Key result
A deep learning-based convolutional neural network model detected unacceptable electrocardiograms with an area under the receiver operating characteristic curve of 0.93 and an F1-score of 0.80.
Why the study?
Noise in biosignal data captured by patient monitoring systems hinders their use for detecting or predicting critical clinical events, prompting the evaluation of deep learning algorithms to screen unacceptable, noisy ECGs.
Effect estimate: AUROC 0.93
A deep learning-based CNN model can efficiently and accurately detect and screen out unacceptable noisy ECG signals, facilitating automated ECG analysis.
No takes yet. Share an insight, caveat, or question.
May support automated ECG preprocessing; leaves open prospective validation before clinical use.
Yoon et al. (2019) studied Electrocardiogram noise (n=3,000). Deep learning-based convolutional neural network model vs. Medical expert interpretation was evaluated on Detection of unacceptable ECGs (AUROC) (AUROC 0.93). A deep learning-based convolutional neural network model detected unacceptable electrocardiograms with an area under the receiver operating characteristic curve of 0.93 and an F1-score of 0.80.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: