Noninvasive Hypoglycemia Detection in People With Diabetes Using Smartwatch Data.
Ontology highlight
ABSTRACT: Objective
To develop a noninvasive hypoglycemia detection approach using smartwatch data.Research design and methods
We prospectively collected data from two wrist-worn wearables (Garmin vivoactive 4S, Empatica E4) and continuous glucose monitoring values in adults with diabetes on insulin treatment. Using these data, we developed a machine learning (ML) approach to detect hypoglycemia (<3.9 mmol/L) noninvasively in unseen individuals and solely based on wearable data.Results
Twenty-two individuals were included in the final analysis (age 54.5 ± 15.2 years, HbA1c 6.9 ± 0.6%, 16 males). Hypoglycemia was detected with an area under the receiver operating characteristic curve of 0.76 ± 0.07 solely based on wearable data. Feature analysis revealed that the ML model associated increased heart rate, decreased heart rate variability, and increased tonic electrodermal activity with hypoglycemia.Conclusions
Our approach may allow for noninvasive hypoglycemia detection using wearables in people with diabetes and thus complement existing methods for hypoglycemia detection and warning.
SUBMITTER: Lehmann V
PROVIDER: S-EPMC10154647 | biostudies-literature | 2023 May
REPOSITORIES: biostudies-literature
Publications
Noninvasive Hypoglycemia Detection in People With Diabetes Using Smartwatch Data.
Lehmann Vera V Föll Simon S Maritsch Martin M van Weenen Eva E Kraus Mathias M Lagger Sophie S Odermatt Katja K Albrecht Caroline C Fleisch Elgar E Zueger Thomas T Wortmann Felix F Stettler Christoph C
Diabetes care 20230501 5
<h4>Objective</h4>To develop a noninvasive hypoglycemia detection approach using smartwatch data.<h4>Research design and methods</h4>We prospectively collected data from two wrist-worn wearables (Garmin vivoactive 4S, Empatica E4) and continuous glucose monitoring values in adults with diabetes on insulin treatment. Using these data, we developed a machine learning (ML) approach to detect hypoglycemia (<3.9 mmol/L) noninvasively in unseen individuals and solely based on wearable data.<h4>Results ...[more]
Similar Datasets
Project description:BackgroundMonitoring glucose excursions is important in diabetes management. This can be achieved using continuous glucose monitors (CGMs). However, CGMs are expensive and invasive. Thus, alternative low-cost noninvasive wearable sensors capable of predicting glycemic excursions could be a game changer to manage diabetes.MethodsIn this article, we explore two noninvasive sensor modalities, electrocardiograms (ECGs) and accelerometers, collected on five healthy participants over two weeks, to predict both hypoglycemic and hyperglycemic excursions. We extract 29 features encompassing heart rate variability features from the ECG, and time- and frequency-domain features from the accelerometer. We evaluated two machine learning approaches to predict glycemic excursions: a classification model and a regression model.ResultsThe best model for both hypoglycemia and hyperglycemia detection was the regression model based on ECG and accelerometer data, yielding 76% sensitivity and specificity for hypoglycemia and 79% sensitivity and specificity for hyperglycemia. This had an improvement of 5% in sensitivity and specificity for both hypoglycemia and hyperglycemia when compared with using ECG data alone.ConclusionsElectrocardiogram is a promising alternative not only to detect hypoglycemia but also to predict hyperglycemia. Supplementing ECG data with contextual information from accelerometer data can improve glucose prediction.
| S-EPMC10973850 | biostudies-literature
Project description:BackgroundNocturnal hypoglycemia poses significant risks to individuals with insulin-treated diabetes, impacting health and quality of life. Although continuous glucose monitoring (CGM) systems reduce these risks, their poor accuracy at low glucose levels, high cost, and availability limit their use. This study examined physiological biomarkers associated with nocturnal hypoglycemia and evaluated the use of machine learning (ML) to detect hypoglycemia during nighttime sleep using data from consumer-grade smartwatches.MethodsThis study analyzed 351 nights of 36 adults with insulin-treated diabetes. Participants wore two smartwatches alongside CGM systems. Linear mixed-effects models compared sleep and vital signs between nights with and without hypoglycemia during early and late sleep. A ML model was trained to detect hypoglycemia solely using smartwatch data.ResultsSixty-six nights with spontaneous hypoglycemia were recorded. Hypoglycemic nights showed increased wake periods, heart rate, stress levels, and activity during early sleep, with weaker effects during late sleep. In nights when hypoglycemia occurred during early sleep, the ML model performed comparable or better than prior studies with an area under the receiver operator curve of 0.78 for level 1 and 0.83 for level 2 hypoglycemia, with sensitivity of 0.78 and 0.89, specificity of 0.64 for both, negative predictive value of 0.94 and 0.99, and positive predictive value of 0.25 and 0.13 for level 1 and level 2 hypoglycemia, respectively.ConclusionsConsumer-grade smartwatches demonstrate promise for detecting nocturnal hypoglycemia, particularly during early sleep. Refining models to reduce false alarms could enhance their clinical utility as low-cost, accessible tools to complement CGM.
| S-EPMC11851596 | biostudies-literature
Project description:Consumer wearable devices that continuously measure vital signs have been used to monitor the onset of infectious disease. Here, we show that data from consumer smartwatches can be used for the pre-symptomatic detection of coronavirus disease 2019 (COVID-19). We analysed physiological and activity data from 32 individuals infected with COVID-19, identified from a cohort of nearly 5,300 participants, and found that 26 of them (81%) had alterations in their heart rate, number of daily steps or time asleep. Of the 25 cases of COVID-19 with detected physiological alterations for which we had symptom information, 22 were detected before (or at) symptom onset, with four cases detected at least nine days earlier. Using retrospective smartwatch data, we show that 63% of the COVID-19 cases could have been detected before symptom onset in real time via a two-tiered warning system based on the occurrence of extreme elevations in resting heart rate relative to the individual baseline. Our findings suggest that activity tracking and health monitoring via consumer wearable devices may be used for the large-scale, real-time detection of respiratory infections, often pre-symptomatically.
| S-EPMC9020268 | biostudies-literature
Project description:Previous literature has demonstrated that hypoglycemic events in patients with type 1 diabetes (T1D) are associated with measurable scalp electroencephalography (EEG) changes in power spectral density. In the present study, we used a dataset of 19-channel scalp EEG recordings in 34 patients with T1D who underwent a hyperinsulinemic-hypoglycemic clamp study. We found that hypoglycemic events are also characterized by EEG complexity changes that are quantifiable at the single-channel level through empirical conditional and permutation entropy and fractal dimension indices, i.e., the Higuchi index, residuals, and tortuosity. Moreover, we demonstrated that the EEG complexity indices computed in parallel in more than one channel can be used as the input for a neural network aimed at identifying hypoglycemia and euglycemia. The accuracy was about 90%, suggesting that nonlinear indices applied to EEG signals might be useful in revealing hypoglycemic events from EEG recordings in patients with T1D.
| S-EPMC7516516 | biostudies-literature
Project description:Background and objectivesAmong people with diabetes mellitus, CKD may promote hypoglycemia through altered clearance of glucose-lowering medications, decreased kidney gluconeogenesis, and blunted counter-regulatory response. We conducted a prospective observational study of hypoglycemia among 105 individuals with type 2 diabetes treated with insulin or a sulfonylurea using continuous glucose monitors.Design, setting, participants & measurementsWe enrolled 81 participants with CKD, defined as eGFR<60 ml/min per 1.73 m2, and 24 control participants with eGFR≥60 ml/min per 1.73 m2 frequency-matched on age, duration of diabetes, hemoglobin A1c, and glucose-lowering medications. Each participant wore a continuous glucose monitor for two 6-day periods. We examined rates of sustained level 1 hypoglycemia (<70 mg/dl) and level 2 hypoglycemia (<54 mg/dl) among participants with CKD. We then tested differences compared with control participants as well as a second control population (n=73) using Poisson and linear regression, adjusting for age, sex, and race.ResultsOver 890 total days of continuous glucose monitoring, participants with CKD were observed to have 255 episodes of level 1 hypoglycemia, of which 68 episodes reached level 2 hypoglycemia. Median rate of hypoglycemic episodes was 5.3 (interquartile range, 0.0-11.7) per 30 days and mean time spent in hypoglycemia was 28 (SD 37) minutes per day. Hemoglobin A1c and the glucose management indicator were the main clinical correlates of time in hypoglycemia (adjusted differences 6 [95% confidence interval, 2 to 10] and 13 [95% confidence interval, 7 to 20] fewer minutes per day per 1% higher hemoglobin A1c or glucose management indicator, respectively). Compared with control populations, participants with CKD were not observed to have significant differences in time in hypoglycemia (adjusted differences 4 [95% confidence interval, -12 to 20] and -12 [95% confidence interval, -29 to 5] minutes per day).ConclusionsAmong people with type 2 diabetes and moderate to severe CKD, hypoglycemia was common, particularly with tighter glycemic control, but not significantly different from groups with similar clinical characteristics and preserved eGFR.
| S-EPMC6556736 | biostudies-literature
Project description:Iatrogenic hypoglycemia activates the immune system and is associated with an increased risk for atherosclerotic disease. We determined acute and long-term effects of insulin-induced hypoglycemia on inflammatory markers in humans with or without type 2 diabetes. A total of 15 adults with type 2 diabetes and 16 matched control subjects (17 men and 14 women, age 59.6 ± 7.1 years, BMI 28.5 ± 4.3 kg/m2) underwent a hyperinsulinemic-euglycemic (5.31 ± 0.32 mmol/L) hypoglycemic (2.80 ± 0.12 mmol/L) glucose clamp. Blood was drawn during euglycemia and hypoglycemia and 1, 3, and 7 days later to determine circulating immune cell composition, function, and inflammatory proteins. In response to hypoglycemia, absolute numbers of circulating lymphocytes and monocytes significantly increased and remained elevated for 1 week. The proportion of CD16+ monocytes increased, and the proportion of CD14+ monocytes decreased, which was sustained for 1 week in people without diabetes. During hypoglycemia, ex vivo stimulated monocytes released more tumor necrosis factor-α and interleukin 1β, and less interleukin 10, particularly in people with diabetes. hs-CRP and 25 circulating inflammatory proteins increased, remaining significantly elevated 1 week after hypoglycemia. While levels at euglycemia differed, responses to hypoglycemia were broadly similar in people with or without type 2 diabetes. We conclude that hypoglycemia induces a proinflammatory response at the cellular and protein level that is sustained for 1 week in people with type 2 diabetes and control subjects.
| S-EPMC9750956 | biostudies-literature
Project description:BackgroundAlthough electrocardiography is the gold standard for heart rate (HR) recording in clinical trials, the increasing availability of smartwatch-based HR monitors opens up possibilities for drug development studies. Smartwatches allow for inexpensive, unobtrusive, and continuous HR estimation for potential detection of treatment effects outside the clinic, during daily life.ObjectiveThe aim of this study is to evaluate the repeatability and sensitivity of smartwatch-based HR estimates collected during a randomized clinical trial.MethodsThe data were collected as part of a multiple-dose, investigator-blinded, randomized, placebo-controlled, parallel-group study of 12 patients with Parkinson disease. After a 6-day baseline period, 4 and 8 patients were treated for 7 days with an ascending dose of placebo and clenbuterol, respectively. Throughout the study, the smartwatch provided HR and sleep state estimates. The HR estimates were quantified as the 2.5th, 50th, and 97.5th percentiles within awake and asleep segments. Linear mixed models were used to calculate the following: (1) the intraclass correlation coefficient (ICC) of estimated sleep durations, (2) the ICC and minimum detectable effect (MDE) of the HR estimates, and (3) the effect sizes of the HR estimates.ResultsSleep duration was moderately repeatable (ICC=0.64) and was not significantly affected by study day (P=.83), clenbuterol (P=.43), and study day by clenbuterol (P=.73). Clenbuterol-induced changes were detected in the asleep HR as of the first night (+3.79 beats per minute [bpm], P=.04) and in the awake HR as of the third day (+8.79 bpm, P=.001). The median HR while asleep had the highest repeatability (ICC=0.70). The MDE (N=12) was found to be smaller when patients were asleep (6.8 bpm to 11.7 bpm) than while awake (10.7 bpm to 22.1 bpm). Overall, the effect sizes for clenbuterol-induced changes were higher while asleep (0.49 to 2.75) than while awake (0.08 to 1.94).ConclusionsWe demonstrated the feasibility of using smartwatch-based HR estimates to detect clenbuterol-induced changes during clinical trials. The asleep HR estimates were most repeatable and sensitive to treatment effects. We conclude that smartwatch-based HR estimates obtained during daily living in a clinical trial can be used to detect and track treatment effects.Trial registrationNetherlands Trials Register NL8002; https://www.trialregister.nl/trial/8002.
| S-EPMC8759015 | biostudies-literature
Project description:Importance:Atrial fibrillation (AF) affects 34 million people worldwide and is a leading cause of stroke. A readily accessible means to continuously monitor for AF could prevent large numbers of strokes and death. Objective:To develop and validate a deep neural network to detect AF using smartwatch data. Design, Setting, and Participants:In this multinational cardiovascular remote cohort study coordinated at the University of California, San Francisco, smartwatches were used to obtain heart rate and step count data for algorithm development. A total of 9750 participants enrolled in the Health eHeart Study and 51 patients undergoing cardioversion at the University of California, San Francisco, were enrolled between February 2016 and March 2017. A deep neural network was trained using a method called heuristic pretraining in which the network approximated representations of the R-R interval (ie, time between heartbeats) without manual labeling of training data. Validation was performed against the reference standard 12-lead electrocardiography (ECG) in a separate cohort of patients undergoing cardioversion. A second exploratory validation was performed using smartwatch data from ambulatory individuals against the reference standard of self-reported history of persistent AF. Data were analyzed from March 2017 to September 2017. Main Outcomes and Measures:The sensitivity, specificity, and receiver operating characteristic C statistic for the algorithm to detect AF were generated based on the reference standard of 12-lead ECG-diagnosed AF. Results:Of the 9750 participants enrolled in the remote cohort, including 347 participants with AF, 6143 (63.0%) were male, and the mean (SD) age was 42 (12) years. There were more than 139 million heart rate measurements on which the deep neural network was trained. The deep neural network exhibited a C statistic of 0.97 (95% CI, 0.94-1.00; P < .001) to detect AF against the reference standard 12-lead ECG-diagnosed AF in the external validation cohort of 51 patients undergoing cardioversion; sensitivity was 98.0% and specificity was 90.2%. In an exploratory analysis relying on self-report of persistent AF in ambulatory participants, the C statistic was 0.72 (95% CI, 0.64-0.78); sensitivity was 67.7% and specificity was 67.6%. Conclusions and Relevance:This proof-of-concept study found that smartwatch photoplethysmography coupled with a deep neural network can passively detect AF but with some loss of sensitivity and specificity against a criterion-standard ECG. Further studies will help identify the optimal role for smartwatch-guided rhythm assessment.
| S-EPMC5875390 | biostudies-literature
Project description:BackgroundPeople with impaired awareness of hypoglycemia (IAH) are at elevated risk of severe, potentially hazardous, hypoglycemia and would benefit from a device alerting to hypoglycemia. Heart rate variability (HRV) changes with hypoglycemia due to sympathetic activity. Since IAH is associated with suppressed sympathetic activity, we investigated whether hypoglycemia elicits a measurable change in HRV in patients with T1D and IAH.MethodEligible participants underwent a modified hyperinsulinemic euglycemic hypoglycemic clamp (glucose nadir, 43.1 ± 0.90 mg/dl), while HRV was measured by a VitalConnect HealthPatch. Measurements of HRV included Root Mean Square of the Successive Differences (RMSSD) and low to high frequency (LF:HF) ratio. Wilcoxon rank-sum test was used for testing within-subject HRV changes.ResultsWe included 12 participants (8 female, mean age 57 ± 12 years, mean HbA1c 57 ± 5 mmol/mol (7.4 ± 0.4%)). Symptoms increased from 4.0 (1.5-7.0) at euglycemia to 7.5 (5.0-11.0) during hypoglycemia (P = .003). In response to hypoglycemia, the LF:HF ratio and RMSSD increased when normalized for data obtained during euglycemia (both P < .01). The LF:HF ratio increased in 6 participants (50%) and declined in one other participant (8%). The RMSSD decreased in 3 (25%) and increased in 4 (33%) participants. In 2 patients, no change in HRV could be detected in response to hypoglycemia.ConclusionsThis study reveals that hypoglycemia-induced changes in HRV are retained in the majority of people with T1D and IAH, and that these changes can be detected by a wearable device. Real-time HRV seems usable for detection of hypoglycemia in patients with IAH.
| S-EPMC9445333 | biostudies-literature
Project description:BackgroundIntensive insulin therapy has documented benefits but may also come at the expense of a higher risk of hypoglycemia. Hypoglycemia is associated with higher all-cause mortality and nocturnal hypoglycemia has been associated with the sudden dead-in-bed syndrome. This proof-of-concept study sought to investigate if nocturnal hypoglycemia can be predicted.MethodContinuous glucose monitoring, meal, insulin, and demographics data from 463 people with type 1 diabetes were obtained from a clinical trial. A total of 4721 nights without or with hypoglycemia (429) were available including data from three consecutive days before the night. Thirty-two features were calculated based on these data. Data were split into 20% participants for evaluation and 80% for training. The optimal feature subset was found from forward selection of the 80% participants with linear discriminant analysis as basis for the classifier.ResultsThe forward selection resulted in a feature subset of four features. The evaluation resulted in an area under the receiver operating characteristics curve (ROC-AUC) of 0.79 leading to a sensitivity and a specificity of, e.g., 75% and 70%.ConclusionsIt was possible to predict nocturnal hypoglycemic episodes with a ROC-AUC of 0.79. A warning at bedtime about nocturnal hypoglycemia could be of great help for people with diabetes to enable preventive actions. Further development of the proposed algorithm is needed for implementation in everyday practice.
| S-EPMC7196854 | biostudies-literature