Detection Of Fibrillation Research Articles

Predictors of atrial fibrillation detection in embolic stroke of undetermined source patients with implantable loop recorder.

Covert atrial fibrillation (AF) is a predominant aetiology of embolic stroke of undetermined source (ESUS). Evidence suggested that AF is more frequently detected by implantable loop recorder (ILR) than by conventional monitoring. However, the predictive factors associated with occult AF detected using ILRs are not well established yet. In this study we aim to investigate the predictors of AF detection in patients with ESUS undergoing an ILR. This observational multi-centre study included consecutive ESUS patients who underwent ILR implantation. The infarcts were divided in deep, cortical infarcts or both. The infarction sites were categorized as anterior and middle cerebral artery, posterior cerebral artery with and without brainstem/cerebellum involvement. Multivariable logistic regression analysis was performed to investigate variables associated with AF detection. Overall, 3,000 patients were initially identified. However, in total, 127 patients who consecutively underwent ILR implantation were included in our analysis. AF was detected in 33 (26%) out of 127 patients. The median follow-up was 411 days. There were no significant differences in clinical characteristics and comorbidities between patients with and without AF detected. AF was detected more often after posterior cerebral artery infarct with brainstem/cerebellum involvement (p < 0.001) whereas less often after infarction in the anterior and middle cerebral artery (p = 0.021). Multivariable regression analysis demonstrated that posterior cerebral artery infarct with brainstem/cerebellum involvement was an independent predictor of AF detection. Our study showed that posterior circulation infarcts with brainstem/cerebellum involvement are associated with AF detection in ESUS patients undergoing ILR. Larger prospective studies are needed to validate our findings.

Detection of Atrial Fibrillation From PPG Sensor Data Using Variational Mode Decomposition

Detection of Atrial Fibrillation From PPG Sensor Data Using Variational Mode Decomposition

Health behavior outcomes in stroke survivors prescribed wearables for atrial fibrillation detection stratified by age.

Smartwatches have become readily accessible tools for detecting atrial fibrillation (AF). There remains limited data on how they affect psychosocial outcomes and engagement in older adults. We examine the health behavior outcomes of stroke survivors prescribed smartwatches for AF detection stratified by age. We analyzed data from the Pulsewatch study, a randomized controlled trial that enrolled patients (≥ 50 years) with a history of stroke or transient ischemic attack and CHA2DS2-VASc ≥ 2. Intervention participants were equipped with a cardiac patch monitor and a smartwatch-app dyad, while control participants wore the cardiac patch monitor for up to 44 days. We evaluated health behavior parameters using standardized tools, including the Consumer Health Activation Index, the Generalized Anxiety Disorder questionnaire, the 12-Item Short Form Health Survey, and wear time of participants categorized into three age groups: Group 1 (ages 50-60), Group 2 (ages 61-69), and Group 3 (ages 70-87). We performed statistical analysis using a mixed-effects repeated measures linear regression model to examine differences amongst age groups. Comparative analysis between Groups 1, 2 and 3 revealed no significant differences in anxiety, patient activation, perception of physical health and wear time. The use of smartwatch technology was associated with a decrease in perception of mental health for Group 2 compared to Group 1 (β = -3.29, P = 0.046). Stroke survivors demonstrated a willingness to use smartwatches for AF monitoring. Importantly, among these study participants, the majority did not experience negative health behavior outcomes or decreased engagement as age increased.

Targeted screening of atrial fibrillation using automated blood pressure measurement device with atrial fibrillation detection function, in patients with type 2 diabetes mellitus in primary care setting

ObjectiveThe prevalence of atrial fibrillation (AF) in type 2 diabetes mellitus (DM) patients under primary care in Hong Kong was yet to be explored. We aimed to evaluate the prevalence of AF in patients with DM so as to provide evidence-based recommendations to incorporate AF screening as a component in regular diabetic risk and complication assessment. The performance of automated BP machine Microlife WatchBP Office AFIB as a screening tool for the detection of AF was also evaluated.MethodThis was a cross-sectional study. Patients with type 2 DM who attended the regular diabetic risk and complication assessment in the participating clinics from 24 August 2021 to 27 January 2022 were recruited. Blood pressure measurement by Microlife WatchBP Office AFIB and 12-lead ECGs were performed for AF screening.ResultsAmong 2015 DM patients in primary care, the prevalence of AF was found to be 1.9% (95% confidence interval [CI] 1.3–2.6). The prevalence of AF increased with age, from 0.5% in patients aged < 65 years, to 2.2% in patients aged 65–74 years and 4.3% in patients aged ≥ 75 years. The sensitivity and specificity of Microlife WatchBP Office AFIB to detect AF were 80% (95% Cl 61.8–92.3) and 97.9%. (95% CI 97.3–98.5), respectively. The positive and negative predictive values were 32.8% (95% CI 21.9–45.1) and 99.7% (95% CI 99.5–99.9) respectively.ConclusionsAF screening with the use of Microlife WatchBP Office AFIB is a simple procedure and can be considered as a standard assessment in the regular comprehensive diabetic risk and complication assessment in primary care setting.

Many Models, Little Adoption-What Accounts for Low Uptake of Machine Learning Models for Atrial Fibrillation Prediction and Detection?

(1) Background: Atrial fibrillation (AF) is a major risk factor for stroke and is often underdiagnosed, despite being present in 13-26% of ischemic stroke patients. Recently, a significant number of machine learning (ML)-based models have been proposed for AF prediction and detection for primary and secondary stroke prevention. However, clinical translation of these technological innovations to close the AF care gap has been scant. Herein, we sought to systematically examine studies, employing ML models to predict incident AF in a population without prior AF or to detect paroxysmal AF in stroke cohorts to identify key reasons for the lack of translation into the clinical workflow. We conclude with a set of recommendations to improve the clinical translatability of ML-based models for AF. (2) Methods: MEDLINE, Embase, Web of Science, Clinicaltrials.gov, and ICTRP databases were searched for relevant articles from the inception of the databases up to September 2022 to identify peer-reviewed articles in English that used ML methods to predict incident AF or detect AF after stroke and reported adequate performance metrics. The search yielded 2815 articles, of which 16 studies using ML models to predict incident AF and three studies focusing on ML models to detect AF post-stroke were included. (3) Conclusions: This study highlights that (1) many models utilized only a limited subset of variables available from patients' health records; (2) only 37% of models were externally validated, and stratified analysis was often lacking; (3) 0% of models and 53% of datasets were explicitly made available, limiting reproducibility and transparency; and (4) data pre-processing did not include bias mitigation and sufficient details, leading to potential selection bias. Low generalizability, high false alarm rate, and lack of interpretability were identified as additional factors to be addressed before ML models can be widely deployed in the clinical care setting. Given these limitations, our recommendations to improve the uptake of ML models for better AF outcomes include improving generalizability, reducing potential systemic biases, and investing in external validation studies whilst developing a transparent modeling pipeline to ensure reproducibility.

Sensitivity and specificity of handheld one lead ECG detecting atrial fibrillation in an outpatient clinic setting

AimsTo validate the sensitivity and specificity of the Zenicor One handheld ECG device for detection of atrial fibrillation in an outpatient clinical setting. Methods and resultsPatients attending outpatient clinics at Danderyd Hospital (n = 220) were examined with one lead handheld ECG immediately after standard care 12‑lead ECG recording. Twelve recordings were excluded (atrial flutter or pacing) or missing. The recordings were dichotomously categorized as “atrial fibrillation” or “not atrial fibrillation” by two senior cardiologists. In cases of diverging interpretations, a third senior cardiologist had the deciding vote. Sensitivity and specificity in diagnosing atrial fibrillation was calculated with 12‑lead ECG as gold standard.Sensitivity and specificity for diagnosis of atrial fibrillation with one lead handheld ECG and 12‑lead ECG as gold standard was 98% and 99% respectively. ConclusionIn a health-care outpatient setting, Zenicor One handheld ECG had high sensitivity and specificity for detection of atrial fibrillation when compared with 12‑lead ECG.

Very-early detection of atrial fibrillation after ablation evaluated by a wearable ECG-patch predicts late blanking period recurrence: Preliminary data from a prospective registry

IntroductionAtrial fibrillation (AF) ablation represents a safe and effective procedure to restore sinus rhythm. The idea that post-procedural AF episodes - during theblanking period-are not considered treatment failure has been increasingly challenged. The E-Patch, a single-use adhesive electrode, facilitates extended continuous ECG monitoring for 120 h. This pilot study aims to assess the effectiveness of this ambulatory monitoring device and investigate whether very-early AF recurrence correlates with delayed blanking period ablation outcomes. MethodsWe conducted a single-center, prospective, longitudinal study, including consecutive post-ablation patients monitored with the E-patch. The ability of the device to continuously record was analyzed, as well as the occurrence of AF episodes during external 7-day loop-recorder in the 2nd-month post-ablation. ResultsWe included 40 patients, median age 62 years (IQR 56–70). E-Patch monitoring was obtained for a median of 118 h (IQR 112–120), with no discomfort nor interpretation artefacts. Very-early AF recurrence was detected in 11 (27.5 %) patients, with a median AF burden of 7 % (IQR 6 %–33 %). Late-blanking period AF was detected in 13 (33 %) of theexternal 7-day looprecordings.Of the 11 patients that had very-early AF recurrence, 10 (91 %) had late-blanking AF. Very-early AF detectionshowed77 % (95 % CI 64 %-90 %) sensitivityand 96 % (95 % CI 90–100 %) specificity inpredictinglate-blanking AF, with a non-parametric ROCcurve AUC of 0.903 (95 % 0.797––1.0). ConclusionThe E-Patch was able to detect very-early AF during an extended period. Very-early AF detection emerges as a predictor of AF recurrence during the late blanking period post-ablation.

Early Detection of Atrial Fibrillation in Chronic Obstructive Pulmonary Disease Patients.

Atrial fibrillation (AF) is an important medical problem, as it significantly affects patients' quality of life and prognosis. AF often complicates the course of chronic obstructive pulmonary disease (COPD), a widespread disease with heavy economic and social burdens. A growing body of evidence suggests multiple links between COPD and AF. This review considers the common pathogenetic mechanisms (chronic hypoxia, persistent inflammation, endothelial dysfunction, and myocardial remodeling) of these diseases and describes the main risk factors for the development of AF in patients with COPD. The most effective models based on clinical, laboratory, and functional indices are also described, which enable the identification of patients suffering from COPD with a high risk of AF development. Thus, AF in COPD patients is a frequent problem, and the search for new tools to identify patients at a high risk of AF among COPD patients remains an urgent medical problem.

An atrial fibrillation detection algorithm based on lightweight design architecture and feature fusion strategy

BackgroundAtrial fibrillation (AF) is one of the common types of cardiac arrhythmias, and its medical burden is continuously increasing. Wearable ECG signal analysis based on deep learning (DL) is an effective approach for screening AF. However, existing DL algorithms require extensive computational resources for AF recognition, hindering their clinical applicability. ObjectiveThis study aims to develop a lightweight DL model to address the challenges of DL algorithms in the clinical AF recognition domain. MethodUsing a distributed approach, layer-by-layer cross-guidance mechanism, and attention fusion mechanism, we designed a lightweight cross-guidance network (LCG-Net). The main path uses lightweight depth-wise separable convolutions to extract deep-level information of AF, while the auxiliary path uses standard convolutions to compensate for the weak feature expression capability of depth-wise separable convolutions. Based on the idea of mutual guidance, a layer-by-layer cross-guidance mechanism is designed to achieve information interaction and fusion between depthwise separable convolutions and standard convolutions. An attention fusion mechanism is developed based on attention mechanism and 2D convolution templates to select and precisely fuse information from different paths and different layers. ResultLCG-Net has only 39.04 K parameters and 8.16 M computations. On a clinical dataset consisting of ECG records from 252 patients, it achieved accuracy and F1 scores of 98.39 % and 98.38 %, respectively. ConclusionThe proposed LCG-Net demonstrates excellent lightweight, stability, and accuracy, holding promising prospects in the clinical diagnosis of AF.

Artificial intelligence for atrial fibrillation detection, prediction, and treatment: A systematic review of the last decade (2013–2023)

AbstractAtrial fibrillation (AF) affects more than 30 million individuals worldwide, making it the most prevalent cardiac arrhythmia on a global scale. This systematic review summarizes recent advancements in applying artificial intelligence (AI) techniques for AF detection, prediction, and guiding treatment selection and risk stratification. In adherence with the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses), a total of 171 pertinent studies conducted between 2013 and 2023 were examined. Studies applying machine learning (ML) and deep learning (DL) to electrocardiogram (ECG), photoplethysmography (PPG), wearable data, and other sources were evaluated. For AF detection, most works employed DL (48 studies) and ML (28 studies) on ECG data. DL methods directly analyzed ECG waveforms and outperformed approaches relying on hand‐crafted features. For prediction and risk stratification, 22 studies used ML while 7 leveraged DL on ECG. An emerging trend showed the growing potential of PPG for AF screening. Overall, AI demonstrated promising capabilities across various AF‐related tasks. However, real‐world implementation faces challenges including a lack of interpretability, the need for multimodal data integration, prospective performance validation, and regulatory compliance. Future research directions involve quantifying model uncertainty, enhancing transparency, and conducting population‐based clinical trials to facilitate translation into practice.This article is categorized under: Application Areas &gt; Health Care Application Areas &gt; Science and Technology Technologies &gt; Artificial Intelligence

Role of Implantable Looper in Atrial Fibrillation Detection

Role of Implantable Looper in Atrial Fibrillation Detection

Protocol for a multicentre and prospective follow-up cohort study of early detection of atrial fibrillation, silent stroke and cognitive impairment in high-risk primary care patients: the PREFA-TE study

BackgroundAtrial fibrillation (AF) is the most common type of cardiac arrhythmia. Future estimations suggest an increase in global burden of AF greater than 60% by 2050. Numerous studies provide growing...

The Rate of Success of CPR in Patients Suffering from Cardiac Arrest in Patients Admitted in CCU in Cardiology Department Ayub Medical Teaching Institute

Cardiopulmonary resuscitation (CPR) is an essential procedure used to treat patients who are in cardiac arrest, particularly in the Critical Care Unit (CCU). Objective: To assess the success rate of CPR admitted to the CCU, revealing insight on the effectiveness of current resuscitation techniques. Methods: This retrospective study was conducted at Department of Cardiology Ayub Medical Teaching Institute, Abbottabad, between 13th April 2023 to 30th November 2023, 110 patients had cardiac arrest while in the intensive care unit. Data were gathered and examined on patient demographics, pre-existing comorbidities, time to start CPR, duration of CPR, and results. CPR success was defined as a sustained restoration of spontaneous circulation (ROSC) for at least 20 minutes. Results: This study included 110 cardiac patients. Most patients (68.18%) were male. Hypertension was the most common comorbidity at 72.7%. CPR started on average 4.8 minutes late, with a 1.2-minute standard deviation. The average CPR time was 18.2 minutes, with a 5.6-minute SD. After cardiac arrest, 60 (54.5%) patients began CPR within 5 minutes. Another subgroup found 96.0% CPR success in 25 (22.7%) ventricular fibrillation patients. Then occurred ventricular tachycardia (80.0%), pulseless electrical activity (76.0%), and asystole (68.9%). These data suggest that CPR works better in ventricular fibrillation patients. Conclusions: This research sheds light on cardiac arrest CCU patients' demographics, care, and outcomes. The data show that early CPR and ventricular fibrillation detection and treatment improve outcomes for these individuals.

Light Weight Residual Convolutional Neural Network for Atrial Fibrillation Detection in Single-lead ECG Recordings

Light Weight Residual Convolutional Neural Network for Atrial Fibrillation Detection in Single-lead ECG Recordings

Diagnostic Performance of Automated Blood Pressure Monitor for Detection of Atrial Fibrillation.

Background Atrial fibrillation (AF) is a type of heart disease characterized by an irregular cardiac rhythm. The complications of AF are associated with significant morbidity, mortality, and medical expenses. This emphasizes the significance of detecting AF early using a feasible device. Methods A total of 123 patients who attended cardiology and INR clinics were enrolled, with 51 of them having AF. The blood pressure of all patients was measured three times using the Rossmax X5, while a single-lead electrocardiogram (ECG) was monitored simultaneously. Following that, a 12-lead ECG was performed on all patients. A cardiologist confirmed the irregular rhythm. Results Compared to the 12-lead ECG method, Rossmax X5 has an accuracy of 99.3%, a sensitivity of 100%, and a specificity of 98.6%. The positive and negative predictive values were also significant, which were 98.1% and 100%, respectively. Conclusion The Rossmax X5 automated blood pressure monitor has a high detection accuracy for AF. Therefore, Rossmax X5 can be recommended for use in the clinical setting as a screening tool for early AF detection.

Minirocket Kullanarak Güçlendirilmiş ve Verimli Atriyal Fibrilasyon Tespiti

Atrial Fibrillation (AF) detection from intracardiac Electrogram (EGM) signals is a critical aspect of cardiovascular health monitoring. This study explores the application of Minirocket, a time series classification (TSC) algorithm, for robust and efficient AF detection. A comparative analysis is conducted against a deep learning approach using a subset of the dataset from Rodrigo et al. (2022). The study investigates the robustness of Minirocket in the face of shorter EGM sequences and varying training sizes, essential for real-world applications such as wearable and implanted devices. Empirical runtime analysis further assesses the efficiency of Minirocket in comparison to conventional machine learning (ML) algorithms. The results showcase Minirocket's notable performance, especially in scenarios with shorter signals and varying training sizes, making it a promising candidate for streamlined AF detection in emerging cardiovascular monitoring technologies. This research contributes to the optimization of AF detection algorithms for increased efficiency and adaptability to dynamic clinical scenarios.

Atrial Fibrillation Detection from Compressed ECG Measurements for Wireless Body Sensor Network

Recent years have witnessed an increasing prevalence of wearable devices in the public, where atrial fibrillation (AF) detection is a popular application in these devices. Generally, AF detection is performed on cloud whereas this paper describes an on-device AF detection method. Technically, compressed sensing (CS) is first used for electrocardiograph (ECG) acquisition. Then QRS detection is proposed to be performed directly on the compressed CS measurements, rather than on the reconstructed signals on the powerful cloud server. Based on the extracted QRS information, AF is determined by quantitatively analyzing the ( RR , dRR ) plot. Databases with ECG samples collected from both medical-level (MIT-BIH afdb) and wearable ECG devices (Physionet Challenge 2017) are introduced for performance validation. The experiment results well demonstrate that our on-device AF detection algorithm can approach the performance of those implemented on the raw signals. Our proposal is suitable for AF screening directly on the wearable devices, without the support of the data center for signal reconstruction and intelligent analysis.

Gender and age characteristics and atrial fibrillation during long-term telemonitoring of the ECG

The aim – to evaluate the possibilities of long-term patch monitoring of the electrocardiogram (ECG) to determine gender and age characteristics of the frequency of detection of atrial fibrillation (AF). Materials and methods. A total of 1,123 patients participated in this 2-year (2021–2023) retrospective multicenter study. Long-term ECG monitoring was performed at the referral of internists and cardiologists of 30 health care institutions according to various clinical indications for this study. For long-term monitoring of the ECG, the innovative device «Mawi Heart Patch» was used – a disposable two-channel device with the possibility of continuous recording of the ECG for up to 7 days. The primary analysis of the records was carried out using Cardio.ai artificial intelligence algorithms, re-decryption and validation by certified doctors. Results. The average age of the study participants was (50.2 ± 16.65) years. There were 599 (53.3 %) women, 524 (46.6 %) men. The average time of ECG monitoring was (81.03 ± 18.5) hours, 74.35 % of the participants had continuous ECG monitoring for 3 days, 20.75 % for 4 days, and 2.58 % for 5 days. In total, AF was diagnosed in 7.93 % of patients, in particular, women and men were 6.51 % and 9.54 % respectively (p &gt; 0.05). In elderly patients, AF was diagnosed statistically significantly more often than in middle-aged patients (18.5 and 3.6 %, respectively, p ˂ 0.001). Among all age groups, the highest frequency of detection of AF was in participants over 75 years old, compared to young (28.8 % vs 3.6 %, p ˂ 0.001) and middle-aged patients (28.8 vs 18.5 %, p ˂ 0.05). Conclusions. The study did not reveal a gender difference in the frequency of diagnosing AF. In patients over 50 years, a significant increase in the frequency of detection of AF was recorded, which tends to increase in the following age periods and reaches a maximum in long-livers (over 90 years). The use of long-term patch ECG monitoring in patients of older age groups expands diagnostic possibilities and makes it possible to prevent complications in time.

Atrial Fibrillation Detection with Single-Lead Electrocardiogram Based on Temporal Convolutional Network-ResNet.

Atrial fibrillation, one of the most common persistent cardiac arrhythmias globally, is known for its rapid and irregular atrial rhythms. This study integrates the temporal convolutional network (TCN) and residual network (ResNet) frameworks to effectively classify atrial fibrillation in single-lead ECGs, thereby enhancing the application of neural networks in this field. Our model demonstrated significant success in detecting atrial fibrillation, with experimental results showing an accuracy rate of 97% and an F1 score of 87%. These figures indicate the model's exceptional performance in identifying both majority and minority classes, reflecting its balanced and accurate classification capability. This research offers new perspectives and tools for diagnosis and treatment in cardiology, grounded in advanced neural network technology.

Directional statistics-inspired end-to-end atrial fibrillation detection model based on ECG rhythm

Numerous algorithms designed for detecting atrial fibrillation (AF) often exhibit limitations in extracting essential rhythm features, leading to challenges in accurately discerning ectopic beats and consequently yielding suboptimal detection performance. In this study, we explored the distribution patterns of R-R intervals (RRIs) in both AF and specific rhythms, such as atrial premature beats (APBs) and normal sinus rhythm (NSR). For the first time, we employed the Sobolev test statistics, a method used in directional statistics to assess spherical uniformity, to quantify the irregularity and variability characteristics of RRIs. We developed an end-to-end learnable model for detecting AF by leveraging this approach. A cross-dataset validation method is employed to train and test the proposed model. This involved the use of a simulated dataset and eight distinct real-world databases. Notably, when trained on the Computing in Cardiology Challenge 2017 (C2017) and tested on the MIT-BIH Atrial Fibrillation Database (AFDB), our model, designed to take only a sequence of 32 RRIs as input, achieved a sensitivity of 96.2%, specificity of 98.2%, and accuracy of 97.3%. The results illustrate its competitive standing against the existing methods for AF detection and its enhanced resilience to ectopic beats. Unlike existing deep learning-based AF detection models, our model prioritizes interpretability, boasts lower computational complexity (with fewer than 2000 learnable parameters), and demonstrates superior generalization capabilities. This will help improve the quality of long-term real-time monitoring and management of AF, reduce the burden on clinicians, and ultimately improve patient outcomes.