Free-living Conditions Research Articles

Hand tremor-based hypoglycemia detection and prediction in adolescents with type 1 diabetes

Diabetes is a chronic disease where blood glucose (BG) concentrations are consistently high. A dangerous diabetes-associated condition is hypoglycemia, where BG level drops below the normal threshold. Hypoglycemia is often accommodated by tremors, sweating, fatigue, anxiety, lightheadedness, disorientation, irritability, and tachycardia. Very few studies are focused on detecting enhanced tremor as a peripheral physiological response to declining BG concentrations. This study undertakes a machine-learning approach to predict hypoglycemia using hand tremors. Tremors were detected and characterized by frequency and amplitude in non-hypoglycemic and hypoglycemic conditions. Accelerometers and continuous glucose monitoring devices recorded the tremor and BG datasets. A home study of 32 T1D adolescents in a free-living condition was conducted. Simultaneously recorded accelerometer and continuous glucose monitor (CGM) data of 194.6 h were collected from 15 participants. These data were utilized for training and testing the predictive model. Four lengths of the sliding window (1, 2, 5, and 10 s) and four machine learning algorithms (decision tree, support vector machine, k-nearest neighbor, and bagged trees ensemble classifier) were applied to classify tremor as non-hypoglycemic or hypoglycemic. The greatest accuracy (86.65%) was achieved by the ensemble classifier (bagged trees based on the subspace k-NN) for 1 s window length. Multiple prediction windows are merged to generate aggregate sequential predictions. Prediction accuracy of 86.05% was achieved for a 15 s batch (3 s window length). This study demonstrates the feasibility of detecting and predicting the onset of hypoglycemia based on hand tremor data collected by wrist-worn accelerometer sensors.

Objective evaluation of the first post-lockdown on physical activity, sedentary behavior and food choice in a sample of French young adult students

From mid-March through early May 2020, France limited outdoor activities to one hour per day because of the Covid-19 epidemic. This inside lockdown could have a lasting impact on post-lockdown sedentary and physical activities and food choice. The aim of this study is to compare behaviors before and after the lockdown in a sample of French young adult students (sex ratio = 1:1, 22 ± 3 y old). Over a period of 4 days in both April and after mid-May 2020, 50 students used the WellBeNet smartphone application to record accelerometry data to evaluate their sedentary behavior and physical activity, and food consumption. Some behaviors remained stable for both periods: sedentary behavior, light-intensity activity, the number of meals per day, the number of portions of fruits, nuts, dairy products, vegetables, legumes and meat-fish-egg. Moderate-intensity activity duration was higher (+1.4%, p < 0.0001), whereas the food balance score was lower (−0.3, p = 0.06) during the return to free-living conditions. Different food behaviors in male and female students were observed between the two periods. The total daily number of food portions and of whole starchy products post lockdown was lower in female students (−1.07 and −0.47 portion/d, p = 0.07 and 0.04) but remained stable in male students, post-lockdown. The consumption of snacks was similar in men and women during the lockdown, whereas it was higher in men post-lockdown (+0.8 portion/d, p = 0.01). During the COVID-19 lockdown, food consumption choice was better overall and the moderate-intensity physical activity level was lower compared with the free-living subsequent period.

Improved screening of fall risk using free-living based accelerometer data

Falls are one of the most costly population health issues. Screening of older adults for fall risks can allow for earlier interventions and ultimately lead to better outcomes and reduced public health spending. This work proposes a solution to limitations in existing fall screening techniques by utilizing a hip-based accelerometer worn in free-living conditions. The work proposes techniques to extract fall risk features from periods of free-living ambulatory activity. Analysis of the proposed techniques is conducted and compared with existing screening methods using Functional Tests and Lab-based Gait Analysis. 1705 Older Adults from Umea (Sweden) were assessed. Data consisted of 1 Week of hip worn accelerometer data, gait measurements and performance metrics for 3 functional tests. Retrospective and Prospective fall data were also recorded based on the incidence of falls occurring 12 months before and after the study commencing respectively. Machine learning based experiments show accelerometer based measures perform best when predicting falls. Prospective falls had a sensitivity and specificity of 0.61 and 0.66 respectively while retrospective falls had a sensitivity and specificity of 0.61 and 0.68 respectively.

257-OR: Pen-Administered Low-Dose Dasiglucagon vs. Usual Care for Prevention and Treatment of Nonsevere Hypoglycemia in People with Type 1 Diabetes during Free-Living Conditions: A Phase 2, Randomized, Two-Period, Crossover Outpatient Study

Objective: To evaluate the efficacy, safety and feasibility of pen-administered low-dose dasiglucagon for prevention and treatment of non-severe hypoglycemia in people with type 1 diabetes during free-living conditions. Methods: Twenty-four adults with insulin pump-treated type 1 diabetes completed a phase 2, randomized, open-label, 2-period cross-over study with 2-week periods. During the usual care (UC) and dasiglucagon (DASI) period, participants managed impending and manifest episodes of hypoglycemia with regular carbohydrate consumption or pen-administered low-dose (80 µg) s.c. dasiglucagon, respectively. Glycemic control was evaluated using continuous glucose monitoring (Dexcom G6) and event-registration of prevention and treatment episodes. Patient-reported outcomes (PRO) were collected following study completion. Results: Participants administered dasiglucagon (IQR: 7-15) times during the 2-week period. Compared with UC, the mean difference (±SEM) in time in range (70-180 mg/dL) was 2.4%-point (±1.5%) (UC: 61.5%, DASI: 63.9%, p=0.129) , difference in time below range (&amp;lt;70 mg/dL) was 0.5%-point (±0.4%) (UC: 3.1%, DASI: 2.6%, p=0.170) , while total daily carbohydrate intake was significantly reduced (UC: 192 g, DASI: 172 g, p=0.007) . Dasiglucagon use was safe and well tolerated with nausea being the most frequent adverse effect (no. of participants: UC: 4, DASI: 8, p=0.206) . PRO demonstrated that 96% (p&amp;lt;0.001) of participants were likely to include dasiglucagon as part of their future routine management of hypoglycemia. Conclusion: Utilization of pen-administered, low-dose dasiglucagon for prevention and treatment of non-severe hypoglycemia during free-living conditions was safe and efficacious while significantly reducing the total daily carbohydrate intake and yielding high treatment satisfaction. Disclosure C.Laugesen: None. A.G.Ranjan: None. S.Schmidt: None. K.Nørgaard: Advisory Panel; Medtronic, Novo Nordisk A/S, Consultant; Novo Nordisk A/S, Research Support; Dexcom, Inc., Medtronic, Novo Nordisk A/S, Zealand Pharma A/S, Speaker's Bureau; Medtronic, Stock/Shareholder; Novo Nordisk A/S. Funding SteDiabetes Center Copenhagen is the sponsor of this investigator-initiated trial. Zealand Pharma provided financial support to the conduct of the study and supplied the study medication.

Restricting Sugar or Carbohydrate Intake Does Not Impact Physical Activity Level or Energy Intake Over 24 Hours Despite Changes in Substrate Use: A Randomised Crossover

ObjectivesTo determine the effects of dietary sugar or carbohydrate restriction on physical activity energy expenditure, energy intake, and physiological outcomes across 24 hours. MethodsIn a randomised, open-label crossover design, twenty-five healthy men (n = 10) and women (n = 15) consumed three diets over a 24-hour period: moderate carbohydrate and sugar content (MODSUG = 50% carbohydrate [20% sugars], 15% protein, 35% fat); low sugar content (LOWSUG = 50% carbohydrate [<5% sugars], 15% protein, 35% fat); and low carbohydrate content (LOWCHO = 8% carbohydrate [<5% sugars], 15% protein, 77% fat). Postprandial metabolic responses to a prescribed breakfast (20% EI) were monitored under laboratory conditions before an ad libitum test lunch, with subsequent diet and physical activity monitoring under free-living conditions until blood sample collection the following morning. ResultsThe MODSUG, LOWSUG and LOWCHO diets resulted in similar mean [95%CI] rates of both physical activity energy expenditure (771 [624,919] vs 677 [565,789] vs 802 [614,991] kcal·d–1; p = 0.29] and energy intake (2071 [1794,2347] vs 2195 [1918,2473] vs 2194 [1890,2498] kcal·d–1; p = 0.34), respectively. The LOWCHO condition elicited the lowest glycaemic and insulinemic responses to breakfast (p < 0.01) but the highest 24-h increase in LDL-cholesterol (p < 0.001), with no differences between the MODSUG and LOWSUG treatments. Leptin concentrations were decreased over 24-h of consuming LOWCHO relative to LOWSUG (p < 0.01). ConclusionsWhen energy density is controlled for, restricting either sugar or total dietary carbohydrate does not modulate physical activity level and energy intake over a 24-hour period despite large changes in metabolism. Funding SourcesThis study was supported by The Rank Prize Funds and Kenniscentrum Suiker and Voeding.

561-P: The Effect of Time-Restricted Eating on Glycemic Control and Insulin Sensitivity in Women with Prediabetes: A Pilot Study

Time-restricted eating (TRE) is a dietary pattern in which daily food intake is restricted to a 6-h period without any caloric intake for the rest of the day. TRE improves glucose homeostasis in animal models even independent of weight loss and calorie intake. However, the weight-loss independent effect of TRE on glucose homeostasis in people remains debatable. We conducted a pilot study to investigate whether TRE without weight loss improves glycemic control and insulin sensitivity in women with prediabetes. Participants were randomized to the TRE (9-h daily eating period) or Control group (15-h daily eating period) for 12 weeks. We studied five participants (TRE n = 3 and Control n = 2) with overweight/obesity, prediabetes, and habitual daily eating period greater than 14 hours ([means and SD] age 50 ± 16 years, BMI 31 ± 5 kg/m2) . The daily eating period decreased in the TRE group (13.8 ± 0.5 vs. 8.5 ± 0.5 h) , but not in the Control group (13.9 ± 0.9 vs. 14.1 ± 1.1 h) . Participants’ weight, diet composition, and number of eating occasions remained unchanged during the study period for both groups. Glycemic control in free-living conditions assessed by using continuous glucose monitoring tended to improve in the TRE group (24-h area under the curve for interstitial glucose concentration: 2572 ± 286 vs. 2161 ± 216 mg/dl/24 h) , but not in the Control group (23± 59 vs. 2374 ± 2mg/dl/24 h) . Skeletal muscle insulin sensitivity assessed by using a hyperinsulinemic-euglycemic clamp procedure tended to increase in the TRE group (median and IQR: 27 (21, 37) vs. 34 (27, 41) μmol/kg/min) , but not in the Control group (41 ± 9 vs. 36 ± μmol/kg/min) . These preliminary results suggest that TRE may lead to weight-loss-independent improvements in glycemic control and insulin sensitivity in women with prediabetes. Further ongoing research will establish the effect of TRE without weight loss on glucose homeostasis and insulin sensitivity in people at high risk for type 2 diabetes mellitus. Disclosure M.Chondronikola: None. S.Zhu: None. P.Surampudi: n/a. B.Roshanravan: None. C.Sanchez: None. Funding NIH-UL1-TR001860, Clinical and Translational Science Center-Highly Innovative Pilot Award and USDA NationalInstitute of Food and Agriculture, Hatch project numberCA-D-NTR-2618-H

Intake monitoring in free-living conditions: Overview and lessons we have learned

The progress in artificial intelligence and machine learning algorithms over the past decade has enabled the development of new methods for the objective measurement of eating, including both the measurement of eating episodes as well as the measurement of in-meal eating behavior. These allow the study of eating behavior outside the laboratory in free-living conditions, without the need for video recordings and laborious manual annotations. In this paper, we present a high-level overview of our recent work on intake monitoring using a smartwatch, as well as methods using an in-ear microphone. We also present evaluation results of these methods in challenging, real-world datasets. Furthermore, we discuss use-cases of such intake monitoring tools for advancing research in eating behavior, for improving dietary monitoring, as well as for developing evidence-based health policies. Our goal is to inform researchers and users of intake monitoring methods regarding (i) the development of new methods based on commercially available devices, (ii) what to expect in terms of effectiveness, and (iii) how these methods can be used in research as well as in practical applications.

Free-living gait does not differentiate chronic mTBI patients compared to healthy controls

BackgroundPhysical function remains a crucial component of mild traumatic brain injury (mTBI) assessment and recovery. Traditional approaches to assess mTBI lack sensitivity to detect subtle deficits post-injury, which can impact a patient’s quality of life, daily function and can lead to chronic issues. Inertial measurement units (IMU) provide an opportunity for objective assessment of physical function and can be used in any environment. A single waist worn IMU has the potential to provide broad/macro quantity characteristics to estimate gait mobility, as well as more high-resolution micro spatial or temporal gait characteristics (herein, we refer to these as measures of quality). Our recent work showed that quantity measures of mobility were less sensitive than measures of turning quality when comparing the free-living physical function of chronic mTBI patients and healthy controls. However, no studies have examined whether measures of gait quality in free-living conditions can differentiate chronic mTBI patients and healthy controls. This study aimed to determine whether measures of free-living gait quality can differentiate chronic mTBI patients from controls.MethodsThirty-two patients with chronic self-reported balance symptoms after mTBI (age: 40.88 ± 11.78 years, median days post-injury: 440.68 days) and 23 healthy controls (age: 48.56 ± 22.56 years) were assessed for ~ 7 days using a single IMU at the waist on a belt. Free-living gait quality metrics were evaluated for chronic mTBI patients and controls using multi-variate analysis. Receiver operating characteristics (ROC) and Area Under the Curve (AUC) analysis were used to determine outcome sensitivity to chronic mTBI.ResultsFree-living gait quality metrics were not different between chronic mTBI patients and controls (all p > 0.05) whilst controlling for age and sex. ROC and AUC analysis showed stride length (0.63) was the most sensitive measure for differentiating chronic mTBI patients from controls.ConclusionsOur results show that gait quality metrics determined through a free-living assessment were not significantly different between chronic mTBI patients and controls. These results suggest that measures of free-living gait quality were not impaired in our chronic mTBI patients, and/or, that the metrics chosen were not sensitive enough to detect subtle impairments in our sample.

0085 SleepInceptionNet: A Deep Learning Algorithm for Real-Time Sleep Stages Scoring Using Single-channel EEG

Abstract Introduction Most of the current automatic polysomnography sleep staging methods use multi-signal data and require a sequence of preceding and following epochs to score the stage of a specific epoch, which may not be desirable for analysis in real-time and/or in free-living conditions. We developed a deep learning-based sleep staging algorithm, namely SleepInceptionNet, that is designed to score each epoch using only single-channel electroencephalogram (EEG) data within that specific epoch. Methods Polysomnography data of 883 participants (937,975 thirty-second epochs) in the Multi-Ethnic Study of Atherosclerosis (MESA) obtained from the National Sleep Research Resource (NSRR) were randomly divided into a separate training/validation set of 194 participants and a test set of 689 participants. Each 30-second raw central EEG channel signal was transformed to time-frequency domain images using continuous wavelet transform method. Sleep stage in each epoch was obtained using SleepInceptionNet, in which the InceptionV3 convolutional neural network structure was trained and tuned on the time-frequency images of the training set to classify each epoch into one of the five stages of Wake, N1, N2, N3, or rapid eye movement (REM) sleep. Results Compared to the ground truth manually scored polysomnography sleep stages, SleepInceptionNet achieved an overall kappa agreement of 0.690 and overall weighted accuracy of 0.897. The model showed accuracy (mean±SD across the test set participants) of 0.940±0.067 in detecting Wake, 0.883±0.047 in detecting N1, 0.845±0.055 in detecting N2, 0.939±0.038 in detecting N3, and 0.930±0.038 in detecting REM epochs, in reference to manually scored polysomnography. Conclusion SleepInceptionNet showed a high agreement with manually scored polysomnography in epoch-by-epoch classification of sleep stages. This study demonstrates the viability of real-time, accurate sleep staging using a single-channel EEG, which could have a variety of applications such as delivery of on-demand interventions during specific sleep stages in free-living conditions. Support (If Any) MESA Sleep Ancillary study was funded by NIH-NHLBI Association of Sleep Disorders with Cardiovascular Health Across Ethnic Groups (RO1 HL098433). MESA is supported by NHLBI (HHSN268201500003I, N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01- HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169) and NCATS (UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420). NSRR was supported by NHLBI (R24 HL114473, 75N92019R002).

Detecting accelerometer non-wear periods using change in acceleration combined with rate-of-change in temperature

BackgroundAccelerometery is commonly used to estimate physical activity, sleep, and sedentary behavior. In free-living conditions, periods of device removal (non-wear) can lead to misclassification of behavior with consequences for research outcomes and clinical decision making. Common methods for non-wear detection are limited by data transformations (e.g., activity counts) or algorithm parameters such as minimum durations or absolute temperature thresholds that risk over- or under-estimating non-wear time. This study aimed to advance non-wear detection methods by integrating a ‘rate-of-change’ criterion for temperature into a combined temperature-acceleration algorithm.MethodsData were from 39 participants with neurodegenerative disease (36% female; age: 45–83 years) who wore a tri-axial accelerometer (GENEActiv) on their wrist 24-h per day for 7-days as part of a multi-sensor protocol. The reference dataset was derived from visual inspection conducted by two expert analysts. Linear regression was used to establish temperature rate-of-change as a criterion for non-wear detection. A classification and regression tree (CART) decision tree classifier determined optimal parameters separately for non-wear start and end detection. Classifiers were trained using data from 15 participants (38.5%). Outputs from the CART analysis were supplemented based on edge cases and published parameters.ResultsThe dataset included 186 non-wear periods (85.5% < 60 min). Temperature rate-of-change over the first five minutes of non-wear was − 0.40 ± 0.17 °C/minute and 0.36 ± 0.21 °C/minute for the first five minutes following device donning. Performance of the DETACH (DEvice Temperature and Accelerometer CHange) algorithm was improved compared to existing algorithms with recall of 0.942 (95% CI 0.883 to 1.0), precision of 0.942 (95% CI 0.844 to 1.0), F1-Score of 0.942 (95% CI 0.880 to 1.0) and accuracy of 0.996 (0.994–1.000).ConclusionThe DETACH algorithm accurately detected non-wear intervals as short as five minutes; improving non-wear classification relative to current interval-based methods. Using temperature rate-of-change combined with acceleration results in a robust algorithm appropriate for use across different temperature ranges and settings. The ability to detect short non-wear periods is particularly relevant to free-living scenarios where brief but frequent removals occur, and for clinical application where misclassification of behavior may have important implications for healthcare decision-making.

Deepaware: A hybrid deep learning and context-aware heuristics-based model for atrial fibrillation detection

BackgroundState-of-the-art automatic atrial fibrillation (AF) detection models trained on RR-interval (RRI) features generally produce high performance on standard benchmark electrocardiogram (ECG) AF datasets. These models, however, result in a significantly high false positive rates (FPRs) when applied on ECG data collected under free-living ambulatory conditions and in the presence of non-AF arrhythmias. MethodThis paper proposes DeepAware, a novel hybrid model combining deep learning (DL) and context-aware heuristics (CAH), which reduces the FPR effectively and improves the AF detection performance on participant-operated ambulatory ECG from free-living conditions. It exploits the RRI and P-wave features, as well as the contextual features from the ambulatory ECG. ResultsDeepAware is shown to be very generalizable and superior to the state-of-the-art models when applied on unseen benchmark ECG AF datasets. Most importantly, the model is able to detect AF efficiently when applied on participant-operated ambulatory ECG recordings from free-living conditions and has achieved a sensitivity (Se), specificity (Sp), and accuracy (Acc) of 97.94%, 98.39%, 98.06%, respectively. Results also demonstrate the effect of atrial activity analysis (via P-waves detection) and CAH in reducing the FPR over the RRI features-based AF detection model. ConclusionsThe proposed DeepAware model can substantially reduce the physician’s workload of manually reviewing the false positives (FPs) and facilitate long-term ambulatory monitoring for early detection of AF.

Detecting sleep outside the clinic using wearable heart rate devices

The adoption of multisensor wearables presents the opportunity of longitudinal monitoring of sleep in large populations. Personalized yet device-agnostic algorithms can sidestep laborious human annotations and objectify cross-cohort comparisons. We developed and tested a heart rate-based algorithm that captures inter- and intra-individual sleep differences in free-living conditions and does not require human input. We evaluated it on four study cohorts using different research- and consumer-grade devices for over 2000 nights. Recording periods included both 24 h free-living and conventional lab-based night-only data. We compared our optimized method against polysomnography, sleep diaries and sleep periods produced through a state-of-the-art acceleration based method. Against sleep diaries, the algorithm yielded a mean squared error of 0.04–0.06 and a total sleep time (TST) deviation of -2.70 (± 5.74) and 12.80 (± 3.89) minutes, respectively. When evaluated with PSG lab studies, the MSE ranged between 0.06 and 0.11 yielding a time deviation between -29.07 and -55.04 minutes. These results showcase the value of this open-source, device-agnostic algorithm for the reliable inference of sleep in free-living conditions and in the absence of annotations.

Generation of post-meal insulin correction boluses in type 1 diabetes simulation models for in-silico clinical trials: More realistic scenarios obtained using a decision tree approach

Background and objectiveIn type 1 diabetes (T1D) research, in-silico clinical trials (ISCTs) notably facilitate the design/testing of new therapies. Published simulation tools embed mathematical models of blood glucose (BG) and insulin dynamics, continuous glucose monitoring (CGM) sensors, and insulin treatments, but lack a realistic description of some aspects of patient lifestyle impacting on glucose control. Specifically, to effectively simulate insulin correction boluses, required to treat post-meal hyperglycemia (BG > 180 mg/dL), the timing of the bolus may be influenced by subjects’ behavioral attitudes. In this work, we develop an easily interpretable model of the variability of correction bolus timing observed in real data, and embed it into a popular simulation tool for ISCTs. MethodsUsing data collected in 196 adults with T1D monitored in free-living conditions, we trained a decision tree (DT) model to classify whether a correction bolus is injected in a future time window, based on predictors collected back in time, related to CGM data, previous insulin boluses and subject's characteristics. The performance was compared to that of a logistic regression classifier with LASSO regularization (LC), trained on the same dataset. After validation, the DT was embedded within a popular T1D simulation tool and an ISCT was performed to compare the simulated correction boluses against those observed in a subset of data not used for model training. ResultsThe DT provided better classification performance (accuracy: 0.792, sensitivity: 0.430, specificity: 0.878, precision: 0.455) than the LC and presented good interpretability. The most predictive features were related to CGM (and its temporal variations), time since the last insulin bolus, and time of the day. The correction boluses simulated by the DT, after implementation in the simulation tool, showed a good agreement with real-world data. ConclusionsThe DT developed in this work represents a simple set of rules to mimic the same timing of correction boluses observed on real data. The inclusion of the model in simulation tools allows investigators to perform ISCTs that more realistically represent the patient behavior in taking correction boluses and the post-prandial BG response. In the future, more complex models can be investigated.

IHearken: Chewing sound signal analysis based food intake recognition system using Bi-LSTM softmax network

Background and ObjectiveFood ingestion is an integral part of health and wellness. Continues monitoring of different food types and observing the amount being consumed prevents gastrointestinal diseases and weight-related issues. Food intake recognition (FIR) systems, thus have significant impact on everyday life. The purpose of this study is to develop an automatic approach for the FIR using a contemporary wearable hardware and machine learning technique. This will assist clinicians and concern person to manage health issues associated with food intake. MethodsIn this work, we present a novel hardware iHearken, a headphone-like wearable sensor-based system to monitor eating activities and recognize food intake type in the free-living condition. State-of-the-art hardware is designed for data acquisition where 16 subjects are recruited and 20 different food items are used for data collection. Further, chewing sound signals are analyzed for FIR using bottleneck features. The proposed model is divided into 4 distinct phases: data acquisition, event detection using a pre-trained model, bottleneck feature extraction, and classification based on bidirectional long short-term memory (Bi-LSTM) softmax model. The Bi-LSTM network with softmax function is applied to calculate the identification score for apiece chewing signal which further classifies the chewing signal data into liquid / solid food classes. ResultsThe results of proposed model performance is evaluated in (%) for accuracy, precision, recall and F-score as 97.422, 96.808, 98.0, and 97.512, respectively, and root mean square error (RMSE), and mean absolute percentage error (MAPE) as 0.160 1.030 respectively for numbers of correct food type recognized. Further, we also evaluated our model's performance for food classification into solid and liquid and achieved an accuracy (96.66%), precision (96.40%), recall (95.230%), F-score (95.79%), RMSE (0.182), and MAPE (2.22). We also demonstrated that the food recognition accuracy of different models with the proposed model differed statistically. ConclusionAn informatics complexity study of the proposed model was subsequently explored to review the effectiveness of the proposed wearable device and the methodology. The medical importance of this investigation is the reliable monitoring of the clinical development of the food intake classification methods via food chew event detection in the ambulatory environment has been justified.

Objectively Measured Vigorous‐Intensity Physical Activity is Related to Endothelial Function in Midlife and Older Men but not in Estrogen‐Deficient Postmenopausal Women

BackgroundAerobic exercise training is consistently associated with enhanced vascular endothelial function via suppression of oxidative stress in midlife/older (ML/O; 50 years) men but not in estrogen‐deficient postmenopausal women (E‐PMW). It is unclear if chronic physical activity (PA) patterns assessed under free‐living conditions, including sedentary behavior and vigorous exercise, are related to endothelial function among ML/O men or E‐PMW. We investigated if objectively measured PA patterns were associated with endothelial function in ML/O adults and the potential mechanisms involved.Methods172 ML/O adults were included for analysis (67 ± 7 years; 85M/87W). Endothelium‐dependent dilation was assessed via brachial artery flow‐mediated dilation (FMDBA), a bioassay of endothelial function. PA was assessed over 3‐7 days via wrist‐worn accelerometers (GENEactiv), which measured sedentary time and light, moderate, and vigorous activity. The relations between endothelial function and PA were examined using simple and stepwise linear regression. Multiple linear regression was used to examine the correlation between endothelial function and PA when controlling for conventional cardiovascular disease risk factors. Plasma oxidized low‐density lipoprotein (OxLDL) concentrations and total antioxidant status were used as markers of oxidative stress and antioxidant defenses, respectively. Interleukin‐6, tumor necrosis factor‐α, and C‐reactive protein concentrations were examined to assess chronic low‐grade inflammation. Mediation analysis was used to determine the potential for changes in oxidative stress, antioxidant defenses, and inflammation as mechanisms mediating the relation between PA and endothelial function.ResultsIn the entire cohort of subjects, FMDBA was negatively associated with sedentary time (R2=0.04; p=0.01), not associated with light PA (R2=0.02; p=0.09), and positively associated with moderate (R2=0.03; p=0.04) and vigorous (R2=0.06; p&lt;0.01) PA. Only sedentary time (R2=0.06; p=0.02) and vigorous PA (R2=0.11; p&lt;0.01) were associated with FMDBA in ML/O men. No PA parameters were associated with FMDBA in E‐PMW (all p&gt;0.05). Stepwise linear regression revealed vigorous PA primarily mediated the effects of PA on FMDBA in ML/O men (R2=0.10; p&lt;0.01). When controlling for age, blood pressure, body mass index, triglycerides, high‐density lipoprotein, low‐density lipoprotein, and fasting glucose, vigorous PA independently predicted FMDBA in ML/O men (R2=0.05; p=0.05) but not in E‐PMW (R2=0.03; p=0.18). Plasma OxLDL concentrations mediated 13.5% of the group effect of vigorous PA on FMDBA (p&lt;0.01); no other plasma markers had a mediating effect (all p&gt;0.05).ConclusionsObjectively measured PA, primarily vigorous PA, is positively associated with endothelial function in ML/O men but not in E‐PMW. This relation is mediated, in part, by oxidative stress. These associations are consistent with findings that aerobic exercise training does not improve endothelial function in E‐PMW. Alternative strategies to improve endothelial function in E‐PMW are needed.

Long-Term Prediction of Blood Glucose Levels in Type 1 Diabetes Using a CNN-LSTM-Based Deep Neural Network.

In this work, we leverage state-of-the-art deep learning-based algorithms for blood glucose (BG) forecasting in people with type 1 diabetes. We propose stacks of convolutional neural network and long short-term memory units to predict BG level for 30-, 60-, and 90-minute prediction horizon (PH), given historical glucose measurements, meal information, and insulin intakes. The evaluation was performed on two data sets, Replace-BG and DIAdvisor, representative of free-living conditions and in-hospital setting, respectively. For 90-minute PH, our model obtained mean absolute error of 17.30 ± 2.07 and 18.23 ± 2.97 mg/dL, root mean square error of 23.45 ± 3.18 and 25.12 ± 4.65 mg/dL, coefficient of determination of 84.13 ± 4.22% and 82.34 ± 4.54%, and in terms of the continuous glucose-error grid analysis 94.71 ± 3.89% and 91.71 ± 4.32% accurate predictions, 1.81 ± 1.06% and 2.51 ± 0.86% benign errors, and 3.47 ± 1.12% and 5.78 ± 1.72% erroneous predictions, for Replace-BG and DIAdvisor data sets, respectively. Our investigation demonstrated that our method achieved superior glucose forecasting compared with existing approaches in the literature, and thanks to its generalizability showed potential for real-life applications.

System Performance and User Feedback Regarding Wearable Bioimpedance System for Multi-Site Knee Tissue Monitoring: Free-Living Pilot Study With Healthy Adults.

Knee-focused wearable devices have the potential to support personalized rehabilitation therapies by monitoring localized tissue alterations related to activities that reduce functional symptoms and pain. However, supporting these applications requires reported data to be reliable and accurate which can be challenging in the unsupervised free-living conditions that wearable devices are deployed. This pilot study has assessed a knee-focused wearable sensor system to quantify 1) system performance (operation, rates of data artifacts, environment impacts) to estimate realistic targets for reliable data with this system and 2) user experiences (comfort, fit, usability) to help inform future designs to increase usability and adoption of knee-focused wearables. Study data was collected from five healthy adult participants over 2 days, with 84.5 and 35.9% of artifact free data for longitudinal and transverse electrode configurations. Small to moderate positive correlations were also identified between changes in resistance, temperature, and humidity with respect to acceleration to highlight how this system can be used to explore relationships between knee tissues and environmental/activity context.

Relationship between Measured Aerobic Capacity and Total Energy Expenditure Obtained by the Doubly Labeled Water Method in Community-Dwelling, Healthy Adults Aged 81-94 Years.

The doubly labeled water method is a gold-standard method for the measurement of total energy expenditure in daily life. We aimed to identify the relationship between measured aerobic capacity and total energy expenditure, activity energy expenditure, or physical activity level using the doubly labeled water method in adults of advanced old age. A total of 12 physically independent older adults (10 men and 2 women), aged 81–94 years, participated in this study. The aerobic capacity was evaluated according to the lactate threshold. Total energy expenditure under free-living conditions was assessed using the doubly labeled water method, and self-reported physical activity was obtained using the Japanese version of the International Physical Activity Questionnaire. The lactate threshold was significantly positively correlated with total energy expenditure, activity energy expenditure, and physical activity level after adjusting for age and sex. We found that the aerobic capacity of the lactate threshold was positively and independently correlated with total energy expenditure, activity energy expenditure, or physical activity level. The present results suggest that maintaining aerobic capacity is an important factor in preventing frailty, although further research is required.

Redox-sensitive fluorescent biosensors detect Sinorhizobium meliloti intracellular redox changes under free-living and symbiotic lifestyles

Reactive oxygen species such as hydrogen peroxide (H2O2) are key signaling molecules that control the setup and functioning of Rhizobium-legume symbiosis. This interaction results in the formation of a new organ, the root nodule, in which bacteria enter the host cells and differentiate into nitrogen (N2)-fixing bacteroids. The interaction between Sinorhizobium meliloti and Medicago truncatula is a genetic model to study N2-fixing symbiosis. In previous work, S. meliloti mutants impaired in the antioxidant defense, showed altered symbiotic properties, emphasizing the importance of redox-based regulation in the bacterial partner. However, direct measurements of S. meliloti intracellular redox state have never been performed. Here, we measured dynamic changes of intracellular H2O2 and glutathione redox potential by expressing roGFP2-Orp1 and Grx1-roGFP2 biosensors in S. meliloti. Kinetic analyses of redox changes under free-living conditions showed that these biosensors are suitable to monitor the bacterial redox state in real-time, after H2O2 challenge and in different genetic backgrounds. In planta, flow cytometry and confocal imaging experiments allowed the determination of sensor oxidation state in nodule bacteria. These cellular studies establish the existence of an oxidative shift in the redox status of S. meliloti during bacteroid differentiation. Our findings open up new possibilities for in vivo studies of redox dynamics during N2-fixing symbiosis.

Pilot experimental study; Effect of environmental stimulation consisting of sound with high-frequency components, aromas, and light exposure from organic light-emitting diodes (OLEDs) toward rest-activity rhythm in institutionalized patients with dementia

Environmental stimulation is expected to have a positive impact on night sleep, psychological or functional states in dementia. The objective of this pilot study was to examine the effects of environmental stimulation consisting of sounds with high-frequency components, aromas, and light exposure from organic light-emitting diodes (OLEDs) to create comfortable living spaces for persons with dementia. Subjects (n =6) were recruited from elderly persons admitted to a single nursing home in Akita Prefecture, Japan, between August and September 2020. The night-time sleep state, the rest-activity rhythm, and the light exposure of the subjects living in environments with or without environmental stimulation consisting of sounds with high-frequency components, aromas, and light exposure from OLEDs were measured for 46.3 consecutive days using wrist activate devices under free-living conditions in a nursing home. In a period of environmental stimulation depending on the presence or absence of sounds with high-frequency components, reduction of the fragmented rest-activity rhythms was significantly observed in the subjects (p &lt; 0.05). However, changes in the night-time sleep state had no significant difference during the study period. In conclusion, these preliminary results suggest that future examinations are warranted not only to inform effective or comfortable living conditions for elderly persons with dementia but also to improve the disruption of rest-activity rhythms in persons with dementia.