Detection of Coupling in Short Physiological Series by a Joint Distribution Entropy Method

Discrimination of Patients with Varying Degrees of Coronary Artery Stenosis by ECG and PCG Signals Based on Entropy

Coronary heart disease (CHD) is the leading cause of cardiovascular death. This study aimed to propose an effective method for mining cardiac mechano-electric coupling information and to evaluate its ability to distinguish patients with varying degrees of coronary artery stenosis (VDCAS). Five minutes of electrocardiogram and phonocardiogram signals was collected synchronously from 191 VDCAS patients to construct heartbeat interval (RRI)-systolic time interval (STI), RRI-diastolic time interval (DTI), HR-corrected QT interval (QTcI)-STI, QTcI-DTI, Tpeak-Tend interval (TpeI)-STI, TpeI-DTI, Tpe/QT interval (Tpe/QTI)-STI, and Tpe/QTI-DTI series. Then, the cross sample entropy (XSampEn), cross fuzzy entropy (XFuzzyEn), joint distribution entropy (JDistEn), magnitude-squared coherence function, cross power spectral density, and mutual information were applied to evaluate the coupling of the series. Subsequently, support vector machine recursive feature elimination and XGBoost were utilized for feature selection and classification, respectively. Results showed that the joint analysis of XSampEn, XFuzzyEn, and JDistEn had the best ability to distinguish patients with VDCAS. The classification accuracy of severe CHD-mild-to-moderate CHD group, severe CHD-chest pain and normal coronary angiography (CPNCA) group, and mild-to-moderate CHD-CPNCA group were 0.8043, 0.7659, and 0.7500, respectively. The study indicates that the joint analysis of XSampEn, XFuzzyEn, and JDistEn can effectively capture the cardiac mechano-electric coupling information of patients with VDCAS, which can provide valuable information for clinicians to diagnose CHD.

Detection of Coronary Artery Disease Using Multi-Domain Feature Fusion of Multi-Channel Heart Sound Signals

Heart sound signals reflect valuable information about heart condition. Previous studies have suggested that the information contained in single-channel heart sound signals can be used to detect coronary artery disease (CAD). But accuracy based on single-channel heart sound signal is not satisfactory. This paper proposed a method based on multi-domain feature fusion of multi-channel heart sound signals, in which entropy features and cross entropy features are also included. A total of 36 subjects enrolled in the data collection, including 21 CAD patients and 15 non-CAD subjects. For each subject, five-channel heart sound signals were recorded synchronously for 5 min. After data segmentation and quality evaluation, 553 samples were left in the CAD group and 438 samples in the non-CAD group. The time-domain, frequency-domain, entropy, and cross entropy features were extracted. After feature selection, the optimal feature set was fed into the support vector machine for classification. The results showed that from single-channel to multi-channel, the classification accuracy has increased from 78.75% to 86.70%. After adding entropy features and cross entropy features, the classification accuracy continued to increase to 90.92%. The study indicated that the method based on multi-domain feature fusion of multi-channel heart sound signals could provide more information for CAD detection, and entropy features and cross entropy features played an important role in it.

Short-Term Effect of Percutaneous Coronary Intervention on Heart Rate Variability in Patients with Coronary Artery Disease

Myocardial ischemia in patients with coronary artery disease (CAD) leads to imbalanced autonomic control that increases the risk of morbidity and mortality. To systematically examine how autonomic function responds to percutaneous coronary intervention (PCI) treatment, we analyzed data of 27 CAD patients who had admitted for PCI in this pilot study. For each patient, five-minute resting electrocardiogram (ECG) signals were collected before and after the PCI procedure. The time intervals between ECG collection and PCI were both within 24 h. To assess autonomic function, normal sinus RR intervals were extracted and were analyzed quantitatively using traditional linear time- and frequency-domain measures [i.e., standard deviation of the normal-normal intervals (SDNN), the root mean square of successive differences (RMSSD), powers of low frequency (LF) and high frequency (HF) components, LF/HF] and nonlinear entropy measures [i.e., sample entropy (SampEn), distribution entropy (DistEn), and conditional entropy (CE)], as well as graphical metrics derived from Poincaré plot [i.e., Porta’s index (PI), Guzik’s index (GI), slope index (SI) and area index (AI)]. Results showed that after PCI, AI and PI decreased significantly (p < 0.002 and 0.015, respectively) with effect sizes of 0.88 and 0.70 as measured by Cohen’s d static. These changes were independent of sex. The results suggest that graphical AI and PI metrics derived from Poincaré plot of short-term ECG may be potential for sensing the beneficial effect of PCI on cardiovascular autonomic control. Further studies with bigger sample sizes are warranted to verify these observations.

Resting Heartbeat Complexity Predicts All-Cause and Cardiorespiratory Mortality in Middle- to Older-Aged Adults From the UK Biobank

Background

Spontaneous heart rate fluctuations contain rich information related to health and illness in terms of physiological complexity, an accepted indicator of plasticity and adaptability. However, it is challenging to make inferences on complexity from shorter, more practical epochs of data. Distribution entropy (DistEn) is a recently introduced complexity measure that is designed specifically for shorter duration heartbeat recordings. We hypothesized that reduced DistEn predicted increased mortality in a large population cohort.

Method and Results

The prognostic value of DistEn was examined in 7631 middle‐older–aged UK Biobank participants who had 2‐minute resting ECGs conducted (mean age, 59.5 years; 60.4% women). During a median follow‐up period of 7.8 years, 451 (5.9%) participants died. In Cox proportional hazards models with adjustment for demographics, lifestyle factors, physical activity, cardiovascular risks, and comorbidities, for each 1‐SD decrease in DistEn, the risk increased by 36%, 56%, and 73% for all‐cause, cardiovascular, and respiratory disease–related mortality, respectively. These effect sizes were equivalent to the risk of death from being >5 years older, having been a former smoker, or having diabetes mellitus. Lower DistEn was most predictive of death in those <55 years with a prior myocardial infarction, representing an additional 56% risk for mortality compared with older participants without prior myocardial infarction. These observations remained after controlling for traditional mortality predictors, resting heart rate, and heart rate variability.

Conclusions

Resting heartbeat complexity from short, resting ECGs was independently associated with mortality in middle‐ to older‐aged adults. These risks appear most pronounced in middle‐aged participants with prior MI, and may uniquely contribute to mortality risk screening.

Bivariate Entropy Analysis of Electrocardiographic RR-QT Time Series

QT interval variability (QTV) and heart rate variability (HRV) are both accepted biomarkers for cardiovascular events. QTV characterizes the variations in ventricular depolarization and repolarization. It is a predominant element of HRV. However, QTV is also believed to accept direct inputs from upstream control system. How QTV varies along with HRV is yet to be elucidated. We studied the dynamic relationship of QTV and HRV during different physiological conditions from resting, to cycling, and to recovering. We applied several entropy-based measures to examine their bivariate relationships, including cross sample entropy (XSampEn), cross fuzzy entropy (XFuzzyEn), cross conditional entropy (XCE), and joint distribution entropy (JDistEn). Results showed no statistically significant differences in XSampEn, XFuzzyEn, and XCE across different physiological states. Interestingly, JDistEn demonstrated significant decreases during cycling as compared with that during the resting state. Besides, JDistEn also showed a progressively recovering trend from cycling to the first 3 min during recovering, and further to the second 3 min during recovering. It appeared to be fully recovered to its level in the resting state during the second 3 min during the recovering phase. The results suggest that there is certain nonlinear temporal relationship between QTV and HRV, and that the JDistEn could help unravel this nuanced property.

Signal Information Prediction of Mortality Identifies Unique Patient Subsets after Severe Traumatic Brain Injury: A Decision-Tree Analysis Approach

Nonlinear physiological signal features that reveal information content and causal flow have recently been shown to be predictors of mortality after severe traumatic brain injury (TBI). The extent to which these features interact together, and with traditional measures to describe patients in a clinically meaningful way remains unclear. In this study, we incorporated basic demographics (age and initial Glasgow Coma Scale [GCS]) with linear and non-linear signal information based features (approximate entropy [ApEn], and multivariate conditional Granger causality [GC]) to evaluate their relative contributions to mortality using cardio-cerebral monitoring data from 171 severe TBI patients admitted to a single neurocritical care center over a 10 year period. Beyond linear modelling, we employed a decision tree analysis approach to define a predictive hierarchy of features. We found ApEn (p = 0.009) and GC (p = 0.004) based features to be independent predictors of mortality at a time when mean intracranial pressure (ICP) was not. Our combined model with both signal information-based features performed the strongest (area under curve = 0.86 vs. 0.77 for linear features only). Although low "intracranial" complexity (ApEn-ICP) outranked both age and GCS as crucial drivers of mortality (fivefold increase in mortality where ApEn-ICP <1.56, 36.2% vs. 7.8%), decision tree analysis revealed clear subsets of patient populations using all three predictors. Patients with lower ApEn-ICP who were >60 years of age died, whereas those with higher ApEn-ICP and GCS ≥5 all survived. Yet, even with low initial intracranial complexity, as long as patients maintained robust GC and "extracranial" complexity (ApEn of mean arterial pressure), they all survived. Incorporating traditional linear and novel, non-linear signal information features, particularly in a framework such as decision trees, may provide better insight into "health" status. However, caution is required when interpreting these results in a clinical setting prior to external validation.

Comparison of QT interval variability of coronary patients without myocardial infarction with that of patients with old myocardial infarction

BACKGROUND

The significant association of myocardial ischemia with elevated QT interval variability (QTV) has been reported in myocardial infarction (MI) patients. However, the influence of the time course of MI on QTV has not been investigated systematically.

METHOD

Short-term QT and RR interval time series were constructed from the 5 min electrocardiograms of 49 coronary patients without MI and 26 patients with old MI (OMI). The QTV, heart rate variability (HRV), and QT-RR coupling of the two groups were analyzed using various time series analysis tools in the time- and frequency-domains, as well as nonlinear dynamics.

RESULTS

Nearly all of the tested QTV indices for coronary patients with OMI were higher than those for patients without MI. However, no significant differences were found between the two groups in any of the variables employed to assess the HRV and QT-RR coupling. All of the markers that showed statistical significances in univariate analyses still possessed the capabilities of distinguishing between the two groups even after adjusting for studied baseline characteristics, including the coronary atherosclerotic burden.

CONCLUSIONS

The results suggested that the QTV increased in coronary patients with OMI compared to those without MI, which might reflect the influence of post-MI remodeling on the beat-to-beat temporal variability of ventricular repolarization. The non-significant differences in the HRV and QT-RR couplings could indicate that there were no differences in the modulation of the autonomic nervous system and interaction of QT with the RR intervals between the two groups.

Network analysis through the use of joint-distribution entropy on EEG recordings of MCI patients during a visual short-term memory binding task

The early diagnosis of Alzheimer's disease (AD) is particularly challenging. Mild cognitive impairment (MCI) has been linked to AD and electroencephalogram (EEG) recordings are able to measure brain activity directly with high temporal resolution. In this context, with appropriate processing, the EEG recordings can be used to construct a graph representative of brain functional connectivity. This work studies a functional network created from a non-linear measure of coupling of beta-filtered EEG recordings during a short-term memory binding task. It shows that the values of the small-world characteristic and eccentricity are, respectively, lower and higher in MCI patients than in controls. The results show how MCI leads to EEG functional connectivity changes. They expect that the network differences between MCIs and control subjects could be used to gain insight into the early stages of AD.

EZ Entropy: a software application for the entropy analysis of physiological time-series

Background

Entropy analysis has been attracting increasing attentions in the recent two or three decades. It assesses complexity, or irregularity, of time-series which is extraordinarily relevant to physiology and diseases as demonstrated by tremendous studies. However, the complexity can hardly be appreciated by traditional methods including time-, frequency-domain analysis, and time-frequency analysis that are the common built-in options in commercialized measurement and statistical software. To facilitate the entropy analysis of physiological time-series, a new software application, namely EZ Entropy, was developed and introduced in this article.

Results

EZ Entropy was developed in MATLAB^® environment. It was programmed in an object-oriented style and was constructed with a graphical user interface. EZ Entropy is easy to operate through its compact graphical interface, thus allowing researchers without knowledge of programming like clinicians and physiologists to perform such kind of analysis. Besides, it offers various settings to meet different analysis needs including (1) processing single data recording, (2) batch processing multiple data files, (3) sliding window calculations, (4) recall, (5) displaying intermediate data and final results, (6) adjusting input parameters, and (7) exporting calculation results after the run or in real-time during the analysis. The analysis results could be exported, either manually or automatically, to comma-separated ASCII files, thus being compatible to and easily imported into the common statistical analysis software. Code-wise, EZ Entropy is object-oriented, thus being quite easy to maintain and extend.

Conclusions

EZ Entropy is a user-friendly software application to perform the entropy analysis of time-series, as well as to simplify and to speed up this useful analysis.

Electronic supplementary material

The online version of this article (10.1186/s12938-019-0650-5) contains supplementary material, which is available to authorized users.

Automated Recognition of Epileptic EEG States Using a Combination of Symlet Wavelet Processing, Gradient Boosting Machine, and Grid Search Optimizer

Automatic recognition methods for non-stationary electroencephalogram (EEG) data collected from EEG sensors play an essential role in neurological detection. The integrated approaches proposed in this study consist of Symlet wavelet processing, a gradient boosting machine, and a grid search optimizer for a three-class classification scheme for normal subjects, intermittent epilepsy, and continuous epilepsy. Fourth-order Symlet wavelets are adopted to decompose the EEG data into five frequencies sub-bands, such as gamma, beta, alpha, theta, and delta, whose statistical features were computed and used as classification features. The grid search optimizer is used to automatically find the optimal parameters for training the classifier. The classification accuracy of the gradient boosting machine was compared with that of a conventional support vector machine and a random forest classifier constructed according to previous descriptions. Multiple performance indices were used to evaluate the proposed classification scheme, which provided better classification accuracy and detection effectiveness than has been recently reported in other studies on three-class classification of EEG data.

Area asymmetry of heart rate variability signal

BACKGROUND

Heart rate fluctuates beat-by-beat asymmetrically which is known as heart rate asymmetry (HRA). It is challenging to assess HRA robustly based on short-term heartbeat interval series.

METHOD

An area index (AI) was developed that combines the distance and phase angle information of points in the Poincaré plot. To test its performance, the AI was used to classify subjects with: (i) arrhythmia, and (ii) congestive heart failure, from the corresponding healthy controls. For comparison, the existing Porta's index (PI), Guzik's index (GI), and slope index (SI) were calculated. To test the effect of data length, we performed the analyses separately using long-term heartbeat interval series (derived from >3.6-h ECG) and short-term segments (with length of 500 intervals). A second short-term analysis was further carried out on series extracted from 5-min ECG.

RESULTS

For long-term data, SI showed acceptable performance for both tasks, i.e., for task i p < 0.001, Cohen's d = 0.93, AUC (area under the receiver-operating characteristic curve) = 0.86; for task ii p < 0.001, d = 0.88, AUC = 0.75. AI performed well for task ii (p < 0.001, d = 1.0, AUC = 0.78); for task i, though the difference was statistically significant (p < 0.001, AUC = 0.76), the effect size was small (d = 0.11). PI and GI failed in both tasks (p > 0.05, d < 0.4, AUC < 0.7 for all). However, for short-term segments, AI indicated better distinguishability for both tasks, i.e., for task i, p < 0.001, d = 0.71, AUC = 0.71; for task ii, p < 0.001, d = 0.93, AUC = 0.74. The rest three measures all failed with small effect sizes and AUC values (d < 0.5, AUC < 0.7 for all) although the difference in SI for task i was statistically significant (p < 0.001). Besides, AI displayed smaller variations across different short-term segments, indicating more robust performance. Results from the second short-term analysis were in keeping with those findings.

CONCLUSION

The proposed AI indicated better performance especially for short-term heartbeat interval data, suggesting potential in the ambulatory application of cardiovascular monitoring.