Key result
Transformer network detecting heart rate from BCG signals achieves 0.97 correlation with ECG.
Why the study?
Longitudinal monitoring of heart rate and heart rate variability aids in tracking cardiovascular diseases and autonomic activity, motivating methods to detect precise heartbeat timing from non-invasive ballistocardiogram signals.
Does a transformer network accurately detect heart rate and mean heart beat interval from ballistocardiogram signals compared to ground truth ECG in healthy and older adults?
Does a transformer network accurately detect heart rate and mean heart beat interval from ballistocardiogram signals compared to ground truth ECG in healthy and older adults?
Effect estimate: Correlation coefficient 0.97
A transformer network can accurately derive heart rate and heart rate variability metrics from non-invasive ballistocardiogram signals, offering potential for long-term home monitoring.
Longitudinal monitoring of heart rate (HR) and heart rate variability (HRV) can aid in tracking cardiovascular diseases (CVDs), sleep quality, and sleep disorders, and reflect autonomic nervous system activity, stress levels, and overall well-being. These metrics are valuable in both clinical and everyday settings. In this paper, we present a transformer network aimed primarily at detecting the precise timing of heart beats from predicted electrocardiogram (ECG) signals that are derived from input Ballistocardiogram (BCG) signals. We compared the performance of segment-based and subject-based models across three datasets: a dataset with 46 young, healthy subjects, a dataset with 28 older adults, and a combined dataset. The segment-based model demonstrated superior performance, with correlation coefficients of 0.97 for both, HR and mean heart beat interval (MHBI) when compared to ground truth ECG. This non-invasive method offers significant potential for long-term, in home HR and HRV monitoring, aiding in the early indication and prevention of cardiovascular issues.
No takes yet. Share an insight, caveat, or question.
Yi et al. (2025) studied Healthy subjects and older adults (n=74). Transformer network for heartbeat detection from BCG signals vs. Ground truth ECG was evaluated on Correlation coefficient for HR and mean heart beat interval (MHBI) (Correlation coefficient 0.97). A segment-based transformer network for detecting heartbeats from ballistocardiogram signals achieved a correlation coefficient of 0.97 for heart rate and mean heart beat interval compared to ECG.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: