Seismocardiography: Interpretation and Clinical Application
Executive Summary
This dissertation explores the physiological underpinnings and clinical applications of seismocardiography (SCG). Study I correlates SCG fiducial points with echocardiographic events in 45 healthy subjects, establishing a robust framework for SCG interpretation. Study II demonstrates SCG's potential in assessing cardiac resynchronization therapy (CRT) in heart failure patients, showing significant changes in cardiac timing intervals and amplitudes with CRT activation. Study III develops a novel non-exercise VO2max estimation model incorporating SCG features, achieving high correlation (R=0.90) with measured VO2max, suggesting clinical utility in cardiorespiratory fitness assessment.
Answer Machine Insights
Q: What physiological events were correlated with SCG fiducial points?
Eight events, including mitral valve closure, aortic valve opening, and aortic valve closure, were correlated with SCG fiducial points.
In the SCG, we identified eight common fiducial points with a short temporal difference and high correlation to eight physiologic events found in echo images.
Q: How does CRT affect SCG signals?
CRT significantly shortens IVCT and increases systolic amplitudes, indicating improved cardiac contractility.
The IVCT was significantly shorter with the bi-ventricular pacemaker turned on compared to off (52.6 ± 14.4 ms compared to 63.2 ± 20.9 ms, p = 0.027).
Q: How accurate is the SCG-based VO2max estimation model?
The model achieved a correlation of 0.90 with measured VO2max and a standard error of 3.18 mL·kg−1·min−1.
Using this feature together with demographic data in a linear prediction model for VO2max, yielded a correlation of 0.90 (95% CI: 0.83-0.94) with a SEE of 3.18 mL·kg−1·min−1.
Key Results
Eight SCG fiducial points were correlated with echocardiographic events, achieving high temporal alignment.
CRT activation reduced iso-volumetric contraction time (IVCT) by ~10 ms and increased systolic amplitude by ~4.1 mg.
A non-exercise VO2max estimation model using SCG achieved R=0.90 with a standard error of 3.18 mL·kg−1·min−1.
Visual Evidence
![Fig. 2.4: Electrocardiogram (top), phonocardiogram (middle) and seismocardiogram (bottom) divided into eight periods: Atrial systole, protosystole, iso-volumetric contraction, maximum ejection, reduced ejection, iso-volumetric relaxation, rapid inflow and diastasis. The slopes are named according to the period in which the reside. ©Reprint with permission from BMJ, [19]](/_next/image?url=%2Fscience-hub%2Fimages%2F10.54337_aau451017905%2F10.54337_aau451017905_p34_i1.png&w=3840&q=75)
Fig. 2.4: Electrocardiogram (top), phonocardiogram (middle) and seismocardiogram (bottom) divided into eight periods: Atrial systole, protosystole, iso-volumetric contraction, maximum ejection, reduced ejection, iso-volumetric relaxation, rapid inflow and diastasis. The slopes are named according to the period in which the reside. ©Reprint with permission from BMJ, [19]
Clinical Snapshot
Evidence Rating
Relevance
high Priority