JP5700964B2 - Medical image processing apparatus, method and program - Google Patents

Description

  The present invention relates to a medical image processing apparatus and method for processing a medical image, and a program for causing a computer to execute the medical image processing method, and in particular, a displayed organ to be diagnosed from among a large number of stored images The present invention relates to an image processing apparatus, a method, and a program for supporting acquisition of a comparative image to be compared with the other image.

  In recent years, in the field of medical images, in addition to an X-ray imaging apparatus, an X-ray CT (Computed Tomography) apparatus, an ultrasound (US) diagnostic apparatus, an MRI (Magnetic Resonance Imaging) apparatus, SPECT (Single photon emission computed tomography) and Modality using various technologies such as a PET (Positron Emission Tomography) apparatus is used. An image to which various photographing methods, image processing methods, and display methods are applied using such various modalities is proposed, and the analysis of the image diagnosis is supported by improving the visibility of the target image. However, when multiple types of images obtained using multiple modalities are prepared as medical images become diverse, a comparative image for comparative interpretation is extracted from the medical image to be diagnosed. For this reason, it has become necessary to select a menu for specifying a comparison image.

  Here, by collecting images showing symptoms necessary for diagnosis from medical images of a plurality of patients and creating a reference image group for comparison, a group name for each body part for each reference image belonging to the reference image group, Create a reference image group setting file by associating the group name, feature, display condition, etc. for each symptom, and according to the reference image group setting file, the body position is determined from the header information of the diagnosis target image, etc. A method has been proposed in which a symptom is specified and a reference image group corresponding to the specified body position and symptom is called (see Patent Document 1). According to the method described in Patent Document 1, since the reference image group is classified for each body part and symptom, the doctor can quickly read out a desired reference image group from a plurality of reference images.

  In Patent Document 2, three-dimensional image data including a recognized part among other three-dimensional image data is recognized in three-dimensional image data including a large number of tomographic images. Describes a method for automatically searching for and displaying.

JP 2003-284690 A JP 2009-232981 A

  However, in the methods described in Patent Document 1 and Patent Document 2, an image obtained by photographing a portion common to the image of the portion displayed as a diagnosis target is extracted as a reference image group, but the common portion When there are many images taken by multiple modalities or various types of images obtained by various image processing methods or photographing methods, it takes time to extract a comparative image from these many images. It will be a thing.

  The present invention has been made in view of the above circumstances, and in particular, a comparison image is selected from among a plurality of images taken by different types of modalities and / or a plurality of images to which different types of image processing methods are applied. The object is to reduce the burden on the user to extract and display.

  According to the medical image processing apparatus of the present invention, an image selection unit that selects a desired image from a plurality of images representing an organ to be diagnosed, and a plurality of parts constituting a part of the organ on the desired image are individually provided. An image display means for displaying the image in a selectable manner, an image information acquisition means for acquiring image information representing a modality for capturing the selected desired image and an image processing method of the image, and the plurality of displayed parts And a comparison unit for determining one or more comparison images from the plurality of images with priority based on the designated part and the acquired image information. And a display means for displaying the determined list of comparison images in a selectable order in the order of priority.

  The medical image processing method according to the present invention selects a desired image from a plurality of images representing an organ to be diagnosed, and individually designates a plurality of parts constituting a part of the organ on the desired image. Image information representing the modality and the image processing method for capturing the selected desired image is obtained, and one of the displayed plurality of parts is designated, and the designated Based on the part and the acquired image information, one or more comparison images are determined from the plurality of images with priorities, and the determined list of comparison images is selected in descending order of priority. It is displayed as possible.

  The image processing program according to the present invention includes a computer, an image selection unit that selects a desired image from a plurality of images representing an organ to be diagnosed, and a plurality of images that constitute a part of the organ on the desired image. Image display means for individually displaying each part so that it can be specified, image information acquisition means for acquiring image information representing a modality for capturing the selected desired image and an image processing method for the image, and the displayed A designation unit for designating one of a plurality of parts, and determining one or more comparison images from the plurality of images with priority based on the designated part and the acquired image information Comparison image determining means, and a display means for selectively displaying the determined list of comparison images in the order of priority.

  Here, “a plurality of parts constituting part of the organ” means parts constituting the heart such as the left and right ventricles, the atrium, the mitral valve, and the coronary artery when the organ is the heart. In addition, a lesion occurring in a part of the site is also included.

  The modality includes various modalities as long as it is a modality for obtaining a medical image. Typical examples include an X-ray imaging apparatus, an X-ray CT (Computed Tomography) apparatus, and an ultrasound (US) diagnosis. There are apparatus, MRI (Magnetic Resonance Imaging) apparatus, SPECT (Single photon emission computed tomography) and PET (Positron Emission Tomography) apparatus.

  The “image processing method” refers to an image generation method applied to an image and / or a display method applied to an image. For example, image generation methods include a method of reconstructing an image from volume data such as MIP, MPR, volume rendering, stretch CPR, straight CPR, projection CPR, and pseudo angiography. The display method includes a function image display method by bullseye mapping, a fusion image in which functional images and morphological images obtained by different modalities or different image processing are superimposed, and display of axial, sagittal, and coronal images. There are methods such as displaying a tomogram parallel to the short axis or the long axis of the heart and reproducing a cine moving image.

  Further, the “list of comparison images” means a list in which serial numbers and symbols for identifying comparison images are displayed, but not only numbers and symbols but also one or more comparison images and each comparison image. As long as the priority can be confirmed, it includes those expressed in various forms. For example, the thumbnails of the comparison images may be displayed in a table, and the comparison images may be displayed in the order of priority.

  In the medical image processing apparatus according to the present invention, the comparison image determination unit estimates a diagnosis purpose of the organ based on the designated part and the image information, and is more suitable for the estimated diagnosis purpose. It is preferable to include a priority determination unit that determines the priority of the comparative image to be higher.

  Further, the priority determination means according to the present invention estimates a further diagnostic purpose based on the estimated diagnostic purpose, and determines to make the priority of the comparison image suitable for the estimated further diagnostic purpose higher. It is preferable that

  In the medical image processing apparatus according to the present invention, the organ may be a heart.

  In the medical image processing apparatus according to the present invention, the image processing method may represent any one of stretch CPR, straight CPR, projection CPR, and pseudo-angiography. CPR (Curved Planar Reformation) is a technique for displaying an arbitrary two-dimensional curved surface on a two-dimensional cross section by applying MPR. Mainly, stretch CPR (Stretched CPR) that projects a curved surface on a plane, straight CPR (Straightened CPR) that projects a meandering blood vessel straight, projection CPR (Projected CPR) that projects a curved surface vertically There are different types. Pseudo-angiography is a display method in which an image obtained by an angiographic examination method for performing radiography on a blood vessel into which a contrast agent has been injected is reconstructed by volume data such as CT and displayed in a pseudo manner. It is used to make it easy to visually recognize the shape and the like.

  In the medical image processing apparatus according to the present invention, the designation means performs an operation of selectively designating a part that is a part of the organ on the displayed image of the organ. It may be operated or a touch panel may be operated with a finger or the like. In addition, the designating unit may limit a part that can be specified to some of a plurality of parts according to the acquired image processing method.

  The medical image processing apparatus according to the present invention further includes a lesion extraction unit that extracts a lesion of the designated part according to the designation of the part of the designation unit, and the image information acquisition unit includes the image information acquisition unit as the image information. Information that represents a lesion may be further acquired.

  In the medical image processing apparatus according to the present invention, it is preferable that the comparative image determination means determines a comparative image by giving a high priority to the time-series images according to the estimated diagnostic purpose.

  The medical image processing apparatus according to the present invention preferably further includes comparison image display means for extracting and displaying a comparison image selected from the displayed comparison image list from the plurality of images. Here, “extract and display” means not only displaying a selected image as a comparative image, for example, when displaying a tomographic image, but also selecting the selected image from image data. It also includes performing appropriate image processing and reconstructing and displaying as a comparative image.

  In the medical image processing apparatus according to the present invention, it is preferable that the comparative image display means further determines a display protocol for displaying the comparative image based on the estimated diagnostic purpose.

  Furthermore, in the medical image processing apparatus according to the present invention, it is preferable that the comparison image display condition is such that a display protocol for displaying the comparison image is determined to be the same as a display protocol for the desired image.

  According to the medical image processing apparatus, the medical image processing apparatus method, and the medical image processing apparatus program according to the present invention, priority is given to one or more comparison images from a plurality of images based on the designated part and the acquired image information. Select a suitable image for comparison from multiple images in order of priority, simply by specifying a region, so that the list of determined comparison images can be selected in descending order of priority. Can do. Therefore, since a user such as an image interpretation doctor can reduce the extraction work for extracting and displaying the comparison image, the burden on the user when selecting the comparison image can be reduced.

  Further, in the medical image processing apparatus according to the present invention, the comparison image determination means estimates the diagnosis purpose of the organ based on the designated part and the image information, and the priority of the comparison image that is more suitable for the estimated diagnosis purpose Since the priority determination means for determining to be higher is provided, an image suitable for comparison can be selected with high priority based on the estimated diagnostic purpose. It is possible to reduce the trouble of selecting a comparison image suitable for the purpose of diagnosis and reduce the burden on the user when selecting a comparison image.

  Further, the priority determination means according to the present invention estimates a further diagnostic purpose based on the estimated diagnostic purpose, and determines to make the priority of the comparison image suitable for the estimated further diagnostic purpose higher. If this is the case, it is possible to estimate the further diagnostic purpose of the next step from the estimated diagnostic purpose, and to increase the priority of the comparison image suitable for the further diagnostic purpose. It is possible to reduce the trouble of selecting an image and reduce the burden on the user when selecting a comparison image.

  In the medical image processing apparatus according to the present invention, when the image processing method represents any one of stretch CPR, straight CPR, projection CPR, and pseudo-angiography, a specific diagnostic purpose Therefore, the diagnostic purpose can be estimated with high probability. Stretch CPR, straight CPR, projection CPR, and pseudoangiography are all image processing methods used for diagnosing vascular abnormalities. For this reason, when an image corresponding to such an image processing method is selected and the image processing method of the present embodiment is applied, it is less likely that an erroneous diagnostic purpose is estimated, and the diagnostic purpose is further improved from the image processing method. Can be displayed with high priority and can be selected with high accuracy, and the user can select a comparative image with high accuracy. Therefore, the burden on the user when selecting the comparative image can be reduced.

  Further, in the medical image processing apparatus according to the present invention, when the designating unit restricts the parts that can be designated according to the acquired image processing method, the diagnosis target of the user is determined based on the image processing method. Because it is possible to limit the specification of a part so that it is not possible to specify a part that is estimated not to be a diagnosis target, it is possible to prevent the user from specifying the part by mistake, and the burden on the user when selecting a part Can be reduced.

  In addition, the medical image processing apparatus according to the present invention further includes a lesion extraction unit that extracts a lesion of the designated part in accordance with the designation of the part of the designation unit, and the image information acquisition unit includes information representing the lesion as the image information. Since the priority of the comparison image can be determined based on the image information including the lesion at the specified site, it is possible to estimate a more probable diagnostic purpose and more suitable for the diagnostic purpose. Since the comparison image is displayed so as to be selectable with high priority and the user can select the comparison image with high accuracy, the burden on the user when selecting the comparison image can be reduced.

  In the medical image processing apparatus according to the present invention, when the comparison image determining means determines the comparison image by giving a high priority to the time-series images according to the estimated diagnosis purpose, it is more suitable for the diagnosis purpose. Since the comparison image can be displayed with high priority and can be selected, and the user can select the comparison image with high accuracy, the burden on the user when selecting the comparison image can be reduced.

  When the comparison image selected from the displayed comparison image list further includes comparison image display means for extracting and displaying the comparison image from the plurality of images, the medical image processing apparatus according to the present invention can generate a desired image from the comparison image list. A desired comparison image can be displayed simply by selecting a comparison image, and a user such as an image interpretation doctor can reduce the extraction work for extracting and displaying the comparison image. The burden can be reduced.

  In the medical image processing apparatus according to the present invention, when the comparison image display means further determines a display protocol for displaying the comparison image based on the estimated diagnosis purpose, the comparison image display protocol is set as the user. Since the comparison image can be displayed with a display protocol suitable for diagnosis without setting each time the image is displayed, high-precision diagnosis can be supported, and the user's comparison image can be saved without having to set the user's display protocol. The burden at the time of selection can be reduced.

  Further, in the medical image processing apparatus according to the present invention, when the comparison image display condition determines that the display protocol for displaying the comparison image is the same as the display protocol for the desired image, the display protocol for the comparison image is changed. Since the comparison image can be displayed with a display protocol suitable for diagnosis without the user having to set it every time it is displayed, high-precision diagnosis can be supported. The burden at the time of image selection can be reduced.

The figure which shows schematic structure of the medical information system to which the image processing apparatus by embodiment of this invention is applied. Functional block diagram of the medical image processing apparatus according to the first embodiment The flowchart which shows the process performed in 1st Embodiment. The figure showing the example of the diagnostic purpose estimation table in 1st Embodiment The figure showing the example of the further diagnostic purpose estimation table and priority determination table in 1st Embodiment The figure showing the example of the related diagnostic objective matching table which matches the related diagnostic objective of 1st Embodiment. The figure which shows the candidate list of the comparison image in 1st Embodiment. The figure which shows the designation | designated method of the site | part in 1st Embodiment, and the displayed comparison image list The figure which shows the comparative image displayed in 1st Embodiment The figure which shows the modification of the comparison image displayed in 1st Embodiment. The figure which shows the further modification of the comparison image displayed in 1st Embodiment. Functional block diagram of the medical image processing apparatus according to the second embodiment The flowchart which shows the process performed in 2nd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, the following embodiment and its modification are applicable also to other embodiment, without changing the essence.

  FIG. 1 is a diagram showing a schematic configuration of a medical information system to which an image processing apparatus according to an embodiment of the present invention is applied. As shown in FIG. 1, the medical information system according to the present embodiment includes a medical image photographing device (modality) 1, an image interpretation workstation (WS) 2, an image server 3, an image database 4, and a part recognition terminal 5. And are connected so as to communicate with each other. Note that each device in the present embodiment is controlled by a program installed from a recording medium such as a CD-ROM. The program may be installed after being downloaded from a server connected via a network such as the Internet.

  The modality 1 is a device that generates an image data of an image representing a part of the subject by photographing the part to be examined, adds the incidental information such as examination information and patient information to the image data, and outputs the image data. Is included. The incidental information is in a format that complies with a standardized standard such as the DICOM standard and a manufacturer-specific standard such as a modality thereof. Specific examples of the modality 1 include a CT apparatus, an MRI apparatus, a PET apparatus, a SPECT apparatus, and an ultrasonic imaging apparatus. In the present embodiment, the modality 1 refers to a plurality of modalities including a plurality of types of apparatuses among a CT apparatus, an MRI apparatus, a PET apparatus, a SPECT apparatus, and an ultrasonic imaging apparatus. In the present embodiment, the image data includes a plurality of three-dimensional images that represent examination target portions of a subject acquired by a plurality of types of apparatuses among a CT apparatus, an MRI apparatus, a PET apparatus, a SPECT apparatus, and an ultrasonic imaging apparatus. It is assumed that the image data is configured as an aggregate of image data of axial discontinuity images (referred to as tomographic images) with a predetermined slice interval and slice thickness.

  The image interpretation workstation 2 is an apparatus used by an image interpretation doctor who is a user for image interpretation and generation of an image interpretation report. One or two high-definition displays for various displays, a keyboard and a mouse for various inputs A hard disk for storing various data such as various programs and image data, including a medical image processing program of the present invention, a CPU for controlling the image diagnosis support apparatus according to the present embodiment by executing the various programs, The memory 9 is a work area at the time of program execution, and is composed of a communication interface connected to a network via a bus.

  The image interpretation workstation 2 requests image browsing to the image server 3, various image processing on the image received from the image server 3, various analysis processing including automatic detection and highlighting of structures and lesion-like portions in the image, Display, interpretation report creation support, interpretation report registration request and browsing request to interpretation report server (not shown), interpretation report received from interpretation report server, and the like.

  Here, the image processing apparatus of the present invention is mounted on the image interpretation workstation 2. The interpretation workstation 2 has a diagnostic purpose estimation table T1 in which the diagnostic purpose, the image information, and the body part are associated with each other for estimating the diagnostic purpose, and a further diagnostic purpose based on the estimated diagnostic purpose. A further diagnostic purpose estimation table T2, a priority determination table T3B that determines the priority of the comparison image for each diagnostic purpose, and a related diagnostic purpose association table T4 that associates the estimated diagnostic purpose with the related diagnostic purpose are stored. Yes. 4A shows an example of a diagnostic purpose estimation table T1, FIG. 4B shows an example of a further diagnostic purpose estimation table T2 and priority determination tables T3A and T3B, and FIG. 4C shows an associated diagnostic purpose determination table T4.

  The interpretation workstation 2 performs various analysis processes instructed by the user, such as CT coronary artery analysis, CT cardiac function analysis, MR coronary artery analysis, MR cardiac function analysis, MR delayed contrast analysis, etc. An analysis application is installed to do this.

  For example, the analysis application recognizes a region of interest, extracts a lesion by analyzing a tomographic image from a tomographic image including the region of interest using, for example, a maximum value projection method (MIP method), or the like. For example, by creating an image obtained by performing various image processing such as stretch CPR for the region of interest, or performing various displays such as a bullseye image of the heart or a tomographic image parallel to the long axis of the heart, It means various techniques for improving the visibility of each part that constitutes a part of an organ, and also a lesion included in the part.

  The interpretation workstation 2 first acquires image information of the designated part and the diagnosis target image in accordance with an instruction from the user when interpreting the target image that is the diagnosis target. Then, with reference to the table T1, table T2, etc. based on the acquired designated part and the image information of the diagnosis target image, the comparison images obtained from the plurality of three-dimensional image data are determined with priorities, and the priorities are determined. The comparative image specifying information is displayed on the display in this order. The image processing method of the present invention will be described in detail later.

  The image server 3 is obtained by installing a software program that provides a function of a database management system (DataBase Management System: DBMS) on a general-purpose computer having a relatively high processing capability. The image server 3 includes a large-capacity storage in which the image database 4 is configured. This storage may be a large-capacity hard disk device connected to the image server 3 by a data bus, or connected to a NAS (Network Attached Storage) and a SAN (Storage Area Network) connected to the network 10. It may be a disk device. The image server 3 also has a communication interface that communicates with the modality 1 and the image interpretation workstation 2 via the network 10.

  When receiving the image registration request from the modality 1, the image server 3 arranges the image in the database format and registers it in the image database 4. In the image database 4, the above-described image data of the three-dimensional image, supplementary information, and part recognition result information described later are registered. The accompanying information includes, for example, an image ID for identifying individual images, a patient ID for identifying a subject, an examination ID for identifying examinations, a unique ID (UID) assigned to each image, Examination date when the image was generated, examination time, type of modality used in the examination to obtain the image, patient information such as patient name, age, gender, examination site (imaging site), imaging information (imaging protocol) , Imaging sequence, imaging method, imaging conditions, use of contrast agent, etc.) Information such as a series number or collection number when a plurality of images are acquired in one examination can be included.

  When the image server 3 receives a browsing request from the image interpretation workstation 2 via the network 10, the image server 3 searches for an image registered in the image database 4 and sends the extracted image to the image interpretation workstation 2 as the request source. Send. When an operation for requesting browsing of an image to be interpreted is performed by a user such as an image interpretation doctor, the image interpretation workstation 2 transmits a view request to the image server 3 and acquires an image necessary for image interpretation. Then, analysis processing such as automatic lesion detection processing is executed on the image in response to a request from the image interpretation doctor.

  The part recognition terminal 5 recognizes the part of the subject included in the three-dimensional image before or after registration in the image database 4 of the three-dimensional image composed of a plurality of tomographic images acquired in the modality 1. The part recognition process is performed. The site recognition terminal 5 also has a communication interface for communicating with the modality 1 and the image interpretation workstation 2 via the network 10. The part recognition terminal 5 starts the recognition process by receiving a notification from the image server 3 that the registration of the three-dimensional image in the image database 4 is completed. The recognition process may be started by an instruction from the interpretation workstation 2.

  As the part recognition process, various kinds of processes that can extract a part constituting a part of an organ can be used. For example, regions such as the heart ventricle, myocardium, and coronary artery can be accurately extracted from the tomographic image to be observed. For example, an automatic recognition technique disclosed in Japanese Patent Application Laid-Open No. 2004-141612 can be used. For example, various methods have been proposed for extracting coronary artery regions from volume data, including the method shown in Andrzej Szymczak, et al. Coronary vessel tree frin 3D imagery: A topological approach. Medical Image Analysis 2006. ing.

Further, for example, the method proposed by the applicant in Japanese Patent Application No. 2009-048679 and Japanese Patent Application No. 2009-069895 can be used for extracting the heart region. Hereinafter, an outline of the processing described in those documents will be described. The site recognition terminal 5 extracts a region corresponding to the heart (hereinafter, heart region) from the volume data based on a predetermined algorithm. In the process of extracting the heart region, the position S _ref that characterizes the shape of the heart, such as the position of the aortic valve, the position of the mitral valve, the position of the apex, etc., is also specified. The coordinates of the specified position are stored in the memory, and are used when defining the reference coordinate system in the processing described later.

  Next, the part recognition terminal 5 sets a rectangular parallelepiped region including the heart region in the volume data as a search range, and searches for a linear structure included in the search range based on a predetermined algorithm. Furthermore, a point estimated as a point on the core line of the coronary artery is detected based on the linear structure detected by the search. In the following description, a point estimated as a point on the coronary artery path is referred to as a candidate point or a node.

  The search for the linear structure is performed by calculating eigenvalues of a 3 × 3 Hessian matrix for each local region within the search range. In the region including the linear structure, one of the three eigenvalues of the Hessian matrix is a value close to 0, and the other two are relatively large values. The eigenvector corresponding to the eigenvalue whose value is close to 0 indicates the principal axis direction of the linear structure. The site recognition terminal 5 determines the likelihood of a linear structure for each local area based on the eigenvalues of the Hessian matrix, and detects the center point of the local area where the linear structure is identified as a candidate point.

  In the search for the linear structure, it is preferable to generate a plurality of resolution-specific data (Gaussian pyramids) having different resolutions by performing resolution conversion on the data in the search range, and to perform repeated search (scanning) at different resolutions. In the above search method, the linear structure cannot be identified when the diameter (width) of the local region is smaller than the blood vessel diameter, but it is possible to identify a linear structure of any size by performing a search with different resolutions. Become. Thereby, candidate points can be detected without omission from a thick blood vessel serving as a backbone to a thin blood vessel at the end.

  Next, the part recognition terminal 5 connects the candidate points detected by the search based on a predetermined algorithm. Thereby, a tree structure composed of candidate points and branches (edges) connecting the candidate points is constructed. The coordinate information of a plurality of detected candidate points and vector information indicating the direction of the branches are stored in a memory together with the candidate points and branch identifiers.

  Subsequently, the part recognition terminal 5 identifies in detail the shape of the coronary artery for each detected candidate point based on the value (CT value) of the surrounding voxel. Specifically, the outline of the coronary artery (the outer wall of the blood vessel) is identified in a cross section perpendicular to the coronary artery path. The shape is identified using a known segmentation technique represented by Graph-Cuts.

Finally, the site recognition terminal 5 defines a reference coordinate system with the position S _ref of the aortic valve, mitral valve, and apex stored in the heart region identification process as the reference position. Also, by defining the length from the apex to the aortic valve as 1, the scale of the coordinate system is normalized. And the coordinate value memorize | stored in memory in each above-mentioned process is converted into the coordinate value of a reference | standard coordinate system. That is, data representing candidate points, branch positions, coronary artery contours, and the like are normalized. The normalized information is associated with the information before normalization and stored in the memory as a recognition result.

  Further, the present invention is not limited to this embodiment. Instead of using the automatic recognition technology by the part recognition terminal 5, the user performs a process of manually extracting a part of the organ, and the extracted part of the organ is used as a recognition result. It may be stored in a memory.

  The site recognition terminal 5 transmits the recognition result to the image server 3. The image server 3 registers the received recognition result information in the image database 4, and transmits the recognition result information from the image server 3 to the interpretation workstation 2 together with the three-dimensional image according to an instruction from the interpretation workstation 2.

  Next, a configuration related to the image processing apparatus of the first embodiment will be described. In the present specification, the same portions are denoted by the same reference numerals, and description thereof is omitted. FIG. 2 is a functional block diagram of the image processing apparatus according to the first embodiment.

  The image selection unit 11 selects an image instructed by a user or the like with an input tool such as a mouse, a pen touch tool, or a touch panel from a plurality of images representing an organ to be diagnosed registered in the image database 4 via the network 10. .

  The image information acquisition means 12 acquires the header information of the selected diagnosis target image, the modality of capturing a desired image selected from display software or the like that displays the diagnosis target image, and image information representing the image processing method of the image To do.

  In the present embodiment, the “image processing method” refers to an image generation method applied to an image and / or a display method applied to an image. For example, image generation methods include a method of reconstructing an image from volume data such as MIP, MPR, volume rendering, stretch CPR, straight CPR, projection CPR, and pseudo angiography. The display method includes a function image display method by bullseye mapping, a fusion image in which functional images and morphological images obtained by different modalities or different image processing are superimposed, and display of axial, sagittal, and coronal images. There are methods such as displaying a tomogram parallel to the short axis or the long axis of the heart and reproducing a cine moving image.

  The image display means 13 displays on a display device such as a display provided in the interpretation workstation 2 so that a plurality of parts constituting a part of the organ can be individually designated. In the present embodiment, it is assumed that a plurality of parts constituting a part of an organ are recognized in advance by the part recognition terminal 5 and stored in the image database 4. The image display means 13 acquires a plurality of parts included in a desired image from the image database 4 and displays each part from which the desired image has been acquired so that it can be individually specified. In addition, it is not limited to this embodiment, You may perform part recognition by the program mounted in the interpretation workstation 2. The part recognition is not performed in advance, and the part recognition process may be performed on the selected image in accordance with the selection of the image by the image selection unit 11.

  The designation unit 14 designates one of the plurality of parts displayed by the image display unit 13 in accordance with an input from an input tool such as a mouse, a pen touch tool, or a touch panel.

  The comparison image determination unit 15 determines one or more comparison images with priority from a plurality of images registered in the image database 4 based on the designated part and the acquired image information.

  The comparison image determining means 15 estimates the purpose of organ diagnosis based on the designated region and image information, and determines the priority of the comparison image that is more suitable for the estimated diagnosis purpose. Determination means 16 is provided. Furthermore, the priority determination means 16 according to the present embodiment estimates a further diagnostic purpose based on the estimated diagnostic purpose, and makes the priority of the comparison image suitable for the estimated further diagnostic purpose higher. To decide.

  The image processing apparatus according to the present embodiment estimates a further diagnostic purpose based on a diagnostic purpose estimation table T1 in which a region, image information, and a diagnostic purpose are associated, and a diagnostic purpose estimated based on the diagnostic purpose estimation table T1. A further diagnostic purpose estimation table T2, a priority determination table T3B for determining the priority of the comparison image for each diagnostic purpose, and a related diagnostic purpose association table T4 are stored. FIG. 4A is a diagram illustrating an example of the diagnostic purpose estimation table T1. FIG. 4B shows a further diagnosis purpose estimation table T2 and priority determination tables T3A and T3B for determining the priority of the comparison image for each diagnosis purpose. FIG. 4C shows a related diagnostic purpose association table T4 that associates related diagnostic purposes. Although details will be described later, the priority determination table T3A is created by the priority determination means 16 based on the image information of the diagnosis target image and the diagnosis purpose D1.

  In this embodiment, the priority determination means 16 first estimates the diagnostic purpose D1 based on the diagnostic purpose estimation table T1 in which the part, the image information, and the diagnostic purpose are associated with each other. Then, the priority determination means 16 estimates further diagnostic objectives D21 and D22 from the estimated diagnostic objective D1 based on the further diagnostic objective estimation table T2. This assumes that the diagnostic purpose to be subsequently diagnosed is predicted as a further diagnostic purpose based on the diagnostic purpose D1 currently diagnosed with the diagnostic image. Then, for each further diagnostic purpose, the modality of the comparative image that should have a higher priority is determined based on the priority determination tables T3A and T3B that determine the priority of the comparative image for each diagnostic purpose. Then, for a plurality of comparison images extracted from the image database 4, the priority of the comparison images taken with the determined modality is increased, and the comparison image list L1a is created. Details will be described later along the flow of image processing of the first embodiment.

  As shown in FIG. 4A, the user associates the diagnostic purpose with the modality of the image used as the diagnostic image of the part and the image processing method in advance based on various findings, creates the diagnostic purpose estimation table T1, and interprets the interpretation work. Store in station 2. Here, in the present embodiment, the diagnostic purpose represents rough abnormality analysis of each part such as abnormality of a specific blood vessel and abnormality of cardiac function of a specific part. However, the present invention is not limited to this embodiment, and the diagnostic purpose may represent a disease name or a more detailed diagnostic purpose.

  Corresponding condition IDs “01” and “02” in the diagnostic purpose estimation table T1 shown in FIG. Various image processing methods are associated with each other. For example, the correlation condition ID “01” in the diagnostic purpose estimation table T1 is associated with the part “coronary artery 51”, the diagnostic purpose “stenosis of the coronary artery 51”, the modality “CT”, and the image processing method “stretch CPR”. It has been. If the part and image information acquired by the image information acquisition unit 12 represent the part “coronary artery 51”, the modality “CT”, and the image processing method “stretch CPR”, the priority determination unit 16 determines the diagnostic purpose estimation table. Based on T1, the diagnosis purpose “stenosis of coronary artery 51” associated with these is estimated.

  Further, as indicated by ID “03” in the diagnostic purpose estimation table T1, image processing for superimposing two types of modalities “CT” and “SPECT” and images obtained by both modalities on the site “coronary artery 54” The method “fusion” is associated with the diagnostic purpose “coronary artery 54 and abnormal cardiac function in the vicinity of coronary artery 54”. In other words, when the image processing method refers to a composite image processing method such as superposition of a plurality of images, the image processing method includes a plurality of abnormalities such as abnormalities of the coronary artery itself and abnormalities of cardiac function as diagnostic purposes. Analysis may be associated, and a plurality of different modalities may be associated as modalities. In this case, the priority determination unit 16 estimates the diagnostic purpose if the site “coronary artery 54”, the modalities “CT” and “SPECT”, and the image processing method “fusion” are acquired by the designation unit 14 and the image information acquisition unit 12. Based on the table T1, the diagnostic purpose “coronary artery 54 and abnormal cardiac function near the coronary artery 54” associated therewith is estimated.

  Further, in the association condition IDs “04” to “08” in the diagnostic purpose estimation table T1, the region “myocardial 52” corresponds to one region “myocardial 52” according to image information such as modality or image processing method. Thus, different types of diagnostic purposes are associated with each other. In this case, the priority determination unit 16 estimates a diagnostic purpose that matches the combination of the part and image information acquired by the image information acquisition unit 12 based on the diagnostic purpose estimation table T1. In other words, the priority determination unit 16 estimates the diagnostic purpose when the region “myocardium 52”, the modality “CT”, and the image processing method “the long axis image of the left ventricle” are acquired by the specifying unit 14 and the image information acquisition unit 12. Based on the table T1, the diagnosis purpose “abnormal cardiac function of the left ventricle” associated therewith is estimated. Further, the priority determination means 16 is based on the diagnostic purpose estimation table T1 if the region “myocardium 52”, modality “CT”, and image processing method “Bullseye” are acquired by the specifying means 14 and the image information acquisition means 12. The diagnostic purpose “necrosis of the myocardium 52” associated with these is estimated.

  Similarly, in the association condition IDs “9” to “11” in the diagnostic purpose estimation table T1, there are various types of diagnostic purposes, different modalities of different types, and various image processing methods for the part “valve 53”. It is associated.

  In diagnosis practice in the medical field, when one abnormality is diagnosed, the progress of the state of the abnormality from the past is confirmed, and the state of the abnormality associated with the abnormality is often diagnosed together. Based on such diagnostic practice, the further diagnostic purpose estimation table T2 of FIG. 4B is based on the estimated current diagnostic purpose D1 and checks past images for the same diagnostic purpose as this diagnostic purpose D1. An estimation rule for further diagnosis purpose estimation is set so that the diagnosis is further estimated as a further diagnosis purpose D21, and an abnormality analysis associated with the diagnosis purpose D1 is estimated as the further diagnosis purpose D22. Yes. It is assumed that such a diagnostic purpose estimation rule is appropriately set in advance by a user such as a diagnostician.

  In order to estimate the further diagnostic purpose D22, the related diagnostic purpose association table T4 associates an arbitrary number of related abnormalities as related diagnostic purposes with respect to a plurality of diagnostic purposes, etc. Created by the user and stored in the interpretation workstation 2.

  The priority determination table T3A prioritizes one or more modalities with which the priority determination means 16 may take a comparison image according to the estimated diagnostic purpose D1 in the step of determining the priority. They are created in association with each other in order and stored in the interpretation workstation 2. The priority determination table T3B is created by associating one or more modalities that a user such as a diagnostician may take a comparative image in order of priority for each diagnosis purpose, and storing them in the interpretation workstation 2 Is. As shown in FIG. 4B, the priority determination table T3B is set to be a SPECT device and an ultrasound (US) diagnostic device in descending order of priority of the modality of the comparison image with respect to the necrosis of the myocardium 52. Yes.

  In the diagnostic purpose estimation table T1 shown in FIG. 4A, for explanation, only one part representing a coronary artery, myocardium, and valve is shown as a part, but a plurality of coronary arteries constituting a part of the heart, a plurality of parts Assume that each part of the myocardium, a plurality of valves, a plurality of ventricles, a plurality of atria, and the diagnostic purpose corresponding to each part are actually registered in the diagnostic purpose estimation table T1. Here, it is assumed that the coronary artery 51 points to the left coronary artery. The part may be defined as long as it defines a part of an organ, and a part obtained by dividing the left coronary artery into finer sections may be defined as one part.

  In the present embodiment, a user such as an interpreting doctor, based on actual diagnosis practice, determines the diagnosis purpose estimation table T1, the estimation rule of the further diagnosis purpose estimation table T2, the related diagnosis purpose association table T4, and the priority determination. A table 3B is created. However, all these tables statistically collect modalities, image processing methods, and data for diagnostic purposes from information such as diagnostic images actually used for diagnosis and interpretation reports associated with the diagnostic images. It is frequently used for the combination of the purpose, diagnosis target part, and image information, the combination of a plurality of subsequent diagnosis purposes, the combination of the diagnosis purposes diagnosed in relation, and the combination of the diagnosis purpose and the modality of the diagnosis target image Some or all of the tables may be created by a computer so that the objects are associated with each other.

  The display means 17 displays the list of comparison images determined by the comparison image determination means 15 in a selectable manner on a display device such as a display provided in the image interpretation workstation 2 in descending order of priority.

  The comparison image display means 18 generates a comparison image selected from the displayed comparison image list L1 from a plurality of images of the same patient registered in the image database 4, and is provided with a display or the like provided in the image interpretation workstation 2. Display on the display device. In the embodiment of the present specification, various displays by the image display unit 12, the display unit 17, and the comparison image display unit 18 are displayed on the same display. However, the display is not limited to this embodiment, and may be displayed on a plurality of display devices.

  Next, processing performed in the present embodiment will be described. FIG. 3 is a flowchart showing processing performed in the present embodiment.

  First, the image selection unit 11 selects an image registered in the image database 4 from the image database 4 in accordance with a user's desired image instruction, and transmits the image to the medical image processing apparatus (S101). Here, the diagnosis target image 71 is selected.

  Next, the image display means 13 displays a plurality of parts constituting a part of the organ so that they can be individually designated (S102). For example, in the diagnosis target image 71, a plurality of parts such as a plurality of coronary arteries, a plurality of ventricles, a plurality of atria, a plurality of valves, etc. are recognized with respect to a tomographic image taken by CT, and these are displayed on the display screen. One of the parts can be specified.

  Then, the image information acquisition unit 12 acquires the image information of the diagnosis target image 71 from the analysis application applied to the header information and the diagnosis target image (S103). For example, the site “coronary artery 51”, modality “CT”, and image processing method “stretch CPR” are acquired from the image information of the diagnosis target image 71.

  And the designation | designated means 14 designates a site | part on a diagnostic object image by a user's etc. input (S104). FIG. 6 shows an image diagram in which a list of comparison images is displayed according to the designation of the part. FIG. 6 is an example of a diagnostic screen displayed when the analysis application for the coronary artery diagnosis support function is selected. As shown in FIG. 6, a straight CPR image 71 and a stretch CPR image 72 showing the coronary artery 51 and a volume rendering (3D) image 73 representing the entire extracted coronary artery region are displayed on the diagnosis screen 91. As shown in FIG. 6, when the coronary artery 51 is designated in the straight CPR image 71 by the arrow P1, a list L1 of comparison images to be described later is displayed.

  And the priority determination means 15 estimates the diagnostic objective corresponding to the designated site | part with reference to the diagnostic objective estimation table T1 mentioned above based on the designated site | part and image information (S105).

  For example, as shown in FIG. 6, in the diagnosis target image 71, when the designation unit 14 receives designation of the user's coronary artery 51, the priority determination unit 16 included in the comparison image determination unit 15 performs the diagnosis shown in FIG. 4A. Based on the association conditions in the purpose estimation table T1, the association that most closely matches the combination of the specified region “coronary artery 51”, the modality “CT” that is the acquired image information, and the image processing method “straight CPR” The diagnostic purpose “stenosis of the coronary artery 51” represented by the condition ID “01” is estimated as the diagnostic purpose D1.

  The priority determination unit 16 determines that the image information of the diagnostic purpose related to the designated part is not included in the diagnostic purpose estimation table T1 if there is no matching condition in the diagnostic purpose estimation table T1 where the designated part and the acquired image information all match. The diagnostic purpose corresponding to the matching condition that matches more frequently is set as the estimated diagnostic purpose D1.

  And the priority determination means 16 estimates further diagnostic objectives D21 and D22 based on the estimated diagnostic objective D1, based on the further diagnostic objective estimation table T2. That is, based on the estimation rule of the further diagnostic purpose estimation table T2 of FIG. 4B, the diagnostic purpose that the user will perform next from the current diagnostic purpose D1 is estimated as further diagnostic purposes D21 and D22.

  Specifically, the priority determination means 16 uses the estimation rule 1 “diagnosis objective same as the diagnostic objective (comparison with the past image)” described in the further diagnostic objective estimation table T2 in FIG. The same “stenosis of the coronary artery 51 (comparison with the past image)” as D1 is estimated as the diagnostic purpose D21. Similarly, based on the estimation rule 2 “diagnostic objective related to diagnostic objective D1” described in the further diagnostic objective estimation table T2, the related objective correspondence table T4 corresponds as a diagnostic objective related to stenosis of the coronary artery 51. The attached “necrosis of the myocardium 52” is estimated as a further diagnostic purpose D22.

  Next, the priority determination unit 16 extracts an image including the organ to be diagnosed from the image database 4 from a plurality of images in which the organ to be diagnosed is recognized by the part recognition terminal 5, and determines the priority for the extracted comparison image. (S106). In addition, the technique of the well-known document 2 may be used for extraction of an image including an organ to be diagnosed, and various organ detection techniques can be applied.

  The priority determination means 16 is based on the priority determination tables T3A and T3B corresponding to the estimated further diagnostic purposes D21 “stenosis of the coronary artery 51 (comparison with the past image)” and D22 “necrosis of the myocardium 52”, respectively. Priorities of a plurality of comparison images are determined.

  Here, the priority determination means 16 is acquired with the same modality as the currently displayed diagnosis target image among the past images based on the diagnostic purpose D21 “stenosis of the coronary artery 51 (comparison with the past images)”. In order to increase the priority of the comparison image, the priority determination table 3A is created by associating the CT apparatus having the same modality as the diagnosis target image with the diagnosis purpose D21 “stenosis of the coronary artery 51 (comparison with the past image)”.

  As shown in FIG. 5, the priority determination means 16 prioritizes the comparison image so that the CT device, SPECT device, and US device associated with the priority determination tables T3A and T3B have higher priorities in this order. Determine the degree. In the present embodiment, when a plurality of further diagnostic purposes are estimated, the modality corresponding to the comparison with the past image is set in advance to have a higher priority than the modalities corresponding to other diagnostic purposes. Yes. In addition, when there are a plurality of further diagnostic purposes, it is preferable that a criterion for giving priority among the plurality of further diagnostic purposes can be appropriately changed by a user such as a doctor using a GUI or the like. For example, after designating a designated part, a plurality of estimated further diagnostic purposes may be displayed in such a manner that priority can be set, and the priority of further diagnostic purposes may be determined in accordance with user input.

  FIG. 5 is a comparison image candidate list L1a representing the result of extracting the comparison image by the priority determination means 16. As shown in FIG. 5, the priority determination means 16 extracts a comparison image candidate from both an image photographed in the past diagnosis of the same patient and an image photographed in the main diagnosis, and each comparison Priorities are assigned to the images, and the comparison image candidate list L1a is associated with information specifying the image, such as the modality in which the comparison image was taken, and the image ID, series name, and shooting date and time for specifying the comparison image. sign up.

  Further, as shown in FIG. 5 as an example of an ultrasound (US) diagnostic apparatus, the priority determination means 16 does not give priority to those that cannot be extracted and displayed, such as no volume data in the image database 4. Remove from comparison image.

  In the present embodiment, the priority determination means 16 estimates a further diagnostic purpose based on the further diagnostic purpose estimation rule set in the further diagnostic purpose estimation table T2, but it is not always necessary to estimate a further diagnostic purpose. Instead, there is provided a priority determination table T3B ′ in which a step for estimating a further diagnostic purpose is omitted and a reference for determining the priority of the comparison image for each diagnostic purpose is set for a plurality of diagnostic purposes. There may be an application example in which priorities of a plurality of comparative images are determined based on the priority determination table T3B ′ corresponding to the diagnostic purpose D1.

  The priority determination means 16 is not limited to this embodiment, and if the priority can be given to the comparison image suitable for the diagnostic practice, the priority determination means 16 rearranges the comparison image candidate list L1a by various methods to obtain the priority. Can be determined. A modification of the priority determination means 16 will be described later.

  The comparison image determination means 15 creates a comparison image list L1 in which only the comparison images for which priority levels have been determined are extracted from the comparison image candidate list L1a.

  Then, the display unit 17 displays the list L1 of comparison images in a selectable order in order of priority (S107). As shown in FIG. 6, a comparison image list L <b> 1 is displayed on the right side of the diagnosis screen 91 so as to be selectable in descending order of priority.

  The list of comparison images is not limited to the present embodiment, and may be expressed in various forms as long as one or more comparison images and the priority order of each comparison image can be confirmed. For example, the thumbnails of the comparison images may be displayed in a table, or the comparison images may be displayed in an order of priority.

  The image processing apparatus preferably includes the comparison image display unit 18 as in the present embodiment. The comparison image display means 18 generates a comparison image selected from the displayed comparison image list L1 from a plurality of images of the same patient registered in the image database 4 or by performing image processing as necessary. Display, and the process ends (S108). FIG. 7 is a diagram in which the comparison image 61a in the comparison image list L1 in FIG. 6 is right-clicked with the mouse as indicated by the arrow P1, and the comparison image 61A is displayed at the lower left of the diagnosis screen 91. Note that the upper left is a diagnosis target image 71 in which a region is designated in FIG.

  As shown in FIG. 7, in the present embodiment, the comparison image display means 18 applies the same image processing method as the image processing method for the diagnosis target image to the comparison image and displays it. When the image processing method for the comparison image cannot be made the same as the image processing method for the diagnosis target image, the comparison image is displayed by a predetermined image processing method. By making the image processing method for displaying the comparison image the same as the image processing method for the diagnosis target image, it is possible to display the comparison image in the same display manner as the diagnosis target image, and to support high-accuracy diagnosis. This saves the user from setting the image processing method and reduces the burden on the user when selecting the comparison image. In particular, this effect is remarkable when a comparison image representing the same part imaged in the past with respect to the target image is selected. Note that the image processing method for the comparison image may be an image processing method designated by the user, such as enabling the image processing method to be designated by a pull-down menu when selecting the comparison image from the list.

  In FIG. 7, the comparison image display means 18 determines that the display protocol for displaying the comparison image is the same as the display protocol for the diagnosis target image. The display protocol here refers to display conditions such as a color template, gradation processing, enhancement processing parameters, window width of a medical image, window level, and display sequence. As shown in FIG. 7, the target image 71 representing the designated part and the comparison image 61A are such that the designated part has the same size, the same image processing, the same arrangement, and the same color processing in the same size window. It is displayed as follows.

In addition, by determining the display protocol for displaying the comparison image to be the same as the display protocol of the diagnosis target image, the comparison image can be displayed in the same display mode as the desired image, and the comparison image is suitable for diagnosis. Therefore, it is possible to support highly accurate diagnosis, and further, it is possible to save the user from setting the display protocol and reduce the burden on the user when selecting the comparison image. In particular, this effect is remarkable when a comparison image representing the same part imaged in the past with respect to the target image is selected.

  Note that, for example, when the comparison image is associated with another priority determination rule based on an image processing method or an image protocol by the comparison image determination unit 15 or the like, or depending on the image processing method or modality, the same display protocol is applied. If this is not appropriate, the already associated image processing method and image protocol are not changed.

  FIG. 8 is a diagram illustrating a modification example of the display of the diagnosis target image 74 and the comparison image 64. A diagnosis target image 74 generated by the volume rendering method is displayed in the window W1 of FIG. 8, and a comparison image 64 generated by the volume rendering method is displayed in the window W2. As shown in FIG. 8, the comparison image display means 18 displays the selected desired image 74 and comparison image 64 in the same color template, gradation processing, enhancement processing parameters, window width of the medical image, window level, etc. It is displayed by protocol, and the display size, view angle, and alignment of the specified part and the part in the comparison image are performed. In particular, when a comparison image representing the same part taken in the past with respect to the target image is selected, the target image and the comparison image are displayed with the same display method, display protocol, and part display mode as much as possible. By doing so, both can be easily understood and compared with high accuracy, and diagnostic efficiency is improved.

  Further, a plurality of comparison images may be displayed according to the designation of the part. By displaying a plurality of comparison images, it is possible to compare the diagnosis target portions of the target image with high accuracy and increase the diagnosis efficiency.

  Further, as shown in W11 and W12 of FIG. 9, when a plurality of diagnosis target images generated from the first image data are displayed by different reconstruction methods (or display methods), the diagnosis selected by the user Not only the comparison image corresponding to the target image but also each of the plurality of displayed images, each comparison image is reconstructed from image data corresponding to the selected image (hereinafter referred to as second image data). May be displayed.

  As an example, FIG. 9 shows an example in which, when images generated from the same image data are displayed by two different display methods, a comparative image is displayed for each image having a different display method. In FIG. 9, as a diagnosis target image generated from the same image data, an image 751 obtained by reconstructing a three-dimensional image acquired by the CT apparatus in the window W11 by the volume rendering method and a three-dimensional image 752 acquired by the SPECT apparatus. A fusion image 75 of a bull's eye image 762 of a three-dimensional image acquired by SPECT and an image 761 of a three-dimensional image acquired by CT and a three-dimensional image acquired by SPECT are displayed in the window W12.

  The priority determination means 16 estimates the diagnostic purpose in accordance with the designation of the part on the diagnostic object image 761 or 762, and obtains the volume data imaged by the CT corresponding to the diagnostic purpose and the volume data imaged by the SPECT. Extracted from the image database and determined as a comparison image for each of the diagnosis target images 761 or 762.

  The comparison image display means 18 uses the same image processing method as the image 75 for the volume data imaged by the extracted CT and the volume data imaged by the SPECT, from the image 651 by the CT apparatus and the image 652 by the SPECT apparatus. A fusion image 65 is generated and displayed on W21, and a fusion image 66 composed of an image 661 by the CT apparatus and an image 662 by the SPECT apparatus, which is the same image processing method as the target image 76, is generated and displayed on W22. In this way, by displaying the plurality of comparative images 65 and 66 with the same image processing method for each of the plurality of target images 75 and 76, the diagnosis target of the target image represented by the plurality of display methods is displayed. The part can be compared with high accuracy, and the diagnostic efficiency is improved.

  Further, the comparison image display means 18 includes a correspondence table in which a display protocol of an image for displaying the comparison image is associated for each diagnosis purpose, and the window size of the image for displaying the comparison image based on the estimated diagnosis purpose Further, a display protocol such as a layout or a color template may be further determined. Since the comparison image can be displayed with a display protocol suitable for diagnosis without the user setting the display protocol of the comparison image every time it is displayed, high-accuracy diagnosis can be supported, and the user has to set the display protocol. And the burden on the user when selecting the comparison image can be reduced.

  As described above, according to the present embodiment, one or more comparison images are determined from a plurality of images with priority based on the designated portion and the acquired image information, and the determined comparison is performed. Since the list of images is displayed so as to be selectable in descending order of priority, an image suitable for comparison can be selected from a plurality of images in order of priority only by designating a part. Therefore, since a user such as an image interpretation doctor can reduce the extraction work for extracting and displaying the comparison image, the burden on the user when selecting the comparison image can be reduced.

  In addition, according to the first embodiment, the comparison image determination unit 15 estimates the purpose of organ diagnosis based on the designated part and image information, and the comparison image determination unit 15 matches the estimated diagnosis purpose. Since priority determination means for determining a higher priority is provided, an image suitable for comparison can be selected with a higher priority based on the estimated diagnostic purpose, and the user's diagnostic purpose can be further improved. It is possible to reduce the trouble of selecting a comparison image suitable for the user and to reduce the burden on the user when selecting a comparison image.

  Furthermore, the priority determination means 16 according to the first embodiment estimates a further diagnostic purpose based on the estimated diagnostic purpose, and increases the priority of the comparison image that matches the estimated further diagnostic purpose. Therefore, it is possible to estimate the further diagnostic purpose of the next step from the estimated diagnostic purpose, and to increase the priority of the comparison image suitable for the further diagnostic purpose. It is possible to reduce the trouble of selecting a comparison image suitable for the user and to reduce the burden on the user when selecting a comparison image.

  As in the first embodiment, when the comparative image determination unit 15 estimates a more relevant diagnostic purpose based on the diagnostic purpose directly estimated from the part and the image information, there is a possibility of the simultaneous occurrence. An image suitable for a related diagnostic purpose such as a lesion can be used as a comparative image, reducing the effort of selecting a comparative image more suitable for the user's diagnostic purpose, and reducing the burden on the user when selecting a comparative image. .

  As shown in the priority determination table T3B, in the present embodiment, the priority of the comparative image photographed with a modality particularly suitable for the purpose of diagnosis is set high. In this case, since the user can easily select the comparison image with high accuracy, the burden on the user when selecting the comparison image can be reduced.

  Further, in the first embodiment, by utilizing the existing information analyzed by the analysis application as the image information, a time for new image processing or the like is not generated, and a lesion extraction processing unit or the like is separately provided. Since it is not necessary to provide the image information, the image information can be acquired easily and efficiently.

  In the present embodiment, since the comparison image selected from the displayed comparison image list L1 is extracted from a plurality of images or generated and displayed by image processing, the comparison image display means 18 is provided. A desired comparison image can be displayed simply by selecting a comparison image, and a user such as an image interpretation doctor can reduce the extraction work for extracting and displaying the comparison image. The burden can be reduced.

  Hereinafter, modifications of the priority determination unit 16 according to the present embodiment will be described.

  When there are a plurality of estimated diagnostic purposes, the priority determination unit 16 preferably determines the priority of the comparative image corresponding to the diagnostic purpose having a high probability.

For example, a still another priority determination table T3C (not shown) in which a highly probable diagnostic purpose is determined for each image processing method and is associated with a higher priority of the comparative image corresponding to the highly probable diagnostic purpose Can be provided. In particular, if the image processing method represents any one of stretch CPR, straight CPR, projection CPR, and pseudo-angiography, these image processing methods all observe vascular abnormalities. Since the image processing method is particularly suitable for the image processing method “stretch CPR”, “straight CPR”, “projection CPR”, and “pseudo-angiography” in the priority determination table T3C. A high diagnostic purpose “stenosis of the coronary artery 51” is created in association with each other. If the image processing method of the diagnosis target image represents any one of the stretch CPR, straight CPR, projection CPR, and pseudo-angiography, the priority determination unit 16 Based on the priority determination table T3C, among the plurality of estimated diagnostic purposes, “stenosis of the coronary artery 51” is determined as a highly probable diagnostic purpose, and the priority of the comparison image corresponding to “stenosis of the coronary artery 51” Can be set higher than the comparative image corresponding to other estimated diagnostic purposes.

  That is, in particular, the image processing method is a diagnostic object corresponding to an image processing method particularly suitable for a specific diagnostic purpose, such as an image processing method of any one of stretch CPR, straight CPR, projection CPR, and pseudo-angiography. When an image is selected and the image processing method of the present embodiment is applied, it is unlikely that an erroneous diagnostic purpose is estimated, and a comparative image that is more suitable for the diagnostic purpose can be selected from the image processing method with higher priority. Since the user can select the comparison image with high accuracy, the burden on the user when selecting the comparison image can be reduced.

  Further, for example, in the diagnosis purpose estimation table T1 of FIG. 4A, for example, a plurality of different types of image processing methods such as tomographic image and moving image reproduction may be applied at the same time. Different types of image processing methods may be associated.

  When a plurality of modalities and / or a plurality of different types of image processing methods are associated with each diagnosis purpose ID, the image information acquisition unit 12 uses one or more modalities and one as the image information of the diagnosis target image. One or more image processing methods are acquired, and the priority determination means 16 compares the plurality of modalities associated with the diagnostic purpose estimation table T1 and / or a plurality of different types of image processing methods with one or more modalities. As the combination of one or more image processing methods matches, it is determined that the probability of the purpose of diagnosis is high. The priority determination means 16 sets the priority of the comparison image corresponding to the diagnosis purpose having a high probability among the plurality of estimated diagnosis purposes higher than the comparison image corresponding to the other estimated diagnosis purposes. To do. Rather than estimating the diagnostic purpose by associating a part, modality, and one image processing method, the correlation between the part, modality, and multiple image processing methods is compared with the associating conditions to estimate the diagnostic purpose. Can often estimate the diagnostic purpose.

  Further, for each diagnosis purpose, a further priority determination table T3D (not shown) associated with the priority of the comparative image that can be displayed by the image processing method suitable for the diagnosis purpose is prepared, and the priority determination means 16 The priority of the comparison image that can be displayed by the image processing method suitable for the diagnosis purpose may be increased from the plurality of comparison images corresponding to the estimated diagnosis purpose based on the priority determination table T3D.

  For example, in the priority determination table T3D (not shown), for the diagnostic purpose “abnormal cardiac function”, in the diagnostic practice, moving image reproduction is suitable for the comparison image of the cardiac function. Set a higher priority for "Play video" in advance. The image processing method suitable for the purpose of diagnosis is statistically extracted from a plurality of interpretation reports used for diagnosis, and the image processing method of the diagnosis target image used frequently for the purpose of diagnosis is manually or The priority determination table T2B can be created in association with the computer.

  When the site “valve” is designated, the valve is a site that is greatly involved in abnormal cardiac function, and therefore the diagnosis purpose “abnormal cardiac function” is estimated with high probability. Of the plurality of extracted comparison images, a high priority can be given to a comparison image that can be reproduced as a moving image, based on the priority determination table T3D. That is, the priority of the time-series image is determined to be high as a comparative image particularly suited to the estimated diagnostic purpose “abnormal cardiac function”.

  As described above, when the priority of the comparative image that can be displayed by the image processing method that is particularly suitable for the diagnostic purpose is set high, the user can easily select the comparative image that is more suitable for the diagnostic purpose with high accuracy. The burden on the user when selecting a comparison image can be reduced.

  As an application of the first embodiment, the priority determination unit 16 may determine a higher priority for a comparative image having image information similar to the acquired image information. For example, among the plurality of extracted comparison images, the same modality and image processing method as the acquired image may be used, and the priority of the comparison image taken by another date or another patient may be determined high. . As a specific example, as shown in the ID “03” of the diagnostic purpose estimation table T1 in FIG. 4A, a fusion image obtained by superimposing the respective images acquired by the CT apparatus and the SPECT apparatus on the coronary artery 54, You may make it determine the priority of the comparative image represented by the "fusion image" which overlap | superposed each image acquired by CT apparatus and SPECT apparatus image | photographed on another date and time highly.

  By increasing the priority of comparison images having image information similar to the acquired image information, it is possible to easily select comparison images represented by similar imaging methods, image processing methods, and image display methods. The burden at the time of image selection can be reduced.

  Further, among the extracted comparison images, a higher priority may be set for the most recently referenced date and time. In addition, regarding the modality and the image processing method, the priority of the comparative image having the modality and the image processing method that are most frequently referred to for each diagnosis purpose may be increased. By deciding the priority of the comparison image according to the user's use result of the comparison image, it is possible to easily select the comparison image reflecting the diagnosis practice of the user, so that the burden on the user when selecting the comparison image is reduced. Can do.

  Based on the diagnosis purpose or image information, the priority determination table and the plurality of prioritization rules as described above are provided, and each priority determination table and each rule are provided for each diagnosis purpose or image information of each comparison image. The priority score may be calculated, and the calculated scores may be added to determine the priority in descending order of score. The priority score may be appropriately weighted for each rule. By determining the priority of the comparison image according to rules provided from various viewpoints, the priority is evaluated more objectively. Therefore, since the comparison image can be easily selected based on the objectively evaluated priority, the burden on the user when selecting the comparison image can be reduced.

  Moreover, the priority determination means 16 does not necessarily need to give a different priority to the some comparison image corresponding to each diagnostic objective. That is, the priority determination means 16 determines the priority of the comparison image for each diagnostic purpose, such as when there are a plurality of estimated diagnostic purposes, and the same priority is assigned to the plurality of comparison images corresponding to one diagnostic purpose. May be determined. For example, in the first embodiment, the modalities “SPECT” and “US” of the priority determination table T3B corresponding to the necrosis of the myocardium 52 may have the same priority.

  In the present embodiment, the priority determination unit 16 also prioritizes an image shot with a modality that is not registered in the priority determination table T2 among the plurality of extracted comparison images. Show it in the comparison image list at a certain time. In this embodiment, since the image condition for increasing the priority is registered in the priority determination table, an image shot with a modality that is not registered in the priority determination table is registered in the priority determination table. The priority is determined to be lower than the image captured by the modality.

  Another modification of the above embodiment is shown below.

  Regardless of this embodiment, various tables such as the diagnostic purpose estimation tables T1, T2, the priority determination table T3B, and the related diagnostic purpose association table T4 can be edited by the user at the time of creation and after creation. The diagnostic purpose estimation tables T1 and T2 may be individually varied depending on the user, location, and time used, such as an interpreting doctor, a clinician, an interpreting room, an operating room, before an operation, during an operation, and after an operation.

  In the image processing of FIG. 3, the acquisition of image information (S103) and the display of a plurality of parts (S102) do not have to be performed in this order. After the display of a plurality of parts (S102), the acquisition of image information is performed. (S103) may be performed.

  Further, the diagnostic purpose estimation tables T1, T2, the priority determination table T3B, the related diagnostic purpose association table T4, etc. according to the first embodiment may be divided or integrated into any number of tables. The table may be a combination of some of the tables.

  The comparison image is preferably an image of the same patient as the target image, but is not limited thereto, and may be an image of another patient or a model image for reference.

  The comparative image determination unit 15 is configured to perform the diagnostic purpose even when the image database 4 does not include an image having the same image information as the image display method or the image processing method associated with the diagnostic purpose in the diagnostic purpose estimation table T1. When there is image data captured with the same modality as that associated with the diagnostic purpose of the estimation table T1, and a comparative image can be obtained by reconstructing the image data, an extractable image can be determined as the comparative image. preferable. Then, the comparison image display means 18 includes various reconstruction means for automatically reconstructing and displaying the comparison image from the image data in accordance with the user's selection of the comparison image, and the comparison means displays the comparison image. It is preferable to generate and display. It is possible to easily select a comparative image having a high priority by the user, and it is possible to reduce the trouble of the user extracting a comparative image having a high priority.

  As a modification of the present embodiment, the designation unit 14 may be configured such that the parts that can be designated are limited according to the acquired image processing method. For example, if the heart being diagnosed is displayed with a cross-section parallel to the short axis of the heart or a bullseye display, it is likely that a myocardial abnormality has been diagnosed, so that a region other than the myocardium cannot be selected. In addition, when the heart image processing method to be diagnosed is pseudoangiography, the possibility of analyzing the blood vessel is high and the possibility of diagnosing a valve abnormality is low, so that the valve cannot be selected. Can be displayed. In addition, if the heart to be diagnosed is displayed in a cross section parallel to the long axis of the heart, and if there is a possibility that any of the blood vessels, myocardium, or valve is subject to analysis, the blood vessels, myocardium It is desirable to display so that all of the valves can be selected.

  As described above, since a selectable part is limited from a plurality of parts constituting a part of the organ based on the image processing method, the user can diagnose by making only the specified part selectable. Only the part corresponding to the purpose can be selected. Therefore, a user such as an image interpretation doctor can prevent the user from specifying a part by mistake, and can reduce the burden on the user when selecting a part.

  Next, a second embodiment will be described.

  FIG. 10 is a functional block diagram of the image processing apparatus according to the second embodiment, and FIG. 11 is a flowchart showing a processing flow of the image processing method according to the second embodiment.

  In the second embodiment, the medical image processing apparatus according to the first embodiment is further provided with a lesion extraction unit 19 that extracts a lesion at a designated site in accordance with the designation of the site of the designation unit. The image information acquisition unit 12 further acquires information representing the lesion extracted by the lesion extraction unit 19 as image information.

  The lesion extraction means 19 is constituted by a plurality of lesion extraction units, that is, CAD, each capable of extracting a plurality of different types of lesions. The lesion extraction means 19 is provided as a program installed in the interpretation workstation 2.

  As the lesion extraction means 19, various automatic lesion extraction methods that extract at least the presence or absence of a lesion can be applied to a designated site. For example, it is possible to extract a lesion at a specific site, such as a technique for detecting a blood vessel abnormality as disclosed in Japanese Patent Application Laid-Open No. 2004-329929, a technique for detecting a known abnormality of cardiac function such as left ventricular ejection fraction (Ejection Fraction), etc. Various techniques can be used. For calculating the left ventricular segmentation rate, for example, various techniques for segmenting the left ventricle diastole and systole as disclosed in US Pat. No. 7,596,258 can be applied.

  When the present embodiment is applied to organs other than the heart, examples of techniques that can be used for the lesion extraction means 19 include “helical” disclosed in JP 2003-225231 A, JP 2003-271924 A, and Kubota et al. Evaluation of lung cancer computer diagnosis support system using CT images, "IEICE Technical Report, IEICE Technical Report, pp.41-46, MI2001-41 (2001-09), Lung Cancer Detection Technology, Wakita et al.," Intellectual CAD for Diffuse Lung Disease ”, Ministry of Education, Culture, Sports, Science and Technology Grant-in-Aid for Scientific Research on Specific Areas“ Support for Intelligent Diagnosis of Multidimensional Medical Images ”, 4th Symposium Proceedings, pp. 45-54, 2007 , Ground-Glass Opacity (GGO), Crazy-Paving, Honeycomb shadow, Emphysema shadow, Granular shadow and other diffuse lung disease detection technology, Wakita et al. Liver cancer detection based on measurement ", computer support Liver cancer detection technology described in Journal of Diagnostic Imaging Society, Vol.10, No.1, Mar 2007, "Intelligent CAD based on understanding of normal structure", MEXT Science Grant-in-Aid for Scientific Research on Specific Areas, Multidimensional The technology for detecting hepatocellular carcinoma, hepatic cysts, hepatic vascular species, hepatic vascular bleeding, cerebral bleed, and cerebral bleed, as described in "Support for Intelligent Diagnosis of Medical Images", pp.55-60, 2007 Can be used.

  Moreover, the abnormal shadow candidate detection technique shown in Japanese Patent Application Laid-Open No. 10-97624 and the calcification region detection technique shown in Japanese Patent Application Laid-Open No. 8-215183, which are the applications of the present applicant, can also be used.

  An image processing method according to the second embodiment will be described with reference to FIG. The steps from S201 to S204 are the same as the steps from S101 to S104, and will be omitted. The difference from the first embodiment is that lesion extraction by the lesion extraction means 19 is performed on the part including the designated position after S204 (S205). By the step of S205, the lesion extraction means 19 acquires the presence / absence of the extracted lesion and information representing the lesion. And the result of lesion extraction is registered in the image database 4. The image information acquisition unit 12 further acquires information representing a lesion from the image database 4 as image information.

  When the coronary artery 51 is designated on the diagnosis target image taken by the CT apparatus and displayed by the straight CPR method, the lesion extraction means 19 is a portion where the diameter of the coronary artery is smaller than a predetermined threshold with respect to the coronary artery 51 Detect if there is any. That is, the presence or absence of coronary artery stenosis is automatically extracted.

  In the step of estimating the diagnostic purpose, if a lesion is extracted, the diagnostic purpose is further estimated using the lesion name (S206).

  The association table T4 according to the second embodiment further associates lesion information with the image information of the first diagnostic purpose estimation table T1. That is, the site, modality, image processing method, and lesion information are associated with each diagnosis purpose. When the lesion extraction unit 19 recognizes the stenosis of the coronary artery, the comparison image determination unit 15 most closely matches the combination of the acquired modality, the image processing method, and the extracted lesion from the association table T4. Estimate the diagnostic purpose associated with the combination of modality, image processing method and lesion information. Here, the diagnostic purpose “stenosis of the coronary artery 51” is estimated.

  Then, a priority order is determined for the comparison image in accordance with the estimated diagnostic purpose (S207). Similar to S105, candidates for comparative images corresponding to both the estimated diagnostic purposes “stenosis of coronary artery 51” and “abnormality of myocardium 52” are extracted, and the priority is determined in the same manner as in S106. If no lesion is extracted, the diagnosis purpose is estimated from the region, the image processing method, and the modality as in S105, and the priority is determined in the same manner as in S106. Since the steps from S208 to S209 are the same as those from S107 to S108, description thereof will be omitted.

  According to the second embodiment, since the priority of the comparison image can be determined based on the image information including the lesion of the designated part, it is possible to estimate a more probable diagnostic purpose and fit the estimated diagnostic purpose. Since the comparison image is displayed with high priority so that the user can select the comparison image with high accuracy, the burden on the user when selecting the comparison image can be reduced.

  Further, in the above description, only the case where an image representing the heart is selected is illustrated, but by applying the above method, it is possible to diagnose a tissue composed of a plurality of parts such as the head (brain) and the lung field. Clearly, a device can be constructed that lists useful comparison images in order of priority.

  The above embodiment is a client / server type system, but one computer has a volume data storage means, a volume data selection means, an observation target extraction means, an observation image generation means, a map generation means, an index value. Functions as a calculation means, a transition information generation means, a prediction means, and a display control means may be provided. In addition, the client-side processing may be distributed by a plurality of computers, for example, causing another computer to execute a part of the processing executed by the image interpretation workstation 2 in the above embodiment.

  Moreover, as a device constituting the system, such as an input device and a display, all known devices can be employed. For example, a joystick can be used instead of a mouse, or a touch panel can be used instead of a display.

DESCRIPTION OF SYMBOLS 1 Modality 2 Interpretation workstation 3 Image server 4 Image database 5 Recognition processing terminal 11 Image selection means 12 Image information acquisition means 13 Image display means 14 Designation means 15 Comparison image determination means 16 Priority determination means 17 Display means 18 Comparison image display means 19 lesion extraction means L1 comparison image list T1, T2 correspondence table

Claims (13)

Image selecting means for selecting a desired image from a plurality of images representing an organ to be diagnosed;
An image display means for displaying at least one or more of a plurality of parts constituting a part of the organ on the desired image so as to be individually designated;
Image information acquisition means for acquiring image information representing a modality for capturing the selected desired image and an image processing method of the image;
Designation means for designating one of the plurality of displayed parts;
A diagnostic purpose of the desired image is estimated based on the designated part and the acquired image information, and a diagnostic purpose different from the estimated diagnostic purpose is estimated based on the estimated diagnostic purpose. A comparison image from the plurality of images by estimating a further diagnostic purpose including at least one and deciding to prioritize an image of the plurality of images that is more suitable for the estimated further diagnostic purpose. A comparative image determining means for determining a candidate for
A medical image processing apparatus, comprising: a display unit configured to display the determined list of comparison image candidates in a selectable order from the highest priority.   The comparison image determination means estimates the estimated diagnostic purpose and a diagnostic purpose corresponding to an abnormality occurring in relation to the abnormality corresponding to the estimated diagnostic purpose as the further diagnostic purpose. The medical image processing apparatus according to claim 1. The comparison image determination means is based on a diagnostic purpose estimation table in which a diagnostic purpose, a target portion of an image diagnosis, an image modality used for the image diagnosis, and an image processing method of the image used for the image diagnosis are associated with each other. , Estimating a diagnostic purpose corresponding to the designated part and the acquired image information as a diagnostic purpose of the desired image,
Based on a further diagnostic purpose estimation table in which a diagnostic purpose and a diagnostic purpose corresponding to an abnormality that occurs in association with the abnormality corresponding to the diagnostic purpose are associated, the related abnormality that corresponds to the estimated diagnostic purpose The medical image processing apparatus according to claim 2, wherein a diagnostic purpose corresponding to the is estimated.   The medical image processing method according to any one of claims 1 to 3, wherein the image processing method represents any one of a stretch CPR, a straight CPR, a projection CPR, and a pseudo angiography. Image processing device.   The image display means designates only a part associated with the image processing method of the desired image based on information associating an image processing method with a part that is highly likely to be diagnosed by the image processing method. The medical image processing apparatus according to any one of claims 1 to 4, wherein the medical image processing apparatus is displayed as a possible part. In accordance with the designation of the part of the designation means, further comprising a lesion extraction means for extracting a lesion of the designated part,
The medical image processing apparatus according to claim 1, wherein the image information acquisition unit further acquires information representing a lesion as the image information.   The comparison image determination means determines a candidate for a comparison image by giving a higher priority to an image that can be reproduced as a moving image than an image that cannot be reproduced as a moving image when the estimated diagnostic purpose is abnormal cardiac function. The medical image processing apparatus according to claim 1, wherein the medical image processing apparatus is a medical image processing apparatus.   8. The medical image processing apparatus according to claim 1, further comprising comparison image display means for displaying an image selected from the displayed list of comparison image candidates as a comparison image. .   9. The medical image processing apparatus according to claim 8, wherein the comparison image display means further determines a display protocol for displaying the comparison image based on the estimated diagnostic purpose.   The medical image processing apparatus according to claim 9, wherein the comparison image display condition is such that a display protocol for displaying the comparison image is determined to be the same as a display protocol for the desired image.   The medical image processing apparatus according to claim 1, wherein the organ is a heart. A method for operating a medical image processing apparatus, comprising:
Select the desired image from multiple images representing the organ to be diagnosed,
Displaying at least one or more parts among a plurality of parts constituting a part of the organ on the desired image so as to be individually designated,
Obtaining image information representing the modality of photographing the selected desired image and the image processing method of the image;
In response to a user input, specify one of the displayed plurality of parts,
A diagnostic purpose of the desired image is estimated based on the designated part and the acquired image information, and a diagnostic purpose different from the estimated diagnostic purpose is estimated based on the estimated diagnostic purpose. A comparison image from the plurality of images by estimating a further diagnostic purpose including at least one and deciding to prioritize an image of the plurality of images that is more suitable for the estimated further diagnostic purpose. Determine the candidates for
A method of operating a medical image processing apparatus, wherein the determined list of comparison image candidates is displayed so as to be selectable in descending order of priority. Computer
Image selecting means for selecting a desired image from a plurality of images representing an organ to be diagnosed;
An image display means for displaying at least one or more of a plurality of parts constituting a part of the organ on the desired image so as to be individually designated;
Image information acquisition means for acquiring image information representing a modality for capturing the selected desired image and an image processing method of the image;
Designation means for designating one of the plurality of displayed parts;
A diagnostic purpose of the desired image is estimated based on the designated part and the acquired image information, and a diagnostic purpose different from the estimated diagnostic purpose is estimated based on the estimated diagnostic purpose. A comparison image from the plurality of images by estimating a further diagnostic purpose including at least one and deciding to prioritize an image of the plurality of images that is more suitable for the estimated further diagnostic purpose. A comparative image determining means for determining a candidate for
A medical image processing apparatus program for functioning as display means for displaying the determined list of comparison image candidates in a selectable order from the highest priority.