Efficient model predicts atrial fibrillation one hour before onset
AF-Mamba: Efficient Long-Term Signal Modeling for Early Prediction of Atrial Fibrillation Onset
Machine Learning
Summary
Atrial fibrillation is a common heart problem that can increase the risk of stroke and heart failure. The authors developed a computer program called AF-Mamba that uses heart signal data collected over an hour to predict when atrial fibrillation might start. Their approach combines two types of artificial intelligence techniques to analyze long sequences of heartbeats efficiently. The model showed good accuracy in tests and worked well across different datasets, suggesting it could help with early, continuous monitoring of heart health using wearable devices.
atrial fibrillationelectrocardiogram (ECG)RR intervalsdeep learningtemporal convolutional networksstate-space modelsAUROCsensitivity and specificitytime-series prediction
Authors
Yongbin Lee, Ki H. Chon
Abstract
Atrial fibrillation (AF) is the most common cardiac arrhythmia and is associated with increased risks of stroke and heart failure. The growing availability of wearable and portable ECG monitoring enables continuous assessment of cardiac rhythm outside clinical settings. Predicting AF before its onset could provide additional lead time for timely clinical assessment and potentially improve the management of patients at risk of AF-related complications. This study focuses on predicting AF onset one hour in advance using long-term RR intervals (RRIs). To address this challenge, we propose a deep learning architecture that integrates temporal convolutional networks (TCNs) for local features encoding with Mamba, a selective state-space model capable of long-range sequence modeling. This hybrid TCN-Mamba design enables efficient training and inference on one-hour input windows, overcoming limitations of Transformers' quadratic scaling and recurrent networks' vanishing gradients. In subject-wise 5-fold testing, the proposed model achieved a sensitivity of 0.889, specificity of 0.943, F1-score of 0.813, AUROC of 0.974, and AUPRC of 0.933. In paired cross-dataset holdout evaluation, AF-Mamba maintained discriminative performance across unseen AF and NSR datasets, achieving a mean AUROC of 0.897. Compared against state-of-the-art AF prediction models and general time-series models, AF-Mamba achieved competitive predictive performance while providing a favorable performance-efficiency trade-off for long RRI sequences. These findings demonstrate the potential of AF-Mamba for accurate AF prediction one hour in advance and real-time continuous ambulatory monitoring.