scholarly journals Development of a Deep Learning Model for Dynamic Forecasting of Blood Glucose Level for Type 2 Diabetes Mellitus: Secondary Analysis of a Randomized Controlled Trial (Preprint)

2019 ◽  
Author(s):  
Syed Hasib Akhter Faruqui ◽  
Yan Du ◽  
Rajitha Meka ◽  
Adel Alaeddini ◽  
Chengdong Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Glucose Level ◽  
Mobile Health ◽  
Learning Model ◽  
Health Data ◽  
Glucose Levels ◽  
Health Lifestyle ◽  
Deep Learning Model

BACKGROUND Type 2 diabetes mellitus (T2DM) is a major public health burden. Self-management of diabetes including maintaining a healthy lifestyle is essential for glycemic control and to prevent diabetes complications. Mobile-based health data can play an important role in the forecasting of blood glucose levels for lifestyle management and control of T2DM. OBJECTIVE The objective of this work was to dynamically forecast daily glucose levels in patients with T2DM based on their daily mobile health lifestyle data including diet, physical activity, weight, and glucose level from the day before. METHODS We used data from 10 T2DM patients who were overweight or obese in a behavioral lifestyle intervention using mobile tools for daily monitoring of diet, physical activity, weight, and blood glucose over 6 months. We developed a deep learning model based on long short-term memory–based recurrent neural networks to forecast the next-day glucose levels in individual patients. The neural network used several layers of computational nodes to model how mobile health data (food intake including consumed calories, fat, and carbohydrates; exercise; and weight) were progressing from one day to another from noisy data. RESULTS The model was validated based on a data set of 10 patients who had been monitored daily for over 6 months. The proposed deep learning model demonstrated considerable accuracy in predicting the next day glucose level based on Clark Error Grid and ±10% range of the actual values. CONCLUSIONS Using machine learning methodologies may leverage mobile health lifestyle data to develop effective individualized prediction plans for T2DM management. However, predicting future glucose levels is challenging as glucose level is determined by multiple factors. Future study with more rigorous study design is warranted to better predict future glucose levels for T2DM management.

scholarly journals Development of a Deep Learning Model for Dynamic Forecasting of Blood Glucose Level for Type 2 Diabetes Mellitus: Secondary Analysis of a Randomized Controlled Trial

10.2196/14452 ◽  
2019 ◽  
Vol 7 (11) ◽  
pp. e14452 ◽  
Author(s):  
Syed Hasib Akhter Faruqui ◽  
Yan Du ◽  
Rajitha Meka ◽  
Adel Alaeddini ◽  
Chengdong Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Blood Glucose ◽  
Glucose Level ◽  
Mobile Health ◽  
Learning Model ◽  
Health Data ◽  
Glucose Levels ◽  
Health Lifestyle ◽  
Deep Learning Model

Background Type 2 diabetes mellitus (T2DM) is a major public health burden. Self-management of diabetes including maintaining a healthy lifestyle is essential for glycemic control and to prevent diabetes complications. Mobile-based health data can play an important role in the forecasting of blood glucose levels for lifestyle management and control of T2DM. Objective The objective of this work was to dynamically forecast daily glucose levels in patients with T2DM based on their daily mobile health lifestyle data including diet, physical activity, weight, and glucose level from the day before. Methods We used data from 10 T2DM patients who were overweight or obese in a behavioral lifestyle intervention using mobile tools for daily monitoring of diet, physical activity, weight, and blood glucose over 6 months. We developed a deep learning model based on long short-term memory–based recurrent neural networks to forecast the next-day glucose levels in individual patients. The neural network used several layers of computational nodes to model how mobile health data (food intake including consumed calories, fat, and carbohydrates; exercise; and weight) were progressing from one day to another from noisy data. Results The model was validated based on a data set of 10 patients who had been monitored daily for over 6 months. The proposed deep learning model demonstrated considerable accuracy in predicting the next day glucose level based on Clark Error Grid and ±10% range of the actual values. Conclusions Using machine learning methodologies may leverage mobile health lifestyle data to develop effective individualized prediction plans for T2DM management. However, predicting future glucose levels is challenging as glucose level is determined by multiple factors. Future study with more rigorous study design is warranted to better predict future glucose levels for T2DM management.

scholarly journals EFFECT OF AEROBIC EXERCISES AND RESISTANCE TRAINING ON GLUCOSE LEVELS IN TYPE 2 DIABETIC SUBJECTS.

2017 ◽  
Vol 10 (7) ◽  
pp. 360
Author(s):  
Aishwarya Pramod Benkar ◽  
Smita Bhimrao Kanase
Keyword(s):  
Diabetes Mellitus ◽  
Blood Glucose ◽  
Resistance Training ◽  
Blood Glucose Level ◽  
Glucose Level ◽  
Postprandial Blood Glucose ◽  
Glucose Levels ◽  
Blood Glucose Levels ◽  
Significant Difference

Objective: Diabetes mellitus is a leading cause of death and disability in the world and its prevalence is predicted to rise to 10% by 2030. Hence, this study is conducted with objectives to find out the effect of aerobic exercises and resisted exercises on blood glucose levels in type 2 diabetes mellitus (T2DM) subjects and to compare the effect of both exercises on blood glucose level.Method: The comparative study was conducted at Krishna Institute of Medical Sciences Deemed University, Physiotherapy department, Karad. 30 participants with age group between 30 and 65 years were taken. Subjects were selected as per inclusion and exclusion criteria. Group A (15) participants were given aerobic exercise on static bicycle, and Group B (15) participants were given resistance training using dumbbells and weight cuffs for 5 days/week for 4 weeks. Diet recommendations were given to every participant.Results: Statistical analysis was performed using paired and unpaired t-test. Analysis showed statistically extremely significant difference in fasting blood glucose level and postprandial blood glucose level in both the groups (p≤0.0001).Conclusion: Thus, this study concludes that both aerobic exercises and resistance training prove to be beneficial in controlling blood glucose levels in T2DM subjects.

scholarly journals A Deep Learning Model for Screening Type 2 Diabetes from Retinal Photographs

2021 ◽  
Author(s):  
Jae-Seung Yun ◽  
Jaesik Kim ◽  
Sang-Hyuk Jung ◽  
Seon-Ah Cha ◽  
Seung-Hyun Ko ◽  
...  
Keyword(s):  
Type 2 Diabetes ◽  
Deep Learning ◽  
Learning Algorithm ◽  
Learning Model ◽  
Retinal Images ◽  
Uk Biobank ◽  
Non Invasive ◽  
Deep Learning Algorithm ◽  
Deep Learning Model

Objective: We aimed to develop and evaluate a non-invasive deep learning algorithm for screening type 2 diabetes in UK Biobank participants using retinal images. Research Design and Methods: The deep learning model for prediction of type 2 diabetes was trained on retinal images from 50,077 UK Biobank participants and tested on 12,185 participants. We evaluated its performance in terms of predicting traditional risk factors (TRFs) and genetic risk for diabetes. Next, we compared the performance of three models in predicting type 2 diabetes using 1) an image-only deep learning algorithm, 2) TRFs, 3) the combination of the algorithm and TRFs. Assessing net reclassification improvement (NRI) allowed quantification of the improvement afforded by adding the algorithm to the TRF model. Results: When predicting TRFs with the deep learning algorithm, the areas under the curve (AUCs) obtained with the validation set for age, sex, and HbA1c status were 0.931 (0.928-0.934), 0.933 (0.929-0.936), and 0.734 (0.715-0.752), respectively. When predicting type 2 diabetes, the AUC of the composite logistic model using non-invasive TRFs was 0.810 (0.790-0.830), and that for the deep learning model using only fundus images was 0.731 (0.707-0.756). Upon addition of TRFs to the deep learning algorithm, discriminative performance was improved to 0.844 (0.826-0.861). The addition of the algorithm to the TRFs model improved risk stratification with an overall NRI of 50.8%. Conclusions: Our results demonstrate that this deep learning algorithm can be a useful tool for stratifying individuals at high risk of type 2 diabetes in the general population.

scholarly journals Deep transfer learning and data augmentation improve glucose levels prediction in type 2 diabetes patients

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Yixiang Deng ◽  
Lu Lu ◽  
Laura Aponte ◽  
Angeliki M. Angelidi ◽  
Vera Novak ◽  
...  
Keyword(s):  
Type 2 Diabetes ◽  
Deep Learning ◽  
Blood Glucose ◽  
Transfer Learning ◽  
Data Augmentation ◽  
Generative Models ◽  
Patient Specific ◽  
Learning Methods ◽  
Glucose Levels

AbstractAccurate prediction of blood glucose variations in type 2 diabetes (T2D) will facilitate better glycemic control and decrease the occurrence of hypoglycemic episodes as well as the morbidity and mortality associated with T2D, hence increasing the quality of life of patients. Owing to the complexity of the blood glucose dynamics, it is difficult to design accurate predictive models in every circumstance, i.e., hypo/normo/hyperglycemic events. We developed deep-learning methods to predict patient-specific blood glucose during various time horizons in the immediate future using patient-specific every 30-min long glucose measurements by the continuous glucose monitoring (CGM) to predict future glucose levels in 5 min to 1 h. In general, the major challenges to address are (1) the dataset of each patient is often too small to train a patient-specific deep-learning model, and (2) the dataset is usually highly imbalanced given that hypo- and hyperglycemic episodes are usually much less common than normoglycemia. We tackle these two challenges using transfer learning and data augmentation, respectively. We systematically examined three neural network architectures, different loss functions, four transfer-learning strategies, and four data augmentation techniques, including mixup and generative models. Taken together, utilizing these methodologies we achieved over 95% prediction accuracy and 90% sensitivity for a time period within the clinically useful 1 h prediction horizon that would allow a patient to react and correct either hypoglycemia and/or hyperglycemia. We have also demonstrated that the same network architecture and transfer-learning methods perform well for the type 1 diabetes OhioT1DM public dataset.

scholarly journals Blood sugar level follows perceived time rather than actual time in people with type 2 diabetes

2016 ◽  
Vol 113 (29) ◽  
pp. 8168-8170 ◽  
Author(s):  
Chanmo Park ◽  
Francesco Pagnini ◽  
Andrew Reece ◽  
Deborah Phillips ◽  
Ellen Langer
Keyword(s):  
Type 2 Diabetes ◽  
Blood Glucose ◽  
Blood Glucose Level ◽  
Glucose Level ◽  
The Body ◽  
Glucose Levels ◽  
Actual Time ◽  
Blood Glucose Levels ◽  
Time Changes

The current study investigates whether perceived time has an effect on blood glucose level in people with type 2 diabetes. The hypothesis is that perceived time will have a greater influence over blood glucose level than actual time. Changes in blood glucose levels were measured in 46 participants with diabetes while they completed simple tasks during a 90-min period. Participants’ perception of time was manipulated by having them refer to clocks that were either accurate or altered to run fast or slow. Blood glucose levels changed in accordance with how much time they believed had passed instead of how much time had actually passed. These results are an example of the influence psychological processes can directly exert on the body.

scholarly journals Indeks Glikemik Bahan Makanan Dengan Kadar Glukosa Darah Pada Penderita Diabetes Mellitus Tipe 2 Di Puskesmas Ranotana Weru

Jurnal GIZIDO ◽  
2019 ◽  
Vol 11 (01) ◽  
pp. 36-41
Author(s):  
Nita R. Momongan ◽  
Phembriah S. Kereh ◽  
Sasaw Sriwartini
Keyword(s):  
Diabetes Mellitus ◽  
Blood Glucose ◽  
Blood Glucose Level ◽  
Health Center ◽  
Glucose Level ◽  
Glycemic Index ◽  
Data Retrieval ◽  
Glucose Levels ◽  
Blood Glucose Levels

The food glycemic index is a scale or number of foodstuffs that if consumed can have an impact on blood glucose levels so it can be used as a way to control blood glucose levels. The purpose of this study to determine the relationship of glycemic index of food with blood glucose level at diabetes mellitus type 2 in working area of Ranotana Weru Health Center. This research is an observational research using cross sectional research design with sample consist of 34 respondents who fill up the criteria of inclusion and exclusion. Data retrieval is done through interviews using food frequency questionnaire form (FFQ) and blood glucose levels obtained from the examination using autocek. Univariate analysis is done by frequency distribution and bivariate analysis using Fisher’s Exact Test. The results showed that of 34 respondents most of the respondents have blood glucose levels when the uncontrolled ≥180 mg/dl and consumed a high food glycemic indexs of 28 respondents (82,3%). While respondents have controled blood glucose levels of <180 mg/dl and consumed a low food glycemic index is 5 respondents (14,7%). Average blood glucose levels of respondents is 237,74% mg/dl. The statistical test is done obtained that there was correlation of food glycemic index with blood glucose level with value p = 0,000 (p <0,05). Conclusion, there is correlation of food glycemic index with blood glucose level in type 2 diabetes mellitus in working area of Ranotana Weru Health Center.

309-OR: Deep Learning Model of Pancreatic Islet Epigenome Refines Association Signals at Type 2 Diabetes Susceptibility Loci

Diabetes ◽  
2019 ◽  
Vol 68 (Supplement 1) ◽  
pp. 309-OR
Author(s):  
AGATA WESOLOWSKA-ANDERSEN ◽  
MATTHIAS THURNER ◽  
ANUBHA MAHAJAN ◽  
FERNANDO ABAITUA ◽  
JASON TORRES ◽  
...  
Keyword(s):  
Type 2 Diabetes ◽  
Deep Learning ◽  
Pancreatic Islet ◽  
Learning Model ◽  
Susceptibility Loci ◽  
Diabetes Susceptibility ◽  
Deep Learning Model

scholarly journals KADAR GLUKOSA DARAH SEWAKTU PADA PASIEN DIABETES MELITUS TIPE 2 DI PUSKESMAS BAHU KOTA MANADO

2015 ◽  
Vol 3 (1) ◽  
Author(s):  
Suci M. J. Amir ◽  
Herlina Wungouw ◽  
Damajanty Pangemanan
Keyword(s):  
Blood Glucose ◽  
Blood Glucose Level ◽  
Glucose Level ◽  
Community Health Center ◽  
World Health Organisation ◽  
World Health ◽  
High Blood Glucose ◽  
Glucose Levels ◽  
Blood Glucose Levels

Abstract: World Health Organisation (WHO) predicts that the number of people with diabetes in Indonesia will increase from 8.4 million in 2000 to 21.3 million in 2030. Riskesdas in 2013 showed that North Sulawesi was one of the provinces with the highest prevalence of diabetes in Indonesia. Therefore, it is necessary to check blood glucose levels regularly for screening and diagnosis of diabetes mellitus. This study aimed to determine blood glucose levels in patients with type 2 diabetes mellitus (T2DM) in Community Health Center Bahu Manado. This study was a descriptive cross sectional study design. Respondents were 22 T2DM patients that had signed the informed consent. The results showed that of the 22 respondents, 11 (50%) had high blood glucose level with an average of 267.8 mg/dL, 4 (18.2%) had moderate high blood glucose level with an average of 153.2 mg/dL, and 7 (31.8%) had normal blood glucose level with an average of 123 mg/dL. Conclusion: Most of T2DM patients in Community Health Center Bahu Manado showed high blood glucose levels with poor blood glucose control.Keywords: type 2 diabetes, blood glucose levelAbstrak: World Health Organisation (WHO) memprediksi kenaikan jumlah penyandang diabetes melitus tipe 2 (DMT2) di Indonesia dari 8,4 juta pada tahun 2000 menjadi 21,3 juta pada tahun 2030. Laporan Riskesdas tahun 2013 menunjukkan bahwa Sulawesi Utara merupakan salah satu provinsi dengan angka prevalensi DMT2 yang tertinggi di Indonesia. Oleh karena itu diperlukan pemeriksaan kadar glukosa darah secara berkala untuk skrining dan diagnosis DMT2, salah satunya pemeriksaan glukosa darah sewaktu. Penelitian ini bertujuan untuk mengetahui kadar glukosa darah sewaktu pada pasien DMT2 di Puskesmas Bahu Kota Manado. Penelitian ini bersifat deskriptif dengan rancangan potong lintang. Didapatkan 22 pasien DMT2 yang bersedia menjadi responden serta menandatangani informed consent. Hasil penelitian menunjukkan bahwa dari 22 responden, 11 (50%) memiliki rerata kadar glukosa darah yang buruk yaitu 267,8 mg/dL, 4 (18,2%) memiliki kadar glukosa darah yang sedang dengan rerata 153,2 mg/dL, dan 7 (31,8%) memiliki kadar glukosa darah yang baik dengan rerata 123 mg/dL. Simpulan: Pasien DMT2 di Puskesmas Bahu Kota Manado menunjukkan sebagian besar memiliki rerata kadar glukosa darah sewaktu yang tinggi dengan kendali glukosa darah yang buruk.Kata kunci: DMT2, glukosa darah sewaktu

Sign in / Sign up

Export Citation Format

Share Document