Seismocardiography-based estimation of hemodynamic parameters during submaximal ergometer test
Executive Summary
This study evaluates the feasibility of using seismocardiography (SCG) to estimate hemodynamic parameters such as stroke volume (SV), heart rate (HR), cardiac output (CO), and cardiac index (CI) during submaximal ergometer exercise across different BMI groups. Using a chest-mounted accelerometer and random forest regression models, the study achieved mean absolute percentage errors (MAPEs) below 8% for all parameters, with the highest reliability observed during post-exercise recovery (median R2 = 0.75 for HR and CO). The findings highlight SCG's potential as a lightweight, non-invasive tool for cardiovascular monitoring in low-motion scenarios, though motion artifacts limit its accuracy during active exercise.
Answer Machine Insights
Q: What was the best-performing phase for SCG-based hemodynamic estimation?
The post-exercise recovery phase provided the most reliable estimates.
Post condition demonstrated significantly higher R2 values compared to all other conditions (p < 0.001 for all parameters).
Q: What were the main limitations of SCG during active exercise?
Motion artifacts and physiological variability reduced estimation reliability during active cycling.
SCG acquisition is technically viable during exercise, but reliable hemodynamic estimation under high-motion conditions remains limited due to motion artifacts and physiological variability.
Key Results
MAPEs below 8% for all hemodynamic parameters across conditions.
Post-exercise recovery phase achieved the highest reliability with median R2 = 0.75 for HR and CO.
Clinical Snapshot
Evidence Rating
Relevance
high Priority