Key result
A cross-validation method using multiple machine learning classifiers effectively identified mislabeled ECG training samples, restoring classification accuracy to noise-free levels when label noise was 20% or less.
Why the study?
Mislabeled training samples significantly affect the classification accuracy of electrocardiogram signals, necessitating methods to identify and mitigate this issue.
Does a cross-validation filtering method improve the classification accuracy of ECG signals by identifying mislabeled training samples?
Does a cross-validation filtering method improve the classification accuracy of ECG signals by identifying mislabeled training samples?
A cross-validation filtering method improves machine learning classification accuracy of ECG signals by removing mislabeled training samples.
May improve ECG ML training robustness to label noise; leaves open prospective clinical validation before practice adoption.
The classification accuracy of electrocardiogram signal is often affected by diverse factors in which mislabeled training samples issue is one of the most influential problems. In order to mitigate this negative effect, the method of cross validation is introduced to identify the mislabeled samples. The method utilizes the cooperative advantages of different classifiers to act as a filter for the training samples. The filter removes the mislabeled training samples and retains the correctly labeled ones with the help of 10-fold cross validation. Consequently, a new training set is provided to the final classifiers to acquire higher classification accuracies. Finally, we numerically show the effectiveness of the proposed method with the MIT-BIH arrhythmia database.
No takes yet. Share an insight, caveat, or question.
Li et al. (2017) studied Arrhythmia (n=20). Cross-validation based mislabeled sample filter vs. No filtering was evaluated on Classification accuracy. A cross-validation method using multiple machine learning classifiers effectively identified mislabeled ECG training samples, restoring classification accuracy to noise-free levels when label noise was 20% or less.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: