Acute ischemia detection based on parameter value range estimation
Inventors
Hopenfeld, Bruce • Johnson, Steven R.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-9375151-B1
Publication Date
2016-06-28
Expiration Date
Abstract
A method for detecting acute ischemia is disclosed. The method comprises the steps of determining ST segment deviations and storing the results in heart rate based histograms. The histogram data is periodically analyzed to estimate a true range of ST deviation for a particular heart rate range. A current threshold for a heart rate range is set by estimating the true range of ST deviations within that bin, determining the median value of ST deviations within the range, and basing thresholds on the range and median value. The true range for a particular heart rate bin is estimated by locating ST deviation values between which reside a predetermined percentage of all ST deviation values collected for that bin.
Core Innovation
The invention relates to detecting acute cardiac ischemia by analyzing a digitized waveform signal derived from sensors, including computing a plurality of ST deviations for singular heartbeats from an electrocardiogram (ECG) over an initial time period. The method stores data pertaining to the plurality of ST deviations and uses heart-rate-binned statistics to model true normal ST deviation behavior per heart-rate bin. The approach supports heart-rate based histograms and optionally ST/RR and ST/QR histograms, using the collected ST deviations within bins for subsequent threshold estimation.
For each heart-rate bin and/or RR bin, the method periodically estimates an upper and a lower normal ST deviation boundary that contain a reduced set of ST deviations between boundaries within a predetermined cumulative percentage range of collected ST deviations. The boundaries are selected using boundary bins such that a predetermined cumulative percentage of ST deviations lies between the boundaries, yielding a reduced set relative to all ST deviations computed in the initial time interval. A dispersion measure is computed based on the reduced set and both boundaries, and the median ST deviation is derived as the center statistic of the estimated range.
After the initial time period, the invention applies a test to detect cardiac event or ischemia based on dispersion and tail parameters relative to the median, producing positive and negative detection thresholds that are heart-rate dependent. The thresholds are formed as functions of dispersion and tail measures and include safety floors and caps, and alarm detection is triggered when the detection thresholds are exceeded persistently. Additional robustness is described, including beat qualification to prevent noisy or abnormal segments from updating histograms and threshold estimates, and ST deviation correction for QRS amplitude.
Claims Coverage
The independent claim is clm-00001. It recites a method that computes per-heartbeat ST deviations over an initial time period, derives upper and lower limits that bound most-but-not-all of those deviations, reduces the set of deviations within the limits, computes a dispersion measure from the reduced set and the limits, and applies a subsequent test based on the dispersion measure; dependent claims refine how the dispersion and threshold test is defined and optionally how heart rate ranges stratify stored deviations.
Most-but-not-all ST deviation bounding for dispersion-based event test
Determine an upper limit by selecting an ST deviation value that exceeds the value of most but not all of the plurality of ST deviations; determine a lower limit by selecting an ST deviation value that is less than the value of most but not all of the plurality of ST deviations, where the upper and lower limits contain therebetween a reduced set of ST deviations with respect to all ST deviations computed within the initial time interval; compute a dispersion measure of the reduced set within the initial time period that is based on both the upper and lower limits; apply a test after the initial time period to detect a cardiac event, wherein the test is based on the dispersion measure.
Singular-heartbeat ST deviations from digitized waveform over initial time period
Compute a plurality of ST deviations from the digitized waveform over an initial time period for a plurality of heart beats, each ST deviation being an ST deviation for a singular heartbeat; store data pertaining to the plurality of ST deviations.
Dispersion measure based on gap between upper and lower limits
Compute the dispersion measure of the reduced set based on the difference between the upper and lower limits.
In-band average or median of reduced ST deviations
Compute an average value of the plurality of ST deviations in between the upper and lower limits; define the average value as a median value of the plurality of ST deviations in between the upper and lower limits.
Upper threshold from median plus multiple of dispersion
Apply an upper threshold equal to the average value plus a multiple of the dispersion measure as part of the test to detect a cardiac event.
Heart-rate-range stratification for computing ST limits
Store the plurality of ST deviations according to heart rate ranges, wherein the upper and lower limits are computed for one of the plurality of heart rate ranges.
Across the independent claim and its dependents, the claimed core is the selection of upper and lower ST deviation limits that bound most-but-not-all of singular-heartbeat ST deviations, reduction of the ST deviation set to those between the limits, computation of a dispersion measure based on the reduced set, and performing an event-detection test after the initial time period based on that dispersion measure, optionally with heart-rate-range stratification and explicit threshold relationships.
Stated Advantages
Provides heart-rate-dependent ischemia or cardiac-event detection thresholds derived from estimated normal ST deviation ranges.
Uses a reduced set of ST deviations confined between upper and lower boundaries based on a predetermined cumulative percentage range.
Applies a dispersion-measure-based test after the initial time period to detect cardiac events.
Supports robustness via beat qualification and ST deviation correction for QRS amplitude.
Supports multi-lead decision logic by combining normalized ST deviation classifiers for left and leftup leads with RR-dependent thresholds.
Documented Applications
Acute cardiac ischemia detection using ECG-derived ST segment deviation statistics.
Cardiac-event detection based on a dispersion-measure test applied after an initial time period.
Use in an implantable medical device (IMD) and external equipment for detecting ischemia or cardiac events.
Interested in licensing this patent?