Algorithm for heart rate extraction in a novel wearable acoustic sensor.

Area of Science:

• Biomedical Engineering
• Cardiovascular Monitoring
• Wearable Technology

Background:

• Phonocardiography is essential for detecting cardiac abnormalities by analyzing heart sounds (S1 and S2).
• Traditional methods use chest-placed acoustic sensors, limiting continuous monitoring applications.
• Extracting heart rate from heart sounds is crucial for cardiovascular health assessment.

Purpose of the Study:

• To develop and validate a novel algorithm for detecting S1 and S2 heart sounds from neck-acquired signals.
• To utilize these detected sounds for accurate heart rate extraction.
• To assess the feasibility of using a single sensor for simultaneous breathing and heart rate monitoring.

Main Methods:

• A new algorithm was developed to detect S1 and S2 heart sounds from acoustic signals acquired by a neck-placed sensor.
• Heart rate was extracted using the detected S1 and S2 sounds.
• The algorithm's performance was evaluated against two commercial devices using over 38 hours of data from ten subjects during sleep.

Main Results:

• The algorithm achieved high accuracy in detecting S1 and S2 heart sounds and extracting heart rate.
• Accuracy rates of 90.73% and 90.69% were recorded for heart rate extraction compared to commercial devices.
• This study utilized the largest dataset to date for acoustic heart sound classification and heart rate extraction.

Conclusions:

• A novel algorithm enables accurate heart rate monitoring using neck-placed sensors, detecting S1 and S2 heart sounds.
• The findings demonstrate the potential for a single wearable sensor to monitor both breathing and heart rate.
• This technology offers significant utility for long-term, non-invasive vital signs monitoring.