JPH07109615B2 - Ultrasonic diagnostic equipment - Google Patents

Description

Description: TECHNICAL FIELD OF THE INVENTION The present invention relates to an ultrasonic diagnostic apparatus configured to obtain an ultrasonic tomographic image by performing ultrasonic transmission / reception by sector scanning, and particularly, to reduce pixel distortion and The present invention relates to an ultrasonic diagnostic apparatus designed to improve S / N.

[Technical background of the invention and its problems]

The so-called sector scanning type ultrasonic diagnostic apparatus is configured to scan the transmitted and received ultrasonic waves in a fan shape by controlling the transmission and reception phase of the ultrasonic transducer group to obtain an ultrasonic tomographic image. In the ultrasonic transmission / reception system in sector scanning,
The reflected sound field distribution for which information is collected is expressed in the polar coordinate system (r, θ), and in the image display system, the image-displayed pixels are expressed in the orthogonal coordinate system (x, y). Therefore, if the information represented by the polar coordinate system (r, θ) is attempted to be represented by the orthogonal coordinate system (x, y), a positional deviation occurs as shown in FIG. , θ) causes so-called pixel distortion in the orthogonal coordinate system (x, y), and biometric information cannot be represented correctly.

As a method for solving such a problem, the fifth method has been conventionally used.
The method shown in FIG. 5B and FIG. 5C is used.
That is, FIG. 5B shows that the data of the sample points in the polar coordinate system (r, θ) is directly allocated to the nearest position in the orthogonal coordinate system (x, y). According to this method, the error in the shaded area in the figure occurs, and the pixel distortion of the image is eliminated to some extent, but the error in the shaded area in the figure occurs, which is not perfect.

On the other hand, FIG. 5 (c) shows that the data at the display point in the Cartesian coordinate system (x, y) is calculated from the plurality of data of the surrounding sample points in the polar coordinate system (r, θ). For example, the calculation formula shown below is used.

According to this method, the error can be eliminated to a considerable extent, but when m >> n, y≈x _k . Therefore, the pixel distortion is eliminated, but the S / N directly related to the display image is not improved, and improvement is desired.

[Object of the Invention]

The present invention has been made based on the above circumstances, and an object thereof is to provide an ultrasonic diagnostic apparatus capable of reducing pixel distortion in sector scanning and improving S / N.

[Outline of Invention]

To achieve such an object, the present invention obtains tomographic image information of the subject by adding received signals obtained by performing ultrasonic wave reception to the subject by sector scanning and detecting the added signal. In the ultrasonic diagnostic apparatus configured to display an image by A / D conversion means for digitizing the detection signal, and echo data for each pixel of the polar coordinate system by the sector scanning output from the A / D conversion means A buffer memory in which the converted address data obtained by converting the polar coordinate system address of the echo data into the rectangular coordinate system address is written into the same address, and a sound based on the converted address data in the buffer memory. The sound field pattern from the first means is defined by first means for determining a field pattern and coefficient data indicating a plurality of filter characteristics stored in advance. Second means for selecting the coefficient data based on the data, the third means for multiplying the corresponding echo data by the coefficient data selected by the second means, and the echo data output from the third means. Fourth means for setting the sum of the coefficient values of the group to 1 and frame data for display by taking in the echo data for each pixel output from the fourth means with the orthogonal coordinate system address set And a frame memory that operates.

Example of Invention

First, the principle of the present invention will be described prior to the description of the embodiments of the present invention. Prior to this application, Nagao et al. Made a detailed study on digital image processing (“Digital Pixel Processing”, translated by Shin Nagao, Ryodai Kagakusha Showa).
Issued December 10, 53). By referring to this known document, the following consideration can be obtained regarding the sampling of an image using a sampling grid and the ultrasonic diagnostic apparatus to which the present invention is directed.

That is, generally, in order to reproduce the original pixel from the sampled value, the convolution of the interpolation function is performed as shown in (1) below.
Done by

Next, when reproducing the sample points of the polar coordinate system (r, θ) of the sector scan as the original image on the frame memory of the display system,
The meaning of each variable in the above equation (1) will be described.

When the above formula (1) is converted into a polar coordinate system (r, θ) formula, the following formula (2) is obtained.

In the above formulas (1) and (2), f (x, y) and f (r, θ) are original image data, f (mX, nY) and f (mR, nΘ) are sampled data, g (X-mX, y-nY), g (r-mR, [theta] -n [Theta]) are interpolation functions, and r, [theta] are orthogonal coordinate system (x, y) on the frame memory, respectively. a) As shown Is.

R and Θ are polar coordinate systems (r, θ), respectively, as shown in Fig. 1 (b).
Is the basic unit of the spatial sampling period.

The interpolation function in the above is a real space expression of the frequency characteristic of the filter, and is the coefficient itself of the digital filter.

When the above equation (2) is Fourier transformed, it becomes the following equation (3).

Here, Q = RΘ. Further, the constants k and l respectively indicate the spatial frequency, and u and v respectively indicate the number of repetitions of the spatial spectrum. Further, as is apparent by referring to the above-mentioned publicly known document, the transformation of the equation from the above equation (2) to the above equation (3) is described in the document “DP Peterson and
D. Middleton, Sampling and reconstruction of wave−nu
mber−limited functionin n−dimensional euclidean
spaces, Informat.contr. 5, 1962, 279-323. ".

In order to completely reproduce the original image, the frequency characteristic of the interpolation function is required to be flat, and it is necessary to satisfy the following expression (4).

G (u, v) = Q = R · Θ (4) In the above, in order to completely reproduce the original image, it is necessary that the frequency characteristic of the interpolation function is flat, but it is flat. A certain bandwidth does not have to cover the entire bandwidth and may be set as follows. That is, it is sufficient to filter characteristics of interpolation function with a bandwidth enough to sufficiently represent the original image, in particular with respect to the original image F _a band is wide as shown in FIG. 2 (a) , An interpolation function G _a having a wide rectangular window is selected, and an interpolation function G a having a narrow rectangular window is applied to the original image F _b having a narrow band as shown in FIG. 2B. Select _b .

Here, in order to reproduce the original image in a perfect form, it suffices to select an interpolation function of a rectangular window having a bandwidth of 1/2 of the sampling frequency, but in this case, the bandwidth of the rectangular window is narrowed. By doing so, it becomes possible to cut excess frequency components and improve S / N.

Inverse Fourier transform of the equation (4) gives the following equation (5).

The above equation (5) is shown in FIGS. 3 (a) and 3 (b), which correspond to FIGS. 2 (a) and 2 (b), respectively. When the width of the rectangular window is narrowed, the proper interpolation function is obtained. Since the bandwidth is widened, the images are sufficiently averaged and noise can be reduced. Moreover, since the cut-off frequencies u _c and v _c exist outside the band of the original image, the original image is not disturbed. Also, in sector scanning, polar coordinate system (r,
If the size of the spread function does not change much regardless of the position at θ), the raster interval spreads at rθ, so the frequency characteristic normalized by the sampling cycle of the original image is higher as r is larger, and smaller as r is smaller. Frequency characteristics are low. That is, the spatial frequency characteristic of the portion A shown in FIG. 1 (a) is low as shown in FIG. 2 (b).
The spatial frequency characteristic of the portion B in the figure is high as shown in FIG.

Therefore, in the sector scanning, the frequency characteristic of the original image changes greatly depending on the position, so that the width of the rectangular window can be narrowed by that much, and the degree of S / N improvement can be increased.

Therefore, in the prior art, the writing process to the pixels on the raster and the interpolation process to the empty pixels that were not written at that time were separately performed, whereas in the present invention, the raw echo is recorded in the frame memory. Instead of writing data, the processing listed below is executed.

That is, the result obtained by performing the convolution operation of the interpolation function and the sample value of the echo data is written in all the pixels of the frame memory (corresponding to the equation (2)).

The interpolation function in is the function represented by the above equations (4) and (5).

The cutoff frequency of the interpolation function is changed according to the spread function, that is, the magnitude of the resolution with respect to the sampling period to improve the S / N.

Here, the data y _n written in the frame memory is represented by the following equation (6) by the echo data x _k and the coefficient a _k .

Here, it is necessary to have a value close to a _o near the writing point, and if the sound field is small, a _k / a _o cannot be too large and the original image described above can be sufficiently expressed. It suffices to apply the filter characteristic of the interpolation function having a bandwidth of the degree.

An embodiment of the ultrasonic diagnostic apparatus according to the present invention based on the above principle will be described with reference to FIG.

In FIG. 4, reference numeral 1 is, for example, a sector scanning array type ultrasonic probe in which a plurality of ultrasonic transducers are arranged in parallel. Reference numeral 2 is a transmitter that applies a delayed excitation pulse in sector scanning to each transducer of the probe 1.

Reference numeral 3 is a receiver in which the reception delay amount is set in correspondence with the excitation delay amount of the transmitter 2 and receives the reception signal in the sector scanning from each transducer of the probe 1. Reference numeral 4 denotes an adder that adds the received signals from the vibrators output from the adder 3. Reference numeral 5 is a detector for detecting the added reception signal from the adder 4 to obtain tomographic image information for one raster in the sector scanning.

Reference numeral 6 is an analog-digital converter (A / D-C) for analog-digital converting the output from the detector 5. 7
Is a Di that converts tomographic image information (echo data) digitized by A / D-C6 from ultrasonic scan to TV scan.
It is a gital scan converter (DSC), and its details are configured as follows.

In other words, this DSC7 is a buffer memory that takes in the echo data in pixel units for each raster and the conversion address (the lower several bits and the upper several bits below the decimal point) obtained by converting the polar coordinate system address into the rectangular coordinate system address. Group (BM) 7A
And a register group (Res) 7B that sequentially takes in echo data in pixel units and its conversion address data from the BM 7A,
One of the subtracter (Sub) 7C that calculates the difference between the previous address data and the current address data that is loaded into this Res 7B, and one of the coefficient data that shows the filter characteristics stored in advance using the output of Sub 7C as an address. Read-on memory (ROM) 7 to select
D, a multiplier (Con) 7E that multiplies (convolves) the echo data with the coefficient data selected by the ROM 7D,
Adder (Add) 7F that adds echo data from multiple convolutions that have been convoluted by this Con 7E, adder (Add) 7G that adds only coefficients from multiple convolutions, and this Add 7F,
Normalization calculator (Nor) 7H that processes addition data from 7G so that the addition value of the coefficient becomes 1, and image data of one frame in which the normalized echo data from this Nor 7H is displayed as an image Frame memory (FM) 7I generated as
And the address setter 7J for address conversion in BM7A
And an address setter 7K for allocating addresses of Sub 7C and FM 7I.

Reference numeral 8 denotes a display system including a CRT (cathode ray tube) and the like for displaying a one-frame tomographic image image generated by FM7I.

Next, the operation of this embodiment configured as described above will be described.

That is, when the transmitter 2 is activated, the ultrasonic probe 1 is excited to perform sector scanning, ultrasonic waves are transmitted to the inside of a subject (not shown), the reception signal is received by the receiver 3, and the adder 4,
The echo data digitized via the detector 5 and A / D-C6 is taken into the DSC7.

In the BM7A of DSC7, the echo data in pixel units and the conversion address obtained by converting the polar coordinate system address of the echo data into the orthogonal coordinate system address by the address setter 7J are sequentially taken in, and in Res 7B, from BM7A. The echo data and the conversion address data in the pixel unit are sequentially fetched.

In Sub 7C, the difference between the address of the pixel to be written, which is taken into Res 7B, and the address of the sample grid point is calculated, and the difference data is given to ROM 7D as a read address while making it correspond to the address by address setter 7K. . Here, the conversion address is given to determine a sound field pattern, and coefficient data showing appropriate filter characteristics is selected from the ROM 7D by this sound field pattern. Then, the coefficient data selected by the ROM 7D is multiplied (convolved) by echo data for each pixel unit by the Con 7E,
Echo data from a plurality of systems that have been convoluted by Con 7H are added by Add 7F, 7G.

Then, the addition data from the Add 7F and 7G is processed by the Nor 7H so that the addition value of the coefficient becomes 1, and the FM7I and the FM7I are respectively associated with the addresses by the address setter 7K.
Is written in the rectangular coordinate system address of, a frame image is generated, and the image is displayed on the display system 8.

As described above, according to the present embodiment, the echo data from the A / D-C6 is not written as it is to the FM7I of the DSC7, but all the pixels of the FM7I show the filter characteristic by the interpolation function. The result obtained by performing the convolution operation of the coefficient data and the echo data is written. In this case, the selected interpolation function is determined based on the conversion address from the polar coordinate system to the rectangular coordinate system. Since the selection is made according to the sound field pattern and the coefficient 1 is further normalized, pixel distortion can be prevented and an image with an improved S / N can be displayed.

In the above embodiment, the conversion address to be written to the BM7A for generating the read address of the ROM 7D uses the lower few bits and the upper few bits below the decimal point of the conversion result to the polar coordinate system or the rectangular coordinate system. However, all bits may be used.

The present invention is not limited to the above embodiments, and various modifications can be carried out without departing from the scope of the present invention.

〔The invention's effect〕

As described above, according to the present invention, the reception signals obtained by performing the ultrasonic transmission / reception to / from the subject by sector scanning are added, and the tomographic image information of the subject is obtained by detecting the added signal. In the ultrasonic wave diagnostic apparatus configured to display an image, A / D conversion means for digitizing the detection signal, and echo data for each pixel of the polar coordinate system by the sector scanning output from the A / D conversion means And a buffer memory in which the converted address data obtained by converting the polar coordinate system address of this echo data into an orthogonal coordinate system address is written to the same address, and a sound field pattern is determined based on the converted address data of this buffer memory. And a coefficient based on the sound field pattern data from the first means, in which coefficient data indicating a plurality of filter characteristics are stored in advance. A second means for selecting the data, a third means for multiplying the corresponding echo data by the coefficient data selected by the second means, and an echo data group output from the third means. Fourth means for setting the total of the numerical values to 1, and frame memory for generating the frame image data to be displayed by taking in the echo data for each pixel output from the fourth means with the orthogonal coordinate system address set So that the result obtained by performing the convolution operation of the interpolation function and the sample value of the echo data is written for all the pixels of the frame memory, and the spread function, that is, the resolution Since the cutoff frequency of the interpolation function is changed according to the size with respect to the sampling period, it is possible to provide an ultrasonic diagnostic apparatus capable of reducing pixel distortion and improving S / N.

[Brief description of drawings]

1 to 3 are for explaining the principle of the present invention. FIG. 1 is a diagram for explaining a polar coordinate system (r, θ) representation in sector scanning, and FIG. 2 is an original image and interpolation. FIG. 3 is a diagram for explaining the relationship with a function, FIG. 3 is a diagram for explaining the characteristics of the interpolation function in FIG. 2, and FIG. 4 is a block showing one embodiment of the ultrasonic diagnostic apparatus according to the present invention. 5 and 5 are diagrams for explaining pixel distortion due to sector scanning. 1 ... Ultrasonic probe, 2 ... Transmitter, 3 ... Receiver, 4
…… Adder, 5 …… Detector, 6 …… Analog / digital converter (A / D-C), 7 …… Digital Scan Converter
(DSC), 7A ... Buffer memory (BM), 7B ... Register group (Res), 7C ... Subtractor (Sub), 7D ... Read-on memory (ROM), 7E ... Multiplier (Con) , 7F, 7G ... Adder (Add), 7H ... Normalization calculator (Nor), 7I ... Frame memory (FM), 8 ... Display system.

Claims (1)

[Claims] 1. A method for adding a received signal obtained by performing ultrasonic wave reception to a subject by sector scanning, and detecting the added signal to obtain tomographic image information of the subject and display the image. In the ultrasonic diagnostic apparatus, the A / D conversion means for digitizing the detection signal, the echo data for each pixel of the polar coordinate system by the sector scanning output from the A / D conversion means, and the polar coordinates of the echo data. A buffer memory in which the conversion address data obtained by converting the system address into an orthogonal coordinate system address is written to the same address; and a first means for determining a sound field pattern based on the conversion address data in the buffer memory. Second coefficient data indicating a plurality of filter characteristics is stored in advance, and coefficient data based on the sound field pattern data from the first means is selected. And a third means for multiplying the corresponding echo data by the coefficient data selected by the second means, and a first means for setting the sum of the coefficient values of the echo data group output from the third means to one. 4 means and a frame memory for generating the frame image data to be displayed by taking in the echo data for each pixel set with the Cartesian coordinate system address and output from the fourth means. Ultrasonic diagnostic equipment.