Dynamic analysis device, dynamic analysis system, program, and classification method

Abstract

To support diagnosis of a disease by using an analysis image obtained by analyzing a dynamic image.SOLUTION: A controller of an analysis apparatus generates an analysis image by analyzing a dynamic image of a chest, and classifies signal values in a lung field region in the analysis image into at least three or more clusters based on magnitudes of the signal values. Then, the control unit causes a display unit to display a cluster map in which a region of each cluster in the analysis image is color-coded and a feature amount for each region of the cluster.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a dynamic analysis device , motion a dynamic analysis system, a program and classification method to and is related thereto.

Background Art

[0002] Conventionally, for example, as a precise examination for diagnosing chest diseases such as pulmonary hypertension, a chest CT image examination has been performed. In the image examination of pulmonary hypertension, for example, enlargement of the pulmonary trunk and narrowing of the peripheral pulmonary vessels are confirmed. That is, the diagnosis of pulmonary hypertension is made by observing changes for each structure (for example, large blood vessels, medium blood vessels, small blood vessels) in the lung field area in the chest CT image.

[0003] However, the CT examination has a problem of radiation exposure. Therefore, studies have been conducted to replace the CT examination with an examination using dynamic imaging with low radiation exposure. For example, Patent Document 1 describes generating a blood flow analysis image having a blood flow feature amount as a signal value and a ventilation analysis image having a ventilation feature amount as a signal value from frame images of dynamic images obtained by dynamic imaging.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the signal value of the analysis image generated by the technique described in Patent Document 1 is a relative signal value based on the signal value of a certain frame image of the dynamic image. Therefore, it has been difficult to compare between patients or to set a threshold value for the signal value to identify the structure area in the lung field area, and it has not been utilized for disease diagnosis.

[0006] The objective of this invention is to support the diagnosis of diseases using analyzed images obtained by analyzing dynamic images. [Means for solving the problem]

[0007] To solve the above problems, the dynamic analysis device according to the present invention is A motion analysis device that performs analysis on motion images obtained by performing motion imaging of a subject using radiation, Frame images constituting the aforementioned dynamic image The signal value of each pixel is generated based on a relative value of the signal value of the corresponding pixel in the reference frame image in the motion image. The system includes a classification means for classifying signal values ​​within a target region in a blood flow analysis image into at least three or more clusters based on the distribution of signal values ​​within the target region, The classification means calculates an initial cluster center for each of the three or more clusters, assigns the signal value within the target area to the cluster closest to each cluster center based on its distance, and assigns the signal value The average position Cluster center of each of the three or more clusters mentioned above as Recalculate. [Effects of the Invention]

[0008] According to the present invention, it is possible to support the diagnosis of diseases using analyzed images obtained by analyzing dynamic images. [Brief explanation of the drawing]

[0009] The advantages and features provided by one or more embodiments of the present invention will be better understood from the following detailed description and accompanying drawings. However, these drawings are for illustrative purposes only and are not intended to limit the scope of the present invention. [Figure 1] This figure shows an example of the overall configuration of the dynamic analysis system according to this embodiment. [Figure 2] This flowchart shows the shooting control process performed by the control unit of the shooting console shown in Figure 1. [Figure 3]It is a flowchart showing the analysis process executed by the control unit of the analyzer in FIG. 1. [Figure 4] It is a diagram schematically showing the branching of the pulmonary artery. [Figure 5A] It is a diagram showing how the signal values on the histogram are classified into clusters. [Figure 5B] It is a diagram showing how the signal values on the histogram are classified into clusters. [Figure 5C] It is a diagram showing how the signal values on the histogram are classified into clusters. [Figure 5D] It is a diagram showing how the signal values on the histogram are classified into clusters. [Figure 5E] It is a diagram showing how the signal values on the histogram are classified into clusters. [Figure 5F] It is a diagram showing how the signal values on the histogram are classified into clusters. [Figure 6A] It is a diagram showing an example of the cluster map of normal cases and the feature amounts of each cluster region. [Figure 6B] It is a diagram showing an example of the cluster map of the lung with an ischemic site in the peripheral blood vessels and the feature amounts of each cluster region.

Embodiments for Carrying Out the Invention

[0010] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. Note that the present invention is not limited to the illustrated examples.

[0011] 〈Configuration of Dynamic Analysis System 100〉 FIG. 1 shows an example of the overall configuration of the dynamic analysis system 100 in the present embodiment. As shown in FIG. 1, the dynamic analysis system 100 includes a photographing device 1, a photographing console 2, and an analysis device 3 (dynamic analysis device). The photographing device 1 and the photographing console 2 are connected by a communication cable or the like. The photographing console 2 and the analysis device 3 are connected via a communication network NT such as a LAN (Local Area Network). It is also possible that a RIS (Radiology Information Systems), HIS (Hospital Information Systems), an electronic medical record system, a PACS (Picture Archiving and Communication System), etc. not shown in the figure are connected to the communication network NT. Each device constituting the dynamic analysis system 100 conforms to the DICOM (Digital Image and Communications in Medicine) standard. The communication between the above-mentioned devices is carried out in accordance with DICOM.

[0012] <Configuration of the photographing device 1> The photographing device 1 is, for example, a photographing unit that photographs the dynamic state of the chest having periodicity (cycle), such as the morphological changes of the lungs during inhalation and exhalation and the pulsation of the heart. Dynamic photography refers to obtaining a plurality of images of a subject by irradiating the subject with radiation such as X-rays in pulses. Pulse irradiation means irradiating the radiation in a pulsed manner at a predetermined time interval. Or, dynamic photography refers to obtaining a plurality of images of a subject by continuously irradiating the subject with radiation. Continuous irradiation means irradiating the radiation at a low dose rate continuously without interruption. A series of images obtained by dynamic photography are called dynamic images. Also, each of the plurality of images constituting the dynamic image is called a frame image. Here, dynamic photography includes video photography, but does not include those that photograph still images while displaying video. Dynamic images include moving images, but do not include images obtained by photographing still images while displaying moving images. In the following embodiments, the case of performing dynamic photography by pulse irradiation will be taken as an example for explanation.

[0013] The radiation source 11 is positioned opposite the radiation detection unit 13, with the subject M, which is the area of ​​the subject being photographed, in between. The radiation source 11 irradiates the subject M with radiation (X-rays) according to the control of the radiation irradiation control device 12. The radiation irradiation control device 12 is connected to the imaging console 2. The radiation irradiation control device 12 controls the radiation source 11 to perform radiography based on the radiation irradiation conditions input from the imaging console 2. The radiation irradiation conditions input from the imaging console 2 include, for example, the pulse rate, pulse width, pulse interval, number of imaging frames per scan, X-ray tube current value, X-ray tube voltage value, and additional filter type. The pulse rate is the number of radiation irradiations per second and is the same as the frame rate, which will be described later. The pulse width is the radiation irradiation time per radiation irradiation. The pulse interval is the time from the start of one radiation irradiation to the start of the next radiation irradiation and is the same as the frame interval, which will be described later.

[0014] The radiation detection unit 13 is composed of a semiconductor image sensor such as an FPD (Flat Panel Detector). The FPD has, for example, a glass substrate. Multiple detection elements (pixels) are arranged in a matrix at predetermined positions on the glass substrate. Each pixel detects radiation irradiated from the radiation source 11 that has passed through at least the subject M according to its intensity, and converts the detected radiation into an electrical signal for storage. Each pixel is equipped with a switching unit such as a TFT (Thin Film Transistor). Note that FPDs can be of the indirect conversion type, which converts X-rays into electrical signals via a scintillator using a photoelectric conversion element, or the direct conversion type, which directly converts X-rays into electrical signals; either type may be used. The radiation detection unit 13 is positioned to face the radiation source 11 with the subject M in between.

[0015] The reading control device 14 is connected to the imaging console 2. Based on the image reading conditions input from the imaging console 2, the reading control device 14 controls the switching unit of each pixel of the radiation detection unit 13 to read the electrical signal accumulated in each pixel. In this way, the reading control device 14 acquires image data. This image data is a frame image. The reading control device 14 then outputs the acquired frame image to the imaging console 2. The image reading conditions are, for example, the frame rate, frame interval, pixel size, image size (matrix size), etc. The frame rate is the number of frame images acquired per second and is the same as the pulse rate. The frame interval is the time from the start of the acquisition operation of one frame image to the start of the acquisition operation of the next frame image and is the same as the pulse interval.

[0016] Here, the radiation irradiation control device 12 and the reading control device 14 are connected to each other and exchange synchronization signals to synchronize the radiation irradiation operation and the image reading operation.

[0017] <Configuration of shooting console 2> The imaging console 2 outputs radiation irradiation conditions and image reading conditions to the imaging device 1, controlling the radiography and image reading operations of the imaging device 1. The imaging console 2 also displays the dynamic images acquired by the imaging device 1 for the imaging technician or other operator to confirm positioning and whether the images are suitable for diagnosis. As shown in Figure 1, the imaging console 2 comprises a control unit 21, a storage unit 22, an operation unit 23, a display unit 24, and a communication unit 25. The various parts of the imaging console 2 are connected by a bus 26.

[0018] The control unit 21 is composed of a CPU (Central Processing Unit), RAM (Random Access Memory), etc. In response to operations by the operation unit 23, the CPU of the control unit 21 reads the system program and various processing programs stored in the memory unit 22, loads them into the RAM, and executes various processes, including the imaging control process described later, according to the loaded programs. In this way, the control unit 21 centrally controls the operation of each part of the imaging console 2, as well as the radiation irradiation and reading operations of the imaging device 1.

[0019] The storage unit 22 is composed of non-volatile semiconductor memory, a hard disk, or the like. The storage unit 22 stores data such as various programs executed by the control unit 21, parameters necessary for executing program processing, or processing results. For example, the storage unit 22 stores a program for executing the imaging control processing shown in Figure 2. The storage unit 22 also stores radiation irradiation conditions and image reading conditions corresponding to the imaging area and imaging direction. Various programs are stored in the storage unit 22 in the form of readable program code. The control unit 21 sequentially executes operations according to the program code.

[0020] The operation unit 23 includes a keyboard equipped with cursor keys, number input keys, and various function keys, and a pointing device such as a mouse. The operation unit 23 outputs instruction signals input by key operations on the keyboard or mouse operations to the control unit 21. The operation unit 23 may also be equipped with a touchscreen on the display screen of the display unit 24. In this case, the operation unit 23 outputs instruction signals input via the touchscreen to the control unit 21.

[0021] The display unit 24 is comprised of monitors such as LCDs (Liquid Crystal Displays) and CRTs (Cathode Ray Tubes). The display unit 24 displays input instructions and data from the operation unit 23 according to the instructions of the display signals input from the control unit 21.

[0022] The communication unit 25 includes a LAN adapter, modem, TA (Terminal Adapter), etc. The communication unit 25 controls data transmission and reception between the imaging console 2 and each device connected to the communication network NT.

[0023] <Configuration of Analysis Device 3> The analysis device 3 acquires dynamic images from the imaging console 2, analyzes the acquired dynamic images, and displays the analysis results. The analysis device 3 corresponds to the dynamic analysis device and display device of the present invention. In this embodiment, the analysis device 3 generates a blood flow analysis image based on the dynamic image of the chest, and classifies and displays the lung field region of the blood flow analysis image into large vessel region, medium vessel region, small vessel region, and no blood flow region. As shown in Figure 1, the analysis device 3 is configured to include a control unit 31, a storage unit 32, an operation unit 33, a display unit 34, and a communication unit 35, with each unit connected by a bus 36.

[0024] The control unit 31 is composed of a CPU, RAM, etc. The CPU of the control unit 31 reads the system program and various processing programs stored in the memory unit 32 in response to operations by the operation unit 33, expands them into RAM, and centrally controls the operation of each part of the analysis device 3 according to the expanded programs. In addition, the CPU of the control unit 31 executes various processes, including the analysis process described later, in cooperation with the programs stored in the memory unit 32. The control unit 31 functions as a classification means, calculation means, display control means, and estimation means.

[0025] The storage unit 32 is composed of non-volatile semiconductor memory, a hard disk, or the like. The storage unit 32 stores data such as various programs, parameters necessary for executing processing by the programs, or processing results. The programs stored in the storage unit 32 include programs for the control unit 31 to execute analysis processing. These various programs are stored in the storage unit 32 in the form of readable program code. The control unit 31 sequentially executes operations according to the program code. Furthermore, the memory unit 32 stores the motion images received from the shooting console 2, associating them with their associated information.

[0026] The operation unit 33 is configured with a keyboard equipped with cursor keys, number input keys, and various function keys, and a pointing device such as a mouse. The operation unit 33 outputs instruction signals input by key operations on the keyboard or mouse operations to the control unit 31. The operation unit 33 may also be equipped with a touchscreen on the display screen of the display unit 34. In this case, the operation unit 33 outputs instruction signals input via the touchscreen to the control unit 31.

[0027] The display unit 34 is composed of a monitor such as an LCD or CRT. The display unit 34 displays various information according to the instructions of the display signals input from the control unit 31. The display unit 34 functions as a presentation means.

[0028] The communication unit 35 includes a LAN adapter, modem, TA, etc. The communication unit 35 controls data transmission and reception between the analysis device 3 and each device connected to the communication network NT.

[0029] <Operation of Dynamic Analysis System 100> Next, the operation of the above-mentioned dynamic analysis system 100 will be described.

[0030] (Operation of imaging device 1 and imaging console 2) First, we will explain the shooting operation using the shooting device 1 and the shooting console 2. Figure 2 shows the shooting control process performed in the control unit 21 of the shooting console 2. The shooting control process is performed in cooperation with the control unit 21 and the program stored in the storage unit 22.

[0031] First, the control unit 21 receives patient information and examination information input by the operator performing the imaging via the operation unit 23 (step S1). Patient information refers to information about the subject being examined. Patient information includes patient ID, name, age, sex, height, weight, etc. Examination information includes examination ID, examination date, imaging site, imaging direction, etc. In this embodiment, the imaging site is the chest. Patient information and test information may also be obtained from a RIS or HIS (not shown) via the communication unit 25.

[0032] Next, the control unit 21 reads the radiation irradiation conditions from the storage unit 22 based on the input patient information and examination information and sets them in the radiation irradiation control device 12. The control unit 21 also reads the image reading conditions from the storage unit 22 and sets them in the reading control device 14 (step S2).

[0033] Next, the control unit 21 waits for an instruction to irradiate (step S3). At this point, the person performing the imaging positions the subject M between the radiation source 11 and the radiation detection unit 13, and when the imaging preparation is complete, they operate the operation unit 23 to input an instruction to irradiate.

[0034] When a radiation irradiation instruction is input by the operation unit 23 (step S3; YES), the control unit 21 outputs an instruction to start imaging to the radiation irradiation control device 12 and the reading control device 14, and starts dynamic imaging (step S4). That is, the control unit 21 irradiates the subject with radiation from the radiation source 11 at pulse intervals set in the radiation irradiation control device 12, and the radiation detection unit 13 acquires frame images. Before or during dynamic imaging, the person performing the imaging instructs the subject on their breathing state. In this embodiment, the subject is instructed to hold their breath in order to suppress movement caused by respiratory motion. The imaging device 1 may also output audio and / or a display instructing the breathing state.

[0035] When the operation unit 23 inputs a signal to end radiation irradiation, the control unit 21 outputs a signal to end imaging to the radiation irradiation control device 12 and the reading control device 14, and stops the imaging operation.

[0036] The frame images of the motion captured by the camera are sequentially input to the shooting console 2. The control unit 21 associates a number indicating the shooting order (frame number) with the input frame images and stores it in the storage unit 22 (step S5). The control unit 21 also displays the input frame images on the display unit 24 (step S6). The person performing the imaging checks the positioning and other aspects based on the displayed dynamic image and determines whether an image suitable for diagnosis was obtained (imaging OK) or whether re-imaging is necessary (imaging NG). Then, the person performing the imaging operates the control unit 23 to input the result of the determination.

[0037] When a judgment result indicating OK for shooting is input by a predetermined operation of the operation unit 23 (step S7; YES), the control unit 21 attaches an identification ID for identifying the dynamic image, patient information, examination information, radiation irradiation conditions, image reading conditions, and a number indicating the shooting order (frame number) as supplementary information to each of the series of frame images acquired by dynamic imaging. The control unit 21 transmits the series of frame images with the supplementary information attached to the analysis device 3 via the communication unit 25 (step S8). Then, the control unit 21 terminates the shooting control process. On the other hand, if a judgment result indicating that shooting is not possible is input through a predetermined operation of the operation unit 23 (step S7; NO), the control unit 21 deletes the series of frame images stored in the storage unit 22 (step S9). Then, the control unit 21 terminates the shooting control process. In this case, reshooting is required.

[0038] (Operation of analysis device 3) Next, we will explain the operation of the analysis device 3. In the analysis device 3, for example, when a motion image is received from the imaging console 2 via the communication unit 35, the control unit 31 associates the received motion image with associated information and stores it in the storage unit 32. When the operation unit 33 instructs the control unit 31 to perform analysis processing on the motion image of the chest stored in the storage unit 32, the control unit 31 executes the analysis processing shown in Figure 3. The analysis processing is performed in cooperation with the control unit 31 and the program stored in the storage unit 32. The analysis processing will be described below with reference to Figure 3.

[0039] First, the control unit 31 analyzes the dynamic image and generates a blood flow analysis image (step S11). Blood flow analysis images can be generated, for example, by the following processes (1) to (4).

[0040] (1) First, the control unit 31 extracts the lung region from each frame image of the dynamic image. Any known method can be used for extracting the lung region. For example, the control unit 31 obtains a threshold from the histogram of the pixel values ​​(signal values) of each pixel in the frame image by discriminant analysis, and extracts regions with higher signals than this threshold as candidate lung regions. Next, edge detection is performed near the boundary of the primary extracted candidate lung region, and the boundary of the lung region can be extracted by extracting the point where the edge is maximized in the small block near the boundary along the boundary.

[0041] (2) Next, the control unit 31 determines a reference frame image. The reference frame image can be, for example, a frame image corresponding to the end of ventricular diastole of the heart. Here, when blood flows into an organ, the transmission of radiation is obstructed by the blood flow. As a result, the amount of radiation transmitted in the radiographic image decreases due to the blood flow, the pixel value (signal value) becomes smaller, and the radiographic image appears whitish. When this is viewed in terms of the relationship between cardiac blood flow flowing into and out of the heart and pulmonary blood flow flowing into and out of the lungs, when the heart is in ventricular diastole and a lot of blood is flowing into the heart, the transmission of radiation is obstructed in the cardiac area, so the signal value on the radiographic image is small and it appears relatively whitish. In contrast, at this timing, there is little blood flow flowing into the lungs, and the amount of radiation transmitted in the lungs increases, so the signal value on the radiographic image becomes larger and it appears relatively dark. In other words, the frame image corresponding to the end of ventricular diastole of the heart is a frame image that shows the state in which pulmonary blood flow is at its lowest. Therefore, for example, the control unit 31 sets an ROI in the ventricular region of the heart for each frame of the dynamic image, and determines the frame image with the minimum ROI signal value (e.g., mean value) as the frame image corresponding to the end of ventricular diastole (reference frame image).

[0042] (3) Next, the control unit 31 calculates the ratio (D' / D) between the signal value difference D' between the frame image and the corresponding pixel in the reference frame image and the signal value D of the corresponding pixel in the reference frame image for each pixel in the lung field region of each frame image. The corresponding pixel in the reference frame image is the pixel located at the same position in the reference frame image. The ratio (D' / D) is a value that represents the rate of change of the signal value of the pixel in each frame image relative to the signal value of the corresponding pixel in the reference frame image. This D' / D is a characteristic quantity that indicates the blood flow in the lungs.

[0043] (4) Next, the control unit 31 aggregates the representative values ​​(maximum absolute values) of multiple ratios (D' / D) calculated for pixels at the same position in each frame image into a single image to generate a blood flow analysis image.

[0044] The control unit 31 may divide the lung field region of each frame image into small blocks consisting of multiple pixels and perform binning, which replaces each signal value within a small block with a representative value (such as the mean value). The control unit 31 may then treat each small block as one pixel and perform the processes described in (3) and (4) above. Furthermore, for example, when dynamic imaging is performed under resting respiration, the control unit 31 may acquire the waveform of the time change of signal value for each pixel in the lung field region of the dynamic image, apply a high-pass filter or band-pass filter in the time direction to extract only the signal component of the heart rate cycle, and then perform the processes described in (3) and (4) above.

[0045] Next, the control unit 31 generates a histogram (probability distribution) of the signal values ​​of the blood flow analysis image (step S12). Specifically, the control unit 31 generates a histogram in which the horizontal axis represents the signal values ​​within the lung field region of the blood flow analysis image and the vertical axis represents the number of pixels having those signal values.

[0046] Next, the control unit 31 normalizes the generated histogram (step S13). For example, it normalizes the range of the horizontal axis (minimum to maximum value) which differs from patient to patient.

[0047] Next, the control unit 31 clusters the signal values ​​from the generated histogram using statistical methods (step S14). For example, the control unit 31 uses the k-means method to classify the signal values ​​within the lung field region into three or more clusters based on their type (in this case, magnitude).

[0048] Here, pulmonary artery blood flow is the flow of blood based on the heartbeat, and the pulmonary artery vessels are deformed by the blood flow caused by the heartbeat. In the signal value (D' / D) of each pixel in the blood flow analysis image, D corresponds to the diameter of the vessel before deformation due to blood flow. In the signal value (D' / D) of each pixel in the blood flow analysis image, D' corresponds to the maximum value of the vessel diameter after deformation due to blood flow. If the elastic modulus of the vessel is K and the intrapulmonary artery pressure is P, then D' / D = K × P. As shown in Figure 4, the pulmonary artery branches and narrows in a geometric progression, and the magnitude of the D' / D value is thought to be a value corresponding to the number of branches. For example, if the signal value (D' / D) of the pulmonary artery trunk in the blood flow analysis image is S, then in a vessel with normal blood flow, the signal value of a vessel that branches once from the pulmonary artery trunk into two will be S / 2, the signal value of a vessel that branches twice into four will be S / 4, and so on. Therefore, in this embodiment, the lung field region of the blood flow analysis image is classified into four clusters based on the magnitude of the signal value. This classifies the lung field region into areas with signal values ​​corresponding to large vessels, areas with signal values ​​corresponding to medium vessels, areas with signal values ​​corresponding to small vessels, and areas with signal values ​​corresponding to no blood flow. Large vessels include, for example, the pulmonary artery trunk and the left and right pulmonary arteries that branch off from it. Small vessels include, for example, capillaries and peripheral vessels. Medium vessels include, for example, the vessels between large and small vessels.

[0049] In step S14, the control unit 31 clusters the signal values ​​of the blood flow analysis image by the following processes (I) to (IV). Figures 5A to 5F show how the signal values ​​on the histogram are classified into clusters. Figures 5A to 5F show an example where the signal values ​​on the histogram are classified into three clusters, clusters A to C. (I) First, the control unit 31 randomly assigns a cluster to each point on the histogram (see Figure 5A). (II) Next, the control unit 31 calculates the average position (cluster center) of the points assigned to each cluster (see the star in Figure 5A). (III) Next, the control unit 31 calculates the distance of each point from the center of each cluster and reassigns it to the cluster closest to it (see Figure 5B). (IV) Repeat steps (II) and (III) above until the assigned cluster no longer changes (see Figures 5C to 5F).

[0050] Note that in methods (I) to (IV) above, clusters are assigned randomly in (I), so the results may depend on the initial values. Therefore, the cluster centers may be found by the following processes (a) to (d) (k-means++ method). This reduces the dependence on the initial values. (a) First, the control unit 31 controls each point x on the histogram i One point is randomly selected from the list and designated as the center of the cluster. (b) Next, the control unit 31 controls each point x on the histogram i Regarding this, we calculate the distance D(x) from the cluster center. (c) Weighted probability distribution (D(x)) for each point 2 / ΣD(x) 2 Use this to select a new cluster center. (d) Repeat steps (b) and (c) above until k cluster centers (in this case, k=4) are selected.

[0051] As mentioned above, the signal value of each pixel in the blood flow analysis image is a relative signal value based on the signal value of the reference frame image. Therefore, it is not possible to compare the magnitude of these signal values ​​between patients. Furthermore, even if a predetermined percentile value of the signal value within the target area is set as a threshold in order to divide the target area (lung field area) of the blood flow analysis image into areas of three or more structures based on the signal value, the threshold is not an absolute standard, making it impossible to identify each structure using a unified standard. On the other hand, in this embodiment, a probability distribution (histogram) of the signal value within the target area is obtained based on the type (magnitude) of the signal value within the target area, and the target area is classified into four clusters using cluster analysis, a statistical method, from the probability distribution. Thus, the target area of ​​the blood flow analysis image, which consists of relative signal values, can be classified into areas of large vessels, medium vessels, small vessels, and areas with no blood flow without variations in standards among patients.

[0052] Next, the control unit 31 labels each pixel within the lung field region of the blood flow analysis image based on the clustering results (step S15). Next, the control unit 31 calculates the area of ​​each cluster region and the ratio of each cluster region to the total area of ​​the lung field region (step S16). In this embodiment, the control unit 31 calculates the area of ​​each cluster by the number of pixels classified into each cluster. Next, the control unit 31 generates a cluster map by color-coding the lung field regions of the blood flow analysis image according to cluster (step S17). In step S17, the lung field regions of the blood flow analysis image should be displayed in a way that allows for identification of each cluster, and this is not limited to color coding.

[0053] Next, the control unit 31 displays the cluster map and the feature quantities of each cluster region on the display unit 34 (step S18). Then, the control unit 31 terminates the analysis process. Examples of feature quantities for each cluster region include the signal value at the cluster center, the area, and the ratio to the entire lung field region.

[0054] Figures 6A and 6B show examples of the cluster map 341 and feature quantities 342 for each cluster region displayed on the display unit 34 in step S18. Cluster 0 is the region of signal value corresponding to no blood flow (no blood flow region). Cluster 1 is the region of signal value corresponding to small blood vessels (small blood vessel region). Cluster 2 is the region of signal value corresponding to medium blood vessels (medium blood vessel region). Cluster 3 is the region of signal value corresponding to large blood vessels (large blood vessel region). Figure 6A shows the cluster map 341 and feature quantities 342 for each cluster region of a normal subject. Figure 6B shows the cluster map 341 and feature quantities 342 for each cluster region of a subject with ischemic areas in peripheral blood vessels. In Figures 6A and 6B, an example is shown in which the region outside the lung region in the cluster map 341 is displayed in black. The cluster map 341 may also be displayed superimposed on one frame image of the dynamic image.

[0055] From the cluster map 341 and the feature quantities 342 of each cluster region displayed on the display unit 34, the physician can obtain information on the shape, area, and proportion of the total area of ​​large vessels, medium vessels, small vessels, and areas with no blood flow in the lung field region. For example, in Figure 6B, an area with no blood flow is observed in an area where small vessels would normally be present, indicating that there is an ischemic area in the peripheral vessels. From the state of the lung field region in this way, the physician can diagnose whether or not there is a disease such as pulmonary hypertension or pulmonary embolism. The control unit 31 may also calculate feature quantities for each vessel region (each cluster region), such as by multiplying the signal value in the lung field region of the blood flow analysis image by the area of ​​the cluster region to calculate the flow velocity, and display them on the display unit 34 together with the cluster map 341, etc.

[0056] Furthermore, the control unit 31 may estimate the presence or absence of disease and / or the type of disease based on the regional pattern of each classified cluster. The type of disease is the disease name or the disease name and its classification (for example, pulmonary hypertension groups 1 to 5). For example, a machine learning model is created and stored in the memory unit 32 based on a large amount of training data that associates clustering results of signal values ​​in the lung field region of blood flow analysis images with the types of diseases diagnosed by physicians. The control unit 31 uses this machine learning model to estimate the presence or absence of a disease and / or the type of disease from the clustering results in step S14. The estimation results are then displayed on the display unit 34. This helps to make diagnosis by physicians easier. Alternatively, artificial intelligence (AI) without training data may be used to estimate the presence or absence of a disease and / or the type of disease from the clustering results in step S14.

[0057] (modified version) In the above embodiment, step S14 describes the case where clustering is performed using a statistical method (k-means method). Alternatively, clustering may be performed using AI such as a machine learning model. For example, a machine learning model is created and stored in the memory unit 32 to cluster the signal values ​​of the lung region in the blood flow analysis images based on image data from numerous blood flow analysis images in which the lung region has been classified by a physician into large vessel region, medium vessel region, small vessel region, and no blood flow region. Alternatively, a machine learning model is created and stored in the memory unit 32 to cluster the signal values ​​of the lung region in the blood flow analysis images based on image data from other modalities (CT, MRI, RI) in which the lung region has been classified into large vessel region, medium vessel region, small vessel region, and no blood vessel region. Then, in step S14, the control unit 31 may cluster the signal values ​​of the lung region in the blood flow analysis images into signal values ​​corresponding to large vessels, signal values ​​corresponding to medium vessels, signal values ​​corresponding to small vessels, and signal values ​​corresponding to no blood flow using the machine learning model stored in the memory unit 32.

[0058] As described above, the control unit 31 of the analysis device 3 classifies the signal values ​​within the lung field region in the blood flow analysis image generated by analyzing the dynamic image of the chest into at least three clusters based on the magnitude of the signal values. Therefore, the signal values ​​within the lung field region of blood flow analysis images can be classified into three or more clusters based on their size, allowing the lung field region to be classified into three or more structures. As a result, this can support physicians in diagnosing diseases using blood flow analysis images.

[0059] Furthermore, the control unit 31 calculates at least one of the area or ratio of the regions for each classified cluster and presents the calculation results. Thus, it is possible to provide a useful feature for each region of each cluster for physicians to diagnose diseases.

[0060] Furthermore, the control unit 31 displays the lung field regions of the blood flow analysis image on the display unit 34, color-coded according to the classified clusters. This allows physicians to intuitively recognize the distribution and size of the regions for each cluster and make a diagnosis.

[0061] Furthermore, the control unit 31 classifies the signal values ​​within the lung field region of the blood flow analysis image based on a probability distribution. Therefore, even relative signal values ​​can be classified with high accuracy.

[0062] Furthermore, the control unit 31 classifies the signal values ​​within the lung field region of the blood flow analysis image into signal values ​​corresponding to large vessels, signal values ​​corresponding to medium vessels, signal values ​​corresponding to small vessels, and signal values ​​corresponding to no blood flow. Therefore, it can assist physicians in diagnosing diseases using blood flow analysis images.

[0063] Furthermore, the control unit 31 classifies the signal values ​​within the lung field region of the blood flow analysis image into four or more clusters. Therefore, it can assist physicians in diagnosing diseases using the blood flow analysis image.

[0064] Furthermore, the control unit 31 estimates the presence or type of disease based on the classified cluster pattern. Therefore, it can help make disease diagnosis easier for physicians.

[0065] The description in the above embodiment refers to the dynamic analysis apparatus according to the present invention. , motionState analysis system, program and How to classify Law This is just one example, and not an exhaustive one. For example, in the above embodiment and its modifications, the case of clustering the signal values ​​of the lung field region of the blood flow analysis image into four groups was described as an example, but the analysis image to be processed may be other analysis images such as ventilation analysis images. As a ventilation analysis image, for example, an image obtained by subtracting the signal value of the corresponding pixel in the frame image at the maximum expiratory position from the signal value of the frame image at the maximum inspiratory position of a dynamic image of the chest taken during deep breathing or resting breathing can be applied.

[0066] Furthermore, although the above embodiment described an example in which the lung field region in the chest analysis image is divided into four regions, the imaging site, the target region for division, and the number of divisions are just examples and can be changed as appropriate.

[0067] Furthermore, in the above embodiment, the control unit 31 calculated and displayed both the area of ​​each classified cluster and its ratio to the entire target area, but it is also possible to calculate only one of them and have the display unit 34 display it.

[0068] Furthermore, while the above description discloses examples using hard disks, semiconductor non-volatile memory, etc., as computer-readable media for the program according to the present invention, the invention is not limited to these examples. Other computer-readable media include portable recording media such as CD-ROMs. Carrier waves can also be used as a medium for providing data for the program according to the present invention via a communication line.

[0069] While embodiments of the present invention have been described and illustrated in detail, the disclosed embodiments are illustrative and for illustrative purposes only and do not limit the present invention. The scope of the present invention should be construed by the appended claims. [Explanation of Symbols]

[0070] 100 Dynamic Analysis System 1. Imaging device 11 Radiation source 12. Radiation irradiation control device 13. Radiation detection unit 14. Reading control device 2. Shooting console 21 Control Unit 22 Memory section 23 Control section 24 Display 25 Communications Department 26 bus 3 Analysis device 31 Control Unit 32 Storage section 33 Operation section 34 Display section 35 Communications Department 36 bus

Claims

1. A motion analysis device that performs analysis on motion images obtained by performing motion imaging of a subject using radiation, The system includes a classification means that classifies the signal values ​​within a target region in a blood flow analysis image, which are generated based on the relative values ​​of the signal values ​​of each pixel in the frame images constituting the dynamic image, with respect to the signal values ​​of the corresponding pixels in a reference frame image in the dynamic image, into at least three or more clusters based on the distribution of signal values ​​within the target region, The classification means is a dynamic analysis device that calculates an initial cluster center for each of the three or more clusters, assigns the signal values ​​within the target area to the nearest cluster based on their distance from each cluster center, and recalculates the average position of the assigned signal values ​​as the cluster center for each of the three or more clusters.

2. The dynamic analysis apparatus according to claim 1, wherein the classification means repeats the assignment and the recalculation of the cluster centers until the assigned clusters no longer change.

3. The dynamic analysis apparatus according to claim 1, wherein the classification means randomly assigns clusters to signal values ​​within the target area, and calculates the cluster center for each cluster based on the assigned signal value.

4. The calculation of the initial cluster centers is performed based on the k-means++ method, as described in claim 1. A dynamic analysis device.

5. The dynamic analysis apparatus according to claim 1, wherein the distribution of signal values ​​within the target region is a histogram showing the distribution of pixels for each signal value within the target region.

6. The aforementioned dynamic image is a dynamic image of the chest. The dynamic analysis apparatus according to claim 1.

7. A calculation means for calculating at least one of the area or ratio of the region for each cluster classified by the classification means, A presentation means for presenting the calculation result by the calculation means, The dynamic analysis device according to claim 1, comprising:

8. Display control means for displaying the target region of the blood flow analysis image in a color-coded manner according to the classified clusters on the display unit. The dynamic analysis device according to claim 1, comprising:

9. The classification means classifies the signal values ​​within the target region into signal values ​​corresponding to large blood vessels, signal values ​​corresponding to medium blood vessels, signal values ​​corresponding to small blood vessels, and signal values ​​corresponding to no blood flow. The aforementioned medium-sized blood vessels are blood vessels that branch off from the aforementioned large blood vessels, and the aforementioned small blood vessels are blood vessels that branch off from the aforementioned medium-sized blood vessels. The dynamic analysis apparatus according to claim 1.

10. The classification means classifies the signal values ​​within the target area into four or more clusters. The dynamic analysis apparatus according to claim 1.

11. The dynamic analysis device according to claim 1, further comprising estimation means for estimating the presence or type of disease based on the classified cluster pattern.

12. A photographic device that performs motion imaging of a subject using radiation, A dynamic analysis device according to any one of claims 1 to 11, A dynamic analysis system equipped with the following features.

13. Computers A classification means that classifies the signal values ​​within a target region in a blood flow analysis image, which is generated based on the relative values ​​of the signal values ​​of each pixel in a frame image constituting a motion image obtained by performing motion imaging of a subject using radiation, with respect to the signal values ​​of the corresponding pixels in a reference frame image in the motion image, into at least three or more clusters based on the distribution of signal values ​​within the target region. To make it function as, The classification means is a program that calculates an initial cluster center for each of the three or more clusters, assigns the signal values ​​within the target area to the cluster closest to each cluster center based on their distance, and calculates the average position of the assigned signal values ​​as the cluster center for each of the three or more clusters.

14. Computers The blood flow analysis image, which is generated based on the relative values ​​of the signal values ​​of each pixel in the frame image constituting the motion image obtained by performing motion imaging of a subject using radiation, with respect to the signal values ​​of the corresponding pixels in a reference frame image in the motion image, has a classification step of classifying the signal values ​​within the target region in the blood flow analysis image into at least three or more clusters based on the distribution of signal values ​​within the target region. The classification step is a classification method that calculates an initial cluster center for each of the three or more clusters, assigns the signal values ​​within the target area to the cluster closest to each cluster center based on their distance, and recalculates the average position of the assigned signal values ​​as the cluster center for each of the three or more clusters.

Images