Radiation image analysis device, radiation image processing method, and recording medium

Abstract

The invention relates to a radiological image analysis apparatus, a radiological image processing method, and a recording medium. A radiation image analysis device is provided with: a first processing unit that performs a first process in which, for each of a plurality of frame images included in a radiation moving image of a target site of a subject, a component of a structure different from the target site in the frame image is reduced; a second processing unit that performs a second process in which positions within the target portion are associated with each other among the plurality of frame images; and an analysis unit that analyzes a radiation moving image including the plurality of frame images that have been subjected to the first processing and the second processing.

Description

Technical Field

[0001] This invention relates to a radiation image analysis apparatus, a radiation image processing method, and a computer-readable, non-transitory recording medium storing a radiation image processing program. Background Technology

[0002] A radiation image analysis apparatus is known to perform dynamic analysis on a dynamic radiation image consisting of multiple frames captured by irradiating a patient's body part with radiation. For example, the radiation image analysis apparatus performs dynamic analysis on a dynamic radiation image obtained by photographing a patient's chest. In this case, because the lungs undergo movements such as breathing, the radiation image analysis apparatus performs accurate dynamic analysis by performing a process called registration across the multiple frames.

[0003] For example, in the device disclosed in Patent Document 1, a first image and a second image obtained by photographing the object part of the patient are arranged in a template space containing standard shape information of the object part, and the first image and the second image arranged in the template space are aligned.

[0004] Patent Document 1: Japanese Patent Application Publication No. 2023-184215

[0005] However, if, as disclosed in Patent Document 1, the image obtained by photographing the patient's chest is deformed to conform to the template space, not only the lung field of the analyzed object, but also the pixel values ​​of the thickness portions before and after the lung field will change. If the pixel values ​​of the thickness portions before and after the lung field also change, it may affect the analysis results of dynamic analysis, resulting in inaccurate analysis results. For example, as an analysis result of dynamic analysis, when acquiring blood flow in the blood vessels of the lung field, even if the blood flow is actually reduced, a higher blood flow value may be obtained. Summary of the Invention

[0006] The purpose of this invention is to provide a radiation image analysis apparatus, a radiation image processing method, and a computer-readable non-transitory recording medium storing a radiation image processing program that can obtain accurate analysis results even when performing registration between frame images.

[0007] The radiation image analysis apparatus of the present invention comprises:

[0008] The first processing unit performs a first processing, in which the components of structures different from the object part are reduced in each of the multiple frame images contained in the radiographic dynamic image of the object part of the subject.

[0009] The second processing unit performs a second processing step, in which a correspondence is established between the positions of the object region among the plurality of frame images; and

[0010] The analysis unit analyzes the aforementioned dynamic image of radiation, which includes multiple frame images that have undergone the first and second processing described above.

[0011] The radiation image processing method involved in this invention,

[0012] In a radiation image analysis device

[0013] A first processing is performed, in which the components of structures different from the object part are reduced in each of the multiple frame images contained in the radiographic dynamic image of the object part of the subject.

[0014] A second process is performed, in which a correspondence is established between the positions of the object region across the plurality of frame images; and

[0015] The aforementioned dynamic image of radiation, which includes multiple frames that have undergone the first and second processing described above, is analyzed.

[0016] The present invention relates to a computer-readable, non-transitory recording medium storing a radiation image processing program.

[0017] The computer of the radiation image analysis device executes:

[0018] In the first process, for each of the multiple frames in the radiographic dynamic image of the object part of the subject, the components of structures different from the object part in the frame image are reduced.

[0019] The second process involves establishing a correspondence between the positions of the object region across the plurality of frame images; and

[0020] The analysis process involves analyzing the aforementioned dynamic image of radiation, which includes multiple frame images that have undergone the first and second processing described above.

[0021] According to the present invention, even when performing registration between frame images, accurate analysis results can be obtained. Attached Figure Description

[0022] Figure 1 This is a diagram illustrating an example of the structure of a radiation image processing system according to an embodiment of the present invention.

[0023] Figure 2This is a block diagram illustrating an example of the functional structure of the imaging control unit in a radiation image imaging system that constitutes a radiation image processing system.

[0024] Figure 3 This is a block diagram illustrating an example of the functional structure of a radiation imaging control device that constitutes a radiation image processing system.

[0025] Figure 4 This is a block diagram illustrating an example of the functional structure of a radiation image analysis device that constitutes a radiation image processing system.

[0026] Figure 5 This is a functional block diagram illustrating the radiation image processing method implemented by the radiation image analysis device.

[0027] Figure 6 This is a diagram that briefly illustrates the existing registration process.

[0028] Figure 7 This is a brief explanation. Figure 5 The diagram shows the registration process in the radiation image processing method.

[0029] Figure 8 This is a graph representing the results of blood flow analysis obtained by dynamically analyzing dynamic images using existing registration processing.

[0030] Figure 9 It means that the use of Figure 5 The figure shows the blood flow analysis results obtained by dynamically analyzing the dynamic image obtained from the registration process in the radiation image processing method.

[0031] Figure 10 This is an explanation Figure 5 The diagram shows a functional block diagram of a variation of the volume thickness reduction processing in the radiographic image processing method.

[0032] Explanation of reference numerals in the attached figures: 1…Radiation image processing system; 10…Radiation image capturing system; 11…Capturing control unit; 12…Radiation irradiation unit; 13…Capturing stage; 14…Radiation detection unit; 15…Display unit; 16…Sound output unit; 20…Radiation capturing control device; 21…Control unit; 22…Storage unit; 23…Operation unit; 24…Display unit; 25…Communication unit; 26…Bus; 30…Radiation image analysis device; 31…Control unit; 32…Storage unit; 33…Operation unit; 34…Display unit; 35…Communication unit; 36…Bus; 40…Image management device; 50…Client terminal; 60…Radiation information terminal; 111…Setting information acquisition unit; 112…Capturing condition determination unit; 113…Image generation unit; 114…Storage unit; 311…Image acquisition unit; 312…First processing unit; 313…Second processing unit; 314…Analysis unit. Detailed Implementation

[0033] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0034] <Structure of a Radiation Image Processing System>

[0035] Figure 1 This is a diagram illustrating the radiation image processing system 1 of this embodiment. The radiation image processing system 1 includes a radiation image capturing system 10, a radiation capturing control device (control console device) 20, a radiation image analysis device 30, an image management device 40, and a client terminal 50.

[0036] exist Figure 1 In the example shown, the radiographic imaging system 10 is located in the imaging room, and the radiographic imaging control device 20 is located in the operating room. The radiographic imaging system 10, the radiographic imaging control device 20, the radiographic image analysis device 30, the image management device 40, and the client terminal 50 are interconnected via a communication network N. For example, the communication network N may be a communication network conforming to the DICOM (Digital Image and Communications in Medicine) standard.

[0037] Additionally, a radiology information terminal 60 is connected to the communication network N. This terminal serves as a radiology information system that transmits information related to radiology examinations, such as patient examination order information, to the radiology image processing system 1. Examples of radiology information terminals 60 include RIS (Radiology Information System).

[0038] The radiographic imaging system 10, under the control of the radiographic imaging control device 20, performs the imaging of dynamic radiographic images (hereinafter, dynamic images), i.e., dynamic radiographic imaging (hereinafter, dynamic imaging). The radiographic imaging control device 20 controls the radiographic imaging system 10 based on examination order information, etc., sent from the radiographic information terminal 60. The dynamic images generated by the radiographic imaging system 10 undergo the processing described later in the radiographic imaging control device 20 and are then sent to the radiographic image analysis device 30. The radiographic image analysis device 30 performs dynamic analysis on the dynamic images. The dynamic images and the results of the dynamic analysis are sent to an image management device 40, which serves as a medical image management system, and are managed there. For example, the image management device 40 may be a PACS (Picture Archiving and Communication System). The dynamic images and the results of the dynamic analysis are sent to a client terminal 50 for viewing by doctors and other medical professionals.

[0039] In this embodiment, dynamic imaging refers to obtaining multiple frame images by repeatedly irradiating (pulsed irradiation) a subject with pulsed radiation (e.g., X-rays) at a predetermined frame rate. Furthermore, a moving image refers to a series of frame images obtained through dynamic imaging. Additionally, dynamic analysis refers to the analysis and processing performed on the moving images, which includes not only analyzing the movement of the subject based on the moving images, but also analyzing the moving images to enhance or weaken (remove) predetermined structures.

[0040] The radiation image capturing system 10, the radiation capturing control device 20, and the radiation image analysis device 30 are all computers that have a processor and a memory, and that perform the specified functions by reading the program stored in the memory and executing it.

[0041] [Radiation Imaging System 10]

[0042] like Figure 1 As shown, the radiation image acquisition system 10 includes: an acquisition control unit 11, a radiation irradiation unit 12, an acquisition stage 13, a radiation detection unit 14, a display unit 15, and a sound output unit 16.

[0043] The imaging control unit 11 obtains setting information related to dynamic imaging settings from the radiation imaging control device 20. Based on the setting information, the imaging control unit 11 sets the imaging conditions for dynamic imaging and controls the radiation irradiation unit 12 to irradiate the patient M (subject) with radiation to perform imaging. The imaging control unit 11 is composed of a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), etc.

[0044] The setting information refers to information related to the settings used for performing dynamic imaging of patient M. For example, the setting information may include at least one of several dynamic analyses that the radiographic image analysis device 30 can perform on the dynamic image. When multiple dynamic analyses are combined, the setting information may also include information related to that combination. The setting information is set by the operator of the radiographic image processing system 1, such as a radiography technician, in the radiographic imaging control device 20 described later.

[0045] The imaging conditions include various factors such as pulse rate, pulse width, pulse interval, number of frames per shot, radiation dose per unit time, and the patient M's physical condition (respiratory status, etc.). The pulse rate is the number of radiation irradiations per second, consistent with the frame rate of the image data. The pulse width is the radiation irradiation time for each irradiation. The pulse interval is the time from the start of one radiation irradiation to the start of the next, consistent with the time interval between multiple image data sets (frame interval). The imaging conditions are automatically determined by the imaging control unit 11 of the radiation imaging system 10 based on the setting information.

[0046] The radiation irradiation unit 12 is positioned opposite the radiation detection unit 14 fixed to the imaging stage 13. The radiation irradiation unit 12 irradiates radiation under the control of the imaging control unit 11.

[0047] The radiation detection unit 14 is composed of a semiconductor image sensor such as an FPD (Flat Panel Detector). The radiation detection unit 14 has a substrate on which multiple detection elements (pixels) arranged in a matrix are configured to detect radiation irradiated from the radiation irradiation unit 12 based on the intensity of the radiation, convert the detected radiation into electrical signals, and store them. Each pixel on the substrate is configured with a switching unit such as a TFT (Thin Film Transistor).

[0048] The radiation detection unit 14 controls the switching units of each pixel to read the electrical signals accumulated in each pixel based on the image reading conditions input from the radiation imaging control device 20, and outputs the intensity information of each pixel to the image generation unit 113. Image reading conditions include, for example, frame rate, frame interval, pixel size, and image size (matrix size). The frame rate is the number of frames acquired per second, consistent with the pulse rate. The frame interval is the time from the start of one image data acquisition operation to the start of the next frame image acquisition operation, consistent with the pulse interval.

[0049] The imaging control unit 11 and the radiation detection unit 14 are interconnected and exchange synchronization signals to synchronize the radiation irradiation action and the image reading action.

[0050] In this way, the radiation image capturing system 10 performs dynamic imaging of radiation images by having the radiation irradiation unit 12 irradiate radiation according to the control of the imaging control unit 11, and the radiation detection unit 14 generating image data based on the intensity of the irradiated radiation.

[0051] When performing dynamic imaging of patient M, display unit 15 and sound output unit 16 provide instructions to patient M regarding the appropriate posture, body condition, and breathing status. Display unit 15 is, for example, a CRT (Cathode Ray Tube), Liquid Crystal Display (LCD), or Electroluminescent (EL) display. Sound output unit 16 is, for example, a speaker. Sound output unit 16 provides instructions to patient M regarding body condition and breathing status, for example, by automatically emitting sound. Display unit 15 and sound output unit 16 can each provide the same instructions to patient M, or only one of them can provide the instructions.

[0052] Figure 2 This is a block diagram illustrating an example of the functional structure of the imaging control unit 11 in the radiation image imaging system 10, which constitutes the radiation image processing system 1. The imaging control unit 11 includes a setting information acquisition unit 111, an imaging condition determination unit 112, an image generation unit 113, and a storage unit 114.

[0053] The setting information acquisition unit 111 acquires setting information from the radiation imaging control device 20.

[0054] The imaging condition determination unit 112 determines the imaging conditions for performing dynamic imaging of patient M based on the setting information. Information indicating the correspondence between various dynamic analyses and imaging conditions suitable for each dynamic analysis is pre-stored in the storage unit 114. In addition, information indicating the correspondence between combinations of various dynamic analyses and imaging conditions suitable for that combination is also pre-stored in the storage unit 114. Regarding the dynamic analysis or combination of various dynamic analyses indicated by the setting information, the imaging condition determination unit 112 reads the information indicating the correspondence from the storage unit 114 and determines the imaging conditions by comparing it with the setting information.

[0055] Furthermore, in cases such as screening and emergency care, it may be impossible to set dynamic analysis as the initial information. In such cases, the imaging condition determination unit 112 determines the imaging condition by having the operator select at least one imaging condition from a plurality of predefined imaging conditions. Additionally, the imaging condition determination unit 112 allows the operator to select inspection order information and determines the imaging condition based on the selected inspection order information. Thus, if dynamic analysis cannot be set before dynamic imaging, it is set after dynamic imaging based on the operator's selected imaging conditions, and the dynamic analysis in the radiographic image analysis device 30 (analysis in the analysis unit 314 described later) is performed.

[0056] The image generation unit 113 performs dynamic imaging of the patient M based on the determined imaging conditions, generating multiple frames of radiation images. Specifically, the image generation unit 113 controls the operation of the radiation irradiation unit 12 and the radiation detection unit 14 based on the imaging conditions, and obtains intensity information related to the radiation intensity transmitted through the subject from the radiation detection unit 14 for each pixel, thereby generating image data.

[0057] As described above, the storage unit 114 stores in advance information indicating the correspondence between various dynamic analyses and shooting conditions suitable for each dynamic analysis, information indicating the correspondence between combinations of various dynamic analyses and shooting conditions suitable for such combinations, etc.

[0058] [Radiation imaging control device 20]

[0059] The radiation imaging control device 20 is, for example, a PC (Personal Computer), workstation, or other computer. The radiation imaging control device 20 can be as follows: Figure 1 The example shown is a desktop computer, but it could also be a portable computer, such as a laptop, tablet, etc.

[0060] The radiation imaging control device 20 receives inspection order information from the radiation information terminal 60 and other devices, and sends it to the radiation image imaging system 10, thereby controlling the dynamic imaging of the radiation image imaging system 10.

[0061] The examination order information includes respiratory-related instructions, patient information, examination information, imaging information, data attributes, and various information related to the upcoming dynamic imaging. Examination information includes the examination ID, the site of examination (e.g., chest, particularly the lungs or heart), and the type of analysis (e.g., ventilation analysis, pulmonary blood flow analysis, measurement of maximum ventilation). For example, examination order information is generated when a doctor or other personnel request a dynamic imaging of patient M from the radiographic image processing system 1.

[0062] Furthermore, the radiography control device 20 generates setting information indicating at least one of a variety of dynamic analyses that the radiography image analysis device 30 can perform, based on operator input. When multiple dynamic analyses are combined, the radiography control device 20 generates setting information indicating the combination of the multiple dynamic analyses. The operator performs an input operation to identify, for example, which dynamic analyses are combined by referring to the contents of an examination order, and generates the setting information accordingly. Alternatively, the operator may identify which dynamic analyses are combined based on information transmitted from a doctor or other means using other methods.

[0063] Figure 3 This is a block diagram illustrating an example of the functional structure of the radiation imaging control device 20 constituting the radiation image processing system 1. The radiation imaging control device 20 includes a control unit 21, a storage unit 22, an operation unit 23, a display unit 24, and a communication unit 25. The various components of the radiation imaging control device 20 are interconnected via a bus 26.

[0064] The radiation imaging control device 20 outputs the setting conditions set by the operator and the inspection order information obtained in advance from the radiation information terminal 60 to the radiation imaging system 10, thereby controlling the imaging process of the radiation imaging system 10. The radiation imaging control device 20 may also display dynamic images generated by the radiation imaging system 10 for operator confirmation.

[0065] The control unit 21 consists of a CPU and RAM, etc. In the control unit 21, the CPU reads the system program and various processing programs stored in the storage unit 22 according to the operation of the operation unit 23, unfolds them in the RAM, and controls the operation of each part of the radiation imaging control device 20 based on the unfolded program.

[0066] The storage unit 22 is composed of a non-volatile semiconductor memory, a hard disk, or the like. The storage unit 22 stores various programs executed by the control unit 21, parameters required for processing through program execution, and data such as processing results (e.g., animated images). Various programs are stored in the form of readable program code, and the control unit 21 executes actions according to that program code sequentially.

[0067] In addition, the storage unit 22 stores image reading conditions used for dynamic imaging. Furthermore, the storage unit 22 stores examination order information sent from the radiology information terminal 60, etc. When the radiology imaging control device 20 controls the dynamic imaging of the radiology image imaging system 10, it reads the image reading conditions and examination order information corresponding to the patient M from the storage unit 22 and sends them.

[0068] The operation unit 23 is an operation device such as a cursor key, a keyboard with numeric input keys and various function keys, a mouse or trackball, or a touch panel. The operation unit 23 generates indication signals based on the operator's input and outputs them to the control unit 21.

[0069] The display unit 24 is composed of display devices such as CRT, liquid crystal display, and organic EL display. The display unit 24 displays input instructions from the operation unit 23 and image data (moving images, etc.) generated by the radiographic imaging system 10, according to the display signal input from the control unit 21.

[0070] The communication unit 25 performs data transmission and reception with the radiation image capturing system 10, the radiation image analysis device 30, the radiation information terminal 60, etc.

[0071] [Radiation Image Analysis Device 30]

[0072] The radiation image analysis device 30 is, for example, a computer such as a PC or workstation. The radiation image analysis device 30 can also be a desktop computer or a portable computer, such as a laptop or tablet computer.

[0073] The radiation image analysis device 30 performs dynamic analysis on the dynamic images captured by the radiation image capturing system 10 based on the setting information set in the radiation capturing control device 20.

[0074] Figure 4 This is a block diagram illustrating an example of the functional structure of the radiation image analysis device 30 constituting the radiation image processing system 1. The radiation image analysis device 30 includes a control unit 31, a storage unit 32, an operation unit 33, a display unit 34, and a communication unit 35. The various components of the radiation image analysis device 30 are connected via a bus 36.

[0075] The control unit 31 is, for example, a computer with one or more hardware processors, consisting of a CPU and RAM. The radiation image processing program is stored in a computer-readable, non-transitory recording medium, and is transferred from this recording medium to the storage unit 32. In the control unit 31, the CPU, based on the operation of the operation unit 33, reads the system program and various processing programs stored in the storage unit 32 and expands them in RAM. Based on the expanded programs, it executes the operation control and dynamic analysis of each part of the radiation image analysis device 30. The control unit 31, for example, executes the radiation image processing program implementing the radiation image processing method described later, as a processing program.

[0076] The control unit 31 includes an image acquisition unit 311, a first processing unit 312, a second processing unit 313, and an analysis unit 314.

[0077] The image acquisition unit 311 acquires multiple frames of radiation images, i.e., dynamic images, generated by the radiation image capturing system 10 and the radiation capturing control device 20.

[0078] The first processing unit 312 performs the following first processing: for each of the multiple frame images contained in the dynamic image of the object part of the patient M, the components of structures in the frame image that are different from the object part of the patient M are reduced.

[0079] The components of structures that differ from the target area are those that overlap with the target area in the direction of radiation irradiation toward the patient M, and are not part of the target area. At this time, the first processing unit 312 can also perform identification processing to recognize components of the patient M other than the target area as structures that differ from the target area, and reduce the number of identified structural components.

[0080] The following explanation uses the component of volume thickness as an example of a component of a structure that differs from the object part. However, the components of a structure that differs from the object part are not limited to the component of volume thickness. For example, they may also include components of artificial objects.

[0081] As a first process, for example, the first processing unit 312 extracts the volume thickness component by calculating the representative value of the frame image in the time direction, and subtracts the volume thickness component in the frame image from the frame image. The details of the calculation process of the representative value in the time direction will be described later. As the volume thickness component, a representative value (e.g., the average, median, maximum or minimum value of the pixel values ​​in the time direction of the multiple frame images) is calculated for each fixed range in the time direction within the multiple frame images.

[0082] Furthermore, the first processing unit 312 may also perform logarithmic transformation processing of the frame image before the first processing, which includes the calculation of the representative value of the time direction.

[0083] After the first processing, the second processing unit 313 performs a second processing step: establishing a correspondence between each position within the object region across multiple frame images, so that the coordinates of each position after establishing the correspondence are consistent. Specifically, the second processing unit 313 ensures that the coordinates (x, y) of the corresponding positions become the same at each position across multiple frame images. The process of establishing a correspondence between each position within the object region across multiple frame images, so that the coordinates of each position after establishing the correspondence are consistent, is called registration processing.

[0084] As a second process, for example, the second processing unit 313 performs optical flow processing across multiple frame images to track each position within the object region and establish a correspondence for each position.

[0085] The analysis unit 314 performs dynamic analysis on a moving image containing multiple frame images that have undergone the first processing and the second processing described above, according to the setting information, and obtains the analysis results. For example, the analysis unit 314 analyzes based on signal changes in the multiple frame images. In this case, if the analysis unit 314 cannot analyze the moving image (cannot obtain analysis results), it determines that it cannot be analyzed.

[0086] The analysis unit 314 includes, for example, blood flow analysis mode, ventilation analysis mode, adhesion analysis mode, diaphragmatic movement analysis mode, and orthopedic related measurement mode, as types of dynamic analysis. Each mode is briefly explained below.

[0087] Blood flow analysis mode is a mode that visualizes signal changes within the lung fields that are synchronized with heart rate.

[0088] The ventilation analysis mode is a mode that extracts signal changes in the time direction within a specific time band, making the behavior of lung tissue during breathing visible.

[0089] Adhesion analysis mode is a mode that visualizes the degree of adhesion in tissues.

[0090] The diaphragm movement analysis model tracks the up-and-down movement of the diaphragm during respiration.

[0091] Plastic surgery-related measurement patterns, such as those used in the limbs, measure changes in the position of specified bones and display the trajectory of movement.

[0092] The storage unit 32 is composed of a non-volatile semiconductor memory, a hard disk, or the like. The storage unit 32 stores various programs executed by the control unit 31, parameters required for processing through program execution, and data such as processing results (moving images, analysis results, etc.). Various programs are stored in the form of readable program code, and the control unit 31 executes actions according to that program code sequentially.

[0093] In addition, the storage unit 32 stores a list of information related to each dynamic image generated by the radiographic imaging system 10, including patient information, examination information, and status (e.g., receiving, dynamic analysis in progress, analysis completed, etc.). Furthermore, the storage unit 32 stores analysis results corresponding to the dynamic images.

[0094] The operation unit 33 can be a pointing device such as a keyboard with cursor keys, numeric input keys, and various function keys, a mouse or trackball, or a touch panel. The operation unit 33 generates instruction signals based on the operator's input and outputs them to the control unit 31. Alternatively, the operation unit 33 may also have a touch panel on the display screen of the display unit 34. In this case, the instruction signals input via the touch panel are output to the control unit 31.

[0095] The display unit 34 is composed of display devices such as CRT, liquid crystal display, and organic EL display. The display unit 34 displays input instructions from the operation unit 33 and image data (moving images, analysis results, etc.) generated by the radiographic imaging system 10, according to the display signal input from the control unit 31.

[0096] The communication unit 35 performs data transmission and reception with the radiation imaging control device 20 and the image management device 40, etc.

[0097] [Radiation Image Processing Methods]

[0098] Figure 5 This is a functional block diagram illustrating the radiation image processing method implemented by the radiation image analysis device 30.

[0099] After the radiographic imaging system 10 and the radiographic imaging control device 20 capture dynamic images of the target area of ​​the patient M, the image acquisition unit 311 of the control unit 31 of the radiographic image analysis device 30 acquires the dynamic images. The multiple frames of radiographic images constituting the dynamic images at this stage are the following unprocessed images, referred to as the ground truth image G11 (processing B11).

[0100] The first processing unit 312 of the control unit 31 performs a logarithmic transformation on each truth image G11 to obtain a logarithmic image G12 after the logarithmic transformation (processes B12 and B13). In other words, the first processing unit 312 performs a logarithmic transformation on the truth image G11 to obtain the logarithmic image G12 before the time-direction representative value calculation processing and optical flow processing described below. Furthermore, the logarithmic transformation is not mandatory; subsequent processing can be performed even on the truth image G11, but performing a logarithmic transformation improves the accuracy of the subsequent processing and dynamic analysis.

[0101] The first processing unit 312 performs time-direction representative value calculation processing on each logarithmic image G12 to obtain a processed logarithmic image G13 (processing B14, B15). By processing the logarithmic image G12 through time-direction representative value calculation processing, an image equivalent to the composition (here, volume thickness composition) of a structure different from the object part can be obtained, that is, the processed logarithmic image G13.

[0102] Here, the temporal representative value calculation process calculates a representative value for each fixed range in the temporal direction across multiple logarithmic images G12, serving as a volume thickness component. The temporal representative value is, for example, a value extracted from a region having width in both the spatial and temporal directions, located at a point of interest within one of the multiple logarithmic images G12. As part of the temporal representative value calculation process, it is possible to use the median, average, maximum, and minimum pixel values ​​from regions of arbitrary width in both the temporal and spatial directions across the multiple logarithmic images G12 as representative values.

[0103] As a process for calculating the representative value of the time direction, for example, a three-dimensional Median filter can be used. Alternatively, the process in process B14 can be any other process that performs the same calculation as the aforementioned representative value of the time direction.

[0104] The processed logarithmic image G13 is preferably an image that contains as few blood flow components as possible and as many components of structures different from the target site as possible. To remove blood flow components as much as possible, the width in the time direction of the processing for calculating the representative value in the time direction is preferably set to an integer multiple of the heart rate cycle. On the other hand, to include as many components of structures different from the target site as possible, the width in the time direction is preferably minimized. Accordingly, the width in the time direction of the processing for calculating the representative value in the time direction is preferably set to a width equivalent to one heart rate cycle. This width in the time direction is equivalent to the time of a logarithmic image G12 containing a predetermined number of logarithmic images G12 of interest.

[0105] Furthermore, regarding the width in the spatial direction for calculating the representative value in the time direction, it is preferable to extend the width in the spatial direction so that the object is included within the region even when a structure different from the object moves within the frame image. This ensures that the object is included within the region even when a structure different from the object moves within the frame image, reducing the difference between the representative value and the actual image value.

[0106] The first processing unit 312 subtracts the corresponding processed logarithmic image G13 from each logarithmic image G12 to obtain a volume thickness reduction image G14 (processing B16, B17).

[0107] The aforementioned treatments B14 to B17 are the first treatments in this invention, namely, the body thickness reduction treatment.

[0108] The second processing unit 313 of the control unit 31 performs optical flow processing between the logarithmic images G12 to obtain the deformation field G15 (processing B18, B19). Specifically, the second processing unit 313 performs optical flow between adjacent logarithmic images G12 in the temporal direction. Furthermore, images that are adjacent in the temporal direction will be referred to as adjacent images thereafter. Moreover, for each small region, the second processing unit 313 calculates the corresponding points between adjacent logarithmic images G12 and calculates the motion vector to obtain the deformation field G15.

[0109] The second processing unit 313 performs registration processing on the multiple volume thickness reduction images G14 based on the deformation field G15 between the logarithmic images G12 corresponding to the adjacent volume thickness reduction images G14. That is, it establishes correspondence for each position within the object part across the multiple volume thickness reduction images G14, making the coordinates of each position consistent after the correspondence is established, and obtains the registration image G16 (processing B20, B21).

[0110] The above-described processes B18 to B21 are the second processes in this invention.

[0111] The analysis unit 314 of the control unit 31 performs dynamic analysis, such as blood flow analysis mode, set by the setting information on the multiple registered images G16 after the second processing, and obtains analysis results G17 (processing B22, B23) that visualize signal changes in the lung field synchronized with the heart rate.

[0112] Here, refer to Figure 6 as well as Figure 7 The existing registration process and the registration process in this invention will be described. Figure 6 This is a diagram that briefly illustrates the existing registration process. Figure 7 This is a brief explanation. Figure 5 The diagram shows the registration process in the radiation image processing method.

[0113] Previously, in the case of registration, for example, in the case of lung field registration, the captured frame image (the frame before deformation) was deformed to match the deformation reference frame for alignment (referencing...). Figure 6 (The left and central images in the document). For example, in Patent Document 1 mentioned above, the captured image is placed in a template space containing standard shape information of the object part for alignment.

[0114] If the existing registration described above is performed, not only the lung field of the analyzed object, but also the thickness of the lung field before and after it will be deformed, and the pixel values ​​of this part will also change (see reference). Figure 6(See the right image in the image). If the pixel values ​​of the thickness portion before and after the lung field also change, the components of the thickness portion (thickness component) may affect the dynamic analysis of the lung field, resulting in inaccurate analysis results.

[0115] Therefore, in the radiographic image processing method of this embodiment, before registration processing, a process is performed to remove the thickness portion outside the lung field of the analysis object, that is, a process is performed to remove the thickness component from the image. Therefore, the frame image after thickness component removal (the frame before deformation) is deformed to match the deformation reference frame, and alignment is performed (refer to...). Figure 7 In this case, because the thickness component is removed, the thickness portions before and after the lung field are not deformed, and the pixel values ​​of these portions do not change. As a result, even with registration processing, the influence of the thickness component on the dynamic analysis of the lung field can be suppressed, leading to more accurate analysis results.

[0116] Figure 8 This is a graph representing the blood flow analysis results obtained by dynamically analyzing dynamic images using existing registration processing. Additionally, Figure 9 It means that the use of Figure 5 The figure shows the blood flow analysis results obtained by dynamically analyzing the dynamic image obtained from the registration process in the radiation image processing method.

[0117] Previously, for example, in the case of lung field region registration, as mentioned above, the body thickness component could potentially affect the dynamic analysis of lung field regions, leading to inaccurate analysis results. For instance, when acquiring lung field blood flow through dynamic analysis, even in areas where blood flow is actually reduced, higher blood flow values ​​might be obtained, such as... Figure 8 As shown in the analysis results, the portion of blood flow reduction could not be visualized.

[0118] In contrast, in the radiographic image processing method of this embodiment, as described above, the volume thickness component is removed before registration processing to suppress its influence on the dynamic analysis of lung field portions. Therefore, for example, when acquiring lung field blood flow through dynamic analysis, if the portion actually has reduced blood flow, then... Figure 9 As shown in the analysis results, the portion of blood flow reduction can be visualized. Figure 9 The black area in the image (indicated by the lung field) allows for more accurate analysis results. By referring to these results, doctors can more accurately assess the condition of the lung fields.

[0119] Furthermore, this example illustrates a dynamic analysis of a blood flow analysis mode, yielding the resulting blood flow analysis results. However, it is not limited to this; for instance, a dynamic analysis of a gas exchange analysis mode can also be performed to obtain the gas exchange state (the behavioral state of lung tissue) as the analysis result.

[0120] As described above, the radiographic image analysis apparatus 30 includes a first processing unit 312, a second processing unit 313, and an analysis unit 314. The first processing unit 312 performs a first processing step: for each of the multiple frame images included in the dynamic image of the object region of the patient M, it reduces the components of structures in the frame image that differ from the object region. After the first processing step, the second processing unit 313 performs a second processing step: between the multiple frame images, it establishes a corresponding position within the object region. The analysis unit 314 analyzes the radiographic dynamic image including the multiple frame images that have undergone the first and second processing steps.

[0121] Thus, the first processing unit 312 performs a first processing step, reducing the body thickness component of patient M in each of the multiple frames contained in the dynamic image. Therefore, the second processing unit 313 can perform a second processing step, establishing a corresponding position within the target area among the multiple frames where the influence of the body thickness component has been suppressed. In other words, the radiographic image analysis apparatus 30 can acquire a dynamic image where the influence of the body thickness component has been suppressed. As a result, even after performing the second processing (registration processing), the analysis unit 314 can obtain more accurate analysis results by performing dynamic analysis of the dynamic image where the influence of the body thickness component has been suppressed.

[0122] This method is particularly preferred for patients M who are unable to hold their breath during imaging, such as patients M admitted to the ICU (Intensive Care Unit). Dynamic imaging of patients M in both natural and artificial respiration states, where they are unable to hold their breath, is performed. The aforementioned dynamic image processing and dynamic analysis are then conducted to obtain analysis results regarding the blood flow and gas exchange status of the patient M's lung fields. Based on the obtained analysis results, doctors can diagnose and understand the condition of the patient M's lung fields, such as the possibility of pulmonary embolism and respiratory status.

[0123] <Variation Example>

[0124] Figure 10 This is an explanation Figure 5 The diagram shows a functional block diagram of a variation of the volume thickness reduction process in the radiation image processing method. Volume thickness reduction processing can also be... Figure 10 The thickness reduction process shown is used to replace... Figure 5 The volume thickness reduction process in the radiation image processing method shown below. (Refer to the following...) Figure 10 The thickness reduction process in this modified example will be explained.

[0125] The first processing unit 312 performs blood vessel removal processing on each logarithmic image G12 to obtain a blood vessel removal dynamic image G31 (the blood vessel removal image in this invention) (processing B31, B32). For example, the first processing unit 312 extracts a portion of the lung field that has signal changes synchronized with the heart rate, identifies the extracted portion as a blood vessel, removes this portion from the logarithmic image G12, thereby performing blood vessel removal processing and obtaining the blood vessel removal dynamic image G31.

[0126] In addition to blood vessels, B31 processing can also remove bone tissue such as the clavicle and ribs. For example, in the logarithmic image G12, B31 identifies bone tissue such as the clavicle and ribs one by one by utilizing anatomical knowledge from two-dimensional X-ray imaging. Then, B31 calculates the signal of the identified bone tissue and removes (or weakens) only that signal, thereby removing bone tissue.

[0127] The first processing unit 312 performs optical flow processing between the vessel removal dynamic images G31 to obtain the deformation field G32 (corresponding information in this invention) (processing B33, B34). Specifically, the first processing unit 312 performs optical flow between adjacent vessel removal dynamic images G31, and for each small region, it calculates the corresponding point between adjacent vessel removal dynamic images G31 and calculates the motion vector to obtain the deformation field G32.

[0128] The first processing unit 312 performs registration processing on multiple logarithmic images G12 based on the deformation field G32 between the blood vessel removal dynamic images G31 corresponding to adjacent logarithmic images G12. That is, it establishes correspondence for each position within the target area across multiple logarithmic images G12 to obtain a dynamic image G33 of the body at rest (the image of the body at rest in this invention) (processing B35, B36).

[0129] The first processing unit 312 performs time averaging processing (processes B37 and B38) to calculate a time-averaged image G34 based on the dynamic image G33 of the body at rest. Blood flow components disappear from the time-averaged image G34.

[0130] The first processing unit 312 performs inverse registration processing (process B39) so that a time-averaged image G34 follows the movement of the logarithmic image G12 (e.g., the movement of the lungs in the logarithmic image G12). Through this inverse registration processing, an inverse-registered dynamic image G35 (the body thickness image in this invention) is obtained (process B40). The first processing unit 312 performs inverse registration processing, for example, based on the aforementioned deformation field G32, that is, it performs a process to establish a correspondence between a time-averaged image G34 and the logarithmic image G12. Thus, it is possible to create an inverse-registered dynamic image G35 (body thickness image) that corresponds to each frame image of the logarithmic image G12 (e.g., following the movement of the lungs in the logarithmic image G12) and is composed of frame images that do not contain blood flow components.

[0131] The first processing unit 312 subtracts the corresponding frame image in the inverse registration dynamic image G35 (body thickness image) from each frame image of the logarithmic image G12, thereby obtaining the body thickness reduction image G36 (processing B41, B42).

[0132] The processes B31 to B42 described above are the first processes in this modified example, namely, the thickness reduction process, which can obtain a thickness reduction image G36 that is equal to or better than the processes B14 to B17 described in the above embodiments.

[0133] Hereinafter, by performing the same processing as processes B18 to B23 described in the above embodiments, the radiation image analysis apparatus 30 is also able to acquire a dynamic image that suppresses the influence of the volume thickness component in this modified example. As a result, even when the second processing (registration processing) is performed, the analysis unit 314 can obtain more accurate analysis results by performing dynamic analysis of the dynamic image that suppresses the influence of the volume thickness component.

[0134] The above embodiments are merely specific examples illustrating the implementation of the present invention and should not be construed as limiting the scope of the invention. That is, the present invention can be implemented in various forms without departing from its spirit or main features.

[0135] For example, in the above embodiment, the radiation image analysis device 30 has a first processing unit 312, but it may also be configured such that the radiation imaging control device 20 has a first processing unit 312. In this case, the radiation imaging control device 20 performs first processing through the first processing unit 312, and the radiation image analysis device 30 performs second processing through the second processing unit 313.

Claims

1. A radiation image analysis device, comprising: The first processing unit performs a first processing, in which the components of structures different from the object part are reduced in each of the multiple frame images contained in the radiographic dynamic image of the object part of the subject. The second processing unit performs a second processing, in which a correspondence is established between the positions within the object region among the plurality of frame images; as well as The analysis unit analyzes the radiation dynamic image, which includes multiple frame images that have undergone the first and second processing.

2. The radiation image analysis device according to claim 1, wherein, The second processing unit performs registration processing, in which the coordinates of the corresponding positions of the multiple frame images that have undergone the second processing are made consistent.

3. The radiation image analysis device according to claim 1, wherein, The analysis unit performs analysis based on signal changes in multiple frames contained in the radiation dynamic image.

4. The radiation image analysis device according to claim 1, wherein, The analysis unit performs analyses related to the blood flow of the subject.

5. The radiation image analysis device according to claim 1, wherein, The target area is the lung field. The analysis unit processes the blood flow in the lung field.

6. The radiation image analysis apparatus according to claim 1, wherein, The target area is the lung field. The analysis unit performs analysis on the gas exchange status of the lung fields.

7. The radiation image analysis apparatus according to claim 1, wherein, As part of the first processing, the first processing unit subtracts from the frame image components containing structures in the frame image that are different from the object region.

8. The radiation image analysis apparatus according to claim 1, wherein, The components of the structure are those of the structure located at the position overlapping with the object region in relation to the direction of radiation irradiation toward the subject.

9. The radiation image analysis apparatus according to claim 1, wherein, The first processing unit performs identification processing to identify parts of the subject other than the object part as components of the structure, and reduces the identified components of the structure.

10. The radiation image analysis apparatus according to claim 1, wherein, The first processing unit subtracts the representative value in the time direction of the plurality of frame images.

11. The radiation image analysis apparatus according to claim 10, wherein, The representative value in the time direction is the average, median, maximum, or minimum value of the pixel values ​​of the multiple frame images in the time direction.

12. The radiation image analysis apparatus according to claim 1, wherein, The first processing unit subtracts a representative value for each fixed range in the time direction within the plurality of frame images.

13. The radiation image analysis apparatus according to claim 12, wherein, The representative value is extracted from a region with width in both the spatial and temporal directions, located at the point of interest within one of the plurality of frame images. The width in the time direction is equivalent to the time of a specified number of frames containing the one frame image.

14. The radiation image analysis apparatus according to claim 13, wherein, The width in the time direction is equivalent to the time of one heart rate cycle.

15. The radiation image analysis apparatus according to claim 14, wherein, The representative value for each fixed range in the time direction is the average, median, maximum, or minimum value in the time direction.

16. The radiation image analysis apparatus according to claim 1, wherein, Before performing the first processing, the first processing unit performs a logarithmic transformation of the frame image.

17. The radiation image analysis apparatus according to claim 7, wherein, The first processing unit performs: The process of removing blood vessels from the lung field of the subject in the frame image to obtain a blood vessel removal image; The system processes multiple blood vessel removal images, tracks the position within the target area and obtains corresponding information for that position. Based on this corresponding information, it processes the multiple frame images to establish the corresponding position within the target area, thereby obtaining the processing of images of a stopped body. Processing of calculating time-averaged images based on the images of the body at rest; Inverse registration processing involves establishing a correspondence between the time-averaged image and the multiple frame images to create a volume thickness image corresponding to the multiple frame images. as well as The process of subtracting the thickness image corresponding to the frame image from the frame image to obtain the thickness reduction image after subtracting the thickness component is performed.

18. The radiation image analysis apparatus according to claim 1, wherein, As a second process, the second processing unit covers the plurality of frame images, tracks the position within the object region, and establishes a correspondence for that position.

19. The radiation image analysis apparatus according to claim 18, wherein, The second processing unit establishes a correspondence between the positions within the object region through optical flow processing.

20. A method for processing radiation images, wherein, In a radiation image analysis device A first processing step is performed, in which the components of structures different from the object region are reduced in each of the multiple frame images contained in the radiographic dynamic image of the object region of the subject. A second process is performed, in which a correspondence is established between the positions within the object region across the plurality of frame images; as well as The radiation dynamic image, which includes multiple frame images that have undergone the first and second processing, is analyzed.

21. A computer-readable, non-transitory recording medium storing a radiographic image processing program, wherein, The radiation image processing program is executed by the computer of the radiation image analysis device: In the first process, for each of the multiple frame images contained in the radiographic dynamic image of the object part of the subject, the components of the frame image that are different from the object part are reduced. In the second process, a correspondence is established between the positions within the object region among the plurality of frame images; as well as The analysis process involves analyzing the dynamic image of radiation, which includes multiple frame images that have undergone the first and second processing.

Images