JP6719259B2 - Medical report creation support device, control method thereof, and program - Google Patents

Description

The present invention relates to a medical report creation support device, a control method thereof , and a program .

An interpretation report has been created in which findings such as CT (Computed Tomography) and MRI (Magnetic Resonance Imaging) observation results of images and diagnostic results derived from the findings are described. Nowadays, with the introduction of information systems, image interpretation reports are electronically recorded and managed instead of paper media, and a large amount of data is accumulated daily as medical information. If the computer can properly understand this large amount of data, it can be secondarily used for data analysis and machine learning.

Generally, the interpretation report is a free description in natural language. In such a general image interpretation report, one report is created for each image examination, and the description and diagnosis columns are prepared therein. In addition, it is rare to handle only one item in one interpretation, and many items are often handled. As a result, a plurality of items will be described in each of the description and diagnosis columns.

However, when multiple items are listed in one column, there is a mechanism that clearly indicates which of the multiple items listed in the finding column is related to which of the multiple items listed in the diagnostic column. Absent. It is possible for humans to understand the correspondences by deciphering the text, but it is difficult for the system to understand. Further, even if the system can recognize which item is connected to which item, it is also important to recognize what kind of relationship these items are connected to. For example, if there is a correspondence between the "massive lesion" in the Finding column and "lung cancer" in the Diagnostic column, and if there is a correspondence between "Ascites" in the Finding column and "Cirrhosis" in the Diagnostic column , Their relationships are different. The former is described in a natural sentence as "this mass lesion is lung cancer". That is, the disease name described in the diagnosis column indicates the content itself described in the finding column. On the other hand, the latter is expressed in a natural sentence as "this ascites is caused by cirrhosis". In other words, it indicates that the content described in the finding column has a secondary onset due to the disease name described in the diagnosis column. As described above, even in the combination of the items described in the same finding column and the diagnostic column, the types of relationships between them are obviously different.

When secondarily using the information described in the report, prepare for the computer to recognize and process the information, and provide information about which item is connected to which item and the type indicating the relationship between the items. It is very useful to store it in advance. In addition, it is preferable that the information to be secondarily used has medical consistency in advance, and a mechanism for checking the consistency of the contents described in the report is also required. There are Patent Document 1 and Patent Document 2 as techniques for checking the consistency of the contents described in the report. The image interpretation report creation support device of Patent Document 1 analyzes the contents input to the report system, and creates the items to be described in the report as an essential requirement list based on the analysis result. It is possible to create a consistent report by notifying the items that are not described in the essential requirement list in advance. The pathological diagnosis report creation support device of Patent Document 2 determines whether the contents of the finding information and the diagnostic information are classified as malignant or benign, and notifies when the classification result is different between the finding information and the diagnostic information. Support the creation of medically consistent reports.

JP-A-2009-93568 JP, 2015-95248, A

In order to clearly distinguish and store the type of relationship between items, the user specifying the type of relationship between items each time leads to a decrease in report creation efficiency. Further, no means has been proposed for checking whether the information between the items is consistent or not. In Patent Document 1, it is possible to prevent the omission of the description of the essential entry requirement, but it is not supported regarding the information regarding the relationship between the items (requirements). Further, in Patent Document 2, it is possible to prevent a contradiction between the description content of the finding information and the description content of the diagnostic information, but it is not possible to perform a prior check on the type of relationship between the description items.

The present invention has been made in view of the above problems, and an object of the present invention is to efficiently create a structured medical report in which a relationship between items is defined.

A medical report creation support device according to an aspect of the present invention for solving the above-mentioned problems has the following configuration. That is,
A medical report generation support apparatus for creating a medical for the report,
The medical report includes a plurality of items corresponding to any of a plurality of predefined attributes,
Selecting means for selecting a first item and a second item from the plurality of items included in the medical report;
And types of attributes, each of the first item and the second item corresponds, lesions are described in the first item or the second item and whether a predetermined type of lesion, the based on a determination means for determining from among a plurality of relationships that are defined with respect to between pre-item relationship between items of the first item and the second item,
And a search means for searching the medical report or the description included in the medical report based on the specified search condition ,
As the search condition, Ru obtained including the relationship between the items.

According to the present invention, it is possible to efficiently create a structured medical report in which the relationship between items is defined.

The figure which shows an example of a structure of the image interpretation report creation system by 1st Embodiment. The figure which shows an example of a structure of the image interpretation report preparation apparatus by 1st Embodiment. The figure which shows an example of the concept of the report produced with the image interpretation report production apparatus. The figure which shows an example of the database storage format of the item and related information which comprise an image interpretation report. The figure which shows an example of the function of the image interpretation report preparation apparatus of 1st Embodiment. The figure which shows an example of GUI of an image interpretation report preparation apparatus. The figure which shows an example of the procedure of the process which automatically determines the kind of relationship between report items. The figure which shows an example of the procedure of the process which classifies the kind of lesion. It is a figure which shows the keyword which shows the lesion for which the description item is described, and an example of the mapping table of the type of lesion. It is a figure which shows an example of the definition of the kind of lesion. It is a figure which shows an example of the procedure which manually specifies the kind of relationship between a finding item and a diagnostic item. It is a figure which shows an example of the procedure of the inconsistency check of the kind of relationship between a finding item and a diagnostic item. The figure which shows the example of the property for every kind of lesion.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The configurations shown in the following embodiments are merely examples, and the present invention is not limited to the configurations shown below. Although an image interpretation report creating apparatus that creates an image interpretation report will be described below, the creation target is not limited to the image interpretation report. INDUSTRIAL APPLICABILITY The present invention can be applied to a medical report creation support device for creating a medical report (interpretation report, diagnostic report, etc.) having a structure in which a plurality of items having different attributes are defined in which the relationship between items is defined.

<First Embodiment>
The image interpretation report creating apparatus according to the first embodiment is built in a medical institution such as a hospital. FIG. 1 is a diagram showing a configuration example of an image interpretation report creating system according to the first embodiment. The image interpretation report creating apparatus 1 is an apparatus for an image reading doctor to browse an image to be imaged and create an image reading report. The image interpretation report creating apparatus 1 is connected to an in-hospital system via a local area network 10 in the hospital. The in-hospital system includes, for example, HIS2, RIS3, medical image capturing apparatus 4, and PACS5. HIS2 (Hospital Information System) is a comprehensive system including a medical office accounting system, a medical treatment appointment system, a medical treatment information system, and the like. The HIS 2 also has an electronic medical record database (hereinafter, electronic medical record DB 6) and an image interpretation report database (hereinafter, image interpretation report DB 7). The electronic medical chart DB 6 stores an electronic medical chart in which medical information of patients is recorded. The image interpretation report DB 7 stores an image interpretation report created by an image interpretation doctor and information related thereto. The RIS3 (Radiology Information System) is a system for performing inspection and treatment reservations by radiological equipment, management of inspection results, material inventory management, and the like. The interpretation report DB 7 may be managed by the RIS 3.

The medical image capturing apparatus 4 is a general term for a CT (Computed Tomography) apparatus, a magnetic resonance tomography apparatus (MRI), an ultrasonic diagnostic apparatus, a PET (Positron Emission Tomography) apparatus, and the like. The PACS 5 (Picture Archiving and Communication System) is a system for electronically storing, retrieving, and communicating medical images captured by the medical image capturing apparatus 4. The PACS 5 has a medical image database (hereinafter, medical image DB 8) as a storage destination of medical images. The lesion type database (hereinafter, lesion type DB) stores data used when classifying lesion types.

FIG. 2 is a block diagram showing a specific configuration example of the image interpretation report creating apparatus 1. The input device 11 is a mouse, a digitizer, or the like, and is used to input a user instruction to the image interpretation report creating apparatus 1. The output device 12 is an LED (Light Emitting Diode) that displays the state of the device or processing contents, a liquid crystal panel, or the like. A CPU (Central Processing Unit) 13 reads a program stored in the secondary storage device 15 into the primary storage device 14. Further, the CPU 13 interprets and executes the read program to perform various controls and calculations in the device and display a UI (User Interface).

The primary storage device 14 is mainly a memory such as a RAM (Random Access Memory). The secondary storage device 15 stores a medical report creation program necessary for operating the image interpretation report creation device 1. The secondary storage device 15 corresponds to a storage medium such as a hard disk or a flash memory, but it goes without saying that the present embodiment does not depend on a specific storage medium. Generally, the capacity of the primary storage device 14 is smaller than the capacity of the secondary storage device 15, and programs and data that cannot be stored in the primary storage device 14 are stored in the secondary storage device 15. In addition, data that must be stored for a long time is also stored in the secondary storage device 15. The communication interface 16 is connected to the local area network 10 in the hospital and realizes communication with various systems/devices in the hospital.

FIG. 3 is a diagram showing the concept of a report created by the image interpretation report creating apparatus 1 of the first embodiment. The report created by the image interpretation report creating apparatus 1 is composed of a plurality of report items and relationship information between the report items. In FIG. 3, the report items are indicated by a frame 71. The report item is classified into any of four types of attributes of a site, a finding, a diagnosis, and a treatment according to the content thereof.

In the report item whose attribute is a body part, a sentence or an image that can identify the body part is input. The image input to the report item of the part is, for example, a soft copy of the image to be read, or address information for accessing the image to be read. At this time, parameters of image processing applied when the image is browsed may be input together. A text explaining one lesion is input to the report item whose attribute is the finding. A text explaining one disease is input to the report item whose attribute is diagnosis. In the report item whose attribute is treatment, a sentence explaining one or more treatments is input.

The relationship information representing the relationship between the report items is classified into three types of “correspondence relationship”, “identical relationship”, and “cause/result relationship (hereinafter, causal relationship)” according to the relationship. In FIG. 3, the correspondence is indicated by a single line 72, the same relation is indicated by a double line 73, and the causal relation is indicated by an arrow line 74. Correspondence means that two report items are related to each other. The same relationship means that two report items indicate the same thing. A causal relationship means that two report items are connected by a cause-effect relationship. A report item can have relationships with other report items. That is, the relationship between report items is a many-to-many relationship.

FIG. 5 is a functional block diagram for explaining main functions of the image interpretation report creating apparatus 1 according to the present embodiment. The image browsing unit 101 reads an image to be an image interpretation target from the medical image DB 8 via the PACS 5 and displays it on the output device 12. In addition, in accordance with an operation of the image interpretation doctor via the input device 11, various processing related to image browsing such as change of display image and application of image processing is executed. The item creation unit 102 creates a report item and automatically generates an ID for uniquely identifying the report item, a report ID for uniquely identifying the report to which the report item belongs, and an item attribute (hereinafter, also referred to as item type). And then saves it in the item table 701 shown in FIG.

The item content editing unit 103 edits the content of the report item according to the user operation, and saves the edited result in the corresponding portion of the item table 701. The relationship information creation unit 104 creates relationship information between arbitrary report items. As shown in FIG. 4B, the relationship information is composed of a relationship ID for uniquely identifying the relationship information, two report item IDs (ID1 and ID2), and a type of relationship between those items. To be done. The relationship information creation unit 104 automatically determines the type of relationship between the two selected report items. The method of determining the type of relationship will be described later with reference to the flowchart of FIG. The relationship information storage unit 105 stores the relationship information created by the relationship information creation unit 104 in the relationship table 702 as shown in FIG. When the type of relationship is causal, the report item ID1 is the cause and the report item ID2 is the result.

The lesion type determination unit 106 classifies the lesion, which is the target of the finding item, into a predetermined type (localized lesion or non-localized lesion in this embodiment). The storage unit 107 temporarily stores the item table 701 and the relationship table 702. It should be noted that, for convenience of description below, when the relationship table 702 is stored, values are assigned so that the relationship IDs are always serial numbers. The relationship information presenting unit 108 displays the relationship between the report items on the output device 12 according to the content of the relationship table 702, for example, by using the above-mentioned single line 72, double line 73, and arrow line 74. The report storage unit 109 stores the contents of the item table 701 and the relation table 702 as report data in the radiology report DB 7 via the HIS 2. The format of the report data does not matter as long as it has the same information as the item table 701 and the relation table 702. For example, it may be converted into the format of report data defined by XML or the like and stored.

FIG. 6 is a diagram showing an outline of an example of a GUI (graphical user interface) provided by the display control of the image interpretation report creating apparatus 1 according to the first embodiment. FIG. 6 shows an example of the report input screen. The image browsing unit 101 displays an image to be interpreted on the medical image browsing screen 131. The image browsing unit 101 realizes various browsing functions according to operations on the browsing screen 131. In the patient information display area 132, patient information of the patient who is the subject of the image to be interpreted is displayed. For example, patient ID, name, age, past medical history, etc. correspond to patient information. In the request information display area 133, the department information of the requester, the requested examination, and the request contents from the requesting doctor are displayed.

The report creation area 134 is divided into areas for parts, findings, diagnoses, and treatments corresponding to the types (attributes) of the items, and an input frame 135 for inputting the contents of the report items in each area. An add button 136 for adding the input frame 135 is displayed. The input frame 135 is associated with one report item, and the item content editing unit 103 receives an editing operation of the content. It should be noted that the description content for each item is displayed in the input frame 135, and the connection between the associated input frames can identify the type of relationship (corresponding relationship, identical relationship, causal relationship) determined for the items. It is displayed in the form. The item content editing unit 103 realizes an image interpretation report editing function according to an operation on the input frame 135. When the add button 136 is pressed, the item creating unit 102 creates a new report item and displays the input frame 135 corresponding to the created report item. The add button 136 is prepared for each item type (site, finding, diagnosis, treatment), and the item creating unit 102 creates a report item of a corresponding type according to the pressed button.

The addition of the input frame may be performed by a key operation via the input device 11. For example, a diagnosis report item may be created by pressing the TAB key while editing the contents of the finding report item. Similarly, the report key of another finding may be created by pressing the return key twice in succession while the content of the report item of the finding is being edited. The input frame 135 to be edited is selected via the input device 11. A plurality of input frames may be selected at the same time. Further, as will be described later, a structured image interpretation report is generated by associating items with each other and defining their types.

When the selected input box is dragged and dropped to another input box, it is considered that the relationship between the report item corresponding to the input frame of the drag source and the report item corresponding to the input frame of the drop destination has been instructed. Be done. At this time, the relationship information creation unit 104 and the relationship information storage unit 105 determine the relationship between the drag-source and drop-destination report items, and execute processing to store the relationship information. In the report creation area 134, the relationship information presenting unit 108 clearly indicates the relationship between the report items determined by the relationship information creating unit 104 by the type of line connecting the input frames 135. As described above, in the present embodiment, when the types of relationships are corresponding relationships, the input frames are connected by a single line, and when the relationship types are the same, the input frames are connected by a double line. It Further, when the type of relationship is a causal relationship, an arrow line is displayed from the input frame of the report item that is the cause to the input frame of the report item that is the result.

When an image is dragged and dropped from the browsing screen 131 to the report creation area 134, the item creation unit 102 creates a report item of a part and displays an input frame corresponding to it. At the same time, the item creating unit 102 creates a report item of a finding having a correspondence relationship with the created report item of the body part, and displays the input frame. Further, when an image is dragged and dropped from the browsing screen 131 to the input frame 135 of the already existing report item, the item content editing unit 103 registers the soft copy of the image as the content of the drop destination report item. When the report save button 137 is pressed, the report save unit 109 registers the report being created in the image interpretation report DB 7 (updates the item table 701 and the relationship table 702).

FIG. 7 is a flowchart showing a procedure of processing for automatically determining the type of relationship between report items by the relationship information creation unit 104. Here, as an example, a case where the relationship between report items is instructed by dragging and dropping the input frame 135 will be described. It should be noted that a plurality of drag-source report items and drop-destination report items are both selected, and the n-th report item in each set can be referred to by DRAG(n) and DROP(n). In the present embodiment, the processing is started in a state where a plurality of drag source report items and a plurality of drop destination report items are specified, but the following processing (relationship between items) is performed every time drag and drop is performed. It is clear that the process of determining the type of) may be performed. Further, the process of determining the type of relationship between items may be performed for all combinations of all report items existing in the report.

In step S701, the relationship information creation unit 104 substitutes 0 for the variable i. In step S702, the relationship information creation unit 104 substitutes the type of DRAG(i) into the variable Ki. In step S703, the relationship information creation unit 104 substitutes 0 for the variable j. In step S704, the relationship information creation unit 104 substitutes the type of DROP(j) into the variable Kj.

As described above, the first item and the second item (DRAG(i) and DROP(j)) for which the relationship between the items is determined are selected from the plurality of items included in the interpretation report. To be done. Then, in the following steps S705 to S715, based on the attributes of the first item and the second item and the lesion described in the first item or the second item, the first item and the second item The type of relationship between the items of is determined. First, in step S705, the relationship information creation unit 104 compares the type Ki of DRAG(i) with the type Kj of DROP(j). If the item types are both diagnostic, the process proceeds to step S706; otherwise, the process proceeds to step S707. In step S706, the relationship information creation unit 104 determines that the relationship between DRAG(i) and DROP(j) is a causal relationship in which the finding of DRAG(i) and DROP(j) is the cause and the diagnosis is the result. To do.

In step S707, the relationship information creation unit 104 determines whether one of the type Ki of DRAG(i) and the type Kj of DROP(j) is a finding and the other is a diagnosis. If one is a finding and the other is a diagnosis, the process proceeds to step S708; otherwise, the process proceeds to step S711. In step S708, the lesion type determination unit 106 determines whether the type of the lesion described in the report item whose type is the finding, of DRAG(i) and DROP(j), is a localized lesion. .. The method of determining the type of lesion will be described later with reference to FIG. If it is determined that the lesion is a localized lesion, the process proceeds to step S709, and otherwise proceeds to step S710. In step S709, the relationship information creation unit 104 sets the same relationship between DRAG(i) and DROP(j). On the other hand, if the type of lesion is not a localized lesion, in step S710, the relationship information creating unit 104 determines whether the type of item is the finding of DRAG(i) and DROP(j), Set a causal relationship that results in the type being diagnosed.

In step S711, the relationship information creation unit 104 determines whether one of the type Ki of DRAG(i) and the type Kj of DROP(j) is a diagnosis and the other is a treatment. If one is a diagnosis and the other is a treatment, the process proceeds to step S712, and otherwise the process proceeds to step S713. In step S712, the relationship information creation unit 104 has a causal relationship in which one of DRAG(i) and DROP(j) whose item type is diagnostic is the cause, and which one is the treatment is the result. And set this between DRAG(i) and DROP(j).

In step S713, the relationship information creation unit 104 determines whether one of the type Ki of DRAG(i) and the type Kj of DROP(j) is a site and the other is a finding. If one is the site and the other is the finding, the process proceeds to step S714, otherwise the process proceeds to step S715. In step S714, the relationship information creation unit 104 determines the relationship between DRAG(i) and DROP(j) as the corresponding relationship. In step S715, the relationship information creation unit 104 determines that there is no connection relationship between DRAG(i) and DROP(j), and does not set the relationship information.

The relationship information creation unit 104 adds 1 to the variable j in step S716, and compares the variable j with the number of drop-destination report items in step S717. If the variable j is equal to or larger than the number of report items to be dropped, the process proceeds to step S718, and otherwise returns to step S704. Also, the relationship information creation unit 104 adds 1 to the variable i in step S718, and compares the variable i with the number of drag-source report items in step S719. If the variable i is equal to or more than the number of report items as the drag source, the process ends, otherwise the process returns to step S702.

The relationship information storage unit 105 generates and updates the relationship table 702 as shown in FIG. 4B based on the relationship information between items determined as described above.

In the image interpretation report creating apparatus 1 according to the first embodiment, report items are connected in a form (for example, a single line, a double line, and an arrow line) according to the type of relationship determined as described above. As a result, it becomes possible to express the flow of logic relating to the connection between the site, the finding, the diagnosis, and the treatment in an easily understandable manner.

FIG. 8 is a flowchart showing a process in which the lesion type determination unit 106 determines the type of lesion. In step S801, the lesion type determination unit 106 analyzes the content described in the finding input frame 135 and extracts a keyword indicating a lesion in which the finding is described. Keywords extracted include, for example, nodules, tumors, infiltrative shadows, ascites, atrophy of the liver, and areas of increased concentration. In step S802, the lesion type determination unit 106 refers to the table in which the keyword indicating the lesion and the lesion type are mapped, and determines whether the lesion type is the localized lesion or the non-local lesion. In step S708 of FIG. 7, the process branches according to the determination result. Although it is determined in step S708 of FIG. 7 whether or not the lesion is a localized lesion, it is clear that whether or not the lesion is a non-local lesion may be determined (however, in this case, the branches of YES and NO are interchanged). ).

FIG. 9 is an example of a mapping table of a keyword indicating a lesion and a lesion type, which is referred to in step S802. There are two types of lesions, localized lesions and non-localized lesions, and the lesions whose findings are to be described are classified into either one. In the present embodiment, the type of lesion is defined by the range of pathological changes.

For example, as shown in FIG. 10A, a pathological change is limited to a narrow range and a lesion that can clearly show the position of the lesion on an image is defined as a localized lesion. Localized lesions are localized abnormalities that occupy a certain volume. The localized lesion surface is solid and the interior can be either solid or gas. On the other hand, as shown in FIG. 10(b), when the pathological change extends over a wide range with respect to the location where the lesion occurs, or when the physical location of the lesion changes depending on the image capturing environment (including the patient's posture). , The lesion is defined as a non-local lesion. Nonlocal lesions are abnormalities in which the abnormal part cannot be fixedly and clearly separated from the normal part.

Although the procedure for determining the type of the relationship between the finding item and the diagnostic item is based on the condition that the finding item and the diagnostic item are created in advance, the present embodiment is not limited to such a configuration. .. For example, while describing the content of the finding in the input frame 135 in the report item of the finding (or at the timing when a punctuation mark indicating the end of the description is input), the keyword indicating the lesion for which the finding is described is changed. You may make it detect. In that case, the relationship information creation unit 104 determines the type of lesion (in this embodiment, localized lesion or non-localized lesion) from the detected keyword, and selects a blank diagnostic item related to the report item being described. Input frame is automatically generated. At that time, the relationship between the finding item and the diagnostic item is automatically classified into one of the same relationship and the causal relationship as described in FIG. 7 (S708 to S710) according to the determined lesion type. , Relationship table 702.

Further, the type of relationship between findings and diagnoses determined as described above can be used, for example, to search for past reports or search for descriptions (report components) contained within a single report. I can. Conventionally, when searching for a report element including a desired disease name, the disease name has been designated as a search condition as a search condition. In that case, the disease name in the search condition is treated as the same as the lesion for which the finding is to be described, and the disease name that is described as the cause of the lesion described in the finding is mixed. There is a possibility that it will be searched. On the other hand, according to the first embodiment, the type of relationship between items can be used as the search condition. Therefore, for example, “disease name”+“relationship with findings” can be specified as the search condition, and it is possible to perform a search with more detailed search conditions than in the past and to improve the accuracy of the search. Can be improved.

As described above, according to the first embodiment, the type of relationship between the description items in the image interpretation report is automatically determined, so that the report can be created efficiently.

<Second Embodiment>
Next, a second embodiment will be described. In the second embodiment, the technique of automatically determining the type of the relationship between the finding and the diagnosis according to the type of the lesion in which the finding is described, which is described in the first embodiment, is applied to the inconsistency check of the existing report. As described in the first embodiment, the type of relationship between items can be automatically determined, but the user may explicitly specify the type of relationship between items. In practice, some doctors may require high accuracy in the automatic determination of the type between items, and many doctors want to give priority to visual input even if it takes time and effort. In such a case, the automatic determination of the type of relationship between items (S705 to S715 in FIG. 7) can be used for the purpose of checking the relationship between items explicitly designated by the user.

FIG. 11 shows an example of a procedure for realizing the association between the finding item and the diagnostic item and the determination of the type of the relationship by the user operation. In FIG. 11A, the user drags and drops the input frame 135a for finding or diagnosis to the other input frame 135b. In FIG. 11B, the finding item and the diagnostic item are connected as a result of the operation of FIG. 11A. In FIG. 11C, for example, by right-clicking a straight line indicating a relationship, a menu for specifying the type of connection relationship is displayed, and the user specifies the type of relationship by mouse operation. The specified relationship is stored in the relationship table 702.

FIG. 12 shows a flowchart of processing for checking the inconsistency of the type of the relationship between the finding item and the diagnostic item for the image interpretation report created by relating the items as shown in FIG. .. In FIG. 12, it is assumed that ITEM(x) can refer to the report item having the item ID x in the item table 701. Further, it is assumed that the relation information of the relation ID x in the relation table 702 can be referred to by RELATION(x).

In step S1201, the relationship information creation unit 104 substitutes 1 for the variable i. In step S1202, the relationship information creation unit 104 refers to the relationship table 702 and substitutes the report item ID1 in the relationship information RELATION(i) with the relationship ID i into the variable m. Then, the relationship information creation unit 104 refers to the item table 701 and substitutes the type in the report item ITEM(m) whose item ID is m into the variable Km. In step S1203, the relationship information creation unit 104 refers to the relationship table 702 and substitutes the report item ID2 in the relationship information RELATION(i) with the relationship ID i into the variable n. Then, the relationship information creation unit 104 refers to the item table 701 and substitutes the type in the report item ITEM(n) whose item ID is n into the variable Kn.

In step S1204, the relationship information creation unit 104 refers to the item table 701, and the report IDs of ITEM(m) and ITEM(n) match the report ID of the report that is the target of the inconsistency check. Check. If they match, the process proceeds to step S1205, and if they do not match, the process proceeds to step S1210. In step S1205, the relationship information creation unit 104 checks whether either Km or Kn is a finding and the other is a diagnosis. If either Km or Kn is a finding and the other is a diagnosis, the process proceeds to step S1206; otherwise, the process proceeds to step S1210.

In step S1206, the kind of relationship between the findings and the items of diagnosis is determined by the procedure of steps S708 to S710 of FIG. 7 described in the first embodiment. In step S1207, the relationship information creation unit 104 compares the type of RELATION(i) with the type of the desired relationship determined in step S1206. As a result of the comparison, if the types of relationships match, the process proceeds from step S1208 to step S1210; otherwise, the process proceeds to step S1209. In step S1209, the relationship information creation unit 104 notifies the user that the type of relationship between the finding and the diagnosis is incorrect. In step S1210, the relationship information creation unit 104 increments the variable i, and in step S1211, checks whether the value of the variable i is larger than the number of records in the relationship table 702. If the value of the variable i is larger than the number of records in the relation table 702, the inconsistency check process ends, otherwise the process returns to step S1202.

According to the second embodiment, by the series of processes described above, it is possible to specify to the user that the type of relationship between the finding and the item of diagnosis explicitly specified by the user is medically incorrect.

In the second embodiment described above, the correctness of the relationship between terms described in the relationship table 702 is verified, but the invention is not limited to this. For example, when the relationship between items is specified by the procedure as shown in FIG. 11, it may be determined whether or not the specified type of relationship is appropriate. By doing so, even when the type of relationship between items is explicitly designated by a user operation, it is possible to prevent designation mistakes in advance. Further, in the above description, the report item type is described as a combination of findings and diagnosis, but the present invention is not limited to this. For example, when the types of report items are diagnosis and treatment, the type of relationship may be determined according to steps S711 to S712 and compared with the type of relationship described in the relationship table 702. Further, in the above, if the items to be associated are present in different reports (NO in step S1204), a warning to that effect may be issued.

<Third Embodiment>
Next, a third embodiment will be described. The third embodiment shows an example in which the consistency check target of the image interpretation report described in the second embodiment is expanded in addition to the type of relationship between findings and diagnosis. In the third embodiment, of a plurality of items included in an image interpretation report, an item describing a lesion having a predetermined attribute (for example, a non-local lesion, a tangible lesion, or an abnormal area as described below) is acquired. To do. Then, it is determined whether or not the acquired item satisfies the restriction defined for the lesion having the predetermined attribute, and when it is determined that the restriction is not satisfied, an error is notified.

In the third embodiment, the following (1) and (2) restrictions are set for the report description format. In addition, below, the item whose attribute is a finding is described as a finding item, and the item of a diagnosis is described as a diagnostic item.
(1) One lesion is described in one finding item (input frame 135).
(2) When a plurality of diagnostic items are associated with one finding item, clearly indicate whether or not the plurality of diagnostic items are simultaneously established. In addition, in the first and second embodiments described above, the types of lesions are roughly classified into localized lesions and non-localized lesions from the range of pathological changes, but in the third embodiment, localized lesions are further divided into two types. Subdivide into two types. The two subdivided types are defined as "tangential lesion" and "abnormal area" for convenience.

Tangible lesions include nodules and tumors. Tangible lesions are those for which there is a diagnosis to be made and the diagnosis is contradictory. For example, a lung nodule is diagnosed with lung cancer but not with lung metastasis. In addition, tangible lesions often occur in the same type. Further, the abnormal area includes shadows and areas (high density area, low density area). The abnormal area is not always an antinomy to the diagnosis attached to it. For example, it is possible that radiation pneumonitis, which is an abnormal region, and bacterial pneumonia overlap in the same region. Moreover, the same type of abnormal regions may frequently occur. The non-local lesion described above basically does not occur frequently in the same kind within the medically distinguishable region.

FIG. 13 shows the characteristics of each type of lesion described above (whether the same type occurs frequently or whether the diagnoses are contradictory to each other). The relationship information creation unit 104 extracts the findings describing the non-local lesions from the interpretation report described according to the report writing method of the third embodiment, and describes the lesions in which the findings are of the same type. Check that When there are a plurality of finding items for the lesion of the same type, the relationship information creating unit 104 gives an error notification regarding the finding item describing the non-local lesion. In addition, the relationship information creation unit 104 extracts a finding in which a tangible lesion is described, and when a plurality of diagnostic items are associated with the finding, whether or not the findings are contradictory. Check. When it is not explicitly stated that the relationship is contradictory, the related information creating unit 104 warns the user to that effect. The above processing is executed, for example, after the processing of FIG. 12 is completed.

<Other embodiments>
Although the embodiments have been described in detail above, the present invention can be embodied as a system, a device, a method, a program, a recording medium (storage medium), or the like. Specifically, it may be applied to a system composed of a plurality of devices (for example, a host computer, an interface device, an imaging device, a web application, etc.), or may be applied to a device composed of one device. good.

Further, the present invention supplies a program that implements one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus execute the program. It can also be realized by a process of reading and executing. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

1: Interpretation report creation device, 2: HIS, 3: RIS, 4: Medical image capturing device, 5: PACS, 6: Electronic medical record DB, 7: Interpretation report DB, 8: Medical image DB, 9: Lesion type data DB 10: local area network, 71: report item, 72: single solid line, 73: double line, 74: arrow line

Claims (11)

A medical report generation support apparatus for creating a medical for the report,
The medical report includes a plurality of items corresponding to any of a plurality of predefined attributes,
Selecting means for selecting a first item and a second item from the plurality of items included in the medical report;
And types of attributes, each of the first item and the second item corresponds, lesions are described in the first item or the second item and whether a predetermined type of lesion, the based on a determination means for determining from among a plurality of relationships that are defined with respect to between pre-item relationship between items of the first item and the second item,
And a search means for searching the medical report or the description included in the medical report based on the specified search condition ,
As the search condition, the medical report generation support device according to claim give Rukoto including the relationship between the items. The determination unit is a kind findings of attributes of the first item, when the type of the attribute of the second item is a diagnostic lesions described in the first item is a predetermined type The medical report creation support device according to claim 1, wherein the relationship between the items is determined from the plurality of relationships based on whether or not there is a lesion . The medical report creation support device according to claim 2, wherein whether or not the lesion is the predetermined type corresponds to a result of classification of whether or not the lesion is a localized lesion. When the type of lesion described in the first item is a localized lesion, the determining unit determines the relationship between the items to be the same, and determines the lesion described in the first item . The medical report creation support apparatus according to claim 3, wherein the relationship between the items is determined to be a causal relationship when the type is not a localized lesion. 5. The selection unit selects two items that are drag and drop targets of a user as the first item and the second item, according to any one of claims 1 to 4. Medical report creation support device. Further comprising display control means for displaying the contents of the medical report on the display device,
6. The display control means displays the description content for each item, and also displays the type of relationship determined between items in a form that can be identified. The described medical report creation support device. The selecting means selects, as the first item and the second item, items for which mutual relationships are defined based on relationship information indicating relationships between items included in the medical report.
Determination means for determining whether or not the relationship between the items determined by the determination means and the relationship defined by the relationship information match.
Wherein when it is determined not to match the determining means, the medical report generation support device according to any one of claims 1 to 6, characterized in that it comprises further a notification unit, a so notify the user .. The medical report creation support apparatus according to claim 1, wherein the plurality of types of attributes include findings and diagnoses. The type of relationship between the items between the first item and the second item is selected from a plurality of types of relationships including a correspondence relationship, the same relationship, and a causal relationship. Item 9. The medical report creation support device according to any one of items 1 to 8. A method of controlling a medical report creation support apparatus for generating a medical for the report,
The medical report includes a plurality of items corresponding to any of a plurality of predefined attributes,
A selection step of selecting a first item and a second item of the plurality of items included in the medical report;
And types of attributes, each of the first item and the second item corresponds, lesions are described in the first item or the second item and whether a predetermined type of lesion, the based on a determination step of determining a plurality of relations defined for advance between items the relationship between items of the first item and the second item,
A search step of searching the description contained in the medical report or medical report based on the specified search conditions, possess,
As the search condition, the control method of the medical report generation support device according to claim give Rukoto including the relationship between the items. A program for causing a computer to function as each unit of the medical report creation support device according to any one of claims 1 to 9.