Key result
CTRhythm outperforms baseline models for single-lead ECG AF detection with an F1 score of ~0.83.
Why the study?
Current CNN and CRNN methods for single-lead ECG AF detection capture only local patterns, failing to capture rich long-range dependencies.
Does the CNN-Transformer Rhythm Classifier (CTRhythm) improve atrial fibrillation detection from single-lead ECGs compared to baseline deep learning models?
Does the CNN-Transformer Rhythm Classifier (CTRhythm) improve atrial fibrillation detection from single-lead ECGs compared to baseline deep learning models?
Absolute Event Rate: 0.831% vs 0.826%
The integration of CNN and Transformer architectures in CTRhythm improves the accuracy of atrial fibrillation detection from single-lead ECGs.
Outperforms CNN/CRNN baselines for single-lead AF detection (F1 0.831); leaves open prospective wearable validation.
Atrial Fibrillation (AF) is a common supraventricular arrhythmia that affects about 30 million people globally. Electrocardiogram (ECG) analysis is the primary diagnostic approach. The widespread adoption of wearable devices monitoring heart rhythm prompted the development of AF detection models for single-lead ECGs, benefitting real-time early diagnosis. Current state-of-the-art methods for AF detection are convolutional neural network (CNN) and convolutional recurrent neural network (CRNN) based models, which only focus on capturing local patterns despite heart rhythms exhibiting rich long-range dependencies. To address this limitation, we propose a novel method for single-lead ECG rhythm classification, termed CNN-Transformer Rhythm Classifier (CTRhythm), which integrates CNN with a Transformer encoder to capture local and global patterns effectively. CTRhythm achieved an overall F1 score of 0.831, outperforming the baseline deep learning models on the golden standard CINC2017 dataset. Moreover, pre-training with additional data improved the overall F1 score to 0.840. In two external validation datasets, CTRhythm showed its strong generalization capabilities. CTRhythm is freely available at https://github.com/labxscut/CTRhythm .
No takes yet. Share an insight, caveat, or question.
Yang et al. (2024) studied Atrial Fibrillation (n=8,528). CTRhythm (CNN-Transformer Rhythm Classifier) vs. Baseline deep learning models was evaluated on Overall F1 score. The CTRhythm model, integrating CNN and Transformer architectures, achieved an overall F1 score of 0.831 for atrial fibrillation detection from single-lead ECGs, outperforming baseline models.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: