Heart Rate Variability Segment Research Articles

Abstract 11895: Decreased Heart Rate Variability Associated With Hopelessness in Patients With Ischemic Heart Disease

Hopelessness, a negative outlook and sense of helplessness toward the future, is present in 27-52% of patients with ischemic heart disease (IHD) and is associated with increased morbidity and mortality. Physiologic dynamics of hopelessness in patients with IHD are largely unknown. Purpose : To examine how stress adaptation, measured by high frequency (HF) heart rate variability (HRV), following an IHD event may be associated with hopelessness. Subgroup differences focusing on characteristics known to be associated with hopelessness and/or HRV will also be examined (i.e., IHD event, marital status, sex). Hypothesis : Decreased HF HRV will be associated with greater levels of hopelessness. Methods : Participants were enrolled while hospitalized for an IHD event after being identified as appropriate for HRV analysis (n=69). Two weeks after hospital discharge, participants wore a Polar H7 heart rate monitor for 10 minutes while supine for HRV data collection. A demographic questionnaire and State-Trait Hopelessness Scale were then completed. From the 10-minutes of HRV data, a 5-minute HRV segment with the least amount of artifact (<5%) was used for analysis. HRV data was log transformed to account for non-normality. Linear regression was used to assess the relationship between HRV and state hopelessness. Results : There was a moderate negative correlation between HF HRV and state hopelessness (r= -0.35, p=0.004). This remained significant in multivariable regression models when controlling for history of depression and factors known to influence HRV (i.e., time of day, age, and smoking status). No sex differences (Cohen’s d=-0.04, p=0.78) were observed for HF HRV, but persons who underwent bypass surgery had lower HF compared to those that did not (Cohen’s d=0.63, p=0.04). Due to high correlation with HF and root mean square of successive differences between normal heartbeats (rMSSD) and SD1 (r=0.85), patterns of association with hopelessness and those variables were similar throughout. Conclusion : Decreased HF was moderately correlated with increased state hopelessness following an IHD event. Assessing HRV in patients following an IHD event could provide promising evidence of a biomarker to assist in identifying patients at risk for hopelessness.

Sleep staging classification based on a new parallel fusion method of multiple sources signals

Objective. In the field of medical informatics, sleep staging is a challenging and time consuming task undertaken by sleep experts. The conventional method for sleep staging is to analyze Polysomnograms (PSGs) recorded in a sleep lab, but the sleep monitoring with polysomnography (PSG) severely degrades the sleep quality. Despite recent significant progress in the development of automatic sleep staging methods, building a good model still remains a big challenge for sleep studies due to the data-variability and data-inefficiency issues. Electrooculograms (EOGs) and electrocardiograms (ECGs) and are much easier to record and may offer an attractive alternative for home sleep monitoring. PSGs from the Sleep Heart Health Study database were used. This study aims to establish an new automatic sleep staging algorithm by using electrooculogram (EOG) and electrocardiogram (ECG). Approach. First, the heart rate variability (HRV) is extracted from EOG with the Weight Calculation Algorithm and an ‘NRRD’ RR interval detection algorithm. Second, three feature sets were extracted from HRV segments and EOG segments: time-domain features, frequency-domain features and nonlinear-domain features. The frequency domain features and nonlinear-domain features were extracted by using Discrete Wavelet Transform, Autoregressive (AR), and Power Spectral entropy, and Refined Composite Multiscale Dispersion Entropy. Third, a new ‘Parallel Fusion Method’ (PFM) for sleep stage classification is proposed. Three kinds of feature sets from EOG and HRV segments are fused by using PFM. Fourth, Extreme Gradient Boosting (XGBoost) is employed for sleep staging. Main results. Our experimental results show significant performance improvement on automatic sleep staging on the target domains achieved with the new sleep staging approach. The performance of the proposed method is tested by evaluating the average accuracy, Kappa coefficient. The average accuracy of sleep classification results by using XGBoost classification model with PFM is 83% and the kappa coefficient is 0.7. Experimental results show that the performance of the proposed method is competitive with the most current methods and results, and the recognition rate of S1 stage is significantly improved. Significance. As a consequence, it would enable one to improve the quality of automatic sleep staging models when the EOG and HRV signals are fused, which can be beneficial for monitor sleep quality and keep abreast of health conditions. Besides, our study provides good research ideas and methods for scholars, doctors and individuals.

Personalized recognition of wake/sleep state based on the combined shapelets and K-means algorithm

BackgroundSleep affects almost all aspects, including health, memory and quality of life. It is important to distinguish wake/sleep correctly in sleep monitoring. Supervised recognition algorithms are trained on expensive and laborious polysomnography (PSG), and individual differences can’t be ignored. Most of the unsupervised recognition algorithms are based on acceleration signal, and body movement seriously affects its accuracy. ObjectiveIn order to improve the generalization ability of wake/sleep recognition and avoid the adverse effects of body movement, we propose an unsupervised method that only needs heart rate variability (HRV) to recognize wake/sleep states, and verify the classification results in a public database. MethodsShapelets algorithm is used to quantify the similarity between HRV segments, and K-means clustering algorithm is used to improve the shapelets algorithm to realize the classification of wake/sleep. The sleep tags in the database are used to verify and compare the classification results before and after the improvement. Besides, the influence of unbalanced samples on the classification performance of the two algorithms is analyzed. ResultsThe accuracy of the combined shapelets and K-means (SLKM) algorithm is 0.7800 ± 0.0692, 0.8826 ± 0.0533 in two databases. Compared with the shapelets algorithm, the accuracy is improved by 11.47% and 15.49% respectively. ConclusionsThis method can effectively realize wake/sleep recognition based on individual HRV. It has stronger robustness, which is very suitable for large-scale and long-time sleep monitoring.

Robust correlation dimension estimator for heart rate variability

Introduction. Non-linear methods of analysis have found widespread use in the Heart Rate Variability (HRV) technology, when the long-term HRV records are available. Using one of the effective nonlinear methods of analysis of HRV correlation dimension D2 for the standard 5-min HRV records is suppressed by unsatisfactory accuracy of available methods in case of short records (usually, doctors have about 500 RRs during standard 5-min HRV record), as well as complexity and ambiguity of choosing additional parameters for known methods of calculating D2. The purpose of the work. Building a robust estimator for calculating correlation dimension D2 with high accuracy for limited se-ries of RR-intervals observed in a standard 5-minute HRV record, i. e. with N  500. As well as demonstrating the capabilities of the D2 formula on a well known attractors (Lorenz, Duffing, Hennon and etc.) and in applications for Normal Sinus Rhythm (NSR), Congestive Heart Failure (CHF) and Atrial Fibrillation (AF). Materials and Methods. We used MIT-BIH long-term HRV records for normal sinus rhythm, congestive heart failure and atrial fibrillation. In order to analyze the accuracy of new robust estimator for D2, we used the known theoretical values for some famous attractors (Lorenz, Duffing, Hennon and etc.) and the most popular Grassberger-Procaccia (G-P) algorithm for D2. The results of the study. We have shown the effectiveness of the developed D2 formula for time series of limited length (N = 500–1000) by some famous attractors (Lorenz, Duffing, Hennon and etc.) and with the most popular Grassberger-Procaccia (G-P) algorithm for D2. It was demonstrated statistically significant difference of D2 for normal sinus rhythm and congestive heart failure by standard 5 min HRV segments from MIT-BIH database. The promised technology for early prediction of atrial fibrillation episodes by current D2 algorithm was shown for standard 5 min HRV segments from MIT-BIH Atrial Fibrillation database. Conclusion. Robust correlation dimension D2 estimator suggested in the article allows for time series of limited length (N ≈ 500) to calculate D2 value that differs at mean from a precise one by 5 ± 4%, as demonstrated for various well known attractors (Lorenz, Duffing, Hennon and etc.). We have shown on the standard 5-min segments from MIT-BIH database of HRV records: - the statistically significant difference of D2 for cases of normal sinus rhythm and congestive heart failure; - D2 drop significantly for the about 30 min. before of AF and D2 growth drastically under AF there was shown for HRV records with Atrial Fibrillation (AF) episodes. The suggested robust correlation dimension D2 estimator is perfect suitable for real time HRV monitoring as accurate, fast and non-consuming for computing resources. Key words: Hearth rate variability; Correlation dimension; Congestive heart failure; Atrial fibrillation.

Heart Rate Variability Assessment by the Entropy Parameters during Sleep

The nonlinear heart rate variability (HRV) parameter quantifies autonomic nervous system (ANS) activity based on the complexity or irregularity of an HRV dataset. At present, among various entropy-related parameters during sleep, approximate entropy (ApEn) and sample entropy (SampEn) are not as well understood as other entropy parameters such as Shannon entropy (SE) and conditional entropy (CE). Therefore, in this study, we investigated the characteristics of ApEn and SampEn to differentiate a rapid eye movement (REM) and nonrapid eye movement (NREM) for sleep stages. For nonlinear sleep HRV analysis, two target 10-minute, long-term HRV segments were obtained from each REM and NREM for 16 individual subjects. The target HRV segment was analyzed by moving the 2-minute window forward by 2 s, resulting in 240 results of each ApEn and SampEn. The ApEn and SampEn were averaged to obtain the mean value and standard deviation (SD) of all the results. SampEn provides excellent discrimination performance between REM and NREM in terms of the mean and SD (p<0.0001 and p=0.1989, respectively; 95% CI), but ApEn was inferior to SampEn (p=0.1980 and p=0.9931). The results indicate that SampEn, but not ApEn could be used to discriminate REM from NREM and detect various sleep-related incidents.

Effect of Motion Artifact on Variation in Heart Rate Variability Parameters

The heart rate variability (HRV) is a noninvasive way properly for investigating the activity of the autonomic nervous system (ANS) as well as to predict cardiovascular diseases. To guarantee an accurate HRV analysis, a motion artifact-free HRV recording must be obtained. However, complete removal of a motion artifact is impossible when measuring heartbeats for 5 min, and the motion artifact due to sudden ANS activity must be taken into consideration for the HRV parameters. And, the ANS balance has thus far been evaluated by each individual HRV parameter calculated for a single 5 min HRV segment, leading to the dynamic activity of the ANS within the same period being ignored. Therefore, to resolve this problem, HRV parameters for ultra-short-term segments that are short enough to reflect a sudden motion artifact must be analyzed. The aim of the present study was to evaluate the effects of a motion artifact on the variation in HRV parameters to provide detailed information on ANS activity. The 121 ultra-short-term HRV segments were created by moving a 1-min window forward by a time shift interval of 2 s for the entire 5 min HRV segment. The ratios of Ln LF to Ln HF in these ultra-short-term segments and a single 5 min segment with a motion artifact were 0.89 and 1.06, respectively, while those in a motion artifact-free HRV segment were 0.75 and 0.93, respectively. This variation test for a short-term motion artifact and motion artifact-free HRV dataset was found to affect the SDNN (7.73 and 2.68), SD2 (11.44 and 4.42), TINN (40.33 and 9.92), and Ln HF (0.37 and 0.13) the most in terms of the standard deviation, respectively. Taken together, the mean HRV parameters of many ultra-short-term segments might play an important role in evaluating dynamic ANS activities within a short-term segment, avoiding the false conclusions made by the traditional HRV analysis.

Artificial Neural Network-Based Automatic Detection of Food Intake for Neuromodulation in Treating Obesity and Diabetes.

Neuromodulation, such as vagal nerve stimulation and intestinal electrical stimulation, has been introduced for the treatment of obesity and diabetes. Ideally, neuromodulation should be applied automatically after food intake. The purpose of this study was to develop a method of automatic food intake detection through dynamic analysis of heart rate variability (HRV). Two experiments were conducted: (1) a small sample series with a standard test meal and (2) a large sample series with varying meal size. Electrocardiograms (ECGs) were collected in the fasting and postprandial states. Each ECG was processed to compute the HRV. For each HRV segment, time- and frequency-domain features were derived and used as inputs to train and test an artificial neural network (ANN). The ANN was trained and tested with different cross-validation methods. The highest classification accuracy reached with leave-one-subject-out-leave-one-sample-out cross-validation was 0.93 in experiment 1 and 0.88 in experiment 2. Retraining the ANN on recordings of a subject drastically increased the achieved accuracy for that subject to values of 0.995 and 0.95 in experiments 1 and 2, respectively. Automatic food intake detection by ANNs, using features from the HRV, is feasible and may have a great potential for neuromodulation-based treatments of meal-related disorders.

Detection of congestive heart failure from short-term heart rate variability segments using hybrid feature selection approach

ObjectivesThe aim of this work is to investigate the accuracy limits of automated detection of congestive heart failure (CHF) from short-term heart rate variability (HRV) series. Short-term HRV analysis uses 5-minute segments from HRV recordings to diagnose a disorder. This work proposes a hybrid feature selection procedure aimed at finding highly accurate models containing only a few highly informative features, which enables physiological interpretation of the features relevant for the model. Materials and methodsShort-term HRV segments are analyzed for CHF diagnosis. Subjects' records from four public PhysioNet databases are considered (66 healthy subjects and 42 CHF subjects). The problem is approached from a machine learning perspective, by extracting 111 linear time domain, frequency domain, time-frequency, nonlinear and symbolic dynamics HRV features. A multistage hybrid feature selection method is proposed that eventually eliminates most features. The method uses a symmetrical uncertainty filter, Naive Bayes wrapper with best first search, and final greedy iterative feature elimination. For classification purposes, we use rotation forest (RTF), radial based support vector machines (SVM), random forest (RF), multilayer perceptron artificial neural network, and k-nearest neighbors’ classifiers in order to evaluate the feature sets at each step of the process and to obtain as accurate model as possible. Leave-one-subject-out cross-validation evaluation method was used, with two variants: subject-level (coarse-grained) and feature vector-level (fine-grained). ResultsThe results show that the feature selection method is capable of either improving or retaining the classification accuracy of the full feature set (RTF: subject-level ACC = 88.9%, feature vector-level ACC = 85.6%; SVM: subject-level ACC = 89.8%, feature vector-level ACC = 83.5%; RF: subject-level ACC = 87.0%, feature vector-level ACC = 85.5%), while greatly reducing the number of included features, to only four HRV features for RTF and RF, and only two HRV features for SVM. The resulting best models for subject-level classification achieved are: RTF: ACC = 90.7%, SENS = 78.6%, SPEC = 98.6%, obtained with features: LF/HF ratio, maximum alphabet entropy, alphabet entropy variance, and HaarWaveletSD (scale = 8); SVM: ACC = 88.0%, SENS = 78.6%, SPEC = 93.9%, obtained with features: LF/HF ratio and Rate_U; RF: ACC = 90.7%, SENS = 78.6%, SPEC = 98.6%, obtained with features: LF/HF ratio, maximum alphabet entropy, Rate_U, and Rate_B. Other classifiers provided similar, but somewhat lower results. A comparison of the procedure with the results of individual filter, wrapper, and simple hybrid approaches is provided, which demonstrates the efficiency of the proposed procedure. ConclusionsThe results suggest that the method can achieve accurate generalizable models for automated diagnosis of CHF from short-term HRV segments in subjects with very few informative features. The choice of the best features and the classification results are similar between the three best classifiers, so the use of any of them with the proposed method is recommended. Nonlinear and symbolic dynamics features are shown to have an important role in the resulting models. The presented methodology may be useful for first-hand screening for CHF as well as for similar diagnostic or automated detection problems in biomedicine.

Paroxysmal atrial fibrillation prediction method with shorter HRV sequences

This paper proposes a method that predicts the onset of paroxysmal atrial fibrillation (PAF), using heart rate variability (HRV) segments that are shorter than those applied in existing methods, while maintaining good prediction accuracy. PAF is a common cardiac arrhythmia that increases the health risk of a patient, and the development of an accurate predictor of the onset of PAF is clinical important because it increases the possibility to stabilize (electrically) and prevent the onset of atrial arrhythmias with different pacing techniques. We investigate the effect of HRV features extracted from different lengths of HRV segments prior to PAF onset with the proposed PAF prediction method. The pre-processing stage of the predictor includes QRS detection, HRV quantification and ectopic beat correction. Time-domain, frequency-domain, non-linear and bispectrum features are then extracted from the quantified HRV. In the feature selection, the HRV feature set and classifier parameters are optimized simultaneously using an optimization procedure based on genetic algorithm (GA). Both full feature set and statistically significant feature subset are optimized by GA respectively. For the statistically significant feature subset, Mann–Whitney U test is used to filter non-statistical significance features that cannot pass the statistical test at 20% significant level. The final stage of our predictor is the classifier that is based on support vector machine (SVM). A 10-fold cross-validation is applied in performance evaluation, and the proposed method achieves 79.3% prediction accuracy using 15-minutes HRV segment. This accuracy is comparable to that achieved by existing methods that use 30-minutes HRV segments, most of which achieves accuracy of around 80%. More importantly, our method significantly outperforms those that applied segments shorter than 30 minutes.

A new algorithm for detection of epileptic seizures based on HRV signal

Epilepsy is one of the most frequent neurological disorders. In a significant number of cases, skilled professionals carry out the detection of the epileptic seizures manually. This necessitates automated epileptic seizure detection. Many researchers have presented computational methods for detecting epileptic seizures based on electroencephalogram signals. In this article, we propose a novel and efficient algorithm for detecting the presence of epileptic seizures in heart rate variability (HRV). This algorithm includes feature extraction and classification. Ten features include time and frequency domain analysis and nonlinear features extracted from one-lead electrocardiogram signal of epileptic patients. Extracted features were used as the input of an artificial neural network, which provides the final classification of the HRV segments (existence of epileptic seizure or not). Multilayer perceptron neural networks with different number of hidden layers and five training algorithms were designed. The results show sensitivity, specificity and accuracy of 88.66%, 90% and 88.33%, respectively, in secondary generalised and 83.33%, 86.11% and 84.72%, respectively, in complex partial seizures. The experimental results portray that the proposed algorithm efficiently detects the presence of epileptic seizure in HRV signals and showed a reasonable accuracy in detection.

Effects of feeding schedule changes on the circadian phase of the cardiac autonomic nervous system and serum lipid levels

The purpose of this study was to investigate whether scheduling meals earlier in the day affects the circadian phase of the cardiac autonomic nervous system as assessed by heart rate variability (HRV) and serum lipid levels. Healthy men aged 21.4±0.5years (n=14) with a habit of regularly skipping breakfast participated in this parallel trial involving altered feeding schedules. Participants in the early mealtime group (EM group, n=8) were asked to eat three meals at 8:00, 13:00, and 18:00, and the control group (n=6) ate at 13:00, 18:00, and 23:00 for 2weeks. On the measurement day before and after intervention, fasting blood samples and 24-h electrocardiograph recordings were collected. Spectral analysis was used for approximate 10-min HRV segments. Low frequency (LF) power, high frequency (HF) power, and the ratio of HF to total power (%HF) were calculated to assess sympathovagal balance. Acrophases of the circadian rhythm of HRV variables were obtained by nonlinear least squares regression. Triglyceride and total and LDL cholesterol levels were significantly decreased in the EM group when compared with the control group (p=0.035, 0.008, and 0.004, respectively). Acrophases for HRV variables were advanced in the EM group and their difference between before and after the intervention in LF power (-3.2±1.2h) and %HF (-1.2±0.5h) reached significant level, respectively (p<0.05). Timing of meals was a key factor in regulating circadian phases of the cardiac autonomic nervous system and lipid metabolism.

Automatic sleep staging using empirical mode decomposition, discrete wavelet transform, time-domain, and nonlinear dynamics features of heart rate variability signals

The conventional method for sleep staging is to analyze polysomnograms (PSGs) recorded in a sleep lab. The electroencephalogram (EEG) is one of the most important signals in PSGs but recording and analysis of this signal presents a number of technical challenges, especially at home. Instead, electrocardiograms (ECGs) are much easier to record and may offer an attractive alternative for home sleep monitoring. The heart rate variability (HRV) signal proves suitable for automatic sleep staging. Thirty PSGs from the Sleep Heart Health Study (SHHS) database were used. Three feature sets were extracted from 5- and 0.5-min HRV segments: time-domain features, nonlinear-dynamics features and time–frequency features. The latter was achieved by using empirical mode decomposition (EMD) and discrete wavelet transform (DWT) methods. Normalized energies in important frequency bands of HRV signals were computed using time–frequency methods. ANOVA and t-test were used for statistical evaluations. Automatic sleep staging was based on HRV signal features. The ANOVA followed by a post hoc Bonferroni was used for individual feature assessment. Most features were beneficial for sleep staging. A t-test was used to compare the means of extracted features in 5- and 0.5-min HRV segments. The results showed that the extracted features means were statistically similar for a small number of features. A separability measure showed that time–frequency features, especially EMD features, had larger separation than others. There was not a sizable difference in separability of linear features between 5- and 0.5-min HRV segments but separability of nonlinear features, especially EMD features, decreased in 0.5-min HRV segments. HRV signal features were classified by linear discriminant (LD) and quadratic discriminant (QD) methods. Classification results based on features from 5-min segments surpassed those obtained from 0.5-min segments. The best result was obtained from features using 5-min HRV segments classified by the LD classifier. A combination of linear/nonlinear features from HRV signals is effective in automatic sleep staging. Moreover, time–frequency features are more informative than others. In addition, a separability measure and classification results showed that HRV signal features, especially nonlinear features, extracted from 5-min segments are more discriminative than those from 0.5-min segments in automatic sleep staging.

Increasing sensitivity in the measurement of heart rate variability: The method of non-stationary RR time–frequency analysis

A novel method of the time–frequency analysis of non-stationary heart rate variability (HRV) is developed which introduces the fragmentary spectrum as a measure that brings together the frequency content, timing and duration of HRV segments. The fragmentary spectrum is calculated by the similar basis function algorithm. This numerical tool of the time to frequency and frequency to time Fourier transformations accepts both uniform and non-uniform sampling intervals, and is applicable to signal segments of arbitrary length. Once the fragmentary spectrum is calculated, the inverse transform recovers the original signal and reveals accuracy of spectral estimates. Numerical experiments show that discontinuities at the boundaries of the succession of inter-beat intervals can cause unacceptable distortions of the spectral estimates. We have developed a measure that we call the “RR deltagram” as a form of the HRV data that minimises spectral errors. The analysis of the experimental HRV data from real-life and controlled breathing conditions suggests transient oscillatory components as functionally meaningful elements of highly complex and irregular patterns of HRV.

Non-stationarities significantly distort short-term spectral, symbolic and entropy heart rate variability indices

The autonomic regulation is non-invasively estimated from heart rate variability (HRV). Many methods utilized to assess autonomic regulation require stationarity of HRV recordings. However, non-stationarities are frequently present even during well-controlled experiments, thus potentially biasing HRV indices. The aim of our study is to quantify the potential bias of spectral, symbolic and entropy HRV indices due to non-stationarities. We analyzed HRV series recorded in healthy subjects during uncontrolled daily life activities typical of 24 h Holter recordings and during predetermined levels of robotic-assisted treadmill-based physical exercise. A stationarity test checking the stability of the mean and variance over short HRV series (about 300 cardiac beats) was utilized to distinguish stationary periods from non-stationary ones. Spectral, symbolic and entropy indices evaluated solely over stationary periods were contrasted with those derived from all the HRV segments. When indices were calculated solely over stationary series, we found that (i) during both uncontrolled daily life activities and controlled physical exercise, the entropy-based complexity indices were significantly larger; (ii) during uncontrolled daily life activities, the spectral and symbolic indices linked to sympathetic modulation were significantly smaller and those associated with vagal modulation were significantly larger; (iii) while during uncontrolled daily life activities, the variance of spectral, symbolic and entropy rate indices was significantly larger, during controlled physical exercise, it was smaller. The study suggests that non-stationarities increase the likelihood to overestimate the contribution of sympathetic control and affect the power of statistical tests utilized to discriminate conditions and/or groups.