Key result
The AMGLVQ algorithm improved ECG classification accuracy compared to SVM or back-propagation neural networks while handling imbalanced datasets and detecting unknown data.
The proposed AMGLVQ algorithm effectively integrates feature extraction and classification, improving accuracy for ECG-based biometrics and arrhythmia classification, particularly in imbalanced datasets.
Electrocardiogram (ECG) signal for human identity recognition is a new area on biometrics research. The ECG is a vital signal of human body, unique, robustness to attack, universality and permanence, difference to others traditional biometrics technic. This study also proposes Adaptive Multilayer Generalized Learning Vector Quantization (AMGLVQ), that integrating feature extraction and classification method. The experiments shown that AMGLVQ can improve the accuracy of classification better than SVM or back-propagation NN and also able to handle some problems of heartbeat classification: imbalanced data set, inconsistency between feature extraction and classification and detecting unknown data on testing phase.
No takes yet. Share an insight, caveat, or question.
Imah et al. (2013) studied Arrhythmia and normal sinus rhythm (ECG biometrics) (n=10). Adaptive Multilayer Generalized Learning Vector Quantization (AMGLVQ) vs. SVM and back-propagation NN was evaluated on Classification accuracy. The AMGLVQ algorithm improved ECG classification accuracy compared to SVM or back-propagation neural networks while handling imbalanced datasets and detecting unknown data.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: