An open-access simultaneous electrocardiogram and phonocardiogram database.

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Simultaneous recording of electrocardiogram (ECG) and phonocardiogram (PCG) offers a multimodal approach to understanding cardiac electro-mechanical activity.
• Existing methods may lack the integration needed to capture subtle interdependencies between electrical and mechanical heart functions.
• The EPHNOGRAM project addresses the need for a cost-effective, low-power device for combined ECG and PCG acquisition.

Purpose of the Study:

• To develop a low-cost, low-power device for simultaneous ECG and PCG recording with noise cancellation.
• To investigate the electro-mechanical coupling of the heart during various physical activities.
• To create a comprehensive database (EPHNOGRAM) for future cardiac research.

Main Methods:

• Development and testing of hardware prototypes for simultaneous ECG-PCG acquisition.
• Collection of multimodal cardiac data from 24 healthy adults during rest and exercise (walking, running, biking).
• Development of software for data annotation and machine learning models for activity classification using joint ECG-PCG features.

Main Results:

• Significant electro-mechanical coupling observed between ECG and PCG, particularly during high-intensity exercise.
• S2-based heart rate exhibited higher volatility than ECG-based heart rate, indicating distinct electrical and mechanical dynamics.
• Hybrid ECG-PCG features, such as the R-S2 interval, proved highly informative for physical activity classification.

Conclusions:

• The EPHNOGRAM database provides a valuable resource for studying cardiac electro-mechanical interrelationships.
• The findings enhance understanding of cardiac function and have potential for improved diagnostic tools.
• The developed hardware is suitable for integration into wearable devices and multimodal stress testing.