Key result
Hybrid CNN-LSTM-Attention network achieves ~99.8% accuracy for arrhythmia classification.
Why the study?
Traditional ECG algorithms based on morphological features suffer from problems of feature extraction and high complexity.
Does a hybrid CNN-LSTM-Attention model accurately classify arrhythmias from ECG data?
Does a hybrid CNN-LSTM-Attention model accurately classify arrhythmias from ECG data?
A novel hybrid CNN-LSTM-Attention model demonstrates high accuracy, sensitivity, and specificity for automatic arrhythmia classification using ECG data.
May aid automated ECG arrhythmia detection; leaves open prospective clinical validation before practice change.
The electrocardiogram (ECG) is a key standard for monitoring the activity regularity of heart. Many cardiac abnormalities will be demonstrated from ECG including arrhythmia, which may be a life-threatening disease. Automatic classification of arrhythmias is one of the most valuable topics in medicine and bioinformatics. To solve the problems of feature extraction and high complexity in traditional Electrocardiograph algorithms based on morphological features, a new hybrid network combining Convolutional Neural Netork and Long Short-term Memory and Attention mechanism was proposed in this paper. And, we used three indicators set by The Association for the Advancement of Medical Instrumentation to evaluate model proposed in this paper. By training and verifying the arrhythmia data from the MIT-BIH database, the mean accuracy rate reached 99.81%, the mean sensitivity reached 99.36%, the mean specificity reached 99.78%. Therefore, we can conclude that the model proposed in this paper can effectively help clinicians detect common arrhythmia on conventional screening Electroencephalograph.
No takes yet. Share an insight, caveat, or question.
Tao Zhang (2021) studied Arrhythmia. Hybrid network combining CNN, LSTM, and Attention mechanism was evaluated on Mean accuracy rate. A hybrid network combining CNN, LSTM, and Attention mechanism achieved a mean accuracy of 99.81%, sensitivity of 99.36%, and specificity of 99.78% for arrhythmia classification.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: