Key result
An ECG beat classification method using multi-kernel ResNet with a transformer encoder achieved an average F1 score of 0.839 for four-class beat classification on the MIT-BIH arrhythmia database.
Why the study?
Diagnosing more complicated cardiovascular disease relies on analyzing beat-unit abnormalities, where beat classification plays a fundamental role in classifying arrhythmia on ECG.
A novel deep learning approach using multi-kernel ResNet and Transformer encoders improves ECG beat classification for arrhythmia detection.
In this work, we investigated the electrocardiogram (ECG) for cardiovascular disease diagnosis using the deep learning method. The diagnosis of a more complicated cardiovascular disease can be performed by analyzing beat-unit abnormalities. Beat classification plays the basic role of classifying arrhythmia in ECG. To classify the four beat classes of the MIT-BIH arrhythmia data base, we extracted features using the multi-kernel ResNet, trained the system to successfully learn the mutual information of continuous beats using the transformer encoder, and enabled the transformer to learn stably by applying position embedding based on the heart rate. We used the icentiallk data set, which is a big public data set, for the transfer-learning and could make a more general model with a variety of patient data. We obtained an average F1 score of 0.839, which is better than any existing score, for the four-class beat classification of the MIT-BIH arrhythmia database.
No takes yet. Share an insight, caveat, or question.
Chon et al. (2023) studied Arrhythmia. Multi-kernel ResNet with Transformer encoder vs. Existing methods was evaluated on Four-class beat classification (average F1 score). An ECG beat classification method using multi-kernel ResNet with a transformer encoder achieved an average F1 score of 0.839 for four-class beat classification on the MIT-BIH arrhythmia database.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: