-- A 3-Classes Dataset:

• The normal sinus rhythm (NSR) class: MIT-BIH database

• The arrhythmia (ARR) class: MIT-BIH database

• The congestive heart failure (CHF) class: BIDMC database

• Each class has 3930 vectors (extracted from 30 recordings)

• A total of 11,790 feature vectors extracted out of the 90 recordings.

• Each row in the file represents one sample of length 500.

• The last value in each row (element 501) represents the respective calss (0, 1, or 2)

Check and cite the paper: https://www.mdpi.com/2682016
Eleyan, A.; Alboghbaish, E. Electrocardiogram Signals Classification Using Deep-Learning-Based Incorporated Convolutional Neural Network and Long Short-Term Memory Framework. Computers 2024, 13, 55. https://doi.org/10.3390/computers13020055