Predictors of atrial fibrillation early recurrence following cryoballoon ablation of pulmonary veins using statistical assessment and machine learning algorithms

Role of Machine Learning and Artificial Intelligence in Arrhythmias and Electrophysiology

Machine learning (ML), a subset of artificial intelligence (AI) centered on machines learning from extensive datasets, stands at the forefront of a technological revolution shaping various facets of society. Cardiovascular medicine has emerged as a key domain for ML applications, with considerable efforts to integrate these innovations into routine clinical practice. Within cardiac electrophysiology, ML applications, especially in the automated interpretation of electrocardiograms, have garnered substantial attention in existing literature. However, less recognized are the diverse applications of ML in cardiac electrophysiology and arrhythmias, spanning basic science research on arrhythmia mechanisms, both experimental and computational, as well as contributions to enhanced techniques for mapping cardiac electrical function and translational research related to arrhythmia management. This comprehensive review delves into various ML applications within the scope of this journal, organized into 3 parts. The first section provides a fundamental understanding of general ML principles and methodologies, serving as a foundational resource for readers interested in exploring ML applications in arrhythmia research. The second part offers an in-depth review of studies in arrhythmia and electrophysiology that leverage ML methodologies, showcasing the broad potential of ML approaches. Each subject is thoroughly outlined, accompanied by a review of notable ML research advancements. Finally, the review delves into the primary challenges and future perspectives surrounding ML-driven cardiac electrophysiology and arrhythmias research.

Validation of a machine learning algorithm to identify pulmonary vein isolation during ablation procedures for the treatment of atrial fibrillation: results of the PVISION study

AIMS

Pulmonary vein isolation (PVI) is the cornerstone of ablation for atrial fibrillation. Confirmation of PVI can be challenging due to the presence of far-field electrograms (EGMs) and sometimes requires additional pacing manoeuvres or mapping. This prospective multicentre study assessed the agreement between a previously trained automated algorithm designed to determine vein isolation status with expert opinion in a real-world clinical setting.

METHODS AND RESULTS

Consecutive patients scheduled for PVI were recruited at four centres. The ECGenius electrophysiology (EP) recording system (CathVision ApS, Copenhagen, Denmark) was connected in parallel with the existing system in the laboratory. Electrograms from a circular mapping catheter were annotated during sinus rhythm at baseline pre-ablation, time of isolation, and post-ablation. The ground truth for isolation status was based on operator opinion. The algorithm was applied to the collected PV signals off-line and compared with expert opinion. The primary endpoint was a sensitivity and specificity exceeding 80%. Overall, 498 EGMs (248 at baseline and 250 at PVI) with 5473 individual PV beats from 89 patients (32 females, 62 ± 12 years) were analysed. The algorithm performance reached an area under the curve (AUC) of 92% and met the primary study endpoint with a sensitivity and specificity of 86 and 87%, respectively (P = 0.005; P = 0.004). The algorithm had an accuracy rate of 87% in classifying the time of isolation.

CONCLUSION

This study validated an automated algorithm using machine learning to assess the isolation status of pulmonary veins in patients undergoing PVI with different ablation modalities. The algorithm reached an AUC of 92%, with both sensitivity and specificity exceeding the primary study endpoints.

Predicting Adverse Outcomes Following Catheter Ablation Treatment for Atrial Flutter/Fibrillation

BACKGROUND & AIM

To develop prognostic survival models for predicting adverse outcomes after catheter ablation treatment for non-valvular atrial fibrillation (AF) and/or atrial flutter (AFL).

METHODS

We used a linked dataset including hospital administrative data, prescription medicine claims, emergency department presentations, and death registrations of patients in New South Wales, Australia. The cohort included patients who received catheter ablation for AF and/or AFL. Traditional and deep survival models were trained to predict major bleeding events and a composite of heart failure, stroke, cardiac arrest, and death.

RESULTS

Out of a total of 3,285 patients in the cohort, 177 (5.3%) experienced the composite outcome-heart failure, stroke, cardiac arrest, death-and 167 (5.1%) experienced major bleeding events after catheter ablation treatment. Models predicting the composite outcome had high-risk discrimination accuracy, with the best model having a concordance index >0.79 at the evaluated time horizons. Models for predicting major bleeding events had poor risk discrimination performance, with all models having a concordance index <0.66. The most impactful features for the models predicting higher risk were comorbidities indicative of poor health, older age, and therapies commonly used in sicker patients to treat heart failure and AF and AFL.

DISCUSSION

Diagnosis and medication history did not contain sufficient information for precise risk prediction of experiencing major bleeding events. Predicting the composite outcome yielded promising results, but future research is needed to validate the usefulness of these models in clinical practice.

CONCLUSIONS

Machine learning models for predicting the composite outcome have the potential to enable clinicians to identify and manage high-risk patients following catheter ablation for AF and AFL proactively.

Atrial fibrillation ablation outcome prediction with a machine learning fusion framework incorporating cardiac computed tomography

BACKGROUND

Structural changes in the left atrium (LA) modestly predict outcomes in patients undergoing catheter ablation for atrial fibrillation (AF). Machine learning (ML) is a promising approach to personalize AF management strategies and improve predictive risk models after catheter ablation by integrating atrial geometry from cardiac computed tomography (CT) scans and patient-specific clinical data. We hypothesized that ML approaches based on a patient's specific data can identify responders to AF ablation.

METHODS

Consecutive patients undergoing AF ablation, who had preprocedural CT scans, demographics, and 1-year follow-up data, were included in the study for a retrospective analysis. The inputs of models were CT-derived morphological features from left atrial segmentation (including the shape, volume of the LA, LA appendage, and pulmonary vein ostia) along with deep features learned directly from raw CT images, and clinical data. These were merged intelligently in a framework to learn their individual importance and produce the optimal classification.

RESULTS

Three hundred twenty-one patients (64.2 ± 10.6 years, 69% male, 40% paroxysmal AF) were analyzed. Post 10-fold nested cross-validation, the model trained to intelligently merge and learn appropriate weights for clinical, morphological, and imaging data (AUC 0.821) outperformed those trained solely on clinical data (AUC 0.626), morphological (AUC 0.659), or imaging data (AUC 0.764).

CONCLUSION

Our ML approach provides an end-to-end automated technique to predict AF ablation outcomes using deep learning from CT images, derived structural properties of LA, augmented by incorporation of clinical data in a merged ML framework. This can help develop personalized strategies for patient selection in invasive management of AF.

An Artificial Intelligence-Enabled ECG Algorithm for Predicting the Risk of Recurrence in Patients with Paroxysmal Atrial Fibrillation after Catheter Ablation

Background: Catheter ablation (CA) is an important treatment strategy to reduce the burden and complications of atrial fibrillation (AF). This study aims to predict the risk of recurrence in patients with paroxysmal AF (pAF) after CA by an artificial intelligence (AI)-enabled electrocardiography (ECG) algorithm. Methods and Results: 1618 ≥ 18 years old patients with pAF who underwent CA in Guangdong Provincial People's Hospital from 1 January 2012 to 31 May 2019 were enrolled in this study. All patients underwent pulmonary vein isolation (PVI) by experienced operators. Baseline clinical features were recorded in detail before the operation and standard follow-up (≥12 months) was conducted. The convolutional neural network (CNN) was trained and validated by 12-lead ECGs within 30 days before CA to predict the risk of recurrence. A receiver operating characteristic curve (ROC) was created for the testing and validation sets, and the predictive performance of AI-enabled ECG was assessed by the area under the curve (AUC). After training and internal validation, the AUC of the AI algorithm was 0.84 (95% CI: 0.78-0.89), with a sensitivity, specificity, accuracy, precision and balanced F Score (F1 score) of 72.3%, 95.0%, 92.0%, 69.1% and 0.707, respectively. Compared with current prognostic models (APPLE, BASE-AF2, CAAP-AF, DR-FLASH and MB-LATER), the performance of the AI algorithm was better (p < 0.01). Conclusions: The AI-enabled ECG algorithm seemed to be an effective method to predict the risk of recurrence in patients with pAF after CA. This is of great clinical significance in decision-making for personalized ablation strategies and postoperative treatment plans in patients with pAF.

Explainable machine learning model reveals its decision-making process in identifying patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation

BACKGROUND

A number of models have been reported for predicting atrial fibrillation (AF) recurrence after catheter ablation. Although many machine learning (ML) models were developed among them, black-box effect existed widely. It was always difficult to explain how variables affect model output. We sought to implement an explainable ML model and then reveal its decision-making process in identifying patients with paroxysmal AF at high risk for recurrence after catheter ablation.

METHODS

Between January 2018 and December 2020, 471 consecutive patients with paroxysmal AF who had their first catheter ablation procedure were retrospectively enrolled. Patients were randomly assigned into training cohort (70%) and testing cohort (30%). The explainable ML model based on Random Forest (RF) algorithm was developed and modified on training cohort, and tested on testing cohort. In order to gain insight into the association between observed values and model output, Shapley additive explanations (SHAP) analysis was used to visualize the ML model.

RESULTS

In this cohort, 135 patients experienced tachycardias recurrences. With hyperparameters adjusted, the ML model predicted AF recurrence with an area under the curve of 66.7% in the testing cohort. Summary plots listed the top 15 features in descending order and preliminary showed the association between features and outcome prediction. Early recurrence of AF showed the most positive impact on model output. Dependence plots combined with force plots showed the impact of single feature on model output, and helped determine high risk cut-off points. The thresholds of CHA₂DS₂-VASc score, systolic blood pressure, AF duration, HAS-BLED score, left atrial diameter and age were 2, 130 mmHg, 48 months, 2, 40 mm and 70 years, respectively. Decision plot recognized significant outliers.

CONCLUSION

An explainable ML model effectively revealed its decision-making process in identifying patients with paroxysmal atrial fibrillation at high risk for recurrence after catheter ablation by listing important features, showing the impact of every feature on model output, determining appropriate thresholds and identifying significant outliers. Physicians can combine model output, visualization of model and clinical experience to make better decision.

Comparison of Myocardial Injury and Inflammation Biomarkers and Their Impact on Recurrence after Cryoballoon and Radiofrequency Ablation for Atrial Fibrillation: A Systematic Review and Meta-Analysis

Background

Biomarkers of myocardial injury and inflammation were found to be different after radiofrequency catheter ablation (RFCA) and cryoballoon ablation (CBA) for atrial fibrillation (AF); however, the results are currently controversial. This study was aimed to systematically compare the differences in myocardial injury and inflammation biomarkers after RFCA and CBA procedures and to investigate their impact on AF recurrence.

Methods

Databases, including PubMed, Embase, the Cochrane Library, ClinicalTrials.gov, China National Knowledge Infrastructure (CNKI) and China Biology Medicine (CBM), were systematically searched from their date of inception to May 2022. The primary outcomes of interest were the differences in myocardial injury and inflammation biomarkers after CBA and RFCA procedures for AF patients, and the impact of the biomarkers on AF recurrence. Secondary outcomes included the total ablation time, the procedure duration and the freedom from atrial tachycardia (AT).

Results

Eighteen studies with a total of 1807 patients were finally enrolled. CBA treatment was associated with significantly greater increases in troponin I (TNI) levels (weighted mean difference [WMD] = 3.13 ug/L, 95% confidence interval [CI] 2.43-3.64) both at 4-6 h (WMD = 3.94 ug/L), 24 h (WMD = 4.23 ug/L), 48 h (WMD = 2.14 ug/L) and 72 h (WMD = 0.56 ug/L), and also creatine kinade MB fraction (CK-MB) levels at 4-6 h (WMD = 33.21 U/L), 24 h (WMD = 35.84 U/L) and 48 h (WMD = 4.62 U/L), while RFCA treatment was associated with greater increases in postablation C-reactive protein (CRP) levels both at 48 h (WMD = -9.32 mg/L) and 72 h (WMD = -10.90 mg/L). The CBA and RFCA treatments had comparable rates of freedom from AT (74.5% vs. 75.2%, RR = 1.08). The CRP levels were significantly higher in patients with early recurrence of AF (ERAF) than in those without ERAF after RFCA treatment (WMD = 3.415 mg/L).

Conclusions

The time-course patterns of postablation myocardial injury and inflammation biomarkers are different between RFCA and CBA procedures. The lower postprocedural elevation of myocardial injury biomarkers and the increased CRP levels may be predictive factors for ERAF.

PROSPERO Registration Number

CRD42021278564.

Machine Learning-Enabled Multimodal Fusion of Intra-Atrial and Body Surface Signals in Prediction of Atrial Fibrillation Ablation Outcomes

BACKGROUND

Machine learning is a promising approach to personalize atrial fibrillation management strategies for patients after catheter ablation. Prior atrial fibrillation ablation outcome prediction studies applied classical machine learning methods to hand-crafted clinical scores, and none have leveraged intracardiac electrograms or 12-lead surface electrocardiograms for outcome prediction. We hypothesized that (1) machine learning models trained on electrograms or electrocardiogram (ECG) signals can perform better at predicting patient outcomes after atrial fibrillation ablation than existing clinical scores and (2) multimodal fusion of electrogram, ECG, and clinical features can further improve the prediction of patient outcomes.

METHODS

Consecutive patients who underwent catheter ablation between 2015 and 2017 with panoramic left atrial electrogram before ablation and clinical follow-up for at least 1 year following ablation were included. Convolutional neural network and a novel multimodal fusion framework were developed for predicting 1-year atrial fibrillation recurrence after catheter ablation from electrogram, ECG signals, and clinical features. The models were trained and validated using 10-fold cross-validation on patient-level splits.

RESULTS

One hundred fifty-six patients (64.5±10.5 years, 74% male, 42% paroxysmal) were analyzed. Using electrogram signals alone, the convolutional neural network achieved an area under the receiver operating characteristics curve (AUROC) of 0.731, outperforming the existing APPLE scores (AUROC=0.644) and CHA2DS2-VASc scores (AUROC=0.650). Similarly using 12-lead ECG alone, the convolutional neural network achieved an AUROC of 0.767. Combining electrogram, ECG, and clinical features, the fusion model achieved an AUROC of 0.859, outperforming single and dual modality models.

CONCLUSIONS

Deep neural networks trained on electrogram or ECG signals improved the prediction of catheter ablation outcome compared with existing clinical scores, and fusion of electrogram, ECG, and clinical features further improved the prediction. This suggests the promise of using machine learning to help treatment planning for patients after catheter ablation.

Artificial Intelligence in Cardiology-A Narrative Review of Current Status

Artificial intelligence (AI) is an integral part of clinical decision support systems (CDSS), offering methods to approximate human reasoning and computationally infer decisions. Such methods are generally based on medical knowledge, either directly encoded with rules or automatically extracted from medical data using machine learning (ML). ML techniques, such as Artificial Neural Networks (ANNs) and support vector machines (SVMs), are based on mathematical models with parameters that can be optimally tuned using appropriate algorithms. The ever-increasing computational capacity of today’s computer systems enables more complex ML systems with millions of parameters, bringing AI closer to human intelligence. With this objective, the term deep learning (DL) has been introduced to characterize ML based on deep ANN (DNN) architectures with multiple layers of artificial neurons. Despite all of these promises, the impact of AI in current clinical practice is still limited. However, this could change shortly, as the significantly increased papers in AI, machine learning and deep learning in cardiology show. We highlight the significant achievements of recent years in nearly all areas of cardiology and underscore the mounting evidence suggesting how AI will take a central stage in the field.

Machine Learning-Predicted Progression to Permanent Atrial Fibrillation After Catheter Ablation

Introduction

We developed a prediction model for atrial fibrillation (AF) progression and tested whether machine learning (ML) could reproduce the prediction power in an independent cohort using pre-procedural non-invasive variables alone.

Methods

Cohort 1 included 1,214 patients and cohort 2, 658, and all underwent AF catheter ablation (AFCA). AF progression to permanent AF was defined as sustained AF despite repeat AFCA or cardioversion under antiarrhythmic drugs. We developed a risk stratification model for AF progression (STAAR score) and stratified cohort 1 into three groups. We also developed an ML-prediction model to classify three STAAR risk groups without invasive parameters and validated the risk score in cohort 2.

Results

The STAAR score consisted of a stroke (2 points, p = 0.003), persistent AF (1 point, p = 0.049), left atrial (LA) dimension ≥43 mm (1 point, p = 0.010), LA voltage <1.109 mV (2 points, p = 0.004), and PR interval ≥196 ms (1 point, p = 0.001), based on multivariate Cox analyses, and it had a good discriminative power for progression to permanent AF [area under curve (AUC) 0.796, 95% confidence interval (CI): 0.753–0.838]. The ML prediction model calculated the risk for AF progression without invasive variables and achieved excellent risk stratification: AUC 0.935 for low-risk groups (score = 0), AUC 0.855 for intermediate-risk groups (score 1–3), and AUC 0.965 for high-risk groups (score ≥ 4) in cohort 1. The ML model successfully predicted the high-risk group for AF progression in cohort 2 (log-rank p < 0.001).

Conclusions

The ML-prediction model successfully classified the high-risk patients who will progress to permanent AF after AFCA without invasive variables but has a limited discrimination power for the intermediate-risk group.

Critical appraisal of technologies to assess electrical activity during atrial fibrillation: a position paper from the European Heart Rhythm Association and European Society of Cardiology Working Group on eCardiology in collaboration with the Heart Rhythm Society, Asia Pacific Heart Rhythm Society, Latin American Heart Rhythm Society and Computing in Cardiology

We aim to provide a critical appraisal of basic concepts underlying signal recording and processing technologies applied for (i) atrial fibrillation (AF) mapping to unravel AF mechanisms and/or identifying target sites for AF therapy and (ii) AF detection, to optimize usage of technologies, stimulate research aimed at closing knowledge gaps, and developing ideal AF recording and processing technologies. Recording and processing techniques for assessment of electrical activity during AF essential for diagnosis and guiding ablative therapy including body surface electrocardiograms (ECG) and endo- or epicardial electrograms (EGM) are evaluated. Discussion of (i) differences in uni-, bi-, and multi-polar (omnipolar/Laplacian) recording modes, (ii) impact of recording technologies on EGM morphology, (iii) global or local mapping using various types of EGM involving signal processing techniques including isochronal-, voltage- fractionation-, dipole density-, and rotor mapping, enabling derivation of parameters like atrial rate, entropy, conduction velocity/direction, (iv) value of epicardial and optical mapping, (v) AF detection by cardiac implantable electronic devices containing various detection algorithms applicable to stored EGMs, (vi) contribution of machine learning (ML) to further improvement of signals processing technologies. Recording and processing of EGM (or ECG) are the cornerstones of (body surface) mapping of AF. Currently available AF recording and processing technologies are mainly restricted to specific applications or have technological limitations. Improvements in AF mapping by obtaining highest fidelity source signals (e.g. catheter-electrode combinations) for signal processing (e.g. filtering, digitization, and noise elimination) is of utmost importance. Novel acquisition instruments (multi-polar catheters combined with improved physical modelling and ML techniques) will enable enhanced and automated interpretation of EGM recordings in the near future.

Predictors of recurrence of atrial fibrillation within the first 3 months after ablation

Aims

Freedom from atrial fibrillation (AF) at 1 year can be achieved in 50–70% of patients undergoing catheter ablation. Recurrent AF early after ablation most commonly terminates spontaneously without further interventional treatment but is associated with later recurrent AF. The aim of this investigation is to identify clinical and procedural factors associated with recurrence of AF early after ablation.

Methods and results

We retrospectively analysed data for recurrence of AF within the first 3 months after catheter ablation from the randomized controlled AXAFA–AFNET 5 trial, which demonstrated that continuous anticoagulation with apixaban is as safe and as effective compared to vitamin K antagonists in 678 patients undergoing first AF ablation. The primary outcome of first recurrent AF within 90 days was observed in 163 (28%) patients, in which 78 (48%) patients experienced an event within the first 14 days post-ablation. After multivariable adjustment, a history of stroke/transient ischaemic attack [hazard ratio (HR) 1.54, 95% confidence interval (CI) 0.93–2.6; P = 0.11], coronary artery disease (HR 1.85, 95% CI 1.20–2.86; P = 0.005), cardioversion during ablation (HR 1.78, 95% CI 1.26–2.49; P = 0.001), and an age:sex interaction for older women (HR 1.01, 95% CI 1.00–1.01; P = 0.04) were associated with recurrent AF. The P-wave duration at follow-up was significantly longer for patients with AF recurrence (129 ± 31 ms vs. 122 ± 22 ms in patients without AF, P = 0.03).

Conclusion

Half of all early AF recurrences within the first 3 months post-ablation occurred within the first 14 days post-ablation. Vascular disease and cardioversion during the procedure are strong predictors of recurrent AF. P-wave duration at follow-up was longer in patients with recurrent AF.

Trial registration

Clinicaltrials.gov identifier NCT02227550

Predictors of Early-Recurrence Atrial Fibrillation after Catheter Ablation in Women and Men with Abnormal Body Weight

Our study aimed to select factors that affect the rate of early recurrence (up to 3 months) of atrial fibrillation (AF) (ERAF) following pulmonary veins isolation (PVI) in obese women and men. The study comprised 114 patients: 54 women (age: 63.8 ± 6.3, BMI 31 ± 4 kg/m²), and 60 men (age: 60.7 ± 6.7; BMI 31 ± 3 kg/m²) with paroxysmal, persistent and long-standing persistent AF. They had been scheduled to undergo cryoballoon (men n = 30; women n = 30) and radiofrequency (RF) ablation (men n = 30; women n = 24) using the CARTO-mapping. The blood was collected at baseline and 24 h after ablation. The rate of ERAF was comparable after cryoballoon and RF ablation and constituted 18% in women and 22% in men. Almost 70 parameters were selected to perform univariate and multivariate analysis and to create a multivariate logistic regression (MLR) model of ERAF in the obese men and women. The MLR analysis was performed by forward stepwise logistic regression with three variables. It was only possible to create the MLR model for the group of obese men. It revealed a poor predictive value with an unsatisfactory sensitivity of 31%. Men with ERAF: smokers (OR 39.25, 95% CI 1.050-1467.8, p = 0.0021), with a higher ST2 elevation (OR 1.68, 95% CI 1.115-2.536, p = 0.0021) who received dihydropyridine calcium channel blockers (OR 0.042, 95% CI 0.002-1.071, p = 0.0021) less frequently. Our results indicate a complex pathogenesis of ERAF dependent on the patients' gender.

Computational models of atrial fibrillation: achievements, challenges, and perspectives for improving clinical care

Despite significant advances in its detection, understanding and management, atrial fibrillation (AF) remains a highly prevalent cardiac arrhythmia with a major impact on morbidity and mortality of millions of patients. AF results from complex, dynamic interactions between risk factors and comorbidities that induce diverse atrial remodelling processes. Atrial remodelling increases AF vulnerability and persistence, while promoting disease progression. The variability in presentation and wide range of mechanisms involved in initiation, maintenance and progression of AF, as well as its associated adverse outcomes, make the early identification of causal factors modifiable with therapeutic interventions challenging, likely contributing to suboptimal efficacy of current AF management. Computational modelling facilitates the multilevel integration of multiple datasets and offers new opportunities for mechanistic understanding, risk prediction and personalized therapy. Mathematical simulations of cardiac electrophysiology have been around for 60 years and are being increasingly used to improve our understanding of AF mechanisms and guide AF therapy. This narrative review focuses on the emerging and future applications of computational modelling in AF management. We summarize clinical challenges that may benefit from computational modelling, provide an overview of the different in silico approaches that are available together with their notable achievements, and discuss the major limitations that hinder the routine clinical application of these approaches. Finally, future perspectives are addressed. With the rapid progress in electronic technologies including computing, clinical applications of computational modelling are advancing rapidly. We expect that their application will progressively increase in prominence, especially if their added value can be demonstrated in clinical trials.

How machine learning is impacting research in atrial fibrillation: implications for risk prediction and future management

There has been an exponential growth of artificial intelligence (AI) and machine learning (ML) publications aimed at advancing our understanding of atrial fibrillation (AF), which has been mainly driven by the confluence of two factors: the advances in deep neural networks (DeepNNs) and the availability of large, open access databases. It is observed that most of the attention has centred on applying ML for dvsetecting AF, particularly using electrocardiograms (ECGs) as the main data modality. Nearly a third of them used DeepNNs to minimize or eliminate the need for transforming the ECGs to extract features prior to ML modelling; however, we did not observe a significant advantage in following this approach. We also found a fraction of studies using other data modalities, and others centred in aims, such as risk prediction, AF management, and others. From the clinical perspective, AI/ML can help expand the utility of AF detection and risk prediction, especially for patients with additional comorbidities. The use of AI/ML for detection and risk prediction into applications and smart mobile health (mHealth) technology would enable ‘real time’ dynamic assessments. AI/ML could also adapt to treatment changes over time, as well as incident risk factors. Incorporation of a dynamic AI/ML model into mHealth technology would facilitate ‘real time’ assessment of stroke risk, facilitating mitigation of modifiable risk factors (e.g. blood pressure control). Overall, this would lead to an improvement in clinical care for patients with AF.

Machine Learning in Arrhythmia and Electrophysiology

Machine learning (ML), a branch of artificial intelligence, where machines learn from big data, is at the crest of a technological wave of change sweeping society. Cardiovascular medicine is at the forefront of many ML applications, and there is a significant effort to bring them into mainstream clinical practice. In the field of cardiac electrophysiology, ML applications have also seen a rapid growth and popularity, particularly the use of ML in the automatic interpretation of ECGs, which has been extensively covered in the literature. Much lesser known are the other aspects of ML application in cardiac electrophysiology and arrhythmias, such as those in basic science research on arrhythmia mechanisms, both experimental and computational; in the development of better techniques for mapping of cardiac electrical function; and in translational research related to arrhythmia management. In the current review, we examine comprehensively such ML applications as they match the scope of this journal. The current review is organized in 3 parts. The first provides an overview of general ML principles and methodologies that will afford readers of the necessary information on the subject, serving as the foundation for inviting further ML applications in arrhythmia research. The basic information we provide can serve as a guide on how one might design and conduct an ML study. The second part is a review of arrhythmia and electrophysiology studies in which ML has been utilized, highlighting the broad potential of ML approaches. For each subject, we outline comprehensively the general topics, while reviewing some of the research advances utilizing ML under the subject. Finally, we discuss the main challenges and the perspectives for ML-driven cardiac electrophysiology and arrhythmia research.

Artificial intelligence for a personalized diagnosis and treatment of atrial fibrillation

Although atrial fibrillation (AF) is the most common cardiac arrhythmia, its early identification, diagnosis, and treatment is still challenging. Due to its heterogeneous mechanisms and risk factors, targeting an individualized treatment of AF demands a large amount of patient data to identify specific patterns. Artificial intelligence (AI) algorithms are particularly well suited for treating high-dimensional data, predicting outcomes, and eventually, optimizing strategies for patient management. The analysis of large patient samples combining different sources of information such as blood biomarkers, electrical signals, and medical images opens a new paradigm for improving diagnostic algorithms. In this review, we summarize suitable AI techniques for this purpose. In particular, we describe potential applications for understanding the structural and functional bases of the disease, as well as for improving early noninvasive diagnosis, developing more efficient therapies, and predicting long-term clinical outcomes of patients with AF.

General anesthesia is not superior to sedation in clinical outcome and cost-effectiveness for ablation of persistent atrial fibrillation

Background

The strategy of anesthesia used during ablation of atrial fibrillation (AF) remains controversial. This study aimed to compare sedation with general anesthesia (GA) for catheter ablation of AF.

Hypothesis

The presence of AF is associated with an increased risk of stroke and heart failure and decreased quality of life and survival.

Methods

We carried out a retrospective single‐centered study with 351 patients undergoing the first ablation procedure for AF under sedation or GA. The main outcome was freedom from recurrence of AF at 1 year. The total time of staying at the ablation laboratory and procedure cost were also calculated.

Results

Freedom from atrial arrhythmia and ablation time did not differ between AF patients under sedation and GA (77.9% vs 79.9% and 42.27 ± 9.84 minutes vs 41.51 ± 9.27 minutes, respectively), while the total procedure time and cost were lower in patients who underwent sedation than GA (171.39 ± 45.09 minutes vs 202.92 ± 43.85 and 8.00 ± 7.02 CNY vs 8.79 ± 11.63 CNY, respectively).

Conclusion

GA is not superior to sedation, in terms of ablation time and freedom from atrial arrhythmia at 1 year, whereas patients with GA had more anesthesia time and procedure cost than sedation.

Predictive role of early recurrence of atrial fibrillation after cryoballoon ablation

Aims

The aims of this study were to determine the rate and the predictors of early recurrences of atrial fibrillation (ERAF) after cryoballoon (CB) ablation and to evaluate whether ERAF correlate with the long-term outcome.

Methods and results

Three thousand, six hundred, and eighty-one consecutive patients (59.9 ± 10.5 years, female 26.5%, and 74.3% paroxysmal AF) were included in the analysis. Atrial fibrillation recurrence, lasting at least 30 s, was collected during and after the 3-month blanking period. Three-hundred and sixteen patients (8.6%) (Group A) had ERAF during the blanking period, and 3365 patients (Group B) had no ERAF. Persistent AF and number of tested anti-arrhythmic drugs ≥2 resulted as significant predictors of ERAF. After a mean follow-up of 16.8 ± 16.4 months, 923/3681 (25%) patients had at least one AF recurrence. The observed freedom from AF recurrence, at 24-month follow-up from procedure, was 25.7% and 64.8% in Groups A and B, respectively (P &lt; 0.001). ERAF, persistent AF, and number of tested anti-arrhythmic drugs ≥2 resulted as significant predictors of AF. In a propensity score matching, the logistic model showed that ERAF 1 month after ablation are the best predictor of long-term AF recurrence (P = 0.042).

Conclusion

In patients undergoing CB ablation for AF, ERAF are rare and are a strong predictor of AF recurrence in the follow-up, above all when occur &gt;30 days after the ablation.

Inflammatory signatures are associated with increased mortality after transfemoral transcatheter aortic valve implantation

Aims

Systemic inflammatory response, identified by increased total leucocyte counts, was shown to be a strong predictor of mortality after transcatheter aortic valve implantation (TAVI). Yet the mechanisms of inflammation‐associated poor outcome after TAVI are unclear. Therefore, the present study aimed at investigating individual inflammatory signatures and functional heterogeneity of circulating myeloid and T‐lymphocyte subsets and their impact on 1 year survival in a single‐centre cohort of patients with severe aortic stenosis undergoing TAVI.

Methods and results

One hundred twenty‐nine consecutive patients with severe symptomatic aortic stenosis admitted for transfemoral TAVI were included. Blood samples were obtained at baseline, immediately after, and 24 h and 3 days after TAVI, and these were analysed for inflammatory and cardiac biomarkers. Myeloid and T‐lymphocyte subsets were measured using flow cytometry. The inflammatory parameters were first analysed as continuous variables; and in case of association with outcome and area under receiver operating characteristic (ROC) curve (AUC) ≥ 0.6, the values were dichotomized using optimal cut‐off points. Several baseline inflammatory parameters, including high‐sensitivity C‐reactive protein (hsCRP; HR = 1.37, 95% CI: 1.15–1.63; P < 0.0001) and IL‐6 (HR = 1.02, 95% CI: 1.01–1.03; P = 0.003), lower counts of Th2 (HR = 0.95, 95% CI: 0.91–0.99; P = 0.009), and increased percentages of Th17 cells (HR = 1.19, 95% CI: 1.02–1.38; P = 0.024) were associated with 12 month all‐cause mortality. Among postprocedural parameters, only increased post‐TAVI counts of non‐classical monocytes immediately after TAVI were predictive of outcome (HR = 1.03, 95% CI: 1.01–1.05; P = 0.003). The occurrence of SIRS criteria within 48 h post‐TAVI showed no significant association with 12 month mortality (HR = 0.57, 95% CI: 0.13–2.43, P = 0.45). In multivariate analysis of discrete or dichotomized clinical and inflammatory variables, the presence of diabetes mellitus (HR = 3.50; 95% CI: 1.42–8.62; P = 0.006), low left ventricular (LV) ejection fraction (HR = 3.16; 95% CI: 1.35–7.39; P = 0.008), increased baseline hsCRP (HR = 5.22; 95% CI: 2.09–13.01; P < 0.0001), and low baseline Th2 cell counts (HR = 8.83; 95% CI: 3.02–25.80) were significant predictors of death. The prognostic value of the linear prediction score calculated of these parameters was superior to the Society of Thoracic Surgeons score (AUC: 0.88; 95% CI: 0.78–0.99 vs. 0.75; 95% CI: 0.64–0.86, respectively; P = 0.036). Finally, when analysing LV remodelling outcomes, ROC curve analysis revealed that low numbers of Tregs (P = 0.017; AUC: 0.69) and increased Th17/Treg ratio (P = 0.012; AUC: 0.70) were predictive of adverse remodelling after TAVI.

Conclusions

Our findings demonstrate an association of specific pre‐existing inflammatory phenotypes with increased mortality and adverse LV remodelling after TAVI. Distinct monocyte and T‐cell signatures might provide additive biomarkers to improve pre‐procedural risk stratification in patients referred to TAVI for severe aortic stenosis.

Preprocedure Application of Machine Learning and Mechanistic Simulations Predicts Likelihood of Paroxysmal Atrial Fibrillation Recurrence Following Pulmonary Vein Isolation

BACKGROUND

Pulmonary vein isolation (PVI) is an effective treatment strategy for patients with atrial fibrillation (AF), but many experience AF recurrence and require repeat ablation procedures. The goal of this study was to develop and evaluate a methodology that combines machine learning (ML) and personalized computational modeling to predict, before PVI, which patients are most likely to experience AF recurrence after PVI.

METHODS

This single-center retrospective proof-of-concept study included 32 patients with documented paroxysmal AF who underwent PVI and had preprocedural late gadolinium enhanced magnetic resonance imaging. For each patient, a personalized computational model of the left atrium simulated AF induction via rapid pacing. Features were derived from pre-PVI late gadolinium enhanced magnetic resonance images and from results of simulations of AF induction. The most predictive features were used as input to a quadratic discriminant analysis ML classifier, which was trained, optimized, and evaluated with 10-fold nested cross-validation to predict the probability of AF recurrence post-PVI.

RESULTS

In our cohort, the ML classifier predicted probability of AF recurrence with an average validation sensitivity and specificity of 82% and 89%, respectively, and a validation area under the curve of 0.82. Dissecting the relative contributions of simulations of AF induction and raw images to the predictive capability of the ML classifier, we found that when only features from simulations of AF induction were used to train the ML classifier, its performance remained similar (validation area under the curve, 0.81). However, when only features extracted from raw images were used for training, the validation area under the curve significantly decreased (0.47).

CONCLUSIONS

ML and personalized computational modeling can be used together to accurately predict, using only pre-PVI late gadolinium enhanced magnetic resonance imaging scans as input, whether a patient is likely to experience AF recurrence following PVI, even when the patient cohort is small.

Applications of Artificial Intelligence in Cardiology. The Future is Already Here

There is currently no other hot topic like the ability of current technology to develop capabilities similar to those of human beings, even in medicine. This ability to simulate the processes of human intelligence with computer systems is known as artificial intelligence (AI). This article aims to clarify the various terms that still sound foreign to us, such as AI, machine learning (ML), deep learning (DL), and big data. It also provides an in-depth description of the concept of AI and its types; the learning techniques and technology used by ML; cardiac imaging analysis with DL; and the contribution of this technological revolution to classical statistics, as well as its current limitations, legal aspects, and initial applications in cardiology. To do this, we conducted a detailed PubMed search on the evolution of original contributions on AI to the various areas of application in cardiology in the last 5 years and identified 673 research articles. We provide 19 detailed examples from distinct areas of cardiology that, by using AI, have shown diagnostic and therapeutic improvements, and which will aid understanding of ML and DL methodology.

Clinical significance of early recurrence of atrial fibrillation after cryoballoon vs. radiofrequency ablation-A propensity score matched analysis

Objectives

One of the mechanisms of early recurrence of atrial fibrillation (ERAF) after AF ablation is considered to be the inflammatory reaction of the atrial tissue. The aim of this study is to compare the clinical significance of ERAF at each stage for true AF recurrence between cryoballoon (CB) and radiofrequency (RF) ablation.

Methods

Among 798 paroxysmal AF patients who underwent an initial ablation, 460 patients (CB, n = 230; RF, n = 230) were selected by propensity score matching. Very ERAF (VERAF), ERAF-1M, ERAF-3M and true AF recurrence were defined as AF recurrence at 0–2, 3–30, 31–90 days and more than 90 days after the procedure, respectively.

Results

The patient characteristics of the two groups were similar. ERAF was observed 21% and 27% in the CB and RF groups, respectively. In both the CB and RF group, VERAF, ERAF-1M and ERAF-3M were more frequently observed in patients with true AF recurrence than in those without. In a multivariable analysis, ERAF-1M and ERAF-3M were found to be independent predictors of true AF recurrence in both the CB (P = 0.04 and P<0.001, respectively) and RF groups (P = 0.02 and P = 0.001, respectively). However, while VERAF was associated with true AF recurrence after RF ablation (P = 0.03), it was not associated with true AF recurrence after CB ablation (P = 0.19).

Conclusion

The relationship between ERAF and true AF recurrence differed between the RF and CB ablation groups. While VERAF was associated with true AF recurrence after RF ablation, it was not a predictor of true AF recurrence after CB ablation.

Machine-learning to stratify diabetic patients using novel cardiac biomarkers and integrative genomics

BACKGROUND

Diabetes mellitus is a chronic disease that impacts an increasing percentage of people each year. Among its comorbidities, diabetics are two to four times more likely to develop cardiovascular diseases. While HbA1c remains the primary diagnostic for diabetics, its ability to predict long-term, health outcomes across diverse demographics, ethnic groups, and at a personalized level are limited. The purpose of this study was to provide a model for precision medicine through the implementation of machine-learning algorithms using multiple cardiac biomarkers as a means for predicting diabetes mellitus development.

METHODS

Right atrial appendages from 50 patients, 30 non-diabetic and 20 type 2 diabetic, were procured from the WVU Ruby Memorial Hospital. Machine-learning was applied to physiological, biochemical, and sequencing data for each patient. Supervised learning implementing SHapley Additive exPlanations (SHAP) allowed binary (no diabetes or type 2 diabetes) and multiple classification (no diabetes, prediabetes, and type 2 diabetes) of the patient cohort with and without the inclusion of HbA1c levels. Findings were validated through Logistic Regression (LR), Linear Discriminant Analysis (LDA), Gaussian Naïve Bayes (NB), Support Vector Machine (SVM), and Classification and Regression Tree (CART) models with tenfold cross validation.

RESULTS

Total nuclear methylation and hydroxymethylation were highly correlated to diabetic status, with nuclear methylation and mitochondrial electron transport chain (ETC) activities achieving superior testing accuracies in the predictive model (~ 84% testing, binary). Mitochondrial DNA SNPs found in the D-Loop region (SNP-73G, -16126C, and -16362C) were highly associated with diabetes mellitus. The CpG island of transcription factor A, mitochondrial (TFAM) revealed CpG24 (chr10:58385262, P = 0.003) and CpG29 (chr10:58385324, P = 0.001) as markers correlating with diabetic progression. When combining the most predictive factors from each set, total nuclear methylation and CpG24 methylation were the best diagnostic measures in both binary and multiple classification sets.

CONCLUSIONS

Using machine-learning, we were able to identify novel as well as the most relevant biomarkers associated with type 2 diabetes mellitus by integrating physiological, biochemical, and sequencing datasets. Ultimately, this approach may be used as a guideline for future investigations into disease pathogenesis and novel biomarker discovery.

The Association Between Diabetes Mellitus and Atrial Fibrillation: Clinical and Mechanistic Insights

A number of clinical studies have reported that diabetes mellitus (DM) is an independent risk factor for Atrial fibrillation (AF). After adjustment for other known risk factors including age, sex, and cardiovascular risk factors, DM remains a significant if modest risk factor for development of AF. The mechanisms underlying the increased susceptibility to AF in DM are incompletely understood, but are thought to involve electrical, structural, and autonomic remodeling in the atria. Electrical remodeling in DM may involve alterations in gap junction function that affect atrial conduction velocity due to changes in expression or localization of connexins. Electrical remodeling can also occur due to changes in atrial action potential morphology in association with changes in ionic currents, such as sodium or potassium currents, that can affect conduction velocity or susceptibility to triggered activity. Structural remodeling in DM results in atrial fibrosis, which can alter conduction patterns and susceptibility to re-entry in the atria. In addition, increases in atrial adipose tissue, especially in Type II DM, can lead to disruptions in atrial conduction velocity or conduction patterns that may affect arrhythmogenesis. Whether the insulin resistance in type II DM activates unique intracellular signaling pathways independent of obesity requires further investigation. In addition, the relationship between incident AF and glycemic control requires further study.