A framework for modeling performers' beat-to-beat heart intervals using music features and Interpretation Maps.

Objective: Music strongly modulates our autonomic nervous system. This modulation is evident in musicians' beat-to-beat heart (RR) intervals, a marker of heart rate variability (HRV), and can be related to music features and structures. We present a novel approach to modeling musicians' RR interval variations, analyzing detailed components within a music piece to extract continuous music features and annotations of musicians' performance decisions. Methods: A professional ensemble (violinist, cellist, and pianist) performs Schubert's Trio No. 2, Op. 100, Andante con moto nine times during rehearsals. RR interval series are collected from each musician using wireless ECG sensors. Linear mixed models are used to predict their RR intervals based on music features (tempo, loudness, note density), interpretive choices (Interpretation Map), and a starting factor. Results: The models explain approximately half of the variability of the RR interval series for all musicians, with R-squared = 0.606 (violinist), 0.494 (cellist), and 0.540 (pianist). The features with the strongest predictive values were loudness, climax, moment of concern, and starting factor. Conclusions: The method revealed the relative effects of different music features on autonomic response. For the first time, we show a strong link between an interpretation map and RR interval changes. Modeling autonomic response to music stimuli is important for developing medical and non-medical interventions. Our models can serve as a framework for estimating performers' physiological reactions using only music information that could also apply to listeners.

An approach to the detection of pain from autonomic and cortical correlates.

OBJECTIVE: To assess the value of combining brain and autonomic measures to discriminate the subjective perception of pain from other sensory-cognitive activations. METHODS: 20 healthy individuals received 2 types of tonic painful stimulation delivered to the hand: electrical stimuli and immersion in 10 Celsius degree (°C) water, which were contrasted with non-painful immersion in 15 °C water, and stressful cognitive testing. High-density electroencephalography (EEG) and autonomic measures (pupillary, electrodermal and cardiovascular) were continuously recorded, and the accuracy of pain detection based on combinations of electrophysiological features was assessed using machine learning procedures. RESULTS: Painful stimuli induced a significant decrease in contralateral EEG alpha power. Cardiac, electrodermal and pupillary reactivities occurred in both painful and stressful conditions. Classification models, trained on leave-one-out cross-validation folds, showed low accuracy (61-73%) of cortical and autonomic features taken independently, while their combination significantly improved accuracy to 93% in individual reports. CONCLUSIONS: Changes in cortical oscillations reflecting somatosensory salience and autonomic changes reflecting arousal can be triggered by many activating signals other than pain; conversely, the simultaneous occurrence of somatosensory activation plus strong autonomic arousal has great probability of reflecting pain uniquely. SIGNIFICANCE: Combining changes in cortical and autonomic reactivities appears critical to derive accurate indexes of acute pain perception.

[LSTM-XGBoost Based RR Intervals Time Series Prediction Method in Hypertensive Patients].

Objective: The prediction of RR intervals in hypertensive patients can help clinicians to analyze and warn patients' heart condition. Methods: Using 8 patients' data as samples, the RR intervals of patients were predicted by long short-term memory network (LSTM) and gradient lift tree (XGBoost), and the prediction results of the two models were combined by the inverse variance method to overcome the disadvantage of single model prediction. Results: Compared with the single model, the proposed combined model had a different degree of improvement in the prediction of RR intervals in 8 patients. Conclusion: LSTM-XGBoost model provides a method for predicting RR intervals in hypertensive patients, which has potential clinical feasibility.

Relationship of Cycling Power and Non-Linear Heart Rate Variability from Everyday Workout Data: Potential for Intensity Zone Estimation and Monitoring.

The short-term scaling exponent alpha1 of detrended fluctuation analysis (DFA-a1) of heart rate variability (HRV) has been shown to be a sensitive marker for assessing global organismic demands. The wide dynamic range within the exercise intensity spectrum and the relationship to established physiologic threshold boundaries potentially allow in-field use and also open opportunities to provide real-time feedback. The present study expands the idea of using everyday workout data from the AI Endurance app to obtain the relationship between cycling power and DFA-a1. Collected data were imported between September 2021 and August 2023 with an initial pool of 3123 workouts across 21 male users. The aim of this analysis was to further apply a new method of implementing workout group data considering representative values of DFA-a1 segmentation compared to single workout data and including all data points to enhance the validity of the internal-to-external load relationship. The present data demonstrate a universal relationship between cycling power and DFA-a1 from everyday workout data that potentially allows accessible and regular tracking of intensity zone demarcation information. The analysis highlights the superior efficacy of the representative-based approach of included data in most cases. Validation data of the performance level and the up-to-date relationship are still pending.

Arrhythmia detection in inter-patient ECG signals using entropy rate features and RR intervals with CNN architecture.

The classification of inter-patient ECG data for arrhythmia detection using electrocardiogram (ECG) signals presents a significant challenge. Despite the recent surge in deep learning approaches, there remains a noticeable gap in the performance of inter-patient ECG classification. In this study, we introduce an innovative approach for ECG classification in arrhythmia detection by employing a 1D convolutional neural network (CNN) to leverage both morphological and temporal characteristics of cardiac cycles. Through the utilization of 1D-CNN layers, we automatically capture the morphological attributes of ECG data, allowing us to represent the shape of the ECG waveform around the R peaks. Additionally, we incorporate four RR interval features to provide temporal context, and we explore the potential application of entropy rate as a feature extraction technique for ECG signal classification. Consequently, the classification layers benefit from the combination of both temporal and learned features, leading to the achievement of the final arrhythmia classification. We validate our approach using the MIT-BIH arrhythmia dataset, employing both intra-patient and inter-patient paradigms for model training and testing. The model's generalization ability is assessed by evaluating it on the INCART dataset. The model attains average accuracy rates of 99.13% and 99.17% for 2-fold and 5-fold cross-validation, respectively, in intra-patient classification with five classes. In inter-patient classification with three and five classes, the model achieves average accuracies of 98.73% and 97.91%, respectively. For the INCART dataset, the model achieves an average accuracy of 98.20% for three classes. The experimental outcomes demonstrate the superiority of the proposed model compared to state-of-the-art models in recognizing arrhythmias. Thus, the proposed model exhibits enhanced generalization and the potential to serve as an effective solution for recognizing arrhythmias in real-world datasets characterized by class imbalances in practical applications.

Predictive value of heart rate variability and electrochemical skin conductance measurements for cardiovascular autonomic neuropathy persistence in type 2 diabetes and prediabetes: A 3-year follow-up study.

OBJECTIVE: The study aimed to explore risk stratification approaches for cardiovascular autonomic neuropathy (CAN) in individuals with prediabetes and type 2 diabetes (T2DM) over a three-year follow-up period. METHODS: Participants underwent evaluations of autonomic function encompassing cardiovascular autonomic reflex tests (CARTs), baroreflex sensitivity (BRS), heart rate variability (HRV) in time domains (standard deviation of all normal RR intervals (SDNN)) and frequency domains (high frequency/low frequency ratio), and electrochemical skin conductance (ESC). The diagnosis of CAN relied on abnormal CART results. Subjects were categorized into 4 groups, based on their assessment of cardiac autonomic function at 3-year follow-up, relative to the presence or absence of CAN at baseline assessment: Persistent absence of CAN; Resolution of CAN; Progression to CAN; and Persistent CAN. RESULTS: Participants with T2DM/prediabetes (n = 91/7) were categorized as: Persistent absence of CAN (n = 25), Resolution of CAN (n = 10), Progression to CAN (n = 18), and Persistent CAN (n = 45) groups. The Persistent absence of CAN group showed significant associations with SDNN. The Resolution of CAN group exhibited notable associations with mean HbA1C (follow-up), while the Progression to CAN group displayed a significant link with baseline estimated glomerular filtration rate. The Persistent CAN group demonstrated significant associations with SDNN and Sudoscan CAN risk score. Screening recommendations involve biennial to annual assessments based on risk levels, aiding in CAN detection and subsequent comprehensive and time-intensive autonomic function tests for confirmation. The study's findings offer improved risk categorization approaches for detecting CAN, which has relevance for shaping public health strategies.

Evaluation of the PSO Metaheuristic Algorithm in Different Types of Sleep Apnea Diagnosis Using RR Intervals.

Background: Sleep apnea is one of the most common sleep disorders that facilitating and accelerating its diagnosis will have positive results on its future trend. Objective: This study aimed to diagnosis the sleep apnea types using the optimized neural network. Material and Methods: This descriptive-analytical study was done on 50 cases of patients referred to the sleep clinic of Imam Khomeini Hospital in Tehran, including 11 normal, 13 mild, 17 moderate and 9 severe cases. At the first, the data were pre-processed in three stages, then The Electrocardiogram (ECG) signal was decomposed to 8 levels using wavelet transform convert and 6 nonlinear features for the coefficients of this level and 10 features were calculated for RR Intervals. For apnea categorizing classes, the multilayer perceptron neural network was used with the backpropagation algorithm. For optimizing Multi-layered Perceptron (MLP) weights, the Particle Swarm Optimization (PSO) evolutionary optimization algorithm was used. Results: The simulation results show that the accuracy criterion in the MLP network is allied with the Backpropagation (BP) training algorithm for different types of apnea. By optimizing the weights in the MLP network structure, the accuracy criterion for modes normal, obstructive, central, mixed was obtained %96.86, %97.48, %96.23, and %96.44, respectively. These values indicate the strength of the evolutionary algorithm in improving the evaluation criteria and network accuracy. Conclusion: Due to the growth of knowledge and the complexity of medical decisions in the diagnosis of the disease, the use of artificial neural network algorithms can be useful to support this decision.

Validity of the Pneumonitor for RR intervals acquisition for short-term heart rate variability analysis extended with respiratory data in pediatric cardiac patients.

BACKGROUND: Breathing pattern alterations change the variability and spectral content of the RR intervals (RRi) on electrocardiogram (ECG). However, there is no method to record and control participants' breathing without influencing its natural rate and depth in heart rate variability (HRV) studies. AIM: This study aimed to assess the validity of the Pneumonitor for acquisition of short-term (5 minutes) RRi in comparison to the reference ECG method for analysis of heart rate (HR) and HRV parameters in the group of pediatric patients with cardiac disease. METHODS: Nineteen patients of both sexes participated in the study. An ECG and Pneumonitor were used to record RRi in 5-minute static rest conditions, the latter also to measure the relative tidal volume and respiratory rate. The validation comprised Student's t-test, Bland-Altman analysis, intraclass correlation coefficient, and Lin's concordance correlation. The possible impact of respiratory activity on the agreement between ECG and the Pneumonitor was also assessed. RESULTS: An acceptable agreement for the number of RRi, mean RR, hazard ratio (HR), and HRV measures calculated based on RRi acquired using the ECG and Pneumonitor was presented. There was no association between the breathing pattern and RRi agreement between devices. CONCLUSIONS: The Pneumonitor might be considered appropriate for cardiorespiratory studies in the group of pediatric cardiac patients in rest condition.

A New Approach to Detecting Atrial Fibrillation Using Count Statistics of Relative Changes between Consecutive RR Intervals.

BACKGROUND: The ratio of the difference between neighboring RR intervals to the length of the preceding RR interval (x%) represents the relative change in the duration between two cardiac cycles. We investigated the diagnostic properties of the percentage of relative RR interval differences equal to or greater than x% (pRRx%) with x% in a range between 0.25% and 25% for the distinction of atrial fibrillation (AF) from sinus rhythm (SR). METHODS: We used 1-min ECG segments with RR intervals with either AF (32,141 segments) or SR (32,769 segments) from the publicly available Physionet Long-Term Atrial Fibrillation Database (LTAFDB). The properties of pRRx% for different x% were analyzed using the statistical procedures and metrics commonly used to characterize diagnostic methods. RESULTS: The distributions of pRRx% for AF and SR differ significantly over the whole studied range of x% from 0.25% to 25%, with particularly outstanding diagnostic properties for the x% range of 1.5% to 6%. However, pRR3.25% outperformed other pRRx%. Firstly, it had one of the highest and closest to perfect areas under the curve (0.971). For pRR3.25%, the optimal threshold for distinction AF from SR was set at 75.32%. Then, the accuracy was 95.44%, sensitivity was 97.16%, specificity was 93.76%, the positive predictive value was 93.85%, the negative predictive value was 97.11%, and the diagnostic odds ratio was 514. The excellent diagnostic properties of pRR3.25% were confirmed in the publicly available MIT-BIH Atrial Fibrillation Database. In a direct comparison, pRR3.25% outperformed the diagnostic properties of pRR31 (the percentage of successive RR intervals differing by at least 31 ms), i.e., so far, the best single parameter differentiating AF from SR. CONCLUSIONS: A family of pRRx% parameters has excellent diagnostic properties for AF detection in a range of x% between 1.5% and 6%. However, pRR3.25% outperforms other pRRx% parameters and pRR31 (until now, probably the most robust single heart rate variability parameter for AF diagnosis). The exquisite pRRx% diagnostic properties for AF and its simple computation make it well-suited for AF detection in modern ECG technologies (mobile/wearable devices, biopatches) in long-term monitoring. The diagnostic properties of pRRx% deserve further exploration in other databases with AF.

Evaluation of a wrist-worn photoplethysmography monitor for heart rate variability estimation in patients recovering from laparoscopic colon resection.

To evaluate the accuracy of heart rate variability (HRV) parameters obtained with a wrist-worn photoplethysmography (PPG) monitor in patients recovering from minimally invasive colon resection to investigate whether PPG has potential in postoperative patient monitoring. 31 patients were monitored for three days or until discharge or reoperation using a wrist-worn PPG monitor (PulseOn, Finland) with a Holter monitor (Faros 360, Bittium Biosignals, Finland) as a reference measurement device. Beat-to-beat intervals (BBI) and HRV information collected by PPG were compared with RR intervals (RRI) and HRV obtained from the ECG reference after removing artefacts and ectopic beats. The beat-to-beat mean error (ME) and mean absolute error (MAE) of good quality heartbeat intervals obtained by wrist PPG were estimated as - 1.34 ms and 10.4 ms respectively. A significant variation in the accuracy of the HRV parameters was found. In the time domain, SDNN (9.11%), TRI (11.4%) and TINN (11.1%) were estimated with low relative MAE, while RMSSD (34.3%), pNN50 (139%) and NN50 (188%) had higher errors. The logarithmic parameters in the frequency domain (VLF Log, LF Log and HF Log) exhibited the lowest relative error, and for non-linear parameters, SD2 (7.5%), DFA α1 (8.25%) and DFA α2 (4.71%) were calculated much more accurately than SD1 (34.3%). The wrist PPG shows some potential for use in a clinical setting. The accuracy of several HRV parameters analyzed post hoc was found sufficient to be used in further studies concerning postoperative recovery of patients undergoing laparoscopic colon resection, although there were large errors in many common HRV parameters such as RMSSD, pNN50 and NN50, rendering them unusable.ClinicalTrials.gov Identifier: NCT04996511, August 9, 2021, retrospectively registered.

Validity of the Polar H10 Sensor for Heart Rate Variability Analysis during Resting State and Incremental Exercise in Recreational Men and Women.

Heart rate variability (HRV) is frequently applied in sport-specific settings. The rising use of freely accessible applications for its recording requires validation processes to ensure accurate data. It is the aim of this study to compare the HRV data obtained by the Polar H10 sensor chest strap device and an electrocardiogram (ECG) with the focus on RR intervals and short-term scaling exponent alpha 1 of Detrended Fluctuation Analysis (DFA a1) as non-linear metric of HRV analysis. A group of 25 participants performed an exhaustive cycling ramp with measurements of HRV with both recording systems. Average time between heartbeats (RR), heart rate (HR) and DFA a1 were recorded before (PRE), during, and after (POST) the exercise test. High correlations were found for the resting conditions (PRE: r = 0.95, rc = 0.95, ICC3,1 = 0.95, POST: r = 0.86, rc = 0.84, ICC3,1 = 0.85) and for the incremental exercise (r > 0.93, rc > 0.93, ICC3,1 > 0.93). While PRE and POST comparisons revealed no differences, significant bias could be found during the exercise test for all variables (p < 0.001). For RR and HR, bias and limits of agreement (LoA) in the Bland−Altman analysis were minimal (RR: bias of 0.7 to 0.4 ms with LoA of 4.3 to −2.8 ms during low intensity and 1.3 to −0.5 ms during high intensity, HR: bias of −0.1 to −0.2 ms with LoA of 0.3 to −0.5 ms during low intensity and 0.4 to −0.7 ms during high intensity). DFA a1 showed wider bias and LoAs (bias of 0.9 to 8.6% with LoA of 11.6 to −9.9% during low intensity and 58.1 to −40.9% during high intensity). Linear HRV measurements derived from the Polar H10 chest strap device show strong agreement and small bias compared with ECG recordings and can be recommended for practitioners. However, with respect to DFA a1, values in the uncorrelated range and during higher exercise intensities tend to elicit higher bias and wider LoA.

Associations between heart rate asymmetry expression and asymmetric detrended fluctuation analysis results.

The relation between recently established asymmetry in Asymmetric Detrended Fluctuation Analysis (ADFA) and Heart Rate Asymmetry is studied. It is found that the ADFA asymmetric exponents are related both to the overall variability and to its asymmetric components at all studied time scales. We find that the asymmetry in scaling exponents, i.e., [Formula: see text] is associated with both variance-based and runs-based types of asymmetry. This observation suggests that the physiological mechanisms of both types are similar, even though their origins and mathematical methods are very different. The graphical abstract demonstrates strong, nonlinear association between the expression of Heart Rate Asymmetry measured using relative descriptors and the Asymmetric Detrended Fluctuation Analysis results. It is clear that there is a strong relation between the two theoretically disparate approaches to signal analysis. The technique to demonstrate the association is loess fit.

A two-step method for paroxysmal atrial fibrillation event detection based on machine learning.

Detection of atrial fibrillation (AF) events is significant for early clinical diagnosis and appropriate intervention. However, in existing detection algorithms for paroxysmal AF (AFp), the location of AF starting and ending points in AFp is not concerned. To achieve an accurate identification of AFp events in the long-term dynamic electrocardiograms (ECGs), this paper proposes a two-step method based on machine learning. In the first step, based on features extracted from the calculated R-to-R intervals (RR intervals, the cycle of heart beat), the rhythm type of the ECG signal is first classified into three classes (AFp rhythm, persistent AF (AFf) rhythm, and non-atrial fibrillation (non-AF, N) rhythm) using support vector machine (SVM). In the second step, the starting and ending points for AF episodes of AFp rhythms predicted in the first step are further located based on heartbeat classification. By training a deep convolutional neural network with phased training, the segmented beats of AFp rhythms are divided into AF beats and non-AF beats to determine the beginning and end of any AF episode. The proposed two-step method is trained and tested on the 4th China Physiological Signal Challenge 2021 databases. A final score U of 1.9310 is obtained on the unpublished test set maintained by the challenge organizers, which demonstrates the advantage of the two-step method in AFp event detection. The work is useful for assessing AF burden index for AFp patients.

Machine Learning Using a Single-Lead ECG to Identify Patients With Atrial Fibrillation-Induced Heart Failure.

Aims: Atrial fibrillation (AF) and heart failure often co-exist. Early identification of AF patients at risk for AF-induced heart failure (AF-HF) is desirable to reduce both morbidity and mortality as well as health care costs. We aimed to leverage the characteristics of beat-to-beat-patterns in AF to prospectively discriminate AF patients with and without AF-HF. Methods: A dataset of 10,234 5-min length RR-interval time series derived from 26 AF-HF patients and 26 control patients was extracted from single-lead Holter-ECGs. A total of 14 features were extracted, and the most informative features were selected. Then, a decision tree classifier with 5-fold cross-validation was trained, validated, and tested on the dataset randomly split. The derived algorithm was then tested on 2,261 5-min segments from six AF-HF and six control patients and validated for various time segments. Results: The algorithm based on the spectral entropy of the RR-intervals, the mean value of the relative RR-interval, and the root mean square of successive differences of the relative RR-interval yielded an accuracy of 73.5%, specificity of 91.4%, sensitivity of 64.7%, and PPV of 87.0% to correctly stratify segments to AF-HF. Considering the majority vote of the segments of each patient, 10/12 patients (83.33%) were correctly classified. Conclusion: Beat-to-beat-analysis using a machine learning classifier identifies patients with AF-induced heart failure with clinically relevant diagnostic properties. Application of this algorithm in routine care may improve early identification of patients at risk for AF-induced cardiomyopathy and improve the yield of targeted clinical follow-up.

Inter-patient arrhythmia identification method with RR-intervals and convolutional neural networks.

Objective. The arrhythmia identification method based on the U-net has the potential for fast application. The RR-intervals have been proven to improve the performance of single-heartbeat identification methods. However, because both the heartbeats number and location in the input of the U-net are unfixed, the approach based on the U-net cannot use RR-intervals directly. To solve this problem, we proposed a novel method. The proposed method also can identify heartbeats of four classes, including non-ectopic (N), supraventricular ectopic beat (SVEB), ventricular ectopic beat (VEB), and fusion beat (F).Approach. Our method consists of the pre-processing and the two-stage identification framework. In the pre-processing part, we filtered input signals with a band-pass filter and created the auxiliary waveforms by RR-intervals. In the first stage of the framework, we designed a network to handle input signals and auxiliary waveforms. We proposed a masking operation to separate the input signal into two signals according to the result of the network. The first signal contains heartbeats of SVEB and VEB. The second signal includes heartbeats of N and F. The second stage consists of two networks and can further identify the heartbeats of SVEB, VEB, N, and F from these two signals.Main result. We validated our method on the MIT-BIH arrhythmia database with the inter-patient model. For classes N, SVEB, VEB, and F, our approach achieved F1 scores of 98.26, 68.61, 95.99, and 47.75, respectively.Significance. Our method not only can effectively utilize RR intervals but also can identify multiple arrhythmias.

Indirect Estimation of Breathing Rate from Heart Rate Monitoring System during Running.

Recent advances in wearable technologies integrating multi-modal sensors have enabled the in-field monitoring of several physiological metrics. In sport applications, wearable devices have been widely used to improve performance while minimizing the risk of injuries and illness. The objective of this project is to estimate breathing rate (BR) from respiratory sinus arrhythmia (RSA) using heart rate (HR) recorded with a chest belt during physical activities, yielding additional physiological insight without the need of an additional sensor. Thirty-one healthy adults performed a run at increasing speed until exhaustion on an instrumented treadmill. RR intervals were measured using the Polar H10 HR monitoring system attached to a chest belt. A metabolic measurement system was used as a reference to evaluate the accuracy of the BR estimation. The evaluation of the algorithms consisted of exploring two pre-processing methods (band-pass filters and relative RR intervals transformation) with different instantaneous frequency tracking algorithms (short-term Fourier transform, single frequency tracking, harmonic frequency tracking and peak detection). The two most accurate BR estimations were achieved by combining band-pass filters with short-term Fourier transform, and relative RR intervals transformation with harmonic frequency tracking, showing 5.5% and 7.6% errors, respectively. These two methods were found to provide reasonably accurate BR estimation over a wide range of breathing frequency. Future challenges consist in applying/validating our approaches during in-field endurance running in the context of fatigue assessment.

Automatic Detection of Atrial Fibrillation in ECG Using Co-Occurrence Patterns of Dynamic Symbol Assignment and Machine Learning.

Early detection of atrial fibrillation from electrocardiography (ECG) plays a vital role in the timely prevention and diagnosis of cardiovascular diseases. Various algorithms have been proposed; however, they are lacking in considering varied-length signals, morphological transitions, and abnormalities over long-term recordings. We propose dynamic symbolic assignment (DSA) to differentiate a normal sinus rhythm (SR) from paroxysmal atrial fibrillation (PAF). We use ECG signals and their interbeat (RR) intervals from two public databases namely, AF Prediction Challenge Database (AFPDB) and AF Termination Challenge Database (AFTDB). We transform RR intervals into a symbolic representation and compute co-occurrence matrices. The DSA feature is extracted using varied symbol-length V, word-size W, and applied to five machine learning algorithms for classification. We test five hypotheses: (i) DSA captures the dynamics of the series, (ii) DSA is a reliable technique for various databases, (iii) optimal parameters improve DSA's performance, (iv) DSA is consistent for variable signal lengths, and (v) DSA supports cross-data analysis. Our method captures the transition patterns of the RR intervals. The DSA feature exhibit a statistically significant difference in SR and PAF conditions (p < 0.005). The DSA feature with W=3 and V=3 yield maximum performance. In terms of F-measure (F), rotation forest and ensemble learning classifier are the most accurate for AFPDB (F = 94.6%) and AFTDB (F = 99.8%). Our method is effective for short-length signals and supports cross-data analysis. The DSA is capable of capturing the dynamics of varied-lengths ECG signals. Particularly, the optimal parameters-based DSA feature and ensemble learning could help to detect PAF in long-term ECG signals. Our method maps time series into a symbolic representation and identifies abnormalities in noisy, varied-length, and pathological ECG signals.

An Improved UNet++ Model for Congestive Heart Failure Diagnosis Using Short-Term RR Intervals.

Congestive heart failure (CHF), a progressive and complex syndrome caused by ventricular dysfunction, is difficult to detect at an early stage. Heart rate variability (HRV) was proposed as a prognostic indicator for CHF. Inspired by the success of 2-D UNet++ in medical image segmentation, in this paper, we introduce an end-to-end encoder-decoder model to detect CHF using HRV signals. The developed model enhances the UNet++ model with Squeeze-and-Excitation (SE) residual blocks to extract deep features hierarchically and distinguish CHF patients from normal subjects. Two open-source databases are utilized for evaluating the proposed method, and three segment lengths of intervals between successive R-peaks are employed in comparison with state-of-the-art methods. The proposed method achieves an accuracy of 85.64%, 86.65% and 88.79% when 500, 1000 and 2000 RR intervals are utilized, respectively. It demonstrates that HRV evaluation based on deep learning can be an important tool for early detection of CHF, and may assist clinicians in achieving timely and accurate diagnoses.

Extended Central Tendency Measure and difference plot for heart rate variability analysis.

Heart rate variability (HRV) is a non-invasive alternative to analyze the role of the autonomic nervous system (ANS) on heart functioning. Many tools have been developed to analyze collected cardiac data. Among them, the Central Tendency Measure (CTM) is a quantitative method for variability analysis of RR intervals. The values of the CTM must be between 0 and 1 (inclusive) for different radius, which follows the intrinsic characteristics of each time series. Using the conventional CTM, the successive differences of the time series may be calculated, and it can classify and differentiate the disturbances in the ANS involving HRV. This method was extended (e-CTM) to analyze the differences between RR interval time series. In this extension, a new parameter is added, which allows analysis of long time intervals, instead of successive and adjacent RR intervals. The ability of the e-CTM to differentiate the groups of the RR interval time series was verified with 145 RR interval time series divided into three groups: subjects with congestive heart failure, healthy subjects, and nurses during one hour of their workday. Results evidence that the new parameter added differentiates the group with pathology (and subsequent impairment of ANS) and group under stress at work (temporary impairment of ANS). These results suggest that the e-CTM is capable of detection long-term variations in the HRV according to the ANS impairment.

How the insula speaks to the heart: Cardiac responses to insular stimulation in humans.

Despite numerous studies suggesting the role of insular cortex in the control of autonomic activity, the exact location of cardiac motor regions remains controversial. We provide here a functional mapping of autonomic cardiac responses to intracortical stimulations of the human insula. The cardiac effects of 100 insular electrical stimulations into 47 epileptic patients were divided into tachycardia, bradycardia, and no cardiac response according to the magnitude of RR interval (RRI) reactivity. Sympathetic (low frequency, LF, and low to high frequency powers ratio, LF/HF ratio) and parasympathetic (high frequency power, HF) reactivity were studied using RRI analysis. Bradycardia was induced by 26 stimulations (26%) and tachycardia by 21 stimulations (21%). Right and left insular stimulations induced as often a bradycardia as a tachycardia. Tachycardia was accompanied by an increase in LF/HF ratio, suggesting an increase in sympathetic tone; while bradycardia seemed accompanied by an increase of parasympathetic tone reflected by an increase in HF. There was some left/right asymmetry in insular subregions where increased or decreased heart rates were produced after stimulation. However, spatial distribution of tachycardia responses predominated in the posterior insula, whereas bradycardia sites were more anterior in the median part of the insula. These findings seemed to indicate a posterior predominance of sympathetic control in the insula, whichever the side; whereas the parasympathetic control seemed more anterior. Dysfunction of these regions should be considered when modifications of cardiac activity occur during epileptic seizures and in cardiovascular diseases.