Automatic ectopic beat elimination in short-term heart rate variability measurement

Automation of classification of sleep stages and estimation of sleep efficiency using actigraphy

Introduction

Sleep is a fundamental and essential physiological process for recovering physiological function. Sleep disturbance or deprivation has been known to be a causative factor of various physiological and psychological disorders. Therefore, sleep evaluation is vital for diagnosing or monitoring those disorders. Although PSG (polysomnography) has been the gold standard for assessing sleep quality and classifying sleep stages, PSG has various limitations for common uses. In substitution for PSG, there has been vigorous research using actigraphy.

Methods

For classifying sleep stages automatically, we propose machine learning models with HRV (heart rate variability)-related features and acceleration features, which were processed from the actigraphy (Maxim band) data. Those classification results were transformed into a binary classification for estimating sleep efficiency. With 30 subjects, we conducted PSG, and they slept overnight with wrist-type actigraphy. We assessed the performance of four proposed machine learning models.

Results

With HRV-related and raw features of actigraphy, Cohen's kappa was 0.974 (p < 0.001) for classifying sleep stages into five stages: wake (W), REM (Rapid Eye Movement) (R), Sleep N1 (Non-Rapid Eye Movement Stage 1, S1), Sleep N2 (Non-Rapid Eye Movement Stage 2, S2), Sleep N3 (Non-Rapid Eye Movement Stage 3, S3). In addition, our machine learning model for the estimation of sleep efficiency showed an accuracy of 0.86.

Discussion

Our model demonstrated that automated sleep classification results could perfectly match the PSG results. Since models with acceleration features showed modest performance in differentiating some sleep stages, further research on acceleration features must be done. In addition, the sleep efficiency model demonstrated modest results. However, an investigation into the effects of HRV-derived and acceleration features is required.

Synergy through integration of digital cognitive tests and wearable devices for mild cognitive impairment screening

Introduction

Advances in mobile computing platforms and the rapid development of wearable devices have made possible the continuous monitoring of patients with mild cognitive impairment (MCI) and their daily activities. Such rich data can reveal more subtle changes in patients' behavioral and physiological characteristics, providing new ways to detect MCI anytime, anywhere. Therefore, we aimed to investigate the feasibility and validity of digital cognitive tests and physiological sensors applied to MCI assessment.

Methods

We collected photoplethysmography (PPG), electrodermal activity (EDA) and electroencephalogram (EEG) signals from 120 participants (61 MCI patients, 59 healthy controls) during rest and cognitive testing. The features extracted from these physiological signals involved the time domain, frequency domain, time-frequency domain and statistics. Time and score features during the cognitive test are automatically recorded by the system. In addition, selected features of all modalities were classified by tenfold cross-validation using five different classifiers.

Results

The experimental results showed that the weighted soft voting strategy combining five classifiers achieved the highest classification accuracy (88.9%), precision (89.9%), recall (88.2%), and F1 score (89.0%). Compared to healthy controls, the MCI group typically took longer to recall, draw, and drag. Moreover, during cognitive testing, MCI patients showed lower heart rate variability, higher electrodermal activity values, and stronger brain activity in the alpha and beta bands.

Discussion

It was found that patients' classification performance improved when combining features from multiple modalities compared to using only tablet parameters or physiological features, indicating that our scheme could reveal MCI-related discriminative information. Furthermore, the best classification results on the digital span test across all tasks suggest that MCI patients may have deficits in attention and short-term memory that came to the fore earlier. Finally, integrating tablet cognitive tests and wearable sensors would provide a new direction for creating an easy-to-use and at-home self-check MCI screening tool.

A Novel Adaptive Noise Elimination Algorithm in Long RR Interval Sequences for Heart Rate Variability Analysis

As heart rate variability (HRV) studies become more and more prevalent in clinical practice, one of the most common and significant causes of errors is associated with distorted RR interval (RRI) data acquisition. The nature of such artifacts can be both mechanical as well as software based. Various currently used noise elimination in RRI sequences methods use filtering algorithms that eliminate artifacts without taking into account the fact that the whole RRI sequence time cannot be shortened or lengthened. Keeping that in mind, we aimed to develop an artifacts elimination algorithm suited to long-term (hours or days) sequences that does not affect the overall structure of the RRI sequence and does not alter the duration of data registration. An original adaptive smart time series step-by-step analysis and statistical verification methods were used. The adaptive algorithm was designed to maximize the reconstruction of the heart-rate structure and is suitable for use, especially in polygraphy. The authors submit the scheme and program for use.

ECG-iCOVIDNet: Interpretable AI model to identify changes in the ECG signals of post-COVID subjects

OBJECTIVE

Studies showed that many COVID-19 survivors develop sub-clinical to clinical heart damage, even if subjects did not have underlying heart disease before COVID. Since Electrocardiogram (ECG) is a reliable technique for cardiovascular disease diagnosis, this study analyzes the 12-lead ECG recordings of healthy and post-COVID (COVID-recovered) subjects to ascertain ECG changes after suffering from COVID-19.

METHOD

We propose a shallow 1-D convolutional neural network (CNN) deep learning architecture, namely ECG-iCOVIDNet, to distinguish ECG data of post-COVID subjects and healthy subjects. Further, we employed ShAP technique to interpret ECG segments that are highlighted by the CNN model for the classification of ECG recordings into healthy and post-COVID subjects.

RESULTS

ECG data of 427 healthy and 105 post-COVID subjects were analyzed. Results show that the proposed ECG-iCOVIDNet model could classify the ECG recordings of healthy and post-COVID subjects better than the state-of-the-art deep learning models. The proposed model yields an F₁-score of 100%.

CONCLUSION

So far, we have not come across any other study with an in-depth ECG signal analysis of the COVID-recovered subjects. In this study, it is shown that the shallow ECG-iCOVIDNet CNN model performed good for distinguishing ECG signals of COVID-recovered subjects from those of healthy subjects. In line with the literature, this study confirms changes in the ECG signals of COVID-recovered patients that could be captured by the proposed CNN model. Successful deployment of such systems can help the doctors identify the changes in the ECG of the post-COVID subjects on time that can save many lives.

Dysautonomia following COVID-19 is not associated with subjective limitations or symptoms but is associated with objective functional limitations

Background

Individuals who contract coronavirus disease 2019 (COVID-19) can suffer with persistent and debilitating symptoms long after the initial acute illness. Heart rate (HR) profiles determined during cardiopulmonary exercise testing (CPET) and delivered as part of a post-COVID recovery service may provide insight into the presence and impact of dysautonomia on functional ability.

Objective

Using an active, working-age, post–COVID-19 population, the purpose of this study was to (1) determine and characterize any association between subjective symptoms and dysautonomia; and (2) identify objective exercise capacity differences between patients classified “with” and those “without” dysautonomia.

Methods

Patients referred to a post–COVID-19 service underwent comprehensive clinical assessment, including self-reported symptoms, CPET, and secondary care investigations when indicated. Resting HR >75 bpm, HR increase with exercise <89 bpm, and HR recovery <25 bpm 1 minute after exercise were used to define dysautonomia. Anonymized data were analyzed and associations with symptoms, and CPET outcomes were determined.

Results

Fifty-one of the 205 patients (25%) reviewed as part of this service evaluation had dysautonomia. There were no associations between symptoms or perceived functional limitation and dysautonomia (P >.05). Patients with dysautonomia demonstrated objective functional limitations with significantly reduced work rate (219 ± 37 W vs 253 ± 52 W; P <.001) and peak oxygen consumption (V̇o₂: 30.6 ± 5.5 mL/kg/min vs 35.8 ± 7.6 mL/kg/min; P <.001); and a steeper (less efficient) V̇e/V̇co₂ slope (29.9 ± 4.9 vs 27.7 ± 4.7; P = .005).

Conclusion

Dysautonomia is associated with objective functional limitations but is not associated with subjective symptoms or limitation.

FLIRT: A feature generation toolkit for wearable data

BACKGROUND AND OBJECTIVE

Researchers use wearable sensing data and machine learning (ML) models to predict various health and behavioral outcomes. However, sensor data from commercial wearables are prone to noise, missing, or artifacts. Even with the recent interest in deploying commercial wearables for long-term studies, there does not exist a standardized way to process the raw sensor data and researchers often use highly specific functions to preprocess, clean, normalize, and compute features. This leads to a lack of uniformity and reproducibility across different studies, making it difficult to compare results. To overcome these issues, we present FLIRT: A Feature Generation Toolkit for Wearable Data; it is an open-source Python package that focuses on processing physiological data specifically from commercial wearables with all its challenges from data cleaning to feature extraction.

METHODS

FLIRT leverages a variety of state-of-the-art algorithms (e.g., particle filters, ML-based artifact detection) to ensure a robust preprocessing of physiological data from wearables. In a subsequent step, FLIRT utilizes a sliding-window approach and calculates a feature vector of more than 100 dimensions - a basis for a wide variety of ML algorithms.

RESULTS

We evaluated FLIRT on the publicly available WESAD dataset, which focuses on stress detection with an Empatica E4 wearable. Preprocessing the data with FLIRT ensures that unintended noise and artifacts are appropriately filtered. In the classification task, FLIRT outperforms the preprocessing baseline of the original WESAD paper.

CONCLUSION

FLIRT provides functionalities beyond existing packages that can address unmet needs in physiological data processing and feature generation: (a) integrated handling of common wearable file formats (e.g., Empatica E4 archives), (b) robust preprocessing, and (c) standardized feature generation that ensures reproducibility of results. Nevertheless, while FLIRT comes with a default configuration to accommodate most situations, it offers a highly configurable interface for all of its implemented algorithms to account for specific needs.

Healthy Lifestyle and Cardiac Vagal Modulation Over 10 Years: Whitehall II Cohort Study

Background Increased vagal modulation is a mechanism that may partially explain the protective effect of healthy lifestyles. However, it is unclear how healthy lifestyles relate to vagal regulation longitudinally. We prospectively examined associations between a comprehensive measure of 4 important lifestyle factors and vagal modulation, indexed by heart rate variability (HRV) over 10 years. Methods and Results The fifth (1997-1999), seventh (2002-2004), and ninth (2007-2009) phases of the UK Whitehall II cohort were analyzed. Analytical samples ranged from 2059 to 3333 (mean age: 55.7 years). A healthy lifestyle score was derived by giving participants 1 point for each healthy factor: physically active, not smoking, moderate alcohol consumption, and healthy body mass index. Two vagally mediated HRV measures were used: high-frequency HRV and root mean square of successive differences of normal-to-normal R-R intervals. Cross-sectionally, a positively graded association was observed between the healthy lifestyle score and HRV at baseline (P_overall≤0.001). Differences in HRV according to the healthy lifestyle score remained relatively stable over time. Compared with participants who hardly ever adhered to healthy lifestyles, those with consistent healthy lifestyles displayed higher high-frequency HRV (β=0.23; 95% CI, 0.10-0.35; P=0.001) and higher root mean square of successive differences of normal-to-normal R-R intervals (β=0.15; 95% CI, 0.07-0.22; P≤0.001) at follow-up after covariate adjustment. These differences in high-frequency HRV and root mean square of successive differences of normal-to-normal R-R intervals are equivalent to ≈6 to 20 years differences in chronological age. Compared with participants who reduced their healthy lifestyle scores, those with stable scores displayed higher subsequent high-frequency HRV (β=0.24; 95% CI, 0.01-0.48; P=0.046) and higher root mean square of successive differences of normal-to-normal R-R intervals (β=0.15; 95% CI, 0.01-0.29; P=0.042). Conclusions Maintaining healthy lifestyles is positively associated with cardiac vagal functioning, and these beneficial adaptations may be lost if not sustained.

Heart rate variability and depressive symptoms: a cross-lagged analysis over a 10-year period in the Whitehall II study

BACKGROUND

People with depression tend to have lower heart rate variability (HRV), but the temporal sequence is poorly understood. In a sample of the general population, we prospectively examined whether HRV measures predict subsequent depressive symptoms or whether depressive symptoms predict subsequent levels of HRV.

METHOD

Data from the fifth (1997-1999) and ninth (2007-2009) phases of the UK Whitehall II longitudinal population-based cohort study were analysed with an average follow-up of 10.5 years. The sample size for the prospective analysis depended on the analysis and ranged from 2334 (644 women) to 2276 (602 women). HRV measures during 5 min of supine rest were obtained. Depressive symptoms were evaluated by four cognitive symptoms of depression from the General Health Questionnaire.

RESULTS

At follow-up assessment, depressive symptoms were inversely associated with HRV measures independently of antidepressant medication use in men but not in women. Prospectively, lower baseline heart rate and higher HRV measures were associated with a lower likelihood of incident depressive symptoms at follow-up in men without depressive symptoms at baseline. Similar but statistically insignificant associations were found in women. Adjustments for known confounders including sociodemographic and lifestyle factors, cardiometabolic conditions or medication did not change the predictive effect of HRV on incident depressive symptoms at follow-up. Depressive symptoms at baseline were not associated with heart rate or HRV at follow-up in either sex.

CONCLUSIONS

These findings are consistent with an aetiological role of the autonomic nervous system in depression onset.

Heart rate-guided, but not dose-guided titration of beta blockers stabilizes ventricular repolarization in patients with chronic heart failure

AIMS

We compared the effects of heart rate-guided and dose-guided beta-blocker titration strategies on QT variability in patients with chronic heart failure (CHF).

METHODS

In a prospective study we recorded 5-minute resting high-resolution ECGs (HRECG) in 100 patients with CHF and measured heart rate (HR) and ventricular repolarization by QT variability index (QTVI). In a subgroup of patients not reaching target HR (<70bpm) we uptitrated beta blockers and repeated HRECG measurements 3months thereafter.

RESULTS

Target HR was present in 46 patients (group A), and in 54 patients HR was above target (group B). The groups did not differ in age, gender, NYHA class, NT pro-BNP, creatinine, or beta blocker dose. Patients in group A displayed significantly lower QTVI than patients in group B (-1.25±0.55 vs. -1.52±0.42, P=0.013). When uptitrating beta-blockers we found a decrease in HR (from 91±15bpm to 71±15bpm, P<0.001), NTpro BNP levels (from 4474±3878pg/ml to 3042±2566pg/ml, P=0.024), and NYHA class (from 3.0±0.8 to 2.5±0.7, P=0.006). With beta-blocker uptitration QTVI decreased in 10 of 24 patients (42%). In these patients HR decreased more than in the remaining cohort (-25±20bpm vs. -15±17bpm, P=0.017). On multivariate analysis, the presence of target HR was a predictor of QTVI decrease (P=0.017), but beta-blocker dose was not.

CONCLUSIONS

In patients with CHF treated by beta-blockers, changes in QT variability appear to occur in parallel with changes of heart rate. This suggests that heart rate-guided titration of beta-blockers may be associated with decreased risk of sudden cardiac death.

Are Changes in Heart Rate Variability in Middle-Aged and Older People Normative or Caused by Pathological Conditions? Findings From a Large Population-Based Longitudinal Cohort Study

Background

No study to date has investigated longitudinal trajectories of cardiac autonomic modulation changes with aging; therefore, we lack evidence showing whether these changes occur naturally or are secondary to disease or medication use. This study tested whether heart rate variability (HRV) trajectories from middle to older age are largely normative or caused by pathological changes with aging in a large prospective cohort. We further assessed whether HRV changes were modified by socioeconomic status, ethnicity, or habitual physical activity.

Methods and Results

This study involved 3176 men and 1238 women initially aged 44 to 69 years (1997–1999) from the UK Whitehall II population‐based cohort. We evaluated time‐ and frequency‐domain HRV measures of short‐term recordings at 3 time points over a 10‐year period. Random mixed models with time‐varying covariates were applied. Cross‐sectionally, HRV measures were lower for men than for women, for participants with cardiometabolic conditions, and for participants reporting use of medications other than beta blockers. Longitudinally, HRV measures decreased significantly with aging in both sexes, with faster decline in younger age groups. HRV trajectories were not explained by increased prevalence of cardiometabolic problems and/or medication use. In women, cardiometabolic problems were associated with faster decline in the standard deviation of all intervals between R waves with normal‐to‐normal conduction, in low‐frequency HRV, and in low‐frequency HRV in normalized units. Socioeconomic status, ethnicity, and habitual physical activity did not have significant effects on HRV trajectories.

Conclusions

Our investigation showed a general pattern and timing of changes in indices of cardiac autonomic modulation from middle to older age. These changes seem likely to reflect the normal aging process rather than being secondary to cardiometabolic problems and medication use.

A preprocessing tool for removing artifact from cardiac RR interval recordings using three-dimensional spatial distribution mapping

Artifact is common in cardiac RR interval data that is recorded for heart rate variability (HRV) analysis. A novel algorithm for artifact detection and interpolation in RR interval data is described. It is based on spatial distribution mapping of RR interval magnitude and relationships to adjacent values in three dimensions. The characteristics of normal physiological RR intervals and artifact intervals were established using 24-h recordings from 20 technician-assessed human cardiac recordings. The algorithm was incorporated into a preprocessing tool and validated using 30 artificial RR (ARR) interval data files, to which known quantities of artifact (0.5%, 1%, 2%, 3%, 5%, 7%, 10%) were added. The impact of preprocessing ARR files with 1% added artifact was also assessed using 10 time domain and frequency domain HRV metrics. The preprocessing tool was also used to preprocess 69 24-h human cardiac recordings. The tool was able to remove artifact from technician-assessed human cardiac recordings (sensitivity 0.84, SD = 0.09, specificity of 1.00, SD = 0.01) and artificial data files. The removal of artifact had a low impact on time domain and frequency domain HRV metrics (ranging from 0% to 2.5% change in values). This novel preprocessing tool can be used with human 24-h cardiac recordings to remove artifact while minimally affecting physiological data and therefore having a low impact on HRV measures of that data.

The recursive combination filter approach of pre-processing for the estimation of standard deviation of RR series

Variation in the interval between the R-R peaks of the electrocardiogram represents the modulation of the cardiac oscillations by the autonomic nervous system. This variation is contaminated by anomalous signals called ectopic beats, artefacts or noise which mask the true behaviour of heart rate variability. In this paper, we have proposed a combination filter of recursive impulse rejection filter and recursive 20% filter, with recursive application and preference of replacement over removal of abnormal beats to improve the pre-processing of the inter-beat intervals. We have tested this novel recursive combinational method with median method replacement to estimate the standard deviation of normal to normal (SDNN) beat intervals of congestive heart failure (CHF) and normal sinus rhythm subjects. This work discusses the improvement in pre-processing over single use of impulse rejection filter and removal of abnormal beats for heart rate variability for the estimation of SDNN and Poncaré plot descriptors (SD1, SD2, and SD1/SD2) in detail. We have found the 22 ms value of SDNN and 36 ms value of SD2 descriptor of Poincaré plot as clinical indicators in discriminating the normal cases from CHF cases. The pre-processing is also useful in calculation of Lyapunov exponent which is a nonlinear index as Lyapunov exponents calculated after proposed pre-processing modified in a way that it start following the notion of less complex behaviour of diseased states.

The effects of cigarette smoking on aerobic and anaerobic capacity and heart rate variability among female university students

Aim

In this study, the effects of cigarette smoking on maximal aerobic capacity, anaerobic capacity, and heart rate variability among female university students were investigated.

Materials and methods

Twelve smokers and 21 nonsmokers participated in this study. All participants performed an intermittent sprint test (IST) and a 20 m shuttle run test to measure their anaerobic capacity and maximal aerobic capacity. The IST was comprised of 6 × 10-second sprints with a 60-second active recovery between each sprint. Heart rate variability was recorded while the participants were in a supine position 20 minutes before and 30 minutes after the IST.

Results

The total work, peak power, and heart rate of the smokers and nonsmokers did not differ significantly. However, the smokers’ average power declined significantly during sprints 4 to 6 (smokers versus nonsmokers, respectively: 95% confidence interval =6.2–7.2 joule/kg versus 6.8–7.6 joule/kg; P<0.05), and their fatigue index increased (smokers versus nonsmokers, respectively: 35.8% ± 2.3% versus 24.5% ± 1.76%; P<0.05) during the IST. The maximal oxygen uptake of nonsmokers was significantly higher than that of the smokers (P<0.05). The standard deviation of the normal to normal intervals and the root mean square successive difference did not differ significantly between nonsmokers and smokers. However, the nonsmokers exhibited a significantly higher normalized high frequency (HF), and significantly lower normalized low frequency (LF), LF/HF ratio, and natural logarithm of the LF/HF when compared with those of the smokers (P<0.05).

Conclusion

Smoking may increase female smokers’ exercise fatigue and decrease their average performance during an IST, while reducing their maximal aerobic capacity. Furthermore, smoking reduces parasympathetic nerve activity and activates sympathetic cardiac control.

Ventricular activity morphological characterization: ectopic beats removal in long term atrial fibrillation recordings

Ectopic beats are early heart beats remarkably different to the normal beat morphology that provoke serious disturbances in electrocardiographic analysis. These beats are very common in atrial fibrillation (AF), causing important residua when ventricular activity has to be removed for atrial activity (AA) analysis. In this work, a method is proposed to cancel out ectopics by discriminating between normal and abnormal beats, with an accuracy higher than 99%, through QRS morphological delineation and characterization. The most similar ectopics to the one under cancellation are clustered to provide a very precise cancellation template. Simulated and real AF recordings were used to validate the method. A new index, able to estimate the presence of ventricular residue after ectopics cancellation, was defined. Results by using the 2, 4, 6, …, 30 most similar ectopics to the one under study yielded optimal cancellation for templates composed of 10 beats. Furthermore, these beats were very likely located close to the ectopic under cancellation, which could facilitate the algorithm implementation. As conclusion, the proposed method is an effective way to remove ectopics from long term AF recordings and get them ready for the application of any QRST cancellation technique able to extract the AA in optimal conditions. Moreover, it could also detect, characterize and remove ectopics in any other type of non-AF recordings.

Detection and removal of ventricular ectopic beats in atrial fibrillation recordings via principal component analysis

Ectopic beats are early heart beats with remarkable large amplitude that provoke serious disturbances in the analysis of electrocardiograms (ECG). These beats are very common in atrial fibrillation (AF) and are the source of important residua when the QRST is intended to be removed. Given that QRST cancellation is a binding step in the appropriate analysis of atrial activity (AA) in AF, a method for ventricular ectopic beats cancellation is proposed as a previous step to the application of any QRST removal technique. First, the method discriminates between normal and ectopic beats with an accuracy higher than 99% through QRS morphological characterization. Next, the most similar ectopic beats to the one under cancellation are clustered and serve to get their eigenvector matrix by principal component analysis. Finally, the highest variance eigenvector is used as cancellation template. The reduction ectopic rate (RER) has been defined to evaluate the method's performance by using templates generated with 5, 10, 20, 40 or 80 ectopics. Optimal results were reached with the 5 most similar complexes, yielding a RER higher than 5.5. In addition, a decreasing RER trend was noticed as the number of considered ectopics for cancellation increased. As conclusion, given that ectopics presented a remarkable variability in their morphology, the proposed cancellation approach is a robust ectopic remover and can notably facilitate the later application of any QRST cancellation technique to extract the AA in the best conditions.

An efficient method of addressing ectopic beats: new insight into data preprocessing of heart rate variability analysis

Heart rate variability (HRV) analysis is affected by ectopic beats. An efficient method was proposed to deal with the ectopic beats. The method was based on trend correlation of the heart timing signal. Predictor of R-R interval (RRI) value at ectopic beat time was constructed by the weight calculation and the slope estimation of preceding normal RRI. The type of ectopic beat was detected and replaced by the predictor of RRI. The performance of the simulated signal after ectopic correction was tested by the standard value using power spectrum density (PSD) estimation, whereas the results of clinical data with ectopic beats were compared with the adjacent ectopic-free data. The result showed the frequency indexes after ectopy corrected had less error than other methods with the test of simulated signal and clinical data. It indicated our method could improve the PSD estimation in HRV analysis. The method had advantages of high accuracy and real time properties to recover the sinus node modulation.

Examining cross-database global training to evaluate five different methods for ventricular beat classification

The detection of ventricular beats in the holter recording is a task of great importance since it can direct clinicians toward the parts of the electrocardiogram record that might be crucial for determining the final diagnosis. Although there already exists a fair amount of research work dealing with ventricular beat detection in holter recordings, the vast majority uses a local training approach, which is highly disputable from the point of view of any practical-real-life-application. In this paper, we compare five well-known methods: a classical decision tree approach and its variant with fuzzy rules, a self-organizing map clustering method with template matching for classification, a back-propagation neural network and a support vector machine classifier, all examined using the same global cross-database approach for training and testing. For this task two databases were used-the MIT-BIH database and the AHA database. Both databases are required for testing any newly developed algorithms for holter beat classification that is going to be deployed in the EU market. According to cross-database global training, when the classifier is trained with the beats from the records of one database then the records from the other database are used for testing. The results of all the methods are compared and evaluated using the measures of sensitivity and specificity. The support vector machine classifier is the best classifier from the five we tested, achieving an average sensitivity of 87.20% and an average specificity of 91.57%, which outperforms nearly all the published algorithms when applied in the context of a similar global training approach.

Development of QRS detection algorithm designed for wearable cardiorespiratory system

An in-home sleep monitoring system was developed previously in our laboratory for monitoring electrocardiography (ECG) and respiratory signals. However, the ECG signal acquired with this system is prone to high-grade noise caused by motion artifact. Since the detection of the QRS complexes with high accuracy is very important in a computer-based analysis of the ECG, a high accuracy QRS detection algorithm is developed and based on the combination of heart rate indicators and morphological ECG features. The proposed algorithm is tested both on 16 h data acquired using the two sensors of our cardiorespiratory belt system, i.e., the polyvinylidene fluoride (PVDF) film and the conductive fabric sheets, and on all 48 records of the MIT/BIH Arrhythmia Database. Satisfying results are obtained for both databases, the sensitivity S(e) and positive predictivity P(+) were calculated for each case and results show S(e)=[96.98%, 93.76%] and P(+)=[97.81%, 99.48%] for conductive fabric and PVDF film sensors, respectively, and S(e)=99.77% and P(+)=99.64% in the case of the MIT/BIH Arrhythmia Database. Further, heart rate variability (HRV) measures were calculated using our system and a commercial system. A comparison between systems' results is done to show the usefulness of our developed algorithm used with our cardiorespiratory belt sensor.

Does Autonomic Function Link Social Position to Coronary Risk?

Background— Laboratory and clinical studies suggest that the autonomic nervous system responds to chronic behavioral and psychosocial stressors with adverse metabolic consequences and that this may explain the relation between low social position and high coronary risk. We sought to test this hypothesis in a healthy occupational cohort.

Methods and Results— This study comprised 2197 male civil servants 45 to 68 years of age in the Whitehall II study who were undergoing standardized assessments of social position (employment grade) and the psychosocial, behavioral, and metabolic risk factors for coronary disease previously found to be associated with low social position. Five-minute recordings of heart rate variability (HRV) were used to assess cardiac parasympathetic function (SD of N-N intervals and high-frequency power [0.15 to 0.40 Hz]) and the influence of sympathetic and parasympathetic function (low-frequency power [0.04 to 0.15 Hz]). Low employment grade was associated with low HRV (age-adjusted trend for each modality, P ≤0.02). Adverse behavioral factors (smoking, exercise, alcohol, and diet) and psychosocial factors (job control) showed age-adjusted associations with low HRV ( P <0.03). The age-adjusted mean low-frequency power was 319 ms 2 among those participants in the bottom tertile of job control compared with 379 ms 2 in the other participants ( P =0.004). HRV showed strong ( P <0.001) linear associations with components of the metabolic syndrome (waist circumference, systolic blood pressure, HDL cholesterol, triglycerides, and fasting and 2-hour postload glucose). The social gradient in prevalence of metabolic syndrome was explained statistically by adjustment for low-frequency power, behavioral factors, and job control.

Conclusions— Chronically impaired autonomic function may link social position to different components of coronary risk in the general population.

Functional assessment of heart rate variability: physiological basis and practical applications

The autonomic nervous system dynamically controls the response of the body to a range of external and internal stimuli, providing physiological stability in the individual. With the progress of information technology, it is now possible to explore the functioning of this system reliably and non-invasively using comprehensive and functional analysis of heart rate variability. This method is already an established tool in cardiology research, and is increasingly being used for a range of clinical applications. This review describes the theoretical basis and practical applications for this emerging technique.