Radiation image processing device and program

Abstract

To shorten processing time related to abnormality determination of a captured dynamic image.SOLUTION: In a radiation image processing apparatus 3, a control unit 31 acquires a dynamic image made of a plurality of frame images captured by a radiographic device, and determines whether the dynamic image has an abnormality by using partial frame images of the acquired dynamic image. For example, the control unit performs thinning-out processing of thinning-out the number of frame images of the dynamic image and determines whether the dynamic image has an abnormality by using the frame images after the thinning-out processing. Or, the control unit determines whether the dynamic image has an abnormality by using the partial frame images within a prescribed range continuous in a time direction of the dynamic image.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a radiological image processing apparatus and a program. [Background technology]

[0002] 2. Description of the Related Art Conventionally, there is known a technique for determining abnormalities in a radiographic image acquired by radiography while the radiographic image is being captured. For example, Patent Document 1 describes that during video capture in which radiation irradiation and generation of radiological images are repeated at high speed, the generated radiological images are acquired, and it is determined whether or not there is an abnormality in the acquired radiological images.If it is determined that there is an abnormality, abnormality actions such as "stopping control for irradiating radiation," "deleting the radiological images," "not outputting to the output unit," and "not performing analysis" are performed. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-48864 Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, during exposure, the presence or absence of an abnormality is repeatedly determined each time several tens of frames of images are acquired per second. However, since many dynamic images (videos) exceed several hundred frames, if an abnormality determination of dynamic images is performed using an image processing device or the like with low processing performance (low specifications), the processing time will be long, and there is a possibility that the user will not be able to check the images smoothly.

[0005] An object of the present invention is to shorten the processing time required to determine an abnormality in a captured dynamic image. [Means for solving the problem]

[0006] In order to solve the above problems, the radiation image processing apparatus of the present invention comprises: an acquisition means for acquiring a dynamic image consisting of a plurality of frame images captured by a radiation imaging device; a selection means for selecting a portion of frame images from the plurality of frame images constituting the dynamic image acquired by the acquisition means; an abnormality determination means for determining whether or not there is an abnormality in the dynamic image by using the part of frame images selected by the selection means; a notification means for notifying a result of the determination by the abnormality determination means; Equipped with 、 the selection means selects frame images obtained by a thinning process performed by a thinning means that thins out the number of frame images of the dynamic image as the portion of frame images; the thinning means determines the number of frame images to be thinned out in the dynamic image captured based on order information on the capturing of the dynamic image, and performs the thinning process; The abnormality determination means determines whether or not there is an abnormality in the dynamic image by using the frame images after the thinning process.

[0007] The program of the present invention also includes: Computer, an acquisition means for acquiring a dynamic image consisting of a plurality of frame images captured by a radiation imaging device; a selection means for selecting a portion of frame images from the plurality of frame images constituting the dynamic image acquired by the acquisition means; an abnormality determination means for determining whether or not there is an abnormality in the dynamic image by using the part of frame images selected by the selection means; a notification means for notifying a result of the determination by the abnormality determination means; It functions as the selection means selects frame images obtained by a thinning process performed by a thinning means that thins out the number of frame images of the dynamic image as the portion of frame images; the thinning means determines the number of frame images to be thinned out in the dynamic image captured based on order information on the capturing of the dynamic image, and performs the thinning process; The abnormality determination means determines whether or not there is an abnormality in the dynamic image by using the frame images after the thinning process. [Effects of the Invention]

[0008] According to the present invention, it is possible to reduce the processing time required to determine abnormalities in captured dynamic images. [Brief explanation of the drawings]

[0009] [Figure 1]1 is a block diagram illustrating an example of a radiation imaging system according to an embodiment of the present invention. [Figure 2] 2 is a block diagram showing the functional configuration of a radiation image processing device included in the radiation imaging system of FIG. 1. FIG. [Figure 3] FIG. 10 is a diagram illustrating an example of data storage in an abnormality determination DB. [Figure 4] FIG. 10 is a diagram showing an example of an inspection screen. [Figure 5] 3 is a flowchart showing the flow of an abnormality determination control process A executed by the control unit in FIG. 2. [Figure 6] 10A and 10B are diagrams for explaining a method for selecting a frame image to be used for abnormality determination; [Figure 7] FIG. 10 is a diagram for explaining extraction of a spine. [Figure 8] FIG. 10 is a diagram for explaining calculation of the amount of change in the apex position. [Figure 9] FIG. 10 is a diagram for explaining setting of a determination target region. [Figure 10] FIG. 10 is a diagram illustrating trimming. [Figure 11] FIG. 10 is a diagram showing an example of an examination screen on which an alert is displayed. [Figure 12] FIG. 10 is a diagram showing an example of an inspection screen on which selection buttons for selecting an abnormality determination algorithm to be used in the next abnormality determination are displayed. [Figure 13] 10 is a flowchart showing the flow of an abnormality determination control process B executed by the control unit of FIG. 2 in the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings, but the technical scope of the present invention is not limited to the following embodiments and illustrated examples.

[0011] First Embodiment (Configuration of Radiography System 100) First, the schematic configuration of a radiation imaging system 100 according to the first embodiment will be described. FIG. 1 is a diagram showing an example of the overall configuration of a radiation imaging system 100. As shown in FIG. 1, the radiation imaging system 100 includes a radiation generating device 1, a radiation detector 2, and a radiation image processing device 3. The radiation image processing device 3 is communicably connected to the radiation generating device 1 and the radiation detector 2. The radiation image processing device 3 is also capable of communicating with a hospital information system (HIS), a radiology information system (RIS), a picture archiving and communication system (PACS), a dynamic analysis device, etc. (not shown) via a communication network installed within the hospital.

[0012] Although not shown, the radiation generating device 1 includes a generator that applies a voltage according to preset radiation irradiation conditions based on operation of an irradiation instruction switch, a radiation source that generates radiation (e.g., X-rays) at a dose according to the applied voltage when a voltage is applied from the generator, etc. The radiation generating device 1 generates radiation in a manner according to the radiographic image to be captured.

[0013] The radiation generating device 1 may be installed in an imaging room, or may be configured as a mobile cart together with the radiation image processing device 3 and the like.

[0014] Although not shown, the radiation detector 2 includes a substrate on which pixels are arranged two-dimensionally (in a matrix), the substrate having radiation detection elements that generate electric charges according to the dose when exposed to radiation and switch elements that store and release the electric charges, a scanning circuit that switches each switch element on and off, a readout circuit that reads out the amount of electric charge released from each pixel as a signal value, a control unit that generates a radiographic image from the multiple signal values ​​read out by the readout circuit, and an output unit that outputs data of the generated radiographic image to the outside. The radiation detector 2 may be of the so-called indirect type, which incorporates a scintillator or the like and converts the irradiated radiation into light of a different wavelength, such as visible light, and generates an electric charge according to the converted light, or it may be of the so-called direct type, which generates an electric charge directly from the radiation without going through a scintillator or the like. The radiation detector 2 may be a dedicated type integrated with an imaging stand, or may be a portable type (cassette type).

[0015] The radiation detector 2 generates a radiographic image corresponding to the irradiated radiation in synchronization with the timing of irradiation of radiation from the radiation generating device 1, and transmits the generated radiographic image to the radiographic image processing device 3. That is, the radiation generating device 1 and the radiation detector 2 constitute a radiographic device, and the radiation source of the radiation generating device 1 and the radiation detector 2 are arranged opposite to each other with a gap between them, and by irradiating the subject S placed between them with radiation from the radiation source, it is possible to radiograph the subject S and obtain a radiographic image. The radiographic imaging may be a still image imaging or a dynamic imaging in which the movement of the subject S is captured.

[0016] Here, still image capture refers to obtaining one radiographic image of the subject S in response to a single capture operation (pressing the irradiation instruction switch). Dynamic image capture refers to obtaining multiple radiographic images of the subject S in response to a single capture operation by repeatedly irradiating the subject S with pulsed radiation such as X-rays at predetermined time intervals (pulse irradiation) or by continuously irradiating the subject S with a low dose rate without interruption (continuous irradiation). A series of radiographic images obtained by dynamic image capture is called a dynamic image. Furthermore, each of the multiple radiographic images that make up a dynamic image is called a frame image. Dynamic photography includes video recording, but does not include taking still images while displaying the video. Dynamic images include video, but do not include images obtained by taking still images while displaying the video.

[0017] The radiation image processing device 3 functions as a console for controlling radiation imaging. The radiation image processing device 3 also has a function of performing an abnormality determination process to determine whether or not there is an abnormality in the radiation image transmitted from the radiation detector 2, and issuing a notification when it is determined that there is an abnormality. In this application, an image with an abnormality (an image having an abnormality) refers to an image that was not captured under normal conditions and has an abnormality that makes it unusable for analysis or diagnosis.

[0018] FIG. 2 is a block diagram showing the functional configuration of the radiation image processing device 3. As shown in FIG. As shown in FIG. 2, the radiological image processing device 3 includes a control unit 31, a communication unit 32, a memory unit 33, a display unit 34, an operation unit 35, a sound output unit 36, and a light output unit 37, and each unit is connected by a bus 38. The radiation image processing device 3 may not be provided with the display unit 34 or the operation unit 35, but may be connected to a display device (such as a tablet terminal) that includes a display unit and an operation unit.

[0019] The control unit 31 includes a CPU (Central Processing Unit), a RAM (Random Access Memory ) etc. The CPU of the control unit 31 reads out various programs stored in the storage unit 33, expands them in the RAM, executes various processes in accordance with the expanded programs, and centrally controls the operations of each unit of the radiation image processing device 3. The control unit 31 also functions as the abnormality determination means and thinning means of the present invention by executing an abnormality determination control process (an abnormality determination control process A) described below.

[0020] The communication unit 32 is composed of a communication module and the like. The communication unit 32 transmits and receives various signals and data between the radiation generating device 1, the radiation detector 2, and other external devices and systems (HIS, RIS, PACS, dynamic analysis device, etc.) connected via a communication network (LAN (Local Area Network), WAN (Wide Area Network), Internet, etc.).

[0021] The storage unit 33 is configured by a non-volatile semiconductor memory, a hard disk, or the like, and stores various programs executed by the control unit 31, parameters required for executing the programs, and the like. The memory unit 33 also stores order information related to the examination (each imaging included in the examination) transmitted from the HIS, RIS, etc. The order information includes the examination ID, examination date, patient information, medical department (requesting medical department), disease name, and information related to each imaging included in the examination (imaging site and imaging direction (chest dynamic PA, abdominal dynamic PA, etc.), body position, type of analysis (ventilation, blood flow, etc.), breathing protocol (breath holding, quiet breathing, deep breathing, etc.)). The storage unit 33 also stores the imaging area and imaging direction in association with imaging conditions (e.g., irradiation conditions (tube current, tube voltage, irradiation time, mAs value, frame rate, etc.), reading conditions (pixel size, frame rate, etc.)).

[0022] The storage unit 33 also stores an abnormality determination database (DB) 331. As shown in Fig. 3, the abnormality determination DB 331 stores information (e.g., respiratory medicine, chest, ventilation, etc.) on at least one item included in the order information (e.g., department, imaging region, type of analysis, etc.), the type of at least one abnormality determination algorithm used to determine abnormalities in dynamic images captured based on the order information including that information, and information on frame images used during abnormality determination (e.g., thinning to 1 / 2, thinning to 1 / 4, frame numbers 100 to 200, etc.). The contents of the abnormality determination DB 331 can be changed by the user, for example, through operations on the operation unit 35. The storage unit 33 may also be capable of storing radiographic images.

[0023] The display unit 34 is configured with an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube), or the like. The display unit 34 displays an examination list screen, an examination screen 341 (see, for example, FIGS. 4 and 11), notification information, etc., based on a control signal input from the control unit 31. The display unit 34 functions as a notification means of the present invention in cooperation with the control unit 31.

[0024] The operation unit 35 is an operation means configured to be operable by a user, such as a keyboard equipped with cursor keys, numeric input keys, various function keys, etc., a pointing device such as a mouse, a touch panel laminated on the surface of the display device, etc. The operation unit 35 outputs a control signal to the control unit 31 in accordance with an operation performed by the user.

[0025] The sound output unit 36 ​​includes a speaker or the like, and outputs sound (voice) according to control from the control unit 31. The sound output unit 36 ​​works in cooperation with the control unit 31 to function as a notification means of the present invention. The light output unit 37 is configured with an LED (Light Emitting Diode) or the like, and outputs light in response to control from the control unit 31. The light output unit 37 works in cooperation with the control unit 31 to function as a notification means of the present invention.

[0026] (Operation of the radiation image processing device 3) Next, the operation of the radiation image processing device 3 will be described. For example, when the control unit 31 receives order information from an HIS, RIS, or the like (not shown) via the communication unit 32, the control unit 31 stores the received order information in the memory unit 33 and displays it on an examination list screen (not shown) on the display unit 34. When order information regarding an examination to be performed is selected from the examination list screen, the control unit 21 displays an examination screen 341 on the display unit 34.

[0027] Fig. 4 is a diagram showing an example of the examination screen 341. As shown in Fig. 4, the examination screen 341 is provided with an imaging condition button 341a, an image display area 341b, a patient information display area 341c, an examination interruption button 341d, an examination end button 341e, and the like. The imaging condition button 341a is a button corresponding to each imaging included in the order information, and is a button for setting imaging conditions (irradiation conditions and reading conditions) corresponding to each imaging in the radiation generating device 1 and the radiation detector 2. The imaging condition button 241a displays, for example, the imaging part and imaging direction of each imaging to identify each imaging included in the order information. The image display area 341b is an area for displaying a radiographic image acquired by radiography. The patient information display area 341c is an area for displaying patient information of the patient (subject S) to be examined. The examination interruption button 341d is a button for the user (imaging implementer) to instruct the interruption of the examination. The test end button 341e is a button for the user to instruct the end of the test.

[0028] The user presses the imaging condition button 341a corresponding to the next radiation imaging to be performed on the examination screen 341 to prepare for imaging. When the user operates the operation unit 35 to press one of the imaging condition buttons 341a on the examination screen 341, the control unit 31 reads out the imaging conditions corresponding to the imaging region and imaging direction of the pressed imaging condition button 341a from the storage unit 33, and transmits the irradiation conditions, among the read imaging conditions, to the radiation generation device 1 via the communication unit 32. In addition, the control unit 31 transmits the reading conditions, among the read imaging conditions, to the radiation detector 2 via the communication unit 32. The radiation generating device 1 sets the irradiation conditions received from the radiation image processing device 3 as the irradiation conditions for the next radiation imaging to be performed. The radiation detector 2 sets the reading conditions received from the radiation image processing device 3 as the reading conditions for the next radiation imaging to be performed.

[0029] The user positions the subject S between the radiation source of the radiation generation device 1 and the radiation detector 2, and when preparations for imaging are complete, operates the irradiation instruction switch. When the irradiation instruction switch is operated, the radiation generation device 1 irradiates the subject S with radiation under the set irradiation conditions. The radiation detector 2 accumulates and reads out the irradiated radiation in synchronization with the radiation irradiation by the radiation generation device 1, generates image data of the radiation image (frame images), and transmits the image data to the radiation image processing device 3. If the imaging condition corresponding to the pressed imaging condition button 341a is dynamic imaging, the radiation generation device 1 and the radiation detector 2 synchronously repeat the irradiation of radiation and the generation and transmission of frame images for a predetermined period of time. The radiation detector 2 assigns a frame number indicating the imaging order to each generated frame image and transmits the frame image to the radiation image processing device 3.

[0030] When the radiological image processing device 3 receives a radiological image from the radiation detector 2 via the communication unit 32, it performs an abnormality determination process to determine whether or not there is an abnormality in the received radiological image, and if it determines that there is an abnormality, it notifies the user of this, for example, by displaying it on the examination screen 341 or by outputting an audio message.

[0031] Here, in the case of still images, only one image needs to be judged, so even if the processing performance (specs) of the radiation image processing device 3 is low, the impact on the user's image confirmation is small. However, in the case of dynamic images, multiple frame images (for example, one dynamic image contains several hundred frames (e.g., 300 frames)) are included, so if abnormality detection processing is performed on all frame images, it takes a long processing time, making it difficult for the user to proceed to the next task, such as image confirmation. Furthermore, not all frame images are necessarily required to detect an abnormality. For example, if there is a lung field defect, several to several tens of frame images are acquired per second, so there is a high possibility that frame images depicting the lung field defect will be depicted in several surrounding frames. When detecting an abnormality, not all frame images are necessarily required. Therefore, when dynamic imaging is started, the control unit 31 of the radiological image processing device 3 of this embodiment starts executing the abnormality determination control process A shown in Figure 5, so that the processing time for the abnormality determination process of dynamic images can be shortened compared to conventional methods.

[0032] The abnormality determination control process A will be described below with reference to Fig. 5. The abnormality determination control process A is executed by the CPU of the control unit 31 in cooperation with a program stored in the ROM or storage unit 33.

[0033] In the abnormality determination control process A, the control unit 31 first acquires order information related to the started dynamic imaging (step S1), and determines the abnormality determination algorithm corresponding to the acquired order information by referring to the abnormality determination DB 331 (step S2). Although the abnormality determination algorithm is determined based on the order information in this example, the abnormality determination algorithm may be determined in response to the user's operation of the operation unit 35 .

[0034] Next, the control unit 31 sequentially acquires the frame images received by the communication unit 32 (step S3), and selects some of the frame images to be used for abnormality determination (step S4). For example, the control unit 31 acquires from the abnormality determination DB 331 information (number of frames to be thinned out and range) of frame images to be used during abnormality determination using the abnormality determination algorithm corresponding to the order information determined in step S2, and selects some frame images to be used for abnormality determination from the multiple frame images of the received dynamic image based on the acquired information. For example, as shown in Fig. 6, based on the acquired information, the control unit 31 selects frame images to be used for abnormality determination by thinning out the number of received frame images by a predetermined number at predetermined intervals, or by acquiring some frame images within a predetermined range that are consecutive in the time direction from the received frame images (i.e., a predetermined range of frame numbers).

[0035] For example, if the information about frame images to be used during abnormality determination obtained from the abnormality determination DB 331 is "thinned out to 1 / 2," the control unit 31 performs a thinning process to acquire one frame image every other frame number (thinning out frame images that are not acquired) to reduce the number of frame images, thereby selecting the frame images to be used during abnormality determination. Also, if the information about frame images to be used during abnormality determination obtained from the abnormality determination DB 331 is "thinned out to 1 / 4," the control unit 31 performs a thinning process to acquire one frame image every three frame numbers (thinning out frame images that are not acquired) to reduce the number of frame images, thereby selecting the frame images to be used during abnormality determination. Note that the number of frame images acquired every N (N is a positive integer) is not limited to one, and may be multiple. Also, for example, if the information on the frame images to be used during abnormality determination obtained from the abnormality determination DB331 is "frame numbers 100 to 200," then frame images in the range of frame numbers 100 to 200 from among the frame images received sequentially from the radiation detector 2 are selected as the frame images to be used for abnormality determination. It is also possible to select frame images to be used when determining an abnormality by further thinning out frame images within a predetermined range of frame numbers.

[0036] Next, the control unit 31 performs an abnormality determination process on the selected frame image using the abnormality determination algorithm determined in step S2 (step S5). For example, when the imaging region is the chest, the abnormality determination algorithms that can be used in the radiation image processing device 3 include at least one of the following abnormality determination algorithms (1) to (9).

[0037] (1) A frame image of the maximum expiratory position and / or maximum inspiratory position is obtained from the selected frame images, and it is determined whether or not there is an abnormality in the dynamic image based on the lung field area of ​​the obtained frame image. For example, the memory unit 33 stores the reference area of ​​a typical lung field at the maximum inspiration position and / or maximum expiration position, and the area of ​​the maximum inspiration position and maximum expiration position calculated from dynamic images previously taken for each patient (associated with patient information), and if the difference between the lung field area at the maximum inspiration position and / or maximum expiration position calculated from the selected frame image and the reference area or the previous lung field area of ​​the same patient is equal to or greater than a predetermined threshold, it is determined that there is an abnormality, and if it is less than the threshold, it is determined that there is no abnormality. For example, if the lung field area at the maximum inspiration position is smaller than the reference area at the maximum inspiration position or the lung field area at the same patient's previous maximum inspiration position by a predetermined threshold or more, it is determined that the patient is not breathing sufficiently (not breathing in sufficiently) and therefore an abnormality exists. Also, if the lung field area at the maximum expiration position is larger than the reference area at the maximum expiration position or the lung field area at the same patient's previous maximum expiration position by a predetermined threshold or more, it is determined that the patient is not breathing sufficiently (not breathing out sufficiently) and therefore an abnormality exists. The lung field area can be calculated, for example, by recognizing the lung field area from the selected frame image and calculating it based on the pixel size and the number of pixels of the recognized lung field area. The method for recognizing the lung field area is not particularly limited, and can be performed using, for example, image processing such as edge detection, deep learning, machine learning using training data, or the like.

[0038] (2) The amount of change (amount of change in position) of the diaphragm or thorax is calculated from the selected frame image, and based on the calculated amount of change, it is determined whether or not there is an abnormality in the dynamic image. For example, the diaphragm or rib cage is extracted from each of the selected frame images, and a predetermined position of the extracted diaphragm, for example, the coordinate (y coordinate) of the highest position, or a predetermined position of the extracted rib cage, for example, the coordinate (x coordinate) of the outermost position, is obtained. Then, for example, the difference between the frame image with the smallest coordinate value and the frame image with the largest coordinate value is taken as the amount of change, and if the amount of change is equal to or less than a predetermined threshold, it is determined that there is an abnormality in the dynamic image, and if it is greater than the threshold, it is determined that there is no abnormality. For example, the lower edge portion of the recognized lung field area can be extracted as the diaphragm, and for example, the edge portion of the outer contour of the recognized lung field area can be extracted as the thorax.

[0039] The abnormality determination algorithms (1) and (2) are used to determine whether a patient is abnormal when the patient is not breathing properly, and are preferably not applied to patients who are unable to breathe properly and who have or are suspected to have a specific disease. For example, if the order information contains the name of a specific disease, the control unit 31 preferably controls so as not to apply the abnormality determination algorithms (1) and (2).

[0040] (3) The tilt angle of the subject S is calculated using the selected frame image, and based on the calculated tilt angle, it is determined whether or not there is an abnormality in the dynamic image. For example, as shown in FIG. 7, a spine (shown enclosed in a rectangle in FIG. 7) is extracted from one of the selected frame images. If extraction of the spine is successful, the tilt angle of the extracted spine with respect to the vertical direction (y direction) is calculated, and if the tilt angle exceeds a predetermined range (for example, ±10°), it is determined that an abnormality exists, and if it is within the range, it is determined that no abnormality exists. If extraction of the spine fails, spine extraction is performed from the next selected frame image, and if extraction of the spine is successful, the tilt angle of the extracted spine with respect to the vertical direction is calculated, and if the tilt angle exceeds a predetermined threshold (for example, ±10°), it is determined that an abnormality exists, and if it is within the range, it is determined that no abnormality exists. Extraction of the spine can be performed using image processing such as edge detection, for example. The abnormality determination algorithm (3) determines that there is an abnormality when the tilt angle of the subject S is equal to or greater than a predetermined threshold, because if the subject S is tilted at a angle greater than a predetermined threshold, an abnormality may occur during dynamic analysis. In this example, the presence or absence of an abnormality is determined from only one frame image in which the spine was successfully extracted, but the presence or absence of an abnormality may also be determined from multiple frame images. Furthermore, if the subject's posture shifts during imaging, this will affect dynamic analysis and diagnosis. Therefore, the spine may be extracted from a specific frame image among the selected frame images, for example, the first frame image, and the tilt angle may be measured (if the spine cannot be extracted, for example, the tilt angle of a frame image with a successively smaller frame number may be measured), and this may be used as the reference tilt angle. The tilt angles of the spine in other selected frame images may then be compared with the reference tilt angle, and if the tilt angle is equal to or greater than a predetermined threshold, it may be determined that an abnormality exists.

[0041] (4) Using the selected frame images, the amount of change in a specified structure (e.g., diaphragm position, lung field area, etc.) or the amount of change in image signal values ​​within a specified structure area (e.g., within the lung field area) is calculated, and the calculated amount of change is compared with a predetermined threshold for the breathing protocol (the breathing state to be photographed (breath holding, deep breathing, quiet breathing, etc.)), and based on the comparison result, it is determined whether or not there is an abnormality in the dynamic image (threshold: breath holding < quiet breathing < deep breathing). This algorithm can determine that there is an abnormality if breathing movement is not performed in accordance with the breathing protocol.

[0042] (5) The position of the apex of the lung (shown by a rectangle in FIG. 8) is identified from the selected frame image, and if the amount of change from the position of the apex of the lung in the frame image with the closest frame number (previous image) is equal to or greater than a predetermined threshold, it is determined that there is an abnormality in the dynamic image, and if it is less than the predetermined threshold, it is determined that there is no abnormality. The position of the apex of the lung can be identified, for example, as the highest point in the lung field region. Since the apex of the lung is a place where movement is unlikely to occur, if there is a positional change equal to or greater than a predetermined threshold, it can be determined that body movement has occurred.

[0043] (6) A process for detecting foreign matter (e.g., gauze) and / or metal artifacts is performed from the selected frame images. If a foreign matter and / or metal artifacts is detected, the dynamic image is determined to have an abnormality; if not, the dynamic image is determined to have no abnormality. Any known method may be used to detect foreign matter and metal artifacts. For example, foreign matter can be detected using deep learning or machine learning using training data. Metal artifacts can be detected, for example, by binarization processing or graph cut processing, which is an advanced region extraction processing. This makes it possible to detect abnormalities due to foreign matter contamination or metal artifacts.

[0044] (7) The lung field and scapula are recognized from the selected frame image, and the presence or absence of abnormalities in the dynamic image is determined based on the positional relationship between the recognized scapula and the lung field (whether they overlap). This is because dynamic analysis cannot be performed accurately if the scapula overlaps the lung field. Any known method can be used to detect the scapula, and for example, it can be detected using deep learning or machine learning using training data.

[0045] (8) Determine whether or not a lung field defect has occurred from the selected frame image. If it is determined that a lung field defect has occurred, determine that there is an abnormality in the dynamic image. If it is determined that there is no lung field defect, determine that there is no abnormality. For example, if the lung field does not fall within the irradiation field area, determine that there is a lung field defect.

[0046] (9) Whether or not there is an overdose or underdose is determined from the selected frame image, and if it is determined that there is an overdose or underdose, it is determined that there is an abnormality in the dynamic image, and if it is determined that there is no overdose or underdose, it is determined that there is no abnormality.Whether or not there is an overdose or underdose is determined is not particularly limited, but can be determined, for example, by comparing the average or median of image signal values ​​in a region of interest (e.g., lung field in the case of the chest) predetermined for each imaging region with a predetermined threshold.

[0047] (10) Determine whether or not there is body movement of the subject from the selected frame image, and if it is determined that there is body movement of the subject, determine that there is an abnormality in the dynamic image, and if it is determined that there is no body movement of the subject, determine that there is no abnormality. The method for determining the presence or absence of body movement of the subject is not particularly limited, but for example, a method of determining based on the amount of change in the position of the apex of the lung, as described above, can be used.

[0048] When performing the abnormality determination process, the control unit 31 may designate a partial region of the selected frame image as a determination target region, as shown in Fig. 9, and perform the abnormality determination process using an image of the determination target region. The determination target region may be determined in advance depending on the type of abnormality determination algorithm, or may be specified by the user by operating the operation unit 35. This can further reduce the time required for the abnormality determination process. The determination target region may be multiple regions.

[0049] Alternatively, the control unit 31 may reduce the size of each selected frame image by trimming it as shown in Fig. 10 before performing the abnormality determination process, and then use the reduced image to perform the abnormality determination process. For example, the image may be trimmed so that only the above-mentioned determination target area remains. This can further reduce the time required for the abnormality determination process.

[0050] Furthermore, the control unit 31 may perform pre-processing such as gradation processing or frequency processing on the selected frame image before carrying out the abnormality determination processing.

[0051] Next, the control unit 31 determines whether or not it has been determined by the abnormality determination process that there is an abnormality in the dynamic image (step S6). If it is determined that there is no abnormality in the dynamic image (step S6; NO), the control unit 31 normally displays the dynamic image on the display unit 34 (step S7), and terminates the abnormality determination control process A. For example, the dynamic image is displayed in the image display area 341b of the examination screen 341. If it is determined that there is an abnormality in the dynamic image (step S6; YES), the control unit 31 displays (notifies) an alert indicating that there is an abnormality in the dynamic image on the display unit 34 (step S8), and terminates the abnormality determination control process A. The alert may be output as sound by the sound output unit 36, or may be output as light by the light output unit 37.

[0052] FIG. 11 is a diagram showing an example of the examination screen 341 on which an alert is displayed. For example, if an abnormality is determined to exist, the control unit 31 displays a message 342a indicating that an abnormality has been detected on the examination screen 341 as an alert to notify the user of the abnormality. This may be output by the sound output unit 36 ​​or the light output unit 37. Alternatively, an icon indicating that an abnormality has been detected may be displayed. This allows the user to easily recognize that an abnormality exists in the dynamic image captured in dynamic shooting. Furthermore, the frame image determined to have an abnormality may be displayed in the image display area 341b of the examination screen 341. Alternatively, the frame number 342b of the frame image determined to have an abnormality may be displayed on the examination screen 341. Alternatively, the seek bar (playback bar) 342c displayed below the image display area 341b may be highlighted in color near the position of the image determined to have an abnormality.

[0053] Furthermore, when it is determined that there is an abnormality in the dynamic image, the control unit 31 may cause the examination screen 341 to display a partial reject button 342d and a total reject button 342e. When the partial reject button 342d is pressed, the control unit 31 automatically rejects only the frame image in which an abnormality is detected (for example, deletes it from the storage unit 33). Alternatively, when the partial reject button 342d is pressed, a designation screen for the user to designate the range to be rejected is displayed as a pop-up, the range designated by the user is designated as rejected, and a dynamic image consisting of the other frame images is attached to the patient information and examination information (information on a predetermined item of the order information), and stored in the storage unit 33 or output to a predetermined output destination such as a PACS or a dynamic analysis device via the communication unit 32. When the all reject button 342e is pressed, the control unit 31 dynamically rejects all frame images of the dynamic image (for example, deletes them from the storage unit 33). In this case, re-imaging becomes possible. For example, when the imaging condition button 341a corresponding to the performed imaging is pressed, the imaging conditions are set again in the radiation generator 1 and the radiation detector 2, and imaging becomes possible in response to pressing the irradiation instruction button. The rejected image may be given a flag indicating that it is a rejected image and output to a predetermined output destination (for example, a management device that manages rejected images).

[0054] Note that the control unit 31 may immediately (in real time) display the alert in step S8 when it is determined that some of the selected frame images contain an abnormality, or may display the alert in step S8 after the abnormality determination process for all of the selected frame images is complete. Furthermore, for example, it may determine whether to display the alert in real time or to display the alert all at once after the abnormality determination process is complete, based on the difference between the value used to determine whether or not an abnormality exists and a threshold value. By immediately displaying the alert in step S8 when it is determined that some of the selected frame images contain an abnormality, the user can immediately (in real time) recognize that there is an abnormality in the captured image.

[0055] Furthermore, after the abnormality determination process is completed, the control unit 31 may display a selection button 341j on the display unit 34 for instructing the execution of abnormality determination process using an abnormality determination algorithm other than the abnormality determination algorithm used in step S5, as shown in Fig. 12. Then, the control unit 31 may additionally execute abnormality determination process on the dynamic image using another abnormality determination algorithm selected in response to pressing of the selection button 341j by the operation unit 35. Alternatively, the control unit 31 may automatically execute abnormality determination process using a preset abnormality determination algorithm other than the abnormality determination algorithm used in step S5. This allows for prioritizing real-time performance and postponing abnormality determination process that takes time to process.

[0056] In this way, the control unit 31 of the radiological image processing device 3 performs abnormality determination on the captured dynamic image using some of the frame images constituting the dynamic image, so abnormality determination can be performed with low load and the processing time required for abnormality determination on the dynamic image can be shortened. As a result, even if the radiological image processing device 3 is a device with low processing performance and low specifications, the processing time required for abnormality determination processing can be reduced, allowing the user to quickly check the image or re-take an image.

[0057] <Second embodiment> Next, a second embodiment of the present invention will be described. In the second embodiment, when the communication unit 32 starts receiving frame images of dynamic images from the radiation detector 2, the control unit 31 of the radiation image processing device 3 thins out the number of frame images (for example, thins out to 2 fps) for image confirmation in parallel with imaging, and performs image processing that is simpler than that of the actual images used for diagnosis, and sequentially displays them on the display unit 34 (such as the image display area 241b of the examination screen 241). These frame images that have undergone simple image processing for image confirmation are called live view images. In the second embodiment, a case will be described in which the live view images are used to perform abnormality determination processing.

[0058] The configuration of the radiation imaging system 100 and each device in the second embodiment is the same as that described in the first embodiment, so the description will be used to describe the operation of the radiation image processing device 3 in the second embodiment.

[0059] 13 is a flowchart showing the flow of an abnormality determination control process (referred to as abnormality determination control process B) in the second embodiment. The abnormality determination control process B is executed by the CPU of the control unit 31 in cooperation with a program stored in the ROM or storage unit 33.

[0060] In the abnormality determination control process B, the control unit 31 first acquires order information related to the started dynamic imaging (step S21), and determines the abnormality determination algorithm corresponding to the acquired order information by referring to the abnormality determination DB 331 (step S22). The processing in steps S21 and S22 is the same as steps S1 and S2 in FIG. 5, and therefore the same explanation will be used.

[0061] Next, the control unit 31 acquires a live view image (step S23). As described above, the control unit 31 generates a live view image in parallel with shooting. The control unit 31 acquires the generated live view image for abnormality determination. As described above, the number of frame images has already been thinned out from the live view image. Note that it is also possible to further thin out the live view image.

[0062] Next, the control unit 31 starts the abnormality determination process for the live view image using the abnormality determination algorithm determined in step S22 (step S24). The processing in step S24 is the same as that explained in step S5 of FIG. 5 except that the frame images used in the processing are live view images, and therefore the explanation therefor will be cited.

[0063] Next, the control unit 31 determines whether or not a predetermined time has elapsed since the start of the abnormality determination process (step S25). If it is determined that the specified time has not elapsed (step S25; NO), the control unit 31 determines whether or not the abnormality determination process for the final live view image has been completed (step S26). When it is determined that the abnormality determination process for the final live view image has not been completed (step S26; NO), the control unit 31 returns to step S25. When it is determined that the abnormality determination process for the final live view image has been completed (step S26; YES), the control unit 31 determines whether or not the abnormality determination process has determined that there is an abnormality in the dynamic image (step S27).

[0064] If it is determined that there is no abnormality in the dynamic image (step S27; NO), the control unit 31 continues the normal display of the live view image on the display unit 34 (step S28) and terminates the abnormality determination control process B. For example, the control unit 31 continues the display of the live view image in the image display area 341b of the examination screen 341, which has already been performed.

[0065] If it is determined that there is an abnormality in the live view image (step S27; YES), the control unit 31 displays an alert indicating that there is an abnormality in the dynamic image together with the live view image on the display unit 34 (step S29), and terminates the abnormality determination control process B. The alert display is similar to that described in the first embodiment, and therefore the description will be used here.

[0066] In the second embodiment, as shown in Fig. 11, a partial reject button 342d and a total reject button 342e may be displayed on the inspection screen 341, and as in the first embodiment, part or all of the dynamic image may be rejected in response to pressing of the partial reject button 342d or the total reject button 342e. Furthermore, as shown in Fig. 12, a selection button 341j may be displayed on the inspection screen 341, and an abnormality determination may be performed using an abnormality determination algorithm different from the abnormality determination process that was performed and selected in response to pressing of the selection button 341j. Furthermore, if the abnormality determination process is interrupted midway, a resume button or the like may be displayed so that the interrupted abnormality determination process can be resumed in response to a user operation.

[0067] On the other hand, in step S25, if it is determined that a predetermined specified time has elapsed since the start of the abnormality determination process (step S25; YES), the control unit 31 interrupts the abnormality determination process (step S30) and terminates the abnormality determination control process B.

[0068] In this way, in the second embodiment, the abnormality determination process is performed using a live view image that has been generated for image confirmation and has undergone simple processing by thinning out the number of frame images of a dynamic image, which contributes to real-time abnormality determination. Also, if the processing time for the abnormality determination process exceeds a specified time, the abnormality determination process is interrupted, so if the abnormality determination process takes too long, the determination can be stopped midway, allowing user operation.

[0069] As described above, according to the radiation image processing device 3, the control unit 31 acquires a dynamic image consisting of a plurality of frame images captured by a radiation imaging device, and determines whether or not there is an abnormality in the dynamic image using some of the frame images from the acquired dynamic image. For example, a thinning process is performed to thin out the number of frame images from the dynamic image, and the frame images after the thinning process are used to determine whether or not there is an abnormality in the dynamic image. Alternatively, a predetermined range of frame images that are consecutive in the time direction of the dynamic image are used to determine whether or not there is an abnormality in the dynamic image. Therefore, abnormality determination can be performed with low load, and the processing time required for abnormality determination of dynamic images can be shortened. As a result, even if the radiation image processing device 3 is a device with low processing performance and low specifications, the processing time required for abnormality determination processing can be reduced, allowing the user to quickly check the image or re-take an image.

[0070] For example, the control unit 31 determines the number of frame images to be thinned out based on order information regarding dynamic image capture. Therefore, the control unit 31 can determine the optimal number of frame images depending on the region to be captured, the medical department, the type of analysis, etc., included in the order information.

[0071] Furthermore, for example, the control unit 31 determines whether or not there is an abnormality in the dynamic image by using a live view image generated to check the dynamic image captured by the radiation imaging device, which can contribute to real-time abnormality determination.

[0072] Furthermore, for example, the control unit 31 suspends the determination process when the processing time for determining whether or not there is an abnormality in the dynamic image exceeds a specified time. Therefore, if the abnormality determination process takes too long, the determination can be stopped midway, allowing the user to perform operations.

[0073] Furthermore, for example, after determining whether or not there is an abnormality in a dynamic image using one of multiple types of abnormality determination algorithms, it is possible to perform additional abnormality determination using other abnormality determination algorithms, so that emphasis is placed on real-time performance and abnormality determination processing that takes time to process can be postponed.

[0074] Furthermore, the control unit 31 determines whether or not there is an abnormality in the dynamic image by using a determination target region of a portion of the frame images of the dynamic image, and therefore the time required for the abnormality determination process can be further reduced.

[0075] Furthermore, the control unit 31 trims and reduces each of the frame images of the dynamic image, and determines whether or not there is an abnormality in the dynamic image using the reduced images, thereby further reducing the time required for the abnormality determination process.

[0076] The radiological image processing device 3 also includes an abnormality determination DB 331 that stores at least one item of information included in the order information related to radiography in association with at least one of a plurality of types of abnormality determination algorithms, and the control unit 31 determines the abnormality determination algorithm to be applied to the dynamic image based on the order information related to radiography of the dynamic image and the abnormality determination DB 331, and determines whether or not there is an abnormality in the dynamic image using the determined abnormality determination algorithm. Therefore, it is possible to implement an abnormality determination algorithm according to, for example, the radiographic body part, the medical department, the type of analysis, etc.

[0077] Furthermore, when it is determined that there is an abnormality in the dynamic image, the control unit 31 outputs a message or icon informing the user of the abnormality, thereby making it possible to make the user aware of the abnormality in the dynamic image.

[0078] Furthermore, the control unit 31 includes an operation means for the user to instruct that some or all of the frame images of the dynamic image be rejected when the dynamic image is determined to be abnormal, so that the user can instruct that some or all of the frame images of the dynamic image be rejected.

[0079] The description of the above embodiment is a preferred example of the radiation image processing apparatus according to the present invention, and the present invention is not limited to this.

[0080] For example, in the above description, examples have been disclosed in which a hard disk or a semiconductor nonvolatile memory is used as a computer-readable medium for the program according to the present invention, but the present invention is not limited to this example. Other computer-readable media include portable recording media such as CD-ROMs. Furthermore, carrier waves can also be used as a medium for providing data for the radiation image display program according to the present invention via a communication line.

[0081] In addition, the detailed configuration and detailed operation of each device constituting the radiation image processing apparatus may be modified as appropriate within the scope of the invention. [Explanation of symbols]

[0082] 100 Radiography System 1. Radiation generator 2. Radiation detectors 3 Radiation image processing device 31 Control Unit 32 Communications Department 33 Storage section 331 Abnormality judgment DB 34 Display section 35 Control section 36 Sound output section 37 Bus S Subject

Claims

1. an acquisition means for acquiring a dynamic image consisting of a plurality of frame images captured by a radiation imaging device; a selection means for selecting a portion of frame images from the plurality of frame images constituting the dynamic image acquired by the acquisition means; an abnormality determination means for determining whether or not there is an abnormality in the dynamic image by using the part of frame images selected by the selection means; a notification means for notifying a result of the determination by the abnormality determination means; Equipped with the selection means selects frame images obtained by a thinning process performed by a thinning means that thins out the number of frame images of the dynamic image as the portion of frame images; the thinning means determines the number of frame images to be thinned out in the dynamic image captured based on order information on the capturing of the dynamic image, and performs the thinning process; The abnormality determination means determines whether or not there is an abnormality in the dynamic image by using the frame images after the thinning process.

2. 2. The radiological image processing device according to claim 1, wherein the abnormality determination means performs the determination using at least one of a plurality of types of abnormality determination algorithms, and after determining whether or not there is an abnormality in the dynamic image using any one of the abnormality determination algorithms, performs additional abnormality determination using another abnormality determination algorithm.

3. 3. The radiographic image processing apparatus according to claim 1, wherein the abnormality determination means determines whether or not the dynamic image has an abnormality by using a determination target region of a portion of the selected frame images.

4. The radiological image processing device according to any one of claims 1 to 3, wherein the abnormality determination means trims and reduces each of the selected frame images, and uses the reduced images to determine whether or not there is an abnormality in the dynamic image.

5. a database that stores at least one item of information included in order information related to imaging in association with at least one of a plurality of types of abnormality determination algorithms; The radiological image processing device according to any one of claims 1 to 4, wherein the abnormality determination means determines an abnormality determination algorithm to be applied to the dynamic image based on order information regarding the capturing of the dynamic image and the database, and determines whether or not there is an abnormality in the dynamic image using the determined abnormality determination algorithm.

6. 6. The radiological image processing device according to claim 1, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm for obtaining a frame image at a maximum expiratory position and / or a maximum inspiratory position from the selected portion of frame images, and determining whether or not there is an abnormality in the dynamic image based on the lung field area of ​​the obtained frame image.

7. The radiological image processing device according to any one of claims 1 to 6, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm that obtains the amount of change in the position of the diaphragm or thorax from the selected portion of frame images, and determines whether or not there is an abnormality in the dynamic image based on the obtained amount of change.

8. The radiological image processing device according to any one of claims 1 to 7, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm that obtains the tilt angle of the subject from the selected portion of frame images and determines whether or not there is an abnormality in the dynamic image based on the obtained tilt angle.

9. The radiological image processing device according to any one of claims 1 to 8, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm that calculates the amount of change in a predetermined structure or the amount of change in image signal values ​​within a predetermined structure area from the selected portion of frame images, compares the calculated amount of change with a threshold value that is predetermined for the breathing protocol at the time of imaging, and determines whether or not there is an abnormality in the dynamic image based on the comparison result.

10. A radiological image processing device according to any one of claims 1 to 9, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm that identifies the position of the apex of the lung from the selected portion of frame images and determines whether or not there is an abnormality in the dynamic image based on changes in the identified position of the apex of the lung.

11. The radiological image processing device according to any one of claims 1 to 10, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm that performs processing to detect foreign matter or metal artifacts from the selected portion of frame images and determines whether or not there is an abnormality in the dynamic image based on the detection results.

12. The radiological image processing device according to any one of claims 1 to 11, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm that recognizes the scapula and lung field area from the selected portion of frame images and determines whether or not there is an abnormality in the dynamic image based on the positional relationship between the recognized scapula and lung field area.

13. A radiological image processing device according to any one of claims 1 to 12, wherein the abnormality determination algorithm usable by the abnormality determination means includes an algorithm for determining whether or not there is an abnormality such as body movement of the subject, a lung field defect, or an excess or deficiency of dose from the selected portion of frame images.

14. The radiological image processing device according to any one of claims 1 to 13, wherein the notification means outputs a message or icon notifying the user that the dynamic image is abnormal when the abnormality determination means determines that the dynamic image is abnormal.

15. A radiological image processing device as described in any one of claims 1 to 14, further comprising an operation means for a user to instruct that some or all of the frame images of the dynamic image be rejected when the dynamic image is determined to be abnormal by the abnormality determination means.

16. Computer, an acquisition means for acquiring a dynamic image consisting of a plurality of frame images captured by a radiation imaging device; a selection means for selecting a portion of frame images from the plurality of frame images constituting the dynamic image acquired by the acquisition means; an abnormality determination means for determining whether or not there is an abnormality in the dynamic image by using the part of frame images selected by the selection means; a notification means for notifying a result of the determination by the abnormality determination means; It functions as the selection means selects frame images obtained by a thinning process performed by a thinning means that thins out the number of frame images of the dynamic image as the portion of frame images; the thinning means determines the number of frame images to be thinned out in the dynamic image captured based on order information on the capturing of the dynamic image, and performs the thinning process; The abnormality determination means determines whether or not there is an abnormality in the dynamic image using the frame images after the thinning process.

Images