JP2952422B2 - Radiation image analyzer - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radiation image analyzing apparatus for obtaining a reading condition, an image processing condition, or a subject part of a radiation image based on image data representing the radiation image. .

(Prior art) Reading a recorded radiation image to obtain image data,
After performing appropriate image processing on the image data, reproduction and recording of the image are performed in various fields. For example, an X-ray image is recorded using a low-gamma X-ray film designed to be compatible with later image processing, and the X-ray image is recorded.
An X-ray image is read from a film on which a line image is recorded, converted into an electric signal, the electric signal (image data) is subjected to image processing, and then reproduced as a visible image in a copy photograph or the like, so that contrast and sharpness are improved. It has been practiced to obtain a reproduced image having good image quality performance such as graininess and the like (see Japanese Patent Publication No. 61-5193).

In addition, the applicant (X-ray, α-ray, β-ray,
Irradiation with gamma rays, electron beams, ultraviolet rays, etc. accumulates a part of this radiation energy, and then irradiation with excitation light, such as visible light, causes a stimulable phosphor (luminous) to emit stimulated emission according to the accumulated energy. Radiation image information of a subject, such as a human body, is temporarily recorded on a sheet-like stimulable phosphor, and the stimulable phosphor sheet is scanned with excitation light such as laser light to radiate the stimulable phosphor. Generates luminescent light, obtains an image signal by photoelectrically reading the obtained stimulating luminescent light, and outputs a radiation image of the subject as a visible image to a recording material such as a photographic photosensitive material or a CRT based on the image data. There has already been proposed a radiation image recording / reproducing system (Japanese Patent Laid-Open No. 55-12429).
Nos. 56-11395, 55-163472, 56-104645
No. 55-116340, etc.).

This system has the practical advantage of being able to record images over a very large radiation exposure area compared to conventional radiographic systems using silver halide photography. That is, in the case of the stimulable phosphor, it has been recognized that the amount of emitted light that is stimulated by excitation after accumulation is proportional to the radiation exposure amount over an extremely wide range. Even if fluctuates considerably, the amount of the stimulating light emitted from the stimulable phosphor sheet is read by the photoelectric conversion means with the reading gain set to an appropriate value and converted into an electric signal. By outputting a radiation image as a visible image on a recording material such as a photographic light-sensitive material or a display device such as a CRT using, a radiation image which is not affected by a change in radiation exposure can be obtained.

In the above-described system, before obtaining an image signal by reading under optimal reading conditions according to the dose of radiation applied to the stimulable phosphor sheet, the stimulable phosphor sheet is scanned with a low-level light beam in advance. A pre-read for reading the outline of the radiation image recorded on the sheet is performed, a pre-read image signal obtained by the pre-read is analyzed, and then the sheet is irradiated with a high-level light beam and scanned, and the radiation image is scanned. There is also a system configured to perform a main reading in which an image signal is obtained by reading under an optimum reading condition.

The reading condition is a general term for various conditions that affect the relationship between the amount of stimulated emission light in reading and the output of the reading device. For example, a reading gain, a scale factor, or a scale factor that determines an input / output relationship. , And the power of the excitation light in reading.

Further, the high level / low level of the light beam means that the energy of the light beam applied per unit area of the sheet is large / small or the energy of the stimulated emission light emitted from the sheet is the light beam. (Having a wavelength sensitivity distribution) means the magnitude of the weighted energy after weighting the energy of the light beam irradiated per unit area of the sheet by the wavelength sensitivity. Methods for changing the beam level include using a light beam of a different wavelength, changing the intensity of the light beam emitted from a laser light source, etc., and inserting or removing an ND filter or the like in the optical path of the light beam. How to change the beam intensity, how to change the scanning density by changing the beam diameter of the light beam, how to change the scanning speed, etc.
Various known methods can be used.

In addition, regardless of whether the system performs the pre-reading or the system which does not perform the pre-reading, the obtained image signal (including the pre-reading image signal) is analyzed, and an optimum image processing condition when performing image processing on the image signal is determined. Some systems let you decide. Here, the image processing condition is a general term for various conditions when a process that affects the gradation and sensitivity of a reproduced image based on an image signal is performed on the image signal. The method for determining the optimum image processing conditions based on this image signal is not limited to a system using a stimulable phosphor sheet, and obtains an image signal from a radiation image recorded on a recording sheet such as a conventional X-ray film. It is also applied to the system.

The calculation for obtaining the reading condition and / or the image processing condition (hereinafter, referred to as a reading condition or the like) based on the image signal (including the pre-read image signal) is performed based on a result of statistically processing a large number of radiation images in advance. The algorithm is defined (for example, Japanese Patent Application Laid-Open No. 60-185944,
-280163).

This conventionally employed algorithm generally obtains a histogram of an image signal, and calculates various values such as the maximum value and the minimum value of the image signal on the histogram, the value of the image signal at the point where the appearance frequency of the image signal is the maximum, and the like. Are obtained, and reading conditions and the like are obtained based on the characteristic points.

However, in recent years, a concept of a neural network completely different from the above algorithm has appeared, and is being applied to various fields.

This neural network inputs information (teacher signal) as to whether an output signal output when a certain input signal is given is a correct signal or an incorrect signal, and thereby, a signal between each unit in the neural network is input. It is equipped with an error reverse propagation learning (back propagation) function of correcting the weight of the connection (weight of the synaptic connection), and by repeatedly 'learning'
It is possible to increase the probability of outputting a correct answer when a new signal is input. (For example, "DERumelh
art, GEHinton and RJWilliams: Learning represent
ations by back-propagating errors, Nature, 323-9,5
33-356, 1986a "," Hideki Aso: Back Propagation Computrol NO.24 53-60 "," Kazuyuki Aihara, Neural Computer, Tokyo Denki University Press ").

This neural network can be applied to determination of reading conditions and the like, and reading conditions and the like can be output by inputting image signals and the like to the neural network.

In addition, if a neural network is used, not only the above reading conditions and image processing conditions, but also, for example, an object part (for example, head, neck, chest, abdomen, etc., and a direction (front, side, )) Can be recognized, and the reading conditions and the image processing conditions can be obtained with reference to the subject portion.

(Problems to be Solved by the Invention) When the neural network is used to determine the reading condition, image processing condition, or subject part of the radiation image, the correct reading condition, image processing condition, or subject is gradually obtained by repeatedly 'learning' in advance. The part can be determined, but the radiographic image
Usually, it is composed of a large number of image data respectively corresponding to a very large number of pixels, and when these image data are input to the neural network as they are, a neural network having a very large number of input points and therefore a very large number of units It is necessary to use a network, and the storage capacity of the storage device for storing the weight of the connection between the units needs to be enormous. Also, in this case, it is necessary to repeat the learning extremely many times when performing the 'learning'.

The present invention has been made in view of the above circumstances, and has as its object to provide a radiation image analysis apparatus that can determine a correct reading condition, image processing condition, or subject portion using a neural network having a relatively small number of units. It is.

(Means for Solving the Problems) A radiation image analyzing apparatus according to the present invention includes a radiation field for recognizing the radiation field based on a large number of image data corresponding to each pixel of a radiation image partially including the radiation field. A recognition unit; and a neural network that receives all or a part of the image data corresponding to each pixel in the recognized irradiation field as input, and outputs a reading condition, an image processing condition, or a subject part of the radiation image. Volume calculation means.

Here, the algorithm for recognizing the irradiation field in the irradiation field recognizing means is not limited to a specific algorithm, and the difference between the average values of the image data inside and outside the irradiation field,
Any algorithm that recognizes the irradiation field based on any of the differences in the degree of variation of the image data inside and outside the irradiation field, the state of the change of the image data near the contour of the irradiation field, or a combination thereof (For example, JP-A-61-39039 and JP-A-63-259).
No. 538, JP-A-1-42436, JP-A-2-67690
Reference).

Further, the reading condition of the radiation image is a reading condition at the time of reading the radiation image, for example, a reading condition determined to be read under the optimal condition when reading the radiation image in the main reading, The image processing condition of a radiation image is an image processing condition when image processing is performed on image data representing the radiation image, and the subject part of the radiation image is an image of any part of the subject in the radiation image. This is a part of the subject in a certain case, for example, a part of the subject that is referred to when the above-described reading conditions and image processing conditions are determined.

(Operation) The radiation image analysis apparatus of the present invention recognizes an irradiation field and inputs only image data corresponding to each pixel in the irradiation field to the neural network. Therefore, the number of input points and the number of units of the neural network are reduced. The storage capacity of the storage device for storing the coefficient representing the weight of the connection between the units can be reduced, and an accurate output can be obtained with a relatively short 'learning'. Becomes

Instead of inputting all the image data of each pixel in the irradiation field to the neural network, for example, taking out every other pixel in the irradiation field and inputting the taken out image data corresponds to each pixel in the irradiation field. If only a part of the image data is input to the neural network, the storage capacity can be further reduced.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. Here, an example in which the above-described stimulable phosphor sheet is used will be described.

FIG. 6 is a perspective view showing an example of an X-ray image reading apparatus and an example of a computer system including an example of a radiation image analyzer of the present invention. This system uses the above-described stimulable phosphor sheet to perform pre-reading.

The stimulable phosphor sheet 11 on which an X-ray image is recorded, obtained by X-ray imaging using an X-ray imaging device (not shown),
First, scanning is performed with a weak light beam, and only a part of the radiation energy stored in the sheet 11 is emitted. The stimulable phosphor sheet 11 set at the predetermined position is provided with a sheet conveying means 13 such as an endless belt driven by a motor 12.
Is conveyed (sub-scan) in the direction of arrow Y. On the other hand, the weak light beam 15 emitted from the laser light source 14
Is reflected and deflected by a rotating polygon mirror 16 which rotates at a high speed in the direction of the arrow and passes through a converging lens 17 such as an fθ lens. Main scanning in the direction of arrow X substantially perpendicular to Sheet 11 irradiated with this light beam 15
From the location, the amount of stimulated emission light 19 corresponding to the amount of radiation image information stored and recorded is diverged.
Is guided by a light guide 20 and is photoelectrically detected by a photomultiplier (photomultiplier tube) 21. The light guide 20 is formed by molding a light-guiding material such as an acrylic plate, and is arranged such that a linear incident end face 20a extends along a main scanning line on the stimulable phosphor sheet 11. The light receiving surface of the photomultiplier 21 is connected to the emission end face 20b formed in an annular shape. Above end face
The stimulated emission light 19 entering the light guide 20 from 20a travels inside the light guide 29 by repeating the pre-reflection, and the emission end face 20
The light emitted from b and received by the photomultiplier 21 is converted into an electric signal by the photomultiplier 21 by the amount of photostimulated light 19 representing a radiation image.

The analog output signal S output from the photomultiplier 21 is logarithmically amplified by a logarithmic amplifier 26, and the A / D converter 2
Digitization is performed in step 7 to obtain a pre-read image signal Sp. The signal level of the pre-read image signal Sp is proportional to the logarithm of the amount of stimulated emission light emitted from each pixel of the sheet 11.

In the pre-reading, the reading conditions, that is, the voltage value applied to the photomultiplier 21 and the amplification factor of the logarithmic amplifier 26 are determined so that the radiation energy stored in the stimulable phosphor sheet 11 can be read over a wide area. Have been.

The obtained look-ahead image signal Sp is transmitted to the computer system 40.
Is input to The computer system 40 includes an example of the radiation image analyzer of the present invention, and includes a CP.
U and a main unit 41 containing an internal memory, a drive unit 42 into which a floppy disk serving as an auxiliary memory is inserted and driven, and an operator
And a CRT display 44 for displaying necessary information.

In the computer system 40, the irradiation field is recognized as described later based on the input pre-read image signal Sp, and then the reading conditions at the time of main reading, that is, the sensitivity and contrast at the time of main reading, are obtained. For example, the voltage value applied to the photomultiplier 21 'and the amplification factor of the logarithmic amplifier 26' are controlled in accordance with the obtained sensitivity and contrast. In this embodiment, the reading condition at the time of the main reading is considered as the reading condition according to the present invention.

Here, the contrast corresponds to the light amount ratio of the strongest stimulated emission light to the weakest stimulated emission light which is converted into an image signal at the time of main reading, and the sensitivity is defined as a predetermined amount of the stimulated emission light. This refers to a photoelectric conversion rate that determines which level of an image signal is the emitted light.

The stimulable phosphor sheet 11 'for which pre-reading has been completed is set at a predetermined position of the main reading means 100', and the sheet 11 'is scanned by a light beam 15' which is stronger than the light beam used for pre-reading, as described above. An image signal is obtained according to the read conditions determined in advance, but since the configuration of the main reading unit 100 ′ is substantially the same as the configuration of the pre-reading unit 100, components corresponding to the components of the pre-reading unit 100 are used. Indicates the number used in the look-ahead means 100 with a dash appended thereto, and a description thereof will be omitted.

Image signal obtained by being digitized by A / D converter 27 'S _Q is input again to the computer system 40. Appropriate image processing on the image signal S _Q is a computer system 40 within are subjected, the image signal subjected to the image processing is sent to a not-shown reproducing apparatus, X-rays image is reproduced and displayed based on the image signal in the playback device Is done.

Next, a method in which the computer system 40 first recognizes the irradiation field based on the pre-read image signal Sp, and then obtains the reading conditions for the main reading will be described.

FIG. 1 is a diagram showing an example of an X-ray image, a pre-read image signal Sp obtained from the X-ray image, and a differential value ΔSp thereof for explaining an embodiment of the recognition means of the present invention. is there.

On the stimulable phosphor sheet 11, a subject image 3 of which the subject is the head of a human body in the irradiation field 2 is recorded. Here, the stimulable phosphor sheet 11 is used as a predetermined point in the irradiation field 2.
Is selected, and along each of the plurality of line segments 5 extending radially from the center C, a differential operation is performed on the pre-read image signal Sp corresponding to each pixel on each line segment, and the pre-read image signal The point where the value of Sp suddenly decreases is obtained as a contour point located on the contour of the irradiation field.

Hereinafter, the case where the contour point on the line segment along the ξ axis among the plurality of line segments 5 will be described.

Graph A is a graph showing the value of the pre-read image signal Sp obtained from each pixel along the ξ axis.

A pre-read image signal of the direct X-ray portion 6 in which X-rays other than the subject image 3 in the irradiation field 2 are directly irradiated on the stimulable phosphor sheet 11
The value of Sp is the highest, and the value of the pre-read image signal Sp sharply drops at the contour of the irradiation field 2.

The graph B is a graph obtained by differentiating the pre-read image signal Sp shown in the graph A from the center C in the negative direction of ξ (left direction in the figure) and the positive direction of ξ (right direction in the figure).

On a line segment extending in the negative direction of the ξ axis from the center C in the graph B, the major peak projecting downward is a _1, therefore the position of the peak a ₁ is defined as a contour point.

In the graph B, there is a peak a ₂ projecting downward on a line segment extending from the center C in the positive direction of the ξ axis.
position of a ₂ is defined as a contour point.

As described above, the contour point 7 is obtained for each of the plurality of line segments 5 connecting the center C and the end of the stimulable phosphor sheet 11. After these contour points 7 are determined, if a line along these contour points 7 is determined, the line becomes the contour of the irradiation field. A line along the contour point 7 is obtained by, for example, a method of connecting the remaining points after smoothing the points, a method of locally obtaining a plurality of straight lines by applying the least squares method, and a method of connecting them. Although it can be obtained by a method of applying a spline curve or the like, in the present embodiment, a plurality of straight lines along the contour point 7 are obtained using Hough transform. Hereinafter, a process for obtaining this straight line will be described.

When the x-axis and the y-axis are defined as shown in the drawing with one end (lower left end of the drawing) of the stimulable phosphor sheet 11 shown in FIG. 1 as the origin, the coordinates of each contour point are (x ₁ , y ₁ ), (x ₂ , y ₂ ), ……,
It is obtained as (x _n , y _n ). Here, these coordinates are represented as coordinates (x ₀ , Y ₀ ). Here, assuming that the coordinates of the contour points are (x ₀ , y ₀ ), a curve represented by ρ = x ₀ cos θ + y ₀ sin θ (1) is obtained by using these x ₀ , y ₀ as constants and all the contour point coordinates. (X ₀ , y ₀ ). This curve is as shown in FIG. 2, and there are as many contour point coordinates (x ₀ , y ₀ ) as there are.

Next, an intersection (ρ ₀ , θ ₀ ) at which the predetermined number Q or more of the plurality of curves intersect is determined. It is rare that many curves intersect exactly at one point due to an error of the contour point coordinates (x ₀ , y ₀ ), etc., and therefore, in practice, for example, the intersections of two curves are separated from each other by an interval smaller than a predetermined value. When they exist, the center of the group of intersections is defined as the intersection (ρ ₀ , θ ₀ ). Next, the intersection (ρ _0, θ ₀₎ is a straight line defined by the following equation _{ρ 0 = x cosθ 0 + y} sinθ 0 ... (2) required in the x-y orthogonal coordinate system. This straight line is a straight line extending along a plurality of contour point coordinates (x ₀ , y ₀ ). When the contour points 7 are arranged as shown in FIG. _1, the straight lines L _{1 to} L ₅ are obtained by extending the line segments forming the contour of the irradiation field 2 (see FIG. 1) as shown in FIG. Is required. next,
Thus a plurality of straight lines L ₁ determined, L _2, L _3, ... L is the region surrounded by the _n sought, this area is recognized as the irradiation field 2. This area is recognized in detail as follows, for example. In the computer system 40, line segments M ₁ , M ₂ , M ₃ ,... M connecting the corners of the stimulable phosphor sheet 11 and the center C are provided.
_m (four if the stimulable phosphor sheet 11 is rectangular) is stored, and the presence or absence of an intersection between each of the line segments M _{1 to} M _m and each of the straight lines L _{1 to} L _n is checked. When this intersection exists, the plane including the sheet corner is cut out of the plane bisected by the straight line. This operation is performed for all straight lines L ₁
~ L _n , by performing on the line segments M ₁ -M _m ,
Region surrounded by L ₁ ~L _n is left. This remaining area is the irradiation field 2 (see FIG. 1).

When the irradiation field 2 is obtained in this manner, the pre-read image signal Sp corresponding to the irradiation field 2 is input to the neural network, and the image signal in the irradiation field 2 is read under appropriate reading conditions at the time of main reading. The reading conditions are determined as described above.

FIG. 4 is a view schematically showing each pixel of a part of the X-ray image in the irradiation field. Each square represents each pixel.

After the irradiation field 2 is obtained, in this embodiment, only the pre-read image signal corresponding to each pixel hatched in FIG. 4 is extracted from the pre-read image signals corresponding to each pixel in the irradiation field 2. Input to the neural network. In the present invention, thinning out the pre-read image signal in the irradiation field 2 is not an essential requirement, but by thinning out the pre-read image signal Sp in the irradiation field 2 and inputting it to the neural network, the neural network Can be further reduced.

Note that, even when the pre-read image signal Sp in the irradiation field 2 is thinned out, in addition to uniformly thinning out the irradiation field 2 as shown in FIG. In consideration of the fact that the main part of the irradiation field 2 is often present, the central part of the irradiation field 2 may be thinned and the end part may be thinned out.

Figure 5 shows backpropagation learning (back propagation)
FIG. 2 is a diagram illustrating an example of a neural network having a function. Error Back Propagation Learning (Back Propagation)
As described above, the “learning” algorithm that sequentially corrects connection weights (synapse connection weights) from the output side to the input side by comparing the output of the neural network with the correct answer (teacher signal), as described above. Say.

As shown in the figure, the first of this neural network
Layer (input layer), second layer (intermediate layer), a third layer (output layer) is _one n respectively, _two n, consists of two units. The signals F ₁ , F ₂ ,..., F _n1 input to the first layer (input layer) are X
Among the pixels in the line image irradiation field 2, as shown in FIG. 4, the pre-read image signals Sp corresponding to the thinned numbers of the respective pixels, and two outputs from the third layer (output layer). Are signals corresponding to the sensitivity and the contrast at the time of main reading.

I-th unit in k-th layer The unit Each output The weight of the connection to Are the same characteristic function Shall be provided. At this time, each unit of Becomes However, each unit constituting the input layer (I = 1,2, ..., n 1) the input F ₁ to, F _2, ..., F _n1 intact units without being weighted (I = 1, 2,..., N ₁ ). The input n ₁ signals F ₁ , F ₂ ,..., F _n1 are weights of each combination. Final output while weighted by The reading conditions (sensitivity and contrast) for the main reading are obtained.

Here, the weight of each connection Will be described. First, weight of each connection by random numbers Is given. In this case, even if the input F ₁ to F _n1 is varied to the maximum, the output y Is preferably limited to a value within a predetermined range or a value close to the predetermined range.

The pre-read image signal Sp is obtained by reading as described above from the stimulable phosphor sheet on which a number of X-ray images of which the optimum reading conditions are known are recorded, and is thinned out as shown in FIG. The above n ₁ inputs F ₁ , F ₂ ,..., F _n1 are obtained. This n ₁
Inputs _{F 1, F 2, ...,} F n1 is input to the neural network shown in FIG. 5, the units Output Is monitored.

Each output Is the final output And the teacher signal (sensitivity) as the correct reading condition for this image And contrast Square error with Is required. In order to minimize these square errors E ₁ and E ₂ , the weight of each connection is calculated as follows. Is corrected. The following About the output of Since this is the same as above, it is omitted here.

To minimize the square error E ₁ , this E ₁ Is a function of Weight of each connection as Is corrected. Here, η is a coefficient called a learning coefficient.

here, And from equation (3) Therefore, equation (9) is Becomes

Here, from equation (6), By transforming equation (11) using equation (5), Here, from equation (3), f ′ (x) = f (x) (1−f (X)) (13) Becomes

In equation (10), we set k = 2 and substitute equations (12) and (14) into equation (10). Substituting equation (15) into equation (8), Becomes According to this equation (16), (I = 1, 2,..., N ₁ ) The weight of each connection is modified.

next, Therefore, substituting equations (4) and (5) into equation (17), Here, from equation (13), Therefore, this equation (19) and equations (12) and (14) are
8) Substituting into the formula, In equation (10), we set k = 1 and substitute equation (20) into equation (10). By substituting this equation (21) into the equation (8), K = 1 and And was corrected by equation (16) (I = 1,2, ..., n ₁ ) is substituted into this equation (22), (I = 1, 2,..., N ₁ ; j = 1, 2,..., N ₂ ) are modified.

Theoretically, by using the equations (16) and (22), the learning coefficient η is made sufficiently small and the number of times of learning is made sufficiently large, so that the weight of each connection is obtained. Can be converged to a predetermined value, but making the learning coefficient η too small is not realistic because the progress of learning is slowed down. On the other hand, if the learning coefficient η is increased, the learning may oscillate (the weight of the connection does not converge to a predetermined value). Therefore, in practice, vibration is suppressed by adding an inertia term such as the following equation to the correction amount of the weight of the connection, and the learning coefficient η is set to a relatively large value.

(E.g., DERumelhart, GEHinton and RJWilliam
s: Learning internal representations by error propa
gation In Parallel Distributed Processing, Volume
1, JLMcClelland, DERumelhart and The PDP Researc
h Group, MIT Press, 1986b ”) Modified connection weight in t-th learning From the weight of the connection before correction Represents the correction amount obtained by subtracting. Α is a coefficient called an inertia term.

As the inertia term α and the learning coefficient η, for example, α = 0.9η = 0.2
Weight of each connection using 5 Is modified (learned), for example, 200,000 times. Is fixed to the final value. At the end of this learning, two outputs Are signals that correctly represent the sensitivity and contrast, respectively, at the time of main reading.

Therefore, after the learning is completed, a pre-read image signal Sp representing an X-ray image whose reading conditions at the time of main reading is unknown is obtained, and the irradiation field of the X-ray image is recognized based on the front image signal Sp. Then, the pre-read image signal Sp of the irradiation field is input to the neural network shown in FIG. 5, and the output obtained thereby is obtained. Is a signal representing the reading conditions (sensitivity and contrast) of the main reading for the X-ray image. Since this signal is obtained after the learning as described above, the signal accurately represents the reading condition at the time of the main reading.

The neural network is not limited to a three-layer structure, but may be a multilayer structure. Also, as for the number of units of each layer, each layer can of course be constituted by an arbitrary number of units according to the number of pixels of the input pre-read image signal Sp and the accuracy of required reading conditions.

As described above, the voltage applied to the photomultiplier 21 'of the main reading means 100' and the amplification factor of the amplifier 26 'are controlled in accordance with the reading conditions obtained by the neural network, and the main reading is performed in accordance with the controlled conditions. It is.

In the above embodiment, the pre-reading means 100 and the main reading means 100 'are separately configured.
0 and the structure of the main reading means 100 'are substantially the same,
The unit 100 and the main reading unit 100 'may be integrally used. In this case, after pre-reading, the stimulable phosphor sheet 11 may be backed once and then scanned again to perform main reading.

When the pre-reading means and the main reading means are used together, it is necessary to switch the intensity of the light beam between the case of pre-reading and the case of main reading. As described above, as a method of this switching, the light from the laser light source is used. Various methods such as a method of switching the intensity itself can be used.

Further, in the above embodiment, the apparatus for obtaining the reading conditions at the time of the main reading in the computer system 40 has been described.
At the time of actual reading, reading is performed under predetermined reading conditions regardless of the pre-read image signal Sp.
Pre-reading on the basis of the image signal Sp, may be calculated image processing conditions for performing image processing on the image signal S _Q, also,
The computer system 40 may obtain both the reading conditions and the image processing conditions.

Further, in the above-described embodiment, the radiation image reading apparatus that performs the pre-reading has been described. However, the present invention can be applied to a radiation image reading apparatus that performs the reading corresponding to the above-mentioned main reading without performing the pre-reading. In this case, when reading, an image signal is obtained by reading under predetermined reading conditions, and a computer system is
Image processing conditions are obtained in 40, and image processing is performed on the image signal according to the obtained image processing conditions.

Further, the present invention can be used not only for a system using a stimulable phosphor sheet but also for an apparatus using a conventional X-ray film.

Further, the present invention is not limited to the method of obtaining the reading condition and the image processing condition, but may be the method of obtaining the subject part by the computer system 40 using a neural network.

(Effects of the Invention) As described in detail above, the radiation image analyzing apparatus of the present invention inputs an irradiation field recognition unit and all or a part of image data corresponding to each pixel in the irradiation field, and Since the apparatus has feature amount calculating means including a neural network that outputs reading conditions, image processing conditions, or a subject portion, the number of input points of the neural network is smaller than when all image data corresponding to the entire radiation image is input to the neural network. Can be reduced, the storage capacity of the storage device for storing the weight of the connection between the units can be reduced, and the number of times of learning can be reduced.

In particular, in a radiographic image, the method of narrowing the irradiation field varies depending on the facility, the site, and the like, and when the entire image data is input to the neural network as in the related art, the various methods of narrowing the irradiation field are supported. Although it is necessary to train the image data, in the present invention, since only the image data of the pixels corresponding to the pixels in the irradiation field are input as described above, the image data corresponding to such various ways of narrowing the irradiation field is input. There is no need for learning, and the number of times of learning can be greatly reduced from this point.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of an X-ray image and a graph of a pre-read image signal obtained from the X-ray image and a differential value thereof. FIG. 2 shows a method for obtaining a straight line along a contour point. FIG. 3 is an explanatory diagram for explaining a method for extracting a region surrounded by a straight line along a contour point. FIG. 4 is a diagram showing each pixel of a part of an X-ray image in an irradiation field. FIG. 5 schematically illustrates an example of a neural network, FIG. 6 illustrates an example of an X-ray image reading apparatus, and an example of a computer system including an example of the present invention. It is a perspective view. 2 ... irradiation field, 3 ... subject part 6 ... direct X-ray part 11,11 '... stimulable phosphor sheet 19,19' ... stimulated luminescence 21,21 '... photomultiplier 26, 26 '... logarithmic amplifier 27, 27' ... A / D converter 40 ... computer system 100 ... pre-reading means, 100 '... main reading means

Claims (1)

(57) [Claims] An irradiation field recognizing means for recognizing the irradiation field based on a large number of image data corresponding to each pixel of a radiation image partially including the irradiation field of radiation, and each pixel in the recognized irradiation field And a feature amount calculating means comprising a neural network that receives all or a part of the image data corresponding to the input, and outputs a reading condition of the radiation image, an image processing condition, or an object part. Image analysis device.