Automated ECG Signals Analysis for Cardiac Abnormality Detection and Classification

The electrocardiogram (ECG) is a critical, non-invasive tool for diagnosing cardiovascular diseases, offering insights into heart function. However, analyzing extended ECG data can be complex, requiring advanced computerized systems for effective diagnosis and classification. These systems must detect arrhythmias, manage data noise, and adapt to individual waveform variations, while ensuring model robustness across different populations and settings. The main goal of this study is to develop an ECG signal processing system that can accurately detect and classify various cardiac conditions. We propose a novel hybrid approach, classifying ECG signals into categories such as normal, left bundle branch block (LBBB), paced beat, right bundle branch block (RBBB), and supraventricular contraction (SVC) using a PhysioNet database. By applying discrete wavelet transform (DWT) and principal component analysis (PCA), we extracted six relevant features from each ECG category. These features were analyzed using an adaptive neuro-fuzzy inference system (ANFIS) classifier, achieving an overall classification accuracy of 99.44%, with average sensitivity and specificity of 99.36% and 99.84%, respectively. This system shows significant promise in enhancing the accuracy and efficiency of diagnosing cardiovascular diseases through ECG analysis.