JP4644011B2 - Ultrasonic Doppler blood flow meter - Google Patents

Description

  The present invention relates to an ultrasonic Doppler blood flow meter that measures blood flow information in the body using an ultrasonic Doppler phenomenon and displays the blood flow information as an image.

  Ultrasound Doppler blood flow meter (color flow device) that uses ultrasonic Doppler phenomenon to display the blood flow velocity distribution in the living body corresponding to the color and superimposed with the black and white two-dimensional tomographic image (B mode image) )It has been known.

  FIG. 6 is a block diagram showing a configuration of an ultrasonic Doppler blood flow meter. The transmission unit 101 in this figure generates a signal for generating an ultrasonic wave. The probe 102 connected to the transmission unit 101 irradiates the living body with an ultrasonic pulse based on a signal from the transmission unit 101, and receives an ultrasonic pulse echo that is a reflected wave thereof and converts it into an electrical signal. The receiving unit 103 amplifies the converted electric signal. The phase detection unit 104 detects the phase of the amplified electrical signal and generates a Doppler shift signal. The A / D converter 105 converts the Doppler shift signal into discrete data.

  The MTI filter unit 106 has a high-pass digital filter, and removes low-frequency signals generated from unnecessary body tissue called clutter from discrete data. The velocity calculation unit 107 calculates the blood flow velocity using the Doppler shift signal from which the clutter component has been removed, and forms blood flow velocity distribution image data. The envelope detection unit 108 performs envelope detection from the electric signal from the reception unit 103 and generates B-mode image data. A digital scan converter (DSC) 109 mixes and scan-converts the blood flow velocity image data from the velocity calculation unit 107 and the B-mode image data from the envelope detection unit 108. The monitor 110 displays a two-dimensional blood flow image based on the signal from the DSC 109.

  The MTI filter unit 6 uses an infinite impulse response filter (IIR filter) because a steep cutoff characteristic can be obtained at a low frequency. The IIR filter has a non-linear type with a remarkable phase characteristic, and has a very large transient response. Therefore, a high frequency component is generated in the data after filtering. This high-frequency component does not reflect blood flow information, and becomes a factor that deteriorates the estimation accuracy of blood flow information.

  As a method for avoiding such deterioration in blood flow velocity estimation accuracy caused by a transient response, there is a method of discarding data of a portion of the filter output where a transient response appears to appear. However, the larger the transient response, the smaller the number of data that can be used in the speed calculation unit 7, leading to a decrease in the signal to noise ratio. In order to discard the data where the transient response appears in the filter output and increase the signal-to-noise ratio, it is necessary to secure a sufficient number of data that can be used in the speed calculator 7. For this reason, a large amount of Doppler data is required, and the number of times of transmission / reception of ultrasonic signals in the same direction has to be increased.

  The cause of this transient response will be described with reference to FIG. FIG. 7 is a diagram showing data input to the filter and data recognized by the filter. If the time when data is input to the MTI filter unit 106 is 0, the input starts from x (0), but the previous data is recognized as a value 0 as a filter. Therefore, at the negative time before time 0, the filtering target data value input to the filter was 0, but at time 0, the target data value suddenly changes from 0 to x (0), that is, at time 0. The discontinuity in the data causes the transient response. In order to mitigate the influence of this discontinuity, in general, initial value subtraction processing (for example, see Patent Document 1), extrapolation of tapering data (for example, see Patent Document 2), and initial value setting of a filter There is a countermeasure (for example, see Patent Document 3).

  The initial value subtraction process performs filtering on data obtained by subtracting the target data value x (0) at time 0 from the target data value x (n). By subjecting the target data to the subtraction process in this way, the discontinuity of the target data value that the data at time 0 changes suddenly in a step shape is alleviated. However, discontinuity remains in the differential coefficient of the target data at time 0, and a high frequency component due to a transient response appears in the filtered data.

  A countermeasure for avoiding this is tapering data extrapolation. This eliminates the discontinuity of the differential coefficient at time 0 by extrapolating a tapered data string in which the differential coefficient of data before and after time 0 is continuous before time 0, Since the start point of the extrapolated data can be regarded as a pseudo time 0, the entire original data string is kept away from the temporal influence range of the transient response.

Since the clutter component is composed of a direct current or low frequency component having a larger amplitude than the blood flow component, the initial value setting of the filter is performed by setting the input data of the filter to a unit step input having the same magnitude as the value at time 0. It is approximated, and it is assumed that it has continued infinitely before time 0, and is set as the initial value of the filter at time 0. However, this means alleviates the discontinuity of the data at time 0, but does not eliminate the discontinuity of the differential coefficient of the data.
JP-A-63-84532 JP 2002-143162 A JP-A-10-127640

  However, the above conventional measures reduce the influence of the transient response by artificially processing the data value to be filtered. Depending on the cutoff characteristic of the filter, the size of the clutter, or the frequency, the transient response is sufficient. In some cases, a part of Doppler shift data used for estimating blood flow information must be discarded. Therefore, there is a problem that the clutter removal capability and the signal-to-noise ratio are improved, and it is insufficient for estimating blood flow information with high accuracy.

  The present invention solves the above-mentioned conventional problems. By reducing the phase distortion in the filter regardless of the frequency and the cutoff characteristic of the filter, the transient response by the filter can be suppressed, and blood flow information can be estimated. Providing an ultrasonic Doppler blood flow meter capable of estimating blood flow information with high accuracy by eliminating the clutter component and improving the signal-to-noise ratio at the time of blood flow information estimation because the data to be used is not discarded. The purpose is to do.

In order to achieve the above object, a first ultrasonic Doppler blood flow meter according to the present invention transmits / receives an ultrasonic pulse and obtains discrete data, and a low-frequency clutter component for the discrete data. Filtering means for removing, and means for estimating blood flow information from the discrete data from which the clutter has been removed, the filtering means being correlated with the discrete data before and after the discrete data, and a data endpoint A data extrapolating means for adding extrapolated data having a value smoothly connected to the data value at the end of the discrete data, a first high-pass filter for filtering the discrete data, and the first A first data reverse order part for inverting the data string of the output data from the high-pass filter with respect to the time axis, and the output data from the first data reverse order part A second high-pass filter for filtering, and a second data reverse order unit for inverting a data string of output data from the second high-pass filter with respect to a time axis, shows the as-pass filter a second high pass filter is the same frequency response, the data extrapolation means, the extrapolation data of point-symmetrical values with respect to data of the joint portion before and after the input data The discrete data to which the extrapolated data is added is transmitted to the first high-pass filter . With this configuration, a transient response due to data filtering can be suppressed.

Further, the second ultrasonic Doppler blood flow meter according to the present invention transmits / receives an ultrasonic pulse, transmits / receives the obtained signal as discrete data, and removes low-frequency clutter from the discrete data. Filtering means and means for estimating blood flow information from the discrete data from which the clutter component has been removed, wherein the filtering means is correlated with the discrete data before and after the discrete data, and at a data end point. Data extrapolation means for adding extrapolated data having values that are smoothly connected to the data values at the ends of the discrete data, a first data reverse order section for inverting the discrete data with respect to the time axis, and inverting A first high-pass filter for filtering the discrete data and a data string of output data from the first high-pass filter that is inverted with respect to the time axis. And a second high-pass filter for filtering the output data from the second data reverse-order part, wherein the first high-pass filter and the second high-pass filter are shows the same frequency response, the data extrapolation means adds the extrapolated data of the point-symmetrical values with respect to data of the joint portion before and after the input data, the discrete data added with the extrapolated data the It transmits to the 1st data reverse order part, It is characterized by the above-mentioned. Also with this configuration, the same effect as that of the first ultrasonic Doppler blood flow meter can be obtained.

  In addition, the filtering unit may include a control unit that sets initial value data of a predetermined value as data before the discrete data is input to the filtering unit. With this configuration, data discontinuity at the time of data input can be prevented.

  The filtering means may include a part or all of the data corresponding to the data added by the data extrapolation means with respect to the data sent from the second data reverse order section or the second high-pass filter. It is also possible to adopt a configuration having a data discarding means for discarding.

  According to the present invention, by reducing the phase distortion in the filter regardless of the frequency and the cutoff characteristic of the filter, the transient response by the filter can be suppressed, and the data used for the estimation of blood flow information can be discarded. Therefore, it is possible to provide an ultrasonic Doppler blood flow meter capable of improving the ability to remove clutter components and the signal-to-noise ratio when estimating blood flow information, and capable of estimating blood flow information with high accuracy.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  First, the configuration of the ultrasonic diagnostic apparatus in the present embodiment will be described. FIG. 1 is a block diagram showing an example of the configuration of the MTI filter unit in the Doppler blood flow meter of the present invention. This MTI filter unit 1 (filtering means) corresponds to the MTI filter unit 106 of the conventional ultrasonic Doppler blood flow meter shown in FIG. The MTI (Moving Target Indicator) filter is a high-pass digital filter. Since the ultrasonic Doppler blood flow meter in the present embodiment is the same as the conventional example shown in FIG. 6 except for the MTI filter unit 1, the description thereof is omitted.

  The data extrapolation unit 2 (data extrapolation means) extrapolates data to the digitized blood flow data detected by the transmission unit 101, the probe 102, and the reception unit 103 (transmission / reception unit) shown in FIG. The first high-pass filter 3 is composed of an IIR filter and cuts a low-frequency component of extrapolated data. The first data reverse order unit 4 reverses the temporal order of the data that has passed through the first high-pass filter 3. The second high-pass filter 5 is composed of an IIR filter, has the same frequency response as the first high-pass filter 3, and the second data reverse order unit 6 is the same as the first data reverse order unit 4. It has the following characteristics. The data discarding unit 7 (data discarding unit) discards the data corresponding to the data added by the data extrapolation unit 2, and uses the remaining data to estimate the blood flow rate. To the means). The control unit 8 (control means) performs control for adding and discarding extrapolated data and setting an initial value. The first high-pass filter 3 and the second high-pass filter 5 are preferably IIR filters. However, the present invention is not limited to this, and other filters can be used.

Next, the principle of a method for suppressing the transient response of the filter regardless of the frequency and the cutoff characteristic of the filter in the configuration like the MTI filter unit 1 will be described. 2A, 2B, and 2C are block diagrams showing the principle of suppressing the transient response of the filter. An example of discrete data of input / output signals x (n), y _f (n), y _f (−n), y _fb (n), and y _fb (−n) of each block is shown below each input / output signal. .

In FIG. 2A, the first filter 11 and the second filter 13 are general filters showing the same frequency response characteristics, and the first data reverse order unit 12 and the second data reverse order unit 14 are data. It reverses the temporal order of. The input data of FIG. 2A is x (n), the impulse responses of the first filter 11 and the second filter 13 are h (n), and the data filtered (forward filtered) by the first filter 11 is y _f ( n) and the respective z transformations are X (z), H (z), and Y _f (z),
Y _f (z) = H (z) X (z)
The relationship holds.

The first data reverse portion 12, y _f (n) is inverted the temporal order of the data data string y _f (-n) is generated. The inverted data string is subjected to filtering (reverse filtering) in the second filter 13, and the z conversion of the output y _fb (n) is performed by converting the y conversion of y _f (−n) to Y _f (1). / Z),
Y _fb (z) = Y _f (1 / z) H (z) = H (z) H (1 / z) X (1 / z)
It becomes. Furthermore, z conversion of the data string y ^R _fb (n) = y _fb (−n) obtained by inverting y _fb (n) with respect to the time axis is
Y ^R _fb (z) = Y _fb (1 / z) = H (z) H (1 / z) X (z)
It becomes.

FIG. 2B is a block diagram in which a series of processing steps of the filter shown in FIG. 2A is regarded as one synthesis filter G15. From the above equation, the transfer function G (z) of the synthesis filter G15 is
G (z) = H (z) H (1 / z)
It becomes. The frequency characteristic G (e ^jw ) of this transfer function obtained by substituting z = e ^jw into the above equation is G (e ^jw ) = H (e ^−jw ) H (e ^jw ) = | H (e ^jw ) | ² and no phase distortion occurs. Therefore, in a series of processing steps of the filter, the phase distortion becomes 0 regardless of the frequency and the cutoff characteristic.

That is, the input data x (n) is filtered by the first filter 11 (forward filtering), and the data string y _f (−n) obtained by temporally inverting the obtained data y _f is converted to the second filter 13. the perform filtering (reverse filtering), the sequence obtained y _fb inverts the data sequence obtained by y ^R _fb is the phase difference between x (n) becomes 0 regardless of the frequency.

  FIG. 2C is a block diagram illustrating a signal processing process of a filter for an input signal having another configuration. 2C, the first filter 11 and the first data reverse order unit 12 and the second filter 13 and the second data reverse order unit 14 are respectively provided to the data input side of the configuration shown in FIG. 2A. It is the replaced structure.

Therefore, the data x (−n) obtained by reversing the temporal order of the input data x (n) is filtered by the first filter 11 (reverse filtering), and the obtained data y _b (n) is inverted. If the data string y _b (−n) is filtered by the second filter 13 (forward filtering), the respective z-transforms X (z) and Y _{bf of the} input data x (n) and y _bf (n) (Z)
Y _bf (z) = H (z) H (1 / z) X (z)
The relationship holds. The frequency characteristic of this transfer function also has no imaginary term. That is, even when the forward filtering is performed after the backward filtering, the phase difference between the input data and the output data becomes 0 regardless of the data frequency and the cutoff characteristic of the filter due to the series of processing of the filter.

However, although the above description holds theoretically, in practice, the phase transition cannot be completely eliminated. This is because the data string y _fb obtained in forward / reverse filtering in which backward filtering is performed after performing forward filtering and the data string y _bf obtained in reverse forward filtering in which forward filtering is performed after performing backward filtering are: Theoretically, if the input data is the same, they should match, but in reality, the symmetry of the phase transition in both filtering is not guaranteed.

  Therefore, in general, the initial value of the filter is set so as to guarantee this symmetry as much as possible. The initial value setting method of the filter is optimally estimated by the least square method using the waveform after forward / reverse filtering and the waveform after reverse forward filtering so as to guarantee the symmetry between the two. The initial value setting method is not limited to this, and any method can be used as long as symmetry can be ensured as much as possible.For example, the initial value is estimated from the filter state equation or the initial input data value is initialized. You may use the method made into a value.

  Next, an example of the MTI filter unit 1 that performs initial value setting using the above principle will be described. FIG. 3 is a block diagram illustrating a configuration example of an IIR filter that is a high-pass filter constituting the MTI filter unit 1.

  The high-pass filter includes multipliers 31 to 35, adders 36 to 38, registers 39 and 42, selectors 40 and 43, and lines 41 and 44. Registers 39 and 42 are connected to the selectors 40 and 43 on the ON side, and initial value lines 47 and 48 from the control unit 8 in FIG. 1 are connected to the OFF side. It is controlled by the control unit 8 via the setting lines 45 and 46. Signals b3, b2, b1, a3, and a2 input to the multipliers 31 to 35 indicate filter coefficients of the IIR filter. The registers 39 and 42 temporarily store signals.

The IIR filter shown in FIG. 3 assumes that the input at time n is x (n) and the output is y (n).
y (n) = b1x (n) + b2x (n-1) + b3x (n-2)-(a2y (n-1) + a3y (n-2))
The transfer function H (z) is
H (z) = (b1 + b2z ⁻¹ + b3z ⁻² ) / (1 + a2z ⁻¹ + a3z ⁻² )
Thus, a desired transfer function can be obtained by setting the filter coefficient.

  The operation of the MTI filter of the ultrasonic Doppler blood flow meter configured as described above will be described with reference to FIGS. 1, 3, 4, and 5. 4 and 5 are timing charts of data in the data extrapolation unit 2 and the first high-pass filter 3, respectively.

  First, as shown in FIG. 4A, the Doppler shift signal data string from the same part in the subject sent from the A / D conversion unit 105 is controlled by the data extrapolation unit 2 from the control unit 8. When the signal is ON, extrapolation data is added (FIG. 4 (c)). When the control signal is OFF, extrapolation data is not added (FIG. 4 (b)). Input to the first high-pass filter 3. Clutter components are removed from the data input to the first high-pass filter 3 according to the cutoff characteristics.

  At that time, as described above, it is necessary to set an initial value of the first high-pass filter 3. As shown in FIG. 3, the first high-pass filter 3 includes selectors 40 and 43 controlled by the control unit 8, initial value setting lines 45 and 46 for transmitting control signals for controlling the selectors 40 and 43, and an initial value. Value lines 47 and 48 are provided, and the data output to the lines 41 and 44 can be switched to the values of the initial value lines 47 and 48 set by the control unit 8 or the values of the registers 39 and 42.

  For example, before the first data received by the ultrasonic diagnostic apparatus is input from the extrapolation unit 2, appropriate initial values are output from the control unit 8 to the initial value lines 47 and 48, and initial value setting is performed. The control unit 8 sets the selectors 40 and 43 to be connected to the signal on the OFF side as shown in FIG. Accordingly, as shown in FIGS. 5B and 5C, the initial values X1 and X2 of the first high-pass filter 3 are output to the lines 41 and 44, thereby setting the initial value of the filter. Note that the present invention does not limit the initial value setting method such as the timing of the control signal described above, and it is sufficient that the initial value of the filter can be set appropriately.

  After completion of the initial setting, data is input, and the selectors 40 and 43 are connected to the ON-side signal as shown in FIG. 5 (d) by the control unit 8 via the initial value setting lines 45 and 46. Thereby, the first high-pass filter 3 performs filtering on the data from the data extrapolation unit 2 as shown in FIG.

  Next, the filtered data is transmitted to the first data reverse order unit 4. The data received by the first data reverse order unit 4 is reversed with respect to the temporal order of the data. The inverted data is transmitted to the second high-pass filter 5 having the same transfer function as that of the first high-pass filter 3. The data filtered by the second high-pass filter 5 is transmitted to the second data reverse order unit 6 and inverted with respect to the temporal order of the data.

  The inverted data is transmitted to the data discard unit 7. The data discarding unit 7 discards the data corresponding to the extrapolated data when extrapolated data is added in the data extrapolating unit 2, but whether to discard the data is the case of extrapolating the data Similarly, when the data is controlled by a signal from the control unit 8 and is not discarded, the inverted data is output as it is.

  It is desirable that the position and the number of data to be discarded are the same as the position and the number of data extrapolated in the data extrapolation unit 2. However, in the present embodiment, it is not limited to the same case, the transient response of the output data can be suppressed, and the signal-to-noise ratio at the time of blood flow information estimation in the speed calculation unit 7 can be appropriately maintained. If it is possible, the position and number of data to be discarded do not have to be the same as the position and number of data added.

  The output data is calculated by the speed calculation unit 107, whereby the blood flow is measured. Hereinafter, it is the same as that of a conventional ultrasonic Doppler blood flow meter, and a description thereof will be omitted.

  The ultrasonic Doppler blood flow meter according to the present invention is configured to invert a data string obtained by performing backward filtering after performing forward filtering in order to suppress a transient response in the MTI filter unit 6. However, this invention is not limited to this, For example, it can also be set as the structure which performs forward direction filtering after performing reverse direction filtering.

  The data added by the data extrapolation unit 2 is related to the discrete data, and can be configured to be smoothly connected to the discrete data at the junction with the discrete data. Here, the term “smoothly” means that data are continuously joined, and the slope formed by adjacent data does not change abruptly at the joined portion of data. Preferably, the data to be added is point symmetric with respect to the discrete data at the joint.

  With the ultrasonic diagnostic apparatus as described above, the transient response of the filter can be suppressed, and data used for blood flow information estimation is not discarded. It is possible to improve the signal-to-noise ratio and estimate blood flow information with high accuracy.

  Although the first high-pass filter 3 and the second high-pass filter 5 have the same frequency response, they do not have to be exactly the same, and can be a filter in a range that can suppress a transient response. That's fine.

  Further, in order to remove a clutter component having a low frequency, an MTI filter that passes a high frequency band is used. However, a similar effect can be obtained by using a bandpass filter that passes a frequency component of Doppler blood flow data. .

  Since the ultrasonic Doppler blood flow meter of the present invention can remove clutter components without wastefully discarding data used for blood flow information estimation, the clutter removal capability and the signal-to-noise ratio when blood flow information is estimated Is improved, and blood flow information can be estimated with high accuracy, which is useful for diagnosis of blood flow dynamics in the medical field.

The block diagram which shows the structure of the MTI filter in embodiment of this invention The conceptual diagram which shows the principle which suppresses the transient response of the filter in this Embodiment Conceptual diagram showing the principle of suppressing the transient response of the filter Conceptual diagram showing the principle of suppressing the transient response of the filter The block diagram which shows the structure of the high-pass filter in embodiment of this invention Timing chart of data extrapolation unit in the embodiment of the present invention Timing chart of high-pass filter in the embodiment of the present invention Block diagram showing the configuration of a conventional ultrasonic Doppler blood flow meter Diagram showing data input to the filter and data recognized by the filter

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 MTI filter 2 Data extrapolation part 3, 11 1st high-pass filter 4, 12 1st data reverse order part 5, 13 2nd high-pass filter 6, 14 2nd data reverse order part 7 Data discard part 8 Control Unit 15 Synthesis Filter G
31-35 Multiplier 36-38 Adder 39, 42 Register 40, 43 Selector 41, 44 Line 45, 46 Initial value setting line 47, 48 Initial value line

Claims (4)

Transmitting and receiving means for transmitting and receiving ultrasonic pulses and obtaining discrete data;
Filtering means for removing low frequency clutter components from the discrete data;
Means for estimating blood flow information from discrete data from which the clutter has been removed,
The filtering means includes
Data extrapolation means for adding extrapolated data having a value that is correlated with the discrete data before and after the discrete data and is smoothly connected to the data value at the end of the discrete data at the data end point;
A first high pass filter for filtering the discrete data;
A first data reverse order unit that inverts a data string of output data from the first high-pass filter with respect to a time axis;
A second high pass filter for filtering the output data from the first reverse data portion;
A second data reverse order part for inverting the data string of the output data from the second high-pass filter with respect to the time axis,
Said first high pass filter the second high pass filter shows the same frequency response,
The data extrapolation means adds extrapolated data having a point-symmetrical value with respect to joint data before and after the input data, and outputs the discrete data to which the extrapolated data is added to the first high-pass filter. An ultrasonic Doppler blood flow meter characterized by transmitting . Transmitting and receiving means for transmitting and receiving ultrasonic pulses and making the obtained signal discrete data;
Filtering means for removing low frequency clutter on the discrete data;
Means for estimating blood flow information from discrete data from which the clutter component has been removed,
The filtering means includes
Data extrapolation means for adding extrapolated data having a value that is correlated with the discrete data before and after the discrete data and is smoothly connected to the data value at the end of the discrete data at the data end point;
A first data reverse order part for inverting the discrete data with respect to the time axis;
A first high pass filter for filtering the inverted discrete data;
A second data reverse order part for inverting the data string of the output data from the first high-pass filter with respect to the time axis;
A second high-pass filter for filtering output data from the second data reverse order part,
Said first high pass filter the second high pass filter shows the same frequency response,
The data extrapolation means adds extrapolated data having a point-symmetrical value to the joint data before and after the input data, and transmits discrete data to which the extrapolated data is added to the first data reverse order section. An ultrasonic Doppler blood flow meter.   The ultrasonic Doppler blood flow meter according to claim 1 or 2, wherein the filtering means includes control means for setting initial value data of a predetermined value as data before the discrete data is input to the filtering means. The filtering means discards a part or all of the data corresponding to the data added by the data extrapolation means for the data sent from the second data reverse order part or the second high-pass filter. The ultrasonic Doppler blood flow meter as described in any one of Claims 1-3 which has a data discard means to perform.

Images