Ultrasonic Doppler Flow Measurement System With Tissue Motion Discrimination

This invention relates to ultrasonic diagnostic systems which measure the flow of fluids through Doppler interrogation and, in particular, to the discrimination of tissue motion by such systems.

Ultrasonic diagnostic systems which measure the flow of fluids through Doppler interrogation are in widespread use for the acquisition of medical patient data concerning the flow of blood and other fluids in the body. In one commonly accepted embodiment of Doppler measurement, an area of the body such as the heart is repetitively interrogated with ultrasonic waves and returning echo signals are compared to a reference to determine the velocity of blood flow in the heart. This interrogation is performed over a two-dimensional sector of the heart to determine flow velocities throughout the interrogated area. The resultant flow velocity values are then displayed in a color image format as a function of measurement location, where different shades and intensities of color represent blood flow of different velocities and directions. Such systems are commercially available as what are known as ultrasonic color flow imaging systems.

Such Doppler systems operate by measuring velocity in terms of relative motion between the fluids being interrogated and the scanhead which produces the ultrasonic waves and detects the resultant echo signals. However, Doppler information signals are returned not only by fluids, but also by tissue structure within the body. These latter signal components are undesired in systems where only fluid motion is being interrogated. But the prior art systems perform well when the scanhead is held stationary and the only substance moving in the interrogated area is the fluid which is being measured. This is because techniques are known for removing signal components returned from stationary tissue, and are generally embodied in what are referred to as clutter rejection filters. However, if the scanhead is moved even momentarily by the user, relative motion will occur between all of the tissue in the body and the scanhead, returning velocity-encoded echoes from throughout the interrogated area. Moreover, even if the scanhead is held stationary, tissue motion caused by the beating heart or breathing will return velocity-encoded echoes. These undesired velocity signals can be detected by the Doppler system even with the use of known clutter rejection filters, and the resulting artifacts will interfere with the fluid flow image, in which only the motion of the fluid being interrogated is desired. Hence it is desirable to eliminate these motion artifact components from the color flow image.

Previous attempts at eliminating these motion artifact effects have focused on their manifestation in the color images, which are displayed as a realtime sequence of fields or frames of Doppler flow velocity data. These efforts have employed image post-processing techniques to look for the occurrence of such artifacts from one image to another, and through frame to frame analytical techniques have attempted to eliminate the effects of the artifacts. Such techniques by their nature allow the ultrasound system to process the Doppler information through Doppler velocity determination and two-dimensional image formatting, thereby processing the undesired tissue motion artifacts along with the desired Doppler velocity information through most of the system data processing network. It would be desirable to eliminate such tissue motion artifact effects prior to image formatting, and most preferably prior to the determination of fluid Doppler velocity values in such systems.

In accordance with the principles of the present invention, an ultrasonic Doppler flow measurement system is provided which discriminates tissue motion artifacts prior to image formatting of detected velocity values and, in a preferred embodiment, prior to determination of Doppler velocity values. An embodiment of the present invention operates on Doppler data prior to image formatting by first discriminating those signal components resulting from tissue motion. The discriminated components are then eliminated from the Doppler spectrum and the resultant data is then processed for the determination of flow velocity information and image formatting. In a first embodiment the data being discriminated is the Doppler data relating to one location being interrogated for flow velocity, and in a preferred embodiment the data being discriminated relates to a plurality of positionally associated interrogation locations. A constructed embodiment of the present invention will obviate the need to perform Doppler velocity estimation and image processing of Doppler data which is contaminated with tissue motion artifacts.

In the drawings

FIG. 1 illustrates in block diagram form an ultrasonic Doppler flow measurement system constructed in accordance with the principles of the present invention;

FIG. 2 illustrates an area being interrogated for Doppler flow information;

FIG. 3 illustrates an array of data values resulting from Doppler interrogation along an ultrasonic vector;

FIGS. 4a and 4b illustrate two Doppler spectra containing tissue motion signal components;

FIGS. 5a-5d illustrate a process for removing tissue motion signal components from a Doppler spectrum in accordance with the principles of the present invention;

FIG. 6 is a phase diagram illustrating the discrimination of tissue motion signal components in accordance with the principles of the present invention;

FIG. 7 illustrates a digital filter for eliminating tissue motion signal components in accordance with the principles of the present invention;

FIGS. 8 and 9 illustrate Doppler spectra in which the absence of tissue motion signal components is discriminated; and

FIG. 10 is a block diagram of a preferred embodiment of a tissue motion signal component eliminator of the present invention.

Referring first to FIG. 1, an ultrasonic diagnostic system for measuring the flow rates of fluids in the body of a patient and constructed in accordance with the principles of the present invention is shown in block diagram form. A scanhead 10 containing one or more piezoelectric transducer elements transmits ultrasonic waves into the body of a patient and receives ultrasonic echoes returned by the tissue and fluid structures of the body. The returning echoes are converted to electrical signals by the transducer(s) and the signals are then digitized by an analog to digital converter 12. A plurality of such digital signal samples are combined in a beamformer 14 to form coherent ultrasonic information signals which are coupled to a digital demodulator 16. The fluid flow velocity information contained in the ultrasonic information signals is encoded as phase shifts of the returning echo signals in relation to a reference signal. Accordingly the demodulation process involves a translation of the ultrasonic information signals to an intermediate frequency range and the resolution of the phase information into two components, an in-phase (I) component and a quadrature (Q) component. The output of the digital demodulator 16 thus is a stream of corresponding I and Q phase information signals containing information as to the velocity of fluids and tissue in the patient's body.

In accordance with the principles of the present invention, the phase information signals are applied to a tissue component eliminator 18. The tissue component eliminator processes the signals by discriminating signal components representing the motion of tissue relative to the scanhead 10, and eliminating these components from the Doppler spectrum. I and Q velocity information signals from which the effects of tissue motion have been removed are produced at the output of the tissue component eliminator.

The I and Q information signals from which the tissue signal components have been removed are applied to a spectral estimator 20, which determines the fluid velocities as represented by the Doppler content of the signals. The velocity information signals produced by the spectral estimator 20 are then applied to a scan converter where, in a preferred embodiment, the velocity information is translated to a color and an intensity and stored locationally as a function of the position in the scanned area of the body at which the measurement was taken. The scan converter thus arranges the flow velocities in an image format suitable for a color flow image, and the image is then displayed on an image display 24. Other image formats such as numerical and graphical representations may also be employed to display the flow velocities to the user.

FIG. 2 depicts the manner in which Doppler data may be acquired to form a color flow image. The scanhead 10 sequentially transmits ultrasonic pulses in a plurality of vector directions over the scanned area 26 of the patient's body. One of these vector directions is indicated by arrow V.sub.n in FIG. 2. As echoes are returned from tissue and fluids along direction V.sub.n, signal samples are taken from echoes returning from sequential positions along the vector. These positions are referred to herein as range cells, some of which are labelled C.sub.1, C.sub.2, C.sub.3, . . . C.sub.n in the Figure. The signal samples are processed by the beamformer 14 and converted by the digital demodulator 16 to I and Q quadrature signal samples.

This sequence of pulse transmission and echo reception is repeated a number of times along vector direction V.sub.n. The transmitted pulses of four of these sequences are labelled P.sub.1, P.sub.2, P.sub.3, and P.sub.4 in the chart of FIG. 3. The I,Q echo samples taken in response to each of these pulse transmissions are shown below each pulse in FIG. 3. Each horizontal row of I,Q samples in FIG. 3 thus represents signal samples taken at a particular range cell location C.sub.1 . . . C.sub.n, but at different points in time. An array of data of the form illustrated in FIG. 3 is operated upon by the tissue component eliminator 18 to remove the effects of tissue motion from the signals.

The Doppler velocity information content of the I,Q signal samples may be understood by reference to a representative Doppler spectrum as shown in FIG. 4a. This spectrum is shown as bounded by end points -PRF/2 and +PRF/2 although as will be subsequently explained the spectrum is a continuum through these end points. The end points -PRF/2 and +PRF/2 are the Nyquist limits of the signal sampling rate at a particular range cell, the sampling rate being the pulse repetition freguency (PRF) of pulses P.sub.1, P.sub.2, P.sub.3, etc. of FIG. 3. In the center of the spectrum is DC, or zero, the Doppler frequency of tissue and fluids at rest. The Doppler phase information contained in the I,Q samples increases in the positive sense (+.phi.) or negative sense (-.phi.) toward the limit values from zero, depending on the velocity of the material at the range cell and its direction.

The waveform of FIG. 4a is seen to contain two peaks, a major one 30 and a lesser one 32. The major peak 30 results from echo information returned by tissue and the minor peak 32 results from echo information returned by a fluid such as blood. Peak 30 may be as much as 60-80 dB greater than peak 32 because tissue is a much stronger reflector of ultrasonic waves than blood. In FIG. 4a peak 30 is located at zero, a condition that exists when the reflecting tissue is substantially at rest, that is, there is substantially no motion component from the tissue. Elimination of the spectral components of peak 30 may be accomplished by high pass filtering the signals of the waveform of FIG. 4a, as the skirts of the rejection band defined by the high pass filter will be displaced symmetrically about zero, the frequency about which the signal components of peak 30 resulting from substantially stationary tissue are displaced. This known technique is in current use, as conventional color flow systems have a need to eliminate signals returned by stationary structures prior to operating on the Doppler flow content of the signals. Signal components resulting from substantially stationary tissue structure are not within the definition of "tissue motion components" as that term is used herein.

The problem confronted by the present invention is spectrally represented by FIG. 4b. In this Figure moving tissue has generated the signal components comprising peak 30'. FIG. 4b differs from FIG. 4a in that the peak 30' is displaced from zero by reason of the motion of the tissue. In this example, the blood velocity peak 32 is displaced on the opposite side of the zero reference by approximately the same magnitude as the displacement of peak 30'. An effort to remove the signal components comprising peak 30', as by notch or high pass filtering, would also eliminate the signal components comprising the blood velocity peak 32 by reason of the spectrally symmetrical effect of the filter. In sum, a process premised on the assumption that the tissue component is closer to zero than the blood flow component is ineffective in dealing with tissue motion in this situation.

A technique for overcoming this problem in accordance with the principles of the present invention is illustrated by the spectra of FIGS. 5a-5d. In the first step of the inventive technique, the frequency location .omega..sub.t of the tissue motion component 30' in FIG. 5a is determined. Once the location of the tissue component is determined, the signals of the entire Doppler spectrum are modulated by the frequency deviation of the tissue component such that the tissue motion component becomes located at zero, as shown in FIG. 5b. Due to the continuity of the spectrum through the positive and negative Nyquist limits, this spectral shift of the signal components results in a "wrap around" of a portion of the spectral content. In the illustrated example, the blood flow component 32 of FIG. 5a is wrapped around +PRF/2 and through -PRF/2 so that the blood flow component appears at the left side of the spectrum in FIG. 5b after modulation.

With the tissue motion component 30' now centered about zero, the spectral data is high pass filtered to remove the tissue component. The skirts 34, 34' of the high pass filter are shown in FIG. 5c, and the spectral content between the skirts as shown in dotted lines is effectively notched out of the spectrum.

With the tissue motion component thus eliminated, the spectral data is remodulated to relocate the blood flow component at its original spectral position. This is accomplished by remodulating the signals by the same deviation .omega..sub.t used above in FIG. 5b, but in an opposite sense.

In this example, the original modulation was by the deviation +.omega..sub.t, and thus the remodulation is by the value -.omega..sub.t. The spectral result of this remodulation is shown in FIG. 5d. With the tissue component thus removed from the flow velocity data, the signal information can now be operated upon to determine the detected flow of blood without contamination by the effects of tissue motion.

Digital apparatus for performing the inventive technique for eliminating the effects of tissue motion may be constructed to operate upon digital I,Q signal data in the following manner. A number of known digital techniques may be used to discriminate the location of the tissue component, a preferred one being an autocorrelator. This autocorrelator operates by creating an offset alignment of the I,Q samples from the range cell location being investigated with the complex conjugate of itself. Taking the I,Q data of range cell C.sub.1 of FIG. 3 for example, the offset alignment would appear as: ##EQU1## Vertically aligned samples are multiplied together and the products aggregated to form a result of the complex form

Using this complex result the arc tangent is calculated using Q'/I', and the result is the frequency of the tissue motion component expressed in radians, the angle .omega..sub.t. Stated another way, the correlator has estimated the mean velocity of the tissue component, .omega..sub.t, and this processing can be expressed mathematically as ##EQU2## assuming X.sub.k =I.sub.k +jQ.sub.k, where n is the number of samples and * indicates complex conjugate. FIG. 6 shows a phase diagram on which the complex value I'+jQ' is indicated, and the tissue component frequency -.omega..sub.t for the example of FIG. 5a. The left-hand end of the abcissa of the diagram is equal to both +PRF/2 and -PRF/2, the end limits of the linearly drawn spectra discussed previously. Through modulation of the I'+jQ' complex value the arrow shown in the Figure could be swept continuously around the diagram in either a clockwise or counterclockwise direction, passing through the limits +PRF/2 and -PRF/2 as it does so. It is this mathematical principle which causes the wrap around of data from one end of the linearly drawn spectrum to the other as mentioned above in conjunction with FIG. 5b.

The spectral data is modulated to shift the tissue motion component to zero by multiplying the I,Q data of the range cell location by a sampled complex sinusoid. This is done by multiplying the I,Q data of the cell by the complex expression cos .theta.+j sin .theta., where .theta. is equal to +.omega..sub.t determined by the correlator above. With the data thus shifted in frequency, it is filtered by processing by a high pass digital filter, a preferred one of which is the infinite impulse response (IIR) high pass filter diagrammatically shown in FIG. 7. In an actual implementation, the I and Q data samples are processed through initialized parallel filters of this form. The input data to the IIR filter, either I or Q, is applied at the input x(n), and the filtered data is produced at the output y(n). In FIG. 7 the a and b terms designate multiplying coefficients by which the data values are multiplied in determination of the filter response characteristics, Z.sup.-1 indicates a single time interval shift delay, and the dotted nodes are points at which data arriving from two paths of the filter are summed. The exact characteristics of the high pass filter cutoff are application dependent, in consideration of the extent of the expected motion artifact.

Following removal of the tissue motion component content of the data by the high pass filter, the data is remodulated in the same way it was modulated, but using a frequency of the opposite sense as above in the complex expression. That is, +.omega..sub.t is replaced by -.omega..sub.t, or the reverse, as appropriate. The spectral signal content is thereby shifted back to its original frequency domain location but with the effects of tissue motion removed.

It has been found that the accuracy of received Doppler information can vary from time to time as a result of random scattering effects and other factors. Such inaccuracies can result in the imprecise determination of the frequency location of the tissue motion component for a particular range cell. However, the present inventors have further found that, on the average, there is only a small deviation in the frequency location of the tissue component from one range cell to an adjacent range cell. The present inventors have taken advantage of this latter finding to overcome the aforementioned problem of tissue component locational inaccuracies. In accordance with a further aspect of the present invention, the discrimination of tissue component location is accomplished by first determining the complex value I'+jQ' of the I,Q data from a number of neighboring range cells. For instance, I'+jQ' can be determined for each of range cells C.sub.1 -C.sub.4 in FIG. 3. The complex values are then accumulated in a summation of the form

and this accumulated complex value is then used in the arc tangent calculation to determine .omega..sub.t for one or more of cells C.sub.1 -C.sub.4. In this way the correlation of tissue component location between neighboring range cells is utilized to resolve inaccuracies in Doppler signals at particular range cell locations.

Another problem that may be encountered in an implementation of the present invention is the reception of signal components containing only desired flow information, devoid of any tissue component contamination. Such information reception may occur, for instance, when the range cell is at the center of a chamber of the heart, where only blood flow is present. This condition is spectrally represented in FIG. 8. In such a case it may be desirable to inhibit operation of a system of the present invention, and it would certainly be desirable not to eliminate the blood flow component as though it were a tissue component. In accordance with a further aspect of the present invention, the present inventors have made use of their recognition of the disparity in characteristics between tissue motion components and fluid flow components (e.g., signal strength, bandwidth) in applying a threshhold comparison before inplementation of the technique of the present invention. As illustrated in FIG. 8, when the components of the spectrum do not exceed the threshhold, the inventive system is inhibited so that the blood flow component 32 is not inadvertently removed from the spectrum.

A preferred technique for implementing this threshhold comparison is to measure the power content of the Doppler spectrum. This may be accomplished digitally by integrating the received data values to effectively find the area under the curve containing the Doppler components. In FIG. 9, this area is shown by the diagonal lines under the spectral curve. The measured magnitude of the area is compared with a threshhold value and if the value is not exceeded, the tissue component eliminator is inhibited from removing any spectral signal content.

A digital embodiment which provides all of the above features is shown in block diagram form in FIG. 10. In this embodiment, an autocorrelator 40 is used to determine .omega..sub.t, and to determine spectral power. The spectral power value is compared with a threshhold value by the comparator 44, and the outcome of the comparison is used to open or close switch 42, controlling the frequency deviation .omega..sub.t by which the signals are modulated and remodulated. If the result of the comparison indicates that a tissue component is present, resulting either from moving or stationary tissue, switch 42 is closed and the frequency deviation of the tissue component .omega..sub.t is applied to complex modulator 46 to shift the spectrum, and to complex modulator 50 to shift the spectrum back after the tissue component has been eliminated. The complex modulator 50 receives a modulation value of the opposite sense as modulator 46 by reason of the inverting component 52. This technique is effective for removing both moving and stationary tissue components, as tissue motion components will be shifted in accordance with their deviation .omega..sub.t, as will stationary tissue components for which .omega..sub.t will be found to be zero.

If the result of the comparison determines that no tissue component is present, the spectrum of FIG. 8, for instance, switch 42 is opened and the modulators 46 and 50 are effectively receiving modulating values of zero. The incoming data is shifted by a deviation of zero; that is, no shift is effected. The high pass filter 48 will filter the data as above, removing signal content about zero, but this will not adversely affect the outcome of the process as the effect is only to remove zero velocity (no flow) information. Modulator 50 will then remodulate the data by zero and the unaffected non-zero velocity flow information 32 is produced at the output for further processing. In a typical color flow implementation, data at the zero (no flow) location is represented in black, a representation which is unaffected by the elimination of data from this spectral location in FIG. 8.