Posture dependent time series filtering system and methods for the detection of cardiac events
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-9414757-B2
Publication Date
2016-08-16
Expiration Date
Abstract
A device for detecting a cardiac event is disclosed. Detection of an event is based on a test applied to a parameter whose value varies according to heart rate. Both the parameter value and heart rate (RR interval) values are filtered with an exponential average filter. The filtered parameter value and heart rate values are stored as separate datasets corresponding to particular body postures. Upper and lower boundary values and detection thresholds of the parameter are computed for each of the datasets. The test to detect the cardiac event depends on the heart rate and the difference between the parameter's value and a corresponding detection threshold.
Core Innovation
The invention is a cardiac event detection system that senses an analog signal from a patient's heart using at least two sensors and digitizes the electrical signal to produce a digitized waveform of a heart rate over a plurality of heart beats. The heart rate defines a first parameter, and a second parameter value is computed from the digitized waveform over the plurality of heart beats. For each of a plurality of body positions, the system computes heart rate time series data and time series data for the second parameter value, thereby generating posture-specific datasets.
For each posture-specific dataset, the system reduces the dispersion of the second parameter value data as a function of heart rate, thereby generating reduced dispersion datasets. From each reduced dispersion dataset, the system generates upper and lower normal boundaries of the value of the second parameter as a function of heart rate, and then generates upper and lower detection thresholds from the upper and lower normal boundaries. The resulting thresholds are associated with heart rate and posture-dependent normal behavior.
The system performs a cardiac event detection test by comparing a value of the second parameter of at least one beat with its associated posture dependent upper or lower detection threshold. The described filtering and boundary generation exclude data associated with rapid heart-rate recovery, ectopic/irregular beats, and transitory events, and use filtered long-/short-term changes of the parameter trajectories relative to the heart-rate dependent normal bands. In the document, the second parameter includes ST segment deviation and related ECG features.
Claims Coverage
The document provides one independent claim directed to posture- and heart-rate-dependent cardiac event detection using digitized analog heart signals, with inventive features centered on generating posture-dependent datasets, reducing dispersion as a function of heart rate, deriving heart-rate-dependent normal boundaries and detection thresholds, and triggering detection by comparing at least one beat’s second-parameter value to the corresponding threshold.
Digitizing analog heart signals for heart rate waveform and second parameter
At least two sensors sense an analog signal from the patient’s heart; an analog-to-digital circuit system digitizes the electrical signal to produce a digitized waveform of a heart rate over a plurality of heart beats, with the heart rate defining a first parameter, and processor computation generates a second parameter value time series from the digitized waveform over the plurality of heart beats.
Generating posture-dependent datasets from heart rate and second parameter time series
For each of a plurality of body positions, the processor computes heart rate time series data and time series data for a second parameter value from the digitized waveform over the plurality of heart beats, generating a plurality of datasets for each of the plurality of postures.
Reducing dispersion of the second parameter as a function of heart rate
For each dataset, the processor reduces the dispersion of the second parameter value data as a function of heart rate, thereby generating a plurality of reduced dispersion datasets.
Deriving heart-rate dependent normal boundaries and detection thresholds for the second parameter
For each reduced dispersion dataset, the processor generates upper and lower normal boundaries of the value of the second parameter as a function of heart rate, and generates upper and lower detection thresholds from the upper and lower normal boundaries.
Comparing at least one beat’s second parameter to posture dependent thresholds for detection
The processor performs a cardiac event detection test by comparing a value of the second parameter of at least one beat with its associated posture dependent upper or lower detection threshold.
Across the independent claim, the core coverage is the pipeline from digitizing analog heart signals, producing posture-dependent heart rate and second-parameter datasets, reducing dispersion of the second parameter as a function of heart rate, creating heart-rate-dependent normal boundaries and thresholds, and detecting cardiac events by threshold comparison for at least one beat.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?