xECGArch: a trustworthy deep learning architecture for interpretable ECG analysis considering short-term and long-term features.

Deep learning-based methods have demonstrated high classification performance in the detection of cardiovascular diseases from electrocardiograms (ECGs). However, their blackbox character and the associated lack of interpretability limit their clinical applicability. To overcome existing limitations, we present a novel deep learning architecture for interpretable ECG analysis (xECGArch). For the first time, short- and long-term features are analyzed by two independent convolutional neural networks (CNNs) and combined into an ensemble, which is extended by methods of explainable artificial intelligence (xAI) to whiten the blackbox. To demonstrate the trustworthiness of xECGArch, perturbation analysis was used to compare 13 different xAI methods. We parameterized xECGArch for atrial fibrillation (AF) detection using four public ECG databases ( n = 9854 ECGs) and achieved an F1 score of 95.43% in AF versus non-AF classification on an unseen ECG test dataset. A systematic comparison of xAI methods showed that deep Taylor decomposition provided the most trustworthy explanations ( + 24 % compared to the second-best approach). xECGArch can account for short- and long-term features corresponding to clinical features of morphology and rhythm, respectively. Further research will focus on the relationship between xECGArch features and clinical features, which may help in medical applications for diagnosis and therapy.

Screening for Unrecognized HFpEF in Atrial Fibrillation and for Unrecognized Atrial Fibrillation in HFpEF.

Because of the bidirectional relationship between atrial fibrillation (AF) and heart failure with preserved ejection fraction (HFpEF), individuals with either condition require consideration of screening for the other. In this review, we summarize current evidence and rationale for screening for occult HFpEF in adults with clinical AF; and occult AF in patients with clinically recognized HFpEF. Assessment of pretest probability for occult HFpEF in symptomatic AF patients may help guide additional testing such as exercise right heart catheterization to diagnose HFpEF and guide HFpEF-specific therapies. In patients with HFpEF, AF screening will identify cases of occult AF where anticoagulation may decrease stroke risk, and correlation of previously unknown AF episodes with paroxysmal symptoms may prompt consideration for rhythm control. Therefore, screening may help clinicians understand the etiology of the often-overlapping symptoms, and it may help guide treatments to slow progression of both conditions and their complications.

Left atrial function in middle-aged men and women with and without paroxysmal atrial fibrillation: Data from the Akershus Cardiac Examination (ACE) 1950 study.

PURPOSE: To assess left atrial (LA) function in individuals with known paroxysmal atrial fibrillation (AF) compared with healthy and nonhealthy individuals without atrial fibrillation. METHODS: The Akershus Cardiac Examination 1950 Study included 3,706 individuals all born in 1950. LA strain assessment of reservoir (LASr), conduit (LAScd) and contractile (LASct) functions were performed in all participants by investigators blinded to clinical data. Participants with cardiovascular disease, obesity, diabetes, pulmonary or renal disease were defined as nonhealthy, and those without as healthy. Patients with paroxysmal AF were identified through medical history and ECG documentation. RESULTS: LA strain assessment was feasible in 3,229 (87%) of the participants (50% women). The healthy group (n = 758) had significantly higher LASr and LAScd than the nonhealthy (n = 2,376), but LASct was similar between the groups. Participants with paroxysmal AF had significantly lower values of all strain parameters than the other groups. Multivariable logistic regression showed a significantly reduced probability of having AF per standard deviation increase in LASr and LASct. A nonlinear restricted cubic spline model fitted better with the association of LASr with paroxysmal AF than the linear model, and LA strain values below the population mean associated with an increased probability of having AF, but for values above the population mean no such association was present. CONCLUSION: Compared to participants without AF, those with known paroxysmal AF had significantly lower values of all LA strain parameters during sinus rhythm. Lower values of LA strain were associated with a significantly increased probability of having AF.

Multiscale dilated convolutional neural network for Atrial Fibrillation detection.

Atrial Fibrillation (AF), a type of heart arrhythmia, becomes more common with aging and is associated with an increased risk of stroke and mortality. In light of the urgent need for effective automated AF monitoring, existing methods often fall short in balancing accuracy and computational efficiency. To address this issue, we introduce a framework based on Multi-Scale Dilated Convolution (AF-MSDC), aimed at achieving precise predictions with low cost and high efficiency. By integrating Multi-Scale Dilated Convolution (MSDC) modules, our model is capable of extracting features from electrocardiogram (ECG) datasets across various scales, thus achieving an optimal balance between precision and computational savings. We have developed three MSDC modules to construct the AF-MSDC framework and assessed its performance on renowned datasets, including the MIT-BIH Atrial Fibrillation Database and Physionet Challenge 2017. Empirical results unequivocally demonstrate that our technique surpasses existing state-of-the-art (SOTA) methods in the AF detection domain. Specifically, our model, with only a quarter of the parameters of a Residual Network (ResNet), achieved an impressive sensitivity of 99.45%, specificity of 99.64% (on the MIT-BIH AFDB dataset), and an [Formula: see text] score of 85.63% (on the Physionet Challenge 2017 AFDB dataset). This high efficiency makes our model particularly suitable for integration into wearable ECG devices powered by edge computing frameworks. Moreover, this innovative approach offers new possibilities for the early diagnosis of AF in clinical applications, potentially improving patient quality of life and reducing healthcare costs.

Observational study protocol for an arrhythmia notification feature.

INTRODUCTION: Screening for atrial fibrillation (AF) in the general population may help identify individuals at risk, enabling further assessment of risk factors and institution of appropriate treatment. Algorithms deployed on wearable technologies such as smartwatches and fitness bands may be trained to screen for such arrhythmias. However, their performance needs to be assessed for safety and accuracy prior to wide-scale implementation. METHODS AND ANALYSIS: This study will assess the ability of the WHOOP strap to detect AF using its WHOOP Arrhythmia Notification Feature (WARN) algorithm in an enriched cohort with a 2:1 distribution of previously diagnosed AF (persistent and paroxysmal) and healthy controls. Recruited participants will collect data for 7 days with the WHOOP wrist-strap and BioTel ePatch (electrocardiography gold-standard). Primary outcome will be participant level sensitivity and specificity of the WARN algorithm in detecting AF in analysable windows compared with the ECG gold-standard. Similar analyses will be performed on an available epoch-level basis as well as comparison of these findings in important subgroups. ETHICS AND DISSEMINATION: The study was approved by the ethics board at the study site. Participants will be enrolled after signing an online informed consent document. Updates will be shared via clinicaltrials.gov. The data obtained from the conclusion of this study will be presented in national and international conferences with publication in clinical research journals. TRIAL REGISTRATION NUMBER: NCT05809362.

Left atrial appendage closure in a patient previously implanted with an interatrial shunt device: a case report.

Patients with previous interatrial shunt device (IASD) implantation may face greater challenges during future left atrial interventional procedures. Herein, we report the first case of left atrial appendage closure (LAAC) in a patient with previous IASD implantation. The patient successfully underwent LAAC using the LAmbre device without complications.

Opportunistic atrial fibrillation screening in primary care in Ireland: results of a pilot screening programme.

BACKGROUND: Atrial fibrillation (AF), a common, frequently asymptomatic cardiac arrhythmia, is a major risk factor for stroke. Identification of AF enables effective preventive treatment to be offered, potentially reducing stroke risk by up to two-thirds. There is international consensus that opportunistic AF screening is valuable though uncertainty remains about the optimum screening location and method. Primary care has been identified as a potential location for AF screening using one-lead ECG devices. METHODS: A pilot AF screening programme is in primary care in the south of Ireland. General practitioners (GPs) were recruited from Cork and Kerry. GPs invited patients ≥65 years to undergo AF screening. The screening comprised a one-lead ECG device, Kardia Mobile, blood pressure check and ascertainment of smoking status. Possible AF on one-lead ECG was confirmed with a 12-lead ECG. GPs also recorded information including medical history, current medication and onward referral. The Keele Decision Support tool was used to assess patients for oral anticoagulation (OAC). RESULTS: 3555 eligible patients, attending 52 GPs across 34 GP practices, agreed to undergo screening. 1720 (48%) were female, 1780 (50%) were hypertensive and 285 (8%) were current smokers. On the one-lead ECG, 3282 (92%) were in normal sinus rhythm, 101 (3%) had possible AF and among 124 (4%) the one-lead ECG was unreadable or unclassified. Of the 101 patients with possible AF, 45 (45%) had AF confirmed with 12-lead ECG, an incidence rate of AF of 1.3%. Among the 45 confirmed AF cases, 27 (60%) were commenced on OAC therapy by their GP. CONCLUSION: These findings suggest that AF screening in primary care may prove useful for early detection of AF cases that can be assessed for treatment. One-lead ECG devices may be useful in the detection of paroxysmal AF in this population and setting. Current OAC of AF may be suboptimal.

Diagnostic accuracy of digital technologies compared with 12-lead ECG in the diagnosis of atrial fibrillation in adults: A protocol for a systematic review.

BACKGROUND: Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia in the world. AF increases the risk of stroke 5-fold, though the risk can be reduced with appropriate treatment. Therefore, early diagnosis is imperative but remains a global challenge. In low-and middle-income countries (LMICs), a lack of diagnostic equipment and under-resourced healthcare systems generate further barriers. The rapid development of digital technologies that are capable of diagnosing AF remotely and cost-effectively could prove beneficial for LMICs. However, evidence is lacking on what digital technologies exist and how they compare in regards to diagnostic accuracy. We aim to systematically review the diagnostic accuracy of all digital technologies capable of AF diagnosis. METHODS: MEDLINE, Embase and Web of Science will be searched for eligible studies. Free text terms will be combined with corresponding index terms where available and searches will not be limited by language nor time of publication. Cohort or cross-sectional studies comprising adult (≥18 years) participants will be included. Only studies that use a 12-lead ECG as the reference test (comparator) and report outcomes of sensitivity, specificity, the diagnostic odds ratio (DOR) or the positive and negative predictive value (PPV and NPV) will be included (or if they provide sufficient data to calculate these outcomes). Two reviewers will independently assess articles for inclusion, extract data using a piloted tool and assess risk of bias using the QUADAS-2 tool. The feasibility of a meta-analysis will be determined by assessing heterogeneity across the studies, grouped by index device, diagnostic threshold and setting. If a meta-analysis is feasible for any index device, pooled sensitivity and specificity will be calculated using a random effect model and presented in forest plots. DISCUSSION: The findings of our review will provide a comprehensive synthesis of the diagnostic accuracy of available digital technologies capable for diagnosing AF. Thus, this review will aid in the identification of which devices could be further trialed and implemented, particularly in a LMIC setting, to improve the early diagnosis of AF. TRIAL REGISTRATION: Systematic review registration: PROSPERO registration number is CRD42021290542. https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021290542.

Coronary Embolism Among Patients With ST-Segment-Elevation Myocardial Infarction and Atrial Fibrillation: An Underrecognized But Deadly Association.

BACKGROUND: The prevalence and impact of coronary emboli (CE) in patients with ST-segment-elevation myocardial infarction (STEMI) and atrial fibrillation (AF) have not been specifically studied. The objective was to describe the clinical characteristics and outcomes of patients with AF and CE in a large series of patients with STEMI. METHODS AND RESULTS: We investigated 2292 consecutive patients with STEMI and among them 225 patients with AF: 46 patients with a STEMI related to CE (group A) and 179 patients with a STEMI related to an atherosclerotic cause (group B). Compared with the 2067 patients without AF and CE (group C), patients with AF and CE were older (73 versus 59 years, P<0.05), more likely to be female (43% versus 22%, P<0.05), and presented more frequently with cardiogenic shock at admission (26% versus 9%, P<0.05). The baseline characteristics of patients with AF (group A versus B) did not differ significantly according to STEMI pathogenesis. In the unadjusted analysis, the 45-day mortality was higher in patients with CE and AF (group A versus group C: 20% versus 4%; P<0.05 and group A versus group B: 20% versus 8%, P=not significant); this trend persisted at 2-year follow-up (group A versus group C: 24% versus 6%; P<0.05 and group A versus group B: 24% versus 17%, P=not significant). After stabilized inverse exposure probability weighting adjustment, a higher 45-day mortality rate was confirmed in patients with CE and AF (group A versus group C: 18% versus 5%, P<0.05). CONCLUSIONS: In patients presenting with STEMI and AF, CE was associated with excess early mortality. REGISTRATION: URL: clinicaltrials.gov. Identifier: NCT05679843.

Association Between Remnant Cholesterol and Risk of Incident Atrial Fibrillation: Population-Based Evidence From a Large-Scale Prospective Cohort Study.

BACKGROUND: Evidence for the relationship between remnant cholesterol (RC) and incident atrial fibrillation (AF) risk remains sparse and limited. METHODS AND RESULTS: Participants were enrolled between 2006 and 2010 and followed up to 2021. The multivariable Cox proportional hazards model was used to examine the relationship between RC quartiles and risk of incident AF. Subgroup analyses and sensitivity analyses were performed to explore the potential modification of the association and the robustness of the main findings. A total of 422 316 participants (mean age, 56 years; 54% women) were included for analyses. During a median follow-up of 11.9 years (first quartile-third quartile, 11.6-13.2 years), there were 24 774 AF events documented with an incidence of 4.92 events per 1000 person-years (95% CI, 4.86-4.98). Participants in higher RC quartiles had a lower risk of incident AF than those in the lowest quartile (first quartile): hazard ratio (HR)=0.96 (95% CI, 0.91-1.00) for second quartile; HR=0.92 (95% CI, 0.88-0.96) for third quartile; and HR=0.85 (95% CI, 0.81-0.89) for fourth quartile (P for trend <0.001). The association between RC quartiles and risk of incident AF was stronger in participants aged ≥65 years, in men, and in participants without history of diabetes when compared with control groups (P<0.001 for interaction). CONCLUSIONS: On the basis of data from this large-scale prospective cohort study, elevated RC was associated with a lower risk of incident AF.

Shifts in Clinical Characteristics, Treatment, and Outcome for Rheumatic Mitral Stenosis: Insights From a 20-Year Multicentre Registry Study in Korea.

BACKGROUND: The rapid economic development of South Korea provides a unique model to study changes in the clinical characteristics, treatment approaches, and clinical outcomes of patients with rheumatic mitral stenosis (MS) relative to socioeconomic growth. METHODS: From the Multicenter mitrAl STEnosis with Rheumatic etiology (MASTER) registry, 2,337 patients diagnosed with moderate or severe rheumatic MS between January 2001 and December 2020 were analyzed. Patients were grouped into consecutive 5-year intervals based on their year of diagnosis. Clinical characteristics, echocardiographic data, and clinical outcomes were assessed. RESULTS: Over 20 years, the severity of mitral stenosis increased from 79.1% to 90.2%; similarly, the average age at diagnosis increased from 54.3 to 63.0 years (all P < 0.001). Comorbidities such as hypertension and atrial fibrillation increased (6.3% to 29.5% and 41.4% to 46.9%, respectively; all P for trend < 0.05). The rate of mitral intervention within five years after diagnosis increased from 31.2% to 47.4% (P for trend < 0.001). However, clinical outcomes of rheumatic mitral stenosis deteriorated over time in the composite outcomes (log-rank test, P < 0.001). Conversely, the incidence of stroke remained stable (60.6-73.7%; P < 0.001), which might be attributed to the increased use of anticoagulation therapy. CONCLUSION: This study observed an increase in patient age, comorbidities, and valve disease severity as the country transitioned from a developing to developed status. Despite a rise in mitral valve interventions, clinical outcomes deteriorated over 20 years, highlighting the need for modified treatment approaches to improve patient outcomes.

New-onset atrial fibrillation detected by ambulatory ECG monitoring after transcatheter aortic valve implantation.

BACKGROUND: Little is known about the occurrence of subclinical new-onset atrial fibrillation (NOAF) after transcatheter aortic valve implantation (TAVI). AIMS: We aimed to evaluate the incidence, predictors, and clinical impact of subclinical NOAF after TAVI. METHODS: This was a multicentre study, including patients with aortic stenosis (AS) and no previous atrial fibrillation undergoing TAVI, with continuous ambulatory electrocardiogram (AECG) monitoring after TAVI. RESULTS: A total of 700 patients (79±8 years, 49% female, Society of Thoracic Surgeons score 2.9% [1.9-4.0]) undergoing transarterial TAVI were included (85% balloon-expandable valves). AECG was started 1 (0-1) day after TAVI (monitoring time: 14121314 days). NOAF was detected in 49 patients (7%), with a median duration of 185 (43-421) minutes (atrial fibrillation burden of 0.7% [0.3-2.8]). Anticoagulation was started in 25 NOAF patients (51%). No differences were found in baseline or procedural characteristics, except for a higher AS severity in the NOAF group (peak gradient: no NOAF: 71.9±23.5 mmHg vs NOAF: 85.2±23.8 mmHg; p=0.024; mean gradient: no NOAF: 44.4±14.7 mmHg vs NOAF: 53.8±16.8 mmHg; p=0.004). In the multivariable analysis, the baseline mean transaortic gradient was associated with a higher risk of NOAF after TAVI (odds ratio 1.04, 95% confidence interval: 1.01-1.06 for each mmHg; p=0.006). There were no differences between groups in all-cause mortality (no NOAF: 4.7% vs NOAF: 0%; p=0.122), stroke (no NOAF: 1.4% vs NOAF: 2.0%; p=0.723), or bleeding (no NOAF: 1.9% vs NOAF: 4.1%; p=0.288) from the 30-day to 1-year follow-up. CONCLUSIONS: NOAF detected with AECG occurred in 7% of TAVI recipients and was associated with a higher AS severity. NOAF detection determined the start of anticoagulation therapy in about half of the patients, and it was not associated with an increased risk of clinical events at 1-year follow-up.

Comparison between catheter ablation versus permanent pacemaker implantation as an initial treatment for tachycardia-bradycardia syndrome patients: a prospective, randomized trial.

BACKGROUND: Clinical outcomes after catheter ablation (CA) or pacemaker (PM) implantation for the tachycardia-bradycardia syndrome (TBS) has not been evaluated adequately. We tried to compare the efficacy and safety outcomes of CA and PM implantation as an initial treatment option for TBS in paroxysmal atrial fibrillation (AF) patients. METHODS: Sixty-eight patients with paroxysmal AF and TBS (mean 63.7 years, 63.2% male) were randomized, and received CA (n = 35) or PM (n = 33) as initial treatments. The primary outcomes were unexpected emergency room visits or hospitalizations attributed to cardiovascular causes. RESULTS: In the intention-to-treatment analysis, the rates of primary outcomes were not significantly different between the two groups at the 2-year follow-up (19.8% vs. 25.9%; hazard ratio (HR) 0.73, 95% confidence interval (CI) 0.25-2.20, P = 0.584), irrespective of whether the results were adjusted for age (HR 1.12, 95% CI 0.34-3.64, P = 0.852). The 2-year rate of recurrent AF was significantly lower in the CA group compared to the PM group (33.9% vs. 56.8%, P = 0.038). Four patients (11.4%) in the CA group finally received PMs after CA owing to recurrent syncope episodes. The rate of major or minor procedure related complications was not significantly different between the two groups. CONCLUSION: CA had a similar efficacy and safety profile with that of PM and a higher sinus rhythm maintenance rate. CA could be considered as a preferable initial treatment option over PM implantation in patients with paroxysmal AF and TBS. TRIAL REGISTRATION: KCT0000155.

Silent brain infarctions in patients with acute cardioembolic stroke.

INTRODUCTION AND AIM: The advances and the wide use of brain imaging have considerably increased the prevalence of silent brain infarctions (SBI). We aim in this study to determine the prevalence of SBI in patients presenting with acute cardioembolic stroke and the predictive cardiovascular risk factors. METHODS: This retrospective study included 267 patients presenting with acute cardioembolic stroke in the emergency and/or neurology departments of the Hassan II University Hospital Center. Clinical, biological and echocardiographic characteristics were recorded. All patients were screened for SBI by brain imaging. RESULTS: The prevalence of SBI in our series was 46%. A group of 203 non-valvular patients and a group of 64 valvular patients were distinguished. In non-valvular group, the average age was 72.97±10.53years. The prevalence of SBI was 45.3%. Forty-four percent of patients with SBI had atrial fibrillation (AF). In multivariate regression analysis, the history of previous stroke, CHA2DS2-VASc Score≥4, enlarged left atrium (LA), the association of AF with enlarged LA and the lability of International Normalized Ratio in patients initially treated with anticoagulants were significantly associated with the occurrence of SBI (P=0.013, P=0.032, P=0.0001, P=0.01, P=0.03, respectively). Territorial location was significantly the most frequent (P=0.007). In valvular group, the average age was 57.19±14.38years. The prevalence of SBI was 48.4%. In multivariate regression analysis, SBI were significantly associated with moderate or severe mitral stenosis (P=0.02) and with the enlarged LA (P=0.02). In all patients, Modified Rankin Scale at 3 months of discharge from the acute stroke was significantly higher (mRS≥3) in patients with SBI (P=0.04). CONCLUSIONS: SBI requires good management of associated cardiovascular risk factors in a population presenting with initial cardioembolic stroke.