Electrocardiography (ECG) analysis and a new feature extraction method using wavelet transform with scalogram...

Area of Science:

• Biomedical Engineering
• Medical Informatics
• Cardiology

Background:

• Electrocardiography (ECG) signal analysis is crucial for diagnosing cardiovascular diseases.
• Accurate ECG interpretation is vital for effective patient management.
• Existing analysis methods can be time-consuming and may lack advanced feature extraction.

Purpose of the Study:

• To develop a user-friendly ECG analysis toolbox with a graphical interface.
• To introduce a novel feature calculation methodology for ECG analysis.
• To improve the accuracy, reliability, and efficiency of ECG interpretation.

Main Methods:

• Development of an integrated ECG analysis toolbox covering all steps from recording to statistical analysis.
• Implementation of a new feature calculation methodology beyond traditional amplitude and duration measurements.
• Validation of the toolbox and new features using the MIT-BIH Arrhythmia ECG Database and an experimental dataset.

Main Results:

• The developed ECG analysis toolbox significantly improves the accuracy and reliability of ECG main wave detection.
• The toolbox substantially reduces analysis time compared to manual methods.
• The proposed new feature set provides distinct information valuable for artificial intelligence-based decision support systems.

Conclusions:

• The developed ECG analysis toolbox offers a comprehensive and efficient solution for ECG analysis.
• The novel feature calculation methodology enhances the diagnostic potential of ECG signals, particularly for AI applications.
• This work contributes to advancing automated cardiovascular disease diagnosis through improved ECG analysis tools and features.