JPH10277032A - Ultrasonic diagnostic device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To exclude a noise influence by referring to echo data of an adjacent voxel adjacent to a notice voxel and setting a smaller value as opacity if it is highly possible that echo data of the notice voxel is noise. SOLUTION: Echo data e1 on a just former ultrasonic beam outputted from a line memory 40 and echo data e2 on an ultrasonic beam are inputted to an arithmetic part 42, which executes average arithmetic and differential value arithmetic. Then, an opacity deciding table 4 decides opacity α based on an average value (e) and a differential value Δe. Then, a nonlinear function where opacity is gradually increased with the raising of the average value (e) and with the raising of the differential value Δe is set. Then, set opacity α is outputted to a three-dimensionally projecting image forming part 34 from an opacity deciding part 32 to utilize opacity α by the voxel arithmetic of a voxel.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an ultrasonic diagnostic apparatus, and more particularly to an ultrasonic diagnostic apparatus for forming a three-dimensional projected image.

[0002]

2. Description of the Related Art In a two-dimensional tomographic image display (so-called B-mode display) in a conventional ultrasonic diagnostic apparatus, a cross section in a living body is displayed as a black and white gray-scale image. However, since only the cut surface of the tissue to be observed is expressed, it is difficult to recognize and grasp the tissue three-dimensionally on the image. On the other hand, an apparatus that transmits and receives ultrasonic waves to and from a three-dimensional region in a living body to form a three-dimensional image of a tissue is being put into practical use. The three-dimensional image is obtained, for example, by shading the tissue surface obtained by performing surface extraction to give a sense of depth, and can express the tissue three-dimensionally. In addition, as a method of imaging a three-dimensional region, an integration method, a projection method, and the like are also known, but an image formed by such a method is planar and has no sense of depth.

In the above-described conventional three-dimensional image processing, all of the echo data captured in a three-dimensional area is temporarily stored in a three-dimensional echo data memory, and then each echo data is processed by software processing or the like. Need to be reconfigured. Therefore, it takes a lot of time to perform a calculation for obtaining one three-dimensional image, and it is extremely difficult to display a three-dimensional image in real time. Further, since the conventional three-dimensional image is basically based on shading of the surface, it has basically been impossible to spatially represent the inside of the tissue through the tissue.

Accordingly, the applicant of the present application has filed a Japanese Patent Application No. 8-185.
No. 781 proposes a new image processing method. Although the principle will be described in detail later, according to such an image processing method, the tissue can be expressed three-dimensionally and transparently, and according to the user's preference, the three-dimensional expression on the tissue surface can be emphasized, or the internal tissue can be expressed. Can be emphasized.

In this image processing method, a voxel operation (described later) is sequentially performed in a time-series order of the received signal, that is, for each echo data existing along the ultrasonic beam. The voxel calculation is executed until a predetermined end condition is satisfied. Then, the voxel operation value at the end point is associated with the pixel value. Therefore, by appropriately setting the termination condition, the voxel operation is terminated in the vicinity of the tissue surface, and a display in which the tissue expression is emphasized can be performed.

[0006]

However, in the above image processing method, a noise source that scatters ultrasonic waves exists before an object to be imaged, that is, between the object and the ultrasonic probe. However, there is a problem that the noise is imaged together with the object. For example, in the case where the surface of a fetus is imaged by the above-described image processing method, if the above-mentioned noise source is present, noise may be mixed as a bright point into a three-dimensional projected image of the fetus, which may hinder diagnosis. Further, in the conventional image processing method,
Since the end condition of the voxel operation along the ultrasonic beam is governed by the echo data on the ultrasonic beam, if noise is included, the voxel operation ends earlier than the originally expected end point. There was a problem.

An object of the present invention is to eliminate the influence of noise as much as possible when forming a three-dimensional projected image.

Another object of the present invention is to reduce noise without impairing real-time performance when forming a three-dimensional projected image.

[0009]

[Means for Solving the Problems]

(1) In order to achieve the above object, the present invention obtains echo data of each voxel in the three-dimensional area by scanning an ultrasonic beam and transmitting and receiving an ultrasonic wave to and from a three-dimensional area. Transmitting and receiving means, along with the ultrasonic beam, by sequentially executing voxel operation based on volume rendering for each voxel, a stereoscopic projection image forming means for forming a stereoscopic projection image of the three-dimensional region, Echo data of the voxel of interest and 1 adjacent to it
Or, opacity determining means for determining opacity used in voxel calculation of the voxel of interest based on echo data of a plurality of adjacent voxels.

According to the above configuration, when voxel calculation is performed for each voxel, echo data of one or more adjacent voxels adjacent to the voxel of interest is referred to, whereby the echo data of the voxel of interest may be noise. Sex is determined. In this case, if the possibility of noise is high, a smaller value is set as the opacity.
When the value of the opacity is set to be small, the contribution of the echo data to the pixel value becomes small, and as a result, noise is reduced.

As described above, in the present invention, attention is paid to the randomness or discontinuity of noise, and the possibility of noise is determined based on the correlation between the echo data of interest and the echo data in the vicinity thereof. Elimination takes place.

In a preferred aspect of the present invention, the opacity determining means uses echo data of adjacent voxels of the same depth on an ultrasonic beam adjacent to the ultrasonic beam in which the voxel of interest exists, and The opacity of the voxel of interest is determined.

By the way, if echo data of the same depth on the immediately preceding ultrasonic beam that has already been acquired is used,
Noise can be suppressed in the voxel operation without impairing the real-time property. In this case, it is desirable to provide a line memory for storing echo data for one ultrasonic beam so that the present echo data and the previous echo data can be used simultaneously.

In a preferred aspect of the present invention, the opacity determining means calculates a difference value of echo data using the echo data of the voxel of interest and the echo data of the adjacent voxel; Average value calculating means for calculating an average value of echo data using voxel echo data and echo data of the adjacent voxel,
And opacity determining means for determining the opacity according to the difference value and the average value.

Since noise generally has a low echo level and is highly likely to be generated individually, if a difference value and an average value are obtained between a plurality of adjacent echo data, the two indexes are used. , The possibility of noise can be determined. Of course, without relying on such two indicators,
Either one may be used. However, in this case, the noise discrimination accuracy decreases. In addition, noise discrimination can be performed using other indices. In any case,
The echo data and the adjacent echo data are compared in some form to determine whether the echo data is noise. In a preferred aspect of the present invention, the transmitting and receiving means for scanning the ultrasonic beam to transmit and receive an ultrasonic wave to and from the three-dimensional area to obtain echo data of each voxel in the three-dimensional area, Along with the sound beam, by sequentially executing voxel calculations based on volume rendering for each voxel, a three-dimensional projection image forming means for forming a three-dimensional projection image of the three-dimensional region, echo data of the voxel of interest and Means for determining, based on echo data of one or more adjacent voxels, parameter values used in the voxel operation of the target voxel so as to suppress noise in the voxel operation of the voxel of interest. And

Here, the parameter used in the above-mentioned voxel calculation is desirably opacity, and other than that, the transparency itself, a coefficient for calculating them, and the like can be considered.

[0017] (2) In a preferred embodiment of the invention, the stereoscopic projection image forming means, and opacity calculating means for calculating the opacity alpha _i voxel i based on the echo data e _i,
And transparency calculating means for calculating a transparency beta _i voxel i based on the echo data e _i, multiplied by the opacity alpha _i in the echo data e _i, a light emission amount calculating means for calculating a light emission amount of the voxel i, 1 previous of multiplying the transparency of beta _i of voxel i to output light amount of the voxel i -1, and transmitted light quantity calculating means for calculating a quantity of transmitted light of the voxel i, adding the the amount of transmitted light and the light emission amount, the output light intensity of the voxel i And a light amount adding means for obtaining the three-dimensional projection image, wherein the output light amount of the end voxel is made to correspond to the pixel value to form the stereoscopic projection image.

In a preferred aspect of the present invention, the three-dimensional projection image forming means includes an opacity calculating means for calculating the opacity α _i of the voxel i based on the echo data ei;
The echo data e _i , the opacity α _i , and the input light amount C corresponding to the output light amount of the previous voxel i−1.
Output light amount calculating means for calculating the output light amount C _OUTi of the voxel i based on _INi , wherein the three-dimensional projection image is formed by making the output light amount of the end voxel correspond to the pixel value.

According to the above arrangement, voxel processing utilizing opacity or the like is performed along the ultrasonic beam. As a result, it is possible to sequentially perform real-time processing on echo data that is sequentially captured in chronological order, and it is possible to eliminate the need for a three-dimensional data memory required in the conventional device. That is, the captured echo data is processed in the capturing order, and data processing can be sufficiently performed without temporarily storing all the echo data in the three-dimensional data memory.

Incidentally, the opacity α _i relates to the degree of diffusion and scattering of ultrasonic waves to the surroundings of the voxel i, and the light emission amount indicates the intensity of the voxel i as a sound source (light source). Seem. On the other hand, the transparency β _i relates to the transmittance of ultrasonic waves, and the amount of transmitted light is considered to correspond to the transmittance when voxel i is viewed as a transmission medium. Transparency is generally defined as 1-opacity.
The amount of light emission and the amount of transmitted light are added to obtain a voxel i.
Is calculated. Here, the output light amount is voxel
It represents the degree of contribution of i to the pixel value. This output light amount is transferred to voxel processing (calculation of transmitted light amount) of the next voxel. Then, when the voxel processing reaches the final voxel, the output light amount of the final voxel is converted into a pixel value. Then, when each pixel value is obtained, one three-dimensional projection image is formed as a set of those pixel values.

It has been confirmed by experiments that this ultrasonic image has both the characteristics as a projected image and the characteristics as a stereoscopic image. That is, a tissue in a living body can be expressed in a transparent manner like an X-ray photograph, while it can be expressed with a sense of depth like an ultrasonic three-dimensional image. Therefore, for example, simultaneous observation of the surface and the inside of a fetus can be performed, and three-dimensional grasp of a tissue can be easily performed in diagnosing a disease.

Of course, by changing the definitions of the opacity and the transparency, it is possible to construct an ultrasonic image of a desired image quality, for example, to enhance the transparency or the three-dimensional effect. Alternatively, the tissue surface can be enhanced or the interior of the tissue can be enhanced.

Such adjustment is performed by changing the definition of opacity and the like, and more specifically, can be performed by appropriately setting an end condition using opacity as a parameter. In this case, if the value of each opacity α _i to be sequentially added is large, the processing ends at a relatively early stage. For example, the image expression ends by seeing through the surface of the tissue. Conversely, if the value of each opacity α _i is small, the processing ends at a relatively late stage. For example, the image processing ends by seeing deep inside the tissue.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Prior to the description of the apparatus configuration, a method of forming a three-dimensional projected image according to an embodiment will be described.

[Explanation of Image Forming Principle] The image processing method according to the present embodiment employs a known volume rendering (Volum
e Rendering) based on real-time image processing (especially ultrasonic image processing). At that time, special conditions are taken into account.

As shown in FIG. 1A, when an ultrasonic beam directed in the Y direction is scanned in the X direction, a scanning surface 10 is formed. When the scanning surface 10 is moved in the Z direction, a three-dimensional echo data capturing space 12 is formed as is well known. For this three-dimensional echo data capture space 12,
The voxel processing according to the present embodiment is performed along each ultrasonic beam, and the three-dimensional echo data capturing space 12 is projected on the projection plane 16 to obtain an ultrasonic image 100 shown in FIG.
It is. In the ultrasonic image 100, one line 100a in the X direction corresponds to one scanning plane 10. In other words,
One ultrasonic beam (perspective line) is one in the ultrasonic image 100.
Corresponds to a pixel.

Here, voxel processing, which will be described in detail below, is performed on the fetched echo data in chronological order, so that each echo data is temporarily stored in a three-dimensional echo data memory to form an image. It is not necessary to read out the echo data in the order necessary for data acquisition, and data processing synchronized with data acquisition can be performed.

FIGS. 2 and 3 show the concept of the voxel 20. FIG. One voxel converts the received signal to A
It corresponds to one echo data obtained by the / D conversion, in other words, it corresponds to a volume (sample point) corresponding to one cycle of the A / D conversion rate. That is,
An ultrasound beam is assumed as a collection of many voxels. FIG. 2 shows each voxel from i-1 to _LLAST . The value obtained by performing the processing sequentially from the first voxel is the luminance value P (x, 1) of one pixel constituting the ultrasonic image.
y).

Here, opacity α and transparency β [β = (1−α) in this embodiment] are defined for each voxel. The opacity α corresponds to spontaneous light emission around the voxel as shown in FIG. The transparency (1−α) corresponds to the degree of transmission of light from the previous voxel in the voxel. In the present embodiment, the opacity α is set in a range of 0 ≦ α ≦ 1. Specifically, opacity is echo data (echo value)
Is defined as a function of

In this case, in the above-mentioned Japanese Patent Application No. 8-1857981, for example, the opacity α _i of voxel _i is

Α _i = k 1 · e _i ^{k 2} (1) Here, e _i is the value of the echo data of the voxel i, also k1 is a coefficient (parameter) is variably set by the user. A numerical value larger than 1 is desirably substituted for k2, for example, k2 = 2 or 3. That is, for the value e _{i of the} echo data, α _i
Changes non-linearly.

However, if the opacity is simply defined as a function in the echo data of the voxel, an image will be formed even if the opacity is noise.
Thus, in the present embodiment, the value of the opacity is reduced for echo data regarded as noise. Specifically, for example, 1
Alternatively, by referring to the echo data of a plurality of voxels, a difference value and an average value between the echo data and the adjacent echo data are calculated, and the opacity is determined as a function of the difference value and the average value. Here, information other than the difference value and the average value may be used as a reference as long as the information can be used for noise determination.

Therefore, in the present embodiment, a small value is determined as the opacity of the echo data regarded as noise, so that the ratio of the contribution of the noise to the output light quantity of the end voxel described later decreases. In other words, the influence of noise can be reduced.

[0033] Now, as shown in FIG. 2, the voxel i, the amount of input light C _INi and the output light amount C _OUTi are defined, the amount of input light C _INi one previous output light amount of the voxel i -1 Equivalent to C _OUTi-1 . That is,

## EQU2 ## There is a relationship of C _INi = C _OUTi-1 (2). However, C _{IN i} = 0 at the start voxel at which the voxel processing is started. Note that the start voxel is set automatically or artificially.

For each voxel, the amount of light emission and the amount of transmitted light are defined based on the opacity α and the transparency (1−α). That is, the light emission amount of voxel i is defined as the product of the opacity and the echo data, and is α _i · e _i . The transmitted light amount of voxel i is defined as the product of the transparency and the input light amount, and is (1−α _i ) · C _INi .

In the present embodiment, as shown in FIG.
The light emission amount and the transmitted light amount are added as follows, and the output light amount C _{OUTi of the} voxel is determined.

[0036]

C _OUTi = (1−α _i ) · C _INi + α _i · e _i (3) where C _INi = C _{OUTi−1 according} to the above second equation. That is, the calculation result of the previous voxel is used for calculation of the next voxel.

While the above equation (3) is sequentially performed from the start voxel to the next voxel and then to the next voxel, the opacity α _i of each voxel is added, and the added value Σα _i becomes 1 At the time when the processing has reached (end condition). However, even when the processing is the last (or the set depth) voxel L _LAST , the processing is terminated (forced termination condition). That is, the conditions under which the processing ends are as follows:

４α _i = 1 or i = L _LAST (4) The termination of the process when Σα _i = 1 means that the process is stopped when the sum of the opacity reaches 1.
Of course, the condition of the fourth expression, in particular, the maximum addition value (end determination value) of α _i may be changed according to the condition.

The output light-amount C _{OU T} of [0038] or more voxels at the time the end determination is made (final voxel) is brightness P (x, y) of the corresponding pixel is used as a. Then, when such pixel value calculation for each ultrasonic beam is performed for all ultrasonic beams, pixel values of all pixels constituting the ultrasonic image can be obtained. That is, one ultrasonic image is formed.

As shown by the above formula (3), the luminance value P of the pixel
(X, y) reflects the values of all echo data from the start voxel to the end voxel. But,
It reflects both scattering and absorption of ultrasonic waves at each voxel, not just simple integration as in the past. Therefore, an ultrasonic image having both the depth (three-dimensional) and transparent properties, such as an image formed by light transmitted from a light source and scattered and absorbed by each voxel, is obtained. Can be configured.

In the above formula (3), the transparency is defined by (1−α _i ), that is, the transparency can be represented by the opacity α _i , so the concept of transparency is apparently deleted from the arithmetic expression. be able to. Therefore, the output light amount C _OUTi can be calculated based on the same principle by _modifying the third expression as follows.

[0041]

C _OUTi = (1−α _i ) · C _INi + α _i · e _i (3) = C _INi + α _i · (e _i −C _INi ) (5-1) = C _INi + Δ _i ··· (5-2) (where Δ _i = α _i · (e _i −C _INi )) The above equation 5-1 is obtained by rewriting the third equation, and the second term is replaced with Δ _i. , 5-2. That is, the output light amount C _OUTi of voxel i is _equal to the input light amount C _OUTi.
It is defined as the sum of the corrected amount delta _i in _INi.
As is clear from the above equation transformation process, the concept of transparency (1−α _i ) is also included in this equation 5-2, and there is no difference in principle.

[Preferred Embodiment] Hereinafter, a preferred embodiment of an ultrasonic diagnostic apparatus according to the present invention to which the above-described principle of image processing is applied will be described. FIG. 5 is a block diagram showing the overall configuration of the ultrasonic diagnostic apparatus according to the present embodiment.

The ultrasonic probe 8 for capturing three-dimensional echo data includes an array transducer 22 and a scanning mechanism. The array vibrator 22 is configured by arranging a plurality of ultrasonic vibrating elements, and an ultrasonic beam is scanned by electronically scanning the array vibrator 22, thereby forming the scanning surface 10. The scanning mechanism includes the drive motor 9 and the encoder 26. The drive motor 9 generates a driving force for driving the array transducer 22 in the Z direction in FIG. The encoder 26 is a detector that detects the position of the array transducer 22 in the Z direction.
By mechanically scanning the array transducer 22 in the Z direction while electronically scanning, a three-dimensional echo data capturing space is formed. 5, the Z direction is the mechanical scanning direction of the array transducer 22, the X direction is the electronic scanning direction of the ultrasonic beam, and the Y direction is the depth direction along the ultrasonic beam.

The drive section 24 is controlled by a control section (not shown) and supplies a drive signal to the drive motor 9. The transmission / reception unit 30 is a unit that supplies a transmission signal to the array transducer 22 and processes a reception signal output from the array transducer 22. The transmission / reception unit 30 includes an amplifier, a detector, and the like.

A three-dimensional projection image forming unit (Vol-mode)
The computing unit 34 executes image processing based on volume rendering based on the above-described image processing method, and a three-dimensional projection image of the three-dimensional echo data capturing space is formed by the image processing. The image data of the stereoscopic projection image is temporarily stored in a DSC (digital scan converter) 36, and finally,
A stereoscopic projection image is displayed on the display 38.

In the present embodiment, the reception signal output from the transmission / reception unit 30 is supplied to the stereoscopic projection image forming unit 34, and separately from the opacity (opacity) determining unit 3
2 has been entered. The opacity determination unit 32 includes a line memory 40, a calculation unit 42, and an opacity determination table 44. The opacity α of the voxel of interest is determined by the opacity determining unit 32.

More specifically, the line memory 40 stores echo data for one ultrasonic beam. The arithmetic unit 42 receives the echo data e1 on the immediately preceding ultrasonic beam output from the line memory 40 and the echo data e2 on the ultrasonic beam. These echo data e1 and e2 are echo data adjacent in the electronic scanning direction at the same depth. FIG. 6 shows two adjacent ultrasonic beams #n among the ultrasonic beams formed by the array transducer 22.
And # n + 1. Echo data e1 of the same depth on such two adjacent ultrasonic beams
And e2 are used in the operation of the operation unit 42. Arithmetic unit 42
In, two calculations, an average value calculation and a difference value calculation, are executed. In calculating the average value,

The calculation of e = (e1 + e2) / 2 (6) is executed, and in the calculation of the difference value,

The calculation of Δe = | e1-e2 | (7) is executed. In addition, e represents an average value,
Δe represents the absolute value of the difference value.

In the opacity determination table 44, the opacity α is determined based on the average value e and the difference value Δe. FIG. 7 shows an example of a specific function. In the example shown in FIG. 7, a non-linear function F is set such that the opacity (opacity) gradually increases as the average value e increases and as the difference value Δe increases.

This is because, as shown in FIG. 8, the echo data from the tissue surface has a relatively large value and is distributed over a wide range, whereas the echo data of the scattered noise basically has a small value and exists locally. It is based on the property of doing. That is, if two coordinate axes of the average value e and the difference value Δe are used, it is possible to set an appropriate opacity in distinction from tissue and noise.

The opacity α thus set is transmitted from the opacity determining section 32 to the above-mentioned three-dimensional projected image forming section 34.
And the opacity is used in the voxel operation of the voxel.

In the above-described embodiment, the average value and the difference value are used. However, it is also possible to determine the opacity using other information as an index. In the above embodiment, the opacity is determined using echo data of another voxel of the same depth adjacent to the voxel. However, for example, the opacity is determined using echo data of voxels adjacent to the left and right. The opacity may be determined. In this case, a plurality of stages of line memories 40 may be provided.

As described above, in the above embodiment, when determining the opacity of a voxel of interest, the opacity is determined with reference to the echo data of the adjacent voxel in addition to the echo data of that voxel. Therefore, the tissue and the noise can be distinguished from each other to minimize the influence of the noise. Therefore, there is an advantage that a stereoscopic projection image more faithful to the actual tissue can be formed, and the diagnostic accuracy can be improved.

The opacity determination function F shown in FIG.
Is an example, and its function can be appropriately changed and used according to the target organ or other conditions.

[0054]

As described above, according to the present invention,
When forming a three-dimensional projected image, the influence of noise can be eliminated as much as possible. Further, according to the present invention, noise can be reduced without impairing the real-time property.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a relationship between a three-dimensional echo data capturing space and a three-dimensional projected image.

FIG. 2 is a diagram illustrating a relationship between an input light amount and an output light amount of each voxel.

FIG. 3 is a diagram showing a light emission amount of each voxel.

FIG. 4 is a diagram for explaining an output light amount of a voxel.

FIG. 5 is a block diagram showing an overall configuration of an ultrasonic diagnostic apparatus according to the present invention.

FIG. 6 is an explanatory diagram showing two adjacent echo data.

FIG. 7 is a diagram showing an opacity determination function.

FIG. 8 is a diagram illustrating properties of echo data on a plane defined by an average value and a difference value.

[Explanation of symbols]

8 Ultrasonic probe for capturing 3D echo data, 10
Scanning plane, 22 array transducer, 30 transmitting / receiving unit, 32
Opacity determining unit, 34 stereoscopic projection image forming unit (Vo
l-mode operation unit), 40 line memory, 42 operation unit, 44 opacity determination table.

Claims (6)

[Claims] A transmitting / receiving unit for obtaining echo data of each voxel in the three-dimensional region by transmitting and receiving an ultrasonic wave to and from a three-dimensional region by scanning an ultrasonic beam; A three-dimensional projection image forming means for forming a three-dimensional projection image of the three-dimensional region by sequentially executing a voxel operation based on volume rendering for each voxel, and echo data of the voxel of interest and one adjacent thereto. Or an opacity determining unit that determines opacity used in voxel calculation of the voxel of interest based on echo data of a plurality of adjacent voxels. 2. The apparatus according to claim 1, wherein the opacity determining means uses echo data of adjacent voxels of the same depth on an ultrasonic beam adjacent to the ultrasonic beam in which the voxel of interest exists. And determining the opacity of the voxel of interest. 3. The apparatus according to claim 1, wherein said opacity determining means calculates a difference value of echo data by using echo data of said voxel of interest and echo data of said adjacent voxel. Average value calculating means for calculating an average value of echo data using the echo data of the voxel of interest and the echo data of the adjacent voxel; and opacity determining means for determining the opacity according to the difference value and the average value. An ultrasonic diagnostic apparatus, comprising: 4. The apparatus according to claim 1, wherein said stereoscopic projection image forming means calculates opacity α _i of voxel i based on echo data e _i , and opacity calculating means based on echo data e _i A transparency calculating means for calculating the transparency β _i of the voxel i, and multiplying the echo data e _i by the opacity α _i to obtain a voxel i
A light emission amount calculating means for calculating the light emission amount of the voxel i; a transmitted light amount calculating means for calculating the transmitted light amount of the voxel i by multiplying the output light amount of the previous voxel i-1 by the transparency β _i of the voxel i; And a light amount adding means for adding the amount of light and the amount of transmitted light to obtain an output light amount of the voxel i, wherein the three-dimensional projection image is formed by making the output light amount of the end voxel correspond to a pixel value. Ultrasound diagnostic equipment. 5. The apparatus of claim 1, wherein the stereoscopic projection image forming means, and opacity calculating means for calculating the opacity alpha _i voxel i based on the echo data e _i, the echo data e _i, The opacity α _i and the input light amount C corresponding to the output light amount of the previous voxel i−1.
Output light amount calculating means for calculating the output light amount C _OUTi of the voxel i based on _INi , wherein the three-dimensional projected image is formed in such a manner that the output light amount of the end voxel corresponds to the pixel value. Ultrasound diagnostic device. 6. A transmission / reception unit for obtaining echo data of each voxel in the three-dimensional region by scanning an ultrasonic beam and transmitting / receiving an ultrasonic wave to / from the three-dimensional region, A three-dimensional projection image forming means for forming a three-dimensional projection image of the three-dimensional region by sequentially executing a voxel operation based on volume rendering for each voxel, and echo data of the voxel of interest and one adjacent thereto. Or, based on echo data of a plurality of adjacent voxels, means for determining a value of a parameter used in the voxel calculation so that noise is suppressed in the voxel calculation of the voxel of interest. Ultrasound diagnostic device.