Makoto Hirama

New method for evaluating left ventricular wall motion by color-coded tissue doppler imaging: In vitro and in vivo studies

OBJECTIVES The aim of this study was to examine the accuracy and validity of a newly developed tissue Doppler imaging system in in vitro and in vivo studies. BACKGROUND Because quantitative measurement of wall motion velocity in real time is still difficult by conventional echocardiography, we developed a new system for evaluating ventricular wall motion by analyzing Doppler signals from cardiac tissue. METHODS We used a modified Doppler color imaging system, omitting the high pass filter to allow Doppler signals from cardiac tissue to enter the auto-correlator. Ultrasound carrier and pulse repetition frequencies were 3.75 MHz and 3.0 to 6.0 kHz, respectively. Under these conditions, the lowest measurable velocity was 0.2 cm/s. RESULTS In the rotating sponge model, the measured velocity correlated well with the actual velocity (y = 0.97x + 2.17, r = 0.99). In clinical settings, the mid-ejection mean velocity at either endocardial or epicardial sites of the left ventricular posterior wall measured by M-mode tissue Doppler imaging correlated well with that measured by conventional M-mode echocardiography (y = 0.94x + 0.64, r = 0.99). During systole, in healthy subjects, the anterior left ventricular wall was color-coded blue and the posterior wall was color-coded red, whereas the akinetic regions associated with myocardial infarction showed no color throughout the cardiac cycle. The ventricular posterior wall excursion velocity, defined as the difference between velocities at the endocardial and epicardial sites, was significantly slower in patients with dilated cardiomyopathy (0.4 +/- 0.3 cm/s) than in normal subjects (2.0 +/- 0.6 cm/s). CONCLUSIONS These results indicate that the present system accurately represents tissue velocity and can create two-dimensional color images that facilitate visual assessment of ventricular wall motion.

Myocardial velocity gradient as a new indicator of regional left ventricular contraction : detection by a two-dimensional tissue Doppler imaging technique

OBJECTIVES This study was performed to assess a new indicator of regional left ventricular contraction determined by a two-dimensional tissue Doppler imaging technique. BACKGROUND Recent studies have demonstrated that instantaneous tissue motion velocity can be noninvasively assessed by tissue Doppler imaging. However, quantitative assessment of regional left ventricular contraction is still difficult because of the effects of the Doppler angle of incidence and parallel motion of the whole heart. METHODS We assessed left ventricular wall motion in 11 normal subjects, 14 patients with an old myocardial infarction (anteroseptal in 7, posterior in 7) and 8 patients with dilated cardiomyopathy. Tissue Doppler velocity was corrected by the Doppler angle of incidence after the hypothetical center of contraction was set. Subsequently, the myocardial velocity gradient between the endocardium and epicardium was determined from the velocity profile along each radial line from the center of contraction by using least squares linear regression. RESULTS In normal subjects, peak myocardial velocity gradient was lower in the anteroseptal wall (mean [+/- SD] 1.69 +/- 0.53 s-1) than in the posterior wall (3.28 +/- 0.67 s-1, p < 0.01). Myocardial velocity gradient in the infarct regions was significantly lower (anteroseptal 0.58 +/- 0.41 s-1, p < 0.05; posterior 0.17 +/- 0.27 s-1, p < 0.01) than that in normal subjects as well as that in the corresponding noninfarct regions (2.84 +/- 0.37 s-1 and 1.48 +/- 0.25 s-1, p < 0.01, respectively). In patients with dilated cardiomyopathy, myocardial velocity gradient was generally lower (anteroseptal 0.72 +/- 0.59 s-1; posterior 0.93 +/- 0.67 s-1) than that in normal subjects (p < 0.01). CONCLUSIONS These results demonstrate that regional left ventricular contraction can be quantitatively assessed by the myocardial velocity gradient derived from two-dimensional tissue Doppler imaging. We suggest that myocardial velocity gradient has potential for the quantitative assessment of regional left ventricular contraction abnormalities in patients.

Analysis of Ventricular Wall Motion Using Color-Coded Tissue Doppler Imaging System

We developed a new color Doppler system by which Doppler signals associated with tissue motion can be determined, and called it the tissue Doppler imaging (TDI) system. Using high-speed scanning, the frame rate was 26–38 F/s, and the pulse repetition frequency was 4.5–6.0 kHz. Under these conditions, the lowest measurable velocity was improved to 2 mm/s. Wall motion toward the transducer was coded as red, and that away from the transducer, as blue. To examine the accuracy and validity of the measured velocity of ventricular wall motion, we performed in vitro and in vivo studies. The results demonstrate that the present TDI system accurately represents the tissue velocity, and is applicable for creating two-dimensional images of the ventricular wall motion in real time, facilitating the visual assessment of abnormal ventricular wall motion.

Quantitative measurement of volume flow rate (cardiac output) by the multibeam Doppler method

A new method has been developed for measuring the volume flow rate of blood flowing through large vessels or outflow tracts of the heart. In this article we describe the principle of a method that can reduce the dependence of the Doppler angle of flow measurement by setting the sample points along a line to which every ultrasound beam is perpendicular. To evaluate the accuracy of this method, flow phantom experiments were made for both steady and pulsatile flows. The volume flow rate measured by this method agrees well with that observed by an ultrasound flowmeter (r = 0.99) when the vessel diameter is large (25 mm). However, this method overestimates by 40% when the vessel diameter is small (8 mm). To make this method applicable to small vessels, an improvement in the lateral resolution of Doppler measurement is necessary. It has been concluded that this method can be used to measure the cardiac output or volume flow rates in large vessels.

Imaging through an inhomogeneous layer by least‐mean‐square error fitting

A new active ultrasonic imaging method through an inhomogeneous layer is proposed. It has the special feature that its effectiveness does not depend on the class of the objects to be imaged. In this method, first, a set of data is acquired by repeating transmission and reception for all possible combinations of pairs of transducers on the array, then the spatial frequency components of the object and the structure of the inhomogeneous layer are estimated from these data by means of least‐mean‐square error fitting. Since the data have redundancy for the parameters to be estimated, this process gives an optimum and stable estimation algorithm even when measurement errors and noise are included. The image is reconstructed from the estimated spatial frequency components through inverse Fourier transform. The effectiveness of this method is ensured by several numerical analyses and experiments.

Adaptive ultrasonic array imaging system through an inhomogeneous layer

An adaptive ultrasonic array imaging technique aiming at reconstruction of the diffraction‐limited images of general objects even through an inhomogeneous layer is proposed. The special feature of the system is the estimation of the complex amplitude distribution of the inhomogeneity by measuring the degree of focusing, that is, by measuring the degree of correlation between the transmitted signals and the signals reflected from the object under observation. The construction of a proper measure function, an iterative procedure for the estimation of the inhomogeneity, and the imaging method that uses the estimated inhomogeneity are discussed in detail. The effectiveness of the method is verified both by numerical analyses and experiments.

An evaluation of an in vivo local sound speed estimation technique by the crossed beam method

An in vivo local sound speed estimation technique, using the crossed beam method, has been proposed and its applicability was evaluated. At first, the potential of this technique was studied by a mapping simulation using the ray tracing technique followed by an experiment with a cylindrical agar phantom. The simulation result showed that an exact measurement of local sound speed values was difficult, but the sound speed information for the local region (its relative magnitude to the surrounding medium) was emphasized as a refraction mapping pattern. The experimental results agreed well with the calculation results. Furthermore, a clinical application was performed, using the clinical system (modified electronic linear scanner), on two liver tumor patients.

Active incoherent ultrasonic imaging through an inhomogeneous layer

In this paper, a method of active incoherent ultrasonic imaging which has been proposed in previous work [Sato et al., Acoustical Imaging (Plenum, New York, 1982), Vol. 11, pp. 289–308; Yokota and Sato, Acoustical Imaging (Plenum, New York, 1983), Vol. 12, pp. 621–634] is extended to the case where an inhomogeneous layer exists between the objects and aperture. Data acquisition is carried out by repeating transmission of burst ultrasonic waves from each transducer on the aperture and reception of the reflected waves on the same aperture. The compensation for the effect of the inhomogeneous layer is then carried out by using the space invariant property of the spatial frequency components of the object. The coherence function of the revised wave field is derived from these data, and the image is reconstructed by means of the nonlinear 2‐D power spectral estimation. An algorithm of image reconstruction which includes cases where the inhomogeneous layer is located some distance from the aperture is demonstra...

New color Doppler technique for detecting turbulent tumor blood flow: A possible aid to hepatocellular carcinoma diagnosis

We created a new imaging technique that detects and emphasizes turbulence, which is a characteristic of blood flow in hepatocellular carcinoma. We devised two indices that determine a characteristic tumor flow, the bi‐directional and low‐peak indices. In the phantom study, both indices of turbulence caused by a stenosis were much higher. In the clinical study, both indices were significantly higher in tumors than in the portal vein or hepatic vein. A turbulent blood flow was detected in 77% of tumors, whereas such detection seldom occurred in the portal or hepatic vein. This technique has the potential to distinguish turbulence in hepatocellular carcinoma.

A Fundamental Evaluation of in Vivo Sound Speed Mapping Technique by Crossed Beam Method

Recently, several kinds of in vivo sound speed measurement techniques, using a pulse echo method, have been developed for the purpose of ultrasound tissue characterization.1,2,3The crossed beam method was proposed as a simple method by Haumschild and Greenleaf4 and Nishimura et al.5. This method uses two single probes. One probe is used for transmitting the ultrasound pulsed wave and the other for receiving the wave scattered from the region where the beams from the two probes cross. From the propagation time of the pulsed wave, the sound speed value is calculated. This method can also be realized using a linear array probe.6