ECG Signal Based Arrhythmia Detection System using Optimized Hybrid Classifier

An essential diagnostic tool in identifying heart rhythm irregularities, known as arrhythmias, is the ECG (Electrocardiogram). Accurate identification of arrhythmias in clinical environments is critical to patient well-being, as both acute and chronic heart conditions are typically reflected in these measurements. This is known to be a severe problem even for human experts, due to variability between individuals and inevitable noise. In this research, we have proposed an effective ECG arrhythmia classification method using a hybrid classifier with SVM (Support vector machine) and ANN (Artificial neural network) which recently shows outstanding performance in the field of pattern recognition. Every ECG beat was transformed into two-dimensional data as input data for the hybrid classifier. Optimization of the proposed hybrid classifier includes various optimization techniques such as GA (Genetic algorithm) and CS (Cuckoo search) algorithm with an optimal objective function. Also, we have compared our proposed hybrid classifier with wellknown optimized ANN based ECG arrhythmia classification models. ECG recordings from the MIT-BIH arrhythmia database are used for the evaluation of the classifier. To precisely validate the hybrid classifier, cross-validation was performed at the evaluation, which involves every ECG recording as a test data with GA and with CS. The experimental results have successfully validated that the proposed hybrid classifier with the GA and CS has achieved excellent classification accuracy without any requirement of manual pre-processing of the ECG signals such as noise filtering, feature extraction, and feature reduction.