Diagnosis Of Paroxysmal Atrial Fibrillation Research Articles

Achieving Real-Time Prediction of Paroxysmal Atrial Fibrillation Onset by Convolutional Neural Network and Sliding Window on R-R Interval Sequences.

Early diagnosis of paroxysmal atrial fibrillation (PAF) could prompt patients to receive timely interventions in clinical practice. Various PAF onset prediction algorithms might benefit from accurate heart rate variability (HRV) analysis and machine learning classification but are challenged by real-time monitoring scenarios. The aim of this study is to present an end-to-end deep learning-based PAFNet model that integrates a sliding window technique on raw R-R intervals of electrocardiogram (ECG) segments to achieve a real-time prediction of PAF onset. This integration enables the deep convolutional neural network (CNN) to be customized as a light-weight architecture that accommodates the size of sliding windows simply by altering the input layer, and specifically its effectiveness in making a new prediction with each new heartbeat. Catering to the potential influence of input sizes, three CNN models were trained using 50, 100, and 200 R-R intervals, respectively. For each model, the performance of the automated algorithms was evaluated for PAF prediction using a ten-fold cross-validation. As a results, a total of 56,381 PAFN-type and 56,900 N-type R-R interval segments were collected from publicly accessible ECG databases, and a promising prediction performance of the automated algorithm with 100 R-R intervals was achieved, with a sensitivity of 97.12%, a specificity of 97.77%, and an accuracy of 97.45%, respectively. Importantly, the automated algorithm with a sliding window step of 1 could process one sample in only 23.1 milliseconds and identify the onset of PAF at least 45 min in advance. The present results suggest that the sliding window technique on raw R-R interval sequences, along with deep learning-based algorithms, may offer the possibility of providing an accurate, real-time, end-to-end clinical tool for mass monitoring of PAF.

Prediction of paroxysmal atrial fibrillation using a convolutional neural network and electrocardiogram signals

Atrial fibrillation (AF) is the most clinically diagnosed arrhythmia in cardiac pathology. The incidence of AF begins at a very early age and its initial state is paroxysmal atrial fibrillation (PAF). This type of heart disease can be detected and predicted by analyzing the spectrogram of a surface electrocardiogram (ECG) signal. In many studies, different ECG signal formats and convolutional neural network (CNN) architectures have been used. However, the lack of good signal preprocessing or signal adequacy may have affected the accuracy, especially on short-term ECG signals. In this study, we analyzed a preprocessed ECG signal, determined the optimal set to predict PAF, and evaluated the accuracy using ECG signals of different durations. The PAF Prediction Challenge–PhysioNet database was used to extract spectrograms in 30-sec and 5-sec windows for two classes (Normal, PAF) and 3 classes (Normal, Close-AF, Distant-AF). Then, the AlexNet architecture was used. The proposed method achieved a two-class accuracy of 99.92% with a 30-sec window and 99.42% with a 5-sec window, improving the PAF prediction performance compared with similar works. In addition, the three-class accuracies were 96.92% and 97.43% with windows of 30-sec, and 5-sec, respectively. These results prove the efficacy of the method for the early diagnosis of PAF, even based on short-term ECG signals.

Single-lead electrocardiogram quality assessment in the context of paroxysmal atrial fibrillation through phase space plots

Current wearable electrocardiogram (ECG) recording systems have great potential to revolutionize early diagnosis of paroxysmal atrial fibrillation (AF). They are able to continuously acquire an ECG signal for long weeks and then increase the probability of detecting first brief, intermittent signs of the arrhythmia. However, the recorded signal is often broadly corrupted by noise and artifacts, and accurate assessment of its quality to avoid automated misdiagnosis and false alarms of AF is still an unsolved challenge. In this context, the present work is pioneer in exploring the usefulness of transforming the single-lead ECG signal into two common phase space (PS) representations, such as the Poincaré plot and the first order difference graph, for evaluation of its quality. Several machine and deep learning models fed with features and images derived from these PS portraits reported a better performance than well-known previous methods, even when they were trained and validated on two separate databases. Indeed, in binary classification of high- and low-quality ECG excerpts, the generated PS-based algorithms reported a discriminant power greater than 85%, misclassifying less than 20% of high-quality AF episodes and non-normal rhythms as noisy excerpts. Moreover, because both PS reconstructions do not require any mathematical transformation, these algorithms also spent much less time in classifying each ECG excerpt in validation and testing stages than previous methods. As a consequence, ECG transformation to both PS portraits enables novel, simple, effective, and computational low-cost techniques, based both on machine and deep learning classifiers, for ECG quality assessment.

Role of POCUS in Assessing an Acute Aortic Thrombus

A 67-year-old female patient presented with abdominal pain with a recent diagnosis of paroxysmal atrial fibrillation. Computed tomography (CT) of the abdomen demonstrated a filling defect concerning for an aortic thrombus. Point of care ultrasound (POCUS) confirmed a mobile thrombus in the proximal abdominal aorta in close proximity to several major arterial branches, leading to urgent surgical consultation due to a concern for mesenteric and end-organ ischemia. POCUS played a role in determining patient management in this novel case, and the patient was anticoagulated and ultimately discharged from the hospital.

The exploratory clinical trial of AI-enabled Holter electrocardiogram for diagnosis of paroxysmal atrial fibrillation from sinus rhythm waveforms

Abstract Backgrounds Machine learning (ML) models using convolutional neural networks (CNNs) have been successfully applied to arrhythmia detection and prediction. The current study aimed to develop a CNN-based AI algorithm for the prediction of paroxysmal atrial fibrillation (PAF) patients during sinus rhythm waveforms from Holter electrocardiograms. Methods A ML model was developed by using 2-leads from Holter ECG waveforms divided into slots of 30 seconds each. As training data, we prepared > 300000 slots with sinus rhythm and they were subsequently converted to image data for application to the CNN-based AF detection algorithm. The performance metrics obtained for the ML model for PAF detection were as follows: sensitivity, 85.2 %; specificity, 64.0 % and AUC for PAF prediction was 0.84. To validate the performance of the ML model, multicenter clinical trial was conducted. The clinical trial was conducted at three independent centers and included 24 patients with PAF and 20 patients in true sinus rhythm. Five factors were matched between the PAF and sinus groups: age, gender, hypertension under treatment, diabetes under treatment, and smoking. Waveform data from validation datasets to Grad-CAM were applied to obtain explainable AI (XAI) images, which allowed for a heat map over the waveform to be produced. Results For the primary endpoint, the estimated sensitivity and specificity [95% CI] were 91.7% [73.0-99.0%] and 65.0% [40.8-84.6%], respectively (Table 1). The sensitivity and specificity of the secondary endpoints using 10 time-independent slots were the same as for the primary endpoint. Regarding safety endpoints, no adverse events attributable to the test device itself were observed. Based on the regions of interest highlighted via Grad-CAM, the most contributing featured regions to predict PAF were indicated by areas on P waves and QT area (Figure 1). Featured regions to rule out the existence of PAF indicated areas just on QRS waves were also present. It was therefore determined that the proposed CNN-based ML model focuses on the P-wave area and atrial repolarization area in the case of PAF. Conclusion The study suggested that 2-lead Holter ECG with 30 second slot could detect PAF patients from their sinus rhythm waveforms and the performance was more effectively than using 12-lead ECGs. The results of this clinical trial are expected to contribute to the screening of patients with PAF that could not be detected by a single time Holter ECG.Table 1Figure 1

Abstract 14753: Cigarette Smoking is Acutely Associated With Increased Cardiac Arrhythmias

Background: Chronic cigarette smoking is a known risk factor for the eventual development of atrial fibrillation (AF). However, the acute effects of cigarette smoking on arrhythmias in patients with AF have not been elucidated. Methods/Approaches: We conducted a prospective observational study enrolling consenting adults with a diagnosis of paroxysmal AF and who actively smoked. Participants were equipped with continuously recording ECG devices (iRhythm, San Francisco, CA) for up to 14 days and used a smart lighter (Quitbit, San Francisco, CA or Slighter, Cary, NC) or button presses on the ECG device to time-stamp cigarette smoking events in real-time. Outcomes included any AF, atrial flutter (AFL), supraventricular tachycardia (SVT), ventricular tachycardia (VT), atrial ectopic rhythms (EAR), and ventricular bigeminy and trigeminy. For this case-crossover study, we used a generalized linear mixed model with a Poisson distribution and accounted for the clustering of repeated measures for each participant to estimate the incidence rate ratio (IRR) between arrhythmias and smoking events. Results: We enrolled 29 participants (45% female, mean 55 ± 13 years old, 76% non-Hispanic white), where 22 used a smart lighter, and 7 relied on real-time button presses to document smoking events. Mean study duration was 12 ± 4 days. While wearing the monitor, 24 participants experienced ≥ 1 arrhythmia, including AF (n = 3), SVT (n = 19), VT (n= 5), EAR (n = 23), and ventricular bigeminy/trigeminy (n =12), and 9 participants had ≥ 1 sustained arrhythmia (arrhythmic events >30 secs). Every additional cigarette smoked per day was associated with a 6% increase in arrhythmia events (IRR 1.06, 95% CI 1.05-1.07, P < 0.001; Figure ). No statistically significant relationships were observed between smoking events and sustained arrhythmias per day. Conclusion: The data suggest smoking acutely increases the chance an arrhythmia will occur.

The left atrial morphologic and functional parameters as predictors of paroxysmal atrial fibrillation in patients after ischemic stroke

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Cardiogenic source of embolism is the most common cause of non-lacunar ischemic stroke (IS). Identifying the patients with undiagnosed paroxysmal atrial fibrillation (PAF) may reduce the risk of recurrent cardiogenic strokes. Reliable predictors of PAF seems to be crucial for selection the patients to prolonged ECG monitoring after IS. Purpose To assess the feasibility of echocardiography in selecting the patients with highest risk of PAF. Methods We present the preliminary data of prospective cohort study of patients after IS or transient ischemic attack of unknown aetiology and possible cardiogenic origin. The examined group comprises of 40 consecutive patients without history of AF, capable to perform ECG recording. All patients included in the study had transthoracic echocardiography (TTE) prior the monitoring period. An offline analysis including left atrial functional and morphologic assessment was performed. Every patient enrolled in the study was equipped for 2 weeks with a mobile, on demand 1-minute single lead ECG recorder and was asked for at least two exams per day. After two weeks all recorded ECGs were checked for AF occurrence. The level of statistical significance was set at p<0.05. Results From 40 patients (aged 64,5 yrs in average) included in the study 45% were males; (n=18). 15% (n=6) of patients were diagnosed with PAF during ECG monitoring and comprised PAF subgroup, the rest of patients formed non PAF subgroup (nPAF). Mean number of ECG records was 56,5 per patient, during the monitoring period. There were no significant differences between PAF/nPAF in age, past medical history, and left ventricular (LV) size or systolic function. Patients from PAF group have statistically significant higher values of left atrial (LA) area and volume see table 1. PAF group has significantly lower (in absolute values) left atrial contraction strain – (LASct). There were no differences in conduit and reservoir LA (LASr) strains nor in LA ejection fraction. The highest PAF diagnosis efficacy according to area under the curve (AUC) of ROC curves was noted for parameters of: LA size: LA atrium volume index (LAVI) in 4 chamber view (AUC=0,82) and LA area index (0,81) LA contraction function: LASct (Avg) (AUC =0,80) LV diastolic function: E/E’ (0,81) For the threshold of 100% sensitivity the highest specificity was found for LA volume before active atrial contraction in 4 chamber view with cut-off value of >=37,6 ml (specificity 64,7%) (Figure 1.) and LASct (Avg) with cut-off value <=17% in absolute values (specificity 47,1%). Conclusions Both functional and morphologic parameters of LA can predict PAF diagnosis during ECG monitoring after IS. TTE after IS can be used to select patients who would benefit the most from prolonged ECG monitoring. High PAF detectability (100% sensitivity) with approximately 1/2 negative results of ECG monitoring can be achieved by both LA volumetric (LAV preA) and strain (LASct) analysis.

Risk of myocardial infarction and ischemic stroke in individuals with first-diagnosed paroxysmal vs. non-paroxysmal atrial fibrillation under anticoagulation.

There is conflicting evidence on whether the type of atrial fibrillation (AF) is associated with risk of cardiovascular events, including acute myocardial infarction (MI) and ischemic stroke. The aim of the present study was to investigate whether the risk of MI and ischemic stroke differs between individuals with first-diagnosed paroxysmal vs. non-paroxysmal AF treated with anticoagulants. De-identified electronic medical records from the TriNetX federated research network were used. Individuals with a new diagnosis of paroxysmal AF who had no evidence of other types of AF in their records were 1:1 propensity score-matched with individuals with non-paroxysmal AF, defined as persistent or chronic AF, who had no evidence of other types of AF in their records. All patients were followed for three years for the outcomes of MI and ischemic stroke. Cox proportional hazard models were used to calculate hazard ratios (HRs) with 95% confidence intervals (CIs). In the propensity-matched cohort, among 24 848 well-matched AF individuals [mean age 74.4 ± 10.4; 10 101 (40.6%) female], 410 (1.7%) were diagnosed with acute MI and 875 (3.5%) with ischemic stroke during the three-year follow-up. Individuals with paroxysmal AF had significantly higher risk of acute MI (HR: 1.65, 95%CI: 1.35-2.01) compared to those with non-paroxysmal AF. First diagnosed paroxysmal AF was associated with higher risk of non-ST elevation MI (nSTEMI) (HR: 1.89, 95%CI: 1.44-2.46). No significant association was observed between the type of AF and risk of ischemic stroke (HR: 1.09, 95%CI: 0.95-1.25). Patients with first-diagnosed paroxysmal AF had higher risk of acute MI compared to individuals with non-paroxysmal AF, attributed to the higher risk of nSTEMI among patients with first-diagnosed paroxysmal AF. There was no significant association between type of AF and risk of ischemic stroke.

Impact of low-dose Brain-Derived Neurotrophic Factor (BDNF) on atrial fibrillation recurrence.

Atrial fibrillation is the most common arrhythmia worldwide and is associated with significant morbidity and mortality. Despite the effectiveness of catheter-based ablation, periprocedural complication and recurrences remain a concern. In this context, we aim to appraise the potential impact of brain-derived neurotrophic factor (BDNF) on reducing episodes of paroxysmal atrial fibrillation (PAF). 22 patients with an established diagnosis of PAF and without structural heart disease were considered. Every patient received 20 drops of GUNA-BDNF administered in the morning. During the 24 months of follow-up, the arrhythmic burden was measured by the average monthly duration of PAF episodes. At the end of the follow-up period (24 months), data from 22 patients, of whom 17 men and five women, were analyzed. The arrhythmic burden, measured in terms of average monthly duration of PAF episodes, was found significantly reduced after the administration of low dose BDNF (9.5 vs. 65.3 minutes per month, P<0.001). A total of 17 out of 22 patients saw their arrhythmic burden eliminated or consistently reduced, furthermore two patients underwent a drastic reduction of the average monthly duration of AF (more than 200 minutes compared to the baseline). Only four patients, despite the administration of BDNF, still experienced an arrhythmic burden of 20 minutes or more. Considering the age groups, the major reduction was observed in people aged 70 or more, who were also the most represented in the sample. These results are coherent with the poor literature currently available. BDNF low dose therapy has shown to have an impacting role in reducing the arrhythmic burden and recurrences of AF, with a particular effectiveness in patients over 70 and without structural heart disease. We should welcome this work, despite it limitations. Further clinical and molecular studies are needed before-considering BDNF low dose as a tool against PAF.

Abstract 9772: Paroxysmal Atrial Fibrillation Among Pulmonary Embolism Patients; Insight From the 2019 National Inpatient Sample

Introduction: Various factors can trigger events of Pulmonary Embolism(PE). Several studies have linked cardiac arrhythmias such as Paroxysmal Atrial Fibrillation (PAF) with a higher risk of multiple cardiovascular and pulmonary complications. As there is a paucity of data on the risk factors for an event of PAF among PE patients and the factors contributing to death, we conducted a retrospective analysis using the 2019 National Inpatient Sample(NIS). Methods: Our study focused on finding patients with a principal diagnosis of Pulmonary Embolism and a diagnosis of Paroxysmal Atrial Fibrillation. A multivariable regression model evaluated several risk factors for PAF incidence among PE patients and their possible mortality risks. Results: A total of 188,850 cases of PE were observed in our study, including 9,960 (5.3%) cases of PAF. Hyperlipidemia (aOR 1.348, 95% CI 1.290-1.409, p&lt;0.01), hypertension (aOR 1.988, 95% CI 1.887-2.095, p&lt;0.01), and old myocardial infarct (aOR 1.574, 95% CI 1.463-1.693, p&lt;0.01) predisposed to PAF among PE, while females (aOR: 0.873, 95% CI 0.837-0.910, p&lt;0.01), patients with obesity (aOR 0.858, 95% CI 0.818-0.900, p&lt;0.01), and those classified as racially Black (aOR 0.603, 95% CI 0.534-0.681, p&lt;0.01), and Hispanic (aOR 0.653, 95% CI 0.562-0.758, p&lt;0.01) (compared to group excluding Whites, Blacks and Hispanics) reduced their risk. An in-hospital mortality rate of 4.0% was also estimated as 395 patients with PAF died. PAF patients with PE had a higher chance of dying if they also had a diagnosis of diabetes (aOR 1.347, 95% CI 1.070-1.695, p=0.011), whereas those who were obese (aOR 0.421, 95% CI 0.311-0.570, p&lt;0.01), with hyperlipidemia (aOR 0.732, 95% CI 0.590-0.909, p&lt;0.01), and smokers (aOR 0.751, 95% CI 0.606-0.931, p&lt;0.01) expressed a lower risk of mortality. PAF patients with PE reported a higher overall risk of death during hospitalization (aOR 1.322, 95% CI 1.190-1.469, p&lt;0.01). Conclusions: Patients with a diagnosis of PAF while admitted with PE showed a higher risk of mortality. Several factors influenced the presence of PAF among PE patients. It is thus essential for physicians to monitor various cardiovascular complications and arrhythmias in patients admitted with PE.

LA strain helps identifying the cardioembolic etiology of transient ischemic attacks

Abstract Background Patients who have transient ischemic attacks (TIA) often suffer from asymptomatic and paroxysmal atrial fibrillation (AF). Due to the fact that AF is located in the atria, we sought to determine whether abnormalities in left atrial (LA) structure and function could help identify the cardioembolic cause of TIA in patients with sinus rhythm but with documented episodes of paroxysmal AF. Methods The research included 190 individuals with TIA and classified them into two groups based on the presence (group I) or absence (group II) of confirmed paroxysmal AF. The diagnosis of paroxysmal AF was established by an assessment of medical records. To prevent assessing atrial stunning, cardiac ultrasonography was conducted in sinus rhythm at least 14 days following the initiation the TIA. Results The results indicated that patients in group I were older, more often female, had a history of stroke or TIA, and had a higher CHA2DS2-VASc score. Additionally, they exhibited increased LA volumes, a decreased LA emptying fraction, and markedly altered LA deformation patterns. Three factors were found as being independently associated with paroxysmal AF using multivariate logistic regression: age, LA reservoir strain, and LA emptying fraction (P&amp;lt;0.0001). The variables had as cut-off values: age &amp;gt;55 years, LA reservoir strain&amp;lt;−17%, and LA emptying fraction &amp;lt;51%. Conclusion The current research establishes that LA strains are independently associated with paroxysmal AF in patients with TIA and may aid in determining the cardioembolic origin of these events. Our findings have considerable clinical implications because, until now, LA 2D-STE was not included in the usual evaluation of TIA patients, but it may be the ideal next step in this regard. Funding Acknowledgement Type of funding sources: None.

PO-712-05 TEMPORAL TRENDS IN USING CATHETER ABLATION AS FIRST-LINE THERAPY IN PATIENTS DIAGNOSED WITH PAROXYSMAL ATRIAL FIBRILLATION IN THE UNITED STATES

More than a decade ago, randomized clinical trials have demonstrated clinical benefits of catheter ablation compared to antiarrhythmic drugs (AAD) in patients with symptomatic paroxysmal atrial fibrillation (AF) refractory or intolerant to AAD. More recently, superiority of catheter ablation as first-line treatment vs AAD has been demonstrated as well. In addition, early AF ablation has been shown to delay AF progression. However, data are limited on contemporary trends of real-world catheter ablation practice as first-line therapy in patients diagnosed with paroxysmal AF. To explore the temporal trends in catheter ablation as first-line therapy in patients diagnosed with paroxysmal AF in the US. A retrospective analysis was conducted in adult patients who had a diagnosis of paroxysmal AF and underwent a catheter ablation of AF by pulmonary vein isolation during 2016-2020 from the IBM MarketScan® Commercial Database. The proportion of catheter ablation as first-line therapy in patients diagnosed with paroxysmal AF was calculated according to the calendar year of the catheter ablation procedure. Logistic regression models evaluated the association between the calendar year of the catheter ablation procedure and first-line therapy choice. There was a significant 61% increase in the proportion of catheter ablation as first-line therapy in patients diagnosed with paroxysmal AF from 2016 (25%) to 2020 (35%) (2020 vs 2016: odds ratio, 1.61; 95% confidence interval, 1.42-1.83) (p for trend <0.0001). The increasing trend was similar by sex, age, and setting of care, but the proportion was slightly higher in younger patients than in older patients and was more pronounced in the inpatient vs outpatient setting of care (Figure 1). There was an increasing trend in using catheter ablation as first-line therapy in patients diagnosed with paroxysmal AF in the US. This finding may reflect increasing comfort level with catheter ablation due to multiple demonstrated clinical benefits compared to AAD as first-line treatment for symptomatic paroxysmal AF.

Decreasing time between first diagnosis of paroxysmal atrial fibrillation and cryoballoon ablation positively affects long-term consequences.

Patients with atrial fibrillation (AF) usually experience a worsening of their AF burden over time. We aimed to investigate timing of pulmonary vein isolation (PVI) by cryoballoon (CB-2) after the first clinical diagnosis of AF on ablation-related outcomes. A total of 132 consecutive patients with paroxysmal AF undergoing PVI by CB-2 were included in the study. The patients were retrospectively sorted into two groups to evaluate differences in AF recurrence risk associated with early ablation (n = 89), defined as within 365 days of first AF diagnosis, and late ablation (n = 365), defined as > 365 days after first AF diagnosis. AF-free survival during follow-up was compared between groups. Although mean procedure times were comparable between groups, mean fluoroscopy times were lower in the early ablation group. For the whole study group, median (interquartile range) time from AF diagnosis to first ablation was 4.0 (2.0-11.3) months [3.0 (1.0-4.0) vs 14.0 (12.0-22.5) months in the early and late ablation groups, respectively]. Median follow-up for the whole population was 12.0 (12.0-18.0) months, and after the blanking period, 14 (10.6%) patients had arrhythmia recurrence (2 in the early and 12 in the late ablation groups). In the univariable Cox regression analysis and propensity score adjusted penalized Cox regression analysis, there was a significant association between delay in ablation time and AF recurrence (unadjusted hazard ratio = 7.74, 95% CI 2.26-40.1, p < 0.001, adjusted hazard ratio = 7.50, 95% CI 2.23-38.6, p < 0.001). Delays in treatment with CB-2 ablation may negatively affect AF-free survival rates among patients with paroxysmal AF.

Detecting paroxysmal atrial fibrillation from normal sinus rhythm in equine athletes using Symmetric Projection Attractor Reconstruction and machine learning

BackgroundAtrial fibrillation (AF) is a common cardiac arrhythmia in both human and equine populations. It is associated with adverse outcomes in humans and decreased athletic performance in both populations. Paroxysmal atrial fibrillation (PAF) presents with intermittent, self-terminating AF episodes, and is difficult to diagnose once sinus rhythm resumes.ObjectiveWe aimed to detect PAF subjects from normal sinus rhythm equine electrocardiograms (ECGs) using the Symmetric Projection Attractor Reconstruction (SPAR) method to encapsulate the waveform morphology and variability as the basis of a machine learning classification.MethodsWe obtained ECG signals from 139 active equine athletes (120 control, 19 with a PAF diagnosis). The SPAR method was applied to 9 short (20-second) ECG strips for each subject. An optimal SPAR feature set was determined by forward feature selection for input to a machine learning model ensemble of 3 different classifiers (k-nearest neighbors, linear support vector machine, and radial basis function kernel support vector machine). Imbalanced data were handled by upsampling the minority (PAF) class. A final subject classification was made by taking a majority vote over results from the 9 ECG strips.ResultsOur final cross-validated classification for a subject gave an accuracy of 89.0%, sensitivity of 94.8%, specificity of 87.1%, and receiver operating characteristic area under the curve of 0.98, taking PAF as the positive class.ConclusionThe SPAR method and machine learning generated a final model with high sensitivity, suggesting that PAF can be discriminated from short equine ECG strips. This preliminary study indicated that SPAR analysis of human ECG could support patient monitoring, risk stratification, and clinical decision-making.

Diagnostic values of different ECG durations in paroxysmal AF diagnosis.

ObjectiveTo evaluate the application value of different‐duration electrocardiograms (ECGs) in paroxysmal atrial fibrillation (AF) diagnosis.MethodsA total of 220 patients with paroxysmal atrial fibrillation diagnosed by 24‐h dynamic electrocardiogram in our hospital from January 2019 to December 2020 were selected as the study subjects. The patients were monitored for 10 s, 5 min, and 24 hr, respectively. The results of the three different‐duration ECGs were compared and analyzed in order to evaluate their paroxysmal AF diagnosis value.ResultsParoxysmal AF was detected in 18 patients (8.2%) with the 10‐s ECG; in 89 patients (40.5%) with the 5‐min ECG; and in 199 patients (90.5%) with the 24‐h dynamic ECG.ConclusionIn patients with paroxysmal AF, ECGs with longer detection times had higher detection rates.

8 Acute pulmonary embolism or paroxysmal atrial fibrillation in patients with heart failure and preserved ejection fraction: prognostic implications

Abstract Aims Heart failure with preserved ejection fraction (HFpEF) is a growing public health problem. Its prevalence among heart failure patients increases over time, accounting for at least 50% of all hospital admissions for HF. Nevertheless, no single guideline exists for diagnosis or treatment for HFpEF, and older age or comorbidities are additional factors that confuse etiology and complicate prognosis. Moreover, there are few data regarding the consequences of HFpEF on other recurrent pathologies. To assess the prognostic impact of a pre-existing HFpEF on patients ospidalized for intercurrent episodes of atrial fibrillation (AF) or acute pulmonary embolism (PE). Methods and results We performed a retrospective evaluation of 222 patients, consecutively hospitalized in our unit of Cardiology with a diagnosis of paroxysmal AF or acute PE, from January 2017 to December 2020. We recruited exclusively patients with normal cardiac function and HFpEF patients. Heart failure with reduced FEVS patients were excluded from the study. We have described for each patient the demographic and clinical characteristics, comorbidities, instrumental test results, and clinical outcomes. In order to assess, for each group, the relationship between patient characteristics and clinical outcomes, the Chi-square test or alternatively the Pearson–Spearman correlation coefficients were calculated. The 222 patients studied had an average age of 73.7 years (min. 27, max: 94). 68 AF patients had pre-existing HFpEF, whereas AF patients without HF were 77. Patients with pre-existing HFpEF and newly-onset AF had a more advanced age (76.7 y vs. 72.9 y), and greater comorbidity (meanly 4 vs. 3) rather than AF patients without HFpEF. Moreover, percentage of converting arrhythmia were significantly higher in AF patients without HFpEF. Patients with acute PE and pre-existing HFpEF were 44, whereas PE patients without HF were 33. Acute PE patients with pre-existing HFpEF had older age, a prevalence for female sex, more comorbidities, an average longer hospitalizations, but no significantly different rates of severe complications (ictus, haemorrhagies, needs for ventilation, pulmonary infarction or deaths) rather than PE patients without HFpEF. Conclusions The patients with AF or PE and concomitant HFpEF that were hospitalized from January 2017 to December 2020, showed an average longer hospitalization, a lower percentage of converting arrhythmia, probably due to the older age and the greater comorbidity.

Diagnosis of paroxysmal atrial fibrillation from thirty-minute heart rate variability data using convolutional neural networks

Paroxysmal atrial fibrillation (PAF) is the initial stage of atrial fibrillation, one of the most common arrhythmia types. PAF worsens with time and affects the patient?s life quality negatively. In this study, we aimed to diagnose PAF early, so patients can start taking precautions before this disease gets worse. We used the atrial fibrillation prediction database, an open data from Physionet and constructed our approach using convolutional neural networks. Heart rate variability (HRV) features are calculated from time-domain measures, frequency-domain measures using power spectral density estimations (fast Fourier transform, Lomb-Scargle, and Welch periodogram), time-frequencydomain measures using wavelet transform, and nonlinear Poincare plot measures. We also normalized these features using min-max normalization and z-score normalization methods. In addition, we also applied alternatively the heart rate normalization (HRN), which gave promising results in a few HRV-based research, before calculating these features. Thus, HRV data, HRN data, and HRV features extracted from six different combinations of these normalizations, in addition to no normalization cases, were applied to the convolutional neural networks classifier. We tuned the classifiers using 90% of samples and tested the classifiers' performances using 10% of data. The proposed approach resulted in 95.92% accuracy, 100% precision, 91.84% recall, and 95.74% f1-score in HRV with z-score feature normalization. When the heart rate normalization was also applied, the proposed approach reached 100% accuracy, 100% precision, 100% recall, and 100% f1-score in HRV with z-score feature normalization. The proposed method with heart rate normalization and z-score normalization methods resulted in better classification performance than similar studies in the literature. In addition, although deep learning models offer no use of separate feature extraction processes, this study reveals that using HRV-specific feature extraction techniques may improve the performance of deep learning algorithms in HRV-based studies. Comparing the existing studies, we concluded that our approach provides a much better tool to diagnose PAF patients.

Paroxysmal atrial fibrillation incidence in acute ischemic stroke or transient ischemic attacks detected by 48-h Holter monitoring

Objective To examine the feasibility of early use of 48-h Holter monitoring to detect the incidence of paroxysmal atrial fibrillation (AF) in patients presenting with acute stroke or transient ischemic attack (TIA) and without present evidence of AF in the Neurology Department (MUST University). Background Many ischemic stroke patients have undiagnosed paroxysmal AF. Paroxysmal AF diagnosis is relevant in individuals who present with cerebral stroke and without AF on admission. Diagnosis of paroxysmal AF is essential because it is a strong risk factor for stroke recurrence and also because oral anticoagulation has an advantage over antiplatelet treatment. Patients and methods The study included 200 consecutive patients who had acute ischemic stroke or TIA symptoms. Each of them had no evidence of AF at the time of presentation, and within 1 week, they performed Holter monitoring for 48 h. Results Two-hundred patients with mean age 65.27 ± 13.61 years old with TIA and acute ischemic stroke were found in 41 (20.5%) patients and 159 (79.5%) patients, respectively. All 200 patients were monitored by 48-h Holter. Among 200 cases, 13 (6.5%) patients had paroxysmal AF. All 13 patients had acute stroke and were older than age 60 years. It represented 8.17% of patients with acute cerebrovascular stroke (13 patients out of 159 patients with acute cerebrovascular stroke). The only factor related to an increased risk of paroxysmal AF was stroke type. AF was responsible for 50% of clinically suspected cardiac embolic stroke cases. Conclusion Paroxysmal AF was diagnosed in ~ 6.5% of acute-stroke or TIA patients having normal ECG rhythm by early use of 48-h Holter monitoring. In acute cerebrovascular stroke patients, AF represented 8.17%. Holter monitoring is a feasible and simple tool for paroxysmal AF diagnosis in TIA or acute-stroke patients, thus allowing proper treatment and better secondary prevention.

Progrostic impact of a pre-existing heart failure with preserved ejection fraction in patients with acute pulmonary embolism or paroxysmal atrial fibrillation

Abstract Funding Acknowledgements Type of funding sources: None. OnBehalf none Introduction heart failure with preserved ejection fraction (HFpEF) is a growing public health problem. Its prevalence among heart failure patients increases over time, accounting for at least 50 % of all hospital admissions for HF. Nevertheless, no single guideline exists for diagnosis or treatment for HFpEF, and older age or comorbidities are additional factors that confuse etiology and complicate prognosis. Moreover, there are few data regarding the consequences of HFpEF on other recurrent pathologies. Aims to assess the prognostic impact of a pre-existing HFpEF on patients ospidalized for intercurrent episodes of atrial fibrillation (AF) or acute pulmonary embolism (PE) Methods We performed a retrospective evaluation of 194 patients, consecutively hospitalized in our unit of Cardiology with a diagnosis of paroxysmal AF or acute PE, from April 2017 to October 2020. We recruited exclusively patients with normal cardiac function and HFpEF patients. Heart failure with reduced FEVS patients were excluded from the study. We have described for each patient the demographic and clinical characteristics, comorbidities, instrumental test results and clinical outcomes. In order to assess, for each group, the relationship between patient characteristics and clinical outcomes, the Chi-square test or alternatively the Pearson-Spearman correlation coefficients were calculated. Results the 194 patients studied had an average age of 73,7 years (min. 27, max 94). 59 AF patients had pre-existing HFpEF, whereas AF patients without HF were 67. Patients with pre-existing HFpEF and newly-onset AF had a more advanced age (76,7 y vs 72,9 y), and greater comorbidity (meanly 4 vs 3) rather than AF patients without HFpEF. Moreover, percentage of converting arrhythmia were significantly higher in AF patients without HFpEF. . Patients with acute PE and pre-existing HFpEF were 38, whereas PE patients without HF were 30. Acute PE patients with pre-existing HFpEF had older age, a prevalence for femal sex, more comorbidities, an average longer hospitalizations, but no significantly different rates of severe complications (ictus, hemorrhagies, needs for ventilation, pulmonary infarction or deaths) rather than PE patients without HFpEF. Conclusions the patients with AF or PE and concomitant HFpEF that were hospitalized from April 2017 to October 2020, showed an average longer hospitalization, a lower percentage of converting arrhythmia, probably due to the older age and the greater comorbidity.

Prediction of Paroxysmal Atrial Fibrillation From Complexity Analysis of the Sinus Rhythm ECG: A Retrospective Case/Control Pilot Study.

Paroxysmal atrial fibrillation (PAF) is the most common cardiac arrhythmia, conveying a stroke risk comparable to persistent AF. It poses a significant diagnostic challenge given its intermittency and potential brevity, and absence of symptoms in most patients. This pilot study introduces a novel biomarker for early PAF detection, based upon analysis of sinus rhythm ECG waveform complexity. Sinus rhythm ECG recordings were made from 52 patients with (n = 28) or without (n = 24) a subsequent diagnosis of PAF. Subjects used a handheld ECG monitor to record 28-second periods, twice-daily for at least 3 weeks. Two independent ECG complexity indices were calculated using a Lempel-Ziv algorithm: R-wave interval variability (beat detection, BD) and complexity of the entire ECG waveform (threshold crossing, TC). TC, but not BD, complexity scores were significantly greater in PAF patients, but TC complexity alone did not identify satisfactorily individual PAF cases. However, a composite complexity score (h-score) based on within-patient BD and TC variability scores was devised. The h-score allowed correct identification of PAF patients with 85% sensitivity and 83% specificity. This powerful but simple approach to identify PAF sufferers from analysis of brief periods of sinus-rhythm ECGs using hand-held monitors should enable easy and low-cost screening for PAF with the potential to reduce stroke occurrence.