Robust off-line heartbeat detection using ECG and pressure-signals

Area of Science:

• Biomedical Engineering
• Signal Processing
• Cardiovascular Monitoring

Background:

• Artefacts in electrocardiogram (ECG) and pressure signals often stem from diverse sources.
• Analyzing both ECG and pressure signals together can improve heartbeat detection efficacy over ECG-only analysis.

Purpose of the Study:

• To present a novel algorithm for robust heartbeat annotation by integrating pressure and ECG signal analysis.
• To introduce a new method for pressure-signal-based heartbeat annotation.
• To develop a mechanism for identifying and correcting paced heart rhythms.
• To implement a noise detection approach for signal quality assessment.

Main Methods:

• Combined analysis of pressure and ECG signals for heartbeat annotation.
• Development of a novel pressure-signal-based annotation approach.
• Implementation of a mechanism to detect and correct paced rhythms.
• Integration of a noise detection strategy.

Main Results:

• The algorithm achieved an overall score of 87.31% on Physionet CINC-2014 challenge datasets.
• Demonstrated improved heartbeat detection effectiveness through combined signal analysis.
• Successfully identified and corrected paced rhythms.
• Incorporated noise detection to enhance signal reliability.

Conclusions:

• Combined analysis of pressure and ECG signals offers superior heartbeat detection.
• The developed algorithm shows significant potential for clinical application in cardiovascular monitoring.
• Further recommendations are proposed for future algorithm enhancement.