JPH0320107B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention converts an X-ray transmission diagnostic image into an optical image via an image intensifier, and captures this optical image with a television camera to improve the contrast resolution of an X-ray television image. The present invention relates to a contrast correction device for X-ray television images that can improve the contrast of X-ray television images.

Recently, an X-ray diagnostic image obtained by X-ray fluoroscopy of a subject (patient) is converted into an optical image via an image intensifier, and this optical image is captured by a television camera and displayed on a monitor television. The image intensifier method, which uses images to perform diagnostic medicine, is attracting attention. According to this image intensifier indirect photography method, the amount of radiation required to obtain one diagnostic image is small, and therefore the load on the X-ray tube is reduced and it is possible to perform small-focus imaging. It has excellent characteristics. Moreover, the resolution of the image intensifier indirect image obtained by this small focus imaging is comparable to the resolution of the direct X-ray image obtained using conventional X-ray film.
Direct X-ray imaging is now being used in a wide range of fields, including diagnostic medicine.

However, image intensifiers have their own problem of reduced image contrast due to scattered radiation.

Therefore, in the past, generally speaking, a light guide effect was introduced to the input fluorescent surface of the image intensifier, and a dark face plate was provided to the output surface of the image intensifier. Efforts have been made to prevent the negative effects of Also, image intensity X
By changing the material of the line entrance window from glass to metal,
Efforts have also been made to improve contrast.
However, the contrast ratio obtained by such a device is about 17:1 at most, which is about 26 dB when converted to S/N. Therefore, it was difficult to obtain high-quality images.

Now, the reduction in contrast in the image intensifier is analyzed as follows. Figure 1 is an experimental configuration diagram for measuring the contrast characteristics of image intensifier 1, in which 2 is an X-ray tube, 3 is a microphotometer, and 4 records the data measured by this microphotometer 3. It is a recorder. Therefore, on the X-ray entrance surface of the image intensifier 1, there is a lead disk 5 with a diameter d.
is provided, and X-rays are made to be incident on this X-ray entrance surface from the X-ray tube 2 via the X-ray aperture 6.

Now, when the lead disk 5 is not provided, the image
If the value detected by the microphotometer 3 via the intensifier 1 is a, and the value detected when the lead disk 5 is provided is b, then the contrast ratio c is as follows: defined.

c=(a-b)/a=1-b/a In other words, ideally, the amount of X-ray transmission at the position where the lead disk 5 of the image intensifier 1 is provided is zero (0), that is, b=0. However, in reality, b>0. This is due to an increase in the background due to stray X-ray light incident from the periphery of the lead disk 5. That is, the image
As shown in the modeled structure of the X-ray entrance surface of the intensifier 1 in FIG. 2, the entrance surface has a structure in which a fluorescent surface 8 is formed on a face plate 7. Therefore, when the X-ray beam 9 is incident, among the light emitted by the fluorescent surface 8, the light component that travels to the output boundary surface of the face plate 7 at an angle equal to or greater than the critical angle is totally reflected. Returning again to the fluorescent surface 8. Then, the light is diffusely reflected by the fluorescent surface 8 and added to the signal light as stray light. As a result, the contrast described above is reduced. This phenomenon of contrast ratio reduction due to scattered rays is caused by the drive voltage of the X-ray tube 2 and the glass structure G and metal structure M of the incident surface of the image intensifier 1 as parameters shown in FIG. For example, if the diameter of the lead disk 5 is 10 mm or less,
Changes in contrast ratio can be kept small. In addition, in FIG. 3, the background increase indicates the reciprocal of the contrast ratio. Therefore, the influence of scattered radiation in the vicinity is small, and 10
It can be seen that the influence of stray light is large at positions that are more than mm away. However, this positional relationship also changes depending on the composition of the face plate 7, its thickness, etc.

The object of the present invention is to obtain a high-contrast image by removing the scattered ray effects from an X-ray television image obtained through such an image intensifier. A contrast correction device for X-ray television images that can improve the contrast of X-ray diagnostic images by correcting scattered radiation through image signal processing, in addition to preventing scattered radiation by improving the configuration and structure of the entrance/exit surface. It is about providing.

The outline of the present invention is that when focusing on a certain area of the screen, the stray light superimposed on the signal light in that area can be considered to be mixed in with a two-dimensional spread from the periphery of the area, Therefore, the signal light is extracted by estimating these stray lights and subtracting the estimated stray lights from the output light of the above region, thereby correcting for the scattered rays (stray light). This effectively achieves the above objectives.

In particular, the image is divided into a plurality of small regions, each small region is subjected to weighting processing to obtain estimated stray light, and these are combined to easily perform the above-mentioned scattered light correction.

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

Now, when we focus on a small area consisting of m x m pixels in a diagnostic image that has been optically converted via an image intensifier, we can see that the small areas surrounding this small area have scattered rays (stray light). It is assumed that the area consists of M×M small area blocks as shown in FIG. That is, assuming that the image signal is extracted from a small area at the center position indicated by diagonal lines in FIG.
The image signal includes signal light obtained in the same small area and stray light due to scattering from the surrounding M×M small area. Therefore, by weighting each signal of these M×M small areas and calculating their sum, the stray light component of the central small area can be canceled out, thereby effectively eliminating the negative effects of scattered radiation. It becomes possible to correct the

Now, if the signal average value of the pixel group of each small area consisting of m×m pixels is Xl,k, and the weighting coefficient determined corresponding to each small area is Al,k, then
A signal output Y obtained by weighting each signal average value of an M×M small area and then combining it is shown as Y= 〓 ^l,k Al,k·Xl,k. Now, if the above output under ideal conditions without the influence of scattered radiation is Yo, then the error ε is expressed as ε=Y−Yo=〓 ^l,k Al, k·Xl, k−Yo. Therefore, if we define this error evaluation value E as E=Σε ^{2 =} Σ[ 〓 ^l,k Al, k It will be done. In other words, in this case, if we partially differentiate the error evaluation value E with respect to the weighting coefficient Al, k and find the condition that the partial differential value becomes zero ∂E/∂Al, k=0, then all l, k The optimal solution will be given. Since the above partial differential value is expressed as ∂E/∂Al, k=Σ2・Xl, k・ε, the corrected change amount of Al, k is △Al, k=α・∂E/∂Al, k ( However, if α is a coefficient) and iteratively corrects the coefficient as Al,k'=Al,k+ΔAl,k, the optimum correction coefficient that minimizes the error can be obtained. Therefore, according to such an algorithm, weighting coefficients for each M×M small area are determined, and the signals of each small area are weighted using these weighting coefficients to obtain estimated stray light, and the estimated stray light is calculated from the small area at the center position. By combining these signals, it becomes possible to effectively correct the effect of scattered radiation.

FIG. 5 shows the configuration of a correction circuit according to an embodiment that corrects the weighting coefficients according to the above-mentioned algorithm and obtains an X-ray television image in which the influence of scattered radiation is corrected. An example in which a small area of 3×3) is targeted for correction processing is shown below.

The mask 11 cuts out a 3×3 adjacent small region of m×m pixels as a unit from an X-ray television image transmitted through an image intensifier. A plurality of weighting circuits 12 that apply predetermined weighting to the image data of each small area cut out by this mask 11 are provided corresponding to each of the small areas, and these weighting circuits 1
The image data subjected to the weighting process via the adder 2 is led to the adder 13, where they are combined and added.

In other words, the averaged image data of each small area is
Let Xl, k (l, k = -1, 0, 1), and the data of the small area at the center position to be processed is Xo, o
In this case, the weighting circuit 12 is given weighting coefficients Al and k corresponding to each of the above-mentioned small areas. Therefore, each output of the weighting circuit 12 is Xl,
k・Al,k, and as the composite addition output Y of the adder 13, Y= 〓 ^l,k Xl,k・Al,k is obtained.

On the other hand, this combined addition output Y is led to a comparator 14 and compared with its ideal value Yo to determine the error ε. The ideal value Yo is calculated from a signal under ideal conditions obtained for each small region using, for example, a lead disc phantom with a known outer diameter, and is calculated from the signal under ideal conditions obtained for each small region. determined each time. Then, the value of the error ε calculated by the comparator 14 in this manner is given to the coefficient setter 15. This coefficient setter 15
For example, as shown in FIG. The above-mentioned weighting coefficients Al and k are obtained by integrating the low-frequency waves (averaging). That is, according to the error ε of the composite output Y obtained according to the weighting coefficient Al,k given in advance, the coefficient variation amount ΔAl,k is calculated by multiplying the average value by a constant α,
The above coefficients Al and k have been corrected. This correction is then iteratively processed using the algorithm described above to minimize the error ε. in this case,
The above correction may be terminated when the error ε becomes less than or equal to a predetermined amount after a predetermined number of corrections.

After setting the weighting coefficients in this way, for example, for all m x m pixels constituting a small area at the center position to be subjected to correction processing, for each pixel under the weighting coefficient conditions set above, By performing data correction processing on the data, it becomes possible to effectively cancel the stray light components included in each of these pixel data and perform contrast correction thereof. and,
After the contrast correction for the m×m pixels is completed, the small area to be processed may be shifted by one small area, information on the surrounding small areas may be obtained, and contrast correction may be performed in the same manner. Therefore, by performing contrast correction while sequentially shifting the position of the mask 11 with respect to the X-ray television image, effective contrast correction can be performed over the entire area of the X-ray television image using image processing technology. becomes possible.

Although the square mean of the error values ε is used here as the evaluation function, it is also possible to use the n-th power mean, the absolute value average, or even the n-th power norm. Furthermore, if uniformity can be assumed for one image block divided into small regions, the same (M×M) coefficient values Al and k can be used for the entire image. Furthermore, if there are shading sensitivities or geometrical differences due to image intensifiers, X-ray characteristics, etc., weighting coefficients Al and k should be determined for each image block cut out by the mask 11. Bye.

As described above, according to the present invention, the average value of the influence of scattered radiation in the divided small regions is estimated, and the influence of each scattering property on the small region to be corrected is estimated and cancelled. Since the contrast is corrected using the image data, a remarkable contrast correction effect can be obtained. In particular, since it is possible to obtain a contrast correction effect that is different from the scattering radiation prevention achieved by structurally improving the input/output surface of the conventional image intensifier, the conventional scattered radiation prevention technology and the scattered radiation correction processing technology of the present invention are used together. As a result, a tremendous contrast improvement effect can be achieved. Further, the contrast correction processing described above can be easily and highly accurately executed using hardware or software as an image processing technique. Therefore, it can be expected to have a much higher contrast improvement effect than conventional structural/configural contrast correction, and has great practical advantages.

Therefore, for example, in X-ray cinegraphy and X-ray photography of the heart and vascular system, contrast is increased by injecting a contrast agent, but even in this case, since the coronary vasculature has a fine structure, , the contrast is greatly reduced. Therefore, if this method is applied to such X-ray diagnosis, the contrast can be easily increased, and the diagnostic medical effect is enormous.

Note that the present invention is not limited to the above embodiments. For example, the number of small areas constituting an image block can be further increased, and the size of the small areas can be changed as appropriate. Furthermore, when the blurring of the target signal component becomes a problem, the averaging process of the image block for the signal light may be stopped. Furthermore, the algorithm for modifying the weighting coefficients Al and k can be modified in various ways. In short, the present invention can be implemented with various modifications without departing from the gist thereof.

[Brief explanation of the drawing]

Figure 1 is an experimental block diagram for measuring X-ray scattered rays from an image intensifier, Figure 2 is a diagram schematically showing the mechanism of X-ray scattering, and Figure 3 is a diagram showing the characteristics of changes in contrast ratio. 4 is a diagram showing divided small areas for explaining the outline of contrast correction processing according to the present invention, FIG. 5 is a block diagram of a contrast correction circuit according to an embodiment of the present invention, and FIG. The figure is a partial configuration diagram of the coefficient setter. 11...Mask, 12...Weighting circuit, 13
... Adder, 14 ... Comparator, 15 ... Coefficient setter, 16 ... Multiplier, 17 ... Integrator.

Claims (1)

[Scope of Claims] 1. A means for converting an X-ray transmission diagnostic image into an optical image via an image intensifier and capturing this optical image with a television camera to obtain an X-ray television image; A means for dividing an X-ray television image into a plurality of small regions and obtaining image signals for each small region, and a means for multiplying each image signal obtained from each of these small regions by a coefficient, and then processing the coefficients. means for determining the total sum value of the image signals of each of the small regions obtained under predetermined conditions; and means for determining the total sum value of the image signals of each of the small regions obtained under predetermined conditions, and the influence of scattered radiation in the image intensifier. means for determining an error evaluation value from an ideal value set as an image signal free of image signals, and repeatedly correcting coefficients for each of the small regions so that the error evaluation value is minimized; and the error evaluation value is minimized. an X-ray television set comprising: means for outputting the sum of the image signals of each of the small areas, which is determined using the coefficient of Contrast correction device for Jiyoung images. 2. The X-ray television image contrast correction apparatus according to claim 1, wherein the image signal of each small area is obtained by averaging pixel signals within the small area. 3. The contrast correction device for X-ray television images according to claim 2, wherein the averaging process of pixel signals within a small area is performed only for a small area that does not contain signal light.