AtrialFibrillation

This is a physionet dataset of two-channel ECG recordings has been created from data used in the Computers in Cardiology Challenge 2004, an open competition with the goal of developing automated methods for predicting spontaneous termination of atrial fibrillation (AF).

The raw instances were 5 second segments of atrial fibrillation, containing two ECG signals, each sampled at 128 samples per second. The class labels are: n, s and t.

class n is described as a non termination artiral fibrilation(that is, it did not terminate for at least one hour after the original recording of the data).

class s is described as an atrial fibrilation that self terminates at least one minute after the recording process.

class t is described as terminating immediately, that is within one second of the recording ending.

This is a pre-processed version of the dataset saved in numpy format. The original dataset is obtained from here.

The data are 3-dimensional arrays of shape [n_samples, time_steps, n_variables]. The data can be loaded as follows:

loaded_data = np.load("AF.npz")
Xtr = loaded_data['Xtr'] # Training data of shape (4823, 45, 2)
Ytr = loaded_data['Ytr'] # Training labels of shape (4823, 1)
Xte = loaded_data['Xte'] # Test data of shape (185, 45, 2)
Yte = loaded_data['Yte'] # Test labels of shape (185, 1)