Search

Standard cardiopulmonary exercise testing (CPET) produces a rich dataset but its current analysis is often limited to a few derived variables such as maximal or peak oxygen uptake (V̇O2). We tested whether breath-by-breath CPET data could be used to determine sample entropy (SampEn) in 81 healthy children and adolescents (age 7-18 years old, equal sex distribution). To overcome challenges of the relatively small time-series CPET data size and its nonstationarity, we developed a Python algorithm for short-duration physiological signals. Comparing pre- and post-ventilatory threshold (VT1) CPET phases, we found: (1) SampEn decreased by 9.46% for V̇O2 and 5.01% for V̇CO2 (p

The multivariate time series classification (MTSC) task aims to predict a class label for a given time series. Recently, modern deep learning-based approaches have achieved promising performance over traditional methods for MTSC tasks. The success of these approaches relies on access to the massive amount of labeled data (i.e., annotating or assigning tags to each sample that shows its corresponding category). However, obtaining a massive amount of labeled data is usually very time-consuming and expensive in many real-world applications such as medicine, because it requires domain experts’ knowledge to annotate data. Insufficient labeled data prevents these models from learning discriminative features, resulting in poor margins that reduce generalization performance. To address this challenge, we propose a novel approach: supervised contrastive learning for time series classification (SupCon-TSC). This approach improves the classification performance by learning the discriminative low-dimensional representations of multivariate time series, and its end-to-end structure allows for interpretable outcomes. It is based on supervised contrastive (SupCon) loss to learn the inherent structure of multivariate time series. First, two separate augmentation families, including strong and weak augmentation methods, are utilized to generate augmented data for the source and target networks, respectively. Second, we propose the instance-level, and cluster-level SupCon learning approaches to capture contextual information to learn the discriminative and universal representation for multivariate time series datasets. In the instance-level SupCon learning approach, for each given anchor instance that comes from the source network, the low-variance output encodings from the target network are sampled as positive and negative instances based on their labels. However, the cluster-level approach is performed between each instance and cluster centers among batches, as opposed to the instance-level approach. The cluster-level SupCon loss attempts to maximize the similarities between each instance and cluster centers among batches. We tested this novel approach on two small cardiopulmonary exercise testing (CPET) datasets and the real-world UEA Multivariate time series archive. The results of the SupCon-TSC model on CPET datasets indicate its capability to learn more discriminative features than existing approaches in situations where the size of the dataset is small. Moreover, the results on the UEA archive show that training a classifier on top of the universal representation features learned by our proposed method outperforms the state-of-the-art approaches.

Physiological response to physical exercise through analysis of cardiopulmonary measurements has been shown to be predictive of a variety of diseases. Nonetheless, the clinical use of exercise testing remains limited because interpretation of test results requires experience and specialized training. Additionally, until this work no methods have identified which dynamic gas exchange or heart rate responses influence an individual's decision to start or stop physical activity. This research examines the use of advanced machine learning methods to predict completion of a test consisting of multiple exercise bouts by a group of healthy children and adolescents. All participants could complete the ten bouts at low or moderate-intensity work rates, however, when the bout work rates were high-intensity, 50% refused to begin the subsequent exercise bout before all ten bouts had been completed (task failure). We explored machine learning strategies to model the relationship between the physiological time series, the participant's anthropometric variables, and the binary outcome variable indicating whether the participant completed the test. The best performing model, a generalized spectral additive model with functional and scalar covariates, achieved 93.6% classification accuracy and an F1 score of 93.5%. Additionally, functional analysis of variance testing showed that participants in the 'failed' and 'success' groups have significantly different functional means in three signals: heart rate, oxygen uptake rate, and carbon dioxide uptake rate. Overall, these results show the capability of functional data analysis with generalized spectral additive models to identify key differences in the exercise-induced responses of participants in multiple bout exercise testing.

Natural patterns of physical activity in youth are characterized by brief periods of exercise of varying intensity interspersed with rest. To better understand systemic physiologic response mechanisms in children and adolescents, we examined five responses [heart rate (HR), respiratory rate (RR), oxygen uptake (V̇O2 ), carbon dioxide production (V̇CO2 ), and minute ventilation (V̇E), measured breath-by-breath] to multiple brief exercise bouts (MBEB). Two groups of healthy participants (early pubertal: 17 female, 20 male; late-pubertal: 23 female, 21 male) performed five consecutive 2-min bouts of constant work rate cycle-ergometer exercise interspersed with 1-min of rest during separate sessions of low- or high-intensity (~40% or 80% peak work, respectively). For each 2-min on-transient and 1-min off-transient we calculated the average value of each cardiopulmonary exercise testing (CPET) variable (Y̅). There were significant MBEB changes in 67 of 80 on- and off-transients. Y̅ increased bout-to-bout for all CPET variables, and the magnitude of increase was greater in the high-intensity exercise. We measured the metabolic cost of MBEB, scaled to work performed, for the entire 15 min and found significantly higher V̇O2 , V̇CO2 , and V̇E costs in the early-pubertal participants for both low- and high-intensity MBEB. To reduce breath-by-breath variability in estimation of CPET variable kinetics, we time-interpolated (second-by-second), superimposed, and averaged responses. Reasonable estimates of τ (

SignificanceGrowing levels of obesity and metabolic syndrome have driven demand for more advanced forms of body composition assessment. While various techniques exist to measure body composition, devices are typically expensive and not portable, involve radiation [in the case of dual-energy x-ray absorptiometry (DXA)], and are limited to analysis of adiposity while metabolic information from blood supply and oxygenation are not considered.AimWe evaluate whether diffuse optical spectroscopic imaging (DOSI) can be used to predict site-specific adiposity and percent fat (whole body) while simultaneously providing information about local tissue hemoglobin levels and oxygenation.ApproachDOSI measures of tissue composition in gastrocnemius, quadriceps, abdomen, and biceps, DXA whole-body composition, and ultrasound-derived skin and adipose tissue thickness (SATT) in the quadriceps were obtained from 99 individuals aged 7 to 34 years old.ResultsVarious DOSI-derived parameters were correlated with SATT and an optical method is proposed for estimating SATT using a newly defined parameter, the optical fat fraction (OFF), which considers all parameters that correlate with SATT. Broadband absorption and scattering spectra from study participants with the thinnest (SATT  ≈  0.25  ±  0.02  cm) and thickest SATT (SATT  ≈  1.55  ±  0.14  cm), representing best estimates for pure in vivo lean and fatty tissue, respectively, are reported. Finally, a trained prediction model is developed which allows DOSI assessment of OFF to predict DXA body-fat percentage, demonstrating that DOSI can be used to quantify body composition.ConclusionsThis study shows that DOSI can be used to assess the adiposity of specific tissues or the entire human body, and the OFF parameter is defined for corroboration and further evaluation in future research.

PurposeThe electronic health record, data science advances, and dynamic environmental and infectious threats to child health highlight the need for harmonized and interoperable approaches to pediatric cardiopulmonary exercise testing (CPET). Accordingly, we developed a terminology harmonization in exercise medicine and exercise science domain analysis model (THEMES DAM) to structure CPET data elements.MethodsTHEMES DAM identified 114 data elements, including participant information, calibration, equipment, protocols, laboratory personnel, encouragement strategies, and analysis procedures. We used the THEMES DAM, vetted by the international data standards organization HL7, to construct a current-state survey of pediatric CPET centers in the United States. Forty-eight of 101 centers responded to a questionnaire covering seven major topic areas (38 items).ResultsCenters predominantly performed between 100 and 500 tests annually. Cardiac disease represented 55% of referrals. Almost all centers calibrated gas concentrations and flow daily, but 42% never calibrated their treadmill or cycle ergometers. All centers measured V̇O2peakbut calculated differently. Centers used a variety of protocols (e.g., for treadmill: 61%, Bruce; 43%, modified Bruce; 59%, other); 44% calculated CPET slopes from submaximal portions of CPET (e.g., V̇O2-HR). All centers verbally encouraged participants, but only 40% used a standardized approach. The interpretation of CPET was done by physicians (60%), exercise physiologists (25%), exercise technicians (10%), nurses (1%), or others (4%). Ninety-one percent would agree to collaborate in multicenter research, 89% to establish dynamic reference values, and 83% to better interpret CPET.ConclusionsThe survey data and the implementation of THEMES DAM could accelerate interoperability across multiple centers. This would facilitate a nimble approach to create pediatric reference values responsive to the constantly changing health environment and stimulate novel approaches to CPET research and clinical application.

IntroductionPhysical activity and fitness are essential for healthy growth in children. The National Health and Nutrition Examination Survey (NHANES) evaluated fitness by estimating V̇O2 max from submaximal measurements of heart rate (HR) during graded treadmill exercise. Our aims were (a) to examine how well NHANES methodology used to estimate V̇O2 max correlated with actual VO2 max and (b) to evaluate a novel fitness metric using actual data collected during exercise and its relationship to physical activity and sedentary time, lipid profiles, and body composition.MethodsFifty-three adolescents completed NHANES submaximal exercise protocol and maximal graded cardiopulmonary exercise testing. We used a novel approach to quantifying fitness (Δvelocity × incline × body mass (VIM)/ΔHR slopes) and evaluated its relationship to physical activity and sedentary time using NHANES data (n = 4498). In a subset (n = 740), we compared ΔVIM/ΔHR slopes to NHANES estimated V̇O2 max and examined their relationship to cardiovascular risk factors (BMI percentiles and lipid levels).ResultsMeasured V̇O2 peak was moderately correlated with NHANES estimated V̇O2 max (r = 0.53, P

Physiol Rep, 7 (17), 2019 e14178. https://doi.org/10.14814/phy2.14178 Dr. Kim D. Lu was inadvertently omitted from the authorship list. Dr. Lu was involved in all aspects of this manuscript including: development of project hypotheses and methods, data collection, analysis, and the writing and review of the final submission. Kim D. Lu MD Pediatric Exercise and Genomics Research Center Department of Pediatrics, University of California, Irvine, School of Medicine.

Current methods for continuous respiration monitoring such as respiratory inductive or optoelectronic plethysmography are limited to clinical or research settings; most wearable systems reported only measures respiration rate. Here we introduce a wearable sensor capable of simultaneously measuring both respiration rate and volume with high fidelity. Our disposable respiration sensor with a Band-Aid© like formfactor can measure both respiration rate and volume by simply measuring the local strain of the ribcage and abdomen during breathing. We demonstrate that both metrics are highly correlated to measurements from a medical grade continuous spirometer on participants at rest. Additionally, we also show that the system is capable of detecting respiration under various ambulatory conditions. Because these low-powered piezo-resistive sensors can be integrated with wireless Bluetooth units, they can be useful in monitoring patients with chronic respiratory diseases in everyday settings.

BackgroundBronchiectasis is associated with morbidity, low exercise capacity and poor quality of life. There is a paucity of data on exercise capacity using cardiopulmonary exercise test (CPET) in non-cystic fibrosis (CF) bronchiectasis. Our aim was to compare exercise capacity using CPET in CF and non-CF bronchiectasis patients.MethodsCross-sectional retrospective/prospective controlled study assessing CPET using cycle ergometer. Exercise parameters and computed tomography (CT) findings were compared. Results: Hundred two patients with bronchiectasis and 88 controls were evaluated; 49 CF (age 19.7 ± 9.7 y/o, FEV1%predicted 70.9 ± 20.5%) and 53 non-CF (18.6 ± 10.6 y/o, FEV1%predicted 68.7 ± 21.5%). Peak oxygen uptake (peak [Formula: see text]) was similar and relatively preserved in both groups (CF 1915.5±702.0; non-CF 1740±568; control 2111.0±748.3 mL/min). Breathing limitation was found in the two groups vs. control; low breathing reserve (49% in CF; 43% non-CF; 5% control) and increased [Formula: see text] (CF 31.4±4.1, non-CF 31.7±4.1 and control 27.2 ± 2.8). Oxygen pulse was lower in the non-CF; whereas a linear relationship between peak [Formula: see text] vs. FEV1 and vs. FVC was found only for CF. CT score correlated with [Formula: see text] and negatively correlated with [Formula: see text] and post exercise oxygen saturation (SpO2).ConclusionsCPET parameters may differ between CF and non-CF bronchiectasis. However, normal exercise capacity may be found unrelated to the etiology of the bronchiectasis. Anatomical changes in CT are associated with functional finding of increased [Formula: see text] and decreased SpO2. Larger longitudinal studies including cardiac assessment are needed to better study exercise capacity in different etiologies of non-CF bronchiectasis.Trial registrationClinicalTrials.gov, registration number: NCT03147651.

ObjectiveTo evaluate the relationship of fitness and obesity on asthma risk in adolescent girls and boys.Study designA cross-sectional assessment of participants 12-19 years of age was conducted by the use of data from the 1999-2004 National Health and Nutrition Examination Survey. Participants completed cardiorespiratory fitness testing, body composition measurements, and respiratory questionnaires.ResultsA total of 4828 participants were included. Overweight/obesity was associated with increased odds of history of asthma (aOR 1.63, 95% CI 1.16-2.30), current asthma (aOR 1.73, 95% CI 1.13, 2.64), and wheezing (aOR 1.40, 95% CI 1.03-1.91) in girls. Overweight/obesity also was associated with increased odds of asthma attacks (aOR 2.67, 95% CI 1.56-4.65) and wheezing related to exercise (aOR 1.60, 95% CI 1.07-2.38) in girls. High fitness was associated with lower odds of asthma-related visits to the emergency department (aOR 0.24, 95% CI 0.07-0.89), wheezing-related medical visits (aOR 0.31, 95% CI 0.13-0.75), wheezing-related missed days (aOR 0.14, 95% CI 0.06-0.33), and wheezing related to exercise (aOR 0.43, 95% CI 0.24-0.76) in boys.ConclusionOverweight/obesity is associated with increased asthma prevalence and morbidity in girls but not in boys, independent of fitness. High fitness is associated with decreased rates of asthma morbidity in boys but not in girls, independent of weight status. Obesity and fitness may each influence asthma onset and severity in different ways for girls compared with boys.

Energy imbalance exposes athletes to relative energy deficiency in sports (REDs) syndrome. Data on energy consumption, REDs, and bone mineral density (BMD) in adolescent acrobatic gymnasts, especially in males, are scarce. Our aim was to examine the eating habits, energy balance, body composition, and BMD of these athletes. In this study, 18 healthy adolescents participating in competitive acrobatic gymnastics completed a questionnaire, underwent a dual-energy X-ray absorptiometry scan (DXA), received a food log, and had their activities monitored for 3 days. Eighteen acrobats were enrolled (mean age: 14.3 ± 1.2 years; males: 6/18). The mean total body BMD Z-score was 0.4 ± 1.0. Top-position acrobats (7/18) had significantly lower total body BMD Z-scores than base-positioned acrobats (−0.2 ± 0.3 vs. 0.8 ± 0.3, p = 0.032), though their forearms were not significantly different (0.2 ± 0.5 vs. 0.8 ± 0.7, p = 0.331). No sex differences were found for BMD Z-scores, BMI, or energy availability. The BMD parameters of the acrobats were within the normal range for a healthy pediatric population, although three had low BMDs (<−1 SD) for healthy athletes. Total body and LS BMD Z-scores were significantly lower in top-position athletes compared to base-position athletes. These findings suggest personalized (top vs. base) training programs (high-impact training) that may achieve better health outcomes. [ABSTRACT FROM AUTHOR]

Author(s): Michal Gur[sup.1] , Karin Yaacoby-Bianu[sup.1] , Anat Ilivitzki[sup.2] , Ronen Bar-Yoseph[sup.1] , Vered Nir[sup.1] , Fahed Hakim[sup.1] [sup.3] , Yazeed Toukan[sup.1] [sup.3] , Lea Bentur[sup.1] [sup.3] Author Affiliations: (1) [...], Introduction Bronchiolitis obliterans (BO) is a chronic airway disease following an insult to the lower respiratory tract. Lung clearance index (LCI) measures ventilation inhomogeneity and has been studied in cystic fibrosis (CF). We aimed to evaluate LCI in BO and to compare it to LCI in CF patients. Methods LCI was measured in BO patients, compared to CF patients, and correlated with spirometry and CT findings. Results Twenty BO patients and 26 CF patients (with similar mean age and BMI) underwent evaluation. FEV1 % and FEF25-75 % predicted were significantly lower in the BO group (60.5 ± 17.8 vs. 72.7 ± 20.7, p = 0.041, and 42.8 ± 22.8 vs. 66.4 ± 37.4, p = 0.017, respectively). In both groups, LCI was inversely correlated with FVC %, FEV1 %, and FEF25-75 % predicted. LCI % was slightly higher (190.4 ± 63.5 vs. 164.9 ± 39.4, p = 0.1) and FRC gas % (measured by multiple breath washout) was significantly higher in the BO group (92.5 ± 35.9 vs. 71.3 ± 18, p = 0.014). The strength of statistical association between the lower FEF25-75 % values and the higher LCI values was stronger in BO patients. Conclusions Similar to CF, LCI may provide estimation of ventilation inhomogeneity in BO. The results indicate greater small airway involvement and air trapping in BO. Further prospective longitudinal studies evaluating the correlation of LCI measurements with multiple clinical and physiological parameters should be performed to assess the clinical benefit of LCI measurement in BO.