Influence of recognition errors of computerised analysis of 24-hour electrocardiograms on the measurement of spectral components of heart rate variability

A REVIEW OF ECG-BASED DIAGNOSIS SUPPORT SYSTEMS FOR OBSTRUCTIVE SLEEP APNEA

Humans need sleep. It is important for physical and psychological recreation. During sleep our consciousness is suspended or least altered. Hence, our ability to avoid or react to disturbances is reduced. These disturbances can come from external sources or from disorders within the body. Obstructive Sleep Apnea (OSA) is such a disorder. It is caused by obstruction of the upper airways which causes periods where the breathing ceases. In many cases, periods of reduced breathing, known as hypopnea, precede OSA events. The medical background of OSA is well understood, but the traditional diagnosis is expensive, as it requires sophisticated measurements and human interpretation of potentially large amounts of physiological data. Electrocardiogram (ECG) measurements have the potential to reduce the cost of OSA diagnosis by simplifying the measurement process. On the down side, detecting OSA events based on ECG data is a complex task which requires highly skilled practitioners. Computer algorithms can help to detect the subtle signal changes which indicate the presence of a disorder. That approach has the following advantages: computers never tire, processing resources are economical and progress, in the form of better algorithms, can be easily disseminated as updates over the internet. Furthermore, Computer-Aided Diagnosis (CAD) reduces intra- and inter-observer variability. In this review, we adopt and support the position that computer based ECG signal interpretation is able to diagnose OSA with a high degree of accuracy.

Measurement fidelity of heart rate variability signal processing: the devil is in the details

Heart rate variability (HRV) is a particularly valuable quantitative marker of the flexibility and balance of the autonomic nervous system. Significant advances in software programs to automatically derive HRV have led to its extensive use in psychophysiological research. However, there is a lack of systematic comparisons across software programs used to derive HRV indices. Further, researchers report meager details on important signal processing decisions making synthesis across studies challenging. The aim of the present study was to evaluate the measurement fidelity of time- and frequency-domain HRV indices derived from three predominant signal processing software programs commonly used in clinical and research settings. Triplicate ECG recordings were derived from 20 participants using identical data acquisition hardware. Among the time-domain indices, there was strong to excellent correspondence (ICC(avg)=0.93) for SDNN, SDANN, SDNNi, rMSSD, and pNN50. The frequency-domain indices yielded excellent correspondence (ICC(avg)=0.91) for LF, HF, and LF/HF ratio, except for VLF which exhibited poor correspondence (ICC(avg)=0.19). Stringent user-decisions and technical specifications for nuanced HRV processing details are essential to ensure measurement fidelity across signal processing software programs.

Automatic filtering of outliers in RR intervals before analysis of heart rate variability in Holter recordings: a comparison with carefully edited data

BACKGROUND

Undetected arrhythmic beats seriously affect the power spectrum of the heart rate variability (HRV). Therefore, the series of RR intervals are normally carefully edited before HRV is analysed, but this is a time consuming procedure when 24-hours recordings are analysed. Alternatively, different methods can be used for automatic removal of arrhythmic beats and artefacts. This study compared common frequency domain indices of HRV when determined from manually edited and automatically filtered RR intervals.

METHODS AND RESULTS

Twenty-four hours Holter recordings were available from 140 healthy subjects of age 1-75 years. An experienced technician carefully edited all recordings. Automatic filtering was performed using a recursive procedure where RR intervals were removed if they differed from the mean of the surrounding RR intervals with more than a predetermined limit (ranging from 10% to 50%). The filtering algorithm was evaluated by replacing 1% of the beats with synthesised ectopic beats. Power spectral analysis was performed before and after filtering of both the original edited data and the noisy data set. The results from the analysis using the noisy data were used to define an age-based filtering threshold. The age-based filtration was evaluated with completely unedited data, generated by removing all annotations from the series of RR intervals, and then comparing the resulting HRV indices with those obtained using edited data. The results showed equivalent results after age-based filtration of both the edited and unedited data sets, where the differences in HRV indices obtained by different preprocessing methods were small compared to the mean values within each age group.

CONCLUSIONS

The study showed that it might not be necessary to perform the time-consuming careful editing of all detected heartbeats before HRV is analysed in Holter recordings.In most subjects, it is sufficient to perform the regular editing needed for valid arrhythmia analyses, and then remove undetected ectopic beats and artefacts by age-based filtration of the series of RR intervals, particularly in subjects older than 30 years.

Heart rate variability as a measure of autonomic regulation of cardiac activity for assessing stress and welfare in farm animals -- a review

Measurement of heart rate variability (HRV) is a non-invasive technique that can be used to investigate the functioning of the autonomic nervous system, especially the balance between sympathetic and vagal activity. It has been proven to be very useful in humans for both research and clinical studies concerned with cardiovascular diseases, diabetic autonomic dysfunction, hypertension and psychiatric and psychological disorders. Over the past decade, HRV has been used increasingly in animal research to analyse changes in sympathovagal balance related to diseases, psychological and environmental stressors or individual characteristics such as temperament and coping strategies. This paper discusses current and past HRV research in farm animals. First, it describes how cardiac activity is regulated and the relationships between HRV, sympathovagal balance and stress and animal welfare. Then it proceeds to outline the types of equipment and methodological approaches that have been adapted and developed to measure inter-beats intervals (IBI) and estimate HRV in farm animals. Finally, it discusses experiments and conclusions derived from the measurement of HRV in pigs, cattle, horses, sheep, goats and poultry. Emphasis has been placed on deriving recommendations for future research investigating HRV, including approaches for measuring and analysing IBI data. Data from earlier research demonstrate that HRV is a promising approach for evaluating stress and emotional states in animals. It has the potential to contribute much to our understanding and assessment of the underlying neurophysiological processes of stress responses and different welfare states in farm animals.

Validation of a cardiac monitor for measuring heart rate variability in adult female pigs: accuracy, artefacts and editing

Autonomic regulation of cardiac activity during stress has not been clearly defined in farm animals. In part, this is due to the limited availability of affordable ambulatory cardiac monitors capable of accurately monitoring and storing large amounts of data that meet the criteria necessary for heart rate variability analysis. Our objectives were to measure the accuracy of a 24-h Polar RR monitor using gold standard ECG, to examine and categorise any occurring anomalies and to ascertain their impact on the outcome of heart rate variability analysis. Five 1-year-old female pigs (gilts) were socially isolated from their pen mates and cardiac activity was simultaneously measured using two systems, a 24-h Polar RR Recorder and a Telemetric ECG system. The Polar data were manually assessed both against and in isolation of the ECG data to identify anomalous beats, which were then assigned to one of five identified error categories. The anomalies in the Polar data were corrected and statistical comparisons were performed among the three data sets to evaluate the effects of anomalies on heart rate variability analysis. Bland-Altman analysis was used to measure the level of agreement among the ECG, Uncorrected Polar and Corrected Polar data. No anomalies or ectopies were found in the ECG data but 46 anomalies (0.81% of total interbeat intervals [IBI]) were found in the Polar Uncorrected data. Manual identification and editing procedures reduced this error to 0.018%. Most mean heart rate and IBI parameters were unaffected by error (P>.05). Standard deviation (S.D.) and root mean square of successive differences (RMSSD) were 45% and 50% higher when anomalies were present in the data. Artefacts affected the magnitude of the frequency domain indices and overestimated total and parasympathetic activity and underestimated sympathetic activity. The mean difference between ECG and Uncorrected Polar data was 1.36 ms (limits of agreement -69.03 to 71.74 ms). This was greatly improved to 0.36 ms (limits of agreement -5.37 to 6.10 ms) after editing. Overall, even a small proportion of error biased the outcome of heart rate variability analysis. This bias was greatly reduced by correcting the anomalous beats. Bland-Altman analysis demonstrated that when there was error present in the Polar data, it could not be used interchangeably with the ECG data. However, if there were no anomalies present in the data or if they were classified and corrected using the approach in this study, then the two systems could be used interchangeably.

Automatic measurement of long-term heart rate variability by implanted single-chamber devices

Heart rate variability (HRV) measurement is an established technology for the assessment of cardiac autonomic status. Recently 24 h HRV has been shown to correlate with disease severity in heart failure. This potentially makes continuous 24 h HRV measurement suitable for monitoring of heart-failure patients. Day-to-day 24 h measurement of HRV is, in principle, feasible when implemented using implanted devices (pacemakers and defibrillators) used in patients who are predominantly in the sinus rhythm. However, a number of such devices used in heart-failure patients are single-chamber devices, in which the distinction between sinus rhythm beats and ectopic beats is problematic. The study investigates whether a reasonably accurate 24 h HRV measurement can be achieved by automatic algorithms, suitable for implementation using implanted devices, without the need for identification of ectopic beats. A set of 5321 nominal 24 h Holter recordings of cardiac patients are used. Each of the recordings contains at least one ectopic beat; approximately 30% of the recordings have more than 1% of ectopic beats. Conventional 24 h measures of HRV, that is the SDNN, HRV index, and SDANN indices, are obtained from each recording after elimination of the ectopic beats and are approximated by HRV measures computed by the same formulas without exclusion of the ectopic beats. The SDANN values are also approximated by the standard deviation of 5 min medians of all RR intervals (SDMRR measure). The errors introduced by including the ectopic beats in the HRV computation were evaluated using the Bland-Altman statistics and by Cohen's kappa statistics investigating the precision of identifying patients with depressed and preserved 24 h HRV. The SDNN measure is very sensitive to the quality of the RR interval sequence and cannot be reasonably used without distinction between sinus rhythm and ectopic beats. The HRV index measure is marginally more acceptable when used without ectopic elimination. The SDANN is rather insensitive, and its replacement by SDMRR values leads to relative errors in the region of 2-5% that are almost independent of the number of ectopic beats included. Even in recordings with a substantial proportion of ectopic beats, a practically acceptable (kappa > 0.9) identification of depressed and preserved SDANN values is possible without ectopic elimination. Thus, continuous monitoring of 24 h HRV is technically feasible within implanted devices, provided the SDANN measure is monitored and either computed from the sequence of all RR intervals or, potentially preferably, replaced by the SDMRR measure.

Heart rate variability: origins, methods, and interpretive caveats

Components of heart rate variability have attracted considerable attention in psychology and medicine and have become important dependent measures in psychophysiology and behavioral medicine. Quantification and interpretation of heart rate variability, however, remain complex issues and are fraught with pitfalls. The present report (a) examines the physiological origins and mechanisms of heart rate variability, (b) considers quantitative approaches to measurement, and (c) highlights important caveats in the interpretation of heart rate variability. Summary guidelines for research in this area are outlined, and suggestions and prospects for future developments are considered.

Preservation of autonomic function following successful reperfusion with streptokinase within 12 hours of the onset of acute myocardial infarction

Successful reperfusion following thrombolysis results in increased heart rate variability in the first 24 hours after administration. Preservation of autonomic function may contribute to improved prognosis when coronary artery patency is restored with intravenous thrombolysis.

Measurement of parasympathetic activity from 24-hour ambulatory electrocardiograms and its reproducibility and sensitivity in normal subjects, patients with symptomatic myocardial ischemia, and patients with diabetes mellitus

The parasympathetic nervous system plays a major role in the pathophysiology of many cardiovascular disease, particularly in modulating myocardial electrical stability. Measurements of heart rate variability have been widely used to assess parasympathetic activity. The reproducibility of measurements obtained from 24-hour ambulatory electrocardiograms has not been well documented. We have developed a technique for measuring parasympathetic activity from clinical quality 24-hour ambulatory electrocardiograms by counting beat-to-beat increases in RR interval that are > 50 ms. To determine the reproducibility and sensitivity of our technique, we analyzed repeated 24-hour electrocardiograms of 173 subjects (19 normal subjects, 67 patients with ischemic heart disease, and 87 diabetics) followed up over periods of 2 to 16 weeks. In all subject groups, mean values for repeated measurements were virtually identical. Measurements were stable in all 3 groups throughout the course of the study, as assessed by intraclass correlation coefficients. This technique is sensitive enough to detect relatively small changes in parasympathetic activity in subjects, as demonstrated by the calculated Bland and Altman coefficients of repeatability. Reproducibility and sensitivity of our technique are particularly good in normal subjects and in patients with ischemic heart disease. The results obtained with this technique imply that other related measurements of parasympathetic activity will show similar excellent short- and long-term reproducibility and sensitivity.