Automated Diagnosis of Myocardial Infarction ECG Signals Using Sample Entropy in Flexible Analytic Wavelet Transform Framework
Key result
An automated diagnostic method using flexible analytic wavelet transform and an LS-SVM classifier achieved a classification accuracy of 99.31% for detecting myocardial infarction from ECG signals.
Why the study?
Does an automated method using flexible analytic wavelet transform and sample entropy accurately diagnose myocardial infarction from ECG signals?
Population
ECG signals for the diagnosis of myocardial infarction
Comparison
Automated diagnosis using flexible analytic… vs Random forest, J48 decision tree, and back…
Design
Other
Authors
Discussion
Loading...
Member takes
Implication
May assist ECG-based MI detection; leaves open prospective validation before clinical use.
Structured PICO
Does an automated method using flexible analytic wavelet transform and sample entropy accurately diagnose myocardial infarction from ECG signals?
P
Population
ECG signals for the diagnosis of myocardial infarction
E
Exposure
Automated diagnosis using flexible analytic wavelet transform (FAWT) and sample entropy with least-squares support vector machine (LS-SVM) classifier
C
Comparator
Random forest (RF), J48 decision tree, and back propagation neural network (BPNN) classifiers
O
Outcome
Classification accuracysurrogate
An automated ECG analysis method using flexible analytic wavelet transform and an LS-SVM classifier achieved 99.31% accuracy in diagnosing myocardial infarction.
Cite This Study
Kumar et al. (2017) studied Myocardial infarction (MI). Automated diagnosis using flexible analytic wavelet transform (FAWT) and LS-SVM classifier vs. Random forest (RF), J48 decision tree, and back propagation neural network (BPNN) classifiers was evaluated on Classification accuracy. An automated diagnostic method using flexible analytic wavelet transform and an LS-SVM classifier achieved a classification accuracy of 99.31% for detecting myocardial infarction from ECG signals.