JP6261218B2 - Display processing device, display processing method and program, and image classification device - Google Patents

Description

  The present invention relates to a technique for classifying image data categories.

  The result of diagnosis by the doctor observing the medical image is compiled into an image diagnosis report. The image diagnosis report describes an image finding regarding an abnormal shadow discovered by a doctor from a medical image and an image diagnosis name for which a medical classification name of the abnormal shadow is determined. For the image findings, the type of image findings (for example, spinous processes, notches, etc.) and the evaluation value (for example, many, strong, etc.) are described. In many cases, both the image finding and the image diagnosis name are determined based on the knowledge and experience of the doctor. Therefore, if the doctor's knowledge and experience are different, different judgments may be made regarding image findings and image diagnosis names.

  In order to reduce the variation of judgments by doctors, doctors need to be educated. In order to conduct education related to image diagnosis, a system that presents image data to which image findings and image diagnosis names have been assigned is necessary, and an image database that stores image data to be presented by this system is required. is necessary. However, it is not easy to assign appropriate image findings and image diagnosis names to a large number of image data. With regard to the name of the image diagnosis, it is possible to obtain a reliable image diagnosis name based on other diagnosis results such as cytodiagnosis and genetic analysis, but the doctor can only judge the image findings one by one. That is, it is a very troublesome task for a doctor to appropriately give a plurality of image findings to a large number of image data. In addition, it is difficult to assign an evaluation value to each image finding based on a uniform standard at all times, and it is very important for enhancing the objectivity of image diagnosis.

  In order to unify the criteria for assigning image findings, there is a work method in which one type of image finding to be noticed is determined and only predetermined evaluation values of image findings are assigned to a large number of image data. . As a technique for supporting this work, there is known a similar image presentation technique for assisting a user in classifying image data whose attribute category is unclassified by presenting similar image data for each category of a certain attribute. ing. Here, the type of image finding to be noticed corresponds to the attribute, and the evaluation value corresponds to the category.

  The similar image presentation technique described above is also useful when actually performing image diagnosis. That is, by constructing an image database storing image data to which evaluation values of image findings are given based on a standard unified in advance, and presenting image data stored in the image database when image diagnosis is performed, Supporting a doctor's image diagnosis can be mentioned.

  Japanese Patent Application Laid-Open No. 2004-151867 discloses a technique for rearranging and displaying image data groups with uncategorized categories according to the degree of similarity with image data groups designated by the user as the same category. Patent Document 1 discloses a technique in which a computer assists a user's judgment by displaying category candidates when a user designates uncategorized image data. That is, in the technique described in Patent Document 1, the unclassified image data is rearranged according to the similarity with the category specified by the user, or the image data in which the category specified by the user is unclassified. Either a category candidate suitable for the display is displayed.

JP 2001-156135 A

  However, the evaluation values of many image findings are given based on a doctor's advanced knowledge and experience. Therefore, even if a category candidate is displayed based on the evaluation value automatically calculated by the computer, there is no guarantee that the candidate is appropriate, and thus it may not be useful for improving the accuracy of the user's classification work. There is sex. In addition, when only a plurality of similar image data is displayed for each category, the user needs to compare all the similar image data with the unclassified image data, which causes a problem that the user's work efficiency is lowered.

  Therefore, an object of the present invention is to improve accuracy and work efficiency when classifying image data categories.

Display processing device of the present invention, with respect to an image belonging to a predetermined category of Jo Tokoro attributes, whether the image is an images similar to other categories close most to the predetermined category in the predetermined attribute not acquisition means for acquiring boundary information indicating whether, by the boundary information when the picture image is shown to be images similar to the other categories of images that belong to the other category other images by remote close that belong to the predetermined categories for the display area, a display processing means for displaying the image image to the boundary picture image of the other categories to the predetermined category have a, the boundary image is characterized by an image including the object.
According to another aspect of the display processing apparatus of the present invention, regarding an image belonging to a predetermined category of a predetermined attribute, is the image similar to another category closest to the predetermined category in the predetermined attribute? Acquisition means for acquiring boundary information indicating whether or not, and when the boundary information indicates that the image is an image similar to the other category, a display area of the other image belonging to the predetermined category Display processing means for displaying the image as a boundary image between the predetermined category and the other category between the display area of the image belonging to the other category and the display area of the image belonging to the other category. It is the image which contains.

  According to the present invention, it is possible to improve accuracy and work efficiency when classifying image data categories.

1 is a diagram illustrating an overall configuration of an image classification support system including an image classification support device according to an embodiment of the present invention. It is a flowchart which shows the read-out process and display process of medical image data in the image classification assistance apparatus which concerns on embodiment of this invention. It is a figure which shows the example of a display screen of the display part in 1st Embodiment. It is a flowchart which shows the category classification | category process and display process of medical image data by the image classification assistance apparatus which concerns on the 1st Embodiment of this invention. It is a figure which shows the example of a display screen of the display part in 2nd Embodiment. It is a flowchart which shows the category classification | category process and display process of medical image data by the image classification assistance apparatus which concerns on 4th Embodiment. It is a figure which shows the example of a display screen of the display part in 4th Embodiment.

  DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments to which the present invention is applied will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a diagram showing an overall configuration of an image classification support system including an image classification support apparatus according to an embodiment of the present invention. As shown in FIG. 1, the image classification support system includes an image classification support device 10 and a database 22, and the image classification support device 10 and the database 22 are connected to each other via a LAN (Local Area Network) 21 so as to communicate with each other. ing.

  The database 22 manages medical image data and information associated therewith. The image classification support apparatus 10 acquires medical image data managed by the database 22 and information associated therewith via the LAN 21.

  The image classification support apparatus 10 includes a communication IF 31, a ROM 32, a RAM 33, a storage unit 34, an operation unit 35, a display unit 36, and a control unit 37 as functional configurations. The communication IF (Interface) 31 is realized by a LAN card or the like, for example, and manages communication between the external device (for example, the database 22) and the image classification support device 10 via the LAN 21. A ROM (Read Only Memory) 32 is realized by a nonvolatile memory or the like, and stores various programs. A RAM (Random Access Memory) 33 is realized by a volatile memory or the like, and temporarily stores various information. The storage unit 34 is realized by, for example, an HDD (Hard Disk Drive) or the like, and stores various types of information. The operation unit 35 is realized by, for example, a keyboard and a mouse, and inputs an instruction from the user into the image classification support apparatus 10. The display part 36 is implement | achieved by the display etc., for example, and displays various information toward a user (for example, doctor).

  The control unit 37 is realized by, for example, a CPU (Central Processing Unit) or the like, and performs overall control of processing in the image classification support device 10. The control unit 37 includes a data reading unit 41, a display processing unit 42, an image movement destination acquisition unit 43, and a category / boundary information setting unit 44 as its functional configuration. These configurations are realized, for example, when the CPU reads and executes a program stored in the ROM 32 or the storage unit 34 using the RAM 33 as a work area. A part or all of these configurations may be realized by a dedicated circuit or the like.

  The data reading unit 41 reads medical image data and information associated therewith from the database 22 via the communication IF 31 and the LAN 21. The information associated with the medical image data includes the attributes of the medical image data (type of image findings), the category of the medical image data (evaluation value of the image findings), and the medical image data having attributes adjacent to the category. Information (hereinafter referred to as boundary information) indicating whether the image data is the boundary image data for the category to be (closest) is included. Hereinafter, medical image data to which attributes, categories, and boundary information are assigned is referred to as classified image data, and medical image data to which these are not yet assigned is referred to as unclassified image data. The display processing unit 42 displays the medical image data read by the data reading unit 41 in a display area on the display unit 36 corresponding to the attribute, category, and boundary information of the medical image data. The data reading unit 41 is an example of an image acquisition unit and a classification acquisition unit, and the display processing unit 42 is an example of a display control unit. The attribute of medical image data is an example of attribute information that is a reference for classification of medical image data.

  The image movement destination acquisition unit 43 selectively acquires an instruction to change (move) the display position of any medical image data displayed on the display processing unit 42 among the instructions input by the user from the operation unit 35. To do. Then, the image movement destination acquisition unit 43 acquires a new display position candidate (hereinafter referred to as a movement destination) of the image data according to the instruction. This instruction input by the user can be realized by a mouse operation known as a drag and drop operation in a general OS using a graphical user interface (GUI), for example. The destination of the image data can be acquired as the position of the mouse pointer on the screen of the display unit 36.

  The category / boundary information setting unit 44 acquires category and boundary information corresponding to the coordinates of the movement destination of the medical image data acquired by the image movement destination acquisition unit 43, and assigns it to the medical image data instructed to move. To do. Thereby, the category and boundary information of the medical image data instructed to move are changed. Then, the display processing unit 42 displays the medical image data instructed to move at the destination according to the changed category and boundary information.

  Next, processing of the image classification support apparatus 10 illustrated in FIG. 1 will be described with reference to FIGS. FIG. 2 is a flowchart showing medical image data read processing and display processing in the image classification support apparatus 10. The processing shown in FIG. 2 is realized by the control unit 37 temporarily reading necessary data and programs from the ROM 32 and the storage unit 34 into the RAM 33 and executing them. The process shown in FIG. 2 is started when the user instructs the attribute (image finding type) An of the medical image data via the operation unit 35.

  In step S <b> 101, the control unit 37 acquires a target attribute An instructed by the user and stores the attribute An in the RAM 33. The display processing unit 42 displays the acquired attribute An in a predetermined display area on the display unit 36.

  In step S <b> 102, the data reading unit 41 acquires the number N of data stored in the database 22 via the LAN 21 and stores it in the RAM 33. Further, the control unit 37 assigns 1 to the data number n of data to be read next, and stores it in the RAM 33.

  In step S <b> 103, the data reading unit 41 reads the data Dn with the data number n from the database 22 via the LAN 21 and stores it in the RAM 33 or the storage unit 34. The data Dn is data including the medical image data In, the category Cn of the attribute An of the medical image data In, and the boundary information Bn of the medical image data In.

  In step S104, the display processing unit 42 displays the medical image data In in the display area of the display unit 36 corresponding to the category Cn and the boundary information Bn. In step S <b> 105, the control unit 37 reads the data number n from the RAM 33, adds 1 to the data number n, and stores it again in the RAM 33. In step S106, the control unit 37 reads the data number n and the data number N from the RAM 33, and compares the data number n with the data number N. When the data number n is smaller than the data number N, medical image data that has not been read from the database 22 remains, so the process returns to step S103, and the next medical image data is read from the database 22. When the data number n reaches the number of data N, all the medical image data has been read out, and the processing shown in FIG. Through the above processing, for example, a screen as shown in FIG.

FIG. 3 is a diagram illustrating an example of a display screen of the display unit 36 in the first embodiment. FIG. 3 shows the display area 301 of the attribute An of interest, the display area 302 of the unclassified image data Ix = {I1, I2,... A display area R _{(Cn, B)} 303 for image data and boundary image data is shown. In the example of FIG. 3, the category Cn = {C1, C2,...} And the boundary information B = {− 1, 0, +1}. Here, B = 0 indicates non-boundary image data. B = −1 indicates that the attribute An of interest in the medical image data in the category Cn is boundary image data most similar to the adjacent category Cn−1. B = + 1 indicates that the attribute An of interest in the medical image data in the category Cn is boundary image data most similar to the adjacent category Cn + 1. Further, as shown in FIG. 3, the display area of the boundary image data is arranged at a position closer to the adjacent category than the display area of the non-boundary image data. Note that the display screen shown in FIG. 3 is a configuration serving as an example of attribute designation means.

In the example of FIG. 3, since the category C0 does not exist, there is no boundary image data of B = −1 in the category C1. Moreover, although only the categories C1 and C2 are illustrated in FIG. 3, there may be three or more categories. Further, in FIG. 3, one boundary image data is illustrated for each of the boundary image data display areas R _{(C1, + 1)} , R _{(C2, −1)} , and R _{(C2, + 1)} . A plurality of boundary image data of a predetermined number or less may be displayed in the display area.

  FIG. 4 is a flowchart showing medical image data category classification processing and display processing by the image classification support apparatus 10. The process shown in FIG. 4 is a process realized by the control unit 37 temporarily reading necessary data and programs from the ROM 32 and the storage unit 34 into the RAM 33 and executing them. The process shown in FIG. 3 is a process that is automatically started after the process shown in FIG.

  In step S201, the control unit 37 acquires the number of categories (number of evaluation values of image findings) M for the attribute (type of image findings) An instructed from step S101 of FIG. When the user instructs the movement destination of the medical image data Ix via the operation unit 35, the image movement destination acquisition unit 43 acquires the coordinate Px of the movement destination of the medical image data Ix and stores it in the RAM 33 in step S202. When the coordinate Px of the movement destination of the medical image data Ix is instructed by a drag and drop operation in a general OS using the above-described GUI, Px is acquired as the coordinate of the mouse pointer on the screen displayed on the display unit 36. be able to. In step S <b> 203, the image movement destination acquisition unit 43 assigns 1 to the category number k and 0 to the boundary information B, and stores each in the RAM 33.

In step S <b> 204, the image movement destination acquisition unit 43 acquires from the display unit 36 the coordinates of the display area R _{(Ck, B)} corresponding to the category Ck specified by the current category number k and the current boundary information B. In step S205, the image movement destination acquisition unit 43 reads the coordinates Px of the movement destination of the medical image data Ix from the RAM 33 and compares it with the coordinates of the display area R _{(Ck, B)} . If the coordinate Px is within the display area R _{(Ck, B)} , the process proceeds to step S210. On the other hand, if the coordinate Px is not within the display region R _{(Ck, B)} , the process proceeds to step S206.

  In step S206, the image movement destination acquisition unit 43 reads the boundary information B from the RAM 33, adds 1 to the boundary information B, and stores the boundary information B in the RAM 33 again. In step S207, the image movement destination acquisition unit 43 determines whether the boundary information B exceeds +1. If B exceeds +1, the process proceeds to step S208. On the other hand, if the boundary information B does not exceed +1, the process returns to step S204.

  In step S208, the image movement destination acquisition unit 43 reads the category number k from the RAM 33, adds 1 to the category number k, and stores it again in the RAM 33. Further, the image movement destination acquisition unit 43 assigns −1 to the boundary information B and stores it in the RAM 33. In step S209, the image destination acquisition unit 43 determines whether the category number k exceeds the number of categories M. If the category number k exceeds the number of categories M, the process ends because the coordinates Px do not belong to any category display area. At this time, the category and boundary information of the medical image data Ix are not changed. On the other hand, when the category number k does not exceed the category number M, the process proceeds to step S204.

  In step S <b> 210, the category / boundary information setting unit 44 reads the boundary information B from the RAM 33 and determines whether the boundary information B is 0. Here, when the boundary information B is not 0, that is, when the boundary information B is −1 or +1, the process proceeds to step S211. On the other hand, when the boundary information B is 0, the process proceeds to step S212.

In step S211, the category / boundary information setting unit 44 sets the boundary information B of the image data I _{(Ck, B) in} the boundary image data display area R _{(Ck, B)} (B = −1 or +1) to 0. change. Then, the display processing unit 42 moves the display position of the image data I _{(Ck, B)} into the display area R _{(Ck, 0)} of the non-boundary image data. In step S211, the category / boundary information setting unit 44 substitutes Ck for the category of the medical image data Ix, and substitutes B for the boundary information. Then, the display processing unit 42 moves the medical image data Ix into the display area R _{(Ck, B)} .

  If the boundary information B is 0 by the processing after step S210, the medical image data Ix is added as non-boundary image data of category Ck. On the other hand, when the boundary information B is not 0, the medical image data Ix is displayed in the display area of the boundary image data of the category Ck (B = −1 or +1) (that is, becomes new boundary image data) and originally exists. The boundary image data is moved to the display area of the non-boundary image data. Through the above processing, the category and boundary information of the medical image data Ix are set. That is, when the medical image data Ix is unclassified image data, it becomes classified image data, and when the medical image data Ix is already classified image data, the classification is changed.

  In the present embodiment, the user can easily determine an appropriate classification destination (category) of the medical image data Ix only by comparing the medical image data Ix with the boundary image data in each category. In addition, the user can quickly perform category classification simply by designating the movement destination of the medical image data Ix. Thereby, both the accuracy of the category classification by the user and the work efficiency can be improved. Furthermore, when performing image diagnosis, it is possible to accurately and efficiently assign an evaluation value of image findings.

  The above is a representative embodiment of the present invention, but the present invention is not limited to the above-described embodiment, and can be appropriately modified and implemented without departing from the scope of the present invention.

  Next, a second embodiment of the present invention will be described. In the second embodiment, the description will be given focusing on differences from the first embodiment.

In the second embodiment, the boundary information B is limited to two of 0 and +1 (or 0 and −1). Further, the display area R _{(Ck, + 1)} (or R _{(Ck, -1)} ) of the boundary image data of the category Ck is displayed outside the display area of the category Ck, for example, between two adjacent categories. Also good.

FIG. 5 is a diagram illustrating a display screen example of the display unit 36 according to the second embodiment. In the second embodiment, the boundary information B is limited to 0 and +1. In the second embodiment, as shown in FIG. 5, the display area R _{(Ck, + 1)} of the boundary image data categories Ck and display area R _(Ck non boundary image data of the two categories of adjacent _{, 0)} and R _{(Ck + 1,0)} . As a result, the number of boundary image data that the user should compare with the medical image data Ix is almost halved, so that the user's work efficiency can be further improved.

  Next, a third embodiment of the present invention will be described. Note that in the third embodiment, description will be given focusing on differences from the first and second embodiments.

In the third embodiment, the boundary image data is not normally displayed, and the boundary image data is displayed in a predetermined display area of the display unit 36 when the movement destination coordinate Px of the medical image data Ix is acquired in step S202. Control the display. Furthermore, when the coordinate Px of the movement destination of the medical image data Ix approaches the display area R _{(Ck + 1,0)} of the non-boundary image data of the category Ck, that is, the coordinate Px and the display area R _{(Ck + 1,0) )} May be displayed when the distance to the representative point (end point or center point) of) is equal to or less than a certain value.

  As a result, the number of boundary image data that the user should compare with the medical image data Ix is further reduced, so that the user's work efficiency can be further improved.

  Next, a fourth embodiment of the present invention will be described. In the first embodiment, a configuration in which the user can change arbitrary boundary image data has been described. In the fourth embodiment, a configuration in which the user cannot change boundary image data is described. Note that the boundary image data is set in advance. In the fourth embodiment, the process shown in FIG. 6 is executed instead of the process shown in FIG. 4, and the other processes are the same.

  FIG. 6 is a flowchart showing medical image data category classification processing and display processing by the image classification support apparatus 10 according to the fourth embodiment. The process illustrated in FIG. 6 is a process that is automatically started after the process illustrated in FIG. In addition, since the structure of the image classification assistance apparatus 10 which concerns on 4th Embodiment is the structure similar to the image classification assistance apparatus 10 shown in FIG. 1, the code | symbol of FIG. 1 shall be diverted also in the following description. To do. The processing shown in FIG. 6 is realized by the control unit 37 temporarily reading necessary data and programs from the ROM 32 and the storage unit 34 into the RAM 33 and executing them.

  In step S301, the control unit 37 acquires the number of categories (number of evaluation values of image findings) M for the attribute (type of image findings) An instructed in step S101 of FIG. When the user instructs the movement destination of the medical image data Ix via the operation unit 35, the image movement destination acquisition unit 43 acquires the movement destination coordinate Px of the medical image data Ix and stores it in the RAM 33 in step S302. In step S <b> 303, the image movement destination acquisition unit 43 assigns 1 to the category number k and stores it in the RAM 33.

In step S304, the image movement destination acquisition unit 43 acquires the coordinates of the display area R _(Ck) corresponding to the category Ck designated by the current category number k from the display unit 36. In step S305, the image movement destination acquisition unit 43 reads the movement destination coordinates Px of the medical image data Ix from the RAM 33 and compares them with the coordinates of the display area R _(Ck) . If the coordinate Px is within the display area R _(Ck) , the process proceeds to step S308. On the other hand, if the coordinate Px is not within the display region R _(Ck) , the process proceeds to step S306.

In step S <b> 306, the image movement destination acquisition unit 43 reads the category number k from the RAM 33, adds 1 to k, and stores it again in the RAM 33. In step S307, the image destination acquisition unit 43 determines whether the category number k exceeds the number of categories M. If the category number k exceeds the number of categories M, the process ends because the coordinates Px do not belong to any category display area. At this time, the category and boundary information of the medical image data Ix are not changed. On the other hand, if the category number k does not exceed the category number M, the process returns to step S304. In step S308, the image movement destination acquisition unit 43 substitutes Ck for the category of the medical image data Ix and substitutes 0 for the boundary information. Then, the display processing unit 42 moves the medical image data Ix into the display area R _(Ck) .

  Through the above processing, the category and boundary information of the medical image data Ix are set. That is, the classification is changed when the medical image data Ix is unclassified image data, and the classification is changed when the medical image data Ix is classified image data.

FIG. 7 is a diagram illustrating a display screen example of the display unit 36 according to the fourth embodiment. In FIG. 7, reference numerals 701, 702, and 703 denote display areas corresponding to 301, 302, and 303 in FIG. In the fourth embodiment, the movement destination of the medical image data Ix is limited to the display area R _(Ck) 704 of the non-boundary image data of each category. There is no risk of changing with care. Furthermore, since it is not necessary to change the boundary image data 705 when performing image diagnosis, it is possible to assign an evaluation value of image findings accurately and more efficiently.

  The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed.

  10: Image classification support device, 21: LAN, 22: Database, 31: Communication IF, 32: ROM, 33: RAM, 34: Storage unit, 35: Operation unit, 36: Display unit, 37: Control unit, 41: Data reading unit, 42: display processing unit, 43: image destination acquisition unit, 44: category / boundary information setting unit

Claims (16)

Acquisition means for acquiring boundary information indicating whether the image belongs to a predetermined category of a predetermined attribute and whether the image is an image similar to another category closest to the predetermined category in the predetermined attribute; ,
When the boundary information indicates that the image is similar to the other category, the display area of the image belonging to the other category is closer than the other image belonging to the predetermined category Display processing means for displaying the image as a boundary image between the predetermined category and the other category at a position;
The display processing apparatus, wherein the boundary image is an image including a subject. Acquisition means for acquiring boundary information indicating whether the image belongs to a predetermined category of a predetermined attribute and whether the image is an image similar to another category closest to the predetermined category in the predetermined attribute; ,
If the boundary information indicates that the image is an image similar to the other category, a display area of another image belonging to the predetermined category and a display area of an image belonging to the other category A display processing unit for displaying the image as a boundary image between the predetermined category and the other category,
The display processing apparatus, wherein the boundary image is an image including a subject.   The display processing device according to claim 1, wherein the display processing unit switches the boundary image from non-display to display in accordance with an operation by a user.   The display processing according to claim 1, wherein the display processing unit displays the boundary image in response to a user operation on an image that is not classified into any category in the predetermined attribute. apparatus.   The display processing means belongs to the distance between the display position of the image specified by the user and the display area of another image belonging to the predetermined category or the display position of the image specified by the user and the other category. The display processing apparatus according to claim 1, wherein a boundary image of the predetermined category is displayed in accordance with a distance from the image display area.   The display processing apparatus according to claim 1, wherein the display processing unit displays an image that is not classified into any category in the predetermined attribute together with the boundary image.   The display processing apparatus according to claim 1, wherein the image is a medical image. The attribute indicates an image finding;
The display processing apparatus according to claim 7, wherein the category indicates an evaluation value of the image findings. A display processing method executed by a display processing device,
An acquisition step of acquiring boundary information indicating whether the image belongs to a predetermined category of a predetermined attribute and whether the image is an image similar to another category closest to the predetermined category in the predetermined attribute; ,
When the boundary information indicates that the image is similar to the other category, the display area of the image belonging to the other category is closer than the other image belonging to the predetermined category A display processing step for displaying the image as a boundary image between the predetermined category and the other category at a position;
The display processing method, wherein the boundary image is an image including a subject. A display processing method executed by a display processing device,
An acquisition step of acquiring boundary information indicating whether the image belongs to a predetermined category of a predetermined attribute and whether the image is an image similar to another category closest to the predetermined category in the predetermined attribute; ,
If the boundary information indicates that the image is an image similar to the other category, a display area of another image belonging to the predetermined category and a display area of an image belonging to the other category And a display processing step for displaying the image as a boundary image between the predetermined category and the other category,
The display processing method, wherein the boundary image is an image including a subject. An acquisition step of acquiring boundary information indicating whether the image belongs to a predetermined category of a predetermined attribute and whether the image is an image similar to another category closest to the predetermined category in the predetermined attribute; ,
When the boundary information indicates that the image is similar to the other category, the display area of the image belonging to the other category is closer than the other image belonging to the predetermined category Causing the computer to execute a display processing step of displaying the image as a boundary image between the predetermined category and the other category at a position;
The boundary image is an image including a subject. An acquisition step of acquiring boundary information indicating whether the image belongs to a predetermined category of a predetermined attribute and whether the image is an image similar to another category closest to the predetermined category in the predetermined attribute; ,
If the boundary information indicates that the image is an image similar to the other category, a display area of another image belonging to the predetermined category and a display area of an image belonging to the other category And causing the computer to execute a display processing step of displaying the image as a boundary image between the predetermined category and the other category,
The boundary image is an image including a subject. An image classification device for classifying medical images with respect to predetermined attribute information according to an operation input,
Image acquisition means for acquiring medical images to be classified;
Attribute designating means for designating attribute information as a basis for the classification;
Classification acquisition means for acquiring a plurality of categories corresponding to the specified attribute information;
A medical image to be classified is displayed, a plurality of first areas indicating a plurality of categories corresponding to the attribute information are displayed side by side on a screen of a display unit in a predetermined order, and two adjacent categories are displayed Display control means for displaying a medical image designated as an image showing a boundary between the two categories in a second region between
Classification means for classifying the medical image to be moved into a category corresponding to the first area of the movement destination in response to an operation input for moving the medical image to be classified to one of the plurality of first areas; An image classification apparatus comprising: The attribute information indicates an image finding,
The image classification device according to claim 13, wherein the category indicates an evaluation value of the image findings.   The image classification apparatus according to claim 13 or 14, wherein the predetermined order is an order based on the attribute information.   The image classification apparatus according to claim 13, wherein the display control unit displays medical images belonging to a category corresponding to the first area in the first area. .

Images