Continuous Glucose Measurement Research Articles

An independent central point OPTICS clustering algorithm for semi-supervised outlier detection of continuous glucose measurements

Continuous glucose monitoring (CGM) collects a host of time-series sensor data and is committed to fully automated systems for glucose control as an essential part. Outlier data in CGM measurements caused by faults may seriously affect the computation of insulin infusion rates and endanger the safety of patients. In this paper, a semi-supervised outlier detection method is proposed for anomaly detection of glucose concentration measurements based on a density-based clustering algorithm, named independent central point OPTICS (ICP-OPTICS). An optimization function is designed to improve the clustering performance and solve the problem of outlier detection with the distance measurement and information entropy of weighted time series. The proposed method in the application can be configured automatically with no prior knowledge except only some clean samples. The UVa/Padova simulator and real dataset are used to evaluate the performance of the method. Compared with the present work, the statistical results show that the method is much better in effectiveness and superiority. Furthermore, the proposed method also provides a possible reference significance for outlier detection in other practice applications.

Incorporating Glucose Variability into Glucose Forecasting Accuracy Assessment Using the New Glucose Variability Impact Index and the Prediction Consistency Index: An LSTM Case Example.

In this work, we developed glucose forecasting algorithms trained and evaluated on a large dataset of free-living people with type 1 diabetes (T1D) using closed-loop (CL) and sensor-augmented pump (SAP) therapies; and we demonstrate how glucose variability impacts accuracy. We introduce the glucose variability impact index (GVII) and the glucose prediction consistency index (GPCI) to assess the accuracy of prediction algorithms. A long-short-term-memory (LSTM) neural network was designed to predict glucose up to 60 minutes in the future using continuous glucose measurements and insulin data collected from 175 people with T1D (41,318 days) and evaluated on 75 people (11,333 days) from the Tidepool Big Data Donation Dataset. LSTM was compared with two naïve forecasting algorithms as well as Ridge linear regression and a random forest using root-mean-square error (RMSE). Parkes error grid quantified clinical accuracy. Regression analysis was used to derive the GVII and GPCI. The LSTM had highest accuracy and best GVII and GPCI. RMSE for CL was 19.8 ± 3.2 and 33.2 ± 5.4 mg/dL for 30- and 60-minute prediction horizons, respectively. RMSE for SAP was 19.6 ± 3.8 and 33.1 ± 7.3 mg/dL for 30- and 60-minute prediction horizons, respectively; 99.6% and 97.6% of predictions were within zones A+B of the Parkes error grid at 30- and 60-minute prediction horizons, respectively. Glucose variability was strongly correlated with RMSE (R≥0.64, P < 0.001); GVII and GPCI demonstrated a means to compare algorithms across datasets with different glucose variability. The LSTM model was accurate on a large real-world free-living dataset. Glucose variability should be considered when assessing prediction accuracy using indices such as GVII and GPCI.

Telemedizin in der Kinderdiabetologie

Der aktuelle Stand telemedizinischer Anwendungen und deren zu erwartendes Potenzial werden aus der Perspektive der pädiatrischen Diabetologie beleuchtet. Kontinuierliche Glukosemesssysteme (CGM [„continuous glucose measurement“]) und Insulinpumpen lassen sich zunehmend auch über interoperable oder herstellerbasierte Datenplattformen auslesen und teilen und ermöglichen es den Diabetesteams, eine Beratung auf Basis dieser Daten auch ohne physischen Kontakt in die kontinuierliche Versorgung einzubauen. Videosprechstunden können dabei zunächst ein ergänzendes Angebot darstellen, unterliegen oft längeren Implementierungsprozessen und sind u. a. abhängig von datenschutzkonformer Videochatsoftware, koordinierten Arbeitsabläufen sowie technischem Support. Telemedizin kann derzeit schon bestehende Beratungsstrukturen in Kliniken und Praxen ergänzen und helfen, das Potenzial digital dokumentierter Daten aus CGM- und automatisierten Insulinabgabesystemen voll auszuschöpfen, wobei telemedizinische Anwendungen bisher kein direktes Einsparungspotenzial bezüglich Zeit und Aufwand des medizinischen Personals und damit der Personalkosten bieten.

Mental and somatic health in university students with type 1 diabetes: new results from DiaSHoT18, a cross sectional national health and well-being survey.

To explore mental and somatic health, quality of life, alcohol-related problems, sleep problems, and diabetes related distress in university students with type 1 diabetes (T1D), compared to students without T1D. Further, we evaluated associations with gender, treatment modalities, and achieved metabolic control. All fulltime Norwegian students aged 18-35 years pursuing higher education in 2018 (n=162.512) were invited into a comprehensive national survey on health and well-being. Students that stated having diabetes was asked further questions about their diabetes care. Of 49,684 participating students, 324 participants stated having T1D. Students with T1D did not show more mental or somatic health symptoms, or report a higher level of loneliness. However, T1D was significantly associated with lower quality of life (QoL). Students with good metabolic control reached the same QoL as students without T1D. Mental disorders and suicidality were associated with lacking metabolic control. The proportion of unhealthy drinking habits was generally low, and even lower in students with T1D. Sleeping patterns were generally good, but students using continuous glucose measurement were awakening more often during sleep. Females with T1D showed higher levels of diabetes related problems and distress, but good metabolic control was associated with lower diabetes distress level. Students with T1D scored equally on most mental and somatic health scales. Their quality of life was significantly worse compared to their healthy peers. Knowledge on the impact of metabolic control, gender and the use of CSII and CGM can be valuable for the caretakers of young adults with T1D.

A protein sandwich enables real-time in vivo biomarker measurement

Biosensors allow rapid, accurate, continuous measurement of glucose and insulin in live rats.

Postprandial Asymptomatic Glycemic Fluctuations after Gastrectomy for Gastric Cancer Using Continuous Glucose Monitoring Device.

PurposeAlthough dumping symptoms are thought to involve postprandial glycemic changes, postprandial glycemic variability without dumping symptoms remains poorly understood due to the lack of a method that allows the easy and continuous measurement of blood glucose levels.Materials and MethodsPatients having undergone distal gastrectomy with Billroth-I (DG-BI) or Roux-en-Y reconstruction (DG-RY), total gastrectomy with RY (TG-RY) and pylorus preserving gastrectomy (PPG) for gastric cancer 3 months to 3 years prior, diagnosed as pathological stage I or II, were prospectively enrolled from March 2018 to January 2020. The interstitial tissue glycemic levels were measured every 15 min, up to 14 days by continuous glucose monitoring. Moreover, using a diary recording the diet and symptoms, asymptomatic glucose profiles without sugar supplementation within 3 h postprandially were compared among the four procedures.ResultsA total of 40 patients were enrolled, 10 patients for each of the four procedures. There were 47 glucose profiles with DG-BI, 46 profiles with DG-RY, 38 profiles with TG-RY, and 46 profiles with PPG. PPG showed the slowest increase with a subsequent gradual decrease in glucose fluctuations, without hyperglycemia or hypoglycemia, among the four procedures. In contrast, TG-RY and DG-RY showed spike-like glycemic variability, sharp rises during meals, and rapid drops. The glucose profiles of DG-BI were milder than those of RY.ConclusionsThe asymptomatic glycemic changes after meals differ among the types of surgical procedures for gastric cancer. Given the mild glycemic fluctuations in PPG and the glucose spikes in TG-RY and DG-RY, pylorus preservation and physiological reconstruction without changes in food pathways may optimize postprandial glucose profiles after gastrectomy.

Errors-in-Variables Modeling of Personalized Treatment-Response Trajectories.

Estimating the impact of a treatment on a given response is needed in many biomedical applications. However, methodology is lacking for the case when the response is a continuous temporal curve, treatment covariates suffer extensively from measurement error, and even the exact timing of the treatments is unknown. We introduce a novel method for this challenging scenario. We model personalized treatment-response curves as a combination of parametric response functions, hierarchically sharing information across individuals, and a sparse Gaussian process for the baseline trend. Importantly, our model accounts for errors not only in treatment covariates, but also in treatment timings, a problem arising in practice for example when data on treatments are based on user self-reporting. We validate our model with simulated and real patient data, and show that in a challenging application of estimating the impact of diet on continuous blood glucose measurements, accounting for measurement error significantly improves estimation and prediction accuracy.

Impact of Meal Timing on Postprandial Interstitial Fluid Glucose Levels in Young Japanese Females.

Flash glucose monitoring (FGM) provides continuous and accessible measurement of the interstitial fluid glucose (ISFG) level and this system is useful for understanding blood glucose fluctuations. We examined differences in postprandial ISFG after the main mealtimes (breakfast, lunch, and dinner) in healthy young Japanese females. Nine healthy young females (aged 21.5±0.6 y old) were enrolled in this study. ISFG was continuously measured by FGM. Participants ate the same meal three times a day consecutively, thereby satisfying their daily energy requirements. Postprandial ISFG fluctuations were evaluated for 4 h after each meal. There were no significant differences in ISFG before the 3 main meals. The postprandial ISFG peak was the lowest after breakfast, increasing in the order of lunch and then dinner. The area under the curve of the 4-h postprandial ISFG was higher after lunch and dinner than after breakfast. The results of this study suggest that postprandial ISFG differ depending on mealtimes in young Japanese females.

Roux-en-Y Gastric Bypass Increases Glycemic Variability and Time in Hypoglycemia in Patients With Obesity and Prediabetes or Type 2 Diabetes: A Prospective Cohort Study.

Roux-en-Y gastric bypass (RYGB) is an established treatment for type 2 diabetes and obesity. The study objective was to establish RYGB's effects on glycemic variability (GV) and hypoglycemia. This was a prospective observational study of 10 participants with obesity and prediabetes or type 2 diabetes who underwent RYGB. Patients were studied before RYGB (Pre) and 1 month, 1 year, and 2 years postsurgery with continuous glucose measurement (CGM). A mixed-meal test (MMT) was conducted at Pre, 1 month, and 1 year. After RYGB, mean CGM decreased (at 1 month, 1 year, and 2 years), and GV increased (at 1 year and 2 years). Five of the 10 participants had a percent time in range (%TIR) <3.0 mmol/L (54 mg/dL) greater than the international consensus target of 1% at 1 or 2 years. Peak glucagon-like peptide-1 (GLP-1) and glucagon area under the curve during MMT were positively and negatively associated, respectively, with contemporaneous %TIR <3.0 mmol/L. Patients undergoing RYGB are at risk for development of postbariatric hypoglycemia due to a combination of reduced mean glucose, increased GV, and increased GLP-1 response.

Enabling equality of employment with insulin treated diabetes in safety‐critical occupations

Enabling equality of employment with insulin treated diabetes in safety‐critical occupations

Fault detection of continuous glucose measurements based on modified k-medoids clustering algorithm

As continuous glucose monitoring (CGM) systems provide critical feedback information of blood glucose concentration to the artificial pancreas for patients with type 1 diabetes (T1D), faults in CGM may seriously affect the computation of insulin infusion rates which can lead to fatal consequences accompany with hypoglycemia or hyperglycemia. In the present work, the k-medoids clustering algorithm is modified by calculating cluster number with a Bayesian Information Criterion (BIC)-based cost function and the SAC (SSE-ASW Criterion) evaluation coefficient which considers both SSE (Sum of Square due to Error) and ASW (Average Silhouette Width) criteria. Then, the modified k-medoids clustering algorithm is proposed to detect sensor failures online with CGM measurements. Different from the qualitative model-based methods and quantitative model-based methods, sufficient clean data are the only requirement of the proposed method. During online monitoring, the new glycemic variability is then tracked against predefined confidence limits during training period to indicate abnormality. The feasibility of the proposed method is successfully assessed using CGM data collected from the UVa/Padova metabolic simulator.

Abbott free style libre for continuous tissue glucose measurement in cynomolgus monkeys and assessment of the effect of training on glucose and cardiovascular parameters

Abbott free style libre for continuous tissue glucose measurement in cynomolgus monkeys and assessment of the effect of training on glucose and cardiovascular parameters

IGLU 2.0: A New Wearable for Accurate Non-Invasive Continuous Serum Glucose Measurement in IoMT Framework

To best of the authors knowledge, this article presents the first-ever non-invasive glucometer that takes into account serum glucose for high accuracy. In case of blood glucose measurement, serum glucose value has always been considered precise blood glucose value during prandial modes. Serum glucose can be measured in laboratory and more stable glucose level compare to capillary glucose. However, this invasive approach is not convenient for frequent measurement. Sometimes, Conventional invasive blood glucose measurement may be responsible for cause of trauma and chance of blood related infections. To overcome this issue, in the current paper, we propose a novel Internet-of-Medical (IoMT) enabled glucometer for non-invasive precise serum glucose measurement. In this work, a near-infrared (NIR) spectroscopic technique has been used for glucose measurement. The novel device called iGLU 2.0 is based on optical detection and precise machine learning (ML) regression models. The optimal multiple polynomial regression and deep neural network models have been presented to analyze the precise measurement. The glucose values of serum are saved on cloud through open IoT platform for endocrinologist at remote location. To validate iGLU 2.0, Mean Absolute Relative Difference (mARD) and Average Error (AvgE) are obtained 6.07% and 6.09%, respectively from predicted blood glucose values for capillary glucose. For serum glucose, mARD and AvgE are found 4.86% and 4.88%, respectively. These results represent that the proposed non-invasive glucose measurement device is more precise for serum glucose compared to capillary glucose.

Nutrition update in gastric cancer surgery.

Patients with gastric cancer are often malnourished during tumor progression. Malnutrition is a risk factor for postoperative complications and a poor prognosis. Early evaluation and management of nutrition can improve these outcomes. Various combined indices in which albumin is the primary component are used to evaluate the nutritional status, including the Prognostic Nutritional Index, Glasgow Prognostic Score, and Controlling Nutritional Status score. Both the American Society for Parenteral and Enteral Nutrition and the European Society for Clinical Nutrition and Metabolism guidelines recommend immediate and early oral/enteral nutrition. However, few reports have described the additional effects of preoperative immunonutrition on clinical outcomes of gastric cancer surgery. Gastrectomy types and reconstruction methods that consider the postoperative nutritional status have been used when oncologically acceptable. Total gastrectomy has recently tended to be avoided because of its negative impact on nutritional status. New findings obtained from the emergence of continuous glucose measurement, such as glucose fluctuation and nocturnal hypoglycemia, may affect nutritional management after gastrectomy. Some prospective clinical studies on perioperative nutritional intervention have set postoperative body weight loss as a primary endpoint. It seems important to continue oral nutritional supplement, even in small doses, to reduce body weight loss after gastrectomy. Evidence generated by prospective, well‐developed randomized controlled studies must be disseminated so that nutritional therapy is widely recognized as an important multimodal therapy in patients undergoing gastric cancer surgery.

72-LB: Novel Epidermal Adhesive Sensors to Enhance Continuous Glucose Measurement in Patients with Diabetes: The EASE Study

By providing a more comfortable and convenient glucose testing experience, non-invasive, wearable biosensors could bolster patients’ engagement with their diabetes mellitus (DM). We investigated the accuracy and acceptability of a novel, needle-free continuous glucose sensor. At 2 g and 5 x 4 mm, this sensor is ultra-light, thin, flexible, and self-adherent to the skin. A mild electrical current applied to the epidermis causes sodium ions, then interstitial fluid (ISF) glucose molecules, to migrate toward an electrode. The sensor gauges glucose by detecting its electrical charge strength. Four individuals with DM participated (age 42-61 years, mean 51.5 years; 1 patient with type 1 DM). Over 4 hours, we obtained 3 sensor and 3 glucometer readings concurrently at each of the following time points: following an 8-hour fast, and every 60 minutes after a standard 500-kcal, 75-g high-carbohydrate meal. The correlation coefficient of sensor and glucometer readings was 0.80 (p = 0.00023). Participants rated on a 1-5 scale, with 5 being the highest: 4.25 for comfort, and 4.0 for being receptive to long-term use of such a sensor. Two individuals denied discomfort; 1 felt tingling; and 1 noted a needle-like sensation. The function and patient-friendly design of a wearable glucose sensor may help substantially advance glucose testing, and in turn, improve clinical outcomes and quality of life. Disclosure E. Chao: None. E. De La Paz Andres: None. A. Barfidokht: None. J. Wang: None. Funding University of California, San Diego; National Institutes of Health (UL1TR001442)

Predicting and Preventing Nocturnal Hypoglycemia in Type 1 Diabetes Using Big Data Analytics and Decision Theoretic Analysis.

Background: Despite new glucose sensing technologies, nocturnal hypoglycemia is still a problem for people with type 1 diabetes (T1D) as symptoms and sensor alarms may not be detected while sleeping. Accurately predicting nocturnal hypoglycemia before sleep may help minimize nighttime hypoglycemia. Methods: A support vector regression (SVR) model was trained to predict, before bedtime, the overnight minimum glucose and overnight nocturnal hypoglycemia for people with T1D. The algorithm was trained on continuous glucose measurements and insulin data collected from 124 people (22,804 valid nights of data) with T1D. The minimum glucose threshold for announcing nocturnal hypoglycemia risk was derived by applying a decision theoretic criterion to maximize expected net benefit. Accuracy was evaluated on a validation set from 10 people with T1D during a 4-week trial under free-living sensor-augmented insulin-pump therapy. The primary outcome measures were sensitivity and specificity of prediction, the correlation between predicted and actual minimum nocturnal glucose, and root-mean-square error. The impact of using the algorithm to prevent nocturnal hypoglycemia is shown in-silico. Results: The algorithm predicted 94.1% of nocturnal hypoglycemia events (<3.9 mmol/L, 95% confidence interval [CI], 71.3-99.9) with an area under the receiver operating characteristic curve of 0.86 (95% CI, 0.75-0.98). Correlation between actual and predicted minimum glucose was high (R = 0.71, P < 0.001). In-silico simulations showed that the algorithm could reduce nocturnal hypoglycemia by 77.0% (P = 0.006) without impacting time in target range (3.9-10 mmol/L). Conclusion: An SVR model trained on a big data set and optimized using decision theoretic criterion can accurately predict at bedtime if overnight nocturnal hypoglycemia will occur and may help reduce nocturnal hypoglycemia.

Wearable multiplexed biosensor system toward continuous monitoring of metabolites

Comprehensive metabolic panels are the most reliable and common methods for monitoring general physiology in clinical healthcare. Translation of this clinical practice to personal health and wellness tracking requires reliable, non-invasive, miniaturized, ambulatory, and inexpensive systems for continuous measurement of biochemical analytes. We report the design and characterization of a wearable system with a flexible sensor array for non-invasive and continuous monitoring of human biochemistry. The system includes signal conditioning, processing, and transmission parts for continuous measurement of glucose, lactate, pH, and temperature. The system can operate three discrete electrochemical cells. The system draws 15 mA under continuous operation when powered by a 3.7 V 150 mAh battery. The analog front-end of the electrochemical cells has four potentiostats and three multiplexers for multiplexed and parallel readout from twelve working electrodes. Utilization of redundant working electrodes improves the measurement accuracy of sensors by averaging chronoamperometric responses across the array. The operation of the system is demonstrated in vitro by simultaneous measurement of glucose and lactate, pH, and skin temperature. In benchtop measurements, the sensors are shown to have sensitivities of 26.31 μA mM−1·cm−2 for glucose, 1.49 μA mM−1·cm−2 for lactate, 54 mV·pH−1 for pH, and 0.002 °C-1 for temperature. With the custom wearable system, these values were 0.84 ± 0.03 mV μM−1·cm−2 or glucose, 31.87 ± 9.03 mV mM−1·cm−2 for lactate, 57.18 ± 1.43 mV·pH−1 for pH, and 63.4 μV·°C−1 for temperature. This miniaturized wearable system enables future evaluation of temporal changes of the sweat biomarkers.

Real-world flash glucose monitoring in Brazil: can sensors make a difference in diabetes management in developing countries?

BackgroundNew technologies are changing diabetes treatment and contributing better outcomes in developed countries. To our knowledge, no previous studies have investigated the comparative effect of sensor-based monitoring on glycemic markers in developing countries like Brazil. The present study aims to evaluate the use of intermittent Continuous Glucose Measurements (iCGM) in a developing country, Brazil, regarding (i) frequency of glucose scans, (ii) its association with glycemic markers and (iii) comparison with these findings to those observed in global population data.MethodsGlucose results were de-identified and uploaded to a dedicated database when Freestyle Libre™ readers were connected to an internet-ready computer. Data between September 2014 and Dec 2018, comprising 688,640 readers and 7,329,052 sensors worldwide, were analysed (including 17,691 readers and 147,166 sensors from Brazil). Scan rate per reader was determined and each reader was sorted into 20 equally-sized rank ordered groups, categorised by scan frequency. Glucose parameters were calculated for each group, including estimated A1c, time above, below and within range identified as 70–180 mg/dL.ResultsIn Brazil, reader users performed an average of 14 scans per day, while around the world, reader users performed an average of 12 scans per day (p < 0.01). In Brazil dataset, those in the lowest and in the highest groups scanned on average 3.6 and 43.1 times per day had an estimated A1c of 7.56% (59 mmol/mol) and 6.71% (50 mmol/mol), respectively (p < 0.01). Worldwide, the lowest group and the highest groups scanned 3.4 times/day and 37.8 times/day and had an eA1c of 8.14% (65 mmol/mol) and 6.70% (50 mmol/mol), respectively (p < 0.01). For the scan groups in both populations, the time spent above 180 mg/dL decreased as the scan frequency increased. In both Brazil and around the world, as scan frequency increased, time in range (TIR) increased. In Brazil, TIR increased from 14.15 to 16.62 h/day (p < 0.01). Worldwide, TIR increased from 12.06 to 16.97 h/day (p < 0.01).ConclusionsWe conclude that Brazilian users have a high frequency of scans, more frequent than global data. Similarly to the world findings, increased scan frequency is associated with better glycemic control.

Measuring group and individual relationship between patterns in sedentary behaviour and glucose in type 2 diabetes adults

AbstractThe aim of this study was to explore the relationship between patterns in sedentary behaviour and glucose in adults with type 2 diabetes.Thirty‐seven adults with type 2 diabetes managed with diet and/or anti‐diabetes medication (not insulin) were recruited. Participants wore an activPAL accelerometer and FreeStyle Libre flash glucose monitor for continuous measurement of activity and glucose for 24 hours/day for three to 14 days while documenting sleep, food and medication. The relationship between daily patterns of sedentary behaviour and sedentary breaks with glucose was investigated using correlation analysis. Regression analysis was used to investigate these relationships at an individual level.Participants (mean age 62.8±10.5 years, BMI 29.6±6.8kg/m2) spent 64% of their day sedentary, 44.7% of sedentary bouts were 30–60 minutes in duration and mean bout duration was 47.2 minutes. No association between mean glucose and sedentary proportion (total sedentary time) was identified. Mean glucose and glucose standard deviation were positively correlated with sedentary bout duration (both p&lt;0.05). Individual regression analysis showed increased sedentary time is associated with increased mean glucose in 25 (68%) of the participants, with a negative association being shown in 12 (32%) of the participants.In analysis of the whole group, sedentary bout duration but not sedentary proportion was associated with mean glucose and glucose variability. Individual regression analysis identified a different relationship pattern for the majority of participants. This is the first study to identify an individualised response to activity behaviour and highlights the importance of conducting individual analysis when using continuous measurement methods.

Fabrication of a biocompatible and continuous glucose biosensor with the poly(3,4-ethylenedioxythiophene) modified electrode

Abstract In this study, we have explored the potentiality of using GOx/AuNP/PEDOT(BSA)/Pt electrode as an implantable, long-lasting, and sensitive glucose biosensor. We have examined the performance of the electrode for glucose calibration with three electrochemical measurements, such as cyclic voltammetry, linear sweep voltammetry, as well as chronoamperometry, which all exhibited strong linear correlation between resulting signal and glucose concentration. In comparison, linear sweep voltammetry gave the best linear sensitivity, whose average was about 3.124 µA/mM/cm2 within a wide glucose concentration range from 0.416 to 50 mM. Meanwhile, the stability of the electrode was confirmed by a continuous glucose measurement in a 7-day period of time and the deviation was about 5%. The electrode was also biocompatibility according to the ISO109930-1 criterion. Furthermore, we have demonstrated that the oxidation of glucose on the GOx/AuNP/PEDOT(BSA)/Pt electrode was oxygen independent, therefore, it was not only easy to make but also feasible for continuous glucose monitoring systems.