Early Detection of Sudden Cardiac Death by Using Ensemble Empirical Mode Decomposition-Based Entropy and Classical Linear Features From Heart Rate Variability Signals

Nonlinear Analysis of the Hand and Foot Force-Time Profiles in the Four Competitive Swimming Strokes

Human locomotion on water depends on the force produced by the swimmer to propel the body forward. Performance of highly complex motor tasks like swimming can yield minor variations that only nonlinear analysis can be sensitive enough to detect. The purpose of the present study was to examine the nonlinear properties of the hand/feet forces and describe their variations across the four competitive swimming strokes performing segmental and full-body swimming. Swimmers performed all-out bouts of 25 m in the four swimming strokes, swimming the full-body stroke, with the arm-pull only and with the leg kicking only. Hand/foot force and swimming velocity were measured. The Higuchi's fractal dimension (HFD) and sample entropy (SampEn) were used for the nonlinear analysis of force and velocity. Both the arm-pull and leg kicking alone were found to produce similar peak and mean hand/foot forces as swimming the full-body stroke. Hand force was more complex in breaststroke and butterfly stroke; conversely, kicking conditions were more complex in front crawl and backstroke. Moreover, the arm-pull and kicking alone tended to be more complex (higher HFD) but more predictable (lower SampEn) than while swimming the full-body stroke. There was no loss of force production from segmental swimming to the full-body counterpart. In conclusion, the number of segments in action influences the nonlinear behavior of the force produced and, when combining the four limbs, the complexity of the hand/foot force tends to decrease.

Heart Rate Variability Measurement through a Smart Wearable Device: Another Breakthrough for Personal Health Monitoring?

Heart rate variability (HRV) is a measurement of the fluctuation of time between each heartbeat and reflects the function of the autonomic nervous system. HRV is an important indicator for both physical and mental status and for broad-scope diseases. In this review, we discuss how wearable devices can be used to monitor HRV, and we compare the HRV monitoring function among different devices. In addition, we have reviewed the recent progress in HRV tracking with wearable devices and its value in health monitoring and disease diagnosis. Although many challenges remain, we believe HRV tracking with wearable devices is a promising tool that can be used to improve personal health.

Performance of heart rate adjusted heart rate variability for risk stratification of sudden cardiac death

PURPOSE

As a non-invasive tool for the assessment of cardiovascular autonomic function, the predictive value of heart rate variability (HRV) for sudden cardiac death (SCD) risk stratification remains unclear. In this study, we investigated the performance of the individualized heart rate (HR) adjusted HRV (HRV_I) for SCD risk stratification in subjects with diverse risks.

METHODS

A total of 11 commonly used HRV metrics were analyzed in 192 subjects, including 88 healthy controls (low risk group), 82 hypertrophic cardiomyopathy (HCM) patients (medium risk group), and 22 SCD victims (high risk group). The relationship between HRV metrics and HR was examined with long-term and short-term analysis. The performance HRV_I was evaluated by area under the receiver operating characteristic curve (AUC) and covariance of variation (CV).

RESULTS

Most of the HRV metrics were exponentially decayed with the increase of HR, while the exponential power coefficients were significantly different among groups. The HRV_I metrics discriminated low, medium and high risk subjects with a median AUC of 0.72[0.11], which was considerably higher than that of the traditional long-term (0.63[0.04]) and short-term (0.58[0.05]) HRV without adjustment. The average CV of the HRV_I metrics was also significantly lower than traditional short-term HRV metrics (0.09 ± 0.02 vs. 0.24 ± 0.13, p < 0.01).

CONCLUSIONS

Subjects with diverse risks of SCD had similar exponential decay relationship between HRV metrics and HR, but with different decaying rates. HRV_I provides reliable and robust estimation for risk stratification of SCD.

Machine learning of electrophysiological signals for the prediction of ventricular arrhythmias: systematic review and examination of heterogeneity between studies

BACKGROUND

Ventricular arrhythmia (VA) precipitating sudden cardiac arrest (SCD) is among the most frequent causes of death and pose a high burden on public health systems worldwide. The increasing availability of electrophysiological signals collected through conventional methods (e.g. electrocardiography (ECG)) and digital health technologies (e.g. wearable devices) in combination with novel predictive analytics using machine learning (ML) and deep learning (DL) hold potential for personalised predictions of arrhythmic events.

METHODS

This systematic review and exploratory meta-analysis assesses the state-of-the-art of ML/DL models of electrophysiological signals for personalised prediction of malignant VA or SCD, and studies potential causes of bias (PROSPERO, reference: CRD42021283464). Five electronic databases were searched to identify eligible studies. Pooled estimates of the diagnostic odds ratio (DOR) and summary area under the curve (AUROC) were calculated. Meta-analyses were performed separately for studies using publicly available, ad-hoc datasets, versus targeted clinical data acquisition. Studies were scored on risk of bias by the PROBAST tool.

FINDINGS

2194 studies were identified of which 46 were included in the systematic review and 32 in the meta-analysis. Pooling of individual models demonstrated a summary AUROC of 0.856 (95% CI 0.755-0.909) for short-term (time-to-event up to 72 h) prediction and AUROC of 0.876 (95% CI 0.642-0.980) for long-term prediction (time-to-event up to years). While models developed on ad-hoc sets had higher pooled performance (AUROC 0.919, 95% CI 0.867-0.952), they had a high risk of bias related to the re-use and overlap of small ad-hoc datasets, choices of ML tool and a lack of external model validation.

INTERPRETATION

ML and DL models appear to accurately predict malignant VA and SCD. However, wide heterogeneity between studies, in part due to small ad-hoc datasets and choice of ML model, may reduce the ability to generalise and should be addressed in future studies.

FUNDING

This publication is part of the project DEEP RISK ICD (with project number 452019308) of the research programme Rubicon which is (partly) financed by the Dutch Research Council (NWO). This research is partly funded by the Amsterdam Cardiovascular Sciences (personal grant F.V.Y.T).

Automated Detection of Sudden Cardiac Death by Discrete Wavelet Transform of Electrocardiogram Signal

Sudden cardiac death (SCD) results in millions of deaths annually; as it is a fatal heart abnormality, early prediction of SCD could save peoples’ lives to the greatest extent. Symmetry and asymmetry play an important role in many fields. Electrocardiograms (ECG) as a noninvasive process for acquiring the electrical activity of the heart, has both asymmetric and non-stationary characteristics; it is frequently employed to diagnose and evaluate the heart’s condition. In this work, we have detected SCD 14 min (separately for each one-minute interval) prior to its occurrence by analyzing ECG signals using discrete wavelet transform (DWT) and locality preserving projection (LPP). In the experiment, we have performed DWT on ECG signals to obtain coefficients, then LPP as a reduction methodology was used to cut down these obtained coefficients. Then, the acquired LPP features were ranked using various methods, including the T-test, Bhattacharyya, Wilcoxon, and entropy. At last, the highly ranked LPP features were subjected to decision tree, k-nearest neighbor (KNN), and support vector machine classifiers for distinguishing normal from SCD ECG signals. Our proposed technique has achieved a highest accuracy of 97.6% for the detection of SCD 14 min prior using the KNN classifier, compared to the existing works. Our proposed method is capable of predicting the people at risk of developing SCD 14 min before its onset, and, hence, clinicians would have enough time to provide treatment in intensive care units (ICU) for a subject at risk of SCD. Thus, this proposed technique as a useful tool can increase the survival rate of many cardiac patients.

CRPMKB: a knowledge base of cancer risk prediction models for systematic comparison and personalized applications

MOTIVATION

In the era of big data and precision medicine, accurate risk assessment is a prerequisite for the implementation of risk screening and preventive treatment. A large number of studies have focused on the risk of cancer, and related risk prediction models have been constructed, but there is a lack of effective resource integration for systematic comparison and personalized applications. Therefore, the establishment and analysis of the cancer risk prediction model knowledge base (CRPMKB) is of great significance.

RESULTS

The current knowledge base contains 802 model data. The model comparison indicates that the accuracy of cancer risk prediction was greatly affected by regional differences, cancer types and model types. We divided the model variables into four categories: environment, behavioral lifestyle, biological genetics and clinical examination, and found that there are differences in the distribution of various variables among different cancer types. Taking 50 genes involved in the lung cancer risk prediction models as an example to perform pathway enrichment analyses and the results showed that these genes were significantly enriched in p53 Signaling and Aryl Hydrocarbon Receptor Signaling pathways which are associated with cancer and specific diseases. In addition, we verified the biological significance of overlapping lung cancer genes via STRING database. CRPMKB was established to provide researchers an online tool for the future personalized model application and developing. This study of CRPMKB suggests that developing more targeted models based on specific demographic characteristics and cancer types will further improve the accuracy of cancer risk model predictions.

AVAILABILITY AND IMPLEMENTATION

CRPMKB is freely available at http://www.sysbio.org.cn/CRPMKB/. The data underlying this article are available in the article and in its online supplementary material.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Optimal Classification of Atrial Fibrillation and Congestive Heart Failure Using Machine Learning

Cardiovascular disorders, including atrial fibrillation (AF) and congestive heart failure (CHF), are the significant causes of mortality worldwide. The diagnosis of cardiovascular disorders is heavily reliant on ECG signals. Therefore, extracting significant features from ECG signals is the most challenging aspect of representing each condition of ECG signal. Earlier studies have claimed that the Hjorth descriptor is assigned as a simple feature extraction algorithm capable of class separation among AF, CHF, and normal sinus rhythm (NSR) conditions. However, due to noise interference, certain features do not represent the characteristics of the ECG signals. This study addressed this critical gap by applying the discrete wavelet transform (DWT) to decompose the ECG signals into sub-bands and extracting Hjorth descriptor features and entropy-based features in the DWT domain. Therefore, the calculation of Hjorth descriptor and entropy-based features performed on each sub-band will produce more detailed information of ECG signals. The optimization of various classifier algorithms, including k-nearest neighbor (k-NN), support vector machine (SVM), random forest (RF), artificial neural network (ANN), and radial basis function network (RBFN), was investigated to provide the best system performance. This study obtained an accuracy of 100% for the k-NN, SVM, RF, and ANN classifiers, respectively, and 97% for the RBFN classifier. The results demonstrated that the optimization of the classifier algorithm could improve the classification accuracy of AF, CHF, and NSR conditions, compared to earlier studies.

HFBD: a biomarker knowledge database for heart failure heterogeneity and personalized applications

MOTIVATION

Heart failure (HF) is a cardiovascular disease with a high incidence around the world. Accumulating studies have focused on the identification of biomarkers for HF precision medicine. To understand the HF heterogeneity and provide biomarker information for the personalized diagnosis and treatment of HF, a knowledge database collecting the distributed and multiple-level biomarker information is necessary.

RESULTS

In this study, the HF biomarker knowledge database (HFBD) was established by manually collecting the data and knowledge from literature in PubMed. HFBD contains 2618 records and 868 HF biomarkers (731 single and 137 combined) extracted from 1237 original articles. The biomarkers were classified into proteins, RNAs, DNAs, and the others at molecular, image, cellular and physiological levels. The biomarkers were annotated with biological, clinical and article information as well as the experimental methods used for the biomarker discovery. With its user-friendly interface, this knowledge database provides a unique resource for the systematic understanding of HF heterogeneity and personalized diagnosis and treatment of HF in the era of precision medicine.

AVAILABILITY

The platform is openly available at http://sysbio.org.cn/HFBD/.

The feasibility of predicting impending malignant ventricular arrhythmias by using nonlinear features of short heartbeat intervals

BACKGROUND AND OBJECTIVE

Malignant ventricular arrhythmias (MAs) occur unpredictably and lead to emergencies. A new approach that uses a timely tracking device e.g., photoplethysmogram (PPG) solely to predict MAs would be irreplaceably valuable and it is natural to expect the approach can predict the occurrence as early as possible.

METHOD

We assumed that with an appropriate metric based on signal complexity, the heartbeat interval time series (HbIs) can be used to manifest the intrinsic characteristics of the period immediately precedes the MAs (preMAs). The approach first characterizes the patterns of preMAs by a new complexity metric (the refined composite multi-scale entropy). The MAs detector is then constructed by checking the discriminability of the MAs against the sinus rhythm and other prevalent arrhythmias (atrial fibrillation and premature ventricular contraction) of three machine-learning models (SVM, Random Forest, and XGboost).

RESULTS

Two specifications are of interest: the length of the HbIs needed to delineate the preMAs patterns sufficiently (l_spec) and how long before the occurrence of MAs will the HbIs manifest specific patterns that are distinct enough to predict the impending MAs (t_spec). Our experimental results confirmed the best performance came from a Random-Forest model with an average precision of 99.99% and recall of 88.98% using a HbIs of 800 heartbeats (the l_spec), 108 seconds (the t_spec) before the occurrence of MAs.

CONCLUSION

By experimental validation of the unique pattern of the preMAs in HbIs and using it in the machine learning model, we showed the high possibility of MAs prediction in a broader circumstance, which may cover daily healthcare using the alternative sensor in HbIs monitoring. Therefore, this research is theoretically and practically significant in cardiac arrest prevention.

Measures of repolarization variability predict ventricular arrhythmogenesis in heptanol-treated Langendorff-perfused mouse hearts

BACKGROUND

Time-domain and non-linear methods can be used to quantify beat-to-beat repolarization variability but whether measures of repolarization variability can predict ventricular arrhythmogenesis in mice have never been explored.

METHODS

Left ventricular monophasic action potentials (MAPs) were recorded during constant right ventricular 8 ​Hz pacing in Langendorff-perfused mouse hearts, in the presence or absence of the gap junction and sodium channel inhibitor heptanol (0.1, 0.5, 1 or 2 ​mM).

RESULTS

Under control conditions, mean action potential duration (APD) was 39.4 ​± ​8.1 ​ms. Standard deviation (SD) of APDs was 0.3 ​± ​0.2 ​ms, coefficient of variation was 0.9 ​± ​0.8% and the root mean square (RMS) of successive differences in APDs was 0.15 ​± ​0.14 ​ms. Poincaré plots of APDn+1 against APDn revealed ellipsoid morphologies with a SD along the line-of-identity (SD2) to SD perpendicular to the line-of-identity (SD1) ratio of 4.6 ​± ​2.1. Approximate and sample entropy were 0.61 ​± ​0.12 and 0.76 ​± ​0.26, respectively. Detrended fluctuation analysis revealed short- and long-term fluctuation slopes of 1.49 ​± ​0.27 and 0.81 ​± ​0.36, respectively. Heptanol at 2 ​mM induced ventricular tachycardia in five out of six hearts. None of the above parameters were altered by heptanol during which reproducible electrical activity was observed (KW-ANOVA, P ​> ​0.05). Contrastingly, SD2/SD1 decreased to 2.03 ​± ​0.41, approximate and sample entropy increased to 0.82 ​± ​0.12 and 1.45 ​± ​0.34, and short-term fluctuation slope decreased to 0.82 ​± ​0.19 during the 20-s period preceding spontaneous ventricular tachy-arrhythmias (n ​= ​6, KW-ANOVA, P ​< ​0.05).

CONCLUSION

Measures of repolarization variability, such as SD2/SD1, entropy, and fluctuation slope are altered preceding the occurrence of ventricular arrhythmogenesis in mouse hearts. Changes in these variables may allow detection of impending arrhythmias for early intervention.

Effects of Cigarette Smoking on Cardiac Autonomic Responses: A Cross-Sectional Study

It has been suggested that some of the adverse, long-term cardiovascular outcomes of smoking are mediated by impaired autonomic nervous system (ANS) activity. Yet, this association is currently inconclusive. Heart rate variability (HRV) and the deep breathing test (DBT) represent common quantitative markers of ANS activity due to their simplicity and reliability. This large cross-sectional study was designed to assess the effect of active smoking on ANS function as manifested by HRV or DBT abnormalities. Electrocardiograms were recorded at rest for 5 min and during forced metronomic breathing. HRV and DBT were calculated according to accepted standards. Participants were divided into two groups based on current smoking status. The study included 242 healthy volunteers (196 nonsmokers and 46 smokers). There were no significant differences in age, sex, and BMI between groups. Cumulative smoking exposure burden (CSEB) for the study group was 5.3 ± 1.3 pack-years. Comparative analysis of HRV and DBT parameters according to smoking status revealed no significant differences between groups. Significant (p < 0.05), yet weak or moderate correlations (r < 0.7) were found between CSEB and abnormal change in HRV parameters consistent with sympathetic overactivity and decreased parasympathetic tone. In conclusion, smoking for a relatively short period in healthy adults does not seem to lead to significant impairment in ANS function. Yet, the consequences of smoking seem to be amplified when cumulative exposure burden increases.