ECG Signal Feature Extraction and Waveform Representation

Abstract

Electrocardiogram (ECG) data processing has become a fascinating study. The essential elements covered in data processing are signal analysis, feature extraction, and classification to detect abnormalities in a patient's ECG signal. Many researchers have described the automated ECG classification system with different ECG characteristics and feature extraction to represent normal ECG & arrhythmia signals. Feature extraction from an electrocardiogram is a necessary procedure for detecting cardiovascular disorders. Furthermore, identifying a group of essential elements that can obtain high accuracy is the main stage in ECG classification. Due to the difficulty in interpreting ECG signals, researchers decided to investigate the automatic detection of cardiac arrhythmia problems. Using data processing techniques and specific computer software, researchers quickly understood complex ECG patterns and predicted the presence or absence of cardiac arrhythmia. Furthermore, finding a group of essential features that can achieve the highest accuracy is the primary stage in ECG classification. This paper analyzes the PTB ECG diagnostic database and ECG parameters analysis extracted & classifications.