cotransporter ◽  
Freestyle Libre

Diabetes has been a crucial medical and social problem worldwide. For adequate nutritional therapy, there have been discussions concerning Calorie Restriction (CR) and Low Carbohydrate Diet (LCD). We have investigated glucose variability of diabetic patients applying CR, LCD, continuous glucose monitoring (CGM) and applied FreeStyle Libre which is flash glucose monitoring (FGM). The patient is a 40-year-old female with type 2 diabetes mellitus (T2DM), who showed BMI 20.7, postprandial blood glucose 257 mg/dL. HbA1c 12.1%, Glycoalbumin 31.6% (11.6-16.4), serum C-peptide 2.0 ng/ml and unremarkable data of liver function, renal, lipids. She was provided the intervention of three stages, which are i) CR with 60% carbohydrate in Day 1-2, ii) LCD meal with 12% carbohydrate in Day 3-5; iii) LCD + Sodium-glucose cotransporter 2 (SGLT2) inhibitor (Ipragliflozin L-Proline 50mg) in Day 6-12. The glucose profile was measured by FreeStyle Libre Pro (Abbott) for 14 days. The daily profile of blood glucose was abruptly decreased on Day 6. Time percentage of satisfactory blood glucose 70-180 mg/dL (/24h) was 0%, 0%, 2%, 14%, 0%, 54%, 100% in Day 1-7, respectively. These results suppose the acute clinical efficacy of SGLT2 inhibitor, and this report would become a reference for future diabetic practice and research.

Sign in / Sign up

Export Citation Format

Share Document