Multiple-Axis Seismocardiography for Cardiac–Respiratory Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring cardiac and respiratory activity (CRA) are invasive, expensive, or lack accuracy, particularly in high-risk subjects or those with congenital heart defects, as they rely on indirect and inferential mechanisms like electrocardiography, echocardiography, and ballistocardiography.
Innovation Solution
A method and apparatus using multiple-axis seismocardiography and gyrocardiography to non-invasively derive biometric data by receiving and processing time-domain samples of vibrations from a vertebrate's body, separating cardiac and respiratory events, and extracting relevant biometric data through autocorrelation and frequency analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If electrocardiography, echocardiography, or ballistocardiography are used to measure cardiac and respiratory activity, then measurement capability is provided, but the methods are invasive, expensive, or lack accuracy
Solution Approach 1:
The patent replaces traditional mechanical and electrical measurement systems (ECG electrodes, echocardiography transducers, ballistocardiography sensors) with a seismocardiographic system that uses accelerometers to directly measure mechanical vibrations of the chest wall. This substitution enables non-invasive measurement while maintaining or improving accuracy by directly capturing the mechanical events of the cardiac cycle without requiring skin penetration, complex imaging, or indirect electrical inference.
2Productivity
If traditional CRA measurement methods are used, then cardiac activity can be monitored, but real-time processing and direct measurement of mechanical heart functions are not achieved
Solution Approach 1:
The patent segments the complex cardiac cycle into distinct mechanical events (atrial contraction, ventricular contraction, valve closures) that can be individually detected and measured by the seismocardiographic sensor. This segmentation allows for real-time processing of specific cardiac phases and enables direct measurement of mechanical functions such as left ventricular ejection time and heart rate variability without requiring complex post-processing or indirect inference from electrical or imaging data.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the accuracy of CRA measurements by directly measuring mechanical heart functions and respiratory rates without invasiveness, allowing real-time processing and derivation of heart rate, variability, left ventricular ejection time, and respiratory parameters.
Implementation Method 1
receiving, from a sensor positioned against the vertebrate, time-domain samples of measurements, by the sensor, of vibrations within a body of the vertebrate, in at least one time-stamped stream that corresponds to at least one associated axis of measurement of at least one of linear and rotational acceleration
Implementation Method 2
at least one associated axis of measurement of at least one of linear and rotational acceleration
Implementation Method 3
extracting the at least one datum from at least one of: the first set of streams by autocorrelating the first set of streams with a time-delayed version thereof to identify at least one characteristic peak of a cardiac cycle of the vertebrate
Data Source
AI summary
An instrument and method for non-invasively deriving at least one biometric datum of a living vertebrate. A sensor positioned against the vertebrate measures vibrations within its body. The sensor provides the samples to the instrument as at least one time-stamped stream that correspond to one or more linear and/or rotational acceleration measurement axes. The instrument organizes the samples into windowed streams, separates them into a first set of streams corresponding to physical events of the vertebrate associated with cardiovascular activity and a second set of streams corresponding to physical events of the vertebrate associated with respiration. A datum can be extracted from the first set by autocorrelating it with a time-delayed version thereof to identify at least one characteristic peak of the vertebrate's cardiac cycle. A datum can be extracted from the second set by determining a characteristic frequency of respiration of the vertebrate.