JP2008108079A - Document editing support system, document editing support method, and document editing support program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a document editing support system, a document editing support method, and a document editing support program which are capable of supporting document editing suitably for document generation of natural language sentences. <P>SOLUTION: The document editing support system includes: a medical image analysis part 14 which analyzes a medical image obtained by photographing an object and acquires areas of respective parts of the object in the medical image and part names showing the parts; a display control part 18 which displays candidates of a part name showing a part corresponding to a designated area, on a monitor on the basis of areas of parts and part names, which are held in an analysis result holding part 16, when this area of the medical image displayed on the monitor is designated by operation of an operation part; a clause generation part 24 which generates a clause including a designated part name when one of candidates of the part name, which are displayed on the monitor, is designated by operation of the operation part; and a compilation part 30 which adds the generated clause to a document during edition of the document. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a document editing support system, a document editing support method, and a document editing support program that support editing of a document, in particular, editing of a document such as an electronic medical record and an interpretation report by a doctor.

  Electronic medical records, CT (Computed Tomography) images, MRI (Magnetic Resonance Imaging) images, interpretation reports that are findings and diagnostic reports on X-ray photographs, etc. include the site and state of the patient's disease in the description. However, these parts and states may be described incorrectly, and such errors can be one of the causes of medical accidents.

  CT and MRI display a group of images obtained by slicing a human body on a specific surface at equal intervals. However, since the human body has a three-dimensional structure, a high level of skill is required to associate the position on the image with the anatomical part name, and the doctor may grasp the diseased part by mistake. In addition, CT images, MRI images, and X-rays may not be displayed at full size, and doctors can intuitively grasp the size of the diseased site even when the scale is displayed. In some cases, for example, a place that should originally be described as 1 mm is described as 1 cm. Furthermore, since the human body is substantially symmetric, even if a CT image or the like is displayed opposite to the left and right, the doctor may not recognize the right and left and may grasp it. In addition, although there may be different types of diseased sites at two or more locations at the same time, a doctor may mistakenly describe each diseased site.

  There are various types of MRI images (T1-weighted images, T2-weighted images, diffusion-weighted images, etc.), and other images are also various types depending on the contrast and imaging conditions. However, each image is similar because the objects are the same, and in many cases, doctors refer to them simultaneously at the time of interpretation, and thus sometimes grasp the wrong type of image.

To solve the above-described problem, the technique described in Patent Document 1 supports interpretation by displaying an image to be interpreted and an image of a normal example at the same time. Moreover, the technique described in patent document 2 and patent document 3 provides the name of the site | part corresponding to the position, when the predetermined position of a medical image is designated.
JP 2001-275986 A JP 2003-260030 A JP 2003-33327 A

  However, it cannot be said that the conventional method described above sufficiently supports document editing by a doctor. Specifically, a doctor creates a finding in a natural language sentence. At this time, only by providing a part name or the like from an image to be interpreted, it is complicated to input a particle.

  An object of the present invention is to solve the above-described problems and provide a document editing support system, a document editing support method, and a document editing support program capable of supporting document editing suitable for creating a document of a natural language sentence. is there.

  The document editing support system of the present invention is displayed on the display screen, first holding means that holds the region of each part of the object in the image of the object and the part name representing the part in association with each other. When an area is designated in the image of the object, the designation is made from the area of each part of the object in the image of the object held in the first holding means and the part name representing the part. First acquisition means for acquiring a part name as a candidate corresponding to the region obtained, and first display control means for displaying the part name candidate acquired by the first acquisition means on the display screen; Generating means for generating a phrase including a part name designated from candidate part names displayed on the display screen by the first display control means; and adding the phrase generated by the generating means to a document to be edited Follow And having a means.

  According to this configuration, when an intuitive action is performed in which the user specifies an area of the object in the displayed image, not only the part name corresponding to the specified area of the object but also the part name A phrase containing is added to the document being edited.

  From the same point of view, the document editing support system of the present invention may be configured such that the generation unit generates a phrase by inserting the specified part name into a phrase pattern representing a predetermined phrase format. Good.

  In addition, the document editing support system of the present invention has second acquisition means for acquiring meta information of the image, in which meta information that is information representing the attribute of the image is associated with the image of the object. The generating unit may generate a phrase including a phrase corresponding to the meta information acquired by the second acquiring unit.

  According to this configuration, a phrase including a word corresponding to the attribute of the image is added to the document being edited, and it becomes easier to create a natural language sentence document.

  In the document editing support system according to the present invention, the second acquisition unit acquires meta information corresponding to each of a plurality of images including the same part with different imaging methods, and the generation unit includes the second acquisition unit. For each piece of meta information acquired by the acquisition means, a phrase including a phrase corresponding to the meta information may be generated.

  According to this configuration, a phrase including a phrase corresponding to each attribute of a plurality of images is added to the document being edited, and it becomes easier to create a natural language sentence document for each image.

  Further, the document editing support system of the present invention is acquired by a third acquisition unit that acquires a phrase to be modified by a phrase including the part name from a predetermined group of words, and the third acquisition unit. You may make it have an addition means to add a phrase to the clause produced | generated by the said production | generation means.

  According to this configuration, a phrase modified by a phrase including a part name is also added to the phrase, making it easier to create a natural language sentence document.

  Further, the document editing support system of the present invention has a table indicating a correspondence relationship between each of the plurality of operations of the operation unit and the phrase, and the third acquisition unit refers to the table to operate the operation unit. May be acquired as a phrase to be modified by a clause including the part name.

  According to this configuration, the phrase modified by the phrase including the part name is quickly added to the document being edited by the user's intuitive operation.

  The document editing support system of the present invention includes a document to which a phrase including the part name is added, an image including a part represented by the part name included in the phrase, and a part included in the phrase in the image. A second holding unit that holds document image related information that is information relating a region of a part represented by a name, and a document and an image included in the document image related information held by the second holding unit You may make it have a 2nd display control means to identify and display the sentence containing the site | part name corresponding to the area | region designated in the displayed image while displaying on a screen.

  According to this configuration, an associated document and an image are displayed, and further, a sentence including a part name corresponding to a region designated in the image is displayed. It becomes possible to easily obtain a description of the part name corresponding to the designated area of the image together with the image.

  In the document editing support system of the present invention, when a sentence included in the displayed document image related information or an area in the image is designated, the part name contained in the designated sentence or the designation A first search means for searching for document image related information including a document to which a phrase including a part name representing a part corresponding to the specified region is added, and a document image related obtained by the search by the first search means You may make it have a 3rd display control means to display at least any one of the document and image contained in information on the said display screen.

  According to this configuration, the user acquires a plurality of document image related information that are related to each other, and in reading and understanding a document included in the document image related information, recognizing an image, and creating a new document, Documents and images included in information can be used.

  From the same viewpoint, the document editing support system of the present invention, when a search key is input, a second search means for searching for document image related information including the search key, and a search by the second search means. And a fourth display control means for displaying on the display screen at least one of a document and an image included in the document image related information obtained by the above.

  The document editing support system according to the present invention further includes a degree-of-association calculation unit that calculates a degree of association between the part name and the image capturing method, and the generation unit has the degree of association equal to or greater than a predetermined value A phrase including the part name and the imaging method may be generated.

  According to this configuration, an appropriate phrase can be generated in consideration of the degree of association between the part name and the image capturing method.

  According to the document editing support method of the present invention, when an area is specified in the object image displayed on the display screen, the region of each part of the object in the object image held by the holding unit, An acquisition step of acquiring a part name as a candidate corresponding to the specified region from a part name representing the part; and a display control step of displaying the candidate of the part name acquired by the acquisition step on the display screen; A generation step of generating a phrase including a part name designated from the part name candidates displayed on the display screen by the display control step, and an addition of adding the phrase generated by the generation step to a document to be edited And a step.

  The document editing support program according to the present invention, when an area is specified in an image of an object displayed on a display screen, the area of each part of the object in the image of the object held by a holding unit, An acquisition step of acquiring a part name as a candidate corresponding to the specified region from a part name representing the part; and a display control step of displaying the candidate of the part name acquired by the acquisition step on the display screen; A generation step of generating a phrase including a part name designated from the part name candidates displayed on the display screen by the display control step, and an addition of adding the phrase generated by the generation step to a document to be edited And causing the computer to execute the steps.

  According to the present invention, not only the part name corresponding to the designated area of the object but also the part name corresponding to the intuitive act of designating the area of the object in the displayed image by the user. Is added to the document being edited, making it easy to create a natural language sentence.

  Embodiments of the present invention will be specifically described with reference to the drawings. FIG. 1 is a diagram showing a hardware configuration of a personal computer (PC) as a document editing support system. The PC 100 as the document editing support system shown in FIG. 1 operates as an operation unit including a CPU 101, a memory 102, a hard disk drive (HDD) 103, an input device such as a keyboard, and an instruction device such as a mouse connected to an internal bus 107. The unit 105 includes a monitor 106 that displays on the display screen.

  FIG. 2 is a functional block diagram of a document editing support system constituted by the PC 100. The document editing support system shown in FIG. 2 supports editing of a document that becomes a finding or interpretation report created by a doctor, and corresponds to the medical image holding unit 12, the medical image analysis unit 14, and the first holding unit. Analysis result holding unit 16, display control unit 18 corresponding to the first to fourth display control units and providing unit, site specifying unit 22 corresponding to the first acquisition unit, generation unit, second acquisition unit and related The phrase generation unit 24, the phrase pattern storage unit 26, the image meta information storage unit 28, the addition unit, the third acquisition unit and the editing unit 30 corresponding to the addition unit, the coordinate acquisition unit 32, the second The document image related information holding unit 34 corresponding to the holding unit and the search unit 36 corresponding to the first and second search units are configured. In the hardware configuration of FIG. 1, each of these functional blocks corresponds to a predetermined program read from the HDD 103 and stored in the memory 102 by the CPU 101 according to the operation of the operation unit 105, for example, a procedure of a flowchart described later. This is realized by executing the program.

  The medical image holding unit 12 holds a medical image such as an MRI image, which is a target for creation of findings and interpretation reports. A medical image is uniquely identified by an examination ID and an image ID that are information included in the medical image. The medical image analysis unit 14 reads out and analyzes the medical image held in the medical image holding unit 12, thereby acquiring the region of the part and the part name representing the part for each part of the target object. For analysis of a medical image, for example, a system described in Japanese Patent Application Laid-Open No. 2003-33327 that detects a lesion of an object and outputs the position and state thereof can be used. The analysis result holding unit 16 holds the medical image and the region and part name of each part of the target corresponding to the medical image in association with each other and holds the result as an analysis result. FIG. 3 is a diagram illustrating an example of a part name and a part region included in the analysis result. The part name and the part region are added with an examination ID and an image ID, and are associated with a medical image including the examination ID and the image ID. The part region is specified by, for example, coordinates in a two-dimensional space corresponding to a medical image, and is represented by a closed region connecting the coordinates.

  When the analysis result holding unit 16 holds the medical image and the region and part name of each part of the target object in association with each other, it is possible to support editing of a document that becomes a finding or interpretation report created by the doctor. . Hereinafter, an operation for supporting document editing will be described with reference to a flowchart. FIG. 4 is a flowchart showing the operation of the document editing support system.

  The display control unit 18 performs control for reading the medical image held in the medical image holding unit 12 and displaying the medical image on the monitor 106 in accordance with the operation of the operation unit 105 (for example, a keyboard) by a doctor who is a user. Perform (S101). FIG. 5A is a diagram illustrating an MRI image of the head, which is an example of a medical image. The display control unit 18 further acquires the examination ID and image ID included in the displayed medical image, and among the analysis results held in the analysis result holding unit 16, the medical image including these examination ID and image ID, in other words, Then, an image having a displayed medical image is acquired and output to the site specifying unit 22.

  Next, the part specifying unit 22 determines whether the region of the object corresponding to the medical image displayed on the display screen of the monitor 106 is designated by the operation of the operation unit 105 (for example, a pointing device such as a mouse) by the doctor. Is determined (S102). For example, in the medical image shown in FIG. 5B, the region of the “X” mark 201 in the head that is the object is designated.

  When an operation for designating an area is performed, the part specifying unit 22 performs a process of acquiring a part name candidate corresponding to the designated area (S103). Specifically, the coordinate acquisition unit 32 acquires the coordinates of the designated area in the medical image being displayed in response to the operation of the operation unit 105 by the doctor, and sends the coordinates to the part specifying unit 22. The part specifying unit 22 compares the coordinates of the designated area in the medical image being displayed with the coordinates of each part area included in the analysis result from the display control unit 18. Then, the part specifying unit 22 specifies a part of the part region included in the analysis result from the display control unit 18 that overlaps the designated region in the medical image being displayed. Further, the part specifying unit 22 sends the part name associated with the specified part region to the display control unit 18 as a candidate for the part name corresponding to the specified region. The display control unit 18 performs control to display these part name candidates on the monitor 106 together with the medical image. For example, FIG. 5C shows a part name candidate list 202 together with a medical image.

  Next, the part specific | specification part 22 determines whether the part name candidate was designated (S104). Specifically, the part specifying unit 22 is suitable for the doctor among the part name candidates displayed in a list format on the monitor 106 by the operation of the operation unit 105 (for example, a pointing device such as a mouse) by the doctor. Is determined, it is determined that the specified part name is specified. When the part name is specified, the part specifying unit 22 obtains the acquired part name, in other words, the part name specified by the doctor, the region of the part associated with the part name in the analysis result, and the medical image. The region designated by the doctor is sent to the phrase generation unit 24.

  The image meta information holding unit 28 holds image meta information including information representing attributes of a medical image such as an imaging method and a scale (%). FIG. 5 is a diagram illustrating an example of image meta information. The image meta information is uniquely specified by the examination ID and the image ID as in the case of the medical image. When the phrase generation unit 24 acquires the medical image, the region of the region, and the region name from the region specifying unit 22, image meta information having the same examination ID and image ID as the medical image is acquired from the image meta information holding unit 28. Obtain (S105).

  The phrase pattern holding unit 26 holds phrase patterns. FIG. 7 is a diagram illustrating an example of a phrase pattern. In these phrase patterns, [] is a portion into which a part name or image metadata is inserted. The phrase generation unit 24 acquires the phrase pattern held in the phrase pattern holding unit 26. Further, the phrase generation unit 24 inserts the phrase corresponding to the part name from the part specifying unit 22 and the image meta information acquired from the image meta information holding unit 28 into [] of the acquired phrase pattern, thereby generating a phrase. Generate (S106).

  On the other hand, if it is determined in S104 that the part name has not been designated, specifically, among the part names displayed in the list format on the monitor 106, what the doctor considers appropriate is not designated. Alternatively, in S103, when the part specifying unit 22 cannot acquire the name of the part corresponding to the designated area, the phrase generating unit 24 operates the operation part 105 (for example, a keyboard) by the doctor. In response to this, a phrase is generated (S107).

  After the phrase generation in S106 and S107, the phrase generation unit 24 includes the generated phrase, the part name included in the generated phrase, the part region and the medical image associated with the part name, and the generated phrase. The image meta information and the region designated by the doctor are output to the editing unit 30. When the editing unit 30 inputs the phrase, part name, part region, medical image, image meta information, and region designated by the doctor, the editing unit 30 adds the phrase to the finding or interpretation report being edited (S108).

  For example, the part name “right thalamus” is designated by the doctor, the information regarding the imaging method included in the image meta information is “T1-weighted image”, and the phrase pattern is “[imaging method] and [part name]”. In this case, the phrase generation unit 24 inserts the part name “right thalamus” into the [part name] of the phrase pattern, and uses the phrase “T1-weighted image” corresponding to the image meta information as the [imaging method] of the phrase pattern. Then, the phrase “T1 weighted image in the right thalamus” is generated, and the editing unit 30 adds the phrase to the finding being edited or the interpretation report, thereby adding “T1 weighted image to the right thalamus. Create a sentence such as "You can see the signal area." Note that the editing unit 30 holds a plurality of words / phrases that are determined to have a high appearance frequency in advance for nouns modified by phrases (here, “high signal area”) and monitors them in a pull-down format or the like. The words designated by the doctor may be inserted.

  As described above, the document editing support system according to the present embodiment has an intuitive operation in which a doctor designates a region of an object in a medical image displayed on the monitor 106 together with the medical image by operating the operation unit 105. If it is done, depending on the operation, not only the part name corresponding to the specified area of the target object, but also the phrase containing the part name is added to the finding being edited and the interpretation report, It is easy to create certain findings and interpretation reports.

  By the way, in the creation of findings and interpretation reports, doctors often refer to medical images obtained by imaging the same part using a plurality of imaging methods at the same time. In such a case, the document editing support system can generate a clause corresponding to each medical image. Specifically, the display control unit 18 performs control to display any one of the medical images obtained by imaging the same part on the monitor 106, and among the analysis results held in the analysis result holding unit 16, An analysis result having the displayed medical image is acquired and output to the region specifying unit 22. Furthermore, the display control unit 18 acquires an analysis result having the same part name as the part name in the analysis result having the displayed medical image from the analysis results held in the analysis result holding unit 16. Output to the identifying unit 22.

  Next, when the region of the target object corresponding to the medical image displayed on the monitor 106 is specified by the operation of the operation unit 105 (for example, a pointing device such as a mouse) by the doctor, The process which acquires the site | part name corresponding to the made area | region is performed. Specifically, the part specifying unit 22 acquires the coordinates of the region designated in the displayed medical image in accordance with the operation of the operation unit 105 by the doctor, and each of the analysis results including the displayed medical image Compare with the coordinates of the region. Then, the part specifying unit 22 specifies a part of the part region from the display control unit 18 that overlaps the region specified in the medical image being displayed. Further, the part specifying unit 22 acquires a part name associated with the specified part region and sends the part name to the display control unit 18.

  The display control unit 18 performs control to display these part names on the monitor 106 together with the medical image as part name candidates. After that, by the operation of the operation unit 105 (for example, a pointing device such as a mouse) as an operation unit by the doctor, among the part names displayed in the list format on the monitor 106, what the doctor considers appropriate is designated. In this case, the part specifying unit 22 specifies the part name, in other words, the part name designated by the doctor, the medical image in the analysis result having the part name, the part region corresponding to the part name, and the doctor designated The area is sent to the phrase generation unit 24. Thereby, the phrase generation unit 24, a part name designated by the doctor, a plurality of medical images obtained by imaging the same part, a part region corresponding to the part name designated by the doctor in each of the medical images, The area designated by the doctor can be acquired.

  When the phrase generation unit 24 acquires the medical image, the region of the region, the region name, and the region designated by the doctor from the region specifying unit 22, the image having the same examination ID and image ID as the medical image is obtained for each medical image. The meta information is acquired from the image meta information holding unit 28 and the phrase pattern held in the phrase pattern holding unit 26 is acquired. Then, the phrase generation unit 24 inserts, for each medical image, a part name from the part specifying unit 22 or a phrase corresponding to the image meta information acquired from the image meta information holding unit 28 in [] of the acquired phrase pattern. To generate a phrase. Furthermore, the editing unit 30 adds the phrase generated for each medical image to the finding or interpretation report being edited.

  For example, there are two medical images that are obtained by photographing the same part and whose photographing methods are “T1-weighted image” and “T2-weighted image”, the part name “right thalamus” is designated by the doctor, and the image meta-information The information regarding the imaging method included is “T1-weighted image” or “T2-weighted image”, the information regarding the scale is “75”, and the phrase pattern is “[imaging method] and [part name] is [size]” Consider the case. In this case, the phrase generation unit 24 inserts the part name “right thalamus” into the [part name] of the phrase pattern, and also uses the phrase “T1-weighted image” or “T2” corresponding to the information about the imaging method included in the image meta information. Insert “enhanced image” into [Shooting method] of the phrase pattern. Further, the phrase generation unit 24 calculates the size of the region designated by the doctor based on the region designated by the doctor and the information related to the scale included in the image meta information, and a phrase (here, “ Insert a range of about 2 mm in diameter ") into the [size] of the phrase pattern. As a result, the phrase “T1-weighted image with a diameter of about 2 mm in the right thalamus” corresponding to the medical image whose photographing method is “T1-weighted image” and the medical image whose photographing method is “T2-weighted image” The phrase “T2-weighted image with a diameter of about 2 mm in the right thalamus” is generated. Then, the editing unit 30 adds these phrases to the finding or interpretation report being edited, thereby displaying a high signal area having a diameter of about 2 mm in the right thalamus in the T1 weighted image shown in FIG. A sentence for each medical image such as “A high-signal region having a diameter of about 2 mm is seen in the right thalamus in the T2-weighted image” shown in FIG. 8B is created. This facilitates the creation of findings and interpretation reports that are natural language sentences for each medical image obtained by imaging the same part.

  Further, as shown in FIG. 9, the editing unit 30 operates the noun (corresponding phrase) modified by the phrase and the operation of the operation unit 105 (for example, a pointing device such as a mouse) by a doctor, specifically, right click, When a table (correspondence table) in which operations such as left click and movement are associated with each other is held and the operation unit 105 is operated by a doctor, findings and interpretation of the noun corresponding to the operation being edited You may make it insert the noun which becomes the modification destination of a clause in the sentence of a report. For example, if the phrase “T1 weighted image in the right thalamus” is obtained, and the doctor moves the drawing device to draw a circle more than once by right-clicking, according to FIG. The phrase “high signal area” is obtained as the phrase, and for example, a sentence such as “A high signal area is seen in the right thalamus in the T1-weighted image” is created. As a result, the phrase modified by the phrase including the part name is quickly added to the finding or interpretation report being edited by an intuitive operation of the doctor.

  Again, referring back to FIG. For each medical image, the editing unit 30 associates the medical image, a sentence including a phrase, a part name, a part area, image meta information, and a region specified by a doctor, and retains document image related information as document image related information. Held by the unit 34. Thereafter, the display control unit 18 reads out the document image related information held in the document image related information holding unit 34 in response to the operation of the operation unit 105 (for example, a pointing device such as a mouse) by the doctor, and the document image related information. Is controlled to display on the monitor 106 medical images and sentences included in the.

  FIG. 10 is a diagram illustrating an example of display of document image related information. The display in FIG. 10 includes a plurality of medical images 211 and 212 obtained by photographing the same part, and a finding 220. This display is obtained by document image related information corresponding to the medical image 211 and document image related information corresponding to the medical image 212. When the doctor operates the operation unit 105 to specify a predetermined position of the medical image 211, the display control unit 18 corresponds to the part specified in the medical image 211 based on the document image related information corresponding to the medical image 211. The region of the region to be identified is specified. Further, the display control unit 18 specifies a part region in the medical image 212 corresponding to the part specified in the medical image 211 based on the document image related information corresponding to the medical image 212. Further, in the finding 220, the display control unit 18 underlines and highlights the sentence 222 including the phrase corresponding to the specified region in the document image related information corresponding to the medical image 211. Furthermore, the display control unit 18 highlights the sentence 224 including the clause corresponding to the specified region in the document image related information corresponding to the medical image 212 in the finding 220 by changing the color to bold. Thereby, the doctor can easily obtain a description of the part name corresponding to the predetermined region of the medical image.

  The search unit 36 searches the document image related information held in the document image related information holding unit 34 according to the operation of the operation unit 105 by the doctor, and documents related to medical images and findings displayed on the monitor 106. Get image related information. Specifically, when a medical image region in the displayed document image related information is designated, the search unit 36 is obtained by the same imaging method as that medical image, and the designated region is included in the designated region. Document image related information having a sentence including a part name representing a corresponding part, document image related information having a sentence including a part name representing a part corresponding to the designated region, and the like are acquired by searching. When a sentence in the displayed document image related information is specified, the search unit 36 acquires document image related information having a sentence including a part name included in the specified sentence by searching. Or, medical image capturing method, part name, part size, independent words included in sentences (nodule, patchy, high signal, lacunar infarction, white matter infarction, ischemia, multiple cerebrovascular disorder, signal change) When the search key is specified as a search key, the search unit 36 acquires document image related information having a sentence including the search key by searching.

  The document image related information acquired by the search is output to the display control unit 18, and the display control unit 18 performs control to display a medical image or sentence included in the input document image related information on the monitor 106. Thereby, the doctor can make use of the documents and images included in the other document image related information for reading and recognizing the documents included in the document image related information, recognizing images, and creating new findings and interpretation reports.

  Further, the phrase generation unit 24 may calculate and retain the degree of association between the part name and the medical image photographing method in advance. FIG. 11 is a diagram illustrating an example of the degree of association between a part name and an imaging method. The phrase generation unit 24 generates the phrase only when the degree of association between the part name and the imaging method is predetermined or greater than a predetermined threshold determined by the machine learning method when generating the phrase. These part names and imaging methods are inserted into the pattern. For example, when the degree of association between the part name and the imaging method is shown in FIG. 11 and the threshold is 0.5, a combination of the part name “cerebellum” and the imaging method “T2 weighted image” is inserted into the phrase pattern. Clauses are not generated. Thereby, it is possible to prevent an inappropriate phrase including a part name and an imaging method having a low degree of association from being generated.

  As described above, the document editing support system, the document editing support method, and the document editing support program according to the present invention are capable of document editing support suitable for creating a natural language sentence document. Useful.

It is a figure which shows the hardware constitutions of a document edit assistance system. It is a figure which shows the functional block diagram of a document edit assistance system. It is a figure which shows an example of a site | part name and a site | part area | region. It is a flowchart which shows operation | movement of a document edit assistance system. It is a figure which shows an example of a medical image and a site | part name candidate list. It is a figure which shows an example of image metadata. It is a figure which shows an example of a phrase pattern. It is a figure which shows an example of the sentence containing a clause. It is a figure which shows an example of the correspondence table of operation and a corresponding phrase. It is a figure which shows an example of a display of document image relevant information. It is a figure which shows an example of the association degree of a site | part name and imaging | photography method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 12 Medical image holding | maintenance part 14 Medical image analysis part 16 Analysis result holding | maintenance part 18 Display control part 22 Region specification part 24 Phrase production | generation part 26 Phrase pattern holding part 28 Image meta-information holding part 30 Editing part 32 Coordinate acquisition part 34 Document image related information Holding unit 36 Search unit 100 PC
102 Memory 103 HDD
105 Operation unit 106 Monitor 107 Internal bus

Claims (12)

A first holding unit that holds an area of each part of the object in the image of the object and a part name representing the part in association with each other;
When an area is specified in the image of the object displayed on the display screen, the area of each part of the object in the image of the object held by the first holding unit and the part representing the part First acquisition means for acquiring a part name as a candidate corresponding to the specified region from the name;
First display control means for displaying the part name candidates acquired by the first acquisition means on the display screen;
Generating means for generating a phrase including a part name designated from candidate part names displayed on the display screen by the first display control means;
A document editing support system, comprising: an adding unit that adds the clause generated by the generating unit to a document to be edited.   The document editing support system according to claim 1, wherein the generation unit generates a phrase by inserting the designated part name into a phrase pattern representing a predetermined phrase format. Meta information, which is information representing the attribute of the image, is associated with the image of the object,
A second acquisition unit configured to acquire meta information of the image;
The document editing support system according to claim 1, wherein the generation unit generates a phrase including a phrase corresponding to the meta information acquired by the second acquisition unit. The second acquisition means acquires meta information corresponding to each of a plurality of images including the same part with different imaging methods,
The document editing support system according to claim 3, wherein the generation unit generates a phrase including a phrase corresponding to the meta information for each piece of meta information acquired by the second acquisition unit. Third acquisition means for acquiring a phrase to be modified by a phrase including the part name from a predetermined phrase group;
5. The document editing support system according to claim 1, further comprising: an adding unit that adds the phrase acquired by the third acquiring unit to the phrase generated by the generating unit. It has a table showing the correspondence between each of a plurality of operations of the operation unit and the phrase,
6. The document according to claim 5, wherein the third acquisition unit acquires a word corresponding to the operation of the operation unit with reference to the table as a word to be modified by a phrase including the part name. Editing support system. A document in which a phrase including the part name is added, an image including a part represented by the part name included in the phrase, and a region of the part represented by the part name included in the phrase in the image are associated with each other. Second holding means for holding document image related information as information;
A document and an image included in the document image related information held by the second holding unit are displayed on the display screen, and a sentence including a part name corresponding to a specified area in the displayed image is specified. 7. The document editing support system according to claim 1, further comprising second display control means for displaying. When a sentence or image area included in the displayed document image related information is designated, a part name included in the designated sentence or a part name representing a part corresponding to the designated area First search means for searching for document image related information including a document to which a phrase including
8. A third display control means for displaying on the display screen at least one of a document and an image included in the document image related information obtained by the search by the first search means. Document editing support system described in 1. Second search means for searching for document image related information including the search key when a search key is input;
8. A fourth display control means for displaying on the display screen at least one of a document and an image included in the document image related information obtained by the search by the second search means. Or the document editing support system according to 8. Relevance calculating means for calculating relevance between the part name and the imaging method of the image,
The document editing support system according to claim 1, wherein the generation unit generates a phrase including the part name and the imaging method in which the relevance is a predetermined value or more. When an area is specified in the image of the object displayed on the display screen, the region of each part of the object in the image of the object held by the holding unit, and the part name representing the part, An acquisition step of acquiring a part name as a candidate corresponding to the specified region;
A display control step of displaying on the display screen the candidate site names acquired by the acquiring step;
A generating step for generating a phrase including a part name designated from candidate part names displayed on the display screen by the display control step;
A document editing support method, comprising: an adding step of adding the clause generated by the generating step to a document to be edited. When an area is specified in the image of the object displayed on the display screen, the region of each part of the object in the image of the object held by the holding unit, and the part name representing the part, An acquisition step of acquiring a part name as a candidate corresponding to the specified region;
A display control step of displaying on the display screen the candidate site names acquired by the acquiring step;
A generating step for generating a phrase including a part name designated from candidate part names displayed on the display screen by the display control step;
A document editing support program for causing a computer to execute an adding step of adding a clause generated by the generating step to a document to be edited.

Images