Dataset Information

Cepstral Analysis for Scoring the Quality of Electrocardiograms for Heart Rate Variability.

ABSTRACT: Mobile-health solutions based on heart rate variability often require electrocardiogram (ECG) recordings by inexperienced operators or real-time automatic analyses of long-term recordings by wearable devices in free-moving individuals. In this context, it is useful to associate a quality index with the ECG, scoring the adequacy of the recording for heart rate variability to identify noise or arrhythmias. Therefore, this work aims to propose and validate a computational method for assessing the adequacy of single-lead ECGs for heart rate variability analysis that may run in real time on wearable systems with low computational power. The method quantifies the ECG pseudo-periodic structure employing cepstral analysis. The cepstrum (spectrum of log-spectrum) is estimated on a running ECG window of 10 s before and after "liftering" (filtering in the cepstral domain) to remove slower noise components. The ECG periodicity generates a dominant peak in the liftered cepstrum at the "quefrency" of the mean cardiac interval. The Cepstral Quality Index (CQI) is the ratio between the cepstral-peak power and the total power of the unliftered cepstrum. Noises and arrhythmias reduce the relative power of the cepstral peak decreasing CQI. We analyzed a public dataset of 6072 single-lead ECGs manually classified in normal rhythm or inadequate for heart rate variability analysis because of noise or atrial fibrillation, and the CQI = 47% cut-off identified the inadequate recordings with 79% sensitivity and 85% specificity. We showed that the performance is independent of the lead considering a public dataset of 1,000 12-lead recordings with quality classified as "acceptable" or "unacceptable" by visual inspection. Thus, the cepstrum describes the ECG periodic structure effectively and concisely and CQI appears to be a robust score of the adequacy of ECG recording for heart rate variability analysis, evaluable in real-time on wearable devices.

SUBMITTER: Castiglioni P

PROVIDER: S-EPMC9247307 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background: Performance of wrist actigraphy in assessing sleep not only depends on the sensor technology of the actigraph hardware but also on the attributes of the interpretative algorithm (IA). The objective of our research was to improve assessment of sleep quality, relative to existing IAs, through development of a novel IA using deep learning methods, utilizing as input activity count and heart rate variability (HRV) metrics of different window length (number of epochs of data). Methods: Simultaneously recorded polysomnography (PSG) and wrist actigraphy data of 222 participants were utilized. Classic deep learning models were applied to: (a) activity count alone (without HRV), (b) activity count + HRV (30-s window), (c) activity count + HRV (3-min window), and (d) activity count + HRV (5-min window) to ascertain the best set of inputs. A novel deep learning model (Haghayegh Algorithm, HA), founded on best set of inputs, was developed, and its sleep scoring performance was then compared with the most popular University of California San Diego (UCSD) and Actiwatch proprietary IAs. Results: Activity count combined with HRV metrics calculated per 5-min window produced highest agreement with PSG. HA showed 84.5% accuracy (5.3-6.2% higher than comparator IAs), 89.5% sensitivity (6.2% higher than UCSD IA and 6% lower than Actiwatch proprietary IA), 70.0% specificity (8.2-34.3% higher than comparator IAs), and 58.7% Kappa agreement (16-23% higher than comparator IAs) in detecting sleep epochs. HA did not differ significantly from PSG in deriving sleep parameters-sleep efficiency, total sleep time, sleep onset latency, and wake after sleep onset; moreover, bias and mean absolute error of the HA model in estimating them was less than the comparator IAs. HA showed, respectively, 40.9% and 54.0% Kappa agreement with PSG in detecting rapid and non-rapid eye movement (REM and NREM) epochs. Conclusions: The HA model simultaneously incorporating activity count and HRV metrics calculated per 5-min window demonstrates significantly better sleep scoring performance than existing popular IAs.

Project description:ObjectiveTo evaluate the heart rate variability (HRV) in patients with multibacillary leprosy using dynamic linear and nonlinear analysis.Material and methodsTwenty-one leprosy patients (mean age: 39.14 ±10.58 years) and 21 healthy subjects (mean age: 36.24 ± 12.64 years) completed the sample. Heart rate variability recording was performed by a Polar RS800 CX heart monitor during a period of 15 min in the supine position and 15 min in a sitting position. Analysis of HRV was performed by frequency domain from high frequency (HF) and low frequency (LF) spectral indexes in absolute and normalized units. The nonlinear analysis of HRV was calculated using symbolic analysis (0V%, 1V%, 2LV% and 2UV% indexes), Shannon entropy (SE) and normalized complexity index (NCI).ResultsLinear analysis: both groups showed higher HF values (p < 0.05) and smaller LF values (p < 0.05) in supine than in sitting position. The leprosy patients showed higher LF values (p < 0.05) and smaller HF values (p < 0.05) compared to the controls on supine position. Symbolic analysis: leprosy patients had higher 0V% values (p < 0.05), smaller 2LV% values (p < 0.05) and 2UV % values compared to healthy subjects on both positions. The 1V % had higher values (p < 0.05) for leprosy patients than for controls in the sitting position. The control subjects had smaller 0V % values (p < 0.05), and higher 2UV % values (p < 0.05) in the supine position compared to the sitting position. Leprosy patients had higher 2UV index values (p < 0.05) in the supine position compared to the sitting position. In the complexity analysis, leprosy patients had smaller SE and NCI values (p < 0.05) than the control in the supine position. There was no difference between the SE and NCI values of leprosy and the control subjects in the sitting position. The control subjects had higher SE and NCI values (p < 0.05) in the supine position than in the sitting position.ConclusionLeprosy patients had higher sympathetic modulation and smaller vagal modulation than controls, indicating less HRV and cardiac modulation with lower complexity. The control group displayed significant HRV differences in response to position changes while leprosy patients had fewer HRV differences after the same postural change. An analysis of HRV with linear and non-linear dynamics proved to be a reliable method and promising for the investigation of autonomic dysfunction in patients with multibacillary leprosy.

Project description:Heart rate variability (HRV) is the variation of cardiac inter-beat intervals over time resulting largely from the interplay between the sympathetic and parasympathetic branches of the autonomic nervous system. Individual differences in HRV are associated with emotion regulation, personality, psychopathology, cardiovascular health, and mortality. Previous studies have shown significant heritability of HRV measures. Here we extend genetic research on HRV by investigating sex differences in genetic underpinnings of HRV, the degree of genetic overlap among different measurement domains of HRV, and phenotypic and genetic relationships between HRV and the resting heart rate (HR). We performed electrocardiogram (ECG) recordings in a large population-representative sample of young adult twins (n=1060 individuals) and computed HRV measures from three domains: time, frequency, and nonlinear dynamics. Genetic and environmental influences on HRV measures were estimated using linear structural equation modeling of twin data. The results showed that variability of HRV and HR measures can be accounted for by additive genetic and non-shared environmental influences (AE model), with no evidence for significant shared environmental effects. Heritability estimates ranged from 47 to 64%, with little difference across HRV measurement domains. Genetic influences did not differ between genders for most variables except the square root of the mean squared differences between successive R-R intervals (RMSSD, higher heritability in males) and the ratio of low to high frequency power (LF/HF, distinct genetic factors operating in males and females). The results indicate high phenotypic and especially genetic correlations between HRV measures from different domains, suggesting that >90% of genetic influences are shared across measures. Finally, about 40% of genetic variance in HRV was shared with HR. In conclusion, both HR and HRV measures are highly heritable traits in the general population of young adults, with high degree of genetic overlap across different measurement domains.

Dataset Information

Cepstral Analysis for Scoring the Quality of Electrocardiograms for Heart Rate Variability.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure