Machine learning in the detection and management of atrial fibrillation

Multidisciplinary management strategies for atrial fibrillation

There has been a significant increase in the prevalence of atrial fibrillation (AF) over the past 30 years. Pulmonary vein isolation (PVI) is an effective treatment for AF, but research investigations have shown that AF recurrence still occurs in a significant number of patients after ablation. Heart rhythm outcomes following catheter ablation are correlated with numerous clinical factors, and researchers developed predictive models by integrating risk factors to predict the risk of recurrence of atrial fibrillation. The purpose of this article is to outline the risk scores for predicting cardiac rhythm outcomes after PVI and to discuss the modifiable factors that increase the risk of recurrence of AF, with the hope of further improving catheter ablation efficacy through preoperative identification of high-risk populations and postoperative management of modifiable risk factors.

Thromboembolic risk and oral anticoagulation in subclinical atrial fibrillation

Availability of devices capable of continuous rhythm monitoring such as smartwatches, implantable loop recorders, or pacemakers/defibrillators is continuously increasing. Importantly, device detected "subclinical" atrial fibrillation seems to convey a significantly lower risk of thromboembolism than "clinical" atrial fibrillation verified by a conventional ECG recording. While current guidelines indicate a possible role of oral anticoagulation in selected high-risk patients with subclinical AF, recent trials show an ambiguous risk/benefit relationship of anticoagulation in this setting. The present review therefore summarizes current data on the role of oral anticoagulation in subclinical AF, aims at aiding in the decision process of anticoagulation, and illustrates current gaps in evidence regarding subclinical AF.

Research Concepts and Opportunities for Early-Career Investigators in Cardiac Surgery

Basic, translational or clinic, research is a key component of cardiac surgery. Understanding basic cellular and molecular mechanisms is key to improving patient outcomes, and cardiac surgical procedures must be compared with nonsurgical alternatives. However, guidance for early-career investigators interested in cardiac surgery research is limited. This opinion piece aims at providing basic guidance and principles based on the authors' experience.

More than 30 years of Brugada syndrome: a critical appraisal of achievements and open issues

Over the last three decades, what is referred to as Brugada syndrome (BrS) has developed from a clinical observation of initially a few cases of sudden cardiac death (SCD) in the absence of structural heart disease with ECG signs of "atypical right bundle brunch block" to a predominantly electrocardiographic, and to a lesser extent genetic, diagnosis. Today, BrS is diagnosed in patients without overt structural heart disease and a spontaneous Brugada type 1 ECG pattern regardless of symptoms. The diagnosis of BrS is less clear in those with an only transient or drug-induced type 1 Brugada pattern, but should be considered in the presence of an arrhythmic syncope, family history of BrS, or family history of sudden death. In addition to survived cardiac arrest, syncope is probably the single most decisive risk marker for future arrhythmias. For asymptomatic BrS, risk stratification remains challenging. General recommendations to lower the risk in BrS include avoidance of drugs/agents known to induce and/or increase right precordial ST-segment elevation, including treatment of fever with antipyretic drugs. Several ECG markers that have been associated with an increased risk of SCD have been incorporated into a recently published risk score for BrS. The aim of this article is to provide an overview of the status of risk stratification and to illustrate open issues und gaps in evidence in BrS.

Knowledge structure and global trends of machine learning in stroke over the past decade: A scientometric analysis

Objective

Machine learning (ML) models have been widely applied in stroke prediction, diagnosis, treatment, and prognosis assessment. We aimed to conduct a comprehensive scientometrics analysis of studies related to ML in stroke and reveal its current status, knowledge structure, and global trends.

Methods

All documents related to ML in stroke were retrieved from the Web of Science database on March 15, 2023. We refined the documents by including only original articles and reviews in the English language. The literature published over the past decade was imported into scientometrics software for influence detection and collaborative network analysis.

Results

2389 related publications were included. The annual publication outputs demonstrated explosive growth, with an average growth rate of 63.99 %. Among the 90 countries/regions involved, the United States (729 articles) and China (636 articles) were the most productive countries. Frontiers in Neurology was the most prolific journal with 94 articles. 234 highly cited articles, each with more than 31 citations, were detected. Keyword analysis revealed a total of 5333 keywords, with a predominant focus on the application of ML models in the early diagnosis, classification, and prediction of "acute ischemic stroke" and "atrial fibrillation-related stroke". The keyword "classification" had the first and longest burst, spanning from 2013 to 2018. 'Upport vector machine' got the strongest burst strength with 6.2. Keywords such as 'mechanical thrombectomy', 'expression', and 'prognosis' experienced bursts in 2022 and have continued to be prominent.

Conclusion

The applications of ML in stroke are increasingly diverse and extensive, with researchers showing growing interest over the past decade. However, the clinical application of ML in stroke is still in its early stages, and several limitations and challenges need to be addressed for its widespread adoption in clinical practice.

COMP Improves Ang-II-Induced Atrial Fibrillation via TGF-β Signaling Pathway

Cartilage oligomeric matrix protein (COMP) regulates transforming growth factor-β (TGF-β) signaling pathway, which has been proved to be associated with skin fibrosis and pulmonary fibrosis. Atrial fibrosis is a major factor of atrial fibrillation (AF). Nevertheless, the interaction between COMP and TGF-β as well as their role in AF remains undefined. The purpose of this study is to clarify the role of COMP in AF and explore its potential mechanism. The hub gene of AF was identified from two datasets using bioinformatics. Furthermore, it was verified by the downregulation of COMP in angiotensin-II (Ang-II)-induced AF in mice. Moreover, the effect on AF was examined using CCK8 assay, ELISA, and western blot. The involvement of TGF-β pathway was further discussed. The expression of COMP was the most significant among all these hub genes. Our experimental results revealed that the protein levels of TGF-β1, phosphorylated Smad2 (P-Smad2), and phosphorylated Smad3 (P-Smad3) were decreased after silencing COMP, which indicated that COMP knockdown could inhibit the activation of TGF-β pathway in AF cells. However, the phenomenon was reversed when the activator SRI was added. COMP acts as a major factor and can improve Ang-II-induced AF via TGF-β signaling pathway. Thus, our research enriches the understanding of the interaction between COMP and TGF-β in AF, and provides reference for the pathogenesis and diagnosis of AF.

Detecting Paroxysmal Atrial Fibrillation From an Electrocardiogram in Sinus Rhythm: External Validation of the AI Approach

BACKGROUND

Atrial fibrillation (AF) may occur asymptomatically and can be diagnosed only with electrocardiography (ECG) while the arrhythmia is present.

OBJECTIVES

The aim of this study was to independently validate the approach of using artificial intelligence (AI) to identify underlying paroxysmal AF from a 12-lead ECG in sinus rhythm (SR).

METHODS

An AI algorithm was trained to identify patients with underlying paroxysmal AF, using electrocardiographic data from all in- and outpatients from a single center with at least 1 ECG in SR. For patients without AF, all ECGs in SR were included. For patients with AF, all ECGs in SR starting 31 days before the first AF event were included. The patients were randomly allocated to training, internal validation, and testing datasets in a 7:1:2 ratio. In a secondary analysis, the AF prevalence of the testing group was modified. Additionally, the performance of the algorithm was validated at an external hospital.

RESULTS

The dataset consisted of 494,042 ECGs in SR from 142,310 patients. Testing the model on the first ECG of each patient (AF prevalence 9.0%) resulted in accuracy of 78.1% (95% CI: 77.6%-78.5%), area under the receiver-operating characteristic curve of 0.87 (95% CI: 0.86-0.87), and area under the precision recall curve (AUPRC) of 0.48 (95% CI: 0.46-0.50). In a low-risk group (AF prevalence 3%), the AUPRC decreased to 0.21 (95% CI: 0.18-0.24). In a high-risk group (AF prevalence 30%), the AUPRC increased to 0.76 (95% CI: 0.75-0.78). This performance was robust when validated in an external hospital.

CONCLUSIONS

The approach of using an AI-enabled electrocardiographic algorithm for the identification of patients with underlying paroxysmal AF from ECGs in SR was independently validated.

Exploring the Intersection of Artificial Intelligence and Clinical Healthcare: A Multidisciplinary Review

Artificial intelligence (AI) plays a more and more important role in our everyday life due to the advantages that it brings when used, such as 24/7 availability, a very low percentage of errors, ability to provide real time insights, or performing a fast analysis. AI is increasingly being used in clinical medical and dental healthcare analyses, with valuable applications, which include disease diagnosis, risk assessment, treatment planning, and drug discovery. This paper presents a narrative literature review of AI use in healthcare from a multi-disciplinary perspective, specifically in the cardiology, allergology, endocrinology, and dental fields. The paper highlights data from recent research and development efforts in AI for healthcare, as well as challenges and limitations associated with AI implementation, such as data privacy and security considerations, along with ethical and legal concerns. The regulation of responsible design, development, and use of AI in healthcare is still in early stages due to the rapid evolution of the field. However, it is our duty to carefully consider the ethical implications of implementing AI and to respond appropriately. With the potential to reshape healthcare delivery and enhance patient outcomes, AI systems continue to reveal their capabilities.

State of the art paper: Cardiac computed tomography of the left atrium in atrial fibrillation

The clinical spectrum of atrial fibrillation means that a patient-individualized approach is required to ensure optimal treatment. Cardiac computed tomography can accurately delineate atrial structure and function and could contribute to a personalized care pathway for atrial fibrillation patients. The imaging modality offers excellent spatial resolution and has been utilised in pre-, peri- and post-procedural care for patients with atrial fibrillation. Advances in temporal resolution, acquisition times and analysis techniques suggest potential expanding roles for cardiac computed tomography in the future management of patients with atrial fibrillation. The aim of the current review is to discuss the use of cardiac computed tomography in atrial fibrillation in pre-, peri- and post-procedural settings. Potential future applications of cardiac computed tomography including atrial wall thickness assessment and epicardial fat volume quantification are discussed together with emerging analysis techniques including computational modelling and machine learning with attention paid to how these developments may contribute to a personalized approach to atrial fibrillation management.

Benefits of early rhythm control of atrial fibrillation

In contrast to current guidelines and earlier trials, recent studies demonstrated superiority of rhythm- over rate-control and challenged the strategy of "rate versus rhythm" therapy in patients with atrial fibrillation. These newer studies have started to shift the use of rhythm-control therapy from the symptom-driven therapy of current guidelines to a risk-reducing strategy aimed at restoring and maintaining sinus rhythm. This review discusses recent data and presents an overview on the current discourse: The concept of early rhythm control seems attractive. Patients with rhythm control may undergo less atrial remodeling compared to those with rate control. In addition, in EAST-AFNET 4 an outcome-reducing effect of rhythm control was achieved by delivering therapy with relatively few complications early after the initial AF diagnosis. Successful rhythm control therapy and most likely reduced AF burden, estimated by the presence of sinus rhythm at 12 months after randomization, explained most of the reduction in cardiovascular outcomes achieved by rhythm control. However, it is too early to call for early rhythm control for all AF patients. Rhythm control may raise concerns regarding the generalizability of trial results in routine practice involving important questions on the definition of "early" and "successful", and the relevant issue of antiarrhythmic drugs versus catheter ablation. Further information is required to select patients who will benefit from an early ablative or non-ablative rhythm management.

A Study of R-R Interval Transition Matrix Features for Machine Learning Algorithms in AFib Detection

Atrial Fibrillation (AFib) is a heart condition that occurs when electrophysiological malformations within heart tissues cause the atria to lose coordination with the ventricles, resulting in "irregularly irregular" heartbeats. Because symptoms are subtle and unpredictable, AFib diagnosis is often difficult or delayed. One possible solution is to build a system which predicts AFib based on the variability of R-R intervals (the distances between two R-peaks). This research aims to incorporate the transition matrix as a novel measure of R-R variability, while combining three segmentation schemes and two feature importance measures to systematically analyze the significance of individual features. The MIT-BIH dataset was first divided into three segmentation schemes, consisting of 5-s, 10-s, and 25-s subsets. In total, 21 various features, including the transition matrix features, were extracted from these subsets and used for the training of 11 machine learning classifiers. Next, permutation importance and tree-based feature importance calculations determined the most predictive features for each model. In summary, with Leave-One-Person-Out Cross Validation, classifiers under the 25-s segmentation scheme produced the best accuracies; specifically, Gradient Boosting (96.08%), Light Gradient Boosting (96.11%), and Extreme Gradient Boosting (96.30%). Among eleven classifiers, the three gradient boosting models and Random Forest exhibited the highest overall performance across all segmentation schemes. Moreover, the permutation and tree-based importance results demonstrated that the transition matrix features were most significant with longer subset lengths.

Application of machine learning in predicting the risk of postpartum depression: A systematic review

BACKGROUND

Postpartum depression (PPD) presents a serious health problem among women and their families. Machine learning (ML) is a rapidly advancing field with increasing utility in predicting PPD risk. We aimed to synthesize and evaluate the quality of studies on application of ML techniques in predicting PPD risk.

METHODS

We conducted a systematic search of eight databases, identifying English and Chinese studies on ML techniques for predicting PPD risk and ML techniques with performance metrics. Quality of the studies involved was evaluated using the Prediction Model Risk of Bias Assessment Tool.

RESULTS

Seventeen studies involving 62 prediction models were included. Supervised learning was the main ML technique employed and the common ML models were support vector machine, random forest and logistic regression. Five studies (30 %) reported both internal and external validation. Two studies involved model translation, but none were tested clinically. All studies showed a high risk of bias, and more than half showed high application risk.

LIMITATIONS

Including Chinese articles slightly reduced the reproducibility of the review. Model performance was not quantitatively analyzed owing to inconsistent metrics and the absence of methods for correlation meta-analysis.

CONCLUSIONS

Researchers have paid more attention to model development than to validation, and few have focused on improvement and innovation. Models for predicting PPD risk continue to emerge. However, few have achieved the acceptable quality standards. Therefore, ML techniques for successfully predicting PPD risk are yet to be deployed in clinical environments.