Detection Of Atrial Fibrillation Research Articles

IMC-ResNet: Atrial fibrillation detection method based on interlayer multiscale coupling

Atrial fibrillation(AF) is one of the most common cardiac arrhythmias, and reliable detection of it is of significant importance. This study proposes a multilayer convolutional neural network for AF detection, which introduces the Multiscale Information Interaction module composed of a Detail Enhancer and a Context Information Extractor to achieve the extraction and coupling of multiscale information between layers. The Detail Enhancer is composed of multiscale transposed convolutions to encode waveform details in the electrocardiogram (ECG). The Context Information Extractor is composed of multiscale dilated convolutions to encode the morphological information of the ECG. Furthermore, the Multiscale Information Interaction module is used in conjunction with residual blocks to model the surface feature correlation and intrinsic abstract structure of AF signals. This approach extracts implicit information of AF from ECG signals, achieving high-precision and robust detection of AF. Five-fold cross-validation on the AFDB dataset resulted in an accuracy of 99.56% and an F1-score of 99.65%. Inter-patient testing on the clinical SPHDB database yielded an accuracy of 96.17% and an F1-score of 96.14%. In addition, this study conducted experiments to evaluate the model’s robustness against noise, yielding favorable results across four types of noise tests. The results indicate that the proposed method has great potential for application in clinical auxiliary diagnosis based on wearable devices.

A Systematic Review on the Effectiveness of Machine Learning in the Detection of Atrial Fibrillation.

Recent endeavors have led to the exploration of Machine Learning (ML) to enhance the detection and accurate diagnosis of heart pathologies. This is due to the growing need to improve efficiency in diagnostics and hasten the process of delivering treatment. Several institutions have actively assessed the possibility of creating algorithms for advancing our understanding of atrial fibrillation (AF), a common form of sustained arrhythmia. This means that artificial intelligence is now being used to analyze electrocardiogram (ECG) data. The data is typically extracted from large patient databases and then subsequently used to train and test the algorithm with the help of neural networks. Machine learning has been used to effectively detect atrial fibrillation with more accuracy than clinical experts, and if applied to clinical practice, it will aid in early diagnosis and management of the condition and thus reduce thromboembolic complications of the disease. In this text, a review of the application of machine learning in the analysis and detection of atrial fibrillation, a comparison of the outcomes (sensitivity, specificity, and accuracy), and the framework and methods of the studies conducted have been presented.

Controversies in Diagnosis and Management of Atrial Fibrillation.

Early detection of atrial fibrillation (AF) plays an important role in decreasing adverse cardiovascular outcomes. It is estimated, however, that one-third of those with AF are asymptomatic and may experience the adverse effects of the arrhythmia prior to being detected clinically. In the past, AF was diagnosed on 12-lead electrocardiogram or medically prescribed external monitors. The development of device-monitoring technologies capable of recording AF or AF-surrogates such as atrial high-rate episodes on cardiovascular implantable electronic devices or photoplethysmography/electrocardiogram on consumer-grade wearable devices, has resulted in increased recognition of device-detected, subclinical, AF. Recent studies reveal information about the stroke risk associated with these subclinical events and the response to anticoagulation and raise important questions about the use of both medical and direct-to-consumer AF detection devices for screening purposes. In addition to screening and detection of AF, emerging studies are also being conducted on different strategies for maintenance of sinus rhythm and stroke prevention including catheter ablation and left atrial appendage occlusion. This review aims to highlight recent developments and future studies in these areas.

Compressed Deep Learning Models for Wearable Atrial Fibrillation Detection through Attention

Deep learning (DL) models have shown promise for the accurate detection of atrial fibrillation (AF) from electrocardiogram/photoplethysmography (ECG/PPG) data, yet deploying these on resource-constrained wearable devices remains challenging. This study proposes integrating a customized channel attention mechanism to compress DL neural networks for AF detection, allowing the model to focus only on the most salient time-series features. The results demonstrate that applying compression through channel attention significantly reduces the total number of model parameters and file size while minimizing loss in detection accuracy. Notably, after compression, performance increases for certain model variants in key AF databases (ADB and C2017DB). Moreover, analyzing the learned channel attention distributions after training enhances the explainability of the AF detection models by highlighting the salient temporal ECG/PPG features most important for its diagnosis. Overall, this research establishes that integrating attention mechanisms is an effective strategy for compressing large DL models, making them deployable on low-power wearable devices. We show that this approach yields compressed, accurate, and explainable AF detectors ideal for wearables. Incorporating channel attention enables simpler yet more accurate algorithms that have the potential to provide clinicians with valuable insights into the salient temporal biomarkers of AF. Our findings highlight that the use of attention is an important direction for the future development of efficient, high-performing, and interpretable AF screening tools for wearable technology.

SiamQuality: a ConvNet-based foundation model for photoplethysmography signals.

Physiological data are often low quality and thereby compromises the effectiveness of related health monitoring. The primary goal of this study is to develop a robust foundation model that can effectively handle low-quality issue in physiological data.
Approach: We introduce SiamQuality, a self-supervised learning approach using convolutional neural networks (CNNs) as the backbone. SiamQuality learns to generate similar representations for both high and low quality photoplethysmography (PPG) signals that originate from similar physiological states. We leveraged a substantial dataset of PPG signals from hospitalized intensive care patients, comprised of over 36 million 30-second PPG pairs.
Main Results: After pre-training the SiamQuality model, it was fine-tuned and tested on six PPG downstream tasks focusing on cardiovascular monitoring. Notably, in tasks such as respiratory rate estimation and atrial fibrillation detection, the model's performance exceeded the state-of-the-art by 75% and 5%, respectively. The results highlight the effectiveness of our model across all evaluated tasks, demonstrating significant improvements, especially in applications for heart monitoring on wearable devices.
Significance: This study underscores the potential of CNNs as a robust backbone for foundation models tailored to physiological data, emphasizing their capability to maintain performance despite variations in data quality. The success of the SiamQuality model in handling real-world, variable-quality data opens new avenues for the development of more reliable and efficient healthcare monitoring technologies.

Detection of atrial fibrillation using a nonlinear Lorenz Scattergram and deep learning in primary care

BackgroundAtrial fibrillation (AF) is highly correlated with heart failure, stroke and death. Screening increases AF detection and facilitates the early adoption of comprehensive intervention. Long-term wearable devices have become increasingly popular for AF screening in primary care. However, interpreting data obtained by long-term wearable ECG devices is a problem in primary care. To diagnose the disease quickly and accurately, we aimed to build AF episode detection model based on a nonlinear Lorenz scattergram (LS) and deep learning.MethodsThe MIT-BIH Normal Sinus Rhythm Database, MIT-BIH Arrhythmia Database and the Long-Term AF Database were extracted to construct the MIT-BIH Ambulatory Electrocardiograph (MIT-BIH AE) dataset. We converted the long-term ECG into a two-dimensional LSs. The LSs from MIT-BIH AE dataset was randomly divided into training and internal validation sets in a 9:1 ratio, which was used to develop and internally validated model. We built a MOBILE-SCREEN-AF (MS-AF) dataset from a single-lead wearable ECG device in primary care for external validation. Performance was quantified using a confusion matrix and standard classification metrics.ResultsDuring the evaluation of model performance based on the LS, the sensitivity, specificity and accuracy of the model in diagnosing AF were 0.992, 0.973, and 0.983 in the internal validation set respectively. In the external validation set, these metrics were 0.989, 0.956, and 0.967, respectively. Furthermore, when evaluating the model’s performance based on ECG records in the MS-AF dataset, the sensitivity, specificity and accuracy of model diagnosis paroxysmal AF were 1.000, 0.870 and 0.876 respectively, and 0.927, 1.000 and 0.973 for the persistent AF.ConclusionsThe model based on the nonlinear LS and deep learning has high accuracy, making it promising for AF screening in primary care. It has potential for generalization and practical application.

Smartwatch a new tool in the assesment of ischaemic heart disease

Abstract Introduction Outpatient ECG tracing is getting more important. Therefore, an everyday tool such as the watch has become "smart" (127.5 M units sold in 2021), and its key application is the heart rate measurement through ECG tracing. Some devices are already approved for atrial fibrillation detection. However, their role in ischemic heart disease (IHD), leading cause of cardiovascular death, has not been appropriately tested. Objetcive The aim of this study was to evaluate the usefulness of smartwatch ECG registry in IHD. Methods We present an observational study of 25 consecutive patients with acute IHD. Conventional ECG and smartwatch tracing were obtained simultaneously at admission. Waves of conventional and smartwatch ECGs were objectively compared. A survey on medical attitude was conducted among 12 physicians (3 cardiologists, 3 intensivists, 3 emergency physicians, and 3 general practitioners) and a score (1–5) of concordance between the records was requested. Figure 1. Results There were no differences in Q-wave, Rwave, ST segment, or T-wave. There was a very strong correlation between ST segments, a strong correlation in Q-waves and R-waves, and a moderate correlation in T-wave measurements. All specialists obtained a high level of agreement (4.45 ± 0.45). Smartwatch tracings would lead to similar management compared to conventional ECG. There were only 6 (2%) discrepant cases due to differences in inferior repolarization, showing an almost perfect agreement (kappa = 0.96). Conclusions In most patients with acute IHD, smartwatch ECG tracing is a reliable tool to make the diagnosis and guide appropriate medical care. However, due to their intrinsic limitations, inferior myocardial infarctions may be missed and require a conventional 12-lead ECG to rule them out.

Early Diagnosis of Atrial Fibrillation Episodes: Comparative Analysis of Different Matrix Architectures

This study presents three different matrix architectures for the analysis of ECG parameters, aimed at detecting atrial fibrillation episodes. The evaluation involves a cohort of 15 individuals, utilizing these matrix architectures across various orders. The findings reveal that the matrix norm delivers significantly better results compared to the large discriminant of the matrix. Detailed analysis of the spatial expansion of each matrix structure indicates that the PMLD architecture excels in terms of expandability compared to the MA1 and MA2 matrices. Consequently, third- and fifth-order PMLD matrix architectures are employed for classification techniques, demonstrating enhanced sensitivity with increased matrix order. These results are validated through the classification of several test candidates, confirming the efficacy of the proposed method. The study suggests that the developed approach holds substantial potential for clinical diagnostics in the early detection of atrial fibrillation.

Artificial intelligence-enabled atrial fibrillation detection using smartwatches: current status and future perspectives.

Atrial fibrillation (AF) significantly increases the risk of stroke and heart failure, but is frequently asymptomatic and intermittent; therefore, its timely diagnosis poses challenges. Early detection in selected patients may aid in stroke prevention and mitigate structural heart complications through prompt intervention. Smartwatches, coupled with powerful artificial intelligence (AI)-enabled algorithms, offer a promising tool for early detection due to their widespread use, easiness of use, and potential cost-effectiveness. Commercially available smartwatches have gained clearance from the FDA to detect AF and are becoming increasingly popular. Despite their promise, the evolving landscape of AI-enabled smartwatch-based AF detection raises questions about the clinical value of this technology. Following the ongoing digital transformation of healthcare, clinicians should familiarize themselves with how AI-enabled smartwatches function in AF detection and navigate their role in clinical settings to deliver optimal patient care. In this review, we provide a concise overview of the characteristics of AI-enabled smartwatch algorithms, their diagnostic performance, clinical value, limitations, and discuss future perspectives in AF diagnosis.

Novel interpretable Feature set extraction and classification for accurate atrial fibrillation detection from ECGs

ObjectiveWe present a novel method for detecting atrial fibrillation (AFib) by analyzing Lead II electrocardiograms (ECGs) using a unique set of features. MethodsFor this purpose, we used specific signal processing techniques, such as proper orthogonal decomposition, continuous wavelet transforms, discrete cosine transform, and standard cross-correlation, to extract 48 features from the ECGs. Thus, our approach aims to more effectively capture AFib signatures, such as beat-to-beat variability and fibrillatory waves, than traditional metrics. Moreover, our features were designed to be physiologically interpretable. Subsequently, we incorporated an XGBoost-based ECG classifier to train and evaluate it using the publicly available ‘Training’ dataset of the 2017 PhysioNet Challenge, which includes ‘Normal,’ ‘AFib,’ ‘Other,’ and ‘Noisy’ ECG categories. ResultsOur method achieved an accuracy of 96 % and an F1-score of 0.83 for ‘AFib’ detection and 80 % accuracy and 0.85 F1-score for ‘Normal’ ECGs. Finally, we compared our method's performance with the top-classifiers from the 2017 PhysioNet Challenge, namely ENCASE, Random Forest, and Cascaded Binary. As reported by Clifford et al., 2017, these three best performing models scored 0.80, 0.83, 0.82, in terms of F1-score for ‘AFib’ detection, respectively. ConclusionDespite using significantly fewer features than the competition's state-of-the-art ECG detection algorithms (48 vs. 150–622), our model achieved a comparable F1-score of 0.83 for successful ‘AFib’ detection. Significance: The interpretable features specifically designed for ‘AFib’ detection results in our method's adaptability in clinical settings for real-time arrhythmia detection and is even effective with short ECGs (<10 heartbeats).

Artificial Intelligence and Heart-Brain Connections: A Narrative Review on Algorithms Utilization in Clinical Practice.

Cardiovascular and neurological diseases are a major cause of mortality and morbidity worldwide. Such diseases require careful monitoring to effectively manage their progression. Artificial intelligence (AI) offers valuable tools for this purpose through its ability to analyse data and identify predictive patterns. This review evaluated the application of AI in cardiac and neurological diseases for their clinical impact on the general population. We reviewed studies on the application of AI in the neurological and cardiological fields. Our search was performed on the PubMed, Web of Science, Embase and Cochrane library databases. Of the initial 5862 studies, 23 studies met the inclusion criteria. The studies showed that the most commonly used algorithms in these clinical fields are Random Forest and Artificial Neural Network, followed by logistic regression and Support-Vector Machines. In addition, an ECG-AI algorithm based on convolutional neural networks has been developed and has been widely used in several studies for the detection of atrial fibrillation with good accuracy. AI has great potential to support physicians in interpretation, diagnosis, risk assessment and disease management.

A Narrowband IoT Personal Sensor for Long-Term Heart Rate Monitoring and Atrial Fibrillation Detection

Long-term patient monitoring is required for detection of episodes of atrial fibrillation, one of the most widespread cardiac pathologies. Today, the most used non-invasive technique is Holter electrocardiographic (ECG) monitoring, which can often prove ineffective because of the short duration of recordings (e.g., one day). Other techniques such as photo-plethysmography are adopted by smartwatches for much longer duration monitoring, but this has the disadvantage of offering only intermittent measurements. This study proposes an Internet of Things (IoT) sensor that can provide a very long period of continuous monitoring. The sensor consists of an ECG-integrated Analog Front End (MAX30003), a microcontroller (STM32F401RE), and an IoT narrowband module (STEVAL-STMODLTE). The instantaneous heart rate is extracted from the ECG recording in real time. At intervals of two minutes, the sequence of inter-beat intervals is transmitted to an IoT cloud platform (ThingSpeak). Settled atrial fibrillation event recognition software runs on the cloud and generates alerts when it recognizes such arrhythmia. Performances of the proposed sensor were evaluated by generating analog ECG signals from a public dataset of ECG signals with atrial fibrillation episodes, the MIT-BIH Atrial Fibrillation Database, each recording lasting approximately 10 h. Software implementing the Lorentz algorithm, one of the best detectors of atrial fibrillation, was implemented on the cloud platform. The accuracy, sensitivity, and specificity in recognizing atrial fibrillation episodes of the proposed system was calculated by comparison with a cardiologist’s reference data. Across all patients, the proposed method achieved an accuracy of 0.88, a sensitivity 0.71, and a specificity 0.99. The results obtained suggest that the developed system can continuously record and transmit heart rhythms effectively and efficiently and, in addition, offers considerable performance in recognizing atrial fibrillation episodes in real time.

A Narrowband IoT Personal Sensor for Long-Term Heart Rate Monitoring and Atrial Fibrillation Detection.

Long-term patient monitoring is required for detection of episodes of atrial fibrillation, one of the most widespread cardiac pathologies. Today, the most used non-invasive technique is Holter electrocardiographic (ECG) monitoring, which can often prove ineffective because of the short duration of recordings (e.g., one day). Other techniques such as photo-plethysmography are adopted by smartwatches for much longer duration monitoring, but this has the disadvantage of offering only intermittent measurements. This study proposes an Internet of Things (IoT) sensor that can provide a very long period of continuous monitoring. The sensor consists of an ECG-integrated Analog Front End (MAX30003), a microcontroller (STM32F401RE), and an IoT narrowband module (STEVAL-STMODLTE). The instantaneous heart rate is extracted from the ECG recording in real time. At intervals of two minutes, the sequence of inter-beat intervals is transmitted to an IoT cloud platform (ThingSpeak). Settled atrial fibrillation event recognition software runs on the cloud and generates alerts when it recognizes such arrhythmia. Performances of the proposed sensor were evaluated by generating analog ECG signals from a public dataset of ECG signals with atrial fibrillation episodes, the MIT-BIH Atrial Fibrillation Database, each recording lasting approximately 10 h. Software implementing the Lorentz algorithm, one of the best detectors of atrial fibrillation, was implemented on the cloud platform. The accuracy, sensitivity, and specificity in recognizing atrial fibrillation episodes of the proposed system was calculated by comparison with a cardiologist's reference data. Across all patients, the proposed method achieved an accuracy of 0.88, a sensitivity 0.71, and a specificity 0.99. The results obtained suggest that the developed system can continuously record and transmit heart rhythms effectively and efficiently and, in addition, offers considerable performance in recognizing atrial fibrillation episodes in real time.

Social drivers in atrial fibrillation occurrence, screening, treatment, and outcomes: systematic-narrative hybrid review.

The importance of social drivers of health (SDOH) in the occurrence, detection, treatment, and outcome of atrial fibrillation (AF) has attracted increasing attention. Addressing SDOH factors may suggest opportunities to prevent AF and its complications. We aimed to conduct a structured narrative review and summarize current knowledge on the association between race and ethnicity, SDOH, including rural vs. urban habitation, education, income, and neighbourhood, and the risk of AF, its management, and complications. We identified 537 references in PubMed and 473 references in Embase. After removal of duplicates, we screened the abstracts of 975 references, resulting in 113 references that were examined for eligibility. Subsequently, 34 references were excluded leaving 79 references for the review. Evidence of a social gradient in AF incidence and prevelance were conflicting. However, we found substantial evidence indicating social inequities in the detection of AF, access to treatment, and outcomes such as healthcare utilization, bleeding, heart failure, stroke, dementia, work disability, and death. Inequities are reported across various health care systems and constitute a global problem affecting several continents, although data from Africa and South America are lacking. Given the documented social inequities in AF detection, management, and outcomes, there is an urgent need for healthcare systems, policymakers, and society to identify and implement effective interventions that can reduce inequities and improve outcomes in individuals with AF.

Screening and detection of atrial fibrillation in primary care: current practice and future perspectives.

Atrial fibrillation (AF) is a common arrhythmia associated with an increased risk of stroke, which can be effectively reduced by prophylaxis initiation and integrated care to reduce cardiovascular risk and AF-related complications. Screening for AF has the potential to improve long-term clinical outcomes through timely AF detection in asymptomatic patients. With the central role of primary care in most European healthcare systems in terms of disease detection, treatment, as well as record keeping, primary care is ideally situated as a setting for AF screening efforts. In this review, we provide an overview of evidence relating to AF screening in primary care. We discuss current practices of AF detection and screening, evidence from AF screening trials conducted in primary care settings, stakeholder views on barriers and facilitators for AF screening in primary care, and important aspects that will likely shape routine primary care AF detection as well as AF screening efforts. Finally, we present a potential outline for a primary care-centred AF screening trial coupled to integrated AF care that could further improve the benefit of AF screening.

Early detection of atrial fibrillation in the digital era, risk factors, treatment options, and the need for new definitions.

Graphical abstract.

Diagnostic Utility of N-Terminal Pro-B-Type Natriuretic Peptide in Identifying Atrial Fibrillation Post-Cryptogenic Stroke: A Systematic Review and Meta-Analysis.

Atrial fibrillation (AF) significantly contributes to acute ischemic stroke, with undetected AF being a common culprit in cryptogenic strokes. N-terminal pro-B-type natriuretic peptide (NT-proBNP), indicative of myocardial stress, has been proposed as a biomarker for AF detection, aiding in the selection of patients for extended cardiac monitoring. However, the diagnostic accuracy of NT-proBNP remains uncertain. We conducted a meta-analysis to evaluate the diagnostic accuracy of NT-proBNP in detecting AF among cryptogenic stroke patients. A comprehensive literature search was conducted across PubMed, Embase, and Cochrane databases to identify relevant studies. Studies reporting NT-proBNP levels in stroke patients and data on the proportion of patients with AF above a specified cut-off were included. Meta-analyses were performed using the midas command in STATA. Seven studies encompassing 2171 patients were included in the analysis, of which five studies contained cohorts with cryptogenic strokes. Among patients with cryptogenic stroke, NT-proBNP demonstrated a diagnostic accuracy of 80% (Area Under the Receiver Operating Curve 0.80 [95% CI 0.76-0.83]), with a sensitivity of 81% (95% CI 0.68-0.89) and a specificity of 68% (95% CI 0.60-0.75). Our meta-analysis indicates that NT-proBNP exhibits a good-to-very-good diagnostic accuracy for detecting AF in patients with cryptogenic stroke. These findings suggest potential implications for utilizing NT-proBNP in guiding the selection of patients for prolonged cardiac monitoring, thereby aiding in the management of cryptogenic stroke cases.

Reliability of single-lead electrocardiogram interpretation to detect atrial fibrillation: insights from the SAFER feasibility study.

Single-lead electrocardiograms (ECGs) can be recorded using widely available devices such as smartwatches and handheld ECG recorders. Such devices have been approved for atrial fibrillation (AF) detection. However, little evidence exists on the reliability of single-lead ECG interpretation. We aimed to assess the level of agreement on detection of AF by independent cardiologists interpreting single-lead ECGs and to identify factors influencing agreement. In a population-based AF screening study, adults aged ≥65 years old recorded four single-lead ECGs per day for 1-4 weeks using a handheld ECG recorder. Electrocardiograms showing signs of possible AF were identified by a nurse, aided by an automated algorithm. These were reviewed by two independent cardiologists who assigned participant- and ECG-level diagnoses. Inter-rater reliability of AF diagnosis was calculated using linear weighted Cohen's kappa (κw). Out of 2141 participants and 162 515 ECGs, only 1843 ECGs from 185 participants were reviewed by both cardiologists. Agreement was moderate: κw = 0.48 (95% confidence interval, 0.37-0.58) at participant level and κw = 0.58 (0.53-0.62) at ECG level. At participant level, agreement was associated with the number of adequate-quality ECGs recorded, with higher agreement in participants who recorded at least 67 adequate-quality ECGs. At ECG level, agreement was associated with ECG quality and whether ECGs exhibited algorithm-identified possible AF. Inter-rater reliability of AF diagnosis from single-lead ECGs was found to be moderate in older adults. Strategies to improve reliability might include participant and cardiologist training and designing AF detection programmes to obtain sufficient ECGs for reliable diagnoses.

The Usefulness of Outpatient Cardiac Telemetry in Patients with Cryptogenic Stroke.

Introduction: Atrial fibrillation (AF), apart from non-stenotic supracardiac atherosclerosis and neoplastic disease, is the leading cause of cryptogenic stroke, including embolic stroke of un-determined source (ESUS). The aim of our study was to determine the prevalence of AF in ESUS patients based on 30-day telemetric heart rate monitoring initiated within three months after stroke onset. Another aim was to identify factors that increase the likelihood of detecting subsequent AF among ESUS patients. Material and Methods: patients with first-ever stroke classified as per the ESUS definition were eligible for this study. All patients underwent outpatient 30-day telemetric heart rate monitoring. Results: In the period between 2020 and 2022, 145 patients were included. The mean age of all qualified patients was 54; 40% of eligible patients were female. Six patients (4.14%), mostly male patients (4 vs. 2), were diagnosed with AF within the study period. In each case, the diagnosis related to a patient whose stroke occurred in the course of large vessel occlusion. Episodes of AF were detected between day 1 and 25 after starting ECG monitoring. Out of the analyzed parameters that increase the probability of, A.F.; only supraventricular extrasystoles proved to be an independent factor regarding an increased risk of AF [OR 1.046, CI 95% 1.016-1.071, p-value < 0.01]. Conclusions: The use of telemetry heart rhythm monitoring in an outpatient setting can detect AF in 4% of ESUS patients who have undergone prior diagnostic procedures for cardiogenic embolism. Supraventricular extrasystoles significantly increases the likelihood of AF detection in patients with ESUS within three months following stroke. Comorbid coronary artery disease, diabetes and hypertension, rather than a single-factor clinical burden, increase the likelihood of AF detection in older ESUS patients. ESUS in the course of large vessel occlusion is probably associated with an increased likelihood of cardiogenic embolism.

Higher-Order Spectral Analysis Combined with a Convolution Neural Network for Atrial Fibrillation Detection-Preliminary Study.

The global burden of atrial fibrillation (AFIB) is constantly increasing, and its early detection is still a challenge for public health and motivates researchers to improve methods for automatic AFIB prediction and management. This work proposes higher-order spectra analysis, especially the bispectrum of electrocardiogram (ECG) signals combined with the convolution neural network (CNN) for AFIB detection. Like other biomedical signals, ECG is non-stationary, non-linear, and non-Gaussian in nature, so the spectra of higher-order cumulants, in this case, bispectra, preserve valuable features. The two-dimensional (2D) bispectrum images were applied as input for the two CNN architectures with the output AFIB vs. no-AFIB: the pre-trained modified GoogLeNet and the proposed CNN called AFIB-NET. The MIT-BIH Atrial Fibrillation Database (AFDB) was used to evaluate the performance of the proposed methodology. AFIB-NET detected atrial fibrillation with a sensitivity of 95.3%, a specificity of 93.7%, and an area under the receiver operating characteristic (ROC) of 98.3%, while for GoogLeNet results for sensitivity and specificity were equal to 96.7%, 82%, respectively, and the area under ROC was equal to 96.7%. According to preliminary studies, bispectrum images as input to 2D CNN can be successfully used for AFIB rhythm detection.