A new heart rate variability analysis method by means of quantifying the variation of nonlinear dynamic patterns

Bidimensional and Tridimensional Poincaré Maps in Cardiology: A Multiclass Machine Learning Study

Heart rate is a nonstationary signal and its variation may contain indicators of current disease or warnings about impending cardiac diseases. Hence, heart rate variation analysis has become a noninvasive tool to further study the activities of the autonomic nervous system. In this scenario, the Poincaré plot analysis has proven to be a valuable tool to support cardiac diseases diagnosis. The study’s aim is a preliminary exploration of the feasibility of machine learning to classify subjects belonging to five cardiac states (healthy, hypertension, myocardial infarction, congestive heart failure and heart transplanted) using ten unconventional quantitative parameters extracted from bidimensional and three-dimensional Poincaré maps. Knime Analytic Platform was used to implement several machine learning algorithms: Gradient Boosting, Adaptive Boosting, k-Nearest Neighbor and Naïve Bayes. Accuracy, sensitivity and specificity were computed to assess the performances of the predictive models using the leave-one-out cross-validation. The Synthetic Minority Oversampling technique was previously performed for data augmentation considering the small size of the dataset and the number of features. A feature importance, ranked on the basis of the Information Gain values, was computed. Preliminarily, a univariate statistical analysis was performed through one-way Kruskal Wallis plus post-hoc for all the features. Machine learning analysis achieved interesting results in terms of evaluation metrics, such as demonstrated by Adaptive Boosting and k-Nearest Neighbor (accuracies greater than 90%). Gradient Boosting and k-Nearest Neighbor reached even 100% score in sensitivity and specificity, respectively. The most important features according to information gain are in line with the results obtained from the statistical analysis confirming their predictive power. The study shows the proposed combination of unconventional features extracted from Poincaré maps and well-known machine learning algorithms represents a valuable approach to automatically classify patients with different cardiac diseases. Future investigations on enriched datasets will further confirm the potential application of this methodology in diagnostic.

Age induced interactions between heart rate variability and systolic blood pressure variability using approximate entropy and recurrence quantification analysis: a multiscale cross correlation analysis

The purpose of this study is to study the effect of age on the correlation between heart rate variability (HRV) and blood pressure variability (BPV). To meet this end, multi-scale cross correlation (CC) analysis of HRV and systolic blood pressure variability (SBPV) was performed. The Approximate Entropy (ApEn) and Recurrence Quantification Analysis (RQA) derived indices, calculated from RR interval series (RRi) and systolic blood pressure (SBP) series at multiple temporal scales, are the basis of this CC analysis. For the computation of ApEn and RQA indices, the tolerance threshold (r) is chosen by either: (i) selecting any arbitrary value (0.2) within the recommended range (0.1-0.25) times standard deviation (SD) of time series, and (ii) taking the 'r' (r_opt) corresponding to maximum ApEn (ApEn_max) as tolerance threshold. It is found that (i) at each time scale (τ), a lower SD is observed when indices are computed using r_opt than [Formula: see text] (r_0.2), for RRi as well as SBP series, (ii) descriptive indices of RRi are found significant (p < 0.05) at all scales (τ), however for SBP, these are found insignificant (p > 0.05) at most of the scales, (iii) CC values of descriptive statistics viz., mean and SD are not significant (p > 0.05) irrespective of τ, barring τ = 1, (iv) CC values of ApEn and RQA indices, found using r_opt, are found significant (p < 0.05) and provide enhanced stratification at τ = 1, 2 and 3, whereas this significant correlation and strong classification is missing for indices calculated using r_0.2, and (v) Lastly as τ increases, ApEn and RQA indices, computed with r_opt, reverse their trend but manage to provide significant difference in elder and younger subjects. It is concluded that HRV and SBPV interactions gets altered with age. Descriptive indicators however are not enough to capture these changes. These complex interactions can only be deciphered using complexity-based methods such as approximate entropy and that too at the multiple scale level.

A unified non-linear approach based on recurrence quantification analysis and approximate entropy: application to the classification of heart rate variability of age-stratified subjects

This paper presents a unified approach based on the recurrence quantification analysis (RQA) and approximate entropy (ApEn) for the classification of heart rate variability (HRV). In this paper, the optimum tolerance threshold (r_opt) corresponding to ApEn_max has been used for RQA calculation. The experimental data length (N) of RR interval series (RR_i) is optimized by taking r_opt as key parameter. r_opt is found to be lying within the recommended range of 0.1 to 0.25 times the standard deviation of the RR_i, when N ≥ 300. Consequently, RQA is applied to the age stratified RR_i and indices such as percentage recurrence (%REC), percentage laminarity (%LAM), and percentage determinism (%DET) are calculated along with ApEn_max, [Formula: see text], [Formula: see text], and an index namely the radius differential (R_D). Certain standard HRV statistical indices such as mean RR, standard deviation of RR (or NN) interval (SDNN), and the square root of the mean squared differences of successive RR intervals (RMSSD) (Eur Hear J 17:354-381, 1996) are also found for comparison. It is observed that (i) R_D can discriminate between the elderly and young subjects with a value of 0.1151 ± 0.0236 (mean ± SD) and 0.0533 ± 0.0133 (mean ± SD) respectively for the elderly and young subjects and is found to be statistically significant with p < 0.05. (ii) Similar significant discrimination was obtained using [Formula: see text] with a value of 0.1827 ± 0.0382 (mean ± SD) and 0.2248 ± 0.0320 (mean ± SD) (iii) other significant indices were found to be %REC, %DET, %LAM, SDNN, and RMSSD; however, ApEn_max was found to be insignificant with p > 0.05. The above features of RR_i time series were tested for classification using support vector machine (SVM) and multilayer perceptron neural network (MLPNN). Higher classification accuracy was achieved using SVM with a maximum value of 99.71%. Graphical abstract.

Resting EEG Microstates and Autonomic Heart Rate Variability Do Not Return to Baseline One Hour After a Submaximal Exercise

Recent findings suggest that an acute physical exercise modulates the temporal features of the EEG resting microstates, especially the microstate map C duration and relative time coverage. Microstate map C has been associated with the salience resting state network, which is mainly structured around the insula and cingulate, two brain nodes that mediate cardiovascular arousal and interoceptive awareness. Heart rate variability (HRV) is dependent on the autonomic balance; specifically, an increase in the sympathetic (or decrease in the parasympathetic) tone will decrease variability while a decrease in the sympathetic (or increase in the parasympathetic) tone will increase variability. Relying on the functional interaction between the autonomic cardiovascular activity and the salience network, this study aims to investigate the effect of exercise on the resting microstate and the possible interplay with this autonomic cardiovascular recovery after a single bout of endurance exercise. Thirty-eight young adults performed a 25-min constant-load cycling exercise at an intensity that was subjectively perceived as “hard.” The microstate temporal features and conventional time and frequency domain HRV parameters were obtained at rest for 5 min before exercise and at 5, 15, 30, 45, and 60 min after exercise. Compared to the baseline, all HRV parameters were changed 5 min after exercise cessation. The mean durations of microstate B and C, and the frequency of occurrence of microstate D were also changed immediately after exercise. A long-lasting effect was found for almost all HRV parameters and for the duration of microstate C during the hour following exercise, indicating an uncompleted recovery of the autonomic cardiovascular system and the resting microstate. The implication of an exercise-induced afferent neural traffic is discussed as a potential modulator of both the autonomic regulation of heart rate and the resting EEG microstate.

The application of information theory for the research of aging and aging-related diseases

This article reviews the application of information-theoretical analysis, employing measures of entropy and mutual information, for the study of aging and aging-related diseases. The research of aging and aging-related diseases is particularly suitable for the application of information theory methods, as aging processes and related diseases are multi-parametric, with continuous parameters coexisting alongside discrete parameters, and with the relations between the parameters being as a rule non-linear. Information theory provides unique analytical capabilities for the solution of such problems, with unique advantages over common linear biostatistics. Among the age-related diseases, information theory has been used in the study of neurodegenerative diseases (particularly using EEG time series for diagnosis and prediction), cancer (particularly for establishing individual and combined cancer biomarkers), diabetes (mainly utilizing mutual information to characterize the diseased and aging states), and heart disease (mainly for the analysis of heart rate variability). Few works have employed information theory for the analysis of general aging processes and frailty, as underlying determinants and possible early preclinical diagnostic measures for aging-related diseases. Generally, the use of information-theoretical analysis permits not only establishing the (non-linear) correlations between diagnostic or therapeutic parameters of interest, but may also provide a theoretical insight into the nature of aging and related diseases by establishing the measures of variability, adaptation, regulation or homeostasis, within a system of interest. It may be hoped that the increased use of such measures in research may considerably increase diagnostic and therapeutic capabilities and the fundamental theoretical mathematical understanding of aging and disease.

Characterization of local complex structures in a recurrence plot to improve nonlinear dynamic discriminant analysis

Structures in recurrence plots (RPs), preserving the rich information of nonlinear invariants and trajectory characteristics, have been increasingly analyzed in dynamic discrimination studies. The conventional analysis of RPs is mainly focused on quantifying the overall diagonal and vertical line structures through a method, called recurrence quantification analysis (RQA). This study extensively explores the information in RPs by quantifying local complex RP structures. To do this, an approach was developed to analyze the combination of three major RQA variables: determinism, laminarity, and recurrence rate (DLR) in a metawindow moving over a RP. It was then evaluated in two experiments discriminating (1) ideal nonlinear dynamic series emulated from the Lorenz system with different control parameters and (2) data sets of human heart rate regulations with normal sinus rhythms (n = 18) and congestive heart failure (n = 29). Finally, the DLR was compared with seven major RQA variables in terms of discriminatory power, measured by standardized mean difference (DSMD). In the two experiments, DLR resulted in the highest discriminatory power with DSMD = 2.53 and 0.98, respectively, which were 7.41 and 2.09 times the best performance from RQA. The study also revealed that the optimal RP structures for the discriminations were neither typical diagonal structures nor vertical structures. These findings indicate that local complex RP structures contain some rich information unexploited by RQA. Therefore, future research to extensively analyze complex RP structures would potentially improve the effectiveness of the RP analysis in dynamic discrimination studies.

NONLINEAR ANALYSIS OF PHYSIOLOGICAL SIGNALS: A REVIEW

This paper reviews various nonlinear analysis methods for physiological signals. The assessment is based on a discussion of chaos-inspired methods, such as fractal dimension (FD), correlation dimension (D2), largest Lyapunov exponet (LLE), Renyi's entropy (REN), Shannon spectral entropy (SEN), and approximate entropy (ApEn). We document that these methods are used to extract discriminative features from electroencephalograph (EEG) and heart rate variability (HRV) signals by reviewing the relevant scientific literature. EEG features can be used to support the diagnosis of epilepsy and HRV features can be used to support the diagnosis of cardiovascular diseases as well as diabetes. Documenting the widespread use of these and other nonlinear methods supports our thesis that the study of feature extraction methods, based on the chaos theory, is an important subject which has been gaining more and significance in biomedical engineering. We adopt the position that pursuing research in the field of biomedical engineering is ultimately a progmatic activity, where it is necessary to engage in features that work. In this case, the nonlinear features are working well, even if we do not have conclusive evidence that the underlying physiological phenomena are indeed chaotic.

Automated quantification of the synchrogram by recurrence plot analysis

Recently, the concept of phase synchronization of two weakly coupled oscillators has raised a great research interest and has been applied to characterize synchronization phenomenon in physiological data. Phase synchronization of cardiorespiratory coupling is often studied by a synchrogram analysis, a graphical tool investigating the relationship between instantaneous phases of two signals. Although several techniques have been proposed to automatically quantify the synchrogram, most of them require a preselection of a phase-locking ratio by trial and error. One technique does not require this information; however, it is based on the power spectrum of phase's distribution in the synchrogram, which is vulnerable to noise. This study aims to introduce a new technique to automatically quantify the synchrogram by studying its dynamic structure. Our technique exploits recurrence plot analysis, which is a well-established tool for characterizing recurring patterns and nonstationarities in experiments. We applied our technique to detect synchronization in simulated and measured infants' cardiorespiratory data. Our results suggest that the proposed technique is able to systematically detect synchronization in noisy and chaotic data without preselecting the phase-locking ratio. By embedding phase information of the synchrogram into phase space, the phase-locking ratio is automatically unveiled as the number of attractors.

A real-time maximum-likelihood heart-rate estimator for wearable textile sensors

This paper presents a real-time maximum-likelihood heart-rate estimator for ECG data measured via wearable textile sensors. The ECG signals measured from wearable dry electrodes are notorious for its susceptibility to interference from the respiration or the motion of wearing person such that the signal quality may degrade dramatically. To overcome these obstacles, in the proposed heart-rate estimator we first employ the subspace approach to remove the wandering baseline, then use a simple nonlinear absolute operation to reduce the high-frequency noise contamination, and finally apply the maximum likelihood estimation technique for estimating the interval of R-R peaks. A parameter derived from the byproduct of maximum likelihood estimation is also proposed as an indicator for signal quality. To achieve the goal of real-time, we develop a simple adaptive algorithm from the numerical power method to realize the subspace filter and apply the fast-Fourier transform (FFT) technique for realization of the correlation technique such that the whole estimator can be implemented in an FPGA system. Experiments are performed to demonstrate the viability of the proposed system.