HSF-IBI: A Universal Framework for Extracting Inter-Beat Interval from Heterogeneous Unobtrusive Sensors

Area of Science:

• Biomedical Engineering
• Signal Processing
• Wearable Technology

Background:

• Accurate heartbeat inter-beat interval (IBI) extraction is vital for unobtrusive vital sign monitoring.
• Current methods face challenges in precision and robustness, especially with heterogeneous sensor data.

Purpose of the Study:

• To introduce HSF-IBI, a novel and universal framework for unobtrusive IBI extraction using heterogeneous sensor fusion.
• To enhance the accuracy and robustness of IBI extraction by integrating signals from diverse sensors.

Main Methods:

• Utilized harmonic summation (HarSum) for average heart rate calculation to guide optimal band selection (OBS).
• Employed a basic sequential algorithmic scheme (BSAS)-based template group extraction and template matching (TM).
• Determined optimal IBIs by evaluating the signal quality index (SQI), ensuring morphology independence.

Main Results:

• The HSF-IBI framework successfully integrated data from heterogeneous sensors, improving detection rates by 6.25% and 5.21% on two datasets.
• Achieved superior accuracy compared to existing methods, with mean absolute errors (MAEs) of 5.25 ms and 4.56 ms.

Conclusions:

• The proposed HSF-IBI framework offers a robust and accurate solution for unobtrusive IBI extraction using heterogeneous sensor fusion.
• This approach significantly enhances vital sign monitoring capabilities by improving detection rates and reducing errors.