High-Resolution Digital Phenotypes from Consumer Wearables Enhance Prediction of Cardiometabolic Risk Markers

Background: Consumer-grade wearable devices enable detailed recordings of heart rate and step counts in free-living conditions. Recent studies have shown that summary statistics from these wearable recordings have potential uses for longitudinal monitoring of health and disease states. However, the relationship between higher resolution physiological dynamics from wearables and known markers of health and disease remains largely uncharacterized. Objective: We aimed to (i) derive high resolution digital phenotypes from observational wearable recordings, (ii) characterize their ability to predict modifiable markers of cardiometabolic disease, and (iii) study their connections with genetic predispositions for cardiometabolic disease and with lifestyle factors. Methods: We introduce a principled framework to extract interpretable high resolution phenotypes from wearable data recorded in free-living conditions. The proposed framework standardizes handling of data irregularities, encodes contextual information about underlying physiological state at any given time, and generates a set of 66 minimally redundant features across active, sedentary and sleep states. We applied our approach on a multimodal dataset, from the SingHEART study (NCT02791152), that comprises of heart rate and step count time series from wearables, clinical screening profiles, whole genome sequences and lifestyle survey responses from 692 healthy volunteers. We employed machine learning to model non-linear relationships between the high resolution phenotypes and clinical risk markers for blood pressure, lipid and weight abnormalities. For each risk type, we performed model comparisons based on Brier Skill Scores (BSS) to assess predictive value of the high resolution features over and beyond typical baselines. We then examined associations between the wearable-derived features, polygenic risk for cardiometabolic disease, and lifestyle habits and health perceptions. Results: Compared to typical summary statistic measures like resting heart rate, we find that the high-resolution features collectively have greater predictive value for modifiable clinical markers associated with cardiometabolic disease risk (average improvement in Brier Skill Score=52.3%, P<.001). Further, we show that heart rate dynamics from different activity states contain distinct information about type of cardiometabolic risk, with dynamics in sedentary states being most predictive of lipid abnormalities and patterns in active states being most predictive of blood pressure abnormalities (P<.001). Finally, our results reveal that subtle heart rate dynamics in wearable recordings serve as physiological correlates of genetic predisposition for cardiometabolic disease, lifestyle habits and health perceptions. Conclusions: High resolution digital phenotypes recorded by consumer wearables in free-living states have the potential to enhance prediction of cardiometabolic disease risk, and could enable more proactive and personalized health management. Clinical Trial Registration ID #NCT02791152. Keywords: Wearable device, heart rate, cardiometabolic disease, risk prediction, digital phenotypes, polygenic risk scores, time series analysis, machine learning, free-living

Heart rate dynamics and lactate following high-intensity race-pace continuous vs interval workouts in highly trained athletes

PurposeThe present study aimed to compare the physiological responses of high-intensity race-pace continuous vs. interval workouts commonly used in middle-distance athletics, by means of analyzing post-exercise cardiac autonomic regulation and lactate.MethodsNineteen highly-trained 800-m male runners were asked to run a 600-m race-pace continuous workout and a 2 × 4 × 200-m interval training, counterbalanced and randomized within one week of difference. Blood lactate jointly with linear and nonlinear heart rate dynamics were assessed during the immediate 15-min recovery. Age-category (Under23-Senior vs. Juvenile-Junior) was considered as an inter-subject factor.ResultsPeak lactate was higher following the interval training (15.51 ± 0.99 vs 13.83 ± 1.77 mmol L−1; P < 0.05) whereas lactate removal was almost nonexistent 15 min after both workouts (between 0 and 16%). Vagal modulation (ln RMSSD and lnRMSSD to RR ratio) remained significantly depressed at the end of recovery following both workouts, although the alteration was larger following the interval training. Detrended Fluctuation Analysis evidenced a more random HR behavior (DFA1 closer to 0.5) during the first 9 min of recovery after the interval training, whereas no significant change was observed in heart rate complexity (SampEn). Neither were differences found in post-exercise lactate and HR dynamics as a function of age-category.ConclusionsHigh-intensity workouts commonly used in middle-distance athletics, both race-pace continuous and intervallic approaches, induce a large depression of vagal modulation in highly trained runners, although interval trainings appear to induce even a greater alteration of both linear and nonlinear HR dynamics and a higher post-exercise peak lactate.

Field assessments of heart rate dynamics during spawning migration of wild and hatchery-reared Chinook salmon

During spawning, adult Pacific salmonids ( Oncorhynchus spp . ) complete challenging upriver migrations during which energy and oxygen delivery must be partitioned into activities such as locomotion, maturation and spawning behaviours under the constraints of an individual's cardiac capacity. To advance our understanding of cardiac function in free-swimming fishes, we implanted migrating adult Chinook salmon ( Oncorhynchus tshawytscha ) collected near the mouth of the Sydenham River, Ontario, with heart rate ( f H ) biologgers that recorded f H every 3 min until these semelparous fish expired on spawning grounds several days later. Fundamental aspects of cardiac function were quantified, including resting, routine and maximum f H , as well as scope for f H (maximum−resting f H ). Predictors of f H were explored using generalized least-squares regression, including water temperature, discharge, fish size and fish origin (wild versus hatchery). Heart rate was positively correlated with water temperature, which aligned closely with daily and seasonal shifts. Wild fish had slower resting heart rates than hatchery fish, which led to significantly higher scope for f H . Our findings suggest that wild salmon may have better cardiac capacity during migration than hatchery fish, potentially promoting migration success in wild fish. This article is part of the theme issue ‘Measuring physiology in free-living animals (Part I)’.

Analysis of Heart Rate Dynamics Before and During Meditation

<p>Heart rate is one of the most important vital signs. People usually face high tension in routine life, and if we found an effective method to control the heart rate, it would be very desirable. One of the goals of this paper is to examine changes in heart rate before and during meditation. Another goal is that what impact could have meditation on the human heartbeat.</p><p>To heart rate analysis before and during meditation, available heart rate signals have been used for the Physionet database that contains 10 normal subjects and 8 subjects that meditation practice has been done on them. In this paper, first is paid to extract linear and nonlinear characteristics of heart rate and then is paid to the best combination of features to identify two intervals before and during meditation using MLP and SVM classifiers with the help of sensitivity, specificity and accuracy measurements.</p><p>The achieved results in this paper showed that choosing the best combination of a feature to make a meaningful difference between two intervals before and during meditation includes two-time features (Mean HR, SDNN), a frequency feature ( ), and three nonlinear characteristics ( ). Also, using the support vector machine had better results than the MLP neural network. The sensitivity, specificity, and accuracy of the mean and standard deviation obtained respectively like 92.73 0.23, 89.05 0.67, 89.97 0.23 by using MLP and respectively like 95.96 0.09, 93.80 0.16, and 94.90 0.14 by using SVM.</p>As a result, using meditation can reduce the stress and anxiety of patients by effects on heart rate, and the treatment process speeds up and have an important role in improving the performance of the system.

PR interval as a predictor of syncope in tilt-up testing in adolescents and young adults

Background Syncope, as the most frequent consciousness disorder, is very common in young individuals. The aim of this study was to analyze ECG parameters and clinical properties obtained during tilt-up testing in 12 to 30-year-old subjects. We enrolled a total of 142 patients from our outpatient clinic (39 males, 103 females) with a true positive tilt-up test and analyzed ECG records obtained during tilt-testing. Data were stratified according to the age, gender, and type of syncope. Results PR interval shortening preceding syncope was found in all syncope types, irrespective of the gender. All types of syncope were more frequent in women (72.5%). Mixed syncope type was found to be the most common (47.18%). Male and female subjects differed in initial heart rate (71.56 vs 76.23/min, p=0.05), as well as heart rate dynamics during tilt-up testing. A gender difference was also found in systolic blood pressure (116.92 vs 110.44 mmHg, p<0.01), time to syncope onset (20.77 vs. 16.44 min, p=0.03), and the total number of syncopal episodes in patient history (2.79 vs. 4.62, p<0.05). Subjects with cardioinhibitory syncope had the longest PR interval (average 154.3 ms). PR interval prolongation and loss of variability during tilt-up testing positively correlated with aging (r=0.22, p<0.05). Nodal rhythm was found in 8 patients. Conclusion PR interval shortening on ECG tracings during a tilt-up test can be found in all subtypes of vasovagal syncope, thereby contrasting previous reports that these changes are a hallmark of the cardioinhibitory type of syncope. PR shortening, if observed during ECG monitoring, could be a potential predictor of syncope.

Non-linear heart rate dynamics during and after three controlled exercise intensities in healthy men

We aimed to analyse the complexity and fractal nature of heartbeat during constant exercise, at three different intensities, and recovery.Fourteen healthy men underwent 4 separate sessions. The first session was an incremental treadmill test to determine ventilatory thresholds (VT1 and VT2) and maximal aerobic speed (MAS). Each subject ran at VT1 and VT2 speeds and MAS (second, third and fourth day). The duration of VT1 and VT2 loads were selected in such a way that the product intensity-duration (training load) was the same. Sample Entropy (SampEn) and slope of Detrended Fluctuation Analysis (DFA α1) were measured during the whole session.DFA α1 declines with exercise, being less in the VT1 trial than in the other two.SampEn shows no significant change during exercise. The three tests induce the same decline in SampEn, but at the highest intensity (MAS) tends to decline during the exercise itself, whereas at lower intensities (VT1, VT2) the decline is delayed (10 min of recovery). Subsequently, SampEn at VT1 gradually recovers, whereas at VT2 and MAS it remains stable during recovery.In conclusion, exercise produces a loss of heartbeat complexity, but not fractal nature, during recovery and it depends on intensity.

Nonlinear Methods Most Applied to Heart-Rate Time Series: A Review

The heart-rate dynamics are one of the most analyzed physiological interactions. Many mathematical methods were proposed to evaluate heart-rate variability. These methods have been successfully applied in research to expand knowledge concerning the cardiovascular dynamics in healthy as well as in pathological conditions. Notwithstanding, they are still far from clinical practice. In this paper, we aim to review the nonlinear methods most used to assess heart-rate dynamics. We focused on methods based on concepts of chaos, fractality, and complexity: Poincaré plot, recurrence plot analysis, fractal dimension (and the correlation dimension), detrended fluctuation analysis, Hurst exponent, Lyapunov exponent entropies (Shannon, conditional, approximate, sample entropy, and multiscale entropy), and symbolic dynamics. We present the description of the methods along with their most notable applications.