AF Classification from a Short Single Lead ECG Recording: the PhysioNet/Computing in Cardiology Challenge 2017
By G. Clifford, Chengyu Liu, Benjamin Moody, Li-wei H. Lehman, Ikaro Silva, Qiao Li, Alistair E. W. Johnson, R. Mark. 2017 Computing in Cardiology (CinC), 2017
847 citations recorded by the source corpus.
The PhysioNet/Computing in Cardiology (CinC) Challenge 2017 focused on differentiating AF from noise, normal or other rhythms in short term (from 9–61 s) ECG recordings performed by patients. A total of 12,186 ECGs were used: 8,528 in the public training set and 3,658 in the private hidden test set. Due to the high degree of inter-expert disagreement between a significant fraction of the expert labels we implemented a mid-competition bootstrap approach to expert relabeling of the data, levering the best performing Challenge entrants' algorithms to identify contentious labels. A total of 75 independent teams entered the Challenge using a variety of traditional and novel methods, ranging from random forests to a deep learning approach applied to the raw data in the spectral domain. Four teams won the Challenge with an equal high F1 score (averaged across all classes) of 0.83, although the top 11 algorithms scored within 2% of this. A combination of 45 algorithms identified using LASSO achieved an F1 of 0.87, indicating that a voting approach can boost performance.
Subfields: Cardiology and Cardiovascular Medicine, Cognitive Neuroscience
View Paper: https://www.semanticscholar.org/paper/1896bab29091826644543ffc038d1633282eabcc
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Cardiology and Cardiovascular Medicine
- Awni Y. Hannun
- Matin Hashemi
- Pranav Rajpurkar
- Theodoros Tsiligkaridis
- Marinka Zitnik
- Codie Bourn
- Mohammadhosein Oveisi
- Andrew Y. Ng
In Cognitive Neuroscience
- Matin Hashemi
- Nils Strodthoff
- Wojciech Samek
- Marinka Zitnik
- Mohammadhosein Oveisi
- Andrew Y. Ng
- Saeed Saadatnejad
- Pieter Abbeel
Citations in the corpus
- Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency (2022, background): “For example, in medical applications, labeling data at scale is often infeasible or can be expensive and noisy (experts can disagree on ground-truth labeling [26, 27], e.”
- CLOCS: Contrastive Learning of Cardiac Signals (2020, neutral): “PhysioNet 2017 (Clifford et al., 2017).”
- Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks (2017, background): “A recently released dataset captured from the AliveCor ECG monitor contains about 7000 records (Clifford et al., 2017).”
- PTB-XL, a large publicly available electrocardiography dataset (2020, background): “The only exceptions in terms of freely accessible datasets with larger samples sizes are the AF classification dataset [14] and the Chinese ICBEB Challenge 2018 dataset [15], which contain, however, either just single-lead ECGs or cover only a very limited set of ECG statements.”
- LSTM-Based ECG Classification for Continuous Monitoring on Personal Wearable Devices (2018, methodology): “Deep learning has also been applied to other problems related to analysis of ECG signals, for instance, ECG-based biometrics [32], [33], detecting atrial fibrillation (AF) [34]–[40] which is normally based on CinC Challenge 2017 dataset [41], and diagnosis based on hospital records [42].”