Loading [a11y]/accessibility-menu.js
Automatic Classification of 12-lead ECG Based on Model Fusion | IEEE Conference Publication | IEEE Xplore

Automatic Classification of 12-lead ECG Based on Model Fusion


Abstract:

Aiming at the growing demand for automatic analysis of standard 12-lead electrocardiogram (ECG) in the medical diagnosis process, an automatic classification of 12-lead E...Show More

Abstract:

Aiming at the growing demand for automatic analysis of standard 12-lead electrocardiogram (ECG) in the medical diagnosis process, an automatic classification of 12-lead ECG based on model fusion is proposed. The algorithm extracts local features by Convolutional Neural Networks (CNN), and then extracts temporal features by Bi-directional Long Short-term Memory (BiLSTM). Finally, eXtreme Gradient Boosting (XGBoost) is used to fuse the 12-lead models to get the classification results. The experiment result shows that, in classifying 9 types of ECG singles, fusion model achieved a mean accuracy of 0.964, a micro-average area under the curve (AUC) of 0.908, an average F1 score of 0.812, a mean precision of 0.836, and a mean recall of 0.788. We demonstrated the feasibility and effectiveness of the fusion model based on XGBoost for interpreting 9 common heart rhythms according to 12-lead ECG. The findings may have clinical relevance for the early diagnosis of cardiac-rhythm disorders.
Date of Conference: 17-19 October 2020
Date Added to IEEE Xplore: 25 November 2020
ISBN Information:
Conference Location: Chengdu, China

References

References is not available for this document.