Medical image processing equipment

Abstract

To efficiently identify the position of a target organ in a medical image.SOLUTION: A medical image processing apparatus according to an embodiment comprises a first extraction unit, a second extraction unit, a first identifying unit, a second identifying unit, and a segmentation unit. The first extraction unit extracts anatomical landmarks indicative of feature points of an anatomical tissue included in a medical image from the medical image. The second extraction unit, on the basis of information which associates a relation between an anatomical landmark and an organ to which the anatomical landmark belongs, extracts the anatomical landmark associated to the organ from among the anatomical landmarks extracted by the first extraction unit. The first identifying unit identifies a target organ to be diagnosed, on the basis of a name of the organ extracted by the second extraction unit. The segmentation unit segments a region of the target organ identified from the medical image.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The embodiments disclosed in the present specification and drawings relate to a medical image processing apparatus.

Background Art

[0002] Conventionally, there is known a technique of extracting a target organ to be diagnosed by segmenting an organ included in a medical image captured by an image diagnostic apparatus.

[0003] In the conventional art, when performing segmentation, the position of the target organ is estimated using preset atlas information or the like. However, there are various variations in the imaging range, and the atlas information may not be applied properly. Further, although it is also practiced that a user specifies the position of the target organ before performing segmentation, this may complicate the user's operation.

Prior Art Literature

Patent Literature

[0004]

Patent Literature 1

Summary of Invention

Problem to be Solved by the Invention

[0005] One of the problems to be solved by the embodiments disclosed in the present specification and drawings is to efficiently specify the position of a target organ in a medical image. However, the problems to be solved by the embodiments disclosed in the present specification and drawings are not limited to the above problem. Problems corresponding to the respective effects achieved by the respective configurations shown in the embodiments described later can also be regarded as other problems.

Means for Solving the Problem

[0006] The medical image processing apparatus according to this embodiment comprises a first extraction unit, a second extraction unit, a first identification unit, a second identification unit, and a segmentation unit. The first extraction unit extracts anatomical landmarks representing characteristic points of anatomical tissue contained in a medical image from the medical image. The second extraction unit extracts anatomical landmarks associated with organs from among the anatomical landmarks extracted by the first extraction unit, based on information relating the relationship between the anatomical landmarks and the organs to which the anatomical landmarks belong. The first identification unit identifies the target organ to be diagnosed based on the name of the organ extracted by the second extraction unit. The segmentation unit divides the region of the target organ identified from the medical image. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 shows an example of the overall configuration of a medical information processing system according to the embodiment. [Figure 2] Figure 2 is an explanatory diagram showing an example of a method for generating a landmark extraction model using machine learning according to the embodiment. [Figure 3] Figure 3 illustrates an example of a method for extracting anatomical landmarks using a trained model according to the embodiment. [Figure 4] Figure 4 is an example of CT image data taken of the chest and abdomen of a subject according to the embodiment. [Figure 5] Figure 5 illustrates an example of the process for extracting the location of organ-specific landmarks according to the embodiment. [Figure 6] Figure 6 is a flowchart showing an example of a process performed by the medical image processing device according to the embodiment. [Modes for carrying out the invention]

[0008] The embodiments of the medical image processing device will be described in detail below with reference to the drawings.

[0009] Figure 1 is a block diagram showing an example of the configuration of a medical information processing system S according to an embodiment. As shown in Figure 1, the medical information processing system S comprises a medical image processing device 100, a medical image diagnostic device 200, and a medical image storage device 500. The medical image processing device 100 is connected to the medical image storage device 500 via a network 300 such as an in-hospital LAN (Local Area Network) to enable communication.

[0010] The medical image storage device 500 stores medical images taken by the medical imaging diagnostic device 200. The medical image storage device 500 also stores medical image data in association with the identification information of the subject (e.g., patient ID).

[0011] The medical image storage device 500 is, for example, a PACS (Picture Archiving and Communication System) server device and stores medical image data in a format compliant with DICOM (Digital Imaging and Communications in Medicine). Medical images include, but are not limited to, CT (Computed Tomography) image data, magnetic resonance imaging image data, and ultrasound diagnostic image data.

[0012] The medical image storage device 500 is implemented, for example, by computer equipment such as a DB (Database) server, and stores medical image data in semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or in storage circuits such as hard disks and optical discs.

[0013] Medical imaging diagnostic equipment 200 is, for example, a device that takes medical images of a subject. Medical imaging diagnostic equipment 200 includes, for example, an MRI (Magnetic Resonance Imaging) device, an X-ray CT (Computed Tomography) device, an X-ray diagnostic device, an ultrasound diagnostic device, a PET (Positron Emission Tomography) device, a SPECT (Single Photon Emission Computed Tomography) device, etc.

[0014] However, the medical imaging diagnostic device 200 is not limited to these. The medical imaging diagnostic device 200 is also called a modality. Although Figure 1 illustrates one medical imaging diagnostic device 200, multiple medical imaging diagnostic devices 200 may be provided.

[0015] Medical images are images of a subject taken by a medical imaging diagnostic device 200. Examples of medical images include magnetic resonance images, X-ray CT images, and ultrasound images. However, medical images are not limited to these.

[0016] The medical image processing device 100 is, for example, an information processing device such as a server or a PC (Personal Computer). The medical image processing device 100 includes a network interface 110, a storage circuit 120, an input interface 130, a display 140, and a processing circuit 150.

[0017] The NW interface 110 is connected to the processing circuit 150 and controls the transmission and communication of various data between the medical image processing device 100, the medical image diagnostic device 200, and the medical image storage device 500. The NW interface 110 is implemented by a network card, network adapter, NIC (Network Interface Controller), etc.

[0018] The memory circuit 120 stores various types of information used by the processing circuit 150. For example, the memory circuit 120 stores organ-specific landmark information representing anatomical landmarks characteristic of each organ. The memory circuit 120 also stores various programs. Note that the memory circuit 120 is, for example, a storage device such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or an integrated circuit storage device.

[0019] Further, the memory circuit 120 may be a drive device that reads and writes various information from and to portable storage media such as a CD (Compact Disc), a DVD (Digital Versatile Disc), and a flash memory, and semiconductor memory elements such as a RAM (Random Access Memory).

[0020] The input interface 130 is implemented by a trackball that accepts user operations, a switch button, a mouse, a keyboard, a touchpad that performs input operation by touching an operation surface, a touch screen in which a display screen and a touchpad are integrated, a non-contact input circuit using an optical sensor, a voice input circuit, and the like.

[0021] The input interface 130 is connected to the processing circuit 150, converts an input operation received from a user into an electric signal, and outputs the electric signal to the processing circuit 150. Note that in the present specification, the input interface is not limited to those including physical operation components such as a mouse and a keyboard. For example, an electric signal processing circuit that receives an electric signal corresponding to an input operation from an external input device provided separately from the apparatus and outputs the electric signal to the processing circuit 150 is also included as an example of the input interface.

[0022] The display 140 displays various types of information under the control of the processing circuit 150. For example, the display 140 outputs a medical image viewer containing medical images generated by the processing circuit 150, and a GUI (Graphical User Interface) for accepting various operations from the user. The display 140 is an example of a display unit.

[0023] The display 140 is specifically an LCD display or a CRT (Cathode Ray Tube) display, etc. The input interface 130 and the display 140 may be integrated. For example, the input interface 130 and the display 140 may be implemented using a touch panel.

[0024] The processing circuit 150 is a processor that reads programs from the memory circuit 120 and executes them to realize functions corresponding to each program. The processing circuit 150 in this embodiment includes an acquisition function 151, a first extraction function 152, a second extraction function 153, a first identification function 154, a second identification function 155, a segmentation function 156, a display control function 157, and a reception function 158.

[0025] The first extraction function 152 is an example of the first extraction unit. The second extraction function 153 is an example of the second extraction unit. The first identification function 154 is an example of the first identification unit. The second identification function 155 is an example of the second identification unit. The segmentation function 156 is an example of the segmentation unit. The display control function 157 is an example of the display control unit. The reception function 158 is an example of the reception unit.

[0026] Here, for example, the processing functions of the processing circuit 150, which are components of the processing circuit 150, such as the acquisition function 151, the first extraction function 152, the second extraction function 153, the first identification function 154, the second identification function 155, the segmentation function 156, the display control function 157, and the reception function 158, are stored in the memory circuit 120 in the form of programs that can be executed by a computer. The processing circuit 150 is a processor. For example, the processing circuit 150 reads the programs from the memory circuit 120 and executes them to realize the functions corresponding to each program. In other words, the processing circuit 150 in the state in which each program has been read will have the functions shown in the processing circuit 150 of Figure 1.

[0027] In Figure 1, the processing functions performed by the acquisition function 151, first extraction function 152, second extraction function 153, first identification function 154, second identification function 155, segmentation function 156, display control function 157, and reception function 158 are explained as being realized by a single processor. However, it is also acceptable to configure the processing circuit 150 by combining multiple independent processors, with each processor executing a program to realize the functions.

[0028] Furthermore, although Figure 1 describes a single memory circuit 120 that stores programs corresponding to each processing function, it is also possible to have multiple memory circuits distributed and the processing circuit 150 read the corresponding programs from individual memory circuits.

[0029] The above description illustrates an example in which a "processor" reads and executes programs corresponding to each function from a memory circuit, but the embodiments are not limited to this. The term "processor" refers to circuits such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), Application Specific Integrated Circuit (ASIC), and Programmable Logic Device (e.g., Simple Programmable Logic Device (SPLD), Complex Programmable Logic Device (CPLD), and Field Programmable Gate Array (FPGA)).

[0030] If the processor is a CPU, for example, it performs its functions by reading and executing programs stored in memory circuits. On the other hand, if the processor is an ASIC, instead of storing the program in memory circuits 120, the function is directly incorporated as a logic circuit within the processor's circuitry.

[0031] In this embodiment, each processor is not limited to being configured as a single circuit; multiple independent circuits may be combined to form a single processor and realize its functions. Furthermore, the multiple components shown in Figure 1 may be integrated into a single processor to realize its functions.

[0032] The acquisition function 151 acquires medical images of a subject from the medical image storage device 500 via the network 300 and the NW interface 110. Alternatively, the acquisition function 151 may acquire medical images from the medical image diagnostic device 200.

[0033] For example, the acquisition function 151 acquires a medical image from the medical image storage device 500 that corresponds to the patient ID of the subject to be diagnosed.

[0034] The first extraction function 152 extracts anatomical landmarks, which represent local characteristic points contained in anatomical tissue, from the medical image data IG acquired by the acquisition function 151. For example, the first extraction function 152 extracts anatomical landmarks based on anatomical information.

[0035] Anatomical information refers to information about the location of characteristic points in anatomical tissues, such as bones and organs. Anatomical information is also called anatomical landmark information. Anatomical landmarks are local characteristic points contained in anatomical tissues, such as "the lower end of the kidney" or "the tips of the 1st to 12th ribs."

[0036] The method for extracting anatomical landmarks can employ known image processing techniques. For example, the first extraction function 152 inputs a medical image into a landmark extraction model (an example of anatomical information) and extracts anatomical landmarks based on the output result.

[0037] A landmark extraction model is an example of anatomical information. A landmark extraction model is a trained model that, for example, takes a medical image as input and outputs coordinate information of anatomical landmarks and identification information (hereinafter also referred to as labels) of those anatomical landmarks. For example, a landmark extraction model can be generated by an external learning device. Alternatively, the medical image processing device 100 may generate the landmark extraction model itself.

[0038] In this embodiment, the landmark extraction model is stored in a memory device provided by an external workstation or the like, but the landmark extraction model may also be stored in the memory circuit 120.

[0039] Here, we will explain how to generate a landmark extraction model. For example, a learning device generates a landmark model by performing machine learning (including deep learning). Figure 2 is an explanatory diagram showing an example of how to generate a landmark extraction model using machine learning.

[0040] For example, as shown in Figure 2, the learning device inputs "medical image data," which is the input training data, and "coordinate information and labels of each anatomical landmark in the medical image," which is the output training data, into a machine learning engine as a training dataset. By performing machine learning, it generates a landmark extraction model (trained model) that is configured to output the coordinate information and labels of each anatomical landmark in the medical image in response to the input medical image data.

[0041] Here, as a machine learning engine, for example, a neural network described in the publicly known non-patent document "Pattern recognition and machine learning" by Christopher M. Bishop (USA), 1st edition, Springer, 2006, pp. 225-290 can be applied.

[0042] In addition to the neural networks mentioned above, the machine learning engine may also utilize various algorithms such as deep learning, logistic regression analysis, nonlinear discriminant analysis, support vector machines (SVMs), random forests, and naive Bayes.

[0043] Here, Figure 3 illustrates an example of a method for extracting anatomical landmarks using a trained model. For example, as shown in Figure 3, the first extraction function 152 inputs medical image data acquired by the acquisition function 151 into the landmark extraction model (trained model). The first extraction function 152 then extracts anatomical landmarks based on the coordinate information and labels of anatomical landmarks present in the input medical image data, which are output from the landmark extraction model.

[0044] The following describes the processing of the first extraction function 152 using the example of inputting CT image data of the chest and abdomen (aortic arch, aortic valve, cardiac apex, anterior liver, and posterior liver, etc.) as medical image data into the landmark extraction model. Figure 4 is an example of CT image data of the chest and abdomen of a subject. The first extraction function 152 inputs the CT image data IG to the landmark extraction model stored in the storage device of an external workstation via the NW interface 110.

[0045] The landmark extraction model takes CT image data IG as input and outputs coordinate information of feature points corresponding to multiple anatomical landmarks present on the CT image data IG, such as the aortic arch, aortic valve, cardiac apex, anterior hepatic region, and posterior hepatic region, along with labels (DK, DB, SS, ZK, and KK, etc.) that represent each of these landmarks.

[0046] For example, a landmark extraction model outputs labels that identify feature points (anatomical landmarks) and text data representing the coordinates of those feature points, such as "DK(x1,y1,z1),DB(x2,y2,z2),SS(x3,y3,z3),ZK(x4,y4,z4),KK(x5,y5,z5))". The coordinate information and labels may also be output as data in a format that can be recorded as supplementary information for DICOM.

[0047] The first extraction function 152 then obtains the coordinate information and labels of each anatomical landmark output from the landmark extraction model via the NW interface 110.

[0048] Returning to Figure 1, let's continue the explanation. The second extraction function 153 extracts organ-specific landmarks from the anatomical landmarks extracted by the first extraction function 152, based on the organ-specific landmark information.

[0049] Specifically, the second extraction function 153 refers to organ-specific landmark information and extracts the coordinate information and labels of anatomical landmarks that are registered as organ-specific landmarks from among the anatomical landmarks extracted by the first extraction function 152.

[0050] Organ-specific landmark information is information that associates labels of anatomical landmarks with the names of organs to which the anatomical landmarks identified by those labels belong. For example, "anatomical landmarks: feature points DK, DB, SS" and "organ: heart," or "anatomical landmarks: feature points ZK, KK" and "organ: liver." Organ-specific landmark information is stored, for example, in the memory circuit 120.

[0051] For example, if "Anatomical Landmark: Feature Points DK, DB, SS" and "Organ: Heart" are registered as organ-specific landmark information, the second extraction function 153 will obtain the coordinate information and labels of the anatomical landmarks corresponding to feature points DK, DB, and SS. Also, for example, if "Anatomical Landmark: Feature Points ZK, KK" and "Organ: Liver" are registered as organ-specific landmark information, the second extraction function 153 will obtain the coordinate information and labels of the anatomical landmarks corresponding to feature points ZK and KK.

[0052] Figure 5 shows an example of the results of extracting organ-specific landmarks. Figure 5 shows an example of extracting organ-specific landmarks from the medical image data IG shown in Figure 4. The second extraction function 153 extracts the coordinate information of organ-specific landmarks, represented as feature points DK, DB, SS, ZK, and KK in Figure 5, and the labels DK, DB, SS, ZK, and KK that identify each of them, from the medical image data IG shown in Figure 4.

[0053] Returning to Figure 1, let's continue the explanation. The first identification function 154 identifies the target organ for diagnosis based on the organ to which the organ-specific landmark belongs. Specifically, the first identification function 154 refers to the organ-specific landmark information and identifies the organ corresponding to the label of the anatomical landmark extracted by the second extraction function 153 as the target organ.

[0054] Furthermore, if multiple target organs are identified, the first identification function 154 identifies the organ with the largest number of extracted organ-specific landmarks as the target organ. If the number of extracted organ-specific landmarks is the same, the first identification function 154 identifies the target organ according to a predetermined priority order, for example.

[0055] Furthermore, the first specific function 154 may identify target organs according to priority only, or it may identify target organs according to other criteria. For example, the first specific function 154 may target organs that have organ-specific landmarks closer to the center position on the medical image data IG.

[0056] Furthermore, for example, if information about the lesion site is recorded as supplementary information for the DICOM of the medical image data IG, the first identification function 154 may identify the target organ according to the presence or absence of the lesion site.

[0057] In this case, first, the first specific function 154 refers to the DICOM supplementary information of the medical image data IG acquired by the acquisition function 151 to check whether information about the lesion site is recorded. If information about the lesion site is recorded, the first specific function 154 refers to the organ-specific landmark information to identify the organ-specific landmark corresponding to the lesion site (organ name).

[0058] Next, the first identification function 154 checks whether the organ-specific landmarks extracted by the second extraction function 153 include an organ-specific landmark corresponding to the lesion site recorded as supplementary information in DICOM. If an organ-specific landmark corresponding to the lesion site is extracted by the second extraction function 153, the first identification function 154 identifies the organ corresponding to that organ-specific landmark as the target organ, regardless of the number of organ-specific landmarks extracted.

[0059] Furthermore, for example, the first identification function 154 may pre-determine a threshold for the average value of multiple pixel values ​​corresponding to each organ-specific landmark, and identify the target organ according to the number of organ-specific landmarks that show pixel values ​​exceeding (or falling below) the threshold.

[0060] In this case, the first identification function 154 first calculates the average value of the pixel values ​​of multiple pixels corresponding to each organ-specific landmark extracted by the second extraction function 153 (for example, all pixels within a radius of n pixels from the center coordinates of the organ-specific landmark). Next, the first identification function 154 checks whether the calculated average value of the pixel values ​​for each organ-specific landmark exceeds (or falls below) a predetermined threshold.

[0061] The first identification function 154 then calculates the number of organ-specific landmarks that show pixel values ​​above (or below) a threshold for each organ. For example, the first identification function 154 identifies the organ with the largest number of organ-specific landmarks that show pixel values ​​above (or below) a threshold as the target organ.

[0062] Furthermore, the first specific function 154 may identify the target organ based on information other than medical image data IG. For example, the first specific function 154 may refer to the subject's electronic medical record stored in the server device of the electronic medical record system and identify the target organ based on the lesion site recorded in the electronic medical record.

[0063] Furthermore, for example, the first specific function 154 may refer to examination order information stored in a server device of a Radiology Information Systems (RIS) and identify the target organ based on the information about the organ to be examined recorded in the examination order information.

[0064] Furthermore, for example, the first specific function 154 may receive input from the user specifying a region on the medical image data IG, and identify the target organ according to the content of said input.

[0065] In the example shown in Figure 5, when identifying the target organ based on the number of extracted organ-specific landmarks, the first identification function 154 refers to the organ-specific landmark information and identifies that three organ-specific landmarks were extracted for the heart and two organ-specific landmarks were extracted for the liver. Then, the first identification function 154 identifies the heart, which has a larger number of organ-specific landmarks, as the target organ.

[0066] Returning to Figure 1, let's continue the explanation. The second identification function 155 identifies the segmentation range and starting point based on the identified target organ. Specifically, the second identification function 155 first extracts the contour of the target organ on the medical image data IG based on the medical image data IG acquired by the acquisition function 151 and the organ-specific landmarks of the target organ identified by the first identification function 154. Known edge detection techniques can be used to extract the contour of the target organ.

[0067] The second identification function 155 determines, for example, the area in which edge detection processing will be performed on the medical image data IG, based on organ-specific landmarks of the target organ identified by the first identification function 154. The second identification function 155 then extracts the contour of the target organ by performing a known edge detection process on the determined area.

[0068] Here, the range in which edge detection processing is performed on the medical image data IG is defined based on the location or region where organ-specific landmarks of the target organ exist. For example, the range encompassing each coordinate of the organ-specific landmark of the target organ is defined as the range in which edge detection processing is performed on the medical image data IG.

[0069] Next, the second specific function 155 identifies the area inside the contour of the extracted target organ as the segmentation range. The second specific function 155 also identifies the position within the identified segmentation range that corresponds to the starting point of the segmentation defined in the standard organ model. The standard organ model is, for example, a shape model provided for each organ. The standard organ model is generated, for example, based on the shape of the organs of an adult male or adult female with a standard body type.

[0070] Returning to Figure 1, let's continue the explanation. The segmentation function 156 performs segmentation processing to divide the region of the target organ identified from the medical image data IG. In this embodiment, segmentation refers to the process of distinguishing between the region of interest depicted in the image (the region in which the target organ is depicted) and regions other than the region of interest.

[0071] The display control function 157 controls the display of various information on the display 140. For example, the display control function 157 controls the display of the segmentation results from the segmentation function 156 on the display 140.

[0072] The reception function 158 receives various operation inputs from the user. For example, the reception function 158 receives input for modifying segmentation results. Specifically, the reception function 158 first receives input from the user indicating whether or not to modify the segmentation results. If the user indicates that they wish to modify the results, the reception function 158 receives input from the user specifying the segmentation range and starting point.

[0073] When the reception function 158 receives input from the user specifying the segmentation range and starting point, the segmentation function 156 performs the segmentation process again according to the specified segmentation range and starting point.

[0074] Next, we will describe the processes performed by the medical image processing device 100 configured as described above. Figure 6 is a flowchart showing an example of the processes performed by the medical image processing device 100.

[0075] First, the acquisition function 151 acquires the medical image data IG of the subject to be diagnosed (S1). Specifically, the acquisition function 151 acquires the medical image data IG corresponding to the patient ID of the subject to be diagnosed from the medical image storage device 500.

[0076] Next, the first extraction function 152 extracts multiple anatomical landmarks from the acquired medical image data IG (step S2). Specifically, the first extraction function 152 inputs the medical image data IG into the landmark extraction model. Then, the first extraction function 152 obtains coordinate information and labels representing the anatomical landmarks output from the landmark extraction model.

[0077] Next, the second extraction function 153 extracts organ-specific landmarks (step S3). Specifically, the second extraction function 153 refers to the organ-specific landmark information and obtains the coordinate information and labels of the anatomical landmarks that are registered as organ-specific landmarks among the anatomical landmarks extracted by the first extraction function 152.

[0078] Next, the first identification function 154 identifies the target organ (step S4). Specifically, the first identification function 154 refers to organ-specific landmark information and identifies the organ corresponding to the label of the anatomical landmark extracted in step S3 as the target organ.

[0079] Next, the second identification function 155 identifies the segmentation range and starting point (step S5). Specifically, the second identification function 155 extracts the contour of the target organ based on the medical image data IG acquired in step S1 and the organ-specific landmarks identified in step S4. Then, the second identification function 155 identifies the segmentation range based on the extracted contour. Furthermore, the second identification function 155 identifies the starting point of the segmentation based on the extracted contour of the target organ and a standard model of the target organ.

[0080] Next, the segmentation function 156 performs segmentation processing based on the identified segmentation range and starting point (step S6). Next, the display control function 157 controls the display of the segmentation processing results on the display 140 (step S7). Next, the reception function 158 checks whether or not it has received input from the user for a correction instruction of the segmentation results within a predetermined period (step S8).

[0081] If no correction instructions are received within the specified time (Step S8: No), this process is terminated. On the other hand, if correction instructions are received within the specified time (Step S8: Yes), the segmentation function 156 identifies the segmentation range and starting point according to the user's correction instructions (Step S9). After that, the process proceeds to Step S6.

[0082] As described above, the medical image processing device 100 according to this embodiment extracts a plurality of anatomical landmarks from medical image data IG, and extracts organ-specific landmarks from the extracted plurality of anatomical landmarks based on organ-specific landmark information that associates anatomical landmarks with organs. The medical image processing device 100 also identifies a target organ based on the organ to which the organ-specific landmark belongs. Furthermore, the medical image processing device 100 identifies a segmentation range and a starting point from the identified target organ, and performs segmentation based on the segmentation range and starting point.

[0083] As a result, the medical image processing device 100 according to this embodiment can identify the target organ using the results of anatomical landmark extraction. Therefore, the medical image processing device 100 according to this embodiment can also identify the target organ from medical image data IG of a special imaging range, for example.

[0084] Furthermore, the medical image processing device 100 according to this embodiment can identify the segmentation range and starting point from the identified target organ. Therefore, the medical image processing device 100 according to this embodiment can perform segmentation processing according to the target organ without receiving the segmentation range and starting point from the user.

[0085] Based on the above, the medical image processing device 100 according to this embodiment can efficiently identify the location of a target organ in a medical image.

[0086] The embodiments described above can also be modified and implemented as appropriate by changing some of the configurations or functions of each device. Therefore, the following describes modifications of the embodiments described above as other embodiments. In the following, we will mainly describe the differences from the embodiments described above, and will omit detailed explanations of points that are common with what has already been described. Furthermore, the modifications described below may be implemented individually or in combination as appropriate.

[0087] (modified version) In the embodiments described above, a configuration in which the medical image processing device 100 performs the anatomical landmark extraction process was described. However, the device that performs the anatomical landmark extraction process is not limited to the medical image processing device 100. For example, a medical diagnostic imaging device 200 may perform the anatomical landmark extraction process.

[0088] If the medical imaging diagnostic device 200 is an X-ray CT scanner, for example, the processing circuit of the console device equipped with the X-ray CT scanner performs the anatomical landmark extraction process. In this case, the processing circuit inputs the CT image data acquired and reconstructed by the X-ray CT scanner into a landmark extraction model stored in an external workstation or the like. The processing circuit then acquires the coordinate information and labels of each anatomical landmark output from the landmark extraction model.

[0089] Furthermore, the processing circuit records the coordinate information and labels of each acquired anatomical landmark as supplementary information to the DICOM of the CT image data. The processing circuit then transmits the CT image data, on which the coordinate information and labels of the anatomical landmarks have been recorded as supplementary information to the medical image storage device 500.

[0090] The processing circuit may also generate text data representing the coordinate information and label information of each acquired anatomical landmark. In this case, the processing circuit transmits this data to the medical image storage device 500 along with the CT image data.

[0091] In this modified example, the acquisition function 151 of the medical image processing device 100 can acquire CT image data (medical image data) in which the coordinate information and labels of anatomical landmarks are recorded as DICOM supplementary information.

[0092] Therefore, the medical image processing device 100 according to this modified version can extract organ-specific landmarks on medical image data IG without communicating with an external workstation or the like that stores the landmark extraction model. In other words, according to this modified version, the processing load on the medical image processing device 100 can be reduced, and the position of the target organ in the medical image can be efficiently estimated.

[0093] According to at least one embodiment described above, the location of a target organ in a medical image can be efficiently identified.

[0094] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be implemented in a variety of other forms, and various omissions, substitutions, modifications, and combinations of embodiments are possible without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0095] 100 Medical Image Processing Equipment 110 NW Interfaces 120 Memory circuit 130 Input Interfaces 140 displays 150 Processing Circuits 151 Acquisition function 152 1st extraction function 153 Second extraction function 154 1st specific function 155 Second specific function 156 Segmentation Function 157 Display control function 158 Reception function 200 Medical imaging diagnostic equipment 500 Medical Image Storage Devices S Medical Information Processing System

Claims

1. A first extraction unit extracts anatomical landmarks representing characteristic points of anatomical tissue contained in a medical image from the medical image, A second extraction unit extracts organ-specific landmarks from the anatomical landmarks extracted by the first extraction unit, based on correspondence information that associates identification information of anatomical landmarks with the names of the organs to which the anatomical landmarks belong, and which represent anatomical landmarks characteristic of each organ. A first identification unit identifies the target organ for diagnosis based on the name of the organ to which the organ-specific landmark extracted in the second extraction unit belongs, A segmentation unit that divides the region of the target organ identified from the medical image, Equipped with, The first identification unit, when multiple target organs are identified, identifies the organ with the largest number of organ-specific landmarks as the target organ, and when there are multiple organs with the largest number of organ-specific landmarks, identifies the target organs according to a predetermined priority order. Medical image processing equipment.

2. The aforementioned medical image includes information indicating the presence or absence of a lesion, The first identification unit identifies the target organ based on the supplementary information. The medical image processing apparatus according to claim 1.

3. The system further includes a display control unit that displays the results of the segmentation of the target organ region by the segmentation unit on a display unit. The medical image processing apparatus according to claim 1 or 2.

4. The system further includes a reception unit that receives input from the user to correct the results of the classification of the target organ region, The segmentation unit, in accordance with the correction input, divides the region of the target organ identified from the medical image again. The medical image processing apparatus according to claim 1.

Images