ECG feature extraction and disease diagnosis.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiology

Background:

• Electrocardiogram (ECG) interpretation is complex due to physiological, technical, and pathophysiological influences on waveforms.
• Accurate feature extraction is crucial for reliable clinical decision-making from ECG data.

Purpose of the Study:

• To develop and validate a method for ECG feature extraction and heart disease diagnosis.
• To utilize wavelet transform (WT) and LabVIEW for enhanced signal processing and analysis.

Main Methods:

• ECG signal denoising using LabVIEW signal processing tools.
• R-peak detection employing Haar wavelet decomposition and thresholding.
• P-wave and T-wave detection using Daubechies (db6) wavelet on low-resolution signals.
• Extraction of diagnostic features including heart rate, wave widths, amplitudes, durations, ST segment, and frontal plane axis.
• Application of a scoring pattern for heart disease diagnosis.

Main Results:

• Successful R-peak, P-wave, and T-wave detection using specified wavelet techniques.
• Extraction of multiple quantitative ECG parameters for diagnostic assessment.
• Consideration of specific heart conditions: tachycardia, bradycardia, left ventricular hypertrophy, right ventricular hypertrophy, and myocardial infarction.

Conclusions:

• The proposed WT and LabVIEW-based method effectively extracts crucial ECG features for heart disease diagnosis.
• This approach offers a robust tool for analyzing ECG signals, aiding in the identification of various cardiac conditions.