JP2021194388A - Radiographic image capturing system, program, and image processing method - Google Patents

Abstract

To omit or simplify a user's setting operation for applying proper image processing to a radiographic image.SOLUTION: A radiographic image capturing system according to this invention comprises: a radiation detection device which detects radiation emitted from a radiation source and having passed through a subject and captures a radiation image; an optical camera for capturing an optical image of at least a partial area of the subject; and a console which applies image processing to the radiation image captured by the radiation detection device. The console sets an image processing condition of the radiation image on the basis of the optical image captured by the optical camera.SELECTED DRAWING: Figure 3

Description

The present invention relates to a radiographic imaging system, a program and an image processing method.

Conventionally, a technique for setting positioning and imaging conditions at the time of radiography using an optical image has been proposed.

For example, Patent Document 1 describes a technique for starting shooting when a shooting portion of a subject is reflected in the outer frame of a camera image of a webcam.
Further, Patent Document 2 describes a technique of discriminating a cassette to be used for radiography from a camera image and automatically setting the cassette to be used on the console.
Further, in Patent Document 3, it is determined whether or not a predetermined change has occurred in the exposed area based on an optical image obtained by capturing at least a part of the exposed area irradiated with radiation, and the predetermined change is obtained. It is described that the radiation generation is stopped when the radiation occurs.

Japanese Unexamined Patent Publication No. 2012-24399 Japanese Unexamined Patent Publication No. 2019-33828 Japanese Unexamined Patent Publication No. 2008-206740

The techniques of Patent Documents 1 to 3 can simplify the work related to radiography and perform appropriate radiography. However, in order to perform appropriate image processing on the captured radiographic image, it is necessary for the user to set information about the subject, processing parameters, and the like on the console or the like.

An object of the present invention is to omit or simplify a user's setting operation for performing appropriate image processing on a radiographic image.

In order to solve the above problems, the radiographic imaging system of the invention according to claim 1 is used.
A radiation image imaging unit that detects radiation emitted from a radiation source and has passed through a subject and captures a radiation image.
An optical image imaging unit that captures an optical image of at least a part of the subject.
An image processing unit that performs image processing on a radiation image captured by the radiation image imaging unit is provided.
The image processing unit sets image processing conditions for the radiographic image based on the optical image captured by the optical image imaging unit.

The invention according to claim 2 is the invention according to claim 1.
The optical image capturing unit captures an optical image of at least a part of the region of the subject to which the radiation from the radiation source is irradiated.
The image processing unit determines a portion of the subject to be imaged in the radiographic image by analyzing the optical image, and sets the image processing condition according to the determined portion.

The invention according to claim 3 is the invention according to claim 1.
The image processing unit determines the physique of the subject by analyzing the optical image, and sets the image processing conditions according to the determined physique of the subject.

The invention according to claim 4 is the invention according to claim 1.
The image processing unit calculates the feature amount of the subject by analyzing the optical image, and sets the image processing condition according to the calculated feature amount.

The invention according to claim 5 is the invention according to claim 4.
The feature amount is a feature amount representing the size or shape of a predetermined structure of the subject.

The invention according to claim 6 is the invention according to any one of claims 1 to 5.
The image processing conditions include dynamic range compression processing, contrast correction processing, density correction processing, LUT conversion processing, frequency enhancement processing, scattered ray correction processing, noise suppression processing, image trimming, image masking, image rotation, and image processing. It is at least one image processing condition of inversion.

The program of the invention according to claim 7 is a computer that detects radiation emitted from a radiation source and transmitted through a subject, and performs image processing on a radiation image captured by a radiation image imaging unit that captures a radiation image.
A setting unit that sets image processing conditions for a radiation image based on an optical image captured by an optical image imaging unit that captures an optical image of at least a part of the subject.
To function as.

The image processing method of the invention according to claim 8 includes a step of detecting radiation emitted from a radiation source and transmitted through a subject, and an image of a radiation image.
The step of capturing an optical image of at least a part of the subject and
A step of setting image processing conditions for the radiographic image based on the captured optical image, and
A step of performing image processing on the radiographic image based on the set image processing conditions, and
including.

According to the present invention, it is possible to omit or simplify the setting operation of the user for performing appropriate image processing on the radiographic image.

It is a figure which shows the whole structure of the radiation imaging system in embodiment of this invention. It is a block diagram which shows the functional structure of the console of FIG. It is a flowchart which shows the flow of the radiation image generation processing A executed by the control part of FIG. 2 in 1st Embodiment. It is a flowchart which shows the flow of the radiation image generation processing B executed by the control unit of FIG. 2 in the 2nd Embodiment. It is a flowchart which shows the flow of the radiation image generation processing C executed by the control unit of FIG. 2 in the 3rd Embodiment. It is a flowchart which shows the flow of the 1st shooting process A executed by the control part of FIG. 2 in 4th Embodiment. It is a flowchart which shows the flow of the 2nd shooting process A executed by the control part of FIG. 2 in 4th Embodiment. It is a flowchart which shows the flow of the 1st shooting process B executed by the control part of FIG. 2 in 5th Embodiment. It is a flowchart which shows the flow of the 2nd shooting process B executed by the control part of FIG. 2 in 5th Embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the scope of the invention is not limited to the illustrated examples.

<First Embodiment>
[Structure of Radiation Imaging System 100]
First, the configuration of the first embodiment of the present invention will be described.
FIG. 1 is a diagram showing an overall configuration example of the radiation imaging system 100 according to the present embodiment. The radiographic image capturing system 100 is a system that performs radiographic imaging on the subject H and performs image processing on the obtained radiographic image.

As shown in FIG. 1, the radiation imaging system 100 includes a radiation irradiation device 1, a radiation detection device 2, a console 3, and an optical camera 4. The console 3 is configured to be connected to a radiation irradiation device 1, a radiation detection device 2, and an optical camera 4 so as to be able to transmit and receive data.

The radiation irradiation device 1 includes a radiation source 11 arranged at a position facing the radiation detection device 2 with the subject H (subject) in between, and emits radiation (X-rays) to the subject H according to the control of the console 3. Irradiate.

The radiation detection device 2 is a radiation image imaging unit that detects radiation emitted from the radiation source 11 and transmitted through the subject H and captures a radiation image, and is configured to include a detector holding unit 22, a radiation detector P, and the like. ing. The radiation detector P is composed of an FPD (Flat Panel Detector) or the like. The radiation detector P has, for example, a glass substrate or the like, and detects radiation (X-rays) irradiated from the radiation irradiating device 1 and transmitted through at least the subject H at a predetermined position on the substrate according to its intensity. However, a plurality of detection elements (pixels) that convert the detected radiation into an electric signal and store it are arranged in a matrix. Each pixel is configured to include, for example, a switching unit such as a TFT (Thin Film Transistor). The radiation detector P controls the switching unit of each pixel based on the image reading condition input from the console 3, switches the reading of the electric signal stored in each pixel, and stores the electric signal in each pixel. Image data (radioimage) is acquired by reading the electric signal. Then, the radiation detector P outputs the acquired image data to the console 3.

The console 3 outputs the radiation irradiation condition to the radiation irradiation device 1 and outputs the image reading condition to the radiation detection device 2, and controls the radiation irradiation by the radiation irradiation device 1 and the operation of reading the radiation image by the radiation detection device 2. .. Further, the console 3 performs image processing on the captured radiographic image as an image processing unit.

As shown in FIG. 2, the console 3 includes a control unit 31, a storage unit 32, an operation unit 33, a display unit 34, and a communication unit 35, and each unit is connected by a bus 36.

The control unit 31 is composed of a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like. The CPU of the control unit 31 reads out the system program and various processing programs stored in the storage unit 32 and expands them in the RAM in response to the operation of the operation unit 33, and operates each unit of the console 3 according to the expanded program. In addition, the operations of the radiation irradiation device 1 and the radiation detection device 2 are centrally controlled.

The storage unit 32 is composed of a non-volatile semiconductor memory, a hard disk, or the like. The storage unit 32 stores data such as parameters or processing results required for processing by various programs and programs executed by the control unit 31. Various programs are stored in the form of a readable program code, and the control unit 31 sequentially executes an operation according to the program code.
Further, the storage unit 32 stores the radiographic image acquired by imaging in association with patient information (patient ID, patient name, etc.) and examination information (examination date, imaging site, etc.).

The operation unit 33 is configured to include a keyboard equipped with cursor keys, number input keys, various function keys, and a pointing device such as a mouse, and controls an instruction signal input by key operation on the keyboard or mouse operation. Output to 31. Further, the operation unit 33 may include a touch panel on the display screen of the display unit 34, and in this case, the instruction signal input via the touch panel is output to the control unit 31. Further, the operation unit 33 is provided with an exposure switch or the like for instructing the radiation irradiation device 1 to perform radiography.

The display unit 34 is composed of a monitor such as an LCD (Liquid Crystal Display) or a CRT (Cathode Ray Tube), and displays input instructions, data, and the like from the operation unit 33 according to the instruction of the display signal input from the control unit 31. do.

The communication unit 35 has an interface for transmitting / receiving data to / from the radiation irradiation device 1, the radiation detection device 2, and the optical camera 4. The communication between the console 3 and the radiation irradiation device 1, the radiation detection device 2, and the optical camera 4 may be wired communication or wireless communication.

The optical camera 4 is composed of, for example, a CCD (Charge Coupled Device) camera, a CMOS (Complementary Metal Oxide Semiconductor Device) camera, an infrared camera, or the like, and is an optical image capturing unit that captures an optical image. In the present embodiment, the optical camera 4 is provided in the vicinity of the radiation source 11 of the radiation irradiation device 1, and according to the instruction from the console 3, at least a part of the region (basically) of the region where the radiation is emitted from the radiation source 11. An optical image of a region slightly larger than the region irradiated with radiation) is captured, and the captured optical image is transmitted to the console 3.

[Operation of Radiation Imaging System 100]
Next, the operation of the radiation imaging system 100 will be described.
When performing radiological photography, the photographer first prepares for photography. That is, the operation unit 33 of the console 3 accepts the input of patient information and the like, and positions the subject H. When the preparation for shooting is completed, the person performing the shooting instructs the operation unit 33 to start shooting. When the operation unit 33 instructs to start imaging, the console 3 executes the radiation image generation process A shown in FIG. The radiation image generation process A is executed in collaboration with the control unit 31 and the program stored in the storage unit 32.

In the radiation image generation process A, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S1).
That is, an optical image of at least a part of the region irradiated with radiation from the radiation source 11 is acquired.

Next, the control unit 31 determines the imaged portion by analyzing the acquired optical image (step S2).
The imaging site is a site that is the target of imaging a radiographic image.
The method for determining the imaged portion from the optical image is not particularly limited. For example, a template image of an optical image for each part that can be an imaged part in a radiological examination was prepared in advance, and template matching was performed between the optical image acquired from the optical camera 4 and each template image, and the degree of similarity was the highest. The part can be determined as an imaging part.

Next, the control unit 31 sets the image processing conditions for the radiographic image based on the determined imaged portion (step S3).
For example, a conversion table in which each imaged portion is associated with the optimum image processing conditions (image processing parameters) corresponding to the imaged portion is stored in the storage unit 32, and the control unit 31 uses this conversion table. Then, the image processing conditions corresponding to the imaged portion determined in step S3 are specified, and the specified image processing conditions are set as the image processing conditions of the radiographic image to be imaged. Here, the image processing conditions set according to the imaged portion include, for example, dynamic range compression processing, contrast correction processing, density correction processing, LUT (Look Up Table) processing, frequency enhancement processing, scattered ray correction processing, and noise. Suppression processing, image trimming (for example, trimming to 6-cut size if determined to be finger bone), image masking, image rotation, image inversion (eg, inverting image if determined to be chest PA) ) At least one image processing condition.

Next, the control unit 31 causes the radiation irradiation device 1 and the radiation detection device 2 to perform radiography in response to pressing the exposure switch, and acquires a radiation image (step S4).
Then, the acquired radiographic image is subjected to image processing under the image processing conditions set in step S3 (step S5), and the image-processed radiographic image is stored in the storage unit 32 in association with the patient information and the examination information. (Step S6), the radiation image generation process A is terminated.

In the above-mentioned radiation image generation processing A, the imaged part is determined based on the optical image, the image processing conditions according to the imaged part are automatically set, and the image processing is performed on the radiation image. Appropriate image processing according to the part to be imaged can be performed without the user having to perform setting operations. That is, it is possible to omit or simplify the user's setting operation for performing appropriate image processing on the radiographic image.

In the above description, the image processing conditions of the radiographic image are automatically set based on the imaged area determined by analyzing the optical image, but the image processing conditions and the image are set based on the determined imaged area. The reading condition may be automatically set, and radiography may be performed to acquire a radiographic image.
Further, in FIG. 3, although it is decided to acquire the radiographic image after setting the image processing conditions, the radiological image may be acquired in parallel with the analysis of the optical image.

<Second embodiment>
Hereinafter, a second embodiment of the present invention will be described.
In the radiographic image, the optimum image processing conditions change depending on the physique of the subject H.
Therefore, in the second embodiment, an example will be described in which the physique of the subject H is determined based on the optical image of the subject H, and appropriate image processing according to the physique is applied to the radiographic image.

In the second embodiment, the optical camera 4 is arranged with its lens facing the side surface of the subject H, and the side surface of the subject H (may be the whole or a part of the side surface of the subject H) according to the instruction of the console 3. The optical image is acquired and transmitted to the console 3.
Since the other configurations of the radiation imaging system 100 are the same as those described in the first embodiment, the description will be incorporated, and the operation of the second embodiment will be described below.

When the preparation for imaging is completed and the operation unit 33 instructs the start of imaging, the console 3 executes the radiation image generation process B shown in FIG. The radiation image generation process B is executed in collaboration with the control unit 31 and the program stored in the storage unit 32.

In the radiation image generation process B, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S11).
That is, an optical image of the side surface of the subject H is acquired.

Next, the control unit 31 determines the physique of the subject H by analyzing the acquired optical image (step S12).
For example, as described in Japanese Patent Application Laid-Open No. 2011-139716, the control unit 31 extracts the contour of the subject H from the optical image in which the side surface of the subject H is captured, and calculates the body thickness from the extracted contour. Then, by comparing the calculated body thickness with a plurality of predetermined reference values, it is determined whether the body shape of the subject H is, for example, an obese type, a standard type, or a thin type.

Next, the control unit 31 sets the image processing conditions for the radiographic image based on the determined physique (step S13).
For example, a conversion table in which each physique is associated with the optimum image processing conditions corresponding to the physique is stored in the storage unit 32, and the control unit 31 is determined in step S12 using this conversion table. The image processing conditions according to the physique are specified, and the specified image processing conditions are set as the image processing conditions to be applied to the captured radiographic image. Here, the image processing conditions set according to the physique include, for example, dynamic range compression processing, contrast correction processing, density correction processing, LUT (Look Up Table) processing, frequency enhancement processing, scattered ray correction processing, and noise suppression. At least one image processing condition of processing, image trimming, image masking, image rotation, and image inversion can be mentioned.

Next, the control unit 31 causes the radiation irradiation device 1 and the radiation detection device 2 to perform radiography in response to pressing the exposure switch, and acquires a radiation image (step S14).
Then, the acquired radiographic image is subjected to image processing under the image processing conditions set in step S13 (step S15), and the image-processed radiographic image is stored in the storage unit 32 in association with the patient information and the examination information. (Step S16), the radiation image generation process B is terminated.

In the radiation image generation process B, the physique of the subject H is determined based on the optical image, and the image processing conditions according to the physique are automatically set, so that the user sets the physique of the subject H and the image processing parameters. Appropriate image processing according to the physique is possible without performing any operation. That is, it is possible to omit or simplify the user's setting operation for performing appropriate image processing on the radiographic image.

In the above description, the image processing conditions for the radiographic image are automatically set based on the physique determined by analyzing the optical image, but the imaging conditions and the image reading conditions are set based on the determined physique. May be set automatically, and radiography may be performed to acquire a radiological image.
Further, in FIG. 4, it is decided to acquire the radiographic image after the image processing conditions are determined, but the radiological image may be acquired in parallel with the analysis of the optical image.

<Third embodiment>
Hereinafter, a third embodiment of the present invention will be described.
For example, for inpatients and the like, radiological images may be acquired on a regular basis to follow up the medical condition. In the case of follow-up observation, if the image processing conditions are different each time, it is not possible to know whether or not the person's medical condition has changed, and an appropriate diagnosis cannot be made.
Therefore, in the third embodiment, an optical image of the subject H is acquired at the time of radiography, a patient is identified based on the acquired optical image, and the patient is applied to a radiographic image (past image) taken in the past. An example of specifying the image processing conditions and applying them to the radiographic image acquired this time will be described.

In the third embodiment, the storage unit 32 contains patient information, examination information, an optical image of the subject taken at the time of taking the radio image, a feature amount calculated from the optical image, and a feature amount calculated from the optical image in the storage unit 32. The image processing conditions of the image processing applied to the radiographic image are stored in association with each other.

Further, the optical camera 4 is arranged at a position where an optical image including a predetermined structure (for example, a face or an eye) of the subject H can be taken, and includes the predetermined structure of the subject H in response to an instruction from the console 3. The optical image is acquired and transmitted to the console 3.

Since the other configurations of the radiation imaging system 100 are the same as those described in the first embodiment, the description will be incorporated, and the operation of the third embodiment will be described below.

When the preparation for shooting is completed and the operation unit 33 instructs to start shooting, the console 3 executes the radiation image generation process C shown in FIG. The radiation image generation process C is executed in collaboration with the control unit 31 and the program stored in the storage unit 32.

In the radiation image generation process C, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S21).
That is, an optical image including a predetermined structure (for example, face, eyes, etc.) of the subject H is acquired.

Next, the control unit 31 calculates the feature amount (image feature amount) of the subject H by analyzing the acquired optical image (step S22).
In step S22, the control unit 31 calculates, for example, a feature amount representing the size and shape of a predetermined structure of the subject H. For example, as described in Japanese Patent Application Laid-Open No. 2015-21979, a known image recognition technique is used to recognize a face region from an optical image and recognize a predetermined structure (eyes, nose, mouth) in the face region. Calculate the feature amount that represents the size and shape of the optics. For example, a plurality of feature points are set in the structure, and the distance between predetermined feature points and the like are obtained as the feature quantity of the size. Further, the position of each feature point, the distance between each feature point, and the like can be obtained as the feature amount of the shape. Alternatively, as described in Japanese Patent Application Laid-Open No. 2012-29894, the feature amount of the pupil may be calculated by a known pupil authentication processing technique. Moreover, you may calculate the feature amount of a plurality of structures.

Next, the control unit 31 sets the image processing conditions for the radiographic image based on the calculated feature amount (step S23).
In step S23, the calculated feature amount is compared with the feature amount (feature amount calculated from the optical image) stored in association with the past radiation image (same imaged part) in the storage unit 32 and matches. When a feature amount exists, the image processing condition used for the past radiation image associated with the feature amount is read out and set as the image processing condition of the radiation image to be captured this time. This makes it possible to set the same image processing conditions as those used for the past radiographic images of the subject H (same patient).
If the matching feature amount does not exist in the storage unit 32, the user may set the image processing condition from the operation unit 33, for example.

Next, the control unit 31 causes the radiation irradiation device 1 and the radiation detection device 2 to perform radiography in response to pressing the exposure switch, and acquires a radiation image (step S24).
Then, the acquired radiographic image is subjected to image processing under the image processing conditions set in step S23 (step S25), and the image-processed radiographic image is subjected to patient information, examination information, acquired optical image, feature amount, and the like. The image processing conditions are associated with each other and stored in the storage unit 32 (step S26), and the radiation image generation processing C is terminated.

In the radiation image generation processing C, the feature amount of the subject H is calculated from the optical image, and the image processing conditions applied to the past radiation image of the subject H are specified and automatically set based on the calculated feature amount. Therefore, it is possible to perform appropriate image processing capable of follow-up observation without performing an operation of specifying and setting the image processing conditions applied to the past radiation image of the subject H by the user. That is, it is possible to omit or simplify the user's setting operation for performing appropriate image processing on the radiographic image.

In the above description, the image processing conditions for the radiographic image are automatically set based on the feature amount calculated by analyzing the optical image, but the imaging conditions and the image are set based on the calculated feature amount. The reading condition may be automatically set, and radiography may be performed to acquire a radiological image.
Further, in FIG. 5, it is decided to acquire the radiographic image after the image processing conditions are determined, but the radiological image may be acquired in parallel with the analysis of the optical image.

<Fourth Embodiment>
Next, a fourth embodiment of the present invention will be described.
In radiography, the photographer performs positioning and takes a picture, but since the state of the body cannot be seen by looking at the surface of the body, when an actual radiological image is obtained, it may not be suitable for diagnosis and may be re-photographed. be.
Therefore, in the fourth embodiment, whether or not the positioning is good (whether or not the positioning is suitable for diagnosis) is determined before the radiography based on the optical image acquired at the time of radiography, and when the positioning is poor. Notify the user.

In the fourth embodiment, the storage unit 32 contains a learning program including a machine learning program for determining the quality of positioning at the time of radiography by machine learning based on an input optical image, an optical image, and a learning of positioning determination results. Various programs including a learning parameter generation program that performs learning based on the data for learning and generates learning parameters used in the above machine learning program are stored.
Further, the storage unit 32 is provided with a learning data storage unit for accumulating and storing learning data and a learning parameter storage unit for storing learning parameters.
Further, the optical camera 4 is provided in the vicinity of the radiation source 11 of the radiation irradiation device 1, and according to an instruction from the console 3, an optical image of a region where radiation is emitted from the radiation source 11 is captured, and the captured optical image is captured. Send to console 3.
Since the configuration of the radiation imaging system 100 in the other fourth embodiment is the same as that described in the first embodiment, the description is incorporated, and the operation of the fourth embodiment will be described below.

When the preparation for shooting is completed and the operation unit 33 instructs to start shooting, the console 3 refers to the learning parameter storage unit, and if the learning parameter is not stored, performs the first shooting process A shown in FIG. Run. When the learning parameters are stored, the second shooting process A shown in FIG. 7 is executed. The first shooting process A and the second shooting process A are executed in collaboration with the program stored in the control unit 31 and the storage unit 32.

(First shooting process A)
First, the first shooting process A will be described with reference to FIG.
In the first shooting process A, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S31).

Next, the control unit 31 causes the radiation irradiation device 1 and the radiation detection device 2 to perform radiography in response to the pressing of the exposure switch, acquires a radiation image, and stores it in the storage unit 32 (step S32).

Next, the control unit 31 determines the positioning from the acquired radiographic image (step S33).
For example, the control unit 31 displays the acquired radiographic image on the display unit 34, accepts the input of the positioning quality by the user by the operation unit 33, and determines the positioning according to the input of the positioning quality. Alternatively, the control unit 31 may analyze the radiographic image and automatically determine the positioning. For example, the region of interest is detected according to the region of interest, and if the region of interest is included without any defects, it is determined that the positioning is good, and if the region of interest is defective, it is determined that the positioning is poor. You may judge.

Next, the control unit 31 associates the optical image acquired in step S31 with the positioning determination result and stores (adds) it in the learning data storage unit of the storage unit 32 as learning data (step S34).

Next, the control unit 31 performs learning based on the learning data stored in the learning data storage unit according to the learning parameter generation program, and performs machine learning for determining the quality of positioning from the input optical image. Generate learning parameters (step S35).
As machine learning, for example, a convolutional neural network or the like can be used, but the machine learning is not limited to this.

Then, the control unit 31 stores the generated learning parameter in the storage unit 32 (step S36), and ends the first shooting process A.
When the positioning determination result determined in step S33 is poor positioning, the user instructs the operation unit 33 to take a picture and retakes the picture.

It is preferable that steps S34 to S36 are performed for each imaging site.
Further, the processing of steps S35 to S36 may be performed when a predetermined number or more of learning data are stored.

(Second shooting process A)
Next, the second shooting process A will be described with reference to FIG. 7.

In the second shooting process A, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S41).

Next, the control unit 31 takes the acquired optical image as an input and determines whether the positioning is good or bad by machine learning using the learning parameters stored in the learning parameter storage unit of the storage unit 32 (step S42).

When it is determined that the positioning is defective (step S43; YES), the control unit 31 notifies that the positioning is defective (step S44), and ends the second shooting process A.
In step S44, the control unit 31 displays, for example, a notification such as "Positioning failure" on the display unit 34. Alternatively, the console 3 may be provided with a voice output unit, and a notification message such as "positioning failure" may be output by voice.

On the other hand, when it is determined that the positioning is not defective (good) (step S43; NO), the control unit 31 performs radiography on the radiation irradiation device 1 and the radiation detection device 2 in response to the pressing of the exposure switch. The radiographic image is acquired and stored in the storage unit 32 (step S45).

Next, the control unit 31 determines the positioning from the acquired radiographic image, updates the learning parameter based on the determination result (step S46), and ends the second imaging process A.
In step S46, the control unit 31 executes the same processing as in steps S33 to S36 of FIG. 6 to update the learning parameters stored in the learning parameter storage unit.
When the positioning determination result determined in step S46 is poor positioning, the user instructs the operation unit 33 to take a picture and retakes the picture.

As described above, in the fourth embodiment, it is possible to determine whether or not the positioning is good (whether or not the positioning is suitable for diagnosis) based on the optical image, so that it is troublesome to perform re-imaging. And the exposure of patients can be reduced.

<Fifth Embodiment>
Next, a fifth embodiment of the present invention will be described.
In radiography, the photographer performs positioning and takes a picture, but since the state of the body cannot be seen by looking at the surface of the body, when an actual radiological image is obtained, it may not be suitable for diagnosis and may be re-photographed. be.
Therefore, in the fifth embodiment, whether or not the positioning is good (whether or not the positioning is suitable for diagnosis) is determined based on the optical image before radiography, and if the positioning is poor, the user adjusts the positioning amount. Notify to.

In the fifth embodiment, the storage unit 32 is used to determine the quality of positioning at the time of radiography by machine learning based on the input optical image, and to estimate the positioning adjustment amount in the case of poor positioning. Starting with a learning parameter generation program that generates learning parameters used in the above machine learning program by learning based on learning data consisting of machine learning programs, optical images, positioning judgment results, and positioning adjustment amounts in the case of poor positioning. Various programs are stored.
Further, the storage unit 32 is provided with a learning data storage unit for accumulating and storing learning data and a learning parameter storage unit for storing learning parameters.
Further, the optical camera 4 is provided in the vicinity of the radiation source 11 of the radiation irradiation device 1, and according to an instruction from the console 3, an optical image of a region where radiation is emitted from the radiation source 11 is captured, and the captured optical image is captured. Send to console 3.
Since the configuration of the radiation imaging system 100 in the other fifth embodiment is the same as that described in the first embodiment, the description is incorporated, and the operation of the fifth embodiment will be described below.

When the preparation for shooting is completed and the operation unit 33 instructs to start shooting, the console 3 refers to the learning parameter storage unit, and if the learning parameter is not stored, performs the first shooting process B shown in FIG. Run. When the learning parameters are stored, the second shooting process B shown in FIG. 9 is executed. The first shooting process B and the second shooting process B are executed in collaboration with the program stored in the control unit 31 and the storage unit 32.

(First shooting process B)
First, the first shooting process B will be described with reference to FIG.

In the first shooting process B, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S51).

Next, the control unit 31 causes the radiation irradiation device 1 and the radiation detection device 2 to perform radiography in response to the pressing of the exposure switch, acquires a radiation image, and stores it in the storage unit 32 (step S52).

Next, the control unit 31 determines the positioning of the acquired radiographic image (step S53).
Since the process of step S53 is the same as that described in step S33 of FIG. 6, the description is incorporated.

Next, the control unit 31 determines whether or not the positioning determination result is a positioning defect (step S54).
When it is determined that the positioning is not defective (step S54; NO), the control unit 31 proceeds to step S60.

On the other hand, if it is determined that the positioning is poor (step S54; YES), the control unit 31 re-photographs.
That is, the control unit 31 causes the optical camera 4 to take an image and acquires an optical image (step S55).
Next, the control unit 31 causes the radiation irradiation device 1 and the radiation detection device 2 to perform radiography in response to the pressing of the exposure switch, acquires a radiation image, and stores it in the storage unit 32 (step S56).

Next, the control unit 31 determines the positioning of the acquired radiographic image (step S57).
Since the process of step S57 is the same as that described in step S33 of FIG. 6, the description is incorporated.

Next, the control unit 31 determines whether or not the positioning determination result is a positioning defect (step S58).
If it is determined that the positioning is poor (step S58; YES), the control unit 31 returns to step S55 and repeatedly executes steps S55 to S58.

When it is determined that the positioning is not defective (good) (step S58; NO), the control unit 31 includes the optical image of the poor positioning acquired in step S51 and the optical image of good positioning acquired in step S55. Based on the above, the positioning adjustment amount is estimated (step S59), and the process proceeds to step S60.
For example, in an optical image with poor positioning and an optical image with good positioning, a local matching process or the like is performed to calculate a movement amount for matching the optical image with poor positioning to an optical image with good positioning, and the movement amount is used as the positioning adjustment amount. (For example, see Akiko Kano and Machiko Morokaku, "Study of Time-Time Difference Processing of Chest Mass Screening X-ray Images", Konica Minolta Technology Report Vol.8 (1995)).

In step S60, when the acquired optical image, the positioning determination result, and the positioning determination result are poor positioning, the control unit 31 stores the learning data of the storage unit 32 as learning data in association with the positioning adjustment amount. The unit is stored (added) (step S60).

Next, the control unit 31 performs learning based on the learning data stored in the learning data storage unit according to the learning parameter generation program, determines whether the positioning is good or bad from the input optical image, and when the positioning is poor. A learning parameter for machine learning for estimating the positioning adjustment amount is generated (step S61).
As machine learning, for example, a convolutional neural network or the like can be used, but the machine learning is not limited to this.

Then, the control unit 31 stores the generated learning parameter in the storage unit 32 (step S62), and ends the first shooting process B.
It is preferable that steps S60 to S62 are performed for each imaging site.
Further, the processes of steps S61 to S62 may be performed when a predetermined number or more of learning data are stored.

(Second shooting process B)
Next, the second shooting process B will be described with reference to FIG.

In the second shooting process B, the control unit 31 first causes the optical camera 4 to take an image and acquires an optical image (step S71).

Next, the control unit 31 takes the acquired optical image as an input and determines whether the positioning is good or bad by machine learning using the learning parameters stored in the learning parameter storage unit of the storage unit 32 (step S72).

When it is determined that the positioning is poor (step S73; YES), the control unit 31 estimates the positioning adjustment amount by machine learning (step S74), notifies the estimated positioning adjustment amount (step S75), and second. The shooting process B of is finished.
In step S75, the control unit 31 displays, for example, a notification such as "Positioning is poor. Please move the subject to the right by 〇 cm" on the display unit 34. Alternatively, the console 3 may be provided with a voice output unit, and a notification message such as "Positioning is poor. Please move the subject to the right by 〇 cm." May be output by voice.

On the other hand, when it is determined that the positioning is not defective (good) (step S73; NO), the control unit 31 performs radiography on the radiation irradiation device 1 and the radiation detection device 2 in response to the pressing of the exposure switch. The radiographic image is acquired and stored in the storage unit 32 (step S76).

Next, the control unit 31 determines the positioning from the acquired radiographic image, updates the learning parameter based on the determination result (step S77), and ends the second imaging process B.
In step S77, the control unit 31 executes the same process as in steps S53 to S62 of FIG. 8 to update the learning parameters stored in the learning parameter storage unit.
If the positioning determination result determined in step S77 is poor positioning, the user instructs the operation unit 33 to take a picture and retakes the picture.

As described above, in the fifth embodiment, it is possible to determine whether or not the positioning is good (whether or not the positioning is suitable for diagnosis) based on the optical image, so that it is troublesome to perform re-imaging. And it is possible to prevent the exposure of patients. Further, in the case of poor positioning, the user can be notified of the adjustment amount of the positioning, so that the user can easily understand how to adjust the positioning, and the adjustment of the positioning becomes easy.

Although the first to fifth embodiments have been described above, the description contents in the above embodiments are suitable examples of the present invention, and the description is not limited thereto.

For example, in the above description, an example in which a hard disk, a non-volatile memory of a semiconductor, or the like is used as a computer-readable medium of the program according to the present invention has been disclosed, but the present invention is not limited to this example. As another computer-readable medium, a portable recording medium such as a CD-ROM can be applied. Further, a carrier wave is also applied as a medium for providing the data of the program according to the present invention via a communication line.

In addition, the detailed configuration and detailed operation of each device constituting the radiation imaging system can be appropriately changed without departing from the spirit of the present invention.

100 Radiation imaging system 1 Radiation irradiation device 11 Radiation source 2 Radiation detection device P Radiation detector 3 Console 31 Control unit 32 Storage unit 33 Operation unit 34 Display unit 35 Communication unit 36 Bus 4 Optical camera

Claims (8)

A radiation image imaging unit that detects radiation emitted from a radiation source and has passed through a subject and captures a radiation image.
An optical image imaging unit that captures an optical image of at least a part of the subject.
An image processing unit that performs image processing on a radiation image captured by the radiation image imaging unit is provided.
The image processing unit is a radiation image capturing system that sets image processing conditions for the radiation image based on an optical image captured by the optical image imaging unit. The optical image capturing unit captures an optical image of at least a part of the region of the subject to which the radiation from the radiation source is irradiated.
The image processing unit determines a portion of the subject to be imaged in the radiographic image by analyzing the optical image, and sets the image processing condition according to the determined portion. The radiographic imaging system described in. The radiation image capturing system according to claim 1, wherein the image processing unit determines the physique of the subject by analyzing the optical image, and sets the image processing conditions according to the determined physique of the subject. .. The radiation image photographing system according to claim 1, wherein the image processing unit calculates a feature amount of the subject by analyzing the optical image, and sets the image processing condition according to the calculated feature amount. The radiographic imaging system according to claim 4, wherein the feature amount is a feature amount representing the size or shape of a predetermined structure of the subject. The image processing conditions include dynamic range compression processing, contrast correction processing, density correction processing, LUT conversion processing, frequency enhancement processing, scattered ray correction processing, noise suppression processing, image trimming, image masking, image rotation, and image processing. The radiographic imaging system according to any one of claims 1 to 5, which is at least one image processing condition of inversion. A computer that performs image processing on the radiation image captured by the radiation image imaging unit that detects the radiation emitted from the radiation source and transmitted through the subject and captures the radiation image.
A setting unit that sets image processing conditions for a radiation image based on an optical image captured by an optical image imaging unit that captures an optical image of at least a part of the subject.
A program to function as. The process of detecting the radiation emitted from the radiation source and passing through the subject, and taking a radiation image,
The step of capturing an optical image of at least a part of the subject and
A step of setting image processing conditions for the radiographic image based on the captured optical image, and
A step of performing image processing on the radiographic image based on the set image processing conditions, and
Image processing methods including.

Images