Processing device, processing program and processing method

Abstract

To provide a processing device, processing program, and processing method capable of further appropriately processing acquired medical imaging data.SOLUTION: A processing device including at least one processor is so configured that the at least one processor executes processing of: acquiring one or a plurality of region imaging data from medical imaging data obtained by imaging at least a part of a subject's body as a subject; regarding the acquired one or plurality of region imaging data, converting each region imaging data on the basis of parameter information corresponding to each region imaging data to acquire one or a plurality of converted images; and synthesizing the acquired one or plurality of converted images to acquire a composite image.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present disclosure relates to a processing device, a processing program, and a processing method used to process medical imaging data. [Background technology]

[0002] It has been known that, in the case of an image having a wide range of pixel values, such as a CT image, the shading in a specific region of interest cannot be sufficiently expressed, resulting in unclear contours, etc. Patent Document 1 describes "a medical image processing device, comprising a processing unit, wherein the processing unit acquires volume data of a subject, generates a first image representing at least a part of tissue in the subject based on the volume data, generates a contour line connecting a first point and a second point on a contour of the tissue on the first image based on at least one of a Laplacian and a gradient of each pixel of the first image, a pixel value of the second point, and pixel values ​​of each pixel on a candidate path connecting the first point and the second point, and displays the first image and the contour line on a display unit." [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-155758 Summary of the Invention [Problem to be solved by the invention]

[0004] Therefore, in consideration of the above-described techniques, an object of the present disclosure is to provide a processing device, a processing program, and a processing method that can more appropriately process acquired medical imaging data through various embodiments. [Means for solving the problem]

[0005] According to one aspect of the present disclosure, there is provided a processing device having at least one processor, wherein the at least one processor is configured to acquire one or more area imaging data from medical imaging data obtained by photographing at least a part of a subject's body as a subject, acquire one or more converted images by converting each area imaging data based on parameter information corresponding to each area imaging data for the acquired one or more area imaging data, and perform processing to acquire a composite image by combining the acquired one or more converted images.

[0006] According to one aspect of the present disclosure, there is provided a processing program for causing a computer having at least one processor to function as follows: to acquire one or more area imaging data from medical imaging data obtained by photographing at least a part of a subject's body as a subject; to acquire one or more converted images by converting each area imaging data based on parameter information corresponding to each area imaging data for the acquired one or more area imaging data; and to acquire a composite image by combining the acquired one or more converted images.

[0007] According to one aspect of the present disclosure, there is provided a processing method executed by at least one processor in a computer having the at least one processor, the processing method including the steps of: acquiring one or more area imaging data from medical imaging data obtained by imaging at least a part of a subject's body as a subject; acquiring one or more converted images by converting each area imaging data based on parameter information corresponding to each area imaging data for the acquired one or more area imaging data; and acquiring a composite image by combining the acquired one or more converted images. [Effects of the Invention]

[0008] According to the present disclosure, it is possible to provide a processing device, a processing program, and a processing method that are capable of more appropriately processing acquired medical imaging data.

[0009] It should be noted that the above effects are merely illustrative for the sake of convenience and are not limiting. In addition to or instead of the above effects, any effect described in this disclosure or an effect obvious to a person skilled in the art may be achieved. [Brief explanation of the drawings]

[0010] [Figure 1A] FIG. 1A is a block diagram showing the configuration of a processing system 1 according to an embodiment of the present disclosure. [Figure 1B] FIG. 1B is a block diagram showing the configuration of a processing device 100 according to an embodiment of the present disclosure. [Figure 2A] FIG. 2A is a diagram conceptually illustrating an image management table stored in the processing device 100 according to an embodiment of the present disclosure. [Figure 2B] FIG. 2B is a diagram conceptually illustrating window values ​​determined by the processing device 100 according to an embodiment of the present disclosure. [Figure 2C] FIG. 2C is a diagram conceptually illustrating an example of parameter information stored in the processing device 100 according to an embodiment of the present disclosure. [Figure 2D] FIG. 2D is a diagram conceptually showing a CT value conversion table stored in the processing device 100 according to an embodiment of the present disclosure. [Figure 3] FIG. 3 is a diagram conceptually illustrating an example of medical imaging data according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a diagram showing a processing flow executed in the processing device 100 according to an embodiment of the present disclosure. [Figure 5A] FIG. 5A is a diagram illustrating an example of medical imaging data according to an embodiment of the present disclosure. [Figure 5B] FIG. 5B is a diagram illustrating an example of area imaging data according to an embodiment of the present disclosure. [Figure 5C]FIG. 5C is a diagram illustrating an example of a converted image according to an embodiment of the present disclosure. [Figure 5D] FIG. 5D is a diagram illustrating an example of a composite image according to an embodiment of the present disclosure. [Figure 6] FIG. 6 is a diagram illustrating an example of medical imaging data, a converted image, and a composite image according to an embodiment of the present disclosure. [Figure 7] FIG. 7 is a diagram illustrating a processing flow for generating a trained determination model, which is an example of an application of a synthetic image according to an embodiment of the present disclosure. [Figure 8] FIG. 8 is a diagram showing a processing flow relating to the generation of medical information, which is an example of a use of a composite image according to an embodiment of the present disclosure. [Figure 9] FIG. 9 is a diagram showing an example of a medical information output screen output by the processing device 100 according to an embodiment of the present disclosure. [Figure 10] FIG. 10 is a diagram illustrating an example of medical imaging data, a converted image, and a composite image according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] 1. Overview of Processing System 1 The processing system 1 according to the present disclosure acquires medical imaging data obtained by photographing at least a portion of a subject's body as a subject, detects one or more regions from the acquired medical imaging data to acquire one or more region imaging data, converts each of the acquired region imaging data based on parameter information corresponding to each region imaging data to acquire one or more converted images, and combines the acquired one or more converted images to acquire a composite image. The composite image thus acquired can be used, for example, as a training medical image for generating a trained judgment model for determining the subject's physical condition (e.g., the presence or absence of symptoms, detailed body parts and body composition, treatment scars, or the presence or absence of foreign bodies), input to the trained judgment model to use as input information for determining the physical condition, or output to a terminal device of a medical professional such as a doctor to use as medical information.

[0012] Here, in the case of an image having a wide range of pixel values, such as a CT image taken by a CT device, even though there is sufficient shading when focusing on only a specific region of interest, it may be difficult to fully express the differences in shading in the region of interest in the image as a whole. In such cases, in order to clearly express the differences in shading in the specific region of interest, a process (e.g., windowing) is performed to adjust the gradation to a level appropriate for the specific region of interest. If there are multiple regions of interest, this process is performed for each region of interest. Therefore, when checking the physical condition of a subject, it is necessary to switch between images and interpret the images for each of the multiple regions of interest.

[0013] Therefore, when the processing system 1 acquires, for example, CT scan data, it acquires one or more regional scan data by performing segmentation processing or the like based on the CT scan data. Then, for each of the acquired one or more regional scan data, the processing system 1 determines a window value based on the attribute information, and acquires one or more converted images by performing conversion processing based on the window value. The processing system 1 acquires a composite image by combining the acquired converted images. In other words, the processing system 1 eliminates the need to switch images for each region of interest for interpretation, and makes it possible to properly check each region of interest at once based on a composite image in which each region of interest is converted to an optimal gradation and the shading is properly expressed.

[0014] FIG. 1A is a block diagram showing the configuration of a processing system 1 according to an embodiment of the present disclosure. According to FIG. 1A, the processing system 1 includes at least a processing device 100 for processing acquired medical imaging data to generate a composite image, and a medical imaging data providing device 200 for providing the medical imaging data to the processing device 100. In the present disclosure, the processing device 100 can perform processes such as generating a trained determination model, making a determination using the trained determination model, and outputting medical information. However, these processes may be performed by other processing devices. Furthermore, the processing system 1 does not necessarily have to be composed of only the processing device 100 and the medical imaging data providing device 200, but may also include other devices, such as a terminal device available to medical professionals, a database device storing medical interview information and medical record information, and a server device involved in generating these. The devices included in the processing system 1, including the processing device 100 and the medical imaging data providing device 200, are communicably connected to each other via at least one of a wireless network and a wired network.

[0015] In the present disclosure, the "medical imaging data providing device 200" may be any device capable of providing medical imaging data, and is not limited to a specific one. Examples of such medical imaging data providing device 200 include preferably a device capable of capturing medical images having a wide range of pixel values, more preferably a medical device such as an MRI device, an X-ray device, an electrocardiogram device, an ultrasound diagnostic device, an endoscope device, or an angiography device, a terminal device capable of receiving medical imaging data from such a medical device or capturing images of the body using an on-board camera, a database device for electronic medical records or a PACS (Picture Archiving and Communication Systems), a server device for processing information stored in the database device, and a combination thereof, and more preferably a CT device. In the processing system 1, any of the above-mentioned medical imaging data providing devices 200 can be suitably used, but for convenience of explanation, the following description will be given of a case where a CT device is used.

[0016] Furthermore, in the present disclosure, a "medical image" may be any image obtained by photographing at least a portion of a subject's body as a subject, and is not limited to a specific one. Examples of such medical images include, preferably, images having a wide range of pixel values, more preferably CT images, MRI images, X-ray images, electrocardiogram data, echo images, endoscopic images, blood pressure measurement data, blood sampling data, respiratory function data, angiography images, and combinations thereof, and even more preferably CT images. CT images are used to evaluate the condition of multiple cross-sectional planes of a subject across a predetermined range of the subject (e.g., multiple regions or the entire body). Therefore, it is possible to evaluate not only the condition of a specific cross-sectional plane, but also the condition of multiple spatially consecutive cross-sectional planes. Based on such CT images, it is possible to obtain information regarding the subject's physical condition, such as changes or symptoms occurring in the body. For convenience of explanation, the following describes the use of CT images.

[0017] In addition, in this disclosure, "medical imaging data" refers to data obtained by imaging to generate a medical image. When the medical image is a CT image, the medical imaging data is, for example, projection data obtained by CT imaging or DICOM data generated from the projection data. A CT image is generated by performing windowing processing on the projection data or DICOM data based on appropriate windowing values. However, the content of the medical imaging data is not limited to this, and the medical imaging data may be any data generated during the process from imaging to image generation, including data representing the final image to be generated.

[0018] Furthermore, in this disclosure, a "subject" may be any person who is the subject of medical imaging and is not limited to only those with a specific attribute. Therefore, such subjects include all people, such as patients, examinees, diagnostic subjects, and healthy individuals. Furthermore, in this disclosure, a "user" may be any person who at least operates the processing device 100 or operates a terminal device for inputting instructions to the processing device 100, and is not limited to only those with a specific attribute.

[0019] 2. Configuration of the Processing Device 100 FIG. 1B is a block diagram showing a configuration of a processing device 100 according to an embodiment of the present disclosure. According to FIG. 1B, the processing device 100 includes a processor 111, a memory 112, and a communication interface 113. These components are electrically connected to each other via control lines and data lines. Note that the processing device 100 does not need to include all of the components shown in FIG. 1B; some components may be omitted, or other components may be added. For example, the processing device 100 may include a battery or the like for driving each component.

[0020] Such processing device 100 can be suitably applied to any device capable of executing the processes according to the present disclosure, such as an on-premise server device, a cloud server device, a laptop computer, a desktop computer, a smartphone, or a tablet terminal. Furthermore, processing device 100 can also distribute at least a portion of the processes to a server device or the like. Therefore, processing device 100 also includes a combination of multiple server devices, a combination of a server device and a database device, a combination of a server device and a laptop computer, and the like.

[0021] In the processing device 100, the processor 111 functions as a control unit that controls the processing device 100 or other components of the processing system 1 based on a program stored in the memory 112. Specifically, the processor 111 executes the following processes based on the program stored in the memory 112: "at least one processor acquires one or more area imaging data from medical imaging data obtained by imaging at least a part of the subject's body as a subject," "acquires one or more converted images by converting each acquired area imaging data based on parameter information corresponding to each area imaging data," and "acquires a composite image by combining the acquired one or more converted images." The processor 111 is mainly composed of one or more CPUs, but may also be combined with a GPU, an FPGA, etc. as appropriate.

[0022] The memory 112 is composed of RAM, ROM, non-volatile memory, HDD, etc., and functions as a storage unit. The memory 112 stores instructions and commands for various control operations of the processing system 1 according to this embodiment as a program. Specifically, the memory 112 stores programs for the processor 111 to execute, for example, "a process in which at least one processor acquires one or more area imaging data from medical imaging data obtained by imaging at least a part of the subject's body as a subject," "a process in which, for the acquired one or more area imaging data, each area imaging data is converted based on parameter information corresponding to each area imaging data to acquire one or more converted images," and "a process in which the acquired one or more converted images are combined to acquire a composite image." In addition to the programs, the memory 112 also stores various information stored in an image management table and a CT value conversion table. Note that the various information stored in the memory 112 does not necessarily have to be stored in the memory of the processing device 100, but may be stored in a remote database device or the like. In other words, the memory 112 may also include a database device in some cases.

[0023] The communication interface 113 functions as a communication unit for transmitting and receiving medical imaging data and the like to and from other devices including the medical imaging data providing device 200 via a wired or wireless communication network. Examples of the communication interface 113 include a wired communication connector such as USB or SCSI, a wireless communication transmitting and receiving device such as wireless LAN, Bluetooth (registered trademark), infrared or LTE, and various connection terminals for printed circuit boards or flexible circuit boards.

[0024] 3. Information stored in memory 112 of processing device 100 2A is a diagram conceptually illustrating an image management table stored in the processing device 100 according to an embodiment of the present disclosure. The information stored in the image management table is updated and stored as needed in accordance with the progress of processing by the processor 111 of the processing device 100. The information stored in the image management table may be stored in a memory provided in the processing device 100, or may be stored in a database device or the like connected to the processing device 100 via a communication network.

[0025] 2A, the image management table stores medical imaging data, area imaging data, attribute information, parameter information, converted images, composite images, etc. in association with image ID information. "Image ID information" is information specific to each medical imaging data and is used to identify each medical imaging data. Image ID information is assigned when medical imaging data is acquired by the medical imaging data providing device 200 or when the processing device 100 receives medical imaging data from the medical imaging data providing device 200.

[0026] The "medical imaging data" refers to at least one of data obtained by imaging at least a portion of a subject's body, provided by the medical imaging data providing device 200, and processed data that has undergone various processes, such as high-resolution imaging, pixel interpolation, compression, and data formatting, as necessary. The medical imaging data is received from the medical imaging data providing device 200 via the communication interface 113. The processed data may be processed by the medical imaging data providing device 200 or the processing device 100. The medical imaging data may correspond to either still images or video, and may be in any image format, such as black-and-white or color. Examples of medical imaging data include CT projection data and DICOM data. The medical imaging data may correspond to each of multiple CT images captured over a predetermined range (e.g., multiple body parts or the entire body), or the entirety of multiple CT images. A specific example is described in FIG. 3.

[0027] "Regional imaging data" refers to images acquired from each medical imaging data by processing by the processor 111. An example of such regional imaging data is data acquired by segmenting each anatomical region included in each medical imaging data. Segmentation may be performed, for example, for tissues or regions constituting human parts, or for combinations of multiple tissues or regions. Anatomical regions such as organs, digestive organs, blood vessels, and bones may be used as units of such tissues or regions. Segmentation may also be performed by dividing a medical image into any number of sections each containing any number of pixels. Although not specifically illustrated in FIG. 2A , when multiple regional imaging data are acquired from medical imaging data B1 by segmentation processing, each of the acquired multiple regional imaging data (C1-1, C1-2, C1-3, ..., C1-n) is stored as regional imaging data C1.

[0028] The "attribute information" is information indicating the attributes of each acquired area imaging data. Such attribute information may preferably be, for example, information based on the distribution of pixel values ​​(CT values ​​in the case of CT images) of each pixel constituting each area imaging data, information indicating the maximum and minimum pixel values ​​of each pixel constituting each area imaging data, information indicating the average pixel value of each pixel constituting each area imaging data, information indicating an anatomical location included in each area imaging data, or any combination thereof; more preferably, information based on the distribution of pixel values ​​(CT values ​​in the case of CT images) of each pixel constituting each area imaging data or information indicating an anatomical location included in each area imaging data, and even more preferably, information based on the distribution of pixel values ​​(CT values ​​in the case of CT images) of each pixel constituting each area imaging data. Although not specifically described in Figure 2A, for example, when multiple regional imaging data are acquired from medical imaging data B1 by segmentation processing, attribute information (D1-1, D1-2, D1-3, ... D1-n) is stored as attribute information D1 in association with each of the acquired multiple regional imaging data (C1-1, C1-2, C1-3, ... C1-n).

[0029] The "parameter information" is information indicating parameters used when converting each area image data to obtain a converted image. Such parameter information is determined based on the attribute information of each area image data, and preferably, information indicating a window value is used.

[0030] Here, FIG. 2B is a diagram conceptually illustrating a window value determined by the processing device 100 according to an embodiment of the present disclosure. Specifically, FIG. 2B illustrates a process in which attribute information is calculated for regional imaging data (e.g., regional imaging data C1-1 and regional imaging data C1-2) acquired from medical imaging data (e.g., medical imaging data B1), and parameter information is determined based on the attribute information. FIG. 2B (a) illustrates the entire range of pixel values ​​(e.g., CT value = Hn to Hm) assigned to each pixel constituting medical imaging data B1. For example, a CT value is used as the pixel value. The CT value indicates the degree of X-ray absorption for each pixel, with water being "0" and air being "-1,000."

[0031] 2B(b) shows a case where regional imaging data C1-1 and regional imaging data C1-2 are acquired from medical imaging data B1. However, if shading is expressed directly based on the CT values ​​obtained as medical imaging data as described above, the difference in shading may not be sufficiently apparent in a certain region of interest (region imaging data C1-1 and regional imaging data C1-2). This is because, when viewed from the perspective of the overall pixel values ​​(e.g., CT values ​​= Hn to Hm), the range in which the pixel values ​​of the regional imaging data C1-1 and the region in which the pixel values ​​of the regional imaging data C1-2 are distributed is narrow, and the difference in shading is small. Therefore, as shown in FIG. 2B(b), the range of pixel values ​​expressing shading is determined as parameter information based on each attribute information identified from the regional imaging data C1-1 and the regional imaging data C1-2 (e.g., information based on the distribution of pixel values ​​of each pixel constituting each regional imaging data, or information indicating an anatomical location included in each regional imaging data). For example, in FIG. 2B(b), a range of CT values ​​= H1-1a to H1-1b is determined as parameter information from the regional imaging data C1-1, and a range of CT values ​​= H1-2a to H1-2b is determined as parameter information from the regional imaging data C1-2.

[0032] As an example of such parameter information, a window value is used. The window value includes information indicating a window width indicating a range of pixel values ​​expressing gradation, and information indicating a window level which is the median value of the window width. This makes it possible to identify the range of pixel values ​​expressing gradation for each area imaging data. Note that, although the present embodiment will mainly describe a case where a CT window value is used as parameter information, the parameter information is not limited to this. If the medical imaging data is data obtained by MRI imaging, the MRI window value may be used as parameter information. Furthermore, the parameter information may be any information for determining the brightness or gradation of a medical image.

[0033] Based on the parameter information determined as described above, the processor 111 performs processing to convert the pixel value of each pixel into, for example, 8-bit gradations (256 gradations), thereby expressing shading. Specifically, when a range of CT values ​​= H1-1a to H1-1b is determined as parameter information from the area imaging data C1-1, gradations of 0 to 255 are assigned to the range in order, such that the minimum CT value H1-1a is assigned to gradation 0 and the maximum CT value H1-1b is assigned to gradation 255. Similarly, when a range of CT values ​​= H1-2a to H1-2b is determined as parameter information from the area imaging data C1-2, gradations of 0 to 255 are assigned to the range in order, such that the minimum CT value H1-2a is assigned to gradation 0 and the maximum CT value H1-2b is assigned to gradation 255. The CT value of each pixel is then replaced with the assigned gradation, thereby performing the conversion. This makes it possible to clearly express the shading in the converted photographic data of each area (converted image), as shown in (c) of FIG. 2B.

[0034] FIG. 2C is a diagram conceptually illustrating an example of parameter information stored in the processing device 100 according to an embodiment of the present disclosure. Specifically, FIG. 2C is a diagram illustrating an example of parameter information calculated from information based on the distribution of pixel values ​​(CT values ​​in the case of CT images) of each pixel constituting each area imaging data, which is one of the attribute information. In FIG. 2C, the distribution of pixel values ​​of each pixel in the area imaging data C1-1 is shown as a curve Dx. For each pixel value of the area imaging data C1-1 exhibiting such a distribution, the processor 111 determines, for example, a range of pixel values ​​belonging to the lowest 3% (range D1-2 in FIG. 2C ) and a range of pixel values ​​belonging to the highest 3% (range D1-3 in FIG. 2C ). Then, the processor 111 assumes that both ranges (range D1-2 in FIG. 2C and range D1-3 in FIG. 2C ) are noise components and deletes them from the distribution of pixel values ​​stored as parameter information. The processor 111 stores the remaining pixel value range D1-1 as parameter information.

[0035] By calculating parameter information from information based on the distribution of pixel values ​​of each pixel that constitutes such area shooting data, it is possible to automatically calculate the parameter information used to obtain a converted image, and further, it becomes possible to set parameter information according to the characteristics (distribution of pixel values) of each area shooting data.

[0036] FIG. 2D is a conceptual diagram illustrating a CT value conversion table stored in the processing device 100 according to an embodiment of the present disclosure. Specifically, FIG. 2D illustrates an example of parameter information determined from information indicating an anatomical region included in each region imaging data, which is one of the attribute information. Each region imaging data acquired from medical imaging data by segmentation processing includes, for example, each segmented anatomical region as a subject. Therefore, the processor 111 can determine parameter information for each region imaging data using a pre-defined CT value conversion table through processing. According to FIG. 2D, the CT value conversion table is pre-stored with a range J1 (e.g., CT value = J1-1a to J1-1b) of CT values ​​suitable for expressing shading in the region imaging data of the "head," and a range J2 (e.g., CT value = J1-2a to J1-2b) of CT values ​​suitable for expressing shading in the region imaging data of the "abdomen." Therefore, the processor 111 refers to the CT value conversion table and sets the range J1 (e.g., CT value = J1-1a to J1-1b) as the parameter information when the area shooting data of the "head" is obtained as the area shooting data, and sets the range J2 (e.g., CT value = J1-2a to J1-2b) as the parameter information when the area shooting data of the "abdomen" is obtained as the area shooting data.

[0037] Note that the methods described with reference to FIGS. 2C and 2D are merely examples of methods for determining parameter information, and naturally, other methods may be used for determination.

[0038] Returning to Figure 2A again, for example, when multiple regional imaging data are acquired from medical imaging data B1 by segmentation processing, parameter information (E1-1, E1-2, E1-3, E1-n) is stored in correspondence with each of the acquired multiple regional imaging data (C1-1, C1-2, C1-3, C1-n).

[0039] The "converted image" is an image obtained by converting each area imaging data based on parameter information corresponding to each area imaging data. For example, the converted image is obtained by converting the CT value of each pixel constituting each area imaging data into a pixel value indicating a gradation of 0 to 255 based on the parameter information. Specifically, as shown in FIG. 2B, for each pixel constituting area imaging data C1-1, the CT value of each pixel is converted so that the pixel value becomes "0" when the CT value is H1-1a and the pixel value becomes "255" when the CT value is H1-1b. This allows the area imaging data C1-1 expressed with CT values ​​of H1-1a to H1-1b to be converted into a converted image F1-1 expressed with a gradation of 0 to 255. Similarly, for each pixel constituting area imaging data C1-2, the CT value of each pixel is converted so that the pixel value becomes "0" when the CT value is H1-2a and the pixel value becomes "255" when the CT value is H1-2b. This makes it possible to convert the area imaging data C1-2 expressed in CT values ​​of H1-2a to H1-2b into a converted image F1-2 expressed in gradations of 0 to 255.

[0040] Although not specifically shown in Figure 2A, for example, when multiple regional imaging data are acquired from medical imaging data B1 by segmentation processing, converted images (F1-1, F1-2, F1-3, ... F1-n) are stored in correspondence with each of the acquired regional imaging data (C1-1, C1-2, C1-3, ... C1-n).

[0041] A "composite image" is an image obtained by combining the converted images. Such a composite image can be generated by various methods, such as pixel replacement based on the converted images, or adding pixel values ​​of the converted images and reconverting them to a predetermined gradation.

[0042] Although not specifically shown in FIG. 2A, in addition to the above information, the image management table may store various other information such as subject attribute information, medical interview information, and findings information.

[0043] 4. Medical image examples As described above, a medical image is an image obtained by photographing at least a part of a subject's body as a subject, and is information generated based on medical imaging data provided from the medical imaging data providing device 200 via the communication interface 113. Fig. 3 is a diagram conceptually illustrating an example of a medical image according to an embodiment of the present disclosure. Fig. 3 illustrates a case where a plurality of CT images taken by a CT device are used as the medical image.

[0044] Such CT images are typically obtained by capturing cross sections in the thickness direction from the head to the legs of the body at a spatially constant interval (for example, slice interval = 5 mm) over a predetermined range in the subject's body (at least a plurality of regions (e.g., head, chest, abdomen, pelvis, etc.) or the entire body). In FIG. 3, a plurality of CT images including at least CT images Dm-7 to Dm and CT images Dn-7 to Dn as measurement data are captured at a spatially constant interval (for example, slice interval = 5 mm) over a plurality of regions A to D (i.e., the entire body). Note that the regions shown in FIG. 3 are examples for ease of explanation and do not necessarily accurately represent specific regions of the human body.

[0045] Since the multiple CT images thus acquired were taken at regular intervals as described above, by referring to multiple consecutive CT images, such as CT image Dm-7 to CT image Dm, it is possible to confirm the symptoms occurring in the area included in each CT image. That is, by focusing on each individual CT image, it is possible to grasp the symptoms included in each CT image, their location, and their degree (severity). Furthermore, by focusing on multiple consecutive CT images, it is possible to grasp the transition of the symptoms, their location, and their degree.

[0046] Furthermore, each of the multiple CT images is a two-dimensional image. However, as described above, each of the multiple CT images is a cross-section in the thickness direction taken at a spatially constant interval (for example, a slice interval of 5 mm) in the direction from the head to the legs of the body. Therefore, by stacking each CT image in the order in which it was taken, it is possible to generate a three-dimensional CT image that adds the direction from the head to the legs to the two-dimensional information of the cross-section in the thickness direction. Therefore, in the present disclosure, a CT image may include both a two-dimensional image and a three-dimensional image of each cross-section.

[0047] 5. Processing flow executed by the processing device 100 Fig. 4 is a diagram showing a processing flow executed in the processing device 100 according to an embodiment of the present disclosure. Specifically, Fig. 4 is a diagram showing a processing flow executed by the processor 111 of the processing device 100 until a composite image is acquired from medical imaging data acquired from the medical imaging data providing device 200. The processing flow is mainly performed by the processor 111 of the processing device 100 reading and executing a program stored in the memory 112.

[0048] The processor 111 acquires the medical imaging data by receiving the medical imaging data and the image ID information from the medical imaging data providing device 200 via the communication interface 113 (S111). When acquiring the medical imaging data, the processor 111 stores the medical imaging data in the image management table in association with the received image ID information. Note that the image ID information may be generated by the processor 111 every time new medical imaging data is received.

[0049] FIG. 5A is a diagram illustrating an example of a medical image according to an embodiment of the present disclosure. Specifically, FIG. 5A is a diagram illustrating an example of a medical image generated from medical imaging data received in S111 of FIG. 4. FIG. 5A illustrates one CT image (corresponding to medical imaging data B1) captured from a specific cross section of a subject's body (abdomen) captured by a CT device functioning as the medical imaging data providing device 200. Therefore, the medical image includes as subjects the liver 11, gallbladder 12, large intestine 13, stomach 14, spleen 15, small intestine 16, kidneys 17, pancreas 18, vertebrae 19, and peritoneum 20, which are found in the abdomen. Here, since the medical image in FIG. 5A is a CT image generated based on a uniform windowing value, the shading of each pixel constituting the medical image is expressed according to the CT value. Therefore, as shown in (a) of Figure 2B, each region is not displayed in the optimal shading, with some tissues being displayed dark overall and other tissues being displayed light overall. Note that in Figure 5A, the shading is indicated by line width, with the vertebrae 19, drawn with a thick line, being displayed dark overall, and the liver 11, drawn with a dashed line, being displayed light overall. Thus, it is difficult to adequately reproduce the differences in shading within each tissue in a CT image generated based on a uniform windowing value.

[0050] Returning to FIG. 4 again, the processor 111 reads the acquired medical imaging data from the image management table and performs segmentation processing on the read medical imaging data (S112). The segmentation processing is a process of dividing the medical imaging data into one or more segments (e.g., anatomical regions) to generate one or more regional imaging data from each medical imaging data. The processor 111 acquires an image of each segment (e.g., anatomical region) and position information (e.g., coordinate information) of each image through the segmentation processing. The segmentation processing can be performed using various methods, such as a method using a trained segmentation model, a method in which each segment (anatomical region) is designated by the user, a binarization method using the CT value of each acquired pixel, a region growing method, a snake method, a graph cut method, a mean shift method, and combinations thereof.

[0051] Although the present invention has been described in terms of segmenting medical imaging data based on anatomical locations, other methods may also be used, such as segmenting a medical image into multiple sections, each section consisting of an arbitrary number of pixels.

[0052] Next, the processor 111 generates one or more area imaging data by segmenting one or more anatomical regions included in the medical imaging data (S113). FIG. 5B is a diagram illustrating an example of area imaging data according to an embodiment of the present disclosure. Specifically, FIGS. 5B(a) and 5B(b) each illustrate an example of an image corresponding to area imaging data C1 generated from the medical imaging data B1 by the process of S113 in FIG. 4. FIG. 5B(a) illustrates that the area imaging data C1 is generated as area imaging data (area imaging data C1-1, area imaging data C1-2, and area imaging data C1-3 to area imaging data C1-n) for each segment (anatomical region) identified by the segmentation process. FIG. 5B(b) illustrates that the area imaging data C1 is generated as images of each segment arranged on a single image (partial images corresponding to area imaging data C1-1, area imaging data C1-2, and area imaging data C1-3 to area imaging data C1-n). The generation of the regional imaging data may be performed by either of the methods shown in Fig. 5B(a) and (b). Returning to Fig. 4 again, the processor 111 stores the generated one or more regional imaging data in association with the image ID information of the original medical imaging data.

[0053] Next, the processor 111 generates attribute information for each of the generated area imaging data (S114). Examples of the attribute information include information based on the distribution of pixel values ​​(e.g., CT values) of each pixel constituting each area imaging data, information indicating the maximum and minimum pixel values ​​of each pixel constituting each area imaging data, information indicating the average pixel values ​​of each pixel constituting each area imaging data, information indicating an anatomical location included in each area imaging data, and combinations thereof. Among these, the information based on the distribution of pixel values ​​(e.g., CT values) of each pixel constituting each area imaging data is generated by calculating a range in which the pixel values ​​are distributed based on the pixel values ​​of each pixel constituting the area imaging data. Furthermore, the information indicating an anatomical location included in each area imaging data is generated by identifying information indicating the name of the anatomical location identified by the segmentation process. Note that the type of attribute information generated may differ for each area imaging data. For example, information indicating an anatomical location may be generated as attribute information for some area imaging data, and information based on the distribution of CT values ​​may be generated as attribute information for the remaining area imaging data. The processor 111 stores each piece of generated attribute information in association with each piece of original area photography data.

[0054] Next, the processor 111 determines parameter information based on the attribute information generated for each piece of regional imaging data (S115). An example of such parameter information is a window value including a window width and a window level (see, for example, FIG. 2B). Specifically, the processor 111 generates parameter information from information based on the distribution of pixel values ​​(e.g., CT values) of each pixel constituting each piece of regional imaging data as attribute information, using the method illustrated in FIG. 2C. The processor 111 also generates parameter information from information indicating an anatomical site included in each piece of regional imaging data as attribute information, using a method that uses a CT value conversion table illustrated in FIG. 2D. Note that, when attribute information corresponding to a certain piece of regional imaging data indicates specific content, the processor 111 may generate a preset fixed numerical value as parameter information. The processor 111 stores the generated parameter information in association with each piece of attribute information.

[0055] Next, the processor 111 generates a converted image by converting the pixel values ​​(e.g., CT values) of the area imaging data into 8-bit pixel values ​​representing gradations of 0 to 255 based on the parameter information determined in S115 (S116). Specifically, with reference to FIG. 2B, the processor 111 converts the CT values ​​of each pixel constituting the area imaging data C1-1 into pixel values ​​of the converted image so that the pixel value of each pixel becomes "0" when the CT value is H1-1a and becomes "255" when the CT value is H1-1b. In this way, the processor 111 converts the area imaging data C1-1 expressed with CT values ​​of H1-1a to H1-1b into a converted image F1-1 expressed in gradations of 0 to 255. Similarly, the processor 111 converts the CT value of each pixel constituting the area imaging data C1-2 into a pixel value of a converted image so that the pixel value of each pixel becomes "0" when the CT value is H1-2a and becomes "255" when the CT value is H1-2b. In this way, the processor 111 converts the area imaging data C1-2 expressed with CT values ​​of H1-2a to H1-2b into a converted image F1-2 expressed in gradations of 0 to 255. The processor 111 stores the generated converted image in association with each area imaging data.

[0056] Here, Fig. 5C is a diagram showing an example of a converted image according to an embodiment of the present disclosure. Specifically, Figs. 5C(a) and 5C(b) each show an example of a converted image F1 generated from area shooting data C1 (area shooting data C1-1, area shooting data C1-2, and area shooting data C1-3 to area shooting data C1-n) by the process of S116 in Fig. 4. According to Fig. 5C(a), the converted image F1 is generated by converting each of the area shooting data C1 (area shooting data C1-1, area shooting data C1-2, and area shooting data C1-3 to area shooting data C1-n) in Fig. 5B(a) into a converted image (converted image F1-1, converted image F1-2, and converted image F1-3 to converted image F1-n) one by one. 5C(b), converted images F1-1, F1-2, and F1-3 to F1-n are generated by converting the area shooting data of each segment (area shooting data C1-1, area shooting data C1-2, and area shooting data C1-3 to C1-n) arranged on one image, and then rearranged on one image based on the position information. The generated converted images may be generated by either method shown in FIGS. 5C(a) and (b).

[0057] 4 again, processor 111 performs a process of synthesizing the generated converted images (S117). Specifically, processor 111 generates a composite image by using one or more converted images associated with the same image ID information from among the acquired converted images, by a method such as pixel replacement, or by adding pixel values ​​of the converted images and reconverting them to a predetermined gradation.

[0058] FIG. 5D is a diagram illustrating an example of a composite image according to an embodiment of the present disclosure. Specifically, FIG. 5D illustrates an example of a composite image G1 obtained from each converted image F1 by the process of S117 in FIG. 4. According to FIG. 5D, the composite image G1 is generated by rearranging converted images F1-1, F1-2, and F1-3 to F1-n, which are generated corresponding to each segment (e.g., anatomical region), on a single image based on the positional information of each converted image. As a result, each segment (anatomical region) included in the composite image G1 is converted and represented using an appropriate pixel value (e.g., a gradation of 0 to 255), allowing each segment to be clearly visualized. In other words, multiple tissues can be simultaneously visualized using a single composite image.

[0059] At this time, various image processing such as anti-aliasing and resizing can also be applied to segments (e.g., anatomical regions) included in each converted image. For example, the processor 111 performs anti-aliasing processing on each converted image to reduce aliasing at the boundary portions of the segments. This anti-aliasing processing makes it possible to more smoothly represent the boundary portions of the segments. Furthermore, when rearranging each converted image based on the position information, the processor 111 performs resizing processing to fill the space formed around each segment (space where no segment is located). This resizing processing reduces wasted space and makes it possible to arrange each segment, which is a region of interest, larger. While anti-aliasing and resizing are given as examples here, various other image processing can also be applied, such as various filter processing such as smoothing and edge enhancement, noise removal processing, data format conversion processing, color correction processing, and combinations thereof.

[0060] 4, the image generated by the process shown in Fig. 5D is acquired as a composite image and stored in association with the image ID information (S118), thereby completing the series of processing flows.

[0061] 6 is a diagram showing an example of medical imaging data, a converted image, and a composite image according to an embodiment of the present disclosure. FIG. 6 is a diagram showing another example of medical imaging data, a converted image, and a composite image acquired by the processing flow of FIG. 4. FIG. 6 shows (a) medical imaging data B2, (b) converted images (converted images F2-1, F2-2, and F2-3 to F2-n) generated based on each area imaging data (area imaging data C2-1, area imaging data C2-2, and area imaging data C2-3 to C2-n) acquired from the medical imaging data B2, and (c) composite image G2 obtained by combining each converted image.

[0062] As explained in FIG. 3, the multiple CT images are obtained by capturing cross sections in the thickness direction of the body from the head to the legs at a fixed spatial interval (e.g., a slice interval of 5 mm). Therefore, by stacking the CT images in the order in which they were captured, it is possible to generate a three-dimensional CT image that adds the direction from the head to the legs to the two-dimensional information of the cross sections in the thickness direction. That is, FIG. 6(a) shows an example of a three-dimensional CT image corresponding to medical imaging data B2. The medical imaging data B2 includes various anatomical regions, but because each pixel value is expressed as a CT value, the differences in shading in each segment (e.g., anatomical region) cannot be fully reproduced.

[0063] 6(b) shows converted images F2-1, F2-2, F2-3 to F2-n generated from regional imaging data obtained by segmentation processing of medical imaging data B2 for each segment (e.g., anatomical region). The values ​​of pixels constituting each converted image are determined based on the CT value and parameter values ​​(e.g., window value), making it possible to clearly grasp the shading in each segment (e.g., anatomical region).

[0064] 6(c) shows a composite image G2 obtained by combining the generated converted images. In the composite image G2, each segment (e.g., anatomical region) included in the composite image is converted into a pixel value indicating a gray scale of 0 to 255 that is optimized for each segment, so that even images of different tissues can be viewed together at once.

[0065] In this embodiment, the description will be focused on the case where each converted image is generated by setting parameter values ​​corresponding to each region imaging data so that the CT image is clear, but the method of determining the parameter values ​​is not limited to this. For example, parameter values ​​may be set so that part of the CT image is blurred. Specifically, by setting the parameters corresponding to the region imaging data of the subject's head to a window width of 0 or a brightness of 0, the converted image of the head becomes an image filled in black. This processing makes it possible to mask the head region in the composite image and perform anonymization. Furthermore, for example, to generate a composite image specialized for diagnosing the abdominal cavity, parameters corresponding to each region imaging data may be set so that regions outside the abdominal cavity are masked.

[0066] In this way, processing using composite images can be suitably applied to three-dimensional CT images, rather than two-dimensional CT images of any cross section in the thickness direction of the subject's body.

[0067] 6. Uses of the acquired composite images The composite image obtained by the processing of Figure 4 can be used, for example, as a training medical image for generating a trained judgment model for determining the subject's physical condition (e.g., the presence or absence of symptoms, detailed body parts and body composition, treatment scars, or the presence or absence of foreign objects), or it can be input into the trained judgment model and used as input information for determining the physical condition, or it can be output to a terminal device of a medical professional such as a doctor and used as medical information.

[0068] (A) Example of using a synthetic image as a medical image for training As described above, the acquired composite image can be used as a training medical image to generate a training judgment model for determining the physical condition of a subject. FIG. 7 is a diagram showing a processing flow related to generation of a trained judgment model, which is an example of a use of a composite image according to an embodiment of the present disclosure. Specifically, FIG. 7 shows a processing flow for generating a trained judgment model using the generated composite image as a training medical image. This processing flow is mainly performed by the processor 111 of the processing device 100 reading and executing a program stored in the memory 112, but may also be performed by another processing device (e.g., a model generation device).

[0069] 7, a step of acquiring a composite image generated by the processing device 100 as a training medical image (S211) may be executed. Specifically, the processor 111 of the processing device 100 reads out one or more generated composite images as training medical images by referring to the image management table.

[0070] Next, a process is executed in which information about the condition of a body part seen in each of the acquired composite images (e.g., the presence or absence of a specific symptom, its location (site) and its severity) is labeled as condition information for each of the acquired composite images (S212). The labeling is performed by the processor 111 accepting an operation input by a medical professional such as a doctor, indicating a determination result such as the type and severity of a surgical symptom, the type and severity (e.g., size) of an internal symptom, and location information of the body part where these are seen as findings (e.g., liver, pancreas, spleen, etc.). Examples of the severity include "applicable" or "not applicable" for the symptom, "Level 1," "Level 2," "Level 3," or "Level 4," or "mild," "moderate," or "severe." Alternatively or in addition to this, prepared CT images showing surgical symptoms and CT images showing internal symptoms may be prepared, and the degree of correspondence with these images may be calculated by image analysis processing, and labeling may be performed based on this degree of correspondence.

[0071] Once the training medical images (composite images) and the associated label information are obtained, the processor 111 executes a step of performing machine learning of body part condition patterns using them (S213). As an example, the machine learning is performed by providing a set of the training medical images (composite images) and the label information to a neural network configured by combining neurons, and repeating learning while adjusting the parameters of each neuron so that the output of the neural network is the same as the label information. Then, a step of acquiring a trained judgment model (e.g., neural network and parameters) is executed (S214). This causes the processor 111 to finish generating the trained judgment model. The acquired trained judgment model may be stored in the memory 112 of the processing device 100.

[0072] The trained decision model in Figure 7 was generated using a neural network or a convolutional neural network, but it can also be generated using other machine learning methods such as the nearest neighbor method, decision tree, regression tree, or random forest.

[0073] In this way, by using a synthetic image as a training medical image, there is no need to prepare a training image for each segment (e.g., anatomical site), and the learning process can be efficiently performed in the trained judgment model.

[0074] (B) Examples of using it as input information for a trained decision model or as medical information As described above, the acquired composite image can be used as input information for generating information about the state of the body (for example, the presence or absence of a predetermined symptom, its location (site) and its severity). To generate such information, for example, a trained judgment model capable of outputting the information is used. The trained judgment model may be generated by, for example, the process shown in FIG. 7, or may be generated by using a CT image or each converted image generated based on the CT image as a training medical image, similar to the process shown in FIG. 7.

[0075] Fig. 8 is a diagram showing a processing flow related to the generation of medical information, which is an example of a use of a synthetic image according to an embodiment of the present disclosure. Specifically, Fig. 8 is a diagram showing a processing flow executed by the processor 111 of the processing device 100 from inputting a synthetic image into a trained determination model to obtain information related to a physical condition to outputting medical information. This processing flow is mainly performed by the processor 111 of the processing device 100 reading and executing a program stored in the memory 112, but may also be executed by another processing device (e.g., a server device).

[0076] 8, when processor 111 receives from a user a designation of a subject whose physical condition is to be determined, processor 111 refers to the image management table and reads out a composite image stored in association with the subject (S311). Processor 111 then inputs the read-out composite image as input information to a trained determination model (S312) to obtain information on the determination result of the physical condition (for example, the presence or absence of a predetermined symptom, its location (site) and its severity) as output information (S313). Processor 111 stores the obtained information in association with the subject in memory 112.

[0077] Next, the processor 111 generates medical information using the information on the assessment result of the physical condition acquired as output information and the composite image (S314), and then outputs the generated medical information to the user's terminal device via the communication interface 113 (S315).

[0078] 9 is a diagram showing an example of a medical information output screen output by the processing device 100 according to an embodiment of the present disclosure. Specifically, Fig. 9 is a diagram showing an example of a medical information output screen 30 output by a terminal device based on the medical information output from the processing device 100 in S315 of Fig. 8. According to Fig. 9, the medical information output screen 30 includes, in addition to subject ID information of the subject whose physical condition is to be determined, a composite image G2 generated by the processing device 100 and information on the determination result of the physical condition obtained by using the composite image G2 as input information (symptom information 32 indicating the presence or absence of a predetermined symptom, location information 33 indicating the location of the symptom, and severity information 31 indicating the severity of the symptom).

[0079] In this way, by using the composite image as medical information, medical personnel and the like can more efficiently assess the physical condition by simultaneously checking the assessment result of the physical condition and the composite image.

[0080] The medical information output by the processor 111 may include information indicating the presence or absence of correlation between abnormalities in multiple anatomical locations and the details of the correlation. The medical information output by the processor 111 may also include information indicating the injury route and injury cause inferred from the correlation between abnormalities in multiple anatomical locations. Since the composite image allows abnormalities in multiple anatomical locations to be confirmed in a single image, displaying the above-mentioned medical information regarding the correlation between multiple anatomical locations together with the composite image makes it possible to provide information that is easier for the user to handle.

[0081] Returning to Fig. 8 again, once the medical information is output as described above, the processor 111 ends the series of processing flows. Note that, in Fig. 8, a trained determination model is used to generate information related to the determination result of the physical condition, but the processor 111 may generate information related to the physical condition by simply outputting the composite image to the user's terminal device and accepting operation input from the user. The processor 111 may also generate information related to the determination result of the physical condition by image processing such as comparing the composite image with a model image prepared in advance for each condition.

[0082] In this way, by using the composite image as input information for determining the state of the body, it is not necessary to input individual images for each segment (e.g., anatomical part) as input information, and the process related to the determination can be performed more efficiently.

[0083] As described above, in this embodiment, it is possible to provide a processing device, a processing program, and a processing method that can more appropriately process acquired medical imaging data.

[0084] 7. Variations (A) Example of using multiple medical imaging data with different time phases In the above embodiment, a case where a composite image is acquired from a CT image acquired from the medical imaging data providing device 200 has been described. FIG. 10 is a diagram showing an example of medical imaging data, a converted image, and a composite image according to an embodiment of the present disclosure. Specifically, FIG. 10 is a diagram showing still another example of medical imaging data, a converted image, and a composite image acquired by the processing flow of FIG. 4. FIG. 10 shows (a) medical imaging data (medical imaging data B3-1 and medical imaging data B3-2), (b) converted images (converted images F3-1 to F3-m) generated based on each regional imaging data (region imaging data C3-1 to C3-m) acquired from the medical imaging data B3-1 and each converted image (converted images F3-m1 to F3-n) generated based on each regional imaging data (region imaging data C3-m1 to C3-n) acquired from the medical imaging data B3-2, and (c) a composite image G3 obtained by combining each converted image.

[0085] Generally, CT images are captured in multiple time phases depending on the time elapsed since the subject ingested the contrast agent: an arterial phase (e.g., an early phase approximately 30 seconds after the injection of the contrast agent) in which the contrast agent is predominantly distributed in the arteries; a venous phase (e.g., a late phase approximately 60 seconds after the injection of the contrast agent) in which the contrast agent is predominantly distributed in the veins and portal vein; and an equilibrium phase (e.g., approximately 3 minutes after the injection of the contrast agent) in which the contrast agent is uniformly distributed throughout the tissues. Generally, CT images are captured at optimal time phases for each tissue or region, such as an anatomical region (e.g., blood vessels and muscle tissues are captured in the arterial phase, and organs are captured in the venous phase). Therefore, in FIG. 10(a), medical imaging data B3-1 (first time-phase image) captured in the venous phase and medical imaging data B3-2 (second time-phase image) captured in the arterial phase are acquired, and these are used together as medical imaging data.

[0086] 10(b), the processor 111 presets an appropriate time phase for each anatomical region and performs segmentation processing on the medical imaging data of the preset time phase to obtain each region imaging data. For example, to obtain region imaging data of organs such as the kidneys and lungs, the processor 111 reads out medical imaging data B3-1 captured in the venous phase and performs segmentation processing on the medical imaging data to obtain region imaging data C3-1 to C3-m. To obtain region imaging data of blood vessels and muscle tissues, the processor 111 reads out medical imaging data B3-2 captured in the arterial phase and performs segmentation processing on the medical imaging data B3-2 to obtain region imaging data C3-m1 to C3-n. Then, the processor 111 obtains converted images F3-1 to F3-m from the area shooting data C3-1 to C3-m, and converts converted images F3-m1 to F3-n from the area shooting data C3-m1 to C3-n, respectively.

[0087] 10(c) shows a composite image G3 obtained by combining the generated converted images. The composite image G3 is obtained by combining the converted images, similar to FIG. 5D and FIG. 6(c).

[0088] In this way, CT images are usually interpreted by switching between multiple images with different time phases or by arranging multiple images side by side, but in the example in Figure 10, converted images are generated for each segment (anatomical region) from medical imaging data taken at the optimal time phase for each anatomical region, and a composite image is then generated from these converted images. This makes it possible to interpret images from multiple time phases together in a single image.

[0089] In the example of FIG. 10, the case where the time phases are divided into two, the venous phase and the arterial phase, has been described. However, the same processing can be performed even if the medical imaging data is prepared by dividing it into three or more time phases, including an equilibrium phase and other phases.

[0090] Although the example using a plurality of medical imaging data sets of different time phases has been described above, medical imaging data sets generated using different setting parameters may be used by a similar process. That is, the medical imaging data sets may include at least first-format data and second-format data sets generated using different setting parameters, and it may be set for each anatomical region which of the first-format data and the second-format data sets to be used for the conversion. For example, a plurality of medical imaging data sets having different CT imaging parameters and reconstruction parameters (specifically, convolution kernels, etc.) may be provided from the medical imaging data providing device 200, and the processor 111 may acquire each converted image using appropriate data for each anatomical region from the medical imaging data providing device 200.

[0091] (B) Example of conversion process For example, in the process for generating a converted image in S116 of Fig. 4, the CT values ​​of each pixel constituting medical imaging data are converted into pixel values ​​represented by 256 gradations from 0 to 255. However, 256 gradations from 0 to 255 are merely an example, and other pixel values ​​may naturally be used. For example, the CT values ​​may be converted into pixel values ​​represented by other gradations, such as 16 gradations or 64 gradations. Furthermore, in the process for generating a converted image in S116 of Fig. 4, the CT values ​​are converted into pixel values ​​represented by a 256-gradation grayscale. However, pixel values ​​of 16 gradations, 64 gradations, or 256 gradations may be set for each of the colors red (R), green (G), and blue (B).

[0092] (C) Other The processes and procedures described herein can be realized not only by those explicitly described in the embodiments, but also by software, hardware, or a combination thereof. Specifically, the processes and procedures described herein can be realized by implementing logic corresponding to the processes in media such as integrated circuits, volatile memory, nonvolatile memory, magnetic disks, and optical storage. Furthermore, the processes and procedures described herein can be implemented as computer programs and executed by various computers, including processing devices and server devices.

[0093] Although processes and procedures described herein are described as being performed by a single device, software, component, or module, such processes or procedures may be performed by multiple devices, multiple software, multiple components, and / or multiple modules. Furthermore, although various information described herein is described as being stored in a single memory or storage unit, such information may be stored in multiple memories within a single device or multiple memories distributed across multiple devices. Furthermore, software and hardware elements described herein may be realized by integrating them into fewer components or by decomposing them into more components. [Explanation of symbols]

[0094] 1 Processing System 100 Processing equipment 200 Medical imaging data providing device

Claims

1. A processing device comprising at least one processor, The at least one processor: acquiring one or more area imaging data from medical imaging data obtained by imaging at least a part of the subject's body as a subject; For the one or more pieces of acquired area photography data, convert each piece of area photography data based on parameter information corresponding to each piece of area photography data to acquire one or more converted images; obtaining a composite image by combining the one or more obtained converted images; a processing unit configured to perform processing for:

2. The processing device according to claim 1 , wherein the parameter information is information for determining brightness or gradation of a medical image.

3. The medical imaging data is data obtained by CT imaging or MRI imaging, The parameter information is information indicating a window value. The processing device of claim 1 .

4. The processing device according to claim 3 , wherein the one or more converted images are obtained by performing a gradation conversion process on the corresponding area photography data of the one or more area photography data based on the window value.

5. The processing device according to claim 1 , wherein the composite image is input to a trained determination model for determining the physical state of the subject, and is used to obtain information indicating the determination result of the state as output.

6. The processing device according to claim 1 , wherein the composite image is used as a training medical image for generating a trained determination model for determining the physical condition of the subject or another subject different from the subject.

7. the medical imaging data includes at least a plurality of first time phase data and second time phase data having different imaging timings; which of the first time phase data and the second time phase data is to be used for the transformation is set for each anatomical site; The processing device of claim 1 .

8. the medical imaging data includes at least first format data and second format data generated using different setting parameters; which of the first format data and the second format data is to be used for the conversion is set for each anatomical part; The processing device of claim 1 .

9. The processing device according to claim 1 , wherein the one or more regional imaging data are acquired by performing segmentation processing on the medical imaging data based on anatomical parts.

10. The processing device according to claim 1, wherein the parameter information is determined based on at least one of information indicating a distribution of CT values ​​for each of the one or more regional imaging data and information indicating an anatomical site included in each of the one or more regional imaging data.

11. The processing device according to claim 1 , wherein the one or more transformed images are obtained by performing at least one of anti-aliasing processing and resizing processing on a region of an anatomical part included as a subject in each of the transformed images.

12. In a computer having at least one processor, the at least one processor acquiring one or more area imaging data from medical imaging data obtained by imaging at least a part of the subject's body as a subject; For the one or more pieces of acquired area photography data, converting each piece of area photography data based on parameter information corresponding to each piece of area photography data to acquire one or more converted images; obtaining a composite image by combining the one or more obtained converted images; A processing program that makes it function like this.

13. A processing method executed by at least one processor in a computer having the at least one processor, comprising: acquiring one or more area imaging data from medical imaging data obtained by imaging at least a part of a subject's body; a step of acquiring one or more converted images by converting each area photography data based on parameter information corresponding to each area photography data for the acquired one or more area photography data; obtaining a composite image by combining the one or more obtained transformed images; A processing method comprising:

Images