Poincaré Plot Image and Rhythm-Specific Atlas for Atrial Bigeminy and Atrial Fibrillation Detection

Semi-supervised learning for automatic atrial fibrillation detection in 24-hour Holter monitoring

Paroxysmal atrial fibrillation (AF) is generally diagnosed by long-term dynamic electrocardiogram (ECG) monitoring. Identifying AF episodes from long-term ECG data can place a heavy burden on clinicians. Many machine-learning-based automatic AF detection methods have been proposed to solve this issue. However, these methods require numerous annotated data to train the model, and the annotation of AF in long-term ECG is extremely time-consuming. Reducing the demand for labeled data can effectively improve the clinical practicability of automatic AF detection methods. In this study, we developed a novel semi-supervised learning method that generated modified low-entropy labels of unlabeled samples for training a deep learning model to automatically detect paroxysmal AF in 24 h Holter monitoring data. Our method employed a 1D CNN-LSTM neural network with RR intervals as input and used few labeled training data with numerous unlabeled data for training the neural network. This method was evaluated using a 24 h Holter monitoring dataset collected from 1000 paroxysmal AF patients. Using labeled samples from only 10 patients for model training, our method achieved a sensitivity of 97.8%, specificity of 97.9%, and accuracy of 97.9% in five-fold cross-validation. Compared to the supervised learning method with complete labeled samples, the detection accuracy of our method was only 0.5% lower, while the workload of data annotation was significantly reduced by more than 98%. In general, this is the first study to apply semi-supervised learning techniques for automatic AF detection using ECG. Our method can effectively reduce the demand for AF data annotations and can improve the clinical practicability of automatic AF detection.

Autonomic Tone Variations Prior to Onset of Paroxysmal Atrial Fibrillation

BACKGROUND Variations of heart rate variability (HRV) before paroxysmal atrial fibrillation (PAF) onset are still controversial. We aimed to observe the autonomic tone variations before PAF onset based on HRV analysis. MATERIAL AND METHODS We prospectively investigated 24-h Holter recordings of 60 patients with PAF (M/F: 34/26) and 40 healthy people in sinus rhythm (M/F: 18/12). According to clinical information and Poincare scatter plot, 60 PAF patients were divided into sympathetic group (n=20) and vagus group (n=40). Time domain and frequency domain parameters of HRV were respectively measured before PAF episodes in 3 subgroups. Five time periods were studied using the ANOVA. RESULTS No significant variations were observed for the HRV parameters during 60 minutes preceding PAF in sympathetic group. A significant and linear change in SDNN, RMSSD, PNN50, HF and LF/HF during 60 minutes preceding PAF onset in vagus group. Compared with controls, RMSSD, LF and HF were significantly longer in patients with PAF during 60 minutes before PAF. Comparing sympathetic group and vagus group, we observed the same pattern of autonomic variations with a progressive decrease in LF and HF. A progressive decrease in PNN50 and LF/HF of sympathetic group and a significant increase in PNN50 and LF/HF of vagus group were also observed. CONCLUSIONS Patients with PAF mediated by different autonomic nerves have HRV variations, especially vagus PAF, there was a progressive increase with vagal tone during 60 minutes before PAF onset. The findings may help clinicians better intervene in PAF.