OBJECTIVES: To design a deep learning model that balances model complexity and performance to enable its integration into wearable ECG monitoring devices for automated diagnosis of atrial fibrillation. METHODS: This study was performed based on data from 84 patients with atrial fibrillation, 25 patients with atrial fibrillation, and 18 subjects without obvious arrhythmia collected from the publicly available datasets LTAFDB, AFDB, and NSRDB, respectively. A lightweight attention network based on depthwise separable convolution and fusion of channel-spatial information, namely DSC-AttNet, was proposed. Depthwise separable convolution was introduced to replace standard convolution and reduce model parameters and computational complexity to realize high efficiency and light weight of the model. The multilayer hybrid attention mechanism was embedded to compute the attentional weights of the channels and spatial information at different scales to improve the feature expression ability of the model. Ten-fold cross-validation was performed on LTAFDB, and external independent testing was conducted on AFDB and NSRDB datasets. RESULTS: DSC-AttNet achieved a ten-fold average accuracy of 97.33% and a precision of 97.30% on the test set, both of which outperformed the other 4 comparison models as well as the 3 classical models. The accuracy of the model on the external test set reached 92.78%, better than those of the 3 classical models. The number of parameters of DSC-AttNet was 1.01M, and the computational volume was 27.19G, both smaller than the 3 classical models. CONCLUSIONS This proposed method has a smaller complexity, achieves better classification performance, and has a better generalization ability for atrial fibrillation classification.
Atrial Fibrillation/diagnosis* ; Humans ; Electrocardiography ; Deep Learning ; Wearable Electronic Devices ; Neural Networks, Computer

INTRODUCTION: Arrhythmias, especially atrial fibrillation (AF) and ventricular arrhythmias, are independent risk factors of mortality in patients with ischaemic heart disease (IHD). While there is a growing body of evidence that suggests an association between obstructive sleep apnoea (OSA) and cardiac arrhythmias, evidence on this relationship in patients with IHD has been scant and inconsistent. We hypothesised that in patients with IHD, severe OSA is associated with an increased risk of nocturnal arrhythmias. METHODS: We studied 103 consecutive patients with IHD who underwent an overnight polysomnography. Exposed subjects were defined as patients who had an apnoea-hypopnoea index (AHI) ≥30/h (severe OSA), and nonexposed subjects were defined as patients who had an AHI <30/h (nonsevere OSA). All electrocardiograms (ECGs) were interpreted by the Somte ECG analysis software and confirmed by a physician blinded to the presence or absence of exposure. Arrhythmias were categorised as supraventricular and ventricular. Arrhythmia subtypes (ventricular, atrial and conduction delay) were analysed as dichotomous outcomes using multiple logistic regression models. RESULTS: Atrial fibrillation and AF/flutter (odds ratio 13.5, 95% confidence interval 1.66-109.83; P = 0.003) were found to be more common in the severe OSA group than in the nonsevere OSA group. This association remained significant after adjustment for potential confounders. There was no significant difference in the prevalence of ventricular and conduction delay arrhythmias between the two groups. CONCLUSION In patients with IHD, there was a significant association between severe OSA and nocturnal AF/flutter. This underscores the need to evaluate for OSA in patients with IHD, as it may have important implications on clinical outcomes.
Humans ; Sleep Apnea, Obstructive/diagnosis* ; Atrial Fibrillation/diagnosis* ; Male ; Female ; Middle Aged ; Polysomnography ; Electrocardiography ; Myocardial Ischemia/complications* ; Aged ; Risk Factors ; Logistic Models

The risk prediction of paroxysmal atrial fibrillation (PAF) is a challenge in the field of biomedical engineering. This study integrated the advantages of machine learning feature engineering and end-to-end modeling of deep learning to propose a PAF risk prediction method based on multimodal feature fusion. Additionally, the study utilized four different feature selection methods and Pearson correlation analysis to determine the optimal multimodal feature set, and employed random forest for PAF risk assessment. The proposed method achieved accuracy of (92.3 ± 2.1)% and F1 score of (91.6 ± 2.9)% in a public dataset. In a clinical dataset, it achieved accuracy of (91.4 ± 2.0)% and F1 score of (90.8 ± 2.4)%. The method demonstrates generalization across multi-center datasets and holds promising clinical application prospects.
Humans ; Atrial Fibrillation/diagnosis* ; Machine Learning ; Deep Learning ; Risk Assessment/methods*

Atrial fibrillation is a common arrhythmia, and its diagnosis is interfered by many factors. In order to achieve applicability in diagnosis and improve the level of automatic analysis of atrial fibrillation to the level of experts, the automatic detection of atrial fibrillation is very important. This study proposes an automatic detection algorithm for atrial fibrillation based on BP neural network (back propagation network) and support vector machine (SVM). The electrocardiogram (ECG) segments in the MIT-BIH atrial fibrillation database are divided into 10, 32, 64, and 128 heartbeats, respectively, and the Lorentz value, Shannon entropy, K-S test value and exponential moving average value are calculated. These four characteristic parameters are used as the input of SVM and BP neural network for classification and testing, and the label given by experts in the MIT-BIH atrial fibrillation database is used as the reference output. Among them, the use of atrial fibrillation in the MIT-BIH database, the first 18 cases of data are used as the training set, and the last 7 cases of data are used as the test set. The results show that the accuracy rate of 92% is obtained in the classification of 10 heartbeats, and the accuracy rate of 98% is obtained in the latter three categories. The sensitivity and specificity are both above 97.7%, which has certain applicability. Further validation and improvement in clinical ECG data will be done in next study.
Humans ; Atrial Fibrillation/diagnosis* ; Support Vector Machine ; Heart Rate ; Algorithms ; Neural Networks, Computer ; Electrocardiography

INTRODUCTION: MyDiagnostick is an atrial fibrillation (AF) screening tool that has been validated in the Caucasian population in the primary care setting. METHODS: In our study, we compared MyDiagnostick with manual pulse check for AF screening in the community setting. RESULTS: In our cohort of 671 candidates from a multi-ethnic Asian population, AF prevalence was found to be 1.78%. Of 12 candidates, 6 (50.0%) had a previous history of AF and another 6 (50.0%) were newly diagnosed with AF. Candidates found to have AF during the screening were older (72.0 ± 11.7 years vs. 56.0 ± 13.0 years, P < 0.0001) and had a higher CHADSVASC risk score (2.9 ± 1.5 vs. 1.5 ± 1.1, P = 0.0001). MyDiagnostick had a sensitivity of 100.0% and a specificity of 96.2%. In comparison, manual pulse check had a sensitivity of 83.3% and a specificity of 98.9%. CONCLUSION MyDiagnostick is a simple AF screening device that can be reliably used by non-specialist professionals in the community setting. Its sensitivity and specificity are comparable and validated across various studies performed in different population cohorts.
Humans ; Atrial Fibrillation/diagnosis* ; Heart Rate ; Sensitivity and Specificity ; Risk Factors ; Electrocardiography ; Mass Screening

Objective: To explore the impact of non-valvular atrial fibrillation (AF) on the global cognitive function and executive function of patients without dementia, and to observe the differences between different types of AF. Methods: This research is a prospective and cross-sectional study. Non-dementia patients admitted to the department of neurology in the third people's hospital of Chengdu from July 2018 to July 2019 were included. Patients with non-valvular AF were included in the AF group and those with sinus rhythm were included in the control group. General clinical data and compared global cognitive function (mini-mental state examination (MMSE) and montreal cognitive assessment (MOCA)) and executive function (shape trails test (STT) and stroop color and word test (SCWT)) data were obtained and compared between 2 groups, and between different AF type groups. Results: A total of 386 participants were included, including 203 in AF group (52.6%), age was 68 (63, 71) years old, 119 were male (58.6%) and 183 in control group, age was 68 (63, 71) years old, 101 were male (55.2%). MMSE(28 (27, 29)) and MOCA (25 (22, 26)) scores were lower in AF group than those in control group (P<0.05), while STT-A time (84 (64, 140) s), STT-B time (248 (184, 351) s), STT time difference((159 (106, 245) s), SCWT-A time (50 (50, 50) s), SCWT-B time (55 (46, 63) s), SCWT-C time (100 (86, 120) s) and SCWT time interference (46 (34, 65) s) were higher than those in control group (P<0.05). Moreover, there was no difference in above indexes between paroxysmal AF and non-paroxysmal AF. Conclusion: The global cognitive function and executive function of patients with non-valvular AF are both decreased, while there is no obvious difference of the global cognitive function and executive function between paroxysmal AF and non-paroxysmal AF patients.
Humans ; Male ; Female ; Atrial Fibrillation/diagnosis* ; Executive Function ; Prospective Studies ; Cross-Sectional Studies ; Cognition Disorders/diagnosis* ; Cognition

Atrial Fibrillation/diagnosis* ; Humans ; Quality of Life ; Self Report ; Symptom Assessment

Atrial fibrillation (AF) is a common arrhythmia, which can lead to thrombosis and increase the risk of a stroke or even death. In order to meet the need for a low false-negative rate (FNR) of the screening test in clinical application, a convolutional neural network with a low false-negative rate (LFNR-CNN) was proposed. Regularization coefficients were added to the cross-entropy loss function which could make the cost of positive and negative samples different, and the penalty for false negatives could be increased during network training. The inter-patient clinical database of 21 077 patients (CD-21077) collected from the large general hospital was used to verify the effectiveness of the proposed method. For the convolutional neural network (CNN) with the same structure, the improved loss function could reduce the FNR from 2.22% to 0.97% compared with the traditional cross-entropy loss function. The selected regularization coefficient could increase the sensitivity (SE) from 97.78% to 98.35%, and the accuracy (ACC) was 96.62%, which was an increase from 96.49%. The proposed algorithm can reduce the FNR without losing ACC, and reduce the possibility of missed diagnosis to avoid missing the best treatment period. Meanwhile, it provides a universal loss function for the clinical auxiliary diagnosis of other diseases.
Algorithms ; Atrial Fibrillation/diagnosis* ; Electrocardiography ; Humans ; Neural Networks, Computer ; Stroke

Infective endocarditis (IE) is not a common cause of stroke. Considering the high mortality rates, however, IE should always be considered as a possible cause of stroke even when the chances are low. Atrioesophageal fistula is a life-threatening condition that can cause IE and subsequent stroke, but the diagnosis is often delayed due to its rarity. We report a case of multiple embolic infarcts caused by infective endocarditis associated with atrioesophageal fistula after radiofrequency catheter ablation for atrial fibrillation.
Atrial Fibrillation ; Catheter Ablation ; Diagnosis ; Endocarditis ; Esophageal Fistula ; Fistula ; Mortality ; Stroke

Atrial fibrillation (AF) is one of the most common arrhythmias, which does great harm to patients. Effective methods were urgently required to prevent the recurrence of AF. Four methods were used to analyze RR sequence in this paper, and differences between Pre-AF (preceding an episode of AF) and Normal period (far away from episodes of AF) were analyzed to find discriminative criterion. These methods are: power spectral analysis, approximate entropy (ApEn) and sample entropy (SpEn) analysis, recurrence analysis and time series symbolization. The RR sequence data used in this research were downloaded from the Paroxysmal Atrial Fibrillation Prediction Database. Supporting vector machine (SVM) classification was used to evaluate the methods by calculating sensitivity, specificity and accuracy rate. The results showed that the comprehensive utilization of recurrence analysis parameters reached the highest accuracy rate (95%); power spectrum analysis took second place (90%); while the results of entropy analyses and time sequence symbolization were not satisfactory, whose accuracy were both only 70%. In conclusion, the recurrence analysis and power spectrum could be adopted to evaluate the atrial chaotic state effectively, thus having certain reference value for prediction of AF recurrence.
Atrial Fibrillation ; diagnosis ; Entropy ; Heart Atria ; physiopathology ; Humans ; Recurrence ; Sensitivity and Specificity ; Support Vector Machine