Myocardial strain by Doppler echocardiography. Validation of a new method to quantify regional myocardial function.

BACKGROUND: Myocardial strain is a measure of regional deformation, and by definition, negative strain means shortening and positive strain, elongation. This study investigates whether myocardial strain can be measured by Doppler echocardiography as the time integral of regional velocity gradients, using sonomicrometry as reference method. METHODS AND RESULTS: In 13 anesthetized dogs, myocardial longitudinal strain was measured on apical images as the time integral of regional Doppler velocity gradients. Ultrasonic segment-length crystals were placed near the left ventricular (LV) apex and near the base. Apical ischemia was induced by occluding the left anterior descending coronary artery (LAD), and preload was increased by saline. Percentage systolic strain by Doppler correlated well with strain by sonomicrometry (y=0.82x-1.79, r=0.92, P<0.01). During LAD occlusion, apical myocardium became dyskinetic, as indicated by positive strain values and negative Doppler velocities. At the LV base, myocardial strain by Doppler, strain by sonomicrometry, and velocity of shortening by sonomicrometry (dL/dt) were unchanged during apical ischemia. However, myocardial Doppler velocities at the base decreased from 4.2+/-0.7 (+/-SEM) to 2.7+/-0. 4 cm/s (P<0.05), probably reflecting loss of motion caused by tethering to apical segments. Volume loading increased myocardial Doppler velocities from 2.2+/-0.3 to 4.1+/-0.8 cm/s (P<0.05) and Doppler-derived strain from -12+/-1% to -22+/-2% (P<0.05), whereas peak LV elastance remained unchanged. CONCLUSIONS: Myocardial strain by Doppler echocardiography may represent a new, powerful method for quantifying regional myocardial function and is less influenced by tethering effects than Doppler tissue imaging. Like myocardial Doppler velocities, strain is markedly load-dependent.

Flow estimation using ultrasound imaging (color M-mode) and computer postprocessing.

We have developed a method to calculate flow noninvasively in blood vessels using color Motion-mode (M-mode) and computer postprocessing. The velocity of each point in the cross-sectional area of the vessel was found from the color M-mode recording by correcting for angle both distances and velocities and by assuming a symmetrical circular velocity field. Volume flow was then found by integrating the velocity field at 5-ms intervals through the cardiac cycle. In a cardiovascular hydromechanical model, a correlation of 0.99 and p value of less than 0.001 were found between estimated and measured flow in the model (n = 8). In 20 healthy individuals, we made 31 investigations in the common carotid (CCA), internal carotid (ICA), and external carotid (ECA) artery, comparing flow in the CCA with the added flow in the ICA and ECA. The values (CCA versus ICA + ECA) correlated with r = 0.91 and p less than 0.01. Repeated investigations (n = 8) in one individual gave flow estimates of 495 +/- 50 ml/min in the CCA, 304 +/- 45 ml/min in the ICA, and 165 +/- 37 ml/min in the ECA (means +/- SD). This article shows that this system can make accurate estimation of blood flow to the brain noninvasively.

Real-time strain rate imaging of the left ventricle by ultrasound.

The regional function of the left ventricle can be visualized in real-time using the new strain rate imaging method. Deformation or strain of a tissue segment occurs over time during the cardiac cycle. The rate of this deformation, the strain rate, is equivalent to the velocity gradient, and can be estimated using the tissue Doppler technique. We present the strain rate as color-coded 2-dimensional cine-loops and color M-modes showing the strain rate component along the ultrasound beam axis. We tested the method in 6 healthy subjects and 6 patients with myocardial infarction. In the healthy hearts, a spatially homogeneous distribution of the strain rate was found. In the infarcted hearts, all the infarcted areas in this study showed up as hypokinetic or akinetic, demonstrating that this method may be used for imaging of regional dysfunction. Shortcomings of the method are discussed, as are some possible future applications of the method.

Volumetric blood flow measurement with the use of dynamic 3-dimensional ultrasound color flow imaging.

We describe a new method for measuring blood volume flow with the use of freehand dynamic 3-dimensional echocardiography. During 10 to 20 cardiac cycles, the ultrasonographic probe was slowly tilted while its spatial position was continuously recorded with a magnetic position sensor system. The ultrasonographic data were acquired in color flow imaging mode, and the separate raw digital tissue and Doppler data were transferred to an external personal computer for postprocessing. From each time step in the reconstructed 3-dimensional data, one cross-sectional slice was extracted with the measured and recorded velocity vector components perpendicular to the slice. The volume flow rate through these slices was found by integrating the velocity vector components, and was independent of the angle between the actual flow direction and the measured velocity vector. Allowing the extracted surface to move according to the movement of anatomic structures, an estimate of the flow through the cardiac valves was achieved. The temporal resolution was preserved in the 3-dimensional reconstruction, and with a frame rate of up to 104 frames/s, the reconstruction jitter artifacts were reduced. Examples of in vivo blood volume flow measurement are given, showing the possibilities of measuring the cardiac output and analyzing blood flow velocity profiles.

Cross-sectional left ventricular outflow tract velocities before and after aortic valve replacement: a comparative study with two-dimensional Doppler ultrasound.

To assess whether aortic valve replacement (AVR) results in changes in the flow velocity distribution in the left ventricular outflow tract (LVOT), 10 patients undergoing AVR for aortic stenosis were studied. By extracting velocity information from color flow maps as digital data, instantaneous cross-sectional velocity profiles were constructed. Velocity profiles obtained 1 to 3 days before AVR were compared with recordings made 3 months later. The LVOT velocity profiles were variably skewed both before and after surgery, and no systematic or uniform changes could be detected after AVR. The highest velocities were most often localized in the region from the center of the outflow tract diameter toward the septum both before and after surgery. At the time of peak flow the ratio of the maximum to the cross-sectional mean velocity was 1.38 +/- 0.13 before and 1.39 +/- 0.08 after AVR (NS), and the ratio of the maximum to the mean velocity time integral was 1.47 +/- 0.10 before and 1.56 +/- 0.10 after (NS). We conclude that AVR in patients with aortic stenosis does not result in a change in LVOT velocity profiles that will influence stroke volume estimates with the Doppler technique.

A new method describing cross-sectional blood flow velocity profiles in the left ventricular outflow tract of patients with atrial fibrillation with the use of high-frame rate 2-dimensional color flow imaging.

A new Doppler method was developed to evaluate the instantaneous cross-sectional velocity profile variability in the left ventricular outlet tract in patients with atrial fibrillation. Blood flow velocities acquired at a high frame rate (>90 frames/s) from a single heart cycle were used to display the velocity profile. In 9 patients, 2 heart cycles with different R-R interval lengths were recorded in color flow mode in a transthoracic apical 5-chamber and long-axis view. Raw digital ultrasound data were analyzed with an external personal computer. The data indicated a variable skew in the profiles with the highest velocities and velocity-time integral (VTI) most often located in the center and toward the septum. The maximum VTI overestimated the mean VTI by approximately 40%. No significant difference existed between the two heartbeats. Thus the VTI can be averaged from heartbeats of different R-R lengths in atrial fibrillation.

Instantaneous cross-sectional flow velocity profiles: a comparative study of two ultrasound Doppler methods applied to an in vitro pulsatile flow model.

Two methods based on different techniques for construction of cross-sectional flow velocity profiles from Doppler ultrasound signals were compared: an intraluminal method using pulsed-wave Doppler echocardiography and an extraluminal method using two-dimensional (color) Doppler ultrasound. The methods were applied to an in vitro pulsatile flow model. With the intraluminal method, pulsed Doppler recordings obtained throughout several flow pulses at different positions across a tube were digitized, and cross-sectional flow velocity profiles were obtained by matching the onset of flow velocity at the various positions. With the extraluminal method, cross-sectional flow velocity profiles were obtained by time interpolation between the digital flow velocity data obtained from several flow velocity maps. The first flow velocity map was recorded at onset of flow and the following maps were incrementally delayed with 20 msec from one flow pulse to the next. The time lag caused by the time needed to update each of the flow velocity maps was compensated for by time interpolation between the sequentially recorded flow velocity maps. The cross-sectional flow velocity profiles obtained with the two methods were compared at identical positions within the tube model at equal flow settings and throughout the pulsatile flow periods. At three different flow settings with peak flow velocity of 0.3, 0.5, and 0.7 m/sec, the difference (mean +/- SD) between the obtained velocities were 0.01 +/- 0.04, -0.01 +/- 0.05, and -0.03 +/- 0.07 m/sec, respectively. The findings suggest that cross-sectional flow velocity profiles from pulsatile flow velocity recordings can be obtained equally well with both methods.

The velocity distribution in the aortic anulus in normal subjects: a quantitative analysis of two-dimensional Doppler flow maps.

The velocity distribution in the aortic anulus is commonly assumed to be uniform. A skewed velocity profile may have consequences for the accuracy of volume flow estimates by the Doppler echocardiographic technique. To assess this issue, the velocity distribution in the aortic anulus in 12 normal subjects was studied by computer analysis of digital velocity data from two-dimensional Doppler ultrasound flow maps. The velocity profiles in the aortic anulus were found to be flat but slightly skewed, with the highest velocities toward the septum. There was little interindividual variation. Our findings imply that the centerline velocity is the best estimate for the spatial mean velocity at the aortic anulus in normal subjects. The importance of this finding in patients is unknown. In normal subjects, the results suggest that stroke volume might be overestimated by approximately 15% by Doppler echocardiography if the cross-sectional velocity profile is not accounted for.

Strain rate imaging by ultrasonography in the diagnosis of coronary artery disease.

Regional strain rate in the left ventricle can be assessed in real time and color mapped. The method is termed strain rate imaging (SRI), and findings correspond well with 2-dimensional echocardiography. This study addresses SRI as a method for localizing coronary lesions, compared with standard echocardiography. Twenty patients with acute myocardial infarction who underwent coronary angiography for clinical reasons were examined with SRI and standard echocardiography. Wall motion was graded by SRI color and separately by wall thickening. Strain rate imaging and 2-dimensional echocardiography results agreed well. An infarct-related artery was identified from angiograms combined with electrocardiograms. Both methods identified an infarct-related artery in 19 possible cases and had equal sensitivity and specificity for wall segments affected by lesion. Combining the information from both methods did not change accuracy. The study validates SRI as a method for assessing regional wall function in coronary artery disease. The advantages of SRI are discussed and measurements of strain rates are given.

Velocity matched spectrum analysis: a new method for suppressing velocity ambiguity in pulsed-wave Doppler.

A new approach to spectrum analysis, which is capable of suppressing velocity ambiguity in pulsed-wave ultrasonic Doppler, is presented. By simultaneous processing of several data samples from a range in depth, the movement of the scatterers along the ultrasonic beam can be tracked from pulse to pulse for each velocity component in the spectrum. In this way the correlation length of the signal component arising from a specific velocity increases when that velocity matches the expected velocity. The resulting velocity/time spectral display shows a more clearly defined spectral envelope of the maximum velocity than with conventional methods based on the discrete Fourier transform of the Doppler signal. This makes it possible to delineate velocity waveforms with peak velocity up to several times the Nyquist limit. Experimental data from human subclavian and aortic arteries are presented, where the new method is compared to conventional spectrum analysis.

Accuracy of in-vitro volume estimation of small structures using three-dimensional ultrasound.

We describe an ultrasound probe for three-dimensional transvaginal imaging. The transducer was an annular array with a center frequency of 7.5 MHz which was rotated with an internal stepper motor. The probe had no external moving parts, and the total volume covered by a full rotation defined a half sphere. The raw digital data from the scanner were transferred to an external PC for three-dimensional reconstruction. We evaluated the three-dimensional imaging system by measuring the volumes of phantoms (range 24.8-3362.5 mm(3)) in a water tank, and found good correlation with true volumes (two observers' measurements gave a linear regression with a slope of 1. 010 and R(2) = 0.993, and a slope of 0.956 and R(2) = 0.993, respectively). The size of the point-spread function was used in the calculations to eliminate the effect of under- or overestimation due to the limited ultrasound beam resolution. An example of data acquisition, volume estimation and imaging of an embryo less than 8 weeks old in vivo with the brain cavities and body is given. We conclude that the three-dimensional reconstruction and volume estimation were accurate and repeatable.

Comparison of time-domain displacement estimators for two-dimensional RF tracking.

Techniques have been described in the literature to enable multidimensional strain rate estimation. They are based on multidimensional velocity estimation. One of the problems in obtaining robust lateral strain rate estimates is the fact that lateral velocity estimates are intrinsically noisier than axial ones. The aim of this study was to find the optimal estimator for tracking of the radiofrequency patterns both in axial and lateral directions. Performances of the following estimators were investigated using simulations: cross-correlation, normalized cross-correlation, sum of absolute differences and sum of squared differences. Two-dimensional (2-D) velocity estimation was not feasible using cross-correlation. However, normalized cross-correlation, sum of absolute differences and sum of squared differences showed accurate axial and lateral results. For smaller window lengths, sum of squared differences was found to be the preferred estimator for 2-D velocity estimation using a 1-D kernel.

Dynamic three-dimensional freehand echocardiography using raw digital ultrasound data.

In this paper, we present a new method for simple acquisition of dynamic three-dimensional (3-D) ultrasound data. We used a magnetic position sensor device attached to the ultrasound probe for spatial location of the probe, which was slowly tilted in the transthoracic scanning position. The 3-D data were recorded in 10-20 s, and the analysis was performed on an external PC within 2 min after transferring the raw digital ultrasound data directly from the scanner. The spatial and temporal resolutions of the reconstruction were evaluated, and were superior to video-based 3-D systems. Examples of volume reconstructions with better than 7 ms temporal resolution are given. The raw data with Doppler measurements were used to reconstruct both blood and tissue velocity volumes. The velocity estimates were available for optimal visualization and for quantitative analysis. The freehand data reconstruction accuracy was tested by volume estimation of balloon phantoms, giving high correlation with true volumes. Results show in vivo 3-D reconstruction and visualization of mitral and aortic valve morphology and blood flow, and myocardial tissue velocity. We conclude that it was possible to construct multimodality 3-D data in a limited region of the human heart within one respiration cycle, with reconstruction errors smaller than the resolution of the original ultrasound beam, and with a temporal resolution of up to 150 frames per second.

Fetal and maternal aortic flow in two different maternal positions. An investigation with combined Doppler-velocimetry and ultrasonic multiple array.

To study aortic velocities in the fetus, a newly developed multirangegated Doppler velocimeter was combined with a commercial B-mode linear-array apparatus. Spectrum analysis of the Doppler shift signals was performed. The signals with the highest frequencies were selected for hard-copy, and integrated with a digitizer. In the first part of the study, the reproducibility of the method was established by serial measurements of aortic velocity in 7 near-term fetuses. It proved sufficient to evaluate 5 successive heart-cycles. The mean coefficient of variation was 5% (mean number of recordings evaluated 7.5). In the second part of the study, possible influence of aortocaval compression on fetal aortic velocity indicative of changes in fetal cardiac output, was investigated in 10 pregnant women with no symptoms of supine hypotension. Doppler monitoring of the material cardiac output from the suprasternal notch was added to the set-up described to study the supine reduction of cardiac output frequently described in pregnant women. No significant change was found either in the maternal or in the fetal circulation.

Estimation of regurgitant volume and orifice in aortic regurgitation combining CW Doppler and parameter estimation in a Windkessel-like model.

A method for noninvasive estimation of regurgitant orifice and volume in aortic regurgitation is proposed and tested in anesthetized open chested pigs. The method can be used with noninvasive measurement of regurgitant jet velocity with continuous wave ultrasound Doppler measurements together with cuff measurements of systolic and diastolic systemic pressure in the arm. These measurements are then used for parameter estimation in a Windkessel-like model which include the regurgitant orifice as a parameter. The aortic volume compliance and the peripheral resistance are also included as parameters measurements in the open chest pigs are used. Electromagnetic flow measurements in the ascending aorta and pulmonary artery are used for control, and a correlation between regurgitant volume obtained from parameter estimation and electromagnetic flow measurements of 0.95 over a range from 2.1 to 17.8 mL is obtained.

Clutter rejection filters in color flow imaging: a theoretical approach.

A general class of linear clutter rejection filters is described, covering the commonly used filter types including FIR/IIR filters with linear initialization, as well as regression filters, where the clutter component is estimated by least square curve fitting. The filter can be described by a complex valued matrix, and a frequency response is defined. However, in contrast to a time invariant filter, the general linear filter may create frequency components which are not present in the input signal. This produces bias in the velocity and velocity spread estimates. It is shown that the clutter filter effect on the autocorrelation estimates can be described by a frequency domain transfer function, but unlike time invariant filters, the transfer function is different for each temporal lag of the autocorrelation function. Using a two dimensional (axial and temporal dimension) model of the received signal, the bias in velocity and velocity spread is quantified, both for the autocorrelation algorithm and the time shift cross-correlation estimator. Theoretical expressions, as well as numerical examples are given.

Interpolation methods for time-delay estimation using cross-correlation method for blood velocity measurement.

The cross-correlation method (CCM) for blood flow velocity measurement using Doppler ultrasound is based on time delay estimation of echoes from pulse-to-pulse. The sampling frequency of the received signal is usually kept as low as possible in order to reduce computational complexity, and the peak in the correlation function is found by interpolating the correlation function. The parabolic-fit interpolation method introduces a bias at low sampling rate to the ultrasound center frequency ratio. In this study, four different methods are suggested to improve the estimation accuracy: (1) Parabolic interpolation with bias-compensation, derived from a theoretical signal model. (2) Parabolic interpolation combined with linear filter interpolation of the correlation function. (3) Parabolic interpolation to the complex correlation function envelope. (4) Matched filter interpolation applied to the correlation function. The new interpolation methods are analyzed both by computer simulated signals and RF-signals recorded from a patient with time delay larger than 1/f(0), where f(0) is the center frequency. The simulation results show that these methods are more accurate than the parabolic-fit method. From the simulation, the worst estimation accuracy is about 1.25% of 1/f(0) for the parabolic-fit interpolation, and it is improved by the above methods to less than 0.5% of 1/f(0) when the sampling rate is 10 MHz, the center frequency is 2.5 MHz and the bandwidth is 1 MHz. This improvement also can be observed in the experimental data. Furthermore, the matched filter interpolation gives the best performance when signal-to-noise ratio (SNR) is low. This is verified both by simulation and experimentation.

Three-stage approach to ultrasound contrast detection.

A new method for detecting ultrasound contrast agents using a three-stage pulsing sequence is proposed. The method is based on observations showing that the scattering properties of contrast agents are modified by ultrasonic insonation at high power, but remain unchanged at low power. The objective of the first stage of the pulsing sequence is to use low power pulses to obtain a high resolution reference image without altering the agent. Higher power pulses in the second stage modify the contrast agent. The third stage detects the changes imposed to the contrast agent using low power pulses. A temporal filter is proposed to discriminate contrast response from clutter signal. The method is similar to power Doppler methods in that it uses several pulses to survey the target while destroying the agent. The new idea is to separate detection and destruction to circumvent a trade-off between sensitivity and resolution. Results from in vitro experiments with three different contrast agents are presented. The results are compared with harmonic power Doppler processed from the same data and show that an improvement in sensitivity is achievable by including the high power burst in the pulsing sequence. The results also show that the proposed filter reduces clutter artifacts from moving tissue.

A new ultrasound contrast imaging approach based on the combination of multiple imaging pulses and a separate release burst.

A new ultrasound contrast imaging technique is described that optimally employs the rupture of the contrast agent. It is based on a combination of multiple high frequency, broadband, imaging pulses and a separate release burst. The imaging pulses are used to survey the target before and after the rupture and release of free gas bubbles. In this way, both processes (imaging and release) can be optimized separately. The presence of the contrast agent is simply detected by correlating or subtracting the signal responses of the imaging pulses. Because the time delay between the imaging pulses can be very short, the subtraction is less affected by tissue motion and can be done in real time. In vitro measurements showed that by using a release burst, the detection sensitivity increased 12 to 43 dB for different types of contrast agents. In the presence of a moving phantom, the increase in sensitivity was 22 dB. This new method is very sensitive for contrast agent detection in fundamental imaging mode and, therefore, non-linear propagation effects do not limit the maximum obtainable agent-to-tissue ratio. However, because of the inherent destruction of the contrast agent, it has to operate in an intermittent way. Through experiments, we have demonstrated the potential of the method to achieve simultaneous high sensitivity for contrast detection, i.e., high agent-to-tissue ratio, and high spatial resolution performance for different types of contrast agents.

Statistical evaluation of clutter filters in color flow imaging.

The filter used to separate blood signals from the tissue clutter signal is an important part of a color flow system. In this paper, statistical detection theory is used to evaluate the quality of the most commonly used clutter filters. The probability of falsely classifying a sample volume as containing blood is kept below a specified threshold. With this constraint, the probability of correctly detecting blood is calculated for all the filters. Using a measured clutter signal, we found that polynomial regression filters and projection-initialized IIR filters are best among the commonly used filters. The probability of correctly detecting blood with velocity 10.1 cm/s was 0.32 for both these filters. The corresponding value for the optimal detector was 0.81, whereas a regression filter that depends on the clutter signal statistics achieved a blood detection probability of 0.72.