Key result
Novel ECG signal decomposition and neural network achieves up to ~0.98 F1 score for AF detection.
Why the study?
Detecting AF using wearable single-lead ECG recordings is challenging, and current deep neural networks often use raw signals while disregarding clinical ECG diagnostic experience.
A novel ECG signal decomposition technique combined with a dilated residual neural network achieved high accuracy in detecting atrial fibrillation from single-lead ECG recordings.
Answers are grounded in this paper and cite their sources.
May improve algorithmic AF detection; hypothesis-generating pending prospective clinical validation.
Objective . Detecting atrial fibrillation (AF) using electrocardiogram (ECG) recordings from wearable devices has been challenging due to factors such as low signal-to-noise ratio and the use of only one lead. The use of deep learning has become a popular approach to tackle this task. However, it has been observed that current methods based on deep neural networks tend to favor raw signals as input, disregarding the valuable clinical experience in ECG diagnosis. Approach. In this study, we proposed a novel feature extraction method that generates a pseudo QRS complex signal and a pseudo T, P wave signal for each raw ECG signal using a temporal mask built upon R peak detection. Then a novel dilated residual neural network was trained on the decomposed signal. Main results. We evaluated the performance of our method on the dataset of PhysioNet/CinC 2017 Challenge, achieving an average F 1 ¯ score of 0.843. The method was further tested on MIT-BIH Atrial Fibrillation Database, and an average F 1 ¯ score of 0.984 was obtained. Significance. Our proposed ECG signal decomposition technique introduces simple and reliable domain knowledge into deep neural networks, and the dilated residual network provides large and flexible receptive fields, thereby enhancing the performance in the detection of AF. Our method can be extended to many other tasks involving ECG signals.
No takes yet. Share an insight, caveat, or question.
Li et al. (2023) studied Atrial fibrillation. Signal decomposition and dilated residual neural network was evaluated on Average F1 score for atrial fibrillation detection. A novel ECG signal decomposition technique and dilated residual neural network achieved an average F1 score of 0.843 on the PhysioNet/CinC 2017 dataset and 0.984 on the MIT-BIH dataset for AF detection.
Synapse has enriched 10 related papers. Consider them for comparative context: