JP6761767B2 - Ultrasound imaging device - Google Patents

Description

The present invention relates to an ultrasonic imaging apparatus and a method of processing a received signal.

Ultrasound imaging is generally performed by repeating transmission and reception of ultrasonic waves in different directions. The time required for one transmission / reception is the time required for the ultrasonic waves to propagate in the imaging range from the ultrasonic probe and for the reflected waves from the imaging target to reach the receiving opening of the ultrasonic probe. It is determined by the propagation distance and sound velocity of ultrasonic waves. Therefore, the imaging rate (frame rate, volume rate) of ultrasonic waves depends on the number of transmissions and receptions of ultrasonic waves. So far, various methods have been proposed to reduce the number of transmissions and receptions in order to increase the imaging rate. Generally, in the case of spatially dense transmission / reception, the reception signal of the ultrasonic probe is sequentially connected to each point on one line along the center of the ultrasonic transmission beam for each transmission / reception event. Receive beamforming to image is performed, thereby obtaining one line data which is a set of imaging data of each point. On the other hand, as a method of reducing the number of transmissions and receptions, a parallel reception method in which multiple lines are set within the range of the transmission beam for one transmission / reception event and image data is obtained by receiving beamforming for points on each line. Alternatively, a method such as constructing a missing line between imaging data obtained with a small amount of transmission / reception by interpolation is used.

On the other hand, when trying to achieve both improvement in imaging rate and imaging image quality, in the parallel reception method, for example, the sensitivity distribution of the transmission and reception varies depending on the positional relationship between the transmission beam and a plurality of set lines. There are problems such as distortion of image formation due to this. Further, in the interpolation method, the reliability of the interpolated data becomes an issue. Patent Document 1 proposes a method of improving the image quality by replacing the interpolated data with the actually acquired line data by sequentially switching the pattern of the position of the transmission / reception beam between the volumes.

In order to solve the problem of the parallel reception method, Patent Document 2 describes the difference in distance between the transmission beam and the plurality of reception lines by arranging a plurality of reception lines on the circumference centered on the transmission beam. It has been proposed to reduce the variation in transmission / reception sensitivity due to the above. Further, Patent Document 3 proposes that the receiving line has a spiral shape wound around the transmitting beam.

Japanese Unexamined Patent Publication No. 2013-414 Japanese Unexamined Patent Publication No. 2010-5138 Japanese Unexamined Patent Publication No. 2008-132342

The technique of Patent Document 1 is limited in application when the position movement of the imaging target is sufficiently small, or when it moves periodically like the heart and the imaging sequence can be synchronized with the movement cycle, and the movement of the ultrasonic probe is limited. Or, when there is movement of the imaging target, the interpolation accuracy of the interpolated data remains an issue. Therefore, in order to improve the imaging rate, parallel reception and highly reliable interpolation data generation technology are required. The line data of the conventional parallel reception method is generally formed linearly along the axis of the transmission beam, and is densely arranged around the center axis of the transmission beam. Therefore, when the line data is interpolated by the technique of Patent Document 1, it is difficult to generate the interpolated data for the interpolation region corresponding to the transmission beam interval and reconstruct a minute structure with respect to the pitch of the transmission beam. .. Further, even when the parallel reception lines are sparsely arranged at a certain distance from the transmission beam by the technique of Patent Document 2, the line data is linear, so that the region between the linear line data is described. The imaging data is not actually acquired. Therefore, the actually acquired imaging data has at least a long data missing region in the depth direction, and the sharper the angle formed by the structure to be imaged with respect to the line data, the lower the interpolation performance. Therefore, it has been difficult to achieve smooth interpolation for any structure by using the technique of Patent Document 2. Further, the technique of Patent Document 3 has an effect of improving the variation in transmission / reception sensitivity by controlling the shape of the receiving line other than a straight line, but does not have the effect of improving the interpolation accuracy.

An object of the present invention is to enable smooth and homogeneous interpolation regardless of the structure of the imaging target when performing interpolation between imaging data obtained by received beamforming.

The ultrasonic imaging apparatus of the present invention includes a transmission beam former that outputs transmission signals to a plurality of transducers in which ultrasonic probes are arranged, converts them into ultrasonic waves, and transmits them to an imaging target, and an ultrasonic imaging target. A sparse pattern that sets a reception focus pattern in which one or more reception focal points are thinned out in the depth direction from a plurality of reception focal points set in a predetermined arrangement at least in the depth direction with respect to the range in which ultrasonic waves are transmitted. At least the reception focus pattern set by the sparse pattern setting unit from the setting unit and the received signal that the multiple transducers of the ultrasonic probe receive and output the ultrasonic waves returned from the imaged object to which the ultrasonic waves are transmitted. A receiving beam former that generates imaging data for the receiving focal point, an interpolation processing unit that generates interpolation data for the thinned receiving focal point using the imaging data, and an imaging data generated by the receiving beam former. It also has an image generation unit that generates an image in the imaging range by using the interpolation data generated by the interpolation processing unit.

The ultrasonic imaging apparatus of the present invention can enable smooth and homogeneous interpolation regardless of the structure of the imaging target when performing interpolation between imaging data obtained by received beamforming.

The block diagram which shows the structure of the ultrasonic image pickup apparatus of embodiment of this invention. Explanatory drawing which shows the transmitted ultrasonic wave of an embodiment and the set reception focus pattern. The explanatory view which shows the example of the 3D reception focus pattern of an embodiment. The explanatory view which shows the reception focus pattern which the reception focus is arranged three-dimensionally of embodiment, and the type of a pattern. The block diagram which shows the structure of the receiving beam former of an embodiment. The block diagram which shows the structure of the receiving beam former of an embodiment. The block diagram of the receiving beam former which incorporates a plurality of delay addition part of an embodiment. The explanatory view which shows the example of the receiving line which selected and connected the receiving focus of embodiment one by one for each depth. An explanatory diagram showing a memory having an area for storing imaging data for a total of four time phases of the current time phase and the past time phase of the embodiment and an area for storing interpolation data. The flowchart which shows the operation of the ultrasonic image pickup apparatus of embodiment. The flowchart which shows the generation process of the reception focus pattern of the sparse pattern setting part of embodiment. The flowchart which shows the generation process of the reception focus pattern of the sparse pattern setting part of embodiment. The flowchart which shows the interpolation data generation processing of embodiment. The explanatory view which shows the process which performs the interpolation data generation process of embodiment by compressed sensing.

The ultrasonic imaging apparatus according to the embodiment of the present invention will be described.

FIG. 1 is a block diagram showing a configuration of an ultrasonic imaging device according to the present embodiment. FIG. 2 is an explanatory diagram showing a transmitted ultrasonic wave and a set reception focus pattern. As shown in FIG. 1, the ultrasonic imaging apparatus of the present embodiment includes a transmission beam former 101, a reception beam former 102, a reception focus pattern setting unit (sparse pattern setting unit) 103, an interpolation processing unit 104, and a memory. A main unit 150 including a signal processing unit 114 including 106, an image generation unit 105, a transmission / reception switching unit 107, a control unit 108, a user interface 109, and a display unit 110, and a cable (not shown). It has a main body device 150 and a detachable ultrasonic probe 100. An ultrasonic probe 100 is connected to the transmission beam former 101 and the reception beam former 102 via a transmission / reception switching unit 107. The ultrasonic probe 100 includes a plurality of oscillators that convert an electric signal into an ultrasonic signal and convert an ultrasonic wave into an electric signal, and the plurality of oscillators are arranged one-dimensionally or two-dimensionally.

The transmission beam former 101 outputs a transmission signal to each of the plurality of oscillators of the ultrasonic probe 100 according to the control from the control unit 108. As a result, the plurality of oscillators convert the transmission signal into ultrasonic waves 2 and transmit the transmission signals toward the imaging target 1. Ultrasonic waves 3 such as reflected waves return to the ultrasonic probe 100 from the image pickup target 1 to which the ultrasonic waves 2 are transmitted. Each of the plurality of oscillators of the ultrasonic probe 100 receives ultrasonic waves and outputs a received signal.

According to the control from the control unit 108, the sparse pattern setting unit 103 is set in a predetermined arrangement at least in the depth direction with respect to the range in which the ultrasonic wave 2 of the imaging target 1 is transmitted, as shown in FIG. A sparse reception focus pattern in which one or more reception focal points 5 are thinned out in the depth direction is generated from the plurality of reception focal points 4 and set in the reception beam former 102.

The reception beamformer 102 performs receive beamforming to form an image of the reception signals output by the plurality of oscillators on the reception focus according to the control from the control unit 108, and at least the reception focus pattern set by the sparse pattern setting unit 103. Imaging data is generated for the receiving focal point 4.

The signal processing unit 114 stores the imaging data from the reception beam former 102 in the memory 106 according to the control from the control unit 108, processes the signal in the interpolation processing unit 104, and passes it to the image generation unit 105. The interpolation processing unit 104 uses the imaging data generated by the reception beam former 102 and stored in the memory 106 to decimate the points at the positions of the reception focus 5 thinned out from the reception focus pattern by the sparse pattern setting unit 103. To generate. The image generation unit 105 generates an image of the image pickup target 1 by using the imaging data generated by the reception beam former 102 from the signal processing unit 114 and the interpolation data generated by the interpolation processing unit 104.

As described above, in the ultrasonic imaging apparatus of the present embodiment, one or more receiving focal points 5 are thinned out in the depth direction from a plurality of receiving focal points 4 set in a predetermined arrangement at least in the depth direction. A pattern is set, and interpolation data for points (reception focus 5) thinned out in the depth direction is generated from the imaging data of the reception focus 4. As a result, as shown in FIG. 2, since the sparse reception focus 4 can be set uniformly in each of the two-dimensional or three-dimensional directions including the depth direction, the reliability of interpolation can be improved, and it is minute. Smooth and uniform interpolation is possible for the imaging target 1 having any structure including a structure and a structure having an acute angle with respect to the depth direction.

For example, the sparse pattern setting unit 103 has one or more reception focal points from a plurality of receiving focal points arranged in the depth direction of the range in which the ultrasonic wave 2 is transmitted and the direction intersecting the depth direction. The reception focus pattern is set by thinning out in each of the depth direction and the direction intersecting the depth direction.

FIG. 3 is an example of a three-dimensional reception focus pattern. In FIG. 3, the receiving focal points are arranged three-dimensionally in the depth direction of the range in which the ultrasonic wave 2 is transmitted and the two directions orthogonal to the depth direction, and the receiving focal points are thinned out in each of the three directions. Is set.

For example, the sparse pattern setting unit 103 intersects the depth direction and the depth direction of the range in which the ultrasonic wave 2 is transmitted so that the number of reception focal points set for each depth of the receiving focal point pattern is the same. A fixed number of receiving focal points may be thinned out for each depth from a plurality of receiving focal points arranged in the direction. FIG. 4 shows a reception focus pattern in which reception focal points are arranged in three dimensions. In the example of FIG. 4, in the reception focus pattern, four reception focal points 4 are set at each of the depths A, B, C, and D, and the other points (reception focus 5) are thinned out. At this time, the sparse pattern setting unit 103 may thin out the reception focus 5 so that the position of the reception focus 4 set in the reception focus pattern is random at each depth A, B, C, and D. It may be thinned out in a certain pattern.

The reception beam former 102 may be configured to generate imaging data only for the reception focus 4 of the reception focus pattern set by the sparse pattern setting unit 103 (FIG. 5), and not only the reception focus 4 of the reception focus pattern but also the reception focus 4. , The image formation data may be generated for the thinned points (reception focus 5), and then only the image formation data of the reception focus 4 may be selected (FIG. 6).

FIG. 5 shows the configuration of the reception beam former 102 for realizing a configuration in which imaging data is generated only for the reception focus 4. The reception beam former 102 of FIG. 5 has a delay addition unit 111 and a delay amount setting unit 112. The delay amount setting unit 112 receives the position of the reception focus 4 of the reception focus pattern from the sparse pattern setting unit 103, and receives the reception signal (received RF (radio frequency) signal) output by each oscillator of the ultrasonic probe 100. A delay amount of each received signal for focusing on the reception focus 4 is generated and set in the delay addition unit 111. Specifically, for example, a sparse pattern is obtained from a delay table showing the relationship between the points at all positions where the reception focus 4 can be arranged and the delay amount, which is obtained in advance by calculation or the like and stored in the built-in memory. The delay amount corresponding to the position of the reception focus 4 received from the setting unit 103 may be read out and set in the delay addition unit 111. Alternatively, the delay amount setting unit 112 may receive the position of the reception focal point 4 from the sparse pattern setting unit 103, obtain the delay amount by calculation by a predetermined calculation method, and set it in the delay addition unit 111. The delay addition unit 111 of the reception focus 4 generates imaging data by delaying and adding each received signal by a set delay amount.

When there is only one delay addition unit 111, the delay addition calculation is serially performed for each reception focus 4 of the reception focus pattern to obtain imaging data. FIG. 7 shows a configuration in which the receiving beam former 102 includes a plurality of delay adding units 111a, 111b, 111c. When the receiving beam former 102 is provided with a plurality of delay addition units in this way, the delay addition units 111a to 111c perform delay addition operations in parallel to obtain imaging data. In either case, the delay amount setting unit 112 sets a reception line connected by selecting one reception focus of the reception focus pattern for each depth, and a plurality of reception focal points 4 on the reception line. The amount of delay of the received signal for focusing the received signal of the vibrator is sequentially set in the delay addition unit 111. FIG. 8 shows an example in which three reception lines are set by selecting one reception focus for each depth. At this time, the delay amount setting unit 112 sets the number of reception lines as the same as the number of reception focal points 4 for each depth of the receiving focal point pattern, thereby simultaneously generating imaging data of a plurality of receiving lines in a parallel manner. it can. However, it is assumed that the delay addition units 111a to 111c of the reception beam former 102 are provided in the same number or more as the number of reception lines.

Next, the reception beam former 102 generates imaging data not only for the reception focus 4 of the reception focus pattern but also for the thinned points (reception focus 5), and then selects only the imaging data of the reception focus 4. The case of the configuration (FIG. 6) will be specifically described. In the case of this configuration, it is desirable to arrange the memory 106 for storing the imaging data output by the reception beam former 102 between the reception beam former 102 and the image generation unit 105. The memory 106 and the interpolation processing unit 104 form a signal processing unit 114. FIG. 6 is a configuration example of the reception beam former 102 for realizing this configuration. The reception beam former 102 of FIG. 6 has a delay addition unit 111 and an imaging data thinning unit 113. The imaging data thinning unit 113 receives the position of the reception focus 4 of the reception focus pattern from the sparse pattern setting unit 103, generates a delay amount not only for the reception focus 4 but also for the thinned reception focus 5, and delay addition unit. Set to 111. For example, the delay amount is read from the delay table showing the relationship between the position of the reception focus 4 and the delay amount, which is obtained in advance and stored in the built-in memory, and is set in the delay addition unit 111. The delay addition unit 111 delays and adds each received signal by a set delay amount to generate imaging data of the receiving focal point 4 and the thinned receiving focal point 5. The image formation data thinning unit 113 receives the image formation data from the delay addition unit 111, selects only the image formation data of each reception focus 4 of the reception focus pattern, and stores it in the memory 106. In this method, as in the conventional method, the delay addition unit 111 performs the delay addition processing for all the reception focal points that have not been thinned out. Therefore, the operation of the delay addition unit 111 is the same as in the conventional method. The imaging data of the reception focal point 4 after the thinning process can be obtained while the configuration of the delay addition unit 111 is the same as the conventional one.

The interpolation processing unit 104 calculates the data (interpolation data) of the thinned points (reception focus 5) by the interpolation calculation using the imaging data of the reception focus 4 of the reception focus pattern. For example, when calculating interpolation data by spatial interpolation, any one of the imaging data of one or more receiving focal points 4 located around the thinned points, or the average value of the imaging data of two or more receiving focal points 4 or the like. Is the interpolated data. The imaging data used for interpolation is read from the memory 106.

It is also possible to use the imaging data obtained at the same position by ultrasonic waves transmitted in different time phases as the interpolated data of the thinned points (reception focus 5). In this case, the memory 106 has a structure having an area for storing the imaging data obtained in each of a predetermined number of time phases in the past and the interpolation data. FIG. 9 has an area 901 for storing the imaging data V1, V2, V3, and V4 for a total of four time phases of the current time phase and the past time phase, and an area 902 for storing the time phase interpolation data Vc, respectively. The memory 106 is shown. Thereby, the interpolation data can be obtained by using any one of the past three-time phase imaging data or a plurality of imaging data. At this time, in the present embodiment, since the imaging data of each time phase is the imaging data of the sparse reception focus 4 after thinning out, the conventional imaging data of the four time phases that have not been thinned out is used. There is an advantage that the memory capacity can be reduced as compared with the memory capacity required for storage.

Specifically, for example, as shown in FIG. 9, the memory 106 stores four consecutive sparse imaging data and the interpolation data V'c (i) subjected to spatial interpolation and interphase interpolation. The memory capacity of the above is provided, and the interpolation processing unit 104 calculates the i-th interpolation data V'c (i) by simultaneously performing spatial interpolation and interphase interpolation using the following equation (1). _{V '1 (i), V} ' 2 (i), V '3 (i), V' 4 (i) , respectively, for example, i-th spatial interpolation data calculated by spatial interpolation method described above, the formula (1) is a calculation formula for calculating the interpolation data V'c (i) by performing interphase interpolation by, for example, obtaining the average value of the spatial interpolation data of consecutive four time phases. Although the case of performing temporal interpolation after performing 4:00 phase all spatial interpolation is required memory capacity for storing the entire volume data _{V '1, V' 2,} V '3, V' 4 In the present embodiment, the memory 106 does not need to hold all of the spatial interpolation data for each time phase, and the capacity of the memory 106 may be smaller than the case where the sparsely thinned imaging data is stored. , Allows interpolation between multiple time phase data.
V'c (i) = mean (V '1 (i) + V' 2 (i) + V '3 (i) + V' 4 (i)) ··· (1)

The image generation unit 105 converts an array of reception focal points 4 and 5 composed of imaging data and interpolation data into image data for display on a display. For example, when the scanning of the ultrasonic wave 2 at the time of transmission is a sector scanning, the arrangement of the receiving focal points 4 and 5 is converted into a fan shape in real space. Further, in the conversion process, further interpolation of pixel data is performed. In addition, a filter process is applied to make the image look smooth. For example, since the imaging data has a luminance distribution peculiar to the reception pattern, it is processed by a different filter or a filter with different parameters depending on the reception pattern.

Hereinafter, the operation of each part when capturing an ultrasonic image will be described with reference to a flowchart. Here, as shown in FIG. 1, the ultrasonic imaging apparatus is further provided with a control unit 108 and a user interface (hereinafter, UI) 109, and the control unit 108 uses ultrasonic waves based on instructions and values received from the UI 109. It is assumed that the configuration is such that each part of the image pickup apparatus is controlled. FIG. 10 is a flowchart showing the operation of the control unit 108.

The control unit 108 is configured to include a processor such as a CPU (Central Processing Unit) and a memory in which a program is stored in advance, and the processor reads and executes a program in the memory, as shown in the flow of FIG. Control each part to operate. The sparse pattern setting unit 103, the interpolation processing unit 104, and the image generation unit 105 may be configured so that their functions are executed by software in the same manner as the control unit 108, and a part or all of them may be ASICs. It is composed of hardware such as a custom IC such as (Application Specific Integrated Circuit) and a programmable IC such as FPGA (Field-Programmable Gate Array), and the IC circuit is designed so that each function is realized. It may be.

First, as shown in the flowchart of FIG. 10, the control unit 108 causes the display unit 110 to display the input screen, and receives the setting of the imaging parameter from the operator via the UI 109 (step 501). As a result, the control unit 108 receives from the operator settings such as the irradiation range of the ultrasonic wave 2 to be transmitted, the thinning method of the reception focus pattern set by the sparse pattern setting unit 103, and the thinning rate as imaging parameters (step 501). ..

The control unit 108 instructs the sparse pattern setting unit 103 of the irradiation range of the ultrasonic wave 2, the thinning method of the reception focus pattern, the thinning rate, and the like, and the sparse pattern setting unit 103 generates the reception focus pattern (step 502).

The process of generating the reception focus pattern of the sparse pattern setting unit 103 will be specifically described with reference to FIGS. 11, 12 (a) and 12 (b). The sparse pattern setting unit 103 sets the start point coordinates (x0, y0, z0) in the irradiation range of the ultrasonic wave 2, and M, N, and L total M in the x, y, and z directions from the start point coordinates, respectively. × N × L reception focal points (this reception focal point corresponds to a pixel of an image and is therefore also referred to as a pixel) are arranged (step 51). Next, the sparse pattern setting unit 103 sets the size of the unit area for thinning out the reception focus. Here, the sparse pattern setting unit 103 sets that the size of the unit area is O × P × Q pixels in the x, y, and z directions, respectively (step 52). Next, the sparse pattern setting unit 103 selects the unit area to be thinned out by the numbers i, j, and k assigned to the O × P × Q pixels in the x, y, and z directions of the unit area. That is, as shown in FIG. 3, when the unit area O × P × Q = 2 × 2 × 2, i = 1, 2 ... In the x direction and 2 pixels in the y direction. = 1, 2, ..., K = 1, 2, ... Are assigned to each of 2 pixels in the z direction. The sparse pattern setting unit 103 first selects a unit area of O × P × Q pixels specified by i = j = k = 1 (steps 52 to 56), and determines the position of the pixels in the unit area. Each is specified (step 57), and the reception focus 4 is arranged by the process of FIG. 12 (a) or FIG. 12 (b) (step 58).

In step 58, when the process of FIG. 12A is performed, the sparse pattern setting unit 103 receives the thinning rate R set by the operator from the control unit 108 (step 41), and has O × P × Q pixels. A process of randomly selecting (O × P × Q) / R pixels (step 42) and arranging the reception focus 4 on the selected pixels is performed (step 43). As a result, the reception focal points 4 can be randomly arranged at a predetermined thinning rate R within the unit area of O × P × Q pixels specified by i = j = k = 1. By repeating this while increasing the values of i, j, and k, the sparse pattern setting unit 103 generates a reception focus pattern in which the reception focus 4 is randomly arranged in all M × N × L pixels. (Steps 58-61).

On the other hand, in step 58, when the process of FIG. 12B is performed, the sparse pattern setting unit 103 arranges (O × P) / R reception focal points in the unit area of O × P × Q pixels. The patterns of the plurality of types A, B, C ... Are stored in the built-in memory for each settable thinning rate R (see FIG. 4). The sparse pattern setting unit 103 receives the thinning rate R set by the operator from the control unit 108 (step 71), and reads the patterns A, B, C ... Corresponding to the received R from the memory (step 72). Patterns A, B, B set in the unit area according to the arrangement number k in the z direction of the unit area so that the patterns A, B, C ... of the unit area are repeatedly arranged in the depth direction in the z direction. , C ... Are selected (steps 73, 74). For example, when the thinning rate R is 4, the reception focus of the pattern B is arranged in O × P pixels in the stage of k = 1, and C is arranged in k = 2. In the x and y directions, the patterns set in steps 73 and 74 of A, B, C ... Are repeatedly arranged. By repeating this while increasing the values of i, j, and k, the sparse pattern setting unit 103 applies the reception focus 4 to all M × N × L pixels with patterns A, B, C, and so on. Is generated (steps 58 to 61).

The sparse pattern setting unit 103 sets the generated reception focus pattern in the reception beam former 102 (step 502). In step 502, the value of M × N × L and the value of O × P × Q may be a predetermined value or a value received from the operator in step 501.

Next, the control unit 108 instructs the transmission beam former 101 to generate a transmission signal for transmitting the ultrasonic wave 2 to the irradiation range received in step 501 and the transmission position thereof, and each of the ultrasonic probe 100 Output to the oscillator (step 503). As a result, the ultrasonic wave 2 is transmitted from the ultrasonic probe 100 to the image pickup target 1, the ultrasonic wave 3 such as the reflected wave is received by each oscillator of the ultrasonic probe 100, and the received signal is output.

Under the control of the control unit 108, the reception beam former 102 generates imaging data for at least the reception focus 4 of the sparse pattern setting unit 103 and stores it in the memory 106 (step 504). The method of generating the imaging data is as described with reference to FIGS. 5 and 6.

Under the control of the control unit 108, the interpolation processing unit 104 generates interpolation data for the point (reception focus 5) where the reception focus 4 of the reception focus pattern is not arranged (step 505).

FIG. 13 is a flowchart showing the details of the interpolation data generation process in step 505. This process is an interpolation method for calculating interpolation data from surrounding imaging data. First, the interpolation processing unit 104 sets which range of imaging data to refer to, centering on the point at which the interpolation data should be generated (step 81). For example, when the position of the pixel to be interpolated is (0,0,0), the range W is set to 8 pixels around -1 to +1 in the x, y, z directions, and exists in these 8 pixels. It is decided to read the imaging data to be performed. This range W may be a predetermined range, or may be a range received from the operator in step 501.

Next, the interpolation processing unit 104 obtains a reception focus pattern from the sparse pattern setting unit 103, reads the position p of the interpolation data (step 82), and forms an imaging data V located in the range W from the position p of the interpolation data (step 82). Wp) is read from the memory 106. For example, when the position p of the pixel to be interpolated is (0,0,0), the imaging data V (Wp) existing in the surrounding 8 pixels within -1 to +1 in the x, y, z directions is read out. (Steps 83, 84).

The interpolation processing unit 104 uses the read imaging data V (Wp), calculates, for example, the average value v thereof, and stores it in the memory 106 as the interpolation data of the position p (steps 85 and 86).

The above process is repeated at the point p where all the interpolated data should be generated (step 87). As a result, the process of generating the interpolation data in step 505 is completed. The processing operation of the interpolation processing unit 104 is not limited to the flow shown in FIG. 13, and it is of course possible to generate the interpolation data by other interpolation processing.

Next, under the control of the control unit 108, the image generation unit 105 generates a three-dimensional image (three-dimensional volume data) of the imaging region 120a by using the imaging data and the interpolation data of the memory 106, and the display unit 110. To display.

Then, returning to step 503, the control unit 108 performs the next transmission. Steps 503 to 506 are repeated until the operator gives an instruction to end imaging via UI109.

In the ultrasonic imaging apparatus of the present embodiment, the sparse pattern setting unit 103 enables the formation of sparse imaging data in two-dimensional or three-dimensional directions including the depth direction, thereby improving the reliability of interpolation and any method. Smooth and uniform interpolation is possible even for structural objects.

The interpolation processing unit 104 may generate the interpolation data by using the compressed sensing technique. When the interpolated data is generated by the compressed sensing technique, the sparse pattern setting unit 103 generates a sparse reception focus pattern suitable for generating the interpolated data by the compressed sensing technique. A general compressed sensing will be described with reference to FIG. The original data (imaging data of all receiving focal points that have not been thinned out) is x, and the observation data acquired by undersampling (imaging data of receiving focal point 4 after thinning out) is y, and undersampling (thinning out). Assuming that the observation matrix is Φ, y is described by Eq. (2). x is expressed by Eq. (3) by converting it to v by the basis Ψ. At this time, most of the distribution of the basis Ψ is 0, and the basis Ψ is selected so that the signals are concentrated in the poles (v is sparse) (v is called a sparse signal). .. As Ψ, a transformation matrix such as Fourier or wavelet is used. As described in equation (4), generally 'can be obtained (v' norm of v, called under l ₀ norm constraints of formula (3) is v as a solution that minimizes the Reconstructed sparse signal). Using v', the data x'reconstructed by the inverse conversion as shown in the equation (5) is obtained. By this series of processing, the data x'in which the missing data is interpolated is restored from the observation data y under undersampling. This gives x'corresponding to the interpolated, undecimated imaging data of all receiving focal points.

The sparse thinning process in the sparse pattern setting unit 103 corresponds to selecting the observation matrix Φ. Generally, Φ is required to have randomness. Further, the optimization of Φ can be performed by evaluating the magnitude of the error of x'recovered by the reconstruction using the known data x. For example, the optimum observation matrix Φ with a small error can be selected from several observation matrices Φ.

4 ... reception focus, 5 ... thinned reception focus, 100 ... ultrasonic probe, 101 ... transmission beam former, 102 ... reception beam former, 103 ... sparse pattern setting unit, 104 ... interpolation processing unit, 105 ... image Generation unit, 106 ... Memory, 108 ... Control unit, 109 ... User interface, 110 ... Display unit

Claims (4)

A transmission beam former that outputs a transmission signal to each of a plurality of oscillators in which ultrasonic probes are arranged, converts them into ultrasonic waves, and transmits the ultrasonic waves to an imaging target.
From the received signals that the plurality of oscillators receive and output the ultrasonic waves returned from the image pickup target to which the ultrasonic waves are transmitted by one transmission by the transmission beam former, the image pickup target is subjected to the parallel reception method. A receiving beam former that generates imaging data for a plurality of receiving focal points on a plurality of receiving lines in the ultrasonic irradiation range,
A sparse pattern setting unit that defines a pattern of arrangement of a plurality of reception focal points for which the imaging data of the plurality of reception lines is generated,
Interpolating processing unit and
It has an image generator
The sparse pattern setting unit sets pixels arranged in three dimensions with the transmission direction of the ultrasonic waves as the depth direction in the irradiation range of the ultrasonic waves to be imaged, and among them, the pixels of the two-dimensional array for each depth. A predetermined number of pixels are randomly selected from the above, or a plurality of predetermined selection patterns are selected in order in the depth direction, and the positions of the selected pixels are generated as the imaging data. Set to the receive beam former as the position of the receive focus to be
The interpolation processing unit generates interpolation data of the pixel positions not selected by the sparse pattern setting unit by an interpolation calculation.
The image generation unit generates a three-dimensional image of pixels arranged in the three dimensions by using the imaging data generated by the reception beam former and the interpolation data generated by the interpolation processing unit. An ultrasonic imaging device characterized in that. The ultrasonic imaging apparatus according to claim 1, wherein the receiving beam former sets the receiving focal point set by the sparse pattern setting unit by selecting a predetermined number of pixels for each depth. Ultrasound imaging characterized in that the same number of reception lines as the predetermined number are set by selecting and connecting one by one for each depth, and the imaging data is generated by the parallel beam reception method. apparatus. The ultrasonic imaging apparatus according to claim 1, wherein the receiving beam former generates imaging data with both pixels selected by the sparse pattern setting unit and pixels not selected as reception focal points, and then the image data is generated. An ultrasonic imaging apparatus characterized in that the sparse pattern setting unit selects and outputs the imaging data of the receiving focal point of the selected pixel . The ultrasonic wave imaging device according to claim 1, wherein the receiving beam former includes the same number or more delay addition units as the receiving line .

Images