JP2006311922A - X-ray equipment - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide X-ray equipment with an automatic gradation processing function, capable of setting the pixel value in a region of interest approximately at a target output value, and obtaining an image with appropriate density and contrast both in a low X-ray absorbent region and in a high X-ray absorbent region. <P>SOLUTION: The equipment comprises: a means for extracting a region for extracting a quantity of characteristics from input image data; a means for normalizing the pixel value of the input image data and converting the tone of the pixel value based on the quantity of image characteristics; a means for storing a plurality of basic tone conversion characteristics; a means for storing the target output value corresponding to the quantity of image characteristics; a means for comparing the output value of the basic tone conversion characteristics and the target output value and selecting the basic tone conversion characteristics with the minimum error for the target output value; a means for preparing the gradation characteristics by synthesizing the normalized tone conversion characteristics and the selected basic tone conversion characteristics; and a means for converting the input image data to a prescribed pixel value based on the tone conversion characteristics. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an X-ray imaging apparatus, and particularly has an automatic gradation processing function that digitizes an image within a certain range even when the X-ray dose varies depending on the imaging region and imaging direction, and always obtains an appropriate image density and contrast. The present invention relates to an X-ray imaging apparatus.

In an X-ray imaging apparatus for general imaging examinations such as the chest and abdomen, digital X-ray imaging image data is obtained using an X-ray detector such as an X-ray flat panel detector (hereinafter referred to as FPD). The acquired image data is subjected to various correction processes such as offset correction, gain correction, and defect correction of the X-ray detector. After image enhancement processing such as dynamic range compression processing and filter processing is performed on the data obtained by this correction processing (hereinafter referred to as input image data), the processed image data is saved on the hard disk and automatically displayed. The display look-up table, display window width, and display window level determined by the gradation processing are stored in the hard disk as supplementary information of the image.
Then, the processed image data is converted into a display image using the supplementary information, and the converted image is displayed on a monitor or output to an imager.

The above input image data undergoes tone conversion in order to digitize the image within a certain range even if the X-ray dose incident on the X-ray detector changes depending on the imaging menu (imaging site and imaging direction). There is a method disclosed in Document 1.
The method according to Patent Document 1 extracts a reference value N1 where the X-ray dose is low and the signal level is minimum in the image, and a reference value N0 where the X-ray dose is high and the signal level is maximum in the image. The density is automatically corrected by converting the image signal N stored in the image memory into d by the logarithm shown in 1) and (Equation 2).
d = β · log (N0 / N) (Formula 1)
(Where d is the corrected image signal, β is a constant)
β = dmax / {log (N0 / N1)} (Formula 2)
(However, dmax is the maximum quantization value after density correction.)
Japanese Patent No. 3375237

  For example, for the chest front image in an imaging device using FPD during general imaging inspection, the minimum and maximum values of the image disclosed in Patent Document 1 are used as image feature amounts, and the input image data is converted into different data. When the conversion process, that is, the gradation process is performed using logarithm, in the image obtained by this conversion, the histogram of areas with high X-ray absorption such as the heart, mediastinum, and bone areas becomes wide, so the contrast is high. On the contrary, since the histogram of the low X-ray absorption region such as the lung field becomes narrow, the contrast may be lowered.

In general radiography, for example, there are various imaging parts such as the front of the chest, the front of the abdomen, the front of the thoracic vertebra, the front of the hands, the front of the legs, the Achilles tendon, etc. When an image is easy to see, and an imaging region is the front of the thoracic vertebra, an image that can easily see a bone region such as the thoracic vertebra is required.
In addition, when the imaging site is an Achilles tendon, an image that allows easy viewing of soft tissue is required.

  As described above, in general imaging, in order to provide an image that can be easily diagnosed by a doctor for various imaging sites, a low-dose region is not a logarithmic gradation characteristic using the minimum and maximum pixel values. It is desirable to have gradation characteristics that can automatically adjust the contrast in the high dose region.

In addition, the input image includes a photographed image of a patient including a metal having high X-ray absorption such as a bolt in the body. Even in such an image, the brightness of the region of interest (for example, lung field, mediastinum) ) And contrast (for example, lung field, mediastinum) and stable gradation processing can be automatically performed.
For this purpose, it is necessary to be able to convert the pixel value of the region of interest into a target output value.

  The object of the present invention has been made in view of the above problems, and can convert the pixel value of the region of interest into a target output value, which is suitable for the low X-ray absorption region and the high X-ray absorption region. It is to provide an X-ray imaging apparatus having an automatic gradation processing function capable of obtaining a contrast image.

The X-ray imaging apparatus of the present invention for solving the above problems is achieved by the following.
(1) X-ray generation means for generating X-rays to be irradiated on the subject, X-ray detection means for detecting transmitted X-rays of the subject that are arranged opposite to the X-ray generation means, and the X-ray detection means Gradation conversion means for correcting the detected digital image data and converting the corrected input image data into output image data within a predetermined range, and image processing for the converted image data converted by the gradation conversion means An X-ray imaging apparatus that generates a desired X-ray imaging image, wherein the gradation conversion unit includes a feature amount extraction unit that extracts an image feature amount from the input image data, and a pixel value of the input image data Normalization gradation converting means for normalizing the image feature quantity with the image feature quantity, normalized feature quantity conversion means for converting the image feature quantity into normalized feature quantity with the normalized gradation conversion means, and a plurality of basic gradation levels Basic gradation conversion characteristic storage means for storing conversion characteristics; A target output value storing means for storing a target output value corresponding to the image feature quantity, and a basic value that minimizes an error with respect to the target output value by comparing the output value of the basic gradation conversion characteristic with the target output value. Basic gradation conversion characteristic selecting means for selecting gradation conversion characteristics, and gradation conversion characteristic creating means for generating gradation conversion characteristics by combining the normalized gradation conversion characteristics and the selected basic gradation conversion characteristics And a gradation processing means for converting the input image data into a predetermined pixel value based on the gradation conversion characteristics.

(2) The feature amount extraction unit includes a region extraction unit that extracts a region for extracting an image feature amount from the input image data.

(3) Further, the area extraction means includes means for setting an area corresponding to the imaging region, and is means for extracting an image feature amount from the set area, and the imaging is performed in the basic gradation conversion characteristic storage means. A plurality of basic gradation conversion characteristics corresponding to a part are stored, and a target output value corresponding to an image feature amount corresponding to an imaging part is stored in the target output value storage unit.

(4) The image feature amount is a maximum value, an average value, or a minimum value of pixel values in the region extracted by the region extraction unit.

By configuring in this way, the gradation characteristics are not a single function using logarithm, but a plurality of piecewise functions are used. Since the compression of the side image data can be automatically adjusted, for example, when the imaging region is the front of the chest, the contrast of the lung field can be improved as compared with the case where the conventional logarithm is used.
In addition, even when comparing the histograms of the image after gradation conversion, the lung field histogram is expanded and the contrast of the lung field shadow, which is the most important for the front image of the chest, is obtained compared to the case where the conventional logarithm is used. An image useful for diagnosis can be provided.
Therefore, in the case where the imaging region is the front of the chest, the image density and contrast of the lung field and mediastinum are appropriate, and the lung field that is particularly important for the chest front image should have a stereoscopic effect. Can do.

According to the present invention, a plurality of functions obtained by connecting a function on the low X-ray dose side and a function on the high X-ray dose side are stored in the basic gradation conversion characteristic storage means, and an imaging region based on the image feature value and the target output value, Basic gradation conversion characteristics suitable for the shooting menu such as the shooting direction are automatically selected and gradation processing is performed, so the image density and contrast on the low X dose side and high X dose side are appropriate. can do.
When the present invention is applied to chest front imaging, the image density and contrast of the lung field and mediastinum are appropriate, and the contrast of the shadow of the lung field, which is most important for the chest front image, is obtained, which is useful for diagnosis. Can provide a good image.

  The following is an automatic gradation processing function that digitizes an image within a certain range even when the X-ray dose incident on the X-ray detector changes depending on the imaging menu (imaging site and imaging direction), and always obtains an appropriate image density and contrast. A preferred embodiment of an X-ray imaging apparatus according to the present invention including the above will be described in detail with reference to the accompanying drawings.

FIG. 1 is an overall configuration diagram of an X-ray imaging apparatus having an automatic gradation processing function according to a first embodiment of the present invention.
This X-ray imaging apparatus includes an X-ray generator 2 having an X-ray tube for generating X-rays to be irradiated to the subject 1, and an X-ray movable diaphragm for limiting the irradiation range of the X-rays to be irradiated to the subject 1 Device 3, X-ray generator 2 and ceiling-supported X-ray generator support device 4 for supporting X-ray generators such as X-ray movable diaphragm device 3 and the like. An X-ray detector 5 that detects the transmitted X-ray of the specimen 1 and converts it into digital image data, an X-ray detector support device 6 that supports the X-ray detector 5 on the floor, and an X corresponding to the imaging conditions An X-ray high voltage device 7 for generating a high voltage to be applied to the X-ray tube of the X-ray generator 2 having an X-ray control function for controlling a dose, and a digital image detected by the X-ray detector 5 The data is subjected to various correction processes such as offset correction and gain correction of the X-ray detector of the X-ray detector 5, and the input image data processed by this correction process is processed. An image processing device 8 that generates a captured image to be displayed from image data obtained by converting the converted image data into a certain range of image data, and performing dynamic range compression processing, filtering processing, and the like on the converted image data, and the image Display device 9 (CRT, LCD monitor, etc.) that displays the captured image generated by the processing device 8, and various parameters necessary for setting the imaging conditions, inputting the imaging part name, inputting the patient name, and imaging and image processing And an input device 10 for performing the setting and the like.
Although not shown, the image processing device 8 includes the image data obtained by performing the image processing, a display lookup table necessary for generating a display image from the image data, a display window width, a display A hard disk for storing incidental information such as a window level is connected.

  The X-ray movable diaphragm device 3 is configured so that the X-ray irradiation field can be limited to an appropriate range using four movable blades (lead plates), and the X-ray generation device 2 is not illustrated. However, an X-ray compensation filter made of copper or aluminum plate for controlling the X-ray intensity distribution restricted by the X-ray movable diaphragm device 3 can be added.

  The X-ray detector of the X-ray detector 5 may be any detector that can convert the X-ray data transmitted through the subject 1 into a digital value, such as an image intensifier and a CCD ( Charge Coupled Device (Charge Coupled Device) A combination with a camera or a detector called an X-ray flat panel detector (Flat Panel Detector) that is a semiconductor digital X-ray detector is used, and these detectors are removed. The X-ray detector 5 is composed of possible X-ray grids.

  As shown in FIG. 2, the image processing device 8 performs various correction processes such as offset correction and gain correction of the X-ray detector on the digital image data detected and output by the X-ray detection unit 5 to obtain input image data. Correction processing unit 8a to obtain, automatic gradation processing unit 8b that is a main part of the present invention that automatically converts the gradation of the input image data processed by the correction processing unit 8a into a certain range of image data, An image processing unit 8c that performs image processing such as dynamic range compression processing and filter processing on the image data converted by the automatic gradation processing unit 8b, and automatic display gradation processing on the image data processed by the image processing unit 8c And a display control processing unit 8d that generates a display image using the display lookup table, display window width, and auxiliary information at the display window level determined by the above.

Although not shown, the hardware of the image processing device 8 is a central processing unit (CPU) that controls the operation of each component, a main memory that stores a control program, etc., processed image data and incidental information Hard disk and temporary storage memory for storing data, etc., mouse and controller for operating soft switches on the screen of the display device (the same as the display device 9 in FIG. 1), and keys and switches for setting various parameters And a common bus for connecting the above-described components, which can be realized by a personal computer.
The input device using the mouse and keyboard is included in the input device 10 shown in FIG.

  FIG. 3 is a block diagram of an automatic gradation processing unit 8b which is a main part of the present invention for automatically converting the gradation of the input image data processed by the correction processing unit 8a into a predetermined range of image data. Is an example realized in the form of an executable program.

As described above, the automatic gradation processing of the X-ray imaging apparatus of the present invention shown in FIG. 3 described below is the relationship between the input image data and output image data obtained by gradation conversion of the input data. This is a process for automatically determining the gradient of the gradation characteristic.
By performing this process, the pixel value of the input image data is converted into an appropriate value regardless of whether the incident X-ray dose to the X-ray detector is small or large, and is generated using this converted pixel value. This makes the density and contrast of the displayed image appropriate.

Specifically, when the incident X-ray dose to the X-ray detector is small, the gradient of the gradation characteristics is automatically increased using the feature amount of the input image, so that the incident X-ray dose can be reduced. Pixel values used for image processing can be converted into appropriate pixel values.
Conversely, when the incident X-ray dose to the X-ray detector is large, the gradient of the gradation characteristic is automatically lowered using the feature amount of the input image, so that the image processing can be performed even if the incident X-ray dose is large. The pixel value to be used can be converted into an appropriate pixel value.
The input image is image data obtained by performing offset correction, gain correction, defect correction, and the like on the data detected by the X-ray detector 5. That is, the X-ray detector gain image obtained by subtracting the offset amount of the X-ray detector from the data detected by the X-ray detector, and further obtaining the subtracted value from the captured image data previously captured with an aluminum plate Is divided by a certain coefficient (for example, the average value of the gain image).
An input image is image data after correcting a defective pixel by performing defect correction after such offset and gain correction.

  In FIG. 3, the automatic gradation processing of the X-ray imaging apparatus according to the first embodiment of the present invention includes an area extraction unit 101 that extracts an image feature amount from an input image, and an image extracted from the area extracted by the area extraction unit 101. A feature extraction unit 102 for extracting feature amounts; a normalized gradation conversion characteristic creation unit 103 for normalizing pixel values of the input image using the image feature amounts extracted by the feature extraction unit 102; A normalized feature value creation unit 104 that normalizes the image feature value obtained by the feature extraction unit 102 using the normalized tone conversion property created by the normalized tone conversion property creation unit 103, and a plurality of basic features A basic gradation conversion characteristic storage unit 105 for storing gradation conversion characteristics; a target output value storage unit 106 for storing a target output value corresponding to the image feature amount extracted by the feature extraction unit 102; and the basic gradation conversion Output of basic gradation conversion characteristics stored in characteristic storage section 105 A basic gradation conversion that compares a value with a target output value corresponding to an image feature value stored in the target output value storage unit 106 and selects a basic gradation conversion characteristic that minimizes an error with respect to the target output value The characteristic selection unit 107, the normalized gradation conversion characteristic created by the normalized gradation conversion characteristic creation unit 103, and the basic gradation conversion characteristic selected by the basic gradation conversion characteristic selection unit 107 are combined and input. A gradation conversion characteristic combining unit 108 that generates gradation conversion characteristics suitable for image data, and the input image is converted into a predetermined pixel value using the gradation conversion characteristics generated by the gradation conversion characteristic combining unit 108. The gradation processing unit 109 is configured.

For example, when the imaging region is the front of the chest, the region extraction unit 101, as shown in FIG. 4, the region 1 is the entire image (d), the region 2 is the mediastinum (c), and the region 3 is the right lung field (a) Region 4 may be set in the left lung field (b). The center coordinates and size of these feature extraction areas can be determined in advance.
Each area to be extracted is a fixed area corresponding to the mediastinum before imaging (the central part of the image) or a fixed area corresponding to the lung field (the right and left areas at the top of the image). The coordinates of the end points are input and stored from the input device to the hard disk or memory of the image processing apparatus.
That is, the position and size of the feature extraction region of the imaging part are stored in the hard disk, memory, etc. according to the imaging part before imaging.

For example, in the case of the above-mentioned chest frontal imaging, the center coordinates and size of the feature extraction area are taken after positioning is performed such that the jaw is placed on the upper end of the X-ray detector, so the position of the lung field is approximately constant. For this reason, it is possible to set this area in a fixed manner.
Therefore, the center coordinates and size of the lung field, which is the feature extraction area, can be determined before imaging. Of course, the threshold value may be determined using the ratio of the direct X-ray pixel value in the lung field region, and the shape of the lung field region may be varied from the two obtained threshold values. Specific examples of this ratio indicate respective ratios of the upper limit value and the lower limit value of the pixel value, and the ratio takes values of 50% and 10%. In addition, since there are regions where the pixel values of the lung field and the skin pixel values are the same, a region obtained by removing a certain length (eg, 2 cm) from the top, bottom, left, and right directions of the threshold-processed image may be used as the lung field region.

In the case of small subjects, if the irradiation field size is 14 x 17 inches (corresponding to the size of a half-cut film), there is an area that is not used for diagnosis, so X-rays can be used to reduce exposure. In some cases, the field of illumination is narrowed down to 14 x 14 inches (large film size).
In this case, the size of the irradiation field is reduced from 14 × 17 inches to 14 × 14 inches by using the position information of the four sides of the diaphragm of the X-ray movable diaphragm device 3. ) Center coordinates may be changed according to the vertical size of the irradiation field.

The feature extraction unit 102 calculates, for example, the minimum value, the average value, and the maximum value of the image for the extracted image region.
For example, when the imaging region is the front of the chest, the minimum value and the maximum value are obtained from the entire image (region d in FIG. 4), and the maximum value of region a and the region b are obtained from the left and right lung fields (regions a and b). The maximum value is acquired, and the average value of each lung field (region a, b) is calculated to obtain the maximum value, and further the minimum value is acquired from the mediastinum (region c).
Thus, by using the maximum value of the average values of the left and right lung fields, it is possible to extract the average value of the lung fields more accurately even in the case of a single lung patient without one lung.

In this way, the feature extraction unit 102 determines that the maximum value of the image data is from the regions a, b, and d, the average value is the maximum value among the average values of the regions a and b, and the minimum value is the regions c and d. Is what you want.
The maximum value, average value, and minimum value are stored in the hard disk in advance (before imaging) because the position and size of the feature extraction area corresponding to the imaging region are known. When the button corresponding to the imaging region is pressed, the coordinates of the feature extraction region corresponding to the imaging region are automatically set in the memory, and each feature amount can be extracted from the designated region immediately after imaging. .

The normalized gradation conversion characteristic creation unit 103 uses the image feature amount extracted by the feature extraction unit 102, for example, the minimum value X1 and the maximum value X5 of the entire image, and linearly sets the pixel value as shown in FIG. The characteristic to be converted may be used.
In FIG. 5, the input x is 0 when the incident X-ray dose to the X-ray detector is 0 [μC / kg], corresponding to the value in which the pixel value increases in proportion to the incident X-ray dose. Yes.
The reason why the output is 0 when the input x is between 0 and x1 is that the output value is clipped to 0 because there is no pixel equal to or less than x1 when the minimum value of the image is x1. Alternatively, x1 may be set to 0 (fixed value) instead of the minimum value of the image.

The normalized feature value creation unit 104 converts the image feature value extracted by the feature extraction unit 102 using the normalized tone conversion characteristic created by the normalized tone conversion characteristic creation unit 103. For example, in the feature extraction unit 102, mediastinal minimum value X2 (mediastinal image feature amount), right lung field average value and left lung field average value X3 (right lung field and left lung field image) (Feature value) and the maximum value X4 of the right lung field and the left lung field (image feature quantity of the right lung field and the left lung field) are extracted, and the normalized gradation conversion characteristic is f (x), Normalized features are created by transforming with the functions of f (X2), f (X3), and f (X4), respectively.
By creating normalized features in this way, for example, when the imaging region is chest front imaging, in the captured image after gradation processing, the mediastinal minimum value, lung field average value, and lung field maximum value Can be converted into a target value to be described later.

The basic gradation conversion characteristic storage unit 105 stores a plurality of basic gradation conversion characteristics, and will be described by taking three types of basic gradation conversion characteristics shown in FIG. 6 as an example.
The basic gradation conversion characteristic storage unit 205 stores basic gradation conversion characteristics of g1 (x), g2 (x), and g3 (x) shown in FIG. The input x is data obtained by performing various corrections on the image data detected by the X-ray detector and converted into a digital value. The input data x is, for example, 14 bits, and the output value g corresponding to the input data x. Is 12 bits.

The basic gradation conversion characteristics are, for example, the function 1 of (Expression 3) represented by a straight line on the low X dose side (when the output value g is Y0 or more), and the high X dose side (the output value g is Y0 or less). (Case) may be expressed as a function 2 of (Equation 4) expressed in logarithm.
gi (x) = aL · x + bL gi (x): [Y0, 4095] (Formula 3)
gi (x) = aH · ln (x + bH) + cH gi (x): [0, Y0] (Formula 4)
Alternatively, in the basic gradation conversion characteristic, the function 1 on the low X-ray dose side can be expressed by a quadratic polynomial of (Expression 5), and the function 2 on the high X-ray dose side can be expressed by the logarithm of (Expression 6).
gi (x) = aL · x ² + bL · x + cL gi (x): [Y0, 4095] (Formula 5)
gi (x) = aH · ln (x + bH) + cH gi (x): [0, Y0] (Formula 6)
Alternatively, in the basic gradation conversion characteristic, the function 1 on the low X-ray dose side can be a cubic polynomial of (Expression 7), and the function 2 on the high X-ray dose side can be a cubic polynomial of (Expression 8).
gi (x) = aL · x ³ + bL · x ² + cL · x + dL gi (x): [Y0, 4095] (Formula 7)
gi (x) = aH · x 3 + bH · x 2 + cH · x + dH gi (x): [0, Y0] ( Eq. 8)
Alternatively, in the basic gradation conversion characteristic, the function 1 on the low X-ray dose side can be the logarithm of (Expression 9), and the function 2 on the high X-ray dose side can be the logarithm of (Expression 10).
gi (x) = aL·ln (x + bL) + cL gi (x): [Y0, 4095] (Equation 9)
gi (x) = aH · ln (x + bH) + cH gi (x): [0, Y0] (Equation 10)

In this way, the basic tone conversion characteristics can be expressed by combining straight lines, logarithms, and polynomials on the low X dose side and the high X dose side.For example, as shown in (Equation 3), the contrast on the low X dose side ( The contrast of the image area such as the heart, mediastinum, and bone areas is kept constant by keeping the basic gradient conversion characteristic slope (absolute value) constant. Can be.
If this is converted by increasing the slope of the low-dose region by logarithm or the like, the distribution of the histogram of the mediastinum becomes wide, and the vertebral body becomes conspicuous, which is not suitable for diagnosis of the front of the chest. In other words, in chest front imaging, it is important to extract the shadow of the lung field, so that tone conversion characteristics are required so that the histogram of the lung field becomes wider.

The value of Y0, which is the Y coordinate of the connection point between the two types of functions 1 and 2 of the basic gradation conversion characteristics shown in Fig. 6 above, is determined from the image quality evaluation of the doctor by experimenting with various gradation characteristics. Thus, for example, when the imaging region is the front of the chest, it has been found that if it is set to a value of 50% of the output value range (Y0 = 4095 × 0.50 = 2048).
The basic gradation conversion characteristic storage unit 105 is not limited to the above three basic gradation conversion characteristics, but can be adapted to various imaging parts and imaging methods (various imaging menus). A plurality of gradation conversion characteristics may be stored, and a basic gradation conversion characteristic suitable for the above-described shooting menu may be selected by a basic gradation conversion characteristic selection unit 107 described later.
This basic gradation conversion characteristic is determined in the factory through phantom photography and volunteer photography, and the slope is changed from 75% to 150%, for example, before delivery to the medical facility. A plurality of gradation conversion characteristics are created.
The created plurality of gradation characteristics are stored in the hard disk, and when an image processing application is activated in the image processing apparatus, the image is read from the hard disk and stored in the memory.

The target output value storage unit 106 stores a target output value corresponding to each image feature amount. For example, when the imaging region is the front of the chest, as shown in FIG. 7, the target output value Y2 corresponds to the mediastinal minimum value X2 (media feature image feature amount) obtained by the function f (X2) above. The target pixel value of the output image processed by the gradation processing unit 109, which will be described later, and the target output value Y3 is the average value of the right lung field and the average value of the left lung field obtained by the function f (X3). The target pixel value of the output image corresponding to the maximum value X3 (the image feature amount of the right lung field and the left lung field) processed by the gradation processing unit 109, which will be described later, is set, and the target output value Y4 is the function f (X4 ) For the maximum value X4 of the right lung field and the left lung field (image feature values of the right lung field and the left lung field) as the target pixel value of the output image processed by the gradation processing unit 109 described later. good.
The target output value is determined, for example, from a photographed image of a volunteer, after determining an appropriate gradation characteristic (e.g., slope 1.2), and then each feature amount (minimum mediastinal value, average value of lung field) of the photographed image of the volunteer The maximum value of the lung field) is calculated in advance and determined by obtaining the corresponding output value of the appropriate gradation characteristics, and this is stored in the hard disk.

The basic gradation conversion characteristic selection unit 107 outputs an output value of the basic gradation conversion characteristic stored in the basic gradation conversion characteristic storage unit 105 corresponding to the normalized image feature amount created by the normalized feature amount creation unit 104 And a target value output value stored in the target output value storage unit 106 corresponding to the normalized image feature amount, and FIG. 7 shows the relationship between the basic gradation conversion characteristics and the target output value. Show.
The target output value Y2 stored in advance in the target output value storage unit 106 for the normalized image feature values f (X2), f (X3), f (X4) created by the normalized feature value creation unit 104 , Y3, Y4 and the output value gi (f (X2)), gi () of the i-th basic gradation conversion characteristic gi (x) for the input x being f (X2), f (X3), f (X4) f (X3)), gi (f (X4)) and the sum of absolute values Ei is calculated by (Equation 11), and the sum of absolute values of errors Ei for all basic gradation conversion characteristics gi (x) Ask for.
Ei = | gi (f (X2))-Y2 | + | gi (f (X3))-Y3 | + | gi (f (X4))-Y4 | (Formula 11)
(Where i = 1, 2, 3)
Then, the c-th basic gradation conversion characteristic gc (x) that minimizes the sum total Ei of the absolute values of the errors is selected.
In the case of FIG. 7, c = 2, that is, the second basic gradation conversion characteristic is selected.

In this manner, the target output value stored in the target output value storage unit 106 is used as a comparison point (for example, a black point in FIG. 7) when selecting the basic gradation conversion characteristics.
Further, the target output value defines an approximate conversion value of the feature amount of the input image in the automatic gradation processing device of the present invention, and is selected with respect to the comparison point as shown in FIG. The basic gradation conversion characteristics to be used are not necessarily the same characteristics, but approximate the values of the comparison points under conditions constrained by the input / output relationship of the basic gradation conversion characteristics.
As a result, since the constraint condition is provided for the input / output relationship of the gradation characteristics using the basic gradation conversion characteristics, stable gradation characteristics can be created.

The gradation conversion characteristic synthesis unit 108 includes the normalized gradation conversion characteristic f (x) created by the normalized gradation conversion characteristic creation part 103 and the c-th basic gradation conversion selected by the gradation conversion characteristic selection part 107. The characteristic gc (x) is synthesized by (Equation 12), whereby the gradation conversion characteristic y (x) of the relationship between the input image data x and the output pixel value y shown in FIG. 8 can be obtained.
y (x) = gc (f (x)) (Formula 12)

  The gradation processing unit 109 performs gradation processing of the input image data using the gradation conversion characteristic y (x) synthesized by the gradation conversion characteristic synthesis unit 108, and outputs the gradation-converted image data .

  Thus, since gradation processing is performed on all pixels after the gradation characteristics are created, gradation processing of the input image can be automatically performed at high speed.

Next, the operation of the X-ray imaging apparatus using the automatic gradation processing apparatus will be described with reference to the flowchart of FIG.
(1) Shooting and image processing parameter setting (step S200)
Before imaging, the input device 10 is used to set imaging conditions, input an imaging region name, patient name, and various parameters necessary for imaging and image processing.
The hard disk (not shown) includes various correction data such as X-ray detector offset correction and gain correction necessary for the correction processing of the correction processing unit 8a of the image processing apparatus 8, and the automatic gradation processing unit 8b according to the present invention. Required for the tone conversion process, center coordinates and size of the feature extraction area of the input image according to the shooting menu in the shooting direction, multiple basic tone conversion characteristics, function 1 of these basic tone conversion characteristics The value of the connection point of function 2 and the target output value corresponding to the image feature amount in the feature extraction area are stored, and when the image processing application is activated, the data is read from the hard disk and stored in the primary storage memory To do. However, the value of the connection point between the plurality of basic gradation conversion characteristics and the function 1 and function 2 of these basic gradation conversion characteristics is stored in the basic gradation conversion characteristic storage unit 105 of the automatic gradation processing unit 8b. The target output value corresponding to the image feature amount in the region is stored in the target output value storage unit 106.

(2) Collection of photographed image data (Step S201)
In response to the imaging start signal from the input device 10, the X-ray imaging apparatus starts imaging, and the X-ray generator 2 irradiates the subject 1 with X-rays.
X-rays transmitted through the subject 1 are detected by an X-ray detector of the X-ray detection unit 5, converted into digital image data, and photographed image data of the subject 1 is collected.
When an X-ray flat panel detector is used as the X-ray detector, the X-ray flat panel detector includes an X-ray conversion unit that converts incident X-ray energy into light or a charge amount, and the X-ray conversion unit. A value obtained by converting the converted light into electric charge or a value obtained by directly converting the X-ray energy into electric charge is detected by a two-dimensional semiconductor detection element array. The two-dimensional intensity distribution of the line is accumulated as a charge, and this charge is read out and converted into a digital value by an analog / digital converter to obtain digital image data.

(3) Correction processing (Step S202)
The digital image data detected and collected in this way is input to the image processing device 8 and input by performing various corrections such as offset correction and gain correction of the X-ray detector in the correction processing unit 8a of the image processing device 8. Obtain image data.

(4) Automatic gradation processing (Step S203)
The input image data is gradation-converted by the automatic gradation processing unit 8b so that the corrected input image data is within a predetermined range even if the X-ray dose changes depending on the imaging menu (imaging site or imaging direction).

  1) Reading of gradation conversion processing parameters and extraction of image feature amount areas (steps S204 and S205) Stored in the primary storage memory (not shown) corresponding to the imaging menu set in the input device 10 and the imaging menu of the imaging direction Then, the center coordinates and the size of the feature extraction region of the input image are read out, and the region for extracting the image feature amount is extracted by the region extraction unit 101.

2) Image feature extraction (step 206)
A feature extraction unit 102 extracts an image feature amount from the region extracted by the region extraction unit 101.

3) Creation of normalized gradation conversion characteristics (step S207)
Using the image feature amount extracted by the feature extraction unit 102, a normalized gradation conversion characteristic creating unit 103 creates a normalized gradation conversion characteristic for the pixel value of the input image.

4) Creating normalized features (step 208)
Using the normalized gradation conversion characteristic created by the normalized gradation conversion characteristic creation unit 103, the image feature quantity extracted by the feature extraction unit 102 is created by the normalized feature quantity creation unit 104.

5) Reading basic gradation conversion characteristics and target output value (step S209)
A plurality of basic gradation conversion characteristics stored in the basic gradation conversion characteristic storage unit 105 and a target output value corresponding to the image feature amount extracted by the feature extraction unit 102 stored in the target output value storage unit 106 Are input to the basic gradation conversion characteristic selection unit 107.

6) Selection of basic gradation conversion characteristics (step 210)
The basic gradation conversion characteristic selection unit 107 compares the output value of the basic gradation conversion characteristic with the target output value corresponding to the image feature amount, and the basic gradation conversion characteristic that minimizes the error with respect to the target output value Select.

7) Synthesis of normalized gradation conversion characteristics and basic gradation conversion characteristics (step S211)
The normalized gradation conversion characteristic created by the normalized gradation conversion characteristic creation unit 103 and the basic gradation conversion characteristic selected by the basic gradation conversion characteristic selection unit 107 are synthesized by the gradation conversion characteristic synthesis unit 108. .

8) Tone processing of input image data (step 212)
The gradation conversion characteristic obtained by the gradation conversion characteristic combining unit 108 and the input image data are input to the gradation conversion processing unit 109, and the input image is converted to a predetermined pixel value by gradation conversion. Output image data is obtained.

(5) Image processing such as image enhancement processing (step 213)
The output image data that has undergone gradation conversion by the automatic gradation processing unit 8b is subjected to image processing such as dynamic range compression processing, filter processing, left-right reversal of the image, and trimming of the photographed image by the image processing unit 8c to be displayed. Is generated.

(6) Display image control processing and image display (steps S214 and S215)
A display image is generated by using the display lookup table, display window width, and display window level incidental information determined by the automatic display gradation processing on the image data processed by the display control processing unit 8d, and this display image Is displayed on the display device 9.

Figures 10 to 13 correspond to image features from the basic logarithmic conversion characteristics according to the present invention when the tone conversion is performed using the conventional logarithmic single gradation conversion characteristics in frontal chest photography. A comparison example between a histogram and a photographed image when the selected basic gradation conversion characteristic is selected and gradation conversion is performed with the gradation characteristic obtained by correcting the selected basic gradation conversion characteristic with the target output value is shown.
FIGS. 10 and 11 show comparisons of histograms, and FIGS. 12 and 13 show comparison examples of captured images.
Histogram conversion with the conventional logarithmic single tone characteristic shown in Fig. 10 converts the low X-ray dose region with high X-ray absorption in the mediastinum, heart, and bone regions. As a result, the distribution of the histogram of the mediastinum is widened, and the vertebral body is conspicuous as shown in FIG. 12, which is not suitable for diagnosis of the front of the chest.
On the other hand, the gradation when gradation conversion is performed with gradation characteristics according to the present invention is gradation-converted so that the histogram of the lung field, which is a low X-ray absorption region, is wide as shown in FIG. Therefore, as shown in FIG. 13, the contrast of the shadow of the lung field, which is most important for the chest front image, can be improved, and an image useful for diagnosis can be provided.

As described above, according to the X-ray imaging apparatus of the first embodiment of the present invention, the gradation characteristic uses not a single function using logarithm but a plurality of piecewise functions. The expansion of the image data on the low X-ray dose side and the compression of the image data on the high X-ray dose side can be automatically adjusted by the Y coordinate of the connection point. This can be improved over the conventional logarithm. Further, even when the histograms of the images after gradation conversion are compared, it can be seen that the lung field histogram is expanded as compared with the case where the conventional logarithm is used.
Therefore, in the case where the imaging region is the front of the chest, the image density and contrast of the lung field and mediastinum are appropriate, and the lung field that is particularly important for the chest front image should have a stereoscopic effect. Can do.

FIG. 14 is a block diagram of a second embodiment of an automatic gradation processing unit which is a main part of the X-ray imaging apparatus of the present invention.
In the automatic gradation processing unit 11 of the second embodiment, the region extraction unit 101 in the automatic gradation processing unit 8 of the first embodiment is changed to a region extraction unit 301 that can extract a feature region according to the imaging region. Similarly, the feature extraction unit 102 is changed to a feature extraction unit 302 that can extract an image feature amount in accordance with the imaging region, and the basic gradation conversion characteristic storage unit 105 includes a plurality of basic features corresponding to the imaging region. The basic gradation conversion characteristic storage unit 305 that stores gradation characteristics, and the target output value storage unit 106 are changed to a target output value storage unit 306 that stores a target output value corresponding to an image feature amount corresponding to an imaging region. Is.

For example, in the case of the front of the chest, the region extraction unit 301 sets the region to be extracted as shown in FIG. 3, and in the case of the front of the abdomen, for example, in the center of the image as shown in FIG. A circular area a may be set.
Note that the size of the feature extraction area may be changed according to the size of the irradiation field (width 14 × length 17 inches, width 14 × length 14 inches, etc.).

  The feature extraction unit 302 changes the image feature amount calculation method according to the imaging region. For example, if the imaging region is the front of the chest, the minimum value, maximum value, mediastinal minimum value, maximum value of the average value of the left and right lung fields, maximum value of the maximum values of the left and right lung fields, In the case of the front, the feature amount calculation method is changed such as the minimum value and maximum value of the entire image and the average value of the circular area at the center of the image.

  For example, as shown in FIG. 16, the basic gradation conversion characteristic storage unit 305 is configured such that the function 1 on the low dose side is a straight line, the function 2 on the high dose side is logarithmically, and the connection point Y between the function 1 and the function 2 is Y. The value of the coordinate (Y0) is changed according to the imaging region such as the skull of (a), the chest of (b), and the hip joint of (c), and these are created and stored in advance.

  The target output value storage unit 306 is changed according to the imaging region. For example, when the imaging region is the front of the abdomen, the average value of the circular area at the center of the image is 50% of the output range (for example, output If the maximum value of 4095 is 4095, 4095 × 0.5 = 2048).

  As described above, the region extraction unit 301, the feature extraction unit 302, the basic gradation conversion characteristic storage unit 305, and the target output value storage unit 306 are changed to those corresponding to the imaging region, and the first embodiment is used by using these. In the same way as the automatic gradation processing, the input image data is gradation-converted into a predetermined range of image data, and the converted image data is subjected to image processing and display control processing, whereby an appropriate density corresponding to the imaging region is obtained. Can be displayed on a display device or output to an imager.

1 is an overall configuration diagram of an X-ray imaging apparatus having an automatic gradation processing function according to a first embodiment of the present invention. 1 is a configuration block diagram of an image processing apparatus having an automatic gradation processing function according to a first embodiment of the present invention. The block diagram of the automatic gradation processing part which is the principal part of this invention. The figure which shows the feature extraction area | region of the chest front imaging | photography site | part. The figure which shows the example of the normalization gradation conversion characteristic produced in the normalization gradation conversion characteristic preparation part. The figure which shows the example of the some basic gradation conversion characteristic memorize | stored in the basic gradation conversion characteristic memory | storage part. FIG. 5 is an explanatory diagram for selecting an appropriate basic gradation conversion characteristic from a plurality of basic gradation conversion characteristics. The figure which shows the synthetic | combination function of the normalized gradation conversion characteristic and the selected basic gradation conversion characteristic. The figure which shows the flowchart explaining operation | movement of the X-ray imaging apparatus by this invention. The schematic diagram of the histogram of the chest image which carried out the gradation process using the gradation characteristic by the conventional logarithm. The schematic diagram of the histogram of the chest image which carried out the gradation process using the gradation characteristic of this invention. The figure which shows the chest image using the gradation characteristic by the conventional logarithm. The figure which shows the chest image using the gradation characteristic of this invention. The block diagram which shows 2nd Embodiment of the automatic gradation process part which is the principal part of the X-ray imaging apparatus of this invention. The figure which shows the feature extraction area | region of the abdominal front imaging | photography site | part. FIG. 10 is a diagram illustrating an example of basic gradation conversion characteristics corresponding to an imaging region stored in a basic gradation conversion characteristic storage unit according to the second embodiment of the present invention.

Explanation of symbols

1 Subject, 2 X-ray generator, 3 X-ray movable diaphragm, 5 X-ray detector, 7 X-ray high-voltage device, 8 Image processor, 8a Correction processor, 8b Automatic gradation processor, 8c Image processing Unit, 8d display control unit, 9 display device, 10 input device, 101 area extraction unit, 102 feature extraction unit, 103 normalized gradation conversion characteristic creation unit, 104 normalized feature value creation unit, 105 basic gradation conversion characteristic storage Unit, 106 target output value storage unit, 107 basic gradation conversion characteristic selection unit, 108 gradation conversion characteristic synthesis unit, 109 gradation processing unit, 301 area extraction unit, 302 feature extraction unit, 305 basic gradation conversion characteristic storage unit 306 Target output value storage

Claims (4)

  X-ray generation means for generating X-rays for irradiating the subject, X-ray detection means for detecting transmitted X-rays of the subject disposed opposite to the X-ray generation means, and digital detected by the X-ray detection means A gradation conversion unit that corrects the image data and converts the corrected input image data into output image data in a predetermined range, and the converted image data converted by the gradation conversion unit is subjected to image processing and desired. An X-ray imaging apparatus for generating an X-ray image of the image, wherein the gradation converting means includes a feature quantity extracting means for extracting an image feature quantity from the input image data, and a pixel value of the input image data as the image value. Normalized gradation conversion means that normalizes and converts gradations with feature amounts, normalized feature amount conversion means that converts the image feature amounts into normalized feature amounts by the normalized gradation conversion means, and a plurality of basic gradations Basic gradation change that memorizes conversion characteristics A characteristic storage means, a target output value storage means for storing a target output value corresponding to the image feature value, an output value of the basic gradation conversion characteristic and the target output value, and an error relative to the target output value A basic gradation conversion characteristic selecting means for selecting a basic gradation conversion characteristic that minimizes the gradation, and a gradation conversion characteristic created by combining the normalized gradation conversion characteristic and the selected basic gradation conversion characteristic. An X-ray imaging apparatus comprising tone conversion characteristic creating means and tone processing means for converting the input image data into a predetermined pixel value by the tone conversion characteristics.   The X-ray imaging apparatus according to claim 1, wherein the feature amount extraction unit includes a region extraction unit that extracts a region having an image feature amount from the input image data.   According to a second aspect of the present invention, the area extracting means includes means for setting an area corresponding to an imaging region, and means for extracting an image feature amount from the set area, and further imaging the basic gradation conversion characteristic storage means. An X-ray imaging apparatus characterized by storing a plurality of basic gradation conversion characteristics corresponding to a part, and storing a target output value corresponding to an image feature amount corresponding to the imaging part in the target output value storage means.   4. The X-ray imaging apparatus according to claim 2, wherein the image feature amount is a maximum value, an average value, or a minimum value of pixel values in the region extracted by the region extraction unit.

Images