U.S. patent application number 13/552712 was filed with the patent office on 2014-01-09 for apparatus and method for detecting myocardial ischemia using analysis of high frequency components of an electrocardiogram.
This patent application is currently assigned to BSP Biological Signal Processing Ltd.. The applicant listed for this patent is Guy AMIT, Tamir BEN-DAVID, Yair GRANOT, Oded LURIA, Eran TOLEDO. Invention is credited to Guy AMIT, Tamir BEN-DAVID, Yair GRANOT, Oded LURIA, Eran TOLEDO.
Application Number | 20140012148 13/552712 |
Document ID | / |
Family ID | 49840950 |
Filed Date | 2014-01-09 |
United States Patent
Application |
20140012148 |
Kind Code |
A1 |
AMIT; Guy ; et al. |
January 9, 2014 |
APPARATUS AND METHOD FOR DETECTING MYOCARDIAL ISCHEMIA USING
ANALYSIS OF HIGH FREQUENCY COMPONENTS OF AN ELECTROCARDIOGRAM
Abstract
ECG apparatus comprises an ECG input for obtaining ECG signals;
a high frequency analyzer configured for obtaining high frequency
QRS components from a QRS complex within said ECG signals and
identifying therein at least one reduced amplitude
zone--RAZ--present within a given QRS complex; and a RAZ quantifier
configured for obtaining a quantification of said at least one RAZ
region within said QRS complex.
Inventors: |
AMIT; Guy; (Ganei-Tikva,
IL) ; LURIA; Oded; (Tel-Aviv, IL) ; TOLEDO;
Eran; (Tel Aviv, IL) ; GRANOT; Yair; (Modiln,
IL) ; BEN-DAVID; Tamir; (Tel-Aviv, IL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
AMIT; Guy
LURIA; Oded
TOLEDO; Eran
GRANOT; Yair
BEN-DAVID; Tamir |
Ganei-Tikva
Tel-Aviv
Tel Aviv
Modiln
Tel-Aviv |
|
IL
IL
IL
IL
IL |
|
|
Assignee: |
BSP Biological Signal Processing
Ltd.
Tel-Aviv
IL
|
Family ID: |
49840950 |
Appl. No.: |
13/552712 |
Filed: |
July 19, 2012 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61667482 |
Jul 3, 2012 |
|
|
|
Current U.S.
Class: |
600/509 |
Current CPC
Class: |
A61B 5/7253 20130101;
A61B 5/0428 20130101; A61B 5/7282 20130101; A61B 5/04014 20130101;
A61B 5/7203 20130101; A61B 5/7264 20130101; A61B 5/0472 20130101;
A61B 5/02028 20130101 |
Class at
Publication: |
600/509 |
International
Class: |
A61B 5/0472 20060101
A61B005/0472 |
Claims
1. ECG apparatus comprising: an ECG input for obtaining wideband
ECG signals; a high frequency analyzer configured for: obtaining
high frequency QRS components from a QRS complex within said ECG
signals and a low frequency envelope around said high frequency QRS
components, and identifying within said envelope at least one
reduced amplitude zone--RAZ--present within a given QRS complex;
and a RAZ quantifier configured for obtaining a quantification of
said at least one RAZ region within said QRS complex, wherein said
quantification is used to compute an index that is proportional to
an area of said RAZ within said QRS complex.
2. (canceled)
3. The ECG apparatus of claim 1, wherein said index comprises a
ratio between an area of the at least one RAZ and an area within
the envelope.
4. The ECG apparatus of claim 3, wherein said area of the at least
one RAZ is an area over a local minimum in said envelope between
two flanking maxima in said envelope.
5. The ECG apparatus of claim 4, wherein said area of at least one
RAZ is measured by connecting respective flanking maxima.
6. The ECG apparatus of claim 4, wherein said area of at least one
RAZ is measured under a horizontal line at a height of one of said
flanking maxima, or by connecting peaks of said flanking maxima or
by estimating an expected peak without RAZ by extrapolation and
calculating a difference in areas.
7. The ECG apparatus of claim 1, wherein said high frequency
analyzer is configured to obtain said high frequency components by
aligning QRS complexes of successive heartbeats to cancel
noise.
8. The ECG apparatus of claim 1, comprising a plurality of leads,
the apparatus configured to obtain a quantification of said RAZ for
each lead respectively and to calculate an overall RAZ
quantification statistically from said respective lead
quantifications.
9. The ECG apparatus of claim 8, wherein said plurality of leads
comprises 12 leads and said statistical calculation comprises
taking an average of one or more highest lead quantifications.
10. The apparatus of claim 1 further comprising an output module
configured to indicate myocardial ischemia in the presence of a
relatively high RAZ quantification.
11. The apparatus of claim 10, wherein said output module is
further configured to use a falling RAZ quantification over time to
indicate successful interventions.
12. The apparatus of claim 1, included at least partly within an
implantable device.
13. ECG method comprising: obtaining ECG signals; obtaining high
frequency QRS components from a QRS complex within said ECG
signals, obtaining a low frequency envelope around said high
frequency components, identifying therein at least one reduced
amplitude zone--RAZ--present within a given QRS complex; and
obtaining a quantification of said at least one RAZ region within
said QRS complex, wherein said quantification is proportional to a
size of said RAZ within said QRS complex.
14. (canceled)
15. ECG method comprising: obtaining ECG signals; obtaining high
frequency QRS components from a QRS complex within said ECG
signals, obtaining a low frequency envelope around said high
frequency components, identifying therein at least one reduced
amplitude zone--RAZ--present within a given QRS complex; and
obtaining a quantification of said at least one RAZ region within
said QRS complex, wherein said obtaining a quantification
comprising obtaining a ratio between an area of the at least one
RAZ and an area within the envelope.
16. ECG method comprising: obtaining ECG signals; obtaining high
frequency QRS components from a QRS complex within said ECG
signals, obtaining a low frequency envelope around said high
frequency components, identifying therein at least one reduced
amplitude zone--RAZ--present within a given QRS complex; and
obtaining a quantification of said at least one RAZ region within
said QRS complex, wherein said area of the at least one RAZ is an
area over a local minimum in said envelope between two flanking
maxima of said envelope, or wherein said area of at least one RAZ
is measured by connecting said respective flanking maxima, or
wherein said area of at least one RAZ is measured under a
horizontal line at a height of one of said flanking maxima, or
wherein said area of at least one RAZ is measured by connecting
peaks of said flanking maxima, or wherein said area of at least one
RAZ is measured by estimating an expected peak without RAZ by
extrapolation and calculating a difference in areas.
17. The ECG method of claim 13, comprising obtaining said high
frequency components by aligning QRS complexes of successive
heartbeats to cancel noise.
18. The ECG method of claim 13, comprising using a plurality of ECG
leads, the method comprising obtaining a quantification of said RAZ
for each lead respectively and calculating an overall RAZ
quantification statistically from said respective lead
quantifications.
19. The method of claim 13, further comprising indicating
myocardial ischemia in the presence of a relatively high RAZ
quantification.
20. The method of claim 19, further comprising using a falling RAZ
quantification over time to indicate successful interventions.
21. The method of claim 13, comprising obtaining a first
quantification relatively soon after onset of a suspected
myocardial infarction and obtaining subsequent quantifications at
succeeding time points thereafter.
22. The method of claim 13, further comprising providing quality
analysis by measuring said quantification successively over a
moving time window to determine stability.
23. The method of claim 13, comprising indicating myocardial
ischemia during a monitoring period on the basis of changes in said
RAZ quantification over time.
Description
RELATED APPLICATION
[0001] This application claims the benefit of priority under 35 USC
.sctn.119(e) of U.S. Provisional Patent Application No. 61/667,482
filed Jul. 3, 2012, the contents of which are incorporated herein
by reference in their entirety.
FIELD AND BACKGROUND OF THE INVENTION
[0002] The present invention, in some embodiments thereof, relates
to an apparatus and method for detecting myocardial ischemia using
analysis of high frequency components of an electrocardiogram, and
more particularly but not exclusively to use of regions within the
QRS complex known as reduced amplitude zones or RAZ.
[0003] ECG is used to measure the rate and regularity of
heartbeats, as well as the size and position of the chambers, the
presence of any damage to the heart, and the effects of drugs or
devices used to regulate the heart.
[0004] ECG may be measured during rest (resting ECG) or when the
heart is under stress (stress ECG).
[0005] Usually more than two electrodes are used, and they can be
combined into a number of pairs (For example: left arm (LA), right
arm (RA) and left leg (LL) electrodes form the three pairs LA+RA,
LA+LL, and RA+LL). The output from each pair is known as a lead.
Each lead looks at the heart from a different angle. Different
types of ECGs can be referred to by the number of leads that are
recorded, for example 3-lead, 5-lead or 12-lead ECGs. A 12-lead ECG
is one in which 12 different electrical signals are recorded at
approximately the same time and will often be used as a one-off
recording of an ECG, traditionally printed out as a paper copy. 3-
and 5-lead ECGs tend to be monitored continuously and viewed only
on the screen of an appropriate monitoring device, for example
during an operation or whilst being transported in an
ambulance.
[0006] An ECG is the best way to measure and diagnose abnormal
rhythms of the heart, particularly abnormal rhythms caused by
damage to the conductive tissue that carries electrical signals, or
abnormal rhythms caused by electrolyte imbalances. In a myocardial
infarction (MI), the ECG can identify if the heart muscle has been
damaged in specific areas, though not all areas of the heart are
covered. Acute coronary syndrome (ACS) refers to any group of
symptoms attributed to obstruction of the coronary arteries.
[0007] The ECG device detects and amplifies the tiny electrical
changes on the skin that are caused when the heart muscle
depolarizes and subsequently repolarizes during each heartbeat. At
rest, each heart muscle cell has a negative charge, which causes
the membrane potential, across its cell membrane. Decreasing this
negative charge towards zero, via the influx of the positive
cations, Na.sup.+ and Ca.sup.++, is called depolarization, which
activates the mechanisms in the cell that cause it to contract.
During each heartbeat, a healthy heart will have an orderly
progression as a wave of depolarization, that is triggered by the
cells in the sinoatrial node, spreads out through the atrium, then
passes through the atrioventricular node and finally spreads all
over the ventricles. The progression is detected as waveforms in
the voltage between two electrodes placed either side of the heart
and may be displayed as a wavy line either on a screen or on paper.
This display indicates the overall rhythm of the heart and
weaknesses in different parts of the heart muscle.
[0008] A typical ECG tracing of the cardiac cycle (heartbeat)
consists of a P wave, a QRS complex, a T wave, and a U wave which
is normally visible in 50 to 75% of ECGs. The baseline voltage of
the electrocardiogram is known as the isoelectric line. Typically
the isoelectric line is measured as the portion of the tracing
following the T wave and preceding the next P wave.
[0009] The standard ECG traces ignore, indeed usually filter out,
high frequency components, for example signals above 100 Hz and in
some case even lower thresholds such as 75 Hz or even 50 Hz. In
general the noise level is such that high frequency components
cannot be reliably isolated from a single ECG trace. In order to
obtain high frequency components one typically needs to align ECG
traces from successive heartbeats so that noise cancels.
[0010] One feature that appears from the high frequency component
following alignment is the reduced amplitude zone or RAZ. The
presence or absence of RAZ is discussed in the following patent
disclosures:
[0011] U.S. Pat. No. 7,113,820 filed Jul. 12, 2001 and U.S. Pat.
No. 7,539,535 filed Jan. 26, 2006 both disclose real time cardiac
electrical data being received from a patient, manipulated to
determine various useful aspects of the ECG signal, and displayed
in real time in a useful form on a computer screen or monitor. The
monitor displays the high to frequency data from the QRS complex in
units of microvolts, juxtaposed with a display of conventional ECG
data in units of millivolts or microvolts. The high frequency data
are analyzed for their root mean square (RMS) voltage values and
the discrete RMS values and related parameters are displayed in
real time. The high frequency data from the QRS complex are
analyzed with imbedded algorithms to determine the presence or
absence of reduced amplitude zones, referred to herein as "RAZs".
RAZs are displayed as "go, no-go" signals on the computer monitor.
The RMS and related values of the high frequency components are
displayed as time varying signals, and the presence or absence of
RAZs may be similarly displayed over time.
[0012] In U.S. Pat. No. 7,386,340 filed Mar. 26, 2003, a system for
the diagnosis and monitoring of coronary artery disease, acute
coronary syndromes, cardiomyopathy and other cardiac conditions is
disclosed. Cardiac electrical data are received from a patient,
manipulated to determine various useful aspects of the ECG signal,
and displayed and stored in a useful form using a computer. The
computer monitor displays various useful information, and in
particular graphically displays various permutations of reduced
amplitude zones and kurtosis that increase the rapidity and
accuracy of cardiac diagnoses. The disclosure provides criteria for
recognizing reduced amplitude zones that enhance the sensitivity
and specificity for detecting cardiac abnormalities.
[0013] The above disclosures have in common that they decide
whether or not a RAZ is present and draw their conclusions from
that.
SUMMARY OF THE INVENTION
[0014] Although several studies have shown the correlation of RAZ
and myocardial ischemia, there is currently no way of quantifying
the RAZ and using such quantity in order to detect a heart
condition such as for example, acute coronary syndrome (ACS) or
infer the seriousness of a condition or changes in the condition.
The present embodiments introduce an apparatus and method for
quantifying the RAZ and thereby allowing early detection of ACS
with an indication of seriousness and also permitting a means of
monitoring for changes in the condition.
[0015] The present embodiments provide an index of the RAZ, which
index can be obtained from a patient at various times to diagnose
some acute cardiac event as well as to indicate the initial state
of the patient undergoing a suspected cardiac event and the
patient's subsequent response to intervention. Typically, under
these circumstances a resting ECG is used.
[0016] According to an aspect of some embodiments of the present
invention there is provided ECG apparatus comprising:
[0017] an ECG input for obtaining wideband ECG signals;
[0018] a high frequency analyzer configured for: [0019] obtaining
high frequency QRS components from a QRS complex within the ECG
signals and a low frequency envelope around the high frequency QRS
components, and [0020] identifying within the envelope at least one
reduced amplitude zone--RAZ--present within a given QRS complex;
and
[0021] a RAZ quantifier configured for obtaining a quantification
of the at least one RAZ region within the QRS complex.
[0022] In an embodiment, the quantification is used to compute an
index that is proportional to an area of the RAZ within the QRS
complex.
[0023] In an embodiment, the index comprises a ratio between an
area of the at least one RAZ and an area within the envelope.
[0024] In an embodiment, the area of the at least one RAZ is an
area over a local minimum in the envelope between two flanking
maxima in the envelope.
[0025] In an embodiment, the area of at least one RAZ is measured
by connecting respective flanking maxima.
[0026] In an embodiment, the area of at least one RAZ is measured
under a horizontal line at a height of one of the flanking maxima,
or by connecting peaks of the flanking maxima or by estimating an
expected peak without RAZ by extrapolation and calculating a
difference in areas.
[0027] In an embodiment, the high frequency analyzer is configured
to obtain the high frequency components by aligning QRS complexes
of successive heartbeats to cancel noise.
[0028] An embodiment may comprise a plurality of leads, the
apparatus configured to obtain a quantification of the RAZ for each
lead respectively and to calculate an overall RAZ quantification
statistically from the respective lead quantifications.
[0029] In an embodiment, the plurality of leads comprises 12 leads
and the statistical calculation comprises taking an average of one
or more highest lead quantifications.
[0030] An embodiment may comprise an output module configured to
indicate myocardial ischemia in the presence of a relatively high
RAZ quantification.
[0031] In an embodiment, the output module is further configured to
use a falling RAZ quantification over time to indicate successful
interventions.
[0032] An embodiment may be included at least partly within an
implantable device.
[0033] According to a second aspect of the present invention there
is provided an ECG method comprising:
[0034] obtaining ECG signals;
[0035] obtaining high frequency QRS components from a QRS complex
within the ECG signals,
[0036] obtaining a low frequency envelope around the high frequency
components,
[0037] identifying therein at least one reduced amplitude
zone--RAZ--present within a given QRS complex; and
[0038] obtaining a quantification of the at least one RAZ region
within the QRS complex.
[0039] Embodiments may involve indicating myocardial ischemia in
the presence of a relatively high RAZ quantification, or using a
falling RAZ quantification over time to indicate successful
interventions.
[0040] Embodiments may involve obtaining a first quantification
relatively soon after onset of a suspected myocardial infarction
and obtaining subsequent quantifications at succeeding time points
thereafter.
[0041] Embodiments may involve providing quality analysis by
measuring the quantification successively over a moving time window
to determine stability.
[0042] Embodiments may involve indicating myocardial ischemia
during a monitoring period on the basis of changes in the RAZ
quantification over time.
[0043] Unless otherwise defined, all technical and/or scientific
terms used herein have the same meaning as commonly understood by
one of ordinary skill in the art to which the invention pertains.
Although methods and materials similar or equivalent to those
described herein can be used in the practice or testing of
embodiments of the invention, exemplary methods and/or materials
are described below. In case of conflict, the patent specification,
including definitions, will control. In addition, the materials,
methods, and examples are illustrative only and are not intended to
be necessarily limiting.
[0044] Implementation of the method and/or system of embodiments of
the invention can involve performing or completing selected tasks
manually, automatically, or a combination thereof. Moreover,
according to actual instrumentation and equipment of embodiments of
the method and/or system of the invention, several selected tasks
could be implemented by hardware, by software or by firmware or by
a combination thereof using an operating system.
[0045] For example, hardware for performing selected tasks
according to embodiments of the invention could be implemented as a
chip or a circuit. As software, selected tasks according to
embodiments of the invention could be implemented as a plurality of
software instructions being executed by a computer using any
suitable operating system. In an exemplary embodiment of the
invention, one or more tasks according to exemplary embodiments of
method and/or system as described herein are performed by a data
processor, such as a computing platform for executing a plurality
of instructions. The data processor may include a volatile memory
for storing instructions and/or data and/or a non-volatile storage,
for example, a magnetic hard-disk, flash memory and/or removable
media, for storing instructions and/or data. A network connection
may be provided and a display and/or a user input device such as a
keyboard or mouse may be available as necessary.
BRIEF DESCRIPTION OF THE DRAWINGS
[0046] The patent or application file contains at least one drawing
executed in color. Copies of this patent or patent application
publication with color drawing(s) will be provided by the Office
upon request and payment of the necessary fee.
[0047] Some embodiments of the invention are herein described, by
way of example only, with reference to the accompanying drawings.
With specific reference now to the drawings in detail, it is
stressed that the particulars shown are by way of example and for
purposes of illustrative discussion of embodiments of the
invention. In this regard, the description taken with the drawings
makes apparent to those skilled in the art how embodiments of the
invention may be practiced.
[0048] In the drawings:
[0049] FIG. 1 is a simplified block diagram schematically
illustrating the main blocks of an ECG device according to
embodiments of the present invention;
[0050] FIG. 2 is a schematic diagram illustrating the positioning
of electrodes for a 12 lead ECG, according to known art;
[0051] FIG. 3 is a schematic flow chart illustrating a procedure
for processing a high resolution ECG to provide a RAZ
quantification according to an embodiment of the present
invention;
[0052] FIGS. 4A to 4E are simplified diagrams showing envelopes
drawn around high frequency components of the QRS complex and
illustrating various ways of quantifying the RAZ area;
[0053] FIG. 5 is a graph showing the RAZ envelope and changes
therein according to the present embodiments as a patient enters
hospital with a complaint and then undergoes treatment;
[0054] FIG. 6 is a simplified graph illustrating HFMI according to
embodiments of the present invention for different groups of
patients at admission, post angio and after 24 hours; and
[0055] FIG. 7 is a simplified set of ECG readings and RAZ envelope
diagrams according to the present embodiments taken for a patient
determined not to have an acute cardiac condition.
DESCRIPTION OF SPECIFIC EMBODIMENTS OF THE INVENTION
[0056] The present invention, in some embodiments thereof, relates
to quantification of the RAZ and use of that quantification to draw
conclusions about the state of a patient's heart.
[0057] Embodiments provide an electrocardiograph with a wide
bandwidth capability able to record and analyze high frequency
signals with high resolution. The QRS complexes are detected in all
the leads and throughout the entire duration of the test. The high
frequency components of the QRS (HFQRS) are filtered to be further
analyzed. If the signal to noise ratio (SNR) of the HFQRS is not
sufficient for reliable analysis, as is many times the case,
several wideband or unfiltered QRS complexes from the same lead are
aligned and subsequently averaged to increase the SNR. The averaged
signal is then filtered to obtain the HFQRS.
[0058] A low frequency envelope is computed for the HFQRS, and
Reduced Amplitude Zones (RAZ) in this envelope are sought. Such RAZ
can be depicted as basins and one or more may exist in the HFQRS
envelope. The area of the basins is computed and the ratio of that
area to the area under the envelope is defined to be a new index.
This new index is named the High Frequency Morphology Index (HFMI)
and may be used for detecting acute coronary syndrome as a RAZ in
fact is believed to arise from changes in the electrophysiological
properties of certain parts of the myocardium, changes which may be
attributed to myocardial ischemia.
[0059] The present embodiments may thus be useful as a signal
processing element in the diagnostic process of patients presenting
with chest pain, who today are monitored or treated in various
ways, all of which suffer from relatively low sensitivity and/or
specificity. By measuring the HFMI in real time, the clinical team
at a hospital, emergency room, ambulance or other clinical
settings, may have an additional valuable indication of possible
acute ischemia. As a direct consequence, better care may be offered
to cardiac patients, who can benefit from an earlier and more
accurate diagnosis and treatment, as well as to non-cardiac
patients, who may avoid false-positive diagnosis and unnecessary
further procedures.
[0060] The present embodiments provide various alternate ways of
computing the HFMI. The signal processing pathway for filtering the
HFQRS affects the obtained envelope and therefore the basins. The
methods of computing the envelope from the filtered HFQRS signal
also affect the area of the basins. In addition, the exact method
of defining the basin's area along with the way in which multiple
basins are treated, both have a direct effect on the HFMI. All of
these methods, which are detailed below, are independent and thus
can be applied independently and then may be combined for enhanced
efficiency.
[0061] Before explaining at least one embodiment of the invention
in detail, it is to be understood that the invention is not
necessarily limited in its application to the details of
construction and the arrangement of the components and/or methods
set forth in the following description and/or illustrated in the
drawings and/or the Examples. The invention is capable of other
embodiments or of being practiced or carried out in various
ways.
[0062] Referring now to the drawings, FIG. 1 is a simplified block
diagram showing an ECG apparatus 10 according to embodiments of the
present invention. An ECG input 12 obtains ECG signals from one or
more ECG leads, typically between 3 and 12. FIG. 2 illustrates a
typical layout for a 12 lead ECG.
[0063] A high frequency analyzer 14 obtains high frequency QRS
components from the QRS complexes within the ECG signals from the
various leads. Within the high frequency components one or more
reduced amplitude zones--RAZ--may be identified within any given
QRS complex, as will be explained in greater detail below.
[0064] A RAZ quantifier 16 obtains a quantification of the RAZ
within the QRS complex. As will be explained in greater detail
below, if there are several RAZ regions then the quantification may
use all of the regions, or an average, or the largest region or any
other reasonable configuration. As long as the quantification is
consistent any configuration of the different RAZ regions may work,
and the skilled person will be able to select between the different
methods to obtain the clearest result.
[0065] An output module 18 uses the quantification of the RAZ to
estimate myocardial ischemia. In the presence of a relatively high
RAZ quantification, myocardial ischemia is inferred. A falling RAZ
quantification over time is taken as a sign of successful
interventions or a natural healing process (spontaneous
reperfusion).
[0066] Reference is now made to FIG. 3, which is a simplified flow
chart showing the quantification process of FIG. 1. In stage A,
alignment and averaging of successive QRS complexes of a high
resolution resting ECG is carried out. Aligning QRS complexes of
successive heartbeats reinforces the signal and cancels noise.
[0067] In stage B the averaged ECG is filtered and in stage C a
high frequency QRS signal is obtained. An envelope is then drawn
around the high frequency QRS signal in stage D and a basin area is
obtained between two adjacent peaks of the envelope, based on the
water hole principle as discussed hereinbelow. The basin area,
shown in hashing, is the RAZ area, and a ratio between the RAZ area
and the area within the envelope is calculated for each lead. Then,
an average or other statistical derivation may be calculated over
all the leads in stage E.
[0068] The envelope is typically a low frequency envelope, and may
be obtained using a Hilbert transform, as discussed in greater
detail below. The hashed area shown in FIG. 3 stage D provides the
basin, and the quantification may be defined as a ratio between the
area of the RAZ basin and an area within the envelope.
Alternatively, as discussed below, the absolute area of the RAZ
could be used. If there is more than one basin then, as will be
explained in greater detail below, the area of the largest basin
may be used, or alternatively a sum of a given number of largest
basins or a sum of all the basins may be used.
[0069] The area of the RAZ, the hashed area of FIG. 3 stage D may
be defined as an area over a local minimum in the RAZ in question
between two flanking maxima. Such an area is indicated as 40 in
FIG. 4A, where 42 and 44 denote the flanking maxima and 46 denotes
the local minimum. The hashed area mentioned above may for example
be measured by taking the upper or lower of the two maxima and
connecting it via a horizontal line to the upsloping segment of the
other maxima. An example of such a line drawn from an upper maxima
is shown in FIG. 4C and one from a lower maxima is depicted in FIG.
3D.
[0070] As shown in FIG. 4B the area may also be measured by
connecting peaks 48 and 50 of respective flanking maxima.
[0071] In FIG. 4C, another alternative is shown in which the area
is measured under a horizontal line at a height of one of the
flanking maxima, 52 and 54. As shown the higher of the two maxima
is used, but the lower could alternatively be used, as shown in
FIG. 3D.
[0072] In an embodiment, the apparatus may be included partly or
wholly within an implantable device, say for long term monitoring
of highly at risk patients. The device uses implantable electrodes
that can be intracardiac or epicardiac electrodes.
A device according to the present embodiments may be useful in
detection of ischemia, and provides measurement of the ECG signal
between two implantable electrodes, or between an implantable
electrode and the can of the implantable device, and identifies
changes in RAZ area.
[0073] In a preferred embodiment of the invention a high resolution
wideband ECG acquisition device is used for sampling the ECG signal
at a rate of 1000 Hz. The analog to digital converter employs high
resolution so that signal changes of at least 1 microvolt can be
reliably received and recorded.
[0074] The analysis algorithm as discussed above with reference to
FIG. 3, uses a template-based correlation to identify valid QRS
complexes and exclude noisy or ectopic beats. Careful beat
alignment, followed by beat averaging, is used to increase the
signal-to-noise ratio. The level of noise in the HFQRS signal may
be calculated as the root-mean-square of high frequency components
within the ST segment. Beat averaging is applied to each of the
leads until the level of noise is for example .ltoreq.1 .mu.V.
Other thresholds may be used and averaging within a fixed window is
an option. Beats with low signal to noise ratio are removed from
further processing. Each valid average QRS complex is filtered by a
band-pass filter in the frequency band of 140 Hz to 250 Hz,
although similar frequency bands, wider, narrower or somewhat
shifted, may also be used. The filtered signal, which includes the
high frequency components of the QRS complex is defined as the
HFQRS. The time-domain envelope of the HFQRS complex is calculated
using a Hilbert transform followed by a low pass filter. In this
envelope, RAZ are quantified by computing the ratio of the area of
the basins to the area under the envelope, as explained. An index,
referred to hereinbelow as HFMI, is calculated for each average
HFQRS complex, and the median index of all valid complexes in a
lead is determined to be the HFMI value of the lead. As
alternatives to the median, points with minimal noise or may be
selected, or any other suitable method may be used. HFMI value per
patient may be defined as the average of a number of leads which
may be chosen to be those with maximal index value, or via any
other suitable statistical measure.
[0075] In order to reliably compute the HFMI, the filtered HFQRS
signal requires a high signal to noise ratio. Since the recorded
ECG signal is usually tainted by noise from various sources, in
most practical scenarios the SNR may be improved by signal
processing methods. These include standard procedures such as
filtering outside noise sources, like the mains power frequency (50
Hz/60 Hz) but also collecting multiple QRS complexes from the same
lead and averaging them to obtain a higher SNR as discussed above.
The averaging process includes detecting the QRS complexes in the
ECG signal, defining a template for a QRS complex and subsequently
correlating each QRS complex with the template to decide whether or
not this specific complex should be included in the averaging
process. The selected complexes are then aligned, possibly using
sub-sampling accuracy and averaged to obtain the wideband QRS
complex which includes the entire sampled bandwidth of the signal,
where the HFQRS is usually negligible compared to the low frequency
QRS signal amplitude. The alignment process may be carried out
using a partial bandwidth, e.g. a mid-range frequency band such as
30 Hz-70 Hz. Filtering out the lower frequencies for the alignment
process results in more accurate alignment, but since most of the
signal's energy is found in the lower frequencies, SNR
considerations inhibit the use of higher frequency ranges--hence
the choice of the mid range frequency band.
[0076] Averaging multiple QRS complexes increases the SNR, but may
distort the signal, for example if the HFQRS changes during the
averaged time span. Thus it is helpful to keep the averaged segment
at a minimum and an optimal averaging duration is sought. Such an
averaging duration can be arbitrarily determined but it can also be
adaptive by specifying the required noise level or SNR. Defining
the noise level can be difficult since multiple sources of noise
exist. One of the possibilities is measuring the high frequency ECG
signal at a certain part of the cycle, where no high frequency
signal is expected so that any signal measured in such a part of
the cycle may be considered as noise. Such a process can be carried
out using the root mean square signal of the ST segment. It is
assumed that the high frequency noise level in the QRS complex is
identical to that of the ST segment. The averaging can thus be
determined to include the minimal number of sequential
template-correlated QRS complexes whose matching signal level in
the ST segment is lower than the noise threshold, for example,
under 1 microvolt.
[0077] Since the HFMI is computed from minute high frequency
components, with respect to the much higher-voltage low frequency
components, it is provided that the analog to digital converter
(ADC) which samples the signal has a high resolution. However, due
to the averaging process described above, the requirements for the
ADC may be relaxed. Even if microvolt levels of the signal are to
be analyzed, lower resolution ADC's may be employed since the
averaged signal will have a better resolution and accuracy than the
original sampled signal, a phenomenon known as super
resolution.
[0078] While QRS complexes from different leads cannot be averaged,
they may still be aligned. Thus, various parameters such as
detecting the QRS, defining its central location in the time domain
and its time domain boundaries may be computed considering more
than a single lead. All of the ECG leads signals may be
synchronized so that every time point in one lead may be directly
matched with a similar time point in a different lead. By taking
advantage of this synchronization, a more accurate estimate of the
QRS parameters mentioned above may be achieved compared with
similar estimates based on a single lead.
[0079] The HFQRS has a typical shape such as the one depicted in
FIG. 3C. It has been previously shown that Reduced Amplitude Zones
(RAZ) are indicative of changes in myocardium conduction, which may
be caused by myocardial ischemia. The high frequency components in
the QRS are attributed to the fractal nature of the electric
activation signal dispersing in the myocardium. This pattern is
brought about by the numerous purkinje fibers which stimulate
multiple groups of myocytes to contract. As a result of ischemia
the myocyte to myocyte conduction is impaired and thus there are
fewer discrete sources of effective signals that propagate through
the tissue, leading to a more uniform and smooth propagation front
and consequently to less high frequency content during
depolarization. The RAZ is a result of the tissue inhomogeneity
which translates to temporal differences in the ECG signal. Thus
signals arriving from ischemic tissue will have a lower component
of high frequencies than signals which emanate from healthy tissue.
The depiction of these temporal differences in the amount of high
frequency components is what is shown in the reduced amplitude
zones. To detect RAZ one may study the envelope of the HFQRS, but
the envelope is not a clearly defined attribute, nor is it
straightforward to compute. The present embodiments use any of
several methods in order to obtain the envelope such as full wave
or half wave rectification, using the Hilbert transform or other
digital processing techniques. Various low pass filters may be used
in order to smooth the envelope.
[0080] Regardless of the method used for detecting the envelope,
RAZ in the envelope is defined and the corresponding area is
computed. One way is to look at the RAZ as if it is a basin filled
with water as depicted in FIG. 3D. The area so to speak filled with
water corresponds to the area of the RAZ. In one method the area is
delimited by the lowest peak of the two peaks delimiting the RAZ.
Another method for computing the area is taking a straight line
between these two peaks as demonstrated in FIG. 4B. Yet another
method considers the highest peak and takes a horizontal line
between it and a vertical line from the lower peak, as depicted in
FIG. 4C. Generally speaking, the area of the RAZ may be computed as
a function of the distance between sequential local minima and
maxima of the envelope and the amplitude differences between them.
In its most crude form this area may be a rectangle, but other more
elaborate schemes, such as these described above may be used.
[0081] FIG. 4E illustrates a further option for quantifying RAZ. An
estimate is made by extrapolation of an expected peak without RAZ.
The difference in areas, shown shaded, is then calculated. As
discussed above, the envelope of the QRS complex may be low pass
filtered.
[0082] When multiple RAZ exist, the area may be defined in several
ways. It may be the sum of all the areas, regardless of how they
were computed. It may also be the largest area, i.e. ignoring small
RAZ.
[0083] Since the HFQRS includes, by definition, only high frequency
components, the signal has no DC component and some of the signal
values will be positive while others will be negative and the
signal in general will be centered around zero. In a similar manner
to the detection of the positive envelope discussed so far, a
negative envelope may be detected for the lower part of the HFQRS
where the signal has negative values. Not all of the methods
described above are suitable for this task. For example, full wave
rectification is insensitive to the sign, but half wave
rectification will yield generally different results for the
positive and negative envelopes, as shown in FIG. 4D. The negative
envelope may be used in two ways. First, it may be used in a
similar manner to the positive envelope with all of the methods
described above. Second, it may be used in conjunction with the
positive envelope yielding roughly double the area since both the
upper and lower (upside down) basins are additionally
considered.
[0084] The area of the basins described above is a parameter to
consider when analyzing the HFQRS. The area is affected by some
circumstances such as the signal amplitude. In order to consider a
more robust index of the HFQRS morphology, we turn to the High
Frequency Morphology Index, HFMI, referred to above, which also
considers the area under the HFQRS envelope. The HFMI is the ratio
of both areas, i.e. the area of the basins divided by the area
under the envelope. In the case of a negative envelope or a double
sided envelope, the area in the denominator may be the area above
the envelope or between the envelopes respectively.
[0085] The above has discussed various methods of computing the
HFMI. The procedure may be carried out for all of the available
leads and for any duration of test, although very short tests may
not have sufficient data to allow acceptable SNR. When considering
the time span of the ECG signal used for extracting a single
measurement of HFMI for a specific lead, it is apparent that any
number of HFMI values may be obtained for every lead, when the ECG
test is long enough. The HFMI value may therefore be studied as a
function of time, say when used in monitoring the progress of a
patient. The diagnosis may be based on changes in the HFMI over
time. The idea is that the one measurement of HFMI may not be
sufficiently high to require attention, but if the patient is
monitored for a long time, this somewhat high value (but not high
enough to pass the threshold) may be indicative of ischemia, when
considering the duration of the high HFMI value, its slope and
similar factors.
[0086] Another possibility is to seek a single result for assessing
a patient's condition at a particular point in time. Any number of
statistical tools may be used to determine the HFMI value from the
set of values obtained for each lead. These include: median, mean,
maxima, minima etc.
[0087] When used for monitoring, the quality of the test may be
assessed by using a moving window. Each window may be treated as a
measurement and the variation of the measurement over time, e.g.
the standard deviation, may indicate the degree of the test quality
where a stable result in a stationary situation may indicate high
quality measurements.
[0088] In a study at the intensive coronary care unit (ICCU) of
Soroka Medical Center, the following results were obtained using
the embodiment described above:
[0089] Out of 32 patients who met the criteria for study inclusion
and had a complete set of three ECG recordings, HFQRS analysis was
available in 30 patients (age 55.+-.11 yrs, 26 men). The remaining
2 patients presented inadequate signal quality and were excluded
from the analysis. Most patients (97%) did not have previously
diagnosed coronary artery disease (CAD). Fifteen patients (50%) had
3 or more coronary risk factors.
[0090] The majority of patients (26 pts, 87%) had ST-elevation MI
(STEMI). Of these patients, 17 were urgently reperfused, with a
door-to-balloon time of 87.+-.23 minutes. In 8 STEMI patients
spontaneous reperfusion occurred prior to angiography, ensuing
resolution of their ST segment changes by the time of admission.
One patient had a recent STEMI. In all patients except one,
angiography indicated significant coronary artery disease CAD
(.gtoreq.70% stenosis in a major coronary artery, or .ltoreq.50%
stenosis in the left-main artery). Successful revascularization by
angioplasty was achieved in 24 pts (80%). Of the patients with no
revascularization, four were referred for coronary artery bypass
graft CABG, one had failed percutaneous coronary intervention PCI
and one had insignificant CAD. These patients were included in the
analysis, although their serial ECG recordings were acquired before
intervention. The TIMI risk score for STEMI 18, calculated for 26
STEMI patients, was 2.2.+-.1.8 (Mean.+-.SD). The average time
between onset of symptoms and acquisition of 1st high-resolution
ECG at the ICCU was 5.8.+-.6 hours. Post-revascularization ECG was
acquired 0.7.+-.0.8 hrs after angiography, and 24 h ECG was
acquired 26.1.+-.15 hrs following angiography.
[0091] The ECG at ICCU admission was interpreted by the blind
observer as `ischemic` in 19 pts (63%), `non-ischemic` in 9 pts
(30%) and `inconclusive` in 2 pts (7%). ECG was non-ischemic in 6
STEMI patients with spontaneous reperfusion and 2 NSTEMI patients.
After 24 hours, 32% of the patients with ischemic admission ECG had
non-ischemic or inconclusive ECG, and 45% of the patients with
non-ischemic or inconclusive admission ECG had indications of
ischemia or infarction.
[0092] A typical example of HFQRS analysis results is given in FIG.
5, for a 41 yr old male patient with STEMI. The patient, without
history of CAD, was diagnosed with acute anterior STEMI in the
emergency department before being admitted to the ICCU. Admission
ECG, acquired 2.25 hours after onset of symptoms, showed
spontaneous ST resolution, although chest pain persisted. HFQRS
exhibited significant ischemic morphology (RAZ pattern) in multiple
leads, with HFMI=15% in a typical lead (V4). Urgent angiography,
performed 2 hours after admission, revealed two-vessel disease with
total occlusion of the mid-LAD and critical occlusion of a first
marginal branch. Both vessels were successfully dilated.
Post-revascularization ECG was normal, and HFQRS signal exhibited
partial resolution of RAZ morphology, with HFMI=7%. At 24 h, both
conventional ECG and HFQRS morphology did not indicate ongoing
ischemia.
[0093] The values of HFMI per patient were higher on the admission
ECG than on the post-angiography ECG (4.6.+-.2.9% vs. 3.4.+-.2.3%,
P<0.05) and the 24 h ECG (4.6.+-.2.9% vs. 2.8.+-.2.1%,
P<0.01), as shown in FIG. 6, which is a graph showing HFMI at
three different times for different groupings of the patients. The
number of leads with HFMI value >3%, was higher on the admission
ECG, compared to post-angiography ECG (3.4.+-.2.7 vs. 2.1.+-.2,
P<0.03) and 24 h ECG (3.4.+-.2.7 vs. 2.1.+-.1.8, P<0.02). The
trend of decrease in average HFMI values, following angiography and
after 24 h was also observed in subgroups of STEMI patients who
were referred for urgent reperfusion, as well as in patients with
spontaneous reperfusion. Compared to patients who underwent urgent
reperfusion, in those with spontaneous reperfusion HFMI values
tended to be lower during admission, post-angiography and 24 h.
This difference was not statistically significant, possibly due to
the small sample size.
[0094] HFMI value decreased from admission ECG to 24 h ECG in 71%
of the patients in the analysis group, and in 79% of the patients
who were revascularized by angioplasty.
[0095] A noteworthy case of a patient excluded from the analysis
group is shown in FIG. 7. The patient, a 30 yr-old male with no
risk factors was admitted due to acute chest pain and elevated
Troponin-T (0.8 ng/mL), following 5 days of viral common cold
disease. Admission ECG showed diffuse ST segment elevations (FIG.
7A) and the patient was referred for urgent angiography, which
demonstrated normal coronary arteries. HFQRS analysis revealed
normal signal morphology, with apparent RAZ in only one of the
leads as shown in FIG. 7B. The discharge diagnosis was
perimyocarditis, and no adverse cardiac events were documented
during a 7-month follow-up period.
[0096] It is expected that during the life of a patent maturing
from this application many relevant pulse shaping and symbol
decoding technologies will be developed and the scope of the
corresponding terms in the present description are intended to
include all such new technologies a priori.
[0097] The terms "comprises", "comprising", "includes",
"including", "having" and their conjugates mean "including but not
limited to".
[0098] The term "consisting of" means "including and limited
to".
[0099] As used herein, the singular form "a", "an" and "the"
include plural references unless the context clearly dictates
otherwise.
[0100] It is appreciated that certain features of the invention,
which are, for clarity, described in the context of separate
embodiments, may also be provided in combination in a single
embodiment. Conversely, various features of the invention, which
are, for brevity, described in the context of a single embodiment,
may also be provided separately or in any suitable subcombination
or as suitable in any other described embodiment of the invention.
Certain features described in the context of various embodiments
are not to be considered essential features of those embodiments,
unless the embodiment is inoperative without those elements.
[0101] Although the invention has been described in conjunction
with specific embodiments thereof, it is evident that many
alternatives, modifications and variations will be apparent to
those skilled in the art. Accordingly, it is intended to embrace
all such alternatives, modifications and variations that fall
within the spirit and broad scope of the appended claims.
[0102] All publications, patents and patent applications mentioned
in this specification are herein incorporated in their entirety by
reference into the specification, to the same extent as if each
individual publication, patent or patent application was
specifically and individually indicated to be incorporated herein
by reference. In addition, citation or identification of any
reference in this application shall not be construed as an
admission that such reference is available as prior art to the
present invention. To the extent that section headings are used,
they should not be construed as necessarily limiting.
* * * * *