Dataset Details

This is a randomly split five fold dataset of the MIMIC-IV-ECG varaint of ECG-QA stratified based on patients (i.e., zero patient overlap between training and test). The name of the dataset repository is ecg-qa-mimic-iv-ecg-250-500 where 250 refers to the sampling frequency and 500 denotes 2 seconds. The code to do the splitting is here. The dataset contents are under the same license as the original ECG-QA's license.

Any questions or issues, please do not hesitate to reach out to the maintainer of ECG-Bench.