JPS62192640A - X-ray image processor - Google Patents

Abstract

PURPOSE:To securely remove a scattered X-ray by calculating scattered X-ray components by an arithmetic means which filters a filter coefficient and an original image in an actual or frequency space, and subtracting the scattered X-ray components from the original image in a spot space. CONSTITUTION:A two-dimensional memory 1 stores the original image.T photographed with an X-ray and a scattered X-ray response function storage memory 2 stores the Fourier-transformed value of a scattered X-ray response function. A filter coefficient arithmetic circuit 3 inputs the output of the memory 2 and a photographic condition coefficient outputted by an X-rays rack 7 and performs arithmetic operation based on a specific expression to calculates the filter coefficient in the frequency space. A inverse Fourier transformer 4 receives the output of the circuit 3 and processes the filter coefficient in the frequency space by inverse Fourier transformer to calculate the filter coefficient F in the actual space. A filter arithmetic circuit 5 convolutes the original image T and filter coefficient in the actual space, namely, perform arithmetic operation based upon a specific expression to calculate scattered X-ray components. A subtracter 6 subtracts the scattered X-rays components from the originals image T in the actual space and outputs an image consisting of direct X-ray components.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to an apparatus for processing an X-ray image constructed based on the amount of X-rays transmitted through an object, and more specifically, The present invention relates to an X-ray image processing device that removes scattered X-rays included in the image.

(Prior Art) In general, an X-ray detector in an X-ray diagnostic device has two types of equipment: direct X-rays that are effective as image information, and scattered Xv scattered by a subject, etc.
A is incident. These scattered X-rays are the main cause of deteriorating the contrast and sharpness of images. For this reason, it is extremely important to remove the scattered XL9< in order to improve image quality.

Conventionally, there have been the following two methods for obtaining an X-ray image from which scattered X-rays have been removed.

In the first method, the subject is photographed by irradiating X-rays through a small lead piece before or after the main irradiation for collecting image data. Then, the X-ray dose of the lead shadow is actually measured as a local scattered dose. Then, spatially find the local values at several points,
From this, the scattered radiation distribution in the entire space is estimated using the interpolation method. Then, each time the scattered ray distribution is subtracted from the X-ray image obtained by the main irradiation, an X-ray image from which scattered X-rays have been removed is obtained.

The first method described above lacks accuracy because it obtains the scattered radiation distribution by interpolation processing, and also has zero B5. In addition to zeta radiation, X-ray irradiation is essential to obtain the scattered ray distribution, which has the disadvantage of increasing the radiation dose to the subject. Furthermore, a mechanism is required to bring the lead pieces out into the irradiation area, which poses an obstacle in reducing the size of the device.

As a second method, a method of removing scattered X-rays by filtering has been proposed, but it has been difficult to put into practical use because the accuracy in the peripheral area of the irradiation area is insufficient.

(Problem to be solved by the invention) As mentioned above, the scattered X
The method of determining the line distribution and subtracting it from the X-ray image has many obstacles in ensuring the removal of scattered X-rays, reducing the exposure dose to the subject, and downsizing the device. However, even methods for removing scattered X-rays could not be put to practical use due to accuracy problems.

Therefore, the present invention aims to eliminate the above-mentioned drawbacks, and aims to provide a highly practical X-ray image processing device that can reliably remove scattered radiation and that does not impede the reduction of radiation exposure and miniaturization of the device. shall be.

[Structure of the invention]

(Means for solving the problem) X′! according to the present invention! In the ffA image processing device, an original image including a direct X-ray component and a scattered X-ray component is stored in the first storage section, and the response function of the scattered X-rays on the detector surface to the X-rays incident on the subject is Fourier-transformed. A first calculation means calculates a filter coefficient in the frequency space from the output of the second storage part and the photographing conditions of the subject, and further calculates the filter coefficient in the frequency space from the output of the second storage part and the photographing conditions of the subject. A second calculation means for filtering the image in real space or frequency space calculates a scattered X-ray component, and a subtraction means subtracts the scattered X-ray component from the original image in point space. There is.

(Function) The present invention utilizes the experimental properties of scattered X-rays shown in (1) and (2) below.

■ Scattered X-rays have a response function PSF(x,
y). This response function P S
F (x, y) has approximately the same spread on the detector surface regardless of the thickness of the object, given (i) the camera tube voltage, (ii) the distance between the object and the detector, and (iii) the grid. . In the case of digital images, this spread is extremely large compared to the size of 1 pixel, and it is
This is convenient in that it allows pulling.

■ The integral value of the above response “j;! IP S F (x, y), that is, the amount of scattered X-rays S, is the amount of directly transmitted X-rays (unknown amount)
P is given by the following empirical formula.

5=A-P'+B-P...(1
) Here, the parameters A and B in the above equation (1) are values determined by the following conditions (iv) to (vii) in addition to the conditions described in section (2).

(iv) Shooting tube current (v) Gain of signal conversion system (including aperture ratio of optical diaphragm in case of digital fluorography (DF) system) (vi) FDD (focal point-detector distance) (vi)
Irradiation field area Also, the index n has a value close to 1, and within the range of the human body targeted by medical X-ray equipment and the irradiation tube voltage value, n = 0.9.
It is given in 5.

For reference, one imaging condition in the DF system and the experimental formula for scattered radiation under this condition are shown below.

Photography conditions> Tube voltage 116kVp Tube current
60mA irradiation time
33m5 (continuous X-ray) aperture aperture ratio
0.024 focus-detector distance 11iIf (FDD)
100 cm object-detector path iT+1! (PDD
) l Ocm Phantom Water Grid 401p/cm.

Height: Bisochi = IO:l Spacer material wood, 2mm thick, parallel grid irradiation field
23cmX 23cm (9 inch l T) <Scattered radiation experimental formula>
=5.83XP'・''-2,69XP Based on the above conditions (1) and (2), a method of removing scattered X-rays from an image on which scattered X-rays are superimposed and extracting only the direct X-ray component will be described.

From the above conditions ■ and ■, the scattered X-ray distribution S(x, y) can be directly
When expressed as a line distribution P(x, y), S (x, y) Yff n (A-p'(x-x
',y-y')+B'P(x-x',y-
y' )). Here, D represents an irradiation field and is a response function of random X-rays to incident X-rays, which is given by:

By the way, since n is close to 1, P'(x, y) can be linearly approximated. Approximation methods include a method of Taylor expansion around the door and a method of approximation using the following equation (4).

Pn(x,y) = to-P(x,y)
= (4) Here, the coefficient is calculated from the maximum concentration T□8 of the entire irradiation field area, K=T□X / T Ill M M
...It can be specified by (5). In order to obtain higher accuracy, K = P, , "/ P IIax may be set from the maximum concentration P wax of the entire irradiation field area by direct X-rays.

Here, P, , , X are unknown, so the above equation (1)
Using, T□X=SIIIX →-pH1,
It is sufficient to obtain P□8 from IX + (B + 1) Pox and determine K.

Below, we will explain using the approximate equation given by equation (4). First, if equation (2) is approximated by equation (4), then S(x, y)
= (A- to B) ・ffoPCx-x'
, y-y') Here, the image T (x
, y) is given by the sum of the scattered X-ray distribution S(x, y) and the direct X-ray distribution P(x, y), so T(x, y) = S(x, y) P(x, y)...
・(7) becomes. The above formula (7) is simplified and expressed as the following 2 (8).

Note that C=AK+B, and X means convolution.

The X-ray distribution P is directly determined from the above equation (8).

First, if we Fourier transform both sides of equation (8), we get "r'
(w)=C-P (w) ・P S F(w)+P
(w)-P (w)(C-P SF(w)+1)...
・(9) becomes. Furthermore, T(w), P(w), P
S F (w) where W is a two-dimensional vector.

On both sides of the above formula (9), C-P S F (w) /
(When multiplied by (1+C-PSF(w)), =C-P (w) ・P SF(w) ”・(10
).

The above formula (10) is, that is, the Fourier plane h (on frequency space)
Multiplying by a value equal to is equivalent to filtering the inverse transformation F(x,y) of F(w) in real space.

Therefore, equation (10) represents the operation. The right side of this equation (12) is
As is clear from equation (8), this provides a scattered radiation component. Therefore, when calculating P by substituting equation (12) into equation (8), P = T - TxF - (13)
That is, by subtracting the image TxF (scattered radiation image) from the original image T, an X-ray image can be directly constructed.

It can be treated as a stationary filter because it does not depend on the thickness of the subject due to the experimental property of scattered radiation. Furthermore, C1 is a multiplier that can be calculated once the imaging conditions are determined.

Therefore, based on the original image T and the Fourier transform of the response function, the first calculation means executes equation (11), the second calculation means executes equation (12) (this is also possible in the frequency space), and then the subtraction means By executing equation (13), an X-ray image can be directly obtained.

(Example) Embodiments of the present invention will be described below with reference to the drawings.

<First Example> FIG. 1 is a block diagram of an X-ray image processing apparatus according to a first example. In the figure, a two-dimensional memory 1, which is a first storage unit, stores an original image T taken by an X-ray imaging device (directly
ray components and scattered X-ray components). Second
It is stored as scatter-transformed P S F (w), which is the storage unit of . The filter coefficient calculation circuit 3, which is the first calculation means, calculates the filter coefficient F (w) in the frequency space by calculating the above-mentioned equation (11). For this purpose, the filter coefficient calculation circuit 3 receives the output P S F (w) of the storage memory 2 and the imaging condition coefficient C output from, for example, the X-ray gantry 7 or the X-ray controller. There is. Inverse Fourier transformer 4
is to perform inverse Fourier transform on the filter coefficients in the frequency space to calculate the filter coefficients F(x, y) in the real space. The filter calculation circuit 5, which is a second calculation means, convolves the original image T and the filter coefficient F in real space, that is, calculates the scattered X-ray component by executing the calculation of equation (12). It is something to do. A subtracter 6, which is a subtracting means, calculates scattered X from the original image T in real space.
Subtracting the line component element HF creates an image P with only the direct X-ray component
This outputs the following.

In this manner, according to the first embodiment, the response function of scattered X-rays, which is determined regardless of the thickness of the subject, is stored in advance in the storage memory 2 in the form of Fourier transform, and the imaging condition coefficient C, which changes depending on the imaging conditions, is stored in the storage memory 2 in advance. Based on the output of , the filter coefficient calculation circuit 3 executes equation (11) to calculate the filter coefficient F (w) in the frequency space. An example of this filter coefficient F (w) is shown in FIG. 2(A). Also,
This filter coefficient F (w) is subjected to inverse Fourier transform to obtain the filter coefficient F (x, y) in real space. An example of this filter coefficient F(x,y) is shown in FIG. 2(B).

Thereafter, by convolving this filter coefficient F and the original image T in the filter calculation circuit 5, the scattered X-ray component S in the real space is obtained based on the above-mentioned action, and this scattered X-ray component S is converted into the original image T. By subtracting from the image T, an X-ray image P consisting only of direct X-ray components can be obtained, as is clear from equation (13).

In this way, in the first embodiment, it is possible to perform reliable removal of scattered X-rays by utilizing the experimental properties (1) and (2) of scattered X-rays, and it can be used as a highly practical X-ray processing device. I can do it. Moreover, since the scattered X-rays are removed, the exposure dose does not increase, and no complicated structure is required.

Second Example> As mentioned above, the convolution in the real space shown by equation (12) converts the filter coefficient F (w) shown by equation (11) into the original image (two-dimensional Fourier transformed).

Therefore, in this second embodiment, the above-mentioned convolution is performed in frequency space.

The apparatus of the second embodiment is constructed as shown in FIG. 3, and among the reference numerals in the figure, members having the same functions as those of the first embodiment are designated by the same reference numerals. The differences from the first embodiment are: (1) a two-dimensional Fourier transformer 10 for two-dimensional Fourier transform of the original image T is provided, (2) the inverse Fourier transformer 4 is removed, and (2) the second calculation means 11 is provided. ■Second calculation means 11 that performs convolution on frequency space
A two-dimensional inverse Fourier transformer 12 is provided to perform two-dimensional Fourier transform on the output.

According to this second embodiment, the calculation of equation (11) is performed in the frequency space in the second calculation means 11, and by inverse Fourier transforming this output, the first
It is possible to obtain the same output T I can do it. Therefore, this second embodiment can also achieve exactly the same effects as the first embodiment.

Note that the second embodiment is different from the first embodiment in that the two-dimensional Fourier transformer 1o and the two-dimensional inverse Fourier transformer 12 are essential, and in that the calculation time in the double-degree transformers 10 and 12 is required. disadvantageous in comparison.

It should be noted that the present invention is not limited to the above two embodiments,
It goes without saying that various modifications can be made within the scope of the invention.

〔Effect of the invention〕

As described in detail above, according to the present invention, scattered X-rays can be reliably removed based on the experimental properties of scattered X-rays, and the exposure dose does not increase during the process, and No complicated mechanical configuration is required.

Therefore, a highly practical X-ray image processing device can be provided.

[Brief explanation of drawings]

FIG. 1 is a block diagram of the device according to the first embodiment of the present invention, and FIGS. 2 (A) and (B) show the filter coefficient F(w) in the frequency space and the filter coefficient F in the real space in the same device. (x
, y), and FIG. 3 is a block diagram of an apparatus according to a second embodiment of the present invention. 1... First storage unit, 2... Second storage unit, 3...
・First calculation means, 5.11...Second calculation means, 6
...Subtraction means. Agent Patent Attorney Noriyuki Chika Yudo Daiko Norio 7 (ηfn(A) Figure 2)

Claims (1)

[Claims] a first storage section that stores an original image including a direct X-ray component and a scattered X-ray component; and a second storage section that stores a response function of the scattered X-rays on the detector surface to the subject incident X-rays in a Fourier transformed format. a storage section; a first calculation means for calculating a filter coefficient in frequency space based on the photographing conditions of the subject and the output of the second storage section; a second calculation means for calculating scattered X-ray components by filtering in frequency space; and subtracting the scattered X-ray components from the original image in real space to output an X-ray image with only direct X-ray components. 1. An X-ray image processing device comprising: subtraction means.