ID: 4349750 SCALABLE AF DETECTION USING CNNS WITH STOCKWELL AND FOURIER TRANSFORMS: A SIGNAL PROCESSING APPROACH FOR CLINICAL AND CONSUMER ECGS
Early detection of atrial fibrillation (AF) is critical for stroke prevention and risk reduction. While most convolutional neural networks (CNNs) are trained on raw time-series ECG signals, alternative signal processing approaches may improve model performance and generalizability.
Authors
- Luiz Vasconcelos (ORCID: https://orcid.org/0000-0003-0013-1759)
- Elan Shahrabani
- Ivan Nenadic Wood
- Hakan Oral
Publication Details
- Journal
- Heart Rhythm O2
- Published
- 2025-09-01
- DOI
- https://doi.org/10.1016/j.hroo.2025.07.085
- Primary Topic
- Wireless Body Area Networks
- Type
- article
- Field-Weighted Citation Impact
- 0.00