PhysioNet/CinC Challenge 2017: AF classification from a short single-lead ECG (AliveCor)
The PhysioNet/Computing in Cardiology Challenge 2017 dataset contains single-lead ECGs recorded by patients on AliveCor handheld devices. The public training set has 8,528 recordings lasting 9 to just over 60 seconds, sampled at 300 Hz, each hand-labelled as normal, atrial fibrillation, other rhythm or too noisy. A further 3,658 recordings form a private test set.
- Recordings: 8,528 training (+3,658 private test)
- Device: AliveCor handheld, single lead
- Sampling: 300 Hz, 9-60 s
- Labels: normal, AF, other rhythm, noisy
- Hosted: PhysioNet (open access)
- Type
- Dataset
- Host
- Not reported
- Licence
- Not reported
- Access
- Not reported
- Subjects
- Not reported
- Signals
- ECG
- PPG sites
- None
- Setting
- Not reported
- Wrist
- Not reported
- Raw waveforms
- Not reported
- Updated
- Oct 16, 2026
Original description
The 2017 PhysioNet/CinC Challenge aims to encourage the development of algorithms to classify, from a single short ECG lead recording (between 30 s and 60 s in length), whether the recording shows normal sinus rhythm, atrial fibrillation (AF), an alternative rhythm, or is too noisy to be classified.
Verbatim fromPhysioNet· Licence: Open Data Commons Attribution License v1.0
How it works
No model in the hub reports results on this dataset yet.
Models trained or evaluated on it
Cite
@misc{challenge20172026,
title = {{PhysioNet/CinC Challenge 2017: AF classification from a short single-lead ECG (AliveCor)}},
year = {2026},
doi = {10.13026/d3hm-sf11},
url = {https://physionet.org/content/challenge-2017/1.0.0/}
}History
Record updated