Key result
Deep learning and traditional classifiers evaluated for arrhythmia prediction accuracy on standard ECGs.
Why the study?
Arrhythmia identification relies on ECG analysis, but performance varies between traditional classifiers requiring pre-filtering and newer classifiers that automatically extract attributes.
Do latest machine learning classifiers (DNN, CNN) improve classification accuracy compared to traditional classifiers (LR, DT, SVC) for cardiovascular arrhythmia prediction using ECG datasets?
Do latest machine learning classifiers (DNN, CNN) improve classification accuracy compared to traditional classifiers (LR, DT, SVC) for cardiovascular arrhythmia prediction using ECG datasets?
This study evaluates the classification accuracy of traditional versus deep learning models for detecting cardiac arrhythmias from ECG signals.
Evaluation of DL vs traditional ML for ECG arrhythmia classification remains hypothesis-generating; leaves open whether DNN/CNN improve accuracy in clinical practice.
Cardiac arrhythmia are aggravations in the pattern of heart, brought about by anomaly or strange quick or slow pulses. The scope of cardiovascular arrhythmias incorporates conditions that are promptly hazardous and guiltless extrasystoles. The exhaustive assessment of an ECG of the greatest conceivable quality gathered during the arrhythmia is fundamental for the precise recognizable proof of heart arrhythmias. In past few years the exploration has been seen from feature grouping approach to deal with classifier approaches. The conduct of traditional classifiers such LR (Logistic Regression), DT (Decision Tree), SVC (Support Vector Classifier) etc. are more depended on pre-filtering and attribute extraction modules, while the performance of latest classifiers like DNN (Deep Neural Networks), CNN (Conventional Neural Networks) can create attributes that includes consequently from ECG signal. This paper concentrates on evaluating classification accuracy analysis of traditional and latest approaches using the PTB Diagnostic ECG dataset and the MIT-BIH Arrhythmia dataset.
No takes yet. Share an insight, caveat, or question.
Keerthana et al. (2023) studied Cardiovascular Arrhythmia. Deep learning models (DNN, CNN) vs. Traditional classifiers (LR, DT, SVC) was evaluated on Classification accuracy. Deep learning models (DNN, CNN) and traditional classifiers (LR, DT, SVC) were evaluated for their classification accuracy in predicting cardiovascular arrhythmias using standard ECG datasets.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: