Synthesized Atrial Fibrillation-Like ECG Using Generative Adversarial Network
This is the dataset associated with the IEEE-JBHI submission "Synthesizing Electrocardiograms With Atrial Fibrillation Characteristics Using Generative Adversarial Networks". This dataset contains 4,768 synthesized atrial fibrillation (AF)-like ECG signals stored in PhysioNet MAT/HEA format.
Decompress the 7z file and use it like any other dataset you downloaded from PhysioNet.
(For programmers) Load this dataset with the package wfdb https://github.com/MIT-LCP/wfdb-python.