U.S. patent application number 16/871355 was filed with the patent office on 2021-03-25 for system and method for electromechanical activation of arrhythmias.
This patent application is currently assigned to THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK. The applicant listed for this patent is THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK. Invention is credited to Elisa E. Konofagou, Jean Provost.
Application Number | 20210085284 16/871355 |
Document ID | / |
Family ID | 1000005253304 |
Filed Date | 2021-03-25 |
View All Diagrams
United States Patent
Application |
20210085284 |
Kind Code |
A1 |
Konofagou; Elisa E. ; et
al. |
March 25, 2021 |
SYSTEM AND METHOD FOR ELECTROMECHANICAL ACTIVATION OF
ARRHYTHMIAS
Abstract
Systems and methods for detecting electromechanical wave
propagation within a body structure of a patient in a series of
image frames representing movement the body structure are provided.
Image data is acquired comprising a series of image frames
corresponding to the movement of a body structure. A correlation
calculation is performed on the image frames to generate a
displacement map representing the relative displacement between the
first and second image frames. A video is generated comprising a
series of displacement maps. The parameters of movement of the body
structure are detected by analysis of the displacement maps. The
image acquisition can detect the movement of the body structure
without inducing such movement.
Inventors: |
Konofagou; Elisa E.; (New
York, NY) ; Provost; Jean; (Paris, FR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW
YORK |
New York |
NY |
US |
|
|
Assignee: |
THE TRUSTEES OF COLUMBIA UNIVERSITY
IN THE CITY OF NEW YORK
New York
NY
|
Family ID: |
1000005253304 |
Appl. No.: |
16/871355 |
Filed: |
May 11, 2020 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15048761 |
Feb 19, 2016 |
10687785 |
|
|
16871355 |
|
|
|
|
14449820 |
Aug 1, 2014 |
|
|
|
15048761 |
|
|
|
|
11433510 |
May 12, 2006 |
8858441 |
|
|
14449820 |
|
|
|
|
60680081 |
May 12, 2005 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01S 15/8956 20130101;
G01S 15/8925 20130101; G01S 7/52042 20130101; A61B 8/5223 20130101;
A61B 8/08 20130101; A61B 8/485 20130101; A61B 8/0883 20130101; G01S
7/52087 20130101; G01S 15/8977 20130101; A61B 8/463 20130101 |
International
Class: |
A61B 8/08 20060101
A61B008/08; G01S 15/89 20060101 G01S015/89; G01S 7/52 20060101
G01S007/52 |
Goverment Interests
STATEMENT REGARDING FEDERALLY-SPONSORED RESEARCH
[0003] This invention was made with government support from the
National Institutes of Health under Grant Nos. R01EB006042,
R21HL096094 and R01HL114358. The government has certain rights in
the invention.
Claims
1. A method for mapping electromechanical activity during an
arrhythmia, comprising: obtaining image information of a heart of a
subject using an imaging device; generating a strain map of the
heart from the image information; determining from the strain map
occurrences of a first electromechanical event of the heart and a
second electromechanical event; generating a spatio-temporal map of
atrial and ventricular mechanics of the heart by tracking the onset
of the first and second event for each pixel of a heart wall of the
subject identified from the image information; and identifying,
using the spatio-temporal map, a representative mechanical cycle
associated with a contraction of the heart; and determining a type
of cardiac arrhythmia present in the heart from the information
collected in the ultrasound scan of the heart.
2. The method of claim 2, further comprising determining that the
cardiac arrhythmia present in the heart comprises focal rhythms,
wherein generating the spatio-temporal map further comprises
determining an onset of ventricular contraction by identifying a
first zero-crossing of an incremental strains occurring after an
onset of a P-wave on an electrocardiogram.
3. A system for mapping electromechanical activity during an
arrhythmia comprising: a processor adapted to: obtain image
information of a heart of a subject using an imaging device;
generate a strain map of the heart from the image information;
determine, from the strain map occurrences of a first
electromechanical event of the heart and a second electromechanical
event; generate a spatio-temporal map of atrial and ventricular
mechanics of the heart by tracking the onset of the first and
second event for each pixel of a heart wall of the subject; and
identify, using the spatio-temporal map, a representative
mechanical cycle associated with a contraction of the heart;
wherein the processor is further configured to determine a type of
cardiac arrhythmia present in the heart from the information
collected in the ultrasound scan of the heart.
4. The system of claim 3, wherein the processor determines that the
cardiac arrhythmia present in the heart comprises focal rhythms,
and wherein the processor generates the spatio-temporal map by
determining an onset of ventricular contraction by identifying a
first zero-crossing of an incremental strains occurring after an
onset of a P-wave on an electrocardiogram.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a divisional of U.S. patent application
Ser. No. 15/048,761, entitled "System and Method for
Electromechanical Activation of Arrhythmias", filed Feb. 19, 2016,
now allowed, which is a continuation in part of U.S. patent
application Ser. No. 14/449,820, entitled "Systems And Methods For
Electromechanical Wave Imaging of Body Structures", filed Aug. 1,
2014, which is a continuation of U.S. patent application Ser. No.
11/433,510, entitled "Systems And Methods For Electromechanical
Wave Imaging of Body Structures", filed May 12, 2006, which issued
as U.S. Pat. No. 8,858,441 on Oct. 14, 2014, which claims priority
to U.S. Provisional Application No. 60/680,081 entitled "Systems
And Methods For Electromechanical Wave Imaging of Body Structures",
filed on May 12, 2005, each of which is incorporated herein by
reference in its entirety and from each of which priority is
claimed. This application also claims priority from U.S.
Provisional Application No. 62/118,402, filed Feb. 19, 2015, which
is incorporated by reference herein in its entirety.
[0002] A portion of the disclosure of this patent document contains
material which is subject to copyright protection. The copyright
owner has no objection to the facsimile reproduction by anyone of
any portion of the patent document, as it appears in any patent
granted from the present application or in the Patent and Trademark
Office file or records available to the public, but otherwise
reserves all copyright rights whatsoever.
BACKGROUND
[0004] This present disclosed subject matter relates to techniques
for imaging and detecting the propagation of mechanical waves
within a body structure of a patient.
[0005] Certain medical conditions, such as diagnosis of myocardial
ischemia, can be difficult to establish in their early stages when
treatment is most effective. Patients suffering from myocardial
ischemia can present to an emergency room or acute care facility
with typical cardiac symptoms such as chest pain, described as
tightness, pressure, or squeezing, but some patients can have other
symptoms such as arm or chin pain, nausea, sweating, or abdominal
pain. Certain techniques such as electrocardiogram often provide
inconclusive findings regarding ischemia, and sometimes can even be
unable to identify situations in which ischemia has progressed to
cell damage and myocardial infarction (MI). Other techniques are
available for diagnosing infarction relative to its predecessor,
ischemia. For example, a blood test to measure the creatine
kinase-MB (CK-MB) enzyme level is used for detection of myocardial
cell damage. Other serum markers include troponin I, and to a
lesser extent, myoglobin. However, the blood levels of certain such
compounds can take several hours to rise, so that diagnosis of MI
can be delayed. Reliance on blood tests alone can result in a
significant loss of time when early aggressive therapy is
warranted.
[0006] Certain less invasive diagnostic techniques have become
available through the observation of mechanical properties of
tissue via imaging techniques. Such evaluation of the function of
the heart, cardiovascular tissue, or other body structures can be
based on the mechanical interpretation of the movement of the these
structures, such as, for example, the active contractions and
passive relaxation of the myocardium.
[0007] Using certain imaging techniques, the evaluation of the
heart function can be based on a single mechanical interpretation
of myocardial deformation. By use of these techniques, the
deformations of the myocardium can be quantified over a complete
cardiac cycle in order to provide some information on the
myocardial viability.
[0008] Certain low frequency mechanical vibrations in the heart are
known in human patients. Certain ultrasound techniques can be used
to obtain pulsive mechanical vibrations around end-systole and
end-diastole in the frequency range of 25 to 100 Hz.
[0009] Additionally, atrial arrhythmias are a known and can cause
of morbidity and mortality. Certain mechanical factors, such as
chamber size and wall tension, can affect the onset and
perpetuation of atrial arrhythmia. Certain echocardiographic
measurements can also be used to characterize atrial arrhythmias.
Yet, systems and techniques to analyze the 2-D spatio-temporal
evolution of the local deformations of the atria during e.g., focal
tachycardia, flutter, and fibrillation, would be beneficial.
[0010] Accordingly, there is a need for a noninvasive imaging
modality which provides insight into the source or focus of an
arrhythmia.
SUMMARY
[0011] The present disclosure provides elasticity imaging
techniques to evaluate mechanical wave propagation, and provide an
estimation of electrical propagation in a noninvasive manner.
[0012] In example embodiments, the disclosed subject matter
provides systems and methods for detecting wave propagation within
the tissue of a patient in a series of image frames representing
movement of such tissue of the body structure. Image data is
acquired comprising a series of image frames corresponding to the
movement of the tissue. In an exemplary embodiment, the tissue can
be the wave propagation in the myocardium. In another exemplary
embodiment, the movement of body tissue can be wave propagation in
the arteries or the aorta.
[0013] A correlation calculation can be performed on the image
frames to generate a matrix with the location of correlation maxima
representing the relative displacement between the first and second
image frames, also referred to as a displacement map. A video can
be generated comprising a series of displacement maps. The
parameters of movement of the cardiac structure can be detected,
such as velocity, attenuation, frequency, etc. The wave can be a
shear wave, representative of the electrical wave propagation
within the body structure.
[0014] According to another aspect of the present disclosure,
systems and methods are provided for mapping electromechanical
activity during an arrhythmia. Image information of a heart of a
subject can be obtained using an imaging device. A strain map of
the heart can be generated from the image information. Occurrences
of a first electromechanical event of the heart and a second
electromechanical event can be determined from the strain map. A
spatio-temporal map of atrial and ventricular mechanics of the
heart can be generated by tracking the onset of the first and
second event for each pixel of a heart wall of the subject
identified from the image information. A representative mechanical
cycle associated with a contraction of the heart can be identified
using the spatio-temporal map.
[0015] For example, electromechanical activation mapping can
characterize propagation patterns of electromechanical strains
during focal and reentrant arrhythmias of the heart.
[0016] Additionally, regions in which the mechanical and electrical
activities are decoupled can be identified by mapping the
electromechanical activity of the heart.
[0017] Furthermore, the spatio-temporal map can be generated by
obtaining isochrones strongly correlated to electrical isochrones
by tracking a propagation front of an end-diastole
electromechanical activation of the heart.
[0018] In addition, a type of cardiac arrhythmia present in the
heart can be determined from the information collected in the
ultrasound scan of the heart. Upon identifying that the cardiac
arrhythmia present in the heart includes focal rhythms, an onset of
ventricular contraction can be determined by identifying a first
zero-crossing of an incremental strains occurring after an onset of
a P-wave on an electrocardiogram. Upon determining that the cardiac
arrhythmia present in the heart is a type of reentrant arrhythmia,
a high-resolution Fourier transform can be performed using a
generalized Goertzel algorithm to interpolate strain signals in
Fourier space for each individual pixel in an atria of the
heart.
[0019] For example, a peak mechanical cycle length (MCL) map can be
generated by selecting a MCL having a highest amplitude within the
physiologically-relevant time range for each pixel of the
ultrasound scan of the heart, such that the MCL map identifies, for
each pixel of the atria, which cycle length is most present in a
Fourier spectrum of cycle lengths. The cycle length best
representing an atrial contraction of the heart can be determined.
A phase corresponding to the determined cycle length can be
determined to map a propagation of a mechanical oscillation of the
heart at the determined cycle length.
[0020] Additionally, the ultrasound scan further can be performed
by emitting a circular ultrasonic wave to instruct an ultrasound
apparatus to perform a motion estimation sequence. A B-mode
acquisition can be performed to capture heart anatomy of the heart.
A plurality of beams can be generated to reconstruct frames from
the motion estimation sequence using a delay-and-sum algorithm with
a reconstructed sampling frequency.
BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The patent or application file contains at least one drawing
executed in color. Copies of this patent or patent application
publication with color drawings will be provided by the U.S. Patent
and Trademark Office upon request and payment of the necessary
fee.
[0022] FIG. 1 is a diagram illustrating the system in accordance
with the present disclosure.
[0023] FIG. 2 is a diagram illustrating exemplary stages in a
method in accordance with the present disclosure.
[0024] FIG. 3 is a diagram illustrating a technique for measuring
movement of structures within an image in accordance with the
present disclosure.
[0025] FIG. 4 is a chart representing the velocity of structures
within an image in accordance with the present disclosure.
[0026] FIG. 5 illustrates a technique of detecting wave propagation
in accordance with a further embodiment of the present
disclosure.
[0027] FIG. 6 illustrates a technique of detecting wave propagation
in accordance with yet another embodiment of the present
disclosure.
[0028] FIGS. 7(a)-11(b) are images illustrating the propagation of
a wave within a body structure in accordance with an exemplary
embodiment of the present disclosure.
[0029] FIGS. 12(a)-18(b) are images illustrating the propagation of
a wave within a body structure in accordance with another exemplary
embodiment of the present disclosure.
[0030] FIG. 19 is an ultrasound image of a mouse left ventricle in
a parasternal long-axis view.
[0031] FIG. 20(a) is an axial displacement map overlaid to the
grayscale B-mode image of the left ventricle during systole in
accordance with the present disclosure.
[0032] FIG. 20(b) is an axial displacement map overlaid to the
grayscale B-mode image of the left ventricle during diastole
(relaxation phase) in accordance with the present disclosure.
[0033] FIG. 20(c) is an ECG indicating the time of the acquisition
during the cardiac cycle of FIG. 20(a) in accordance with the
present disclosure.
[0034] FIG. 20(d) is an ECG indicating the time of the acquisition
during the cardiac cycle of FIG. 20(b) in accordance with the
present disclosure.
[0035] FIG. 21(a) is a time plot illustrating the temporal
variation of the axial displacements estimated on one central
RF-line as line plotted on FIG. 20(b) in accordance with the
present disclosure.
[0036] FIG. 21(b) illustrates the frequency content of the
displacement variation in the septum at the depth of 12.5 mm
plotted as a function of time in accordance with the present
disclosure.
[0037] FIG. 21(c) is a time plot illustrating the temporal
variation of the axial displacements after bandpass filtering of
the plot illustrated in FIG. 21(a) showing the transient and high
frequency components in accordance with the present disclosure.
[0038] FIG. 21(d) illustrates the ECG signal acquired
simultaneously with the data illustrated in FIGS. 21(a)-(c) in
accordance with the present disclosure.
[0039] FIGS. 22(a)-(d) illustrate a sequence of axial displacement
maps overlaid to the grayscale B-mode image of the left ventricle
around end-systole taken every 0.6 ms showing the propagation of a
first mechanical wave front in the septum in accordance with the
present disclosure. The arrows indicate the progression of the wave
front in the septum.
[0040] FIGS. 22(e)-(f) illustrate a sequence of axial displacement
maps overlaid to the grayscale B-mode image of the left ventricle
around end-systole taken every 0.6 ms showing the propagation of a
second mechanical wave front in the septum in accordance with the
present disclosure. The arrows indicate the progression of the wave
front in the septum.
[0041] FIGS. 22(g)-(l) illustrate the ECG signal plotted below each
respective image of FIGS. 22(a)-(f) indicating the time t of the
acquisition during the cardiac cycle in accordance with the present
disclosure.
[0042] FIG. 23 is a plot illustrating the distance of propagation
as a function of the phase of the end-systolic wave at three
frequencies in accordance with the present disclosure.
[0043] FIGS. 24(a)-(f) illustrate a sequence of axial displacement
maps overlaid to the grayscale B-mode image of the left ventricle
around the beginning of systole taken every 2.8 ms, showing the
propagation of a strong mechanical wave in the posterior wall in
accordance with the present disclosure. The arrows indicate the
progression of the wave front in the posterior wall.
[0044] FIGS. 24(g)-(l) illustrate the ECG signal plotted below each
respective image of FIGS. 24(a)-(f) indicating the time t of the
acquisition during the cardiac cycle in accordance with the present
disclosure.
[0045] FIG. 25 is a plot illustrating the distance of propagation
as a function of the phase of the wave at the frequency of 80 Hz
during the beginning of systole transient motion in accordance with
the present disclosure.
[0046] FIGS. 26(a)-(e) illustrate a sequence of axial displacement
maps overlaid to the grayscale image (0.12 ms between successive
frames) indicating an electromechanical wave propagating in the
posterior wall of the mouse from the apex towards the base during
pacing in the right atrium close to the sinoatrial node in
accordance with the present disclosure.
[0047] FIGS. 26(f)-(j) illustrate the ECG signal plotted below each
respective image of FIGS. 26(a)-(e) indicating the time t of the
acquisition during the cardiac cycle in accordance with the present
disclosure.
[0048] FIGS. 27(a)-(e) illustrate a sequence of axial displacement
maps overlaid to the grayscale image (0.07 ms between successive
frames) indicating an electromechanical wave propagating in the
posterior wall of the mouse from the base towards the apex during
pacing in the right ventricle close to the base in accordance with
the present disclosure.
[0049] FIGS. 27(f)-(j) illustrate the ECG signal plotted below each
respective image of FIGS. 27(a)-(e) indicating the time t of the
acquisition during the cardiac cycle in accordance with the present
disclosure.
[0050] FIGS. 28(a)-(f) illustrate a sequence of axial displacement
maps overlaid to the grayscale B-mode image of the aorta taken
every 0.7 ms. Sequence of images showing the propagation of the
pulse wave in the aorta. The arrows indicate the progression of the
wave front in the aorta
[0051] FIGS. 28(g)-(l) illustrate the ECG signal plotted below each
respective image of FIGS. 26(a)-(f) indicating the time t of the
acquisition during the cardiac cycle in accordance with the present
disclosure.
[0052] FIG. 29 illustrates the distance of propagation as a
function of the phase of pulse wave at the frequency of 80 Hz. The
slope of the curve gives the pulse wave velocity in accordance with
the present disclosure.
[0053] FIGS. 30(a)-(c) illustrates examples of high temporal
resolution strains during different types of arrhythmia in
accordance with the present disclosure.
[0054] FIGS. 31(a)-(d) illustrates examples of propagating
electromechanical activation in atria of normal subjects and a
patient undergoing focal atrial tachycardia in accordance with the
present disclosure.
[0055] FIGS. 32(a)-(d) illustrates analysis of reentrant
arrhythmias using a single-frequency atrial flutter case using a
peak cycle length map, cycle length histogram, a phase map, and
electrogram in accordance with the present disclosure.
[0056] FIG. 33(a) illustrates peak cycle length maps in two atrial
flutter patients in accordance with the present disclosure. FIG.
33(b) illustrates a correlation between mechanical cycle length
(MCL) and electrical cycle length in five atrial flutter patients
in accordance with the present disclosure. FIG. 33(c) illustrates a
peak cycle length map during atrial fibrillation in accordance with
the present disclosure. FIG. 33(d) illustrates a histogram
depicting spatial fragmentation of peak cycle length during atrial
fibrillation in accordance with the present disclosure.
[0057] Throughout the figures, the same reference numerals and
characters, unless otherwise stated, are used to denote like
features, elements, components or portions of the illustrated
embodiments. Moreover, while the present disclosure will now be
described in detail with reference to the figures, it is done so in
connection with the illustrative embodiments.
DETAILED DESCRIPTION
[0058] The system and methods described herein can be useful for
analyzing data obtained by an image generating device, such as an
ultrasound transducer. The systems and methods can also be useful
for measuring mechanical properties and estimating the electrical
characteristics of a body tissue structure or organ, such as, for
example, the myocardium or the aorta.
[0059] For example, the disclosed subject matter can be used in
connection with imaging and characterizing the propagation of
electromechanical waves in the heart. During the cardiac cycle,
electrical waves propagate in the myocardium in order to induce its
contraction. Contraction of the myocardial fibers results in a
strong mechanical wave, which, since it results from the coupling
of the electrical excitation and the mechanical properties of the
myocardium, is referred to herein as an "electromechanical wave."
The speed of this wave is a function of the electrical and
mechanical properties of the myocardium, and, according to the
present disclosure, can be used to detect changes in these
properties to diagnose heart diseases.
[0060] An exemplary embodiment of the system is illustrated in FIG.
1 and designated system 100. System 100 can include an image
detection device, such as ultrasound probe 102, which is used to
create images of the heart H or other organ or structure of the
patient P. The image detection device does not induce discernible
vibration in the body structure, and merely detects pre-existing
motion. The signals detected by the probe 102 can be transferred to
an ultrasound scanner 104. The exemplary embodiments described
herein are designed to work with conventional ultrasound scanners.
For example, commercial portable scanners, such as Terason 2000,
high frequency scanners, such as Visualsonics Vevo 770, and
routinely used clinical scanners, such as GE System Five or GE
Vivid Five or Seven, are useful for image acquisition in accordance
with the exemplary embodiments. The raw data produced by the
scanner 104 can be transferred to a computer 106 having a CPU 108
for processing the data. In the exemplary embodiment, the computer
and CPU would be Dell PC with a 2 GHz processor. It is understood
that the computer and CPU can also be integrated with the
ultrasound scanner 104. Also useful in the system would be storage
such as disk drives, for storing data on input files 110 and for
writing output onto output files 112. As will be described herein,
input files 110 can include information such as thresholds. Output
files 112 can include the displacement maps, videos of myocardium
displacements, or computed data, such as electromechanical wave
properties. It is understood that a preprogrammed chip can be used
to execute the algorithms described herein. Typically, an output
device, such as monitor 114, and an input device, such as keyboard
116, are also components of the system.
[0061] In accordance with an exemplary embodiment, the methods
described herein are particularly useful for imaging the
propagation of electromechanical waves in the heart. A method for
detecting the properties of the electromechanical wave are
described herein and represented in FIG. 2. In an early stage in
the procedure, raw imaging data of the body structure is acquired
by image acquisition equipment such as the ultrasound probe 102 and
scanner 104. In the exemplary embodiment, a set of N frames of raw
ultrasound data of the heart is acquired during a cardiac cycle at
high frame rate, e.g., higher than 100 fps, although frame rates of
about 56 fps and 170 fps, etc., yield useful results (202). The
selected frame rate should be commensurate with the speed of the
propagation of the movement, such as the wave, being studied. The
electrocardiogram (EKG) can also be recorded. The raw data can be
digitized and stored in real-time in the scanner memory.
[0062] In a subsequent stage, the data can be transferred to a
computer for processing (204). In an exemplary embodiment, the
transfer can occur using a protocol such as Ethernet TCP IP. This
is optional, as the computer can be integrated with the scanner
104.
[0063] At 206, the raw data received from the image acquisition
equipment is processed. In the exemplary embodiment, the data
processing computes an estimation of the displacement of particular
objects in the images, such as the myocardium, between consecutive
frames. Typically this processing occurs off-line; however, it is
understood that this procedure can occur sequentially subsequent to
receiving two consecutive frames. According to the exemplary
embodiment, axial displacements (in the direction of the
transducer) are computed. Lateral, or elevational, displacements
(perpendicular to the transducer) can also be computed using a
similar technique, for example, as disclosed in Konofagou E. E. and
Ophir, J., (1998), A New Elastographic Method for Estimation and
Imaging of Lateral Strains, Corrected Axial Strains and Poison's
Ratios in Tissues," Ultrasound in Medicine and Biology 24(8),
1183-1199 (1998); Konofagou et al. (1998), Three-dimensional Motion
estimation in Elastography, IEEE Proceedings of the Symposium of
Ultrasonics, Ferroelectrics and Frequency Control in Sendai, Japan,
1745-1748. Korteweg, D. Uber die Fortpflanzungsgeschwindigkeit des
Schalles in elastichen Rohren. Ann. Phys. Chem. (1879) 5: 525-37,
the contents of which are incorporated herein.
[0064] N-1 displacement 2D maps (also referred to as correlation
matrices) are computed through the correlation of two consecutive
frames i and i+1 (1<i<N-1). Each frame is represented by a
matrix of pixel values. The displacement maps provide an indication
of the local axial movements between frames. Estimation of the
axial displacements from the two consecutive frames is performed
using a speckle tracking algorithm. In this algorithm, the
time-shifts in the backscattered signals are determined between two
consecutive frames through cross-correlation of small sliding
windows over the entire ultrasound image. For each window, the
signal of the frame i and the frame i+1 are cross-correlated. The
maximum of the correlation coefficient gives an estimation of the
time-shift between the two signals. This time-shift can be
converted to a spatial displacement by assuming a constant speed of
sound for the tissue. This technique can detect displacements on
the order of 10 .mu.m. Using small correlation windows of 7.5 mm,
the resolution of the displacement maps is in the millimeter range.
The cross-correlation algorithm suitable for estimating
displacement between consecutive image frames is described in U.S.
Provisional Patent Application No. 60/619,247, filed Oct. 15, 2004,
which is incorporated by reference herein. In the exemplary
embodiment, a Matlab program Multiframe is used to compute the
displacement maps for the complete sequence of frames obtained at
202, above. Multiframe calls the Matlab routine FunCalculDispl to
compute the displacements for the sequence of frames.
FunCalculDispl in turn calls the routine Correlation2D.cpp which is
a C program that computes the displacement map between consecutive
frames. As discussed above, Correlation2D.cpp uses small sliding
windows to find the displacement which maximizes the correlation
coefficient for each part of the image. In accordance with other
embodiments of the present disclosure, auto-correlation
calculations or coherence calculations, as are known in the art,
can be performed.
[0065] Two optional threshold procedures can be executed in the
procedure 200. At 208, a threshold can be applied on the energy of
the signal, in order to remove the noise that is below a
predetermined signal-to-noise ratio. Low energy ultrasound signals
(e.g., noise in the cavity of the heart) can be removed from the
displacement map according to this method. At 210, a threshold can
be applied on the correlation coefficient to remove erroneous
estimates in the displacements. In the exemplary embodiment, the
noise threshold and correlation-coefficient threshold can be
implemented within the routine Correlation2D.cpp. The levels of the
thresholds are determined experimentally and can be stored in an
input data file 110 for processing on the CPU 108. Procedures 206,
208 and 210 are illustrated sequentially; however, it is understood
that they can occur simultaneously or any other order to
appropriately process the data. Moreover, one or more of these
procedures can be omitted from the process described herein.
[0066] A video of the sequence of N-1 displacement maps can be
assembled to create a video of the displacements of the body
structure or tissue (212). In the exemplary embodiment, a video of
the myocardium displacements is created by this technique.
[0067] The video of the displacement map of the myocardium will
depict the propagation of the electromechanical wave. Next, an
observation and tracking of the wave propagation (214). Although
such tracking can be done manually, it can be difficult to discern
the wave by visual observation and thus make accurate measurements.
Accordingly, wave tracking can be performed by an algorithm, such
as TrackPositionWave, a Matlab program which locates the position
of the wave front by performing a zero-crossing calculation on
consecutive displacement maps.
[0068] The parameters of the electromechanical wave, e.g.,
velocity, amplitude, attenuation, frequency, etc., can be analyzed
at 216. For example, the velocity of the electromechanical wave can
be computed as a function of its position in the myocardium. As
illustrated in FIG. 3, the wall of the myocardium is approximated
as circular with a radius R, and the origin of the spherical
coordinate system was chosen at the center of the cavity. The
wavefront of the electromechanical wave was then tracked by its
angular coordinate .theta.. The Matlab function Overlay can be used
to compute the transformation of the raw image into polar
coordinates. This routine can also display the displacement map
superimposed on the ultrasound grayscale data. As an example, the
speed of the electromechanical wave is shown on the FIG. 4 as a
function of the angular position.
[0069] The ultrasound imaging method described herein has the
advantage of being completely non-invasive. In an exemplary
embodiment, the system described herein can be implemented in
real-time on commercial scanners. It has been shown that the
electrical conductivity is transversely isotropic with respect to
fiber direction, with a longitudinal velocity of about 0.6 m/s and
a transverse velocity of about 0.2 m/s (Roth, B. J. (2000),
Influence of a perfusing bath on the foot of the cardiac action
potential. Circulation Research 86, E19-E22; Spach, M. S.,
Heidlage, J. F., Dolber, P. C., and Barr, R. C. (1998),
Extracellular discontinuities in cardiac muscle--Evidence for
capillary effects on the action potential foot. Circulation
Research 83, 1144-1164). The electromechanical wave velocity noted
herein was very close to the longitudinal velocity of the
mechanical wave. The transverse velocity can be measured by using
ultrasound imaging and displacement estimation using a 3D imaging
probe or a rotational 2D imaging probe. FIG. 5 illustrates a
transducer setup for 2D imaging of the longitudinal waves, and FIG.
6 illustrates a transducer setup for 2D imaging of transverse
waves.
[0070] The mechanical component of the electromechanical wave is
related to the viscoelastic properties of the soft tissue. The
elastic properties of the myocardium have been widely investigated.
The stiffness of the myocardium has been shown to increase during
ischemia and recovers after reperfusion. Thus, early detection of
cardiovascular diseases such as ischemia and infarction can be
strongly improved through non-invasive characterization of the
local myocardial elasticity.
[0071] Low frequency shear (mechanical) waves propagate in soft
tissue at low velocity (0.5 to 50 m/s). For an isotropic and
infinite medium, it has been shown that the velocity of the shear
wave is related to the shear modulus p and the density p by:
V S = .mu. .rho. ( 1 ) ##EQU00001##
(Bercoff, J., Tanter, M., and Fink, M. (2004), Supersonic shear
imaging: A new technique for soft tissue elasticity mapping. IEEE
Transactions on Ultrasonics Ferroelectrics and Frequency Control
51, 396-409; Sarvazyan, A. P., O. V. Rudenko, S. D. Swanson, J. B.
Fowlkes and S. Y. Emelianov, Shear wave elasticity imaging: A new
ultrasonic technology of medical diagnostics. Ultrasound Med Biol
(1998) 24(9): 1419-1435.) According to another exemplary embodiment
of the disclosed subject matter, a system can be implemented to
provide early detection of ischemia through the measure of the
velocity of the mechanical wave.
[0072] However, the myocardium has also anisotropic mechanical
properties and can be considered as a transverse isotropic medium.
As a consequence, two shear waves of different velocities can
propagate in the myocardium. Fast mechanical (shear) waves
propagate in the direction of the fibers, and slow mechanical
(shear) waves propagate in the direction perpendicular to the
fibers. The measure of the two wave velocities can be achieved by
using 3D Ultrasound imaging systems or multiple acquisitions of 2D
images with a rotation of the transducer (see FIGS. 5-6). The wave
velocities are related to two elastic constants, .mu..sub.// the
shear modulus in direction of the fibers and .mu..sub..perp. the
cross-fiber shear modulus.
V // = .mu. / / .rho. ( 2 ) V .perp. = .mu. .perp. .rho. ( 3 )
##EQU00002##
[0073] The systems and methods described herein can potentially
have different applications in the field of early detection of
cardiovascular diseases and cardiac imaging.
[0074] For example, the measure of the electrical excitation
propagation is of high interest in cardiology for early detection
of heart diseases but also for pacing the heart when heartbeat is
too slow or irregular. The purpose of an artificial pacemaker is to
stimulate the heart when either the heart's natural pacemaker is
not fast enough or if there are blocks in the electrical conduction
system preventing the propagation of electrical impulses. Thus, in
order to implant the artificial pacemaker at the correct location,
the electrical propagation must be determined accurately. In vivo
imaging of the electrical propagation in the heart can require
implanting an electrode matrix (up to 500 electrodes) to measure
extracellular potentials at the surface of the heart. This invasive
and potentially precarious surgical procedure cannot be performed
on human patients for diagnostic purposes. The present disclosure
provides a means for determining electrical propagation in the
myocardium of a subject, in the context of achieving an effective
position of a pacemaker in the subject. Other methods known in the
art involve optically-based techniques which also require invasive
procedures, such as open-chest surgery.
[0075] The present disclosure can be further used to create images
and thereby detect myocardial ischemia in a subject either having
symptoms (e.g., chest, arm, or chin pain, nausea, shortness of
breath, and/or sweating) or a subject subjectively lacking such
symptoms (e.g., "silent ischemia"), whereby a finding of increased
electromechanical wave velocity (relative to control values) in a
region of the myocardium of a subject is consistent with and
supportive of a diagnosis of myocardial ischemia in that region.
The present disclosure can also be used to diagnose, or assist in
surgical intervention in, (i) conduction disturbances, such as
re-entry phenomena, or associated with pharmaceutical agents, such
as antidepressants or hyperkalemia, (ii) arrhythmias and
dysrhythmias (e.g., surgical treatment of ventricular dysrhythmias,
diagnosis of low-amplitude atrial fibrillation); and (iii) tissue
abnormalities associated with cardiomyopathies or trauma, etc.
Example A--Imaging of Canine Heart
[0076] The procedure described hereinabove was performed in an
anesthetized open-chested dog. The transducer was placed on the
anterior wall of the left ventricle of the heart, to obtain a short
axis view. Approximately every two minutes, a sequence of three
cardiac cycles was acquired during the experiment, with a frame
rate of 56 fps. The 2D displacement maps were estimated using the
cross-correlation method (window size: 5 mm, 90% overlap). The
axial displacements were processed for the different sequences. On
the displacement video, two electromechanical waves were clearly
detected, propagating in the posterior wall of the left
ventricular, from the septum (left side of the images) to the
lateral wall (right side). The propagation of the mechanical wave
corresponds to the electrical activity shown on an associated
EKG.
[0077] The first electromechanical wave is found at the
end-diastolic phase of the cardiac cycle (which corresponds to the
beginning of the contraction). FIGS. 5(a) through 9(a) show five
consecutive frames of the propagation of the wave. The location of
the electromechanical wavefront is indicated by arrow W in each of
FIGS. 7(a)-11(a). The displacements maps are overlaid to the
grayscale ultrasound images (see, FIGS. 7(b)-11(b)). Blue
displacements are in the direction of the transducer (top of the
image), and red displacements are in the opposite direction. As
shown in these images, the contraction of the myocardium starts on
the left side (septum) and propagates to the right side of the
image. In the figures, the blue region appears on the left side of
the images (behind the wavefront), and the red region appears on
the right side of the image (in front of the wavefront). The
maximum displacements shown are 75 .mu.m (dark blue and dark red),
and the wave propagates within a few milliseconds. Therefore it is
impossible to visually detect this electromechanical wave on the
grayscale images. The wave speeds as measured using the techniques
described above are represented in FIG. 4. The wave velocity was
found to be approximately 0.6 m/s in the posterior wall, which was
corroborated by invasive electrophysiological measurements using a
matrix of electrodes. Temporary regional ischemia was then induced
by coronary artery ligation. The velocity of the electromechanical
wave was found to increase up to approximately 1.7 m/s in the
ischemic region. Although not entirely understood, this strong
increase is believed to be due to an increase of the shear modulus
in the ischemic region or a change in the conduction velocity, or
both. (A second electromechanical wave has also been detected at
the end-systole phase. However, due to its high propagation speed
(related to the high contraction of the myocardium), the
propagation was not caught with a sufficiently high frame rate.
Some evidences of its propagation are detected in the human
experiments, described herein.)
Example B--Imaging of Human Heart
[0078] The procedure 200 described hereinabove was performed on a
young healthy patient. The transducer was placed on the patient's
thorax in order to image the heart in the short axis view. A
sequence of approximately four cardiac cycles was acquired at a
very high frame rate of 170 fps using a Vingmed System Five for RF
image acquisition. In order to reach such a high frame rate, only a
small part of the heart (the left ventricle) was imaged
(80.times.40 mm). The axial displacements were processed for each
frame. On the displacement video, 2 electromechanical waves were
clearly seen, propagating in the posterior wall of the left
ventricular (not shown). FIGS. 12(a)-18(a), which are consecutive
displacement maps superimposed on the grayscale images (FIGS.
12(b)-18(b)), illustrate the propagation of the electromechanical
wave at the end-systole phase. The speed was found to be 0.65 m/s
in the posterior wall. The location of the electromechanical
wavefront is indicated by arrow W in each of FIGS. 12(a)-18(a).
Example C--Imaging of Cardiovascular Tissue in Mice
Animal Preparation
[0079] The procedure described hereinabove was performed on
anesthetized mice. The mice were anesthetized with tribromoethanol.
The hair was removed using potassium thioglycolate and the mouse
was placed in the supine position on a heating stage (VisualSonics,
Toronto ON, Canada) in order to keep the body temperature steady.
ECG signal was obtained from the extremities. The ultrasound probe
was placed on the chest or the abdominal wall using degassed
ultrasound gel (Aquasonic 100, Parker Laboratories Inc., Fairfield
N.J., USA) as a coupling medium.
RF Signal Acquisition
[0080] An ultrasound scanner specifically developed for imaging
small animals (Vevo 770, Visualsonics, Toronto ON, Canada) was used
in this exemplary embodiment. The high frequency ultrasound probe
was composed of a single focused transducer working at 30 MHz, with
a focal depth of 12.7 mm. The transducer was mechanically rotated
and real-time 2D images could be acquired at a frame rate of up to
60 Hz. The field of view was 12.times.12 mm, the axial resolution
was 50 microns, and the lateral resolution was 100 microns.
[0081] A digitizer (2 channels, 200 MS/s, 14 bits, CS14200, Gage
Applied Technologies, Lachine QC, Canada) mounted on a PC computer
slot was connected to the analog RF-output of the ultrasound
scanner. In addition, two TTL outputs were used to trigger the
digitizer on the 2D frames. This setup allows the real-time
acquisition of more than one thousand 2D RF-data, e.g., images.
[0082] In the exemplary embodiment, the ultrasound probe was placed
on the chest in the parasternal position to obtain a longitudinal
(long-axis) view of the left ventricle of the heart. The probe
could also be positioned over the abdomen to obtain a longitudinal
view of the abdominal aorta.
Frame Rate Acquisition
[0083] In addition to the real-time scanning mode, a high frame
rate acquisition mode (EKV) was provided on the scanner in the
exemplary embodiment in order to allow detailed visualization of
the heart contraction. The equipment can operate as quickly as 8000
frames per minute, although the user can see 1000 frames per minute
due to dropped calls. Using this technique, the ultrasound
acquisition of each RF-line was triggered on the mouse ECG. The
transducer was slowly rotated and for each position of the
transducer, ultrasound echo signals were recorded with a pulse
repetition frequency (PRF) of 8000 pulses/s during several cardiac
cycles. The ECG was simultaneously recorded and thus allowed for
the synchronization of the RF-lines based on the R-wave peak, a
reliable peak of the ECG during the cardiac cycle. The complete
acquisition duration was approximately 5 min.
[0084] To compute the tissue motion, RF-signals and ECG signals
were digitized during the EKV acquisition and transferred to the
computer in real-time. The data were then processed off-line,
RF-lines were synchronized using the R-wave peak of the ECG signal,
and a complete set of 2D ultrasound RF-data was reconstructed at
8000 fps for one complete cardiac cycle (approximately 150 ms).
Motion Estimation
[0085] The motion of the tissue was estimated off-line using a
well-known classical speckle tracking algorithm (Bonnefous, O. and
P. Pesque. Time domain formulation of pulse-Doppler ultrasound and
blood velocity estimation by cross correlation. Ultrason Imaging
(1986) 8(2): 73-85). This technique was based on detecting the
small local displacements of the tissue that occur between two
consecutive frames. With the current method, only axial
displacements (in the direction of the transducer) were computed.
In this algorithm, the time-shifts in the backscatterered signals
were determined between the two consecutive frames through
cross-correlation of small sliding windows over the entire
ultrasound image. This technique allowed the detection of very
small displacements on the order of 1 .mu.m or less (correlation
windows of 150 .mu.m, overlapping 90%). Finally, the movie of the
axial displacements was processed at a frame rate up to 8000
frame/s for the entire cardiac cycle. It is understood the lateral
displacement can be obtained using the same technique.
Frequency Analysis
[0086] The axial displacements were analyzed in the frequency
domain as a function of the time. A sliding Blackman window (100
points, 25 ms) as is known in the art, was moved along the
displacement variation at a fixed depth, in steps of 2 ms. The
windowed signals were zero-padded to 8192 points and their FFT was
calculated. The frequency content of the displacements was
evaluated graphically by plotting these spectra as a function of
time. Based on this frequency analysis, the transient and the slow
motions of the tissues were separated using a digital filter. The
displacement estimates were temporally filtered using an FIR
band-pass filter with cut-frequencies of f.sub.1=50 Hz and
f.sub.2=500 Hz, which allows the removal of the low frequency
components but also the high frequency noise.
Wave Velocity
[0087] To analyze the propagation of the mechanical waves, the
phase velocity of the vibration was determined for an angular
frequency co. The wave was assumed to propagate with a velocity c
in a direction r that was arbitrarily determined on the image by
the direction of the wall, and a set of measurement points was
selected on this direction. The wave number is k=.omega./c, and the
phase of the wave is .phi.(r)=kr along the direction of
propagation. The phase was measured as a function of the
propagation distance r, using the Fourier Transform of the temporal
displacements at the location r computed at the angular frequency
co. Finally, the derivative of the phase of the wave with respect
to distance was estimated using a linear regression fit on the set
of measurements points, and the velocity of the wave at the
frequency f was calculated:
c(f)=2.pi.f/(.differential..phi./.differential.r) (4)
Modulus Estimation
[0088] The theory of elastic wave propagation in soft biological
tissue was considered in order to derive the Young's modulus of the
tissue. Assuming that the medium is infinite and isotropic, the
speed of shear waves propagation could be derived from general
equations of the dynamic theory of elasticity. However, it is
understood that the propagation of elastic waves in the myocardium
can optionally take into account additional characteristics such as
the active properties of the muscle fibers, the strong anisotropy
of the tissue, and/or the geometry of the ventricles.
[0089] For the transverse wave on the artery wall, a simple model
of the propagation of a pressure wave in a viscoelastic infinite
thin conduit filled with an incompressible fluid is well described
by the Moens-Korteweg equation:
c = E h 2 R .rho. ( 5 ) ##EQU00003##
where c is the velocity of the wave, E is the Young's modulus of
the conduit wall, h is the wall thickness, p is the density of the
fluid and R the radius of the tube. According to this equation, the
elasticity of the vessel wall can be derived from the measurement
of the pulse wave velocity in the artery.
Results of Example C
In Vivo Cardiac Imaging
[0090] FIG. 19 shows a B-mode image 1910 of a typical parasternal
long-axis view obtained in a normal mouse. Image 1910 shows the
main structures of the left ventricle: the intraventricular septum
1912, the cavity of the left ventricle 1914, the papillary muscle
inside the cavity 1916 and the posterior wall 1918 which is visible
due to strong reflections at the epicardium-lung interface. Also
shown in image 1910 is the right ventricle 1920, aortic valve 1922,
aorta 1924, mitral valve 1926, and left atrium 1928. In this
embodiment, the duration of the average cardiac cycle was 138 ms.
Axial displacements were estimated for the complete set of data. In
order to keep the displacements at appropriate magnitudes for the
estimation (on the order of 1 .mu.m and to reduce the amount of
data, the number of frames was halved, which also reduced the frame
rate to 4000 fps.
[0091] FIGS. 20(a) and 20(b) show the color-coded axial
displacements overlaid onto the grayscale B-mode image for two
different phases of the cardiac cycle. During the systolic phase,
the contraction of the myocardium is shown by positive
displacements (red region) of the posterior wall 2018 and negative
displacements of the septum 2012 (blue region) (FIG. 20(a)). In the
diastolic phase, the directions of the displacements of the
posterior wall 2018 and the septum 2012 (and the colors associated
with the direction of movement) are reversed during the relaxation
(FIG. 20(b)). It should be noted that even if a large part of the
myocardium of the posterior wall is not visible, the motion of the
epicardium undergoes similar motion. The time of acquisition of
FIG. 20(a) is indicated at point t of FIG. 20(c). The time of
acquisition of FIG. 20(b) is indicated at point t of FIG.
20(d).
[0092] A temporal analysis of the motion was performed for single
RF lines of the image. The axial displacement along one central
line of the image (indicated by the white, dotted vertical line
2050 on FIG. 20(b)) is shown as a function of time in FIG. 21(a)
with the corresponding ECG signal (FIG. 21(d)). (FIGS. 2(a)-(d) are
aligned on a temporal basis.) On this line 2050, the displacements
of the septum, the papillary muscle and the posterior wall are
shown in a M-mode format over two cardiac cycles. It shows the
successive main phases of the cardiac cycle: the contraction of the
myocardium (systole) indicated by arrow 2120 initiated at the
R-wave peak of the ECG, followed by the relaxation phase (diastole)
indicated by arrow 2130. The duration of the active contraction was
approximately 50 ms, and that of the relaxation 35 ms. In addition
to this slow and large motion, some rapid transient variations of a
few ms were observed at the beginning and at the end of the
systolic phase, in the septum and the posterior wall.
[0093] In order to separate the electromechanical wave from other
mechanical waves generated by vibrations resulting from valve
functions or blood flow, high-pass filtering was performed. The
frequency content of the tissue displacements resulting from
vibrations in the septum (at depth of 12.5 mm) was analyzed as a
function of time and is shown in FIG. 21(b). During the contraction
and the relaxation of the heart, the motion of tissue was found to
be in the low frequency range of up to 60 Hz. However, during the
transient motion at the end of systole, much larger frequency
components were found that ranged between 50 Hz and 500 Hz. The
same effect was found for the transient motion at the beginning of
systole, but the frequency range was limited between 50 Hz and 250
Hz. Thus, it was possible to almost completely separate the
transient part of the displacement by filtering out the low
frequency component of the motion. After filtering the
displacements using a FIR band-pass filter with cut-off frequencies
f.sub.1=50 Hz and f.sub.2=500 Hz, the two vibrations were clearly
visible and are shown on FIG. 21(c) as regions 2150 and 2152. These
rapid variations occurred within less than 3 ms around the
beginning of systole and end-systole.
End of Systole
[0094] In order to analyze spatially the vibration around
end-systole, we considered the data between 52 ms and 70 ms after
the peak of the R-wave. FIGS. 22(a)-(d) show a sequence of axial
displacements overlaid onto the grayscale B-mode images every 0.6
ms around end-systole. This sequence uncovers a strong mechanical
wave W propagating in the longitudinal direction of the ventricle
along the myocardium, from the base (right side of the images) to
the apex (left side). In other words, as the tissue locally
vibrates along the axial direction of the beam (i.e., along the
beam axis), a transverse wave propagates along the lateral
direction (i.e., in-plane, perpendicular to the beam axis). A
second wave W' is shown in FIGS. 22(e)-(f).
[0095] The mechanical wave, i.e., generated by localized vibrations
in the muscle (FIG. 21(c)), was visible both in the posterior wall
and the septum. Its amplitude was found to be eight times higher in
the septum. Only the mechanical wave propagating in the septum is
described herein. A set of 60 samples was selected in the septum
along the propagation direction (lateral direction of the image),
and the phase of the wave was computed at different frequencies.
Three frequencies were selected for which the displacement
amplitude was large enough to detect, e.g. with respect to noise
level: (*) 82 Hz (.circle-solid.) 246 Hz (.diamond.) 410 Hz. The
phase velocity of the wave was computed for these frequencies and a
large dispersion was found. The distance of propagation was plotted
in FIG. 23 as a function of the phase of the wave divided by the
angular frequency. The phase velocity was found to be 1.20 m/s at
82 Hz, 3.02 m/s at 246 Hz and 4.21 m/s at 410 Hz.
Beginning of Systole
[0096] The same analysis was performed at the beginning of systole.
The filtered data were processed between 0 ms and 20 ms from the
peak of the R-wave. FIGS. 24(a)-(f) show a sequence of axial
displacements overlaid to the grayscale B-mode images every 2.8 ms
around the beginning of systole. A strong vibration was found in
the septum, but no wave propagation was visible in the image plane.
Therefore, a mechanical wave can propagate in the perpendicular
direction, but was not being observed with the equipment described
herein.
[0097] However, the FIG. 24 shows a wave propagating in the
posterior wall (see the white arrows W). The displacements were
initiated at the apex (left side of the images) and then propagated
towards the base (right side). The phase velocity was determined
using the method previously described at the frequency of 80 Hz.
The distance of propagation was plotted in FIG. 25 as a function of
the phase of the wave divided by the angular frequency. The phase
velocity of the wave was obtained using a linear regression fit and
was estimated to be 0.44 m/s.
Imaging Under Different Electrical Pacing Conditions
[0098] In order to determine that the origin and direction of the
wave W were electrically induced and driven, mice were also scanned
during right-atrial pacing (at 90 ms corresponding to a heart cycle
at sinus rhythm of 100 ms period; FIGS. 26(a)-(e)) and
right-ventricular pacing (also at 90 ms; FIGS. 27(a)-(e)). Pacing
was achieved using catheterization through the right side of the
heart, in which the catheter carried nine electrodes that could be
separately activated for varying the pacing location. In some of
the scans, the catheter C was within the imaging field-of-view and
allowed for imaging of the pacing wave during ventricular pacing
(FIGS. 26(c) and 27(c)).
[0099] The most pronounced wave propagating during atrial pacing
was the contraction wave, or wave originating at the isovolumic
contraction phase, that propagated along the longitudinal direction
of the myocardium initiating radial thickening (or, positive (red)
displacement) in its path. At atrial pacing (FIG. 26(a)-(e)), the
contraction wave was very similar to the one during sinus rhythm
(FIGS. 24(a)-(f)), starting at the apex right at the QRS peak and
then propagating along the posterior wall (generally from right to
left in the figure.) Right-ventricular pacing (FIGS. 27(a)-(e))
induced a reverse direction on the contraction wave that now
started from the tip of the catheter (close to the base) with two
waves propagating from base to apex, one along the septum and one
along the posterior wall (generally from left to right in the
figure) (FIG. 27(a)-(e)). Since pacing occurred using the same
mouse, same sonographic view and without affecting the function of
the valves or the blood flow, the reverse direction of the
propagation of the wave is concluded to be induced by the change in
the origin of the electrical stimulus; thereby, confirming that the
wave measured is electrically induced.
In Vivo Vascular Imaging
[0100] A longitudinal view of the abdominal aorta of a mouse was
imaged using the high frame rate technique. Axial displacements
were calculated, and the movie of the motion was processed at 8000
fps for a complete cardiac cycle. During the cardiac cycle, the
displacements of the artery wall were found to be very small except
after the beginning of systole. Strong displacements of the wall
started 10.3 ms after the R-wave peak of the ECG. FIGS. 28(a)-(f)
show a sequence of the axial displacements in color overlaid onto
the grayscale B-mode image. A transverse wave W started propagating
on the right side of the images (heart side) and then propagated
towards the left side in less than 3 ms. This transverse wave was
generated from the sudden pressure change of the blood bolus
traveling through the vessel, known as the arterial pulsive wave
(Nichols, W. and M. F. O'Rourke (1998). Vascular impedance. In
McDonald's: blood flow in arteries: theoretical, experimental and
clinical principles. E. Arnold. London).
[0101] The phase velocity of the pulse wave was computed at the
frequency of 200 Hz. The distance of propagation was plotted in
FIG. 29 as a function of the phase of the wave divided by the
angular frequency, the phase velocity was obtained using a linear
regression fit and was found to be 3.08 m/s. The radius of the
vessel R=0.47 mm and the wall thickness h=0.12 mm were
approximately estimated from the B-mode images, and the blood
density was assumed to be 1060 kg/m.sup.3 (Cutnell, J. and W.
Kenneth (1998). Physics, Fourth Edition. New York). Using these
parameters, the Young's modulus of the aorta wall E=78.8 kPa was
derived from the Moens-Korteweg equation (Eq. 5), which is what has
been typically reported for thoracic aorta moduli in biomechanics
literature (Fung, Y. C. (1993). Biomechanics--Mechanical Properties
of Living Tissues. New York).
Electromechanical Activation of Arrhythmias
[0102] According to another aspect of the disclosed subject matter,
systems and techniques are provided for electromechanical
activation of arrhythmias, including non-transient
electromechanical activation of paroxysmal and periodic arrhythmias
in humans in vivo.
[0103] Certain treatments of cardiac arrhythmias, such as
radio-frequency ablation, can be utilized in clinical practice but
can lack a suitable noninvasive imaging modality to provide insight
into the source or focus of an arrhythmia. Cardiac deformations can
be imaged at high temporal and spatial resolution to elucidate
electrical activation sequences in normal and paced human subjects
non-invasively. In this manner, such imaging can be used to improve
planning and monitoring of ablation-based arrhythmia
treatments.
[0104] Aspects of the disclosed subject matter include techniques
to quantitatively characterize focal and reentrant arrhythmias. For
purpose of illustration and not limitation, and as embodied herein,
spatio-temporal maps of a full-view of the atrial and ventricular
mechanics can be obtained in a single heartbeat. Such maps can
illustrate with suitable detail the electromechanical patterns of
atrial flutter, fibrillation, and tachycardia. For example and
without limitation, during focal arrhythmias, such as premature
ventricular complex and focal atrial tachycardia, the
electromechanical wave imaging techniques can be utilized to
identify the location of the focal zone and the subsequent
propagation of cardiac activation. For purpose of illustration and
not limitation, exemplary electromechanical wave imaging techniques
are described in International Application No. PCT/US13/64377,
filed Oct. 10, 2013, which is incorporated by reference herein in
its entirety. Additionally or alternatively, and as embodied
herein, during reentrant arrhythmias, such as atrial flutter and
fibrillation, Fourier analysis of the strains can show correlated
mechanical and electrical cycle lengths and propagation
patterns.
[0105] For purpose of illustration and application of the disclosed
subject matter, high frame rate ultrasound imaging of the heart can
be used non-invasively and in real time to characterize
lesser-known mechanical aspects of atrial and ventricular
arrhythmias. Such techniques can also be used to assist treatment
planning for intraoperative and longitudinal monitoring of
arrhythmias.
[0106] Certain imaging systems, such as software-based systems can
allow ultra-high frame rates, and thus ultrasound imaging can be
used to allow unprecedented temporal resolution. For example, such
ultrasound imaging systems can provide a five-fold improvement in
the signal-to-noise ratio of cardiac motion and deformation
mapping. For example and without limitation, as embodied herein,
frame rates up to 2000-5000 frames/s can be achieved by using
defocussed transmissions, which can be suitable for depths utilized
in transthoracic cardiac applications. According to exemplary
embodiments of the disclosed subject matter, ultrasound imaging
techniques described herein can be used to map transient strains
occurring in response to the electrical activation, (e.g., the
electromechanical wave). For example and without limitation, and as
embodied herein, such techniques can be used to map transmural
activation sequences of normal and abnormal hearts and to locate
pacing sites in patients undergoing cardiac resynchronization
therapy.
[0107] According aspects of the disclosed subject matter, systems
and techniques are provided to analyze and characterize the
mechanical behavior of the atria. For purpose of illustration and
not limitation, and as embodied herein, systems and techniques
disclosed herein can be utilized to analyze and characterize the
atria during certain types of cardiac arrhythmia, including and not
limitation to, premature ventricular complex, focal tachycardia,
atrial flutter, and atrial fibrillation. While Electromechanical
Wave Imaging (EWI) can be suitable to characterize focal rhythms
such as premature ventricular complex and focal tachycardia, EWI
can have difficulty describing and/or characterizing reentrant
rhythms such as atrial flutter and fibrillation. Accordingly,
systems and techniques described herein are provided to
characterize electromechanical strains, including and without
limitation, during reentrant rhythms based on Fourier analysis.
Exemplary embodiments of the disclosed subject matter can include a
single acquisition sequence of electromechanical activation mapping
that can be used for standard EWI and/or for Fourier analysis of
electromechanical strains. Electromechanical activation mapping can
characterize electromechanical strain propagation patterns during
both focal and reentrant arrhythmias. In this manner, systems and
techniques described herein can determine that local deformations
of the atria can be closely correlated with their electrical
activation. As such, systems and techniques described herein can be
used to determine characteristics of cardiac mechanics in
arrhythmia, to plan ablation treatments, and to monitor the
efficacy of such treatments non-invasively, longitudinally and in
real-time.
Example D--Imaging Electromechanical Activity During Arrhythmia
[0108] For purpose of illustration and confirmation of the
disclosed subject matter, exemplary techniques for imaging
electromechanical activation of arrhythmias are described. The
systems and techniques described herein can be performed, for
purpose of illustration and not limitation, on human subjects. The
human subjects can undergo a diagnostic ultrasound scan, and as
embodied herein, can occur a few minutes to a few hours prior to
electroanatomic mapping and ablation. The cardiac arrhythmias of
the patients can be confirmed during electroanatomic mapping and
ablation to be, for example and without limitation, one or more of
premature ventricular complex (n=1), atrial flutter (n=5), focal
atrial tachycardia (n=1), and atrial fibrillation (n=1).
Additionally, a normal human subject can be imaged as a control for
purpose of comparison.
[0109] Additionally, and as embodied herein, strain maps can be
generated, for example and without limitation using similar
techniques as described herein for single-heartbeat
electromechanical wave imaging (EWI). For example and without
limitation, as embodied herein, a Verasonics system with a 2.5-MHz
probe can be calibrated and customized to adhere to FDA standards,
including measurements of mechanical index and of peak
spatio-temporal-average intensity. The Verasonics system can be
calibrated to have an acoustic power output that is similar to
conventional clinical scanners. Such calibration can be performed
by measuring the peak pressure and/or intensity (e.g., spatial-peak
temporal average intensity, also referred to as Ispta) of the
Verasonics system to ensure that its mechanical index (MI) is
within FDA guidelines. The ultrasound scan can include two
sequences. As embodied herein, in a motion-estimation sequence, a
circular ultrasonic wave can be emitted with a virtual focus of
10.2 mm behind the probe at 2000 fps during 2 seconds. Additionally
or alternatively, as embodied herein, a standard B-mode acquisition
can be performed during 1.5 seconds to depict the heart anatomy.
Frames from the motion-estimation sequence can be reconstructed by
generating a plurality of beams, for example and as embodied herein
128 beams, in post-processing using a delay-and-sum algorithm with
a reconstructed sampling frequency of 20 MHz. As embodied herein,
the motion-estimation rate and the motion-sampling rate can be set
to 1000 and 2000 fps, respectively. The window for
motion-estimation can be 9.2 mm with an overlap of 95.8% (window
shift of 0.3 mm), and the kernel strain estimation can be set to
4.9 mm. For purpose of illustration and not limitation, and as
embodied herein, the techniques described herein for beamforming,
motion-estimation, strain estimation, spatial moving-average of the
strains (12 mm by 10 lines), and the automated contour tracking
technique can be performed off-line on a graphics processing
circuit (embodied herein as a Tesla graphics processing unit) and a
Matlab parallel processing toolbox at a computing speed of 2.4
frames/s.
[0110] Furthermore, and as embodied herein, focal and reentrant
arrhythmias can be analyzed differently for patients with different
types of rhythms. For purpose of illustration and not limitation,
FIG. 30A illustrates strains mapped in subjects having sinus
rhythm. FIG. 30B illustrates strains mapped in subjects having
atrial flutter. FIG. 30C illustrates strains mapped in subjects
having atrial fibrillation.
[0111] As illustrated for example in FIG. 30A, for subjects with
sinus rhythm, the strains in one location, (e.g., one pixel in the
left atrium) can present two representative events over time that
correspond approximately to the beginning and the end of systole:
end-systole, and end-diastole. By tracking the onset (e.g., the
first zero-crossing) of these representative events for each pixel
of the heart walls, isochrones maps can be generated. Isochrones
correlated to electrical isochrones can be obtained, for example
and without limitation, by tracking the propagation front of the
end-diastole electromechanical activation. As illustrated for
example in FIG. 30B, for atrial flutter patients, a similar
location in the left atrium (LA) illustrates that strains can be
periodic, which, in some cases, can be represented by a single
frequency. Alternatively, as illustrated for example in FIG. 30C, a
plurality of frequencies can be observed in a patient with atrial
fibrillation, and as such, analysis based on the Fourier transform
might can be utilized. During atrial fibrillation, the strains can
be chaotic and no period of zero strains, similar to the one in
FIG. 30A, can be observed.
[0112] In addition, and as embodied herein, the onset of
contraction can be determined, for example in subjects who have
focal rhythms such as sinus rhythm and focal tachycardia, as the
first zero-crossing of the incremental strains occurring after the
onset of the P-wave on the electrocardiogram (ECG), which can
utilize the EWI techniques described herein. Additionally or
alternatively, as embodied herein, in atria with reentrant
arrhythmia, during flutter and fibrillation, a high-resolution
Fourier transform can be performed using a generalized Goertzel
algorithm for interpolation in Fourier space on 1.5-second long
incremental strains signals for each individual pixel in the atria.
For purpose of illustration and comparison with conventional ECG
measurements, and without limitation, frequencies can be converted
to cycle lengths, hereinafter referred to as mechanical cycle
length (MCL). As embodied herein, peak MCL maps can be generated by
selecting the MCL with the highest amplitude within the
physiologically-relevant 100-330 ms range for each pixel.
Additionally or alternatively, and as embodied herein, peak cycle
lengths histograms can be constructed and compared to the
electrical cycle length measured directly during the
electroanatomic mapping and ablation.
Focal Rhythms
[0113] FIG. 31 illustrates an EWI cine-loop and isochrones during
focal rhythms. FIG. 31A illustrates the atria of a normal subject,
with propagation from the right atrium (RA) to the LA. The
electromechanical activation regions 3110 can originate in the
right atrium and propagate towards the left atrium as illustrated
in the exemplary EWI cine-loop in FIG. 31A.
[0114] FIG. 31B illustrates an EWI cine-loop depicting the atria of
a subject undergoing a focal atrial tachycardia, which can have a
focus located high in the left atrium (LA). With reference to FIG.
31B, as embodied herein, electrical mapping of this patient has not
been completed in the LA. The EWI in FIG. 31B illustrates
electromechanical activation 3120 originating high in the LA and
propagating into both atria, and further activation can be detected
in the ventricles.
[0115] FIGS. 31C and 31D each illustrate isochrones obtained from a
patient with ventricular tachycardia. FIG. 31C illustrates an
isochrone of the ventricular tachycardia patient during sinus
rhythm. FIG. 31D illustrates an isochrone of the ventricular
tachycardia patient during premature ventricular complex. EWI was
performed during sinus rhythm and during pre-ventricular
contraction. The EWI isochrones obtained during sinus rhythm, as
shown for example in FIG. 31C, depict propagation from the RA, into
the LA and into the ventricles, as previously shown for purpose of
illustration and comparison in normal patients. When this patient
underwent premature ventricular complex, as illustrated for example
in FIG. 31D, the region that was activated early in the ventricle
during sinus rhythm (e.g., from the ventricles to the atria)
triggered the entire electromechanical activation sequence. During
premature ventricular complex, electromechanical activation can
originate from the lateral wall, and can propagate toward the atria
and into the atria. Early activation of the septum can indicate a
potential recruitment of the Purkinje network.
Reentrant Rhythms
[0116] FIG. 32 illustrates the electromechanical behavior of a
heart undergoing atrial flutter, including the analysis of
reentrant arrhythmias using a single frequency flutter case. FIG.
32A illustrates an exemplary peak MCL map, and as depicted, a
single MCL is representative. The peak cycle length map of FIG. 32A
indicates, for each pixel of the atria, which cycle length is
representative in the Fourier spectrum. FIG. 32B illustrates a
histogram of the cycle length which can be used to determine, among
all the pixels of the atria, which cycle length represents atrial
contraction. With reference to FIG. 32B, as embodied herein, one
peak cycle length of 294 ms can be identified. FIG. 32C illustrates
a phase map analyzing the phase of the MCL of FIG. 32B in Fourier
space. As shown for example in FIG. 32C, as embodied herein, a
propagation pattern originates from the right atrium (RA) near the
tricuspid valve towards the LA. The phase corresponding to the 294
ms cycle length can be retrieved from the Fourier spectrum and used
to map the propagation of the mechanical oscillation at 294 ms. In
this manner, the propagation direction can be determined. With
reference to FIG. 32C, as embodied herein, the electromechanical
activation propagated from the RA to the LA. FIG. 32D illustrates
the corresponding intracardiac electrograms obtained a few hours
after the imaging procedure. With reference to FIG. 32D, as
embodied herein, the electrical cycle length was 283 ms.
[0117] Additionally or alternatively, and as embodied herein,
atrial flutter cases can exhibit different patterns. Indeed,
certain cases presented with two dominant frequencies can be
separated between the left and right atria, whereas certain
electrophysiological data can indicate that only one reentrant
circuit was present.
[0118] FIG. 33A illustrates two examples of such atrial flutter
cases. FIG. 33A illustrates peak MCL maps of two exemplary atrial
flutter patients. As shown for example in FIG. 33A, two
representative frequencies can be identified in each patient, with
the shorter cycle length located in the RA. A relationship between
the MCL and the electrical cycle length can be obtained, for
example and as embodied herein, by performing this analysis in
multiple patients, and choosing the peak cycle length closest to
the electrical cycle length. For purpose of illustration and
confirmation of the disclosed subject matter, such an analysis of
MCL and electrical cycle length was conducted for five exemplary
patients. FIG. 33B is a graph illustrating the results of the five
exemplary patients. With reference to FIG. 33B, and as embodied
herein, at least one representative MCL was very close to the
electrical cycle length. FIG. 33B illustrates that
electromechanical cycle length and/or MCL can be correlated with
the electrical cycle length with a correlation of 0.96 and a slope
of 1.1.
[0119] FIGS. 33C and 33D together illustrate results from one
exemplary patient undergoing atrial fibrillation. FIG. 33C
illustrates a peak MCL map depicting multiple clustered dominant
frequencies. For purpose of illustration and not limitation, the
separation into these dominant frequencies is illustrated by the
histogram shown in FIG. 33D. With reference to FIG. 33D, the peak
cycle length map during atrial fibrillation depicts further spatial
fragmentation of the peak cycle length.
[0120] Aspects of the present disclosed subject matter illustrate
electromechanical activation mapping to identify the site of
cardiac rhythm mechanisms during arrhythmia in humans and to
characterize such cardiac rhythm mechanisms, which can lead to
improved treatments and clinical management. Certain clinical
practices utilize minimally invasive techniques to obtain precise
maps of the activation of the atria and ventricles. Such techniques
can be costly, time-consuming, and carry some degree of risk, and
hence can be challenging to provide complete activation maps before
and after treatment, as well as during catheter procedures.
[0121] Aspects of the present disclosed subject matter illustrate
methods for electromechanical activation mapping during reentrant
and focal arrhythmias. For purpose of illustration and not
limitation, and as embodied herein, exemplary techniques are
provided for imaging spatiotemporal mechanics of arrhythmias with
high accuracy and spatial and temporal resolutions in a full field
of view in humans. These exemplary techniques can provide for
characterization of an electromechanical propagation pattern and/or
representative mechanical cycle lengths, which can correspond with
their electrophysiological equivalents.
[0122] For purpose of illustration and not limitation, and as
embodied herein, focal rhythms can behave similarly to paced
rhythms. For example and without limitation, focal rhythms can have
a single source of electromechanical activation located in the
vicinity of the earliest electrical activation. As embodied herein,
EWI can be used to characterize the propagation of
electromechanical activation, which can propagate from an atria's
sinus node and from the bundle branch and which can terminate in
the ventricles during ventricular pacing. Electromechanical
activation propagation patterns similar to pacing can occur in a
patient during premature ventricular complexes. The
electromechanical activation sequence of the same patient during
sinus rhythm can be similar to that of normal subjects. For
example, and as embodied herein, in a patient with atrial
tachycardia, the electromechanical activation propagation pattern
can indicate a source located near the roof of the LA, in
accordance with electrical mapping. As such, an exemplary
application of non-invasive, ultrasound-based, electromechanical
activation mapping is provided, which can be performed during or
prior to invasive procedures. For purpose of illustration and not
limitation, prior knowledge of an electromechanical source located
in the LA can allow for clinical preparation. For example, and as
embodied herein, such prior knowledge can be used to determine
whether transseptal access would be obtained during treatment and
to perform risk-benefit analysis to determine the best course of
treatment (e.g., pharmacological vs. ablation treatment).
[0123] Additionally, and as embodied herein, the electromechanical
activation maps can be correlated with their electrical
counterpart, at least in part of the atrial tissue, during atrial
flutter. For example, and as embodied herein, a single
representative frequency can be identified, and the phase of that
frequency can indicate a propagation direction from the
cavotricuspid isthmus region to the RA and LA, which can occur
during typical atrial flutters. Additionally or alternatively, and
as embodied herein, other behaviors can be identified in the atria,
including and without limitation, one part of the atria that
contracts with the same frequency as the electrical activation and
another region that does not contract with the same frequency. As
such, mapping the mechanics of the heart can identify regions of
the heart in which the mechanical and electrical activities appear
to be decoupled. Further spatial fragmentation of the periodicity
of the mechanics of the atria can be observed during fibrillation.
Such techniques can determine characteristics of the atrial
mechanics during arrhythmia, including in the progression from
flutter to fibrillation and vice versa.
[0124] For purpose of illustration and not limitation, deformation
of the atria caused by the onset of ventricular contraction and
relaxation can affect certain aspects of the techniques described
herein. For example, and as embodied herein, such a deformation can
affect frequency analyses based on multiple activation cycles,
which can be due at least in part to the relatively short
acquisition time of these processes. Filtering and the development
of longer acquisition sequences can inhibit or prevent such atrial
deformation.
[0125] Certain non-invasive electrical mapping techniques can be
utilized to examine the epicardium, and can assume an immobilized
heart function. A mechanical assessment of the atria can be
utilized, for example, by electrophysiologists or interventional
cardiologists to achieve the advantages described herein. According
to certain clinical practices, echocardiograms can be performed on
arrhythmia patients. Other non-invasive electrical mapping
techniques can utilize on time-consuming and costly high resolution
CT or MRI scans. The electromechanical activation mapping systems
and techniques described herein can be obtained separately from, or
in conjunction with, echocardiograms.
[0126] As embodied herein, mapping the electromechanical activity
during arrhythmias non-invasively with real-time feedback can be
used determine characteristics of atrial mechanics in the evolution
and perpetuation of arrhythmias. Furthermore, and as embodied
herein, such a mapping can be used to predict the origin site of
arrhythmias and the mechanism and monitoring of intervention
outcomes.
[0127] It will be understood that the foregoing is only
illustrative of the principles of the present disclosure, and that
various modifications can be made by those skilled in the art
without departing from the scope and spirit of the present
disclosure.
* * * * *