JP6643433B2 - Supporting apparatus for creating an interpretation report and its control method - Google Patents

Description

  The present invention relates to a support device that supports creation of an interpretation report and a control method thereof.

  2. Description of the Related Art In the medical field, image diagnosis is performed in which a doctor interprets medical image data obtained by an imaging device such as an X-ray CT device or an MRI to make a diagnosis. In image diagnosis, in response to a request for interpretation from the attending physician, a doctor comprehensively judges findings (interpretation findings) and various measurement values obtained from the image, and specifies symptoms of a lesion appearing in the image. Then, the doctor summarizes the circumstances leading to the diagnosis as an image interpretation report to the requesting attending physician using the image interpretation findings and measured values.

  In creating such an image interpretation report, a technique is known in which the gaze of the image interpreter is detected to grasp the confirmation status of medical image data at the time of creating the image interpretation report. Patent Literature 1 describes a technique in which input information at the time of creating a report and a portion (attention information) in medical image data that is of interest when the input information is input are stored in association with each other. According to this technique, it is possible to grasp which part of the medical image data was focused on and created the image interpretation report. Patent Literature 2 discloses a technique in which an observation point to be observed is specified using test information or a computer-aided diagnosis function, and a determination is made as to whether the specified observation point has been confirmed based on line-of-sight information. According to this technique, it is possible to grasp whether or not a predetermined site to be observed has been confirmed.

JP 2009-045131 A JP 2006-271528 A

Kamiyama et al .: “Simultaneous Extraction of Multiple Organs from Abdominal CT Image Based on Fusion Movement Graph Cut”, IEICE Technical Report, MI2010-122, 2011-01

  However, in Patent Literature 1, input information and attention information are stored in association with each other, and an attention area where no input is performed (an area which is noticed by the user but is not described in the report) or an input information and attention information is not stored. No consideration was given to consistency with the information. Further, according to the technique of Patent Literature 2, it is possible to determine whether an input has been performed at a predetermined observation point to be observed, but whether an input has been performed for an area that is not predetermined. Cannot be determined. In other words, if the doctor had a finding about an area that was not prompted to observe but forgot to enter it (hereinafter, referred to as omission of description), he could not call attention. Also, since the consistency between the input content and the attention information is not taken into consideration, for example, attention should be paid to a mismatch (hereinafter, referred to as omission of confirmation) in which the area pointed out in the input content does not exist in the attention information. I couldn't prompt.

  The present invention has been made in view of the above-described problems, and has as its object to provide support for reducing the occurrence of omission of description and / or omission of confirmation and creating an appropriate interpretation report.

A support device according to one embodiment of the present invention for achieving the above object has the following configuration. That is,
A support device for supporting creation of an interpretation report,
Display means for displaying medical image data to be interpreted,
A first acquisition means for acquiring a target area that is determined as the target area during the interpretation in the medical image data displayed by the display means,
A second acquisition means for acquiring the description area is a medical area described in the report statement corresponding to the medical image data,
Determining means for determining consistency of the description area and the target area,
Presentation means for presenting information based on the result of the determination by the determination means.

  ADVANTAGE OF THE INVENTION According to this invention, generation | occurrence | production of a description omission or a confirmation omission can be reduced in preparation of an interpretation report.

FIG. 2 is a block diagram illustrating a functional configuration example of an image interpretation report creation support device. FIG. 2 is a block diagram illustrating an example of a hardware configuration of an image interpretation report creation support device. 9 is an exemplary flowchart illustrating the procedure of an interpretation report creation support process according to the embodiment. The figure which showed the data structural example of area information. FIG. 6 is a diagram showing a display example at the time of image interpretation on a monitor. The figure explaining extraction of a gaze area. FIG. 4 is a diagram illustrating a display example of presentation information according to the first embodiment. The figure which shows the example of a display of the presentation information by 2nd Embodiment. The figure which shows the other example of an attention area.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings. However, the scope of the invention is not limited to the illustrated example.

[First Embodiment]
The interpretation report creation support device according to the first embodiment includes medical information (medical image data and electronic medical record information) relating to a case to be diagnosed and input information from a user (user's line-of-sight information and report text). To support the creation of an interpretation report for the case. The user is, for example, an interpreter who interprets medical image data. Hereinafter, a description will be given using an example in which three-dimensional chest X-ray CT image data is handled as an example of medical image data. Obviously, the subject of the interpretation support according to the present invention is not limited to this, and can be applied to, for example, support for interpretation of a two-dimensional image or an ultrasonic tomographic image by simple X-ray imaging. The embodiments described below are merely examples for describing the processing steps of the image interpretation report creation support device.

  FIG. 1 shows an example of a functional configuration of an interpretation report creation support device (hereinafter, support device 100) according to the first embodiment. The support device 100 in the first embodiment is connected to the display unit 200 and the case information terminal 300. The display unit 200 has a monitor 1005 (FIG. 2) and displays various information according to a system or user's instruction. Specific contents displayed on the monitor 1005 will be described later.

  The case information terminal 300 acquires medical information on a case to be diagnosed, for example, medical information (such as three-dimensional medical image data and electronic medical record information) on an abnormal shadow of a lung from a server (not shown). Alternatively, the case information terminal 300 may connect an external storage device, for example, an FDD, an HDD, a CD drive, a DVD drive, an MO drive, a ZIP drive, and the like, and acquire medical information from these drives. Further, the case information terminal 300 acquires accompanying data (clinical data such as a tumor marker value, a report of a past case of the target patient, and the like) related to an abnormal shadow of the lung according to a request of the user. The acquired medical information and accompanying data are transmitted to the support device 100 via a LAN or the like.

  The support device 100 includes a report creation unit 101, an area information acquisition unit 102, a line-of-sight information acquisition unit 104, a report sentence acquisition unit 106, an attention area acquisition unit 108, a description area acquisition unit 110, a determination unit 112, , A presentation information creating unit 114. Although the support device 100 and the case information terminal 300 are shown as separate devices, they may be integrated.

  The report creation unit 101 performs overall processing when creating a report. For example, the report creation unit 101 displays an image interpretation report creation screen (described later with reference to FIG. 5) on the monitor 1005 of the display unit 200. On the interpretation report creation screen, medical information (medical image data) to be interpreted received from the case information terminal 300 is displayed, and the user interprets the displayed medical image data and interprets the image using a keyboard 1007 or the like to be described later. Enter the report. Thus, the user can perform image interpretation by observing the displayed medical information (eg, medical image data).

  The area information acquisition unit 102 acquires information (area information) that associates the coordinate position of medical image data transmitted from the case information terminal 300 with an area. More specifically, correspondence information between a coordinate position of medical image data and a medically divided area is acquired. In the present embodiment, an anatomical structure (eg, heart, liver, upper right lobe, etc.) is used as a medically divided region. The acquired association information is output to the attention area acquisition unit 108 and the description area acquisition unit 110. Note that such area information can be obtained by applying a known segmentation method to medical image data and associating the segmentation result with the coordinate position of the image. As such a segmentation technique, for example, a technique for simultaneous extraction of a plurality of organs in the abdomen described in Non-Patent Document 1 can be mentioned. The region information may be obtained by analyzing the medical image data to be interpreted by the region information obtaining unit 102, or may be obtained by previously storing the region information in the medical image data transmitted from the case information terminal 300. Good. Alternatively, the user may manually specify the region and name corresponding to the anatomical structure.

  The line-of-sight information acquisition unit 104 acquires line-of-sight information indicating which position of the medical image data the user is looking at. In the present embodiment, the line-of-sight information acquisition unit 104 detects a line-of-sight position on the monitor 1005 and, based on the display position of the medical image data on the monitor 1005 and the coordinates of the detected line-of-sight position, displays the corresponding medical image data. Is obtained, and this is used as line-of-sight information. The acquired line-of-sight information is output to the attention area acquisition unit 108. Note that the gaze information acquisition unit 104 can be realized by detecting a gaze position on a monitor using a gaze tracking device 1045 (FIG. 5) such as a video camera or an eye tracker capable of synchronously shooting from a plurality of viewpoints. Specifically, the user's face and eyes are photographed from a plurality of viewpoints, and a predetermined image recognition process is performed on the stereo image obtained by the photographing, so that the coordinate position of the place where the line of sight is turned on the monitor is determined. Can be detected. The line-of-sight tracking device 1045 is disposed, for example, at the end of a monitor (such as directly below the monitor).

  The report sentence obtaining unit 106 obtains a report sentence (text data), which is a result of reading the medical image data by the user, which is input using the interpretation report creation screen. The obtained report sentence is output to the description area obtaining unit 110.

  The attention area acquisition unit 108 is an attention area, which is an area that the user pays attention to at the time of interpretation, among medically divided areas (for example, areas divided based on an anatomical structure) of the displayed medical image data. Get the set of For example, based on the area information acquired by the area information acquisition unit 102 and the line-of-sight information acquired by the line-of-sight information acquisition unit 104, the attention area acquisition unit 108 is a region that is an area of interest on the displayed medical image data. Get the region. Specific contents of the attention area will be described later. The acquired attention area is output to the determination unit 112.

  The description area acquisition unit 110 acquires a set of description areas that are medical areas (for example, areas based on anatomical structures) described in a report sentence corresponding to medical image data. For example, the description area acquisition unit 110 acquires a description area based on the area information acquired by the area information acquisition unit 102 and the report sentence acquired by the report sentence acquisition unit 106. The acquired description area is output to the determination unit 112.

  The determination unit 112 determines consistency between the attention area and the description area based on the set of attention areas acquired by the attention area acquisition unit 108 and the set of description areas acquired by the description area acquisition unit 110. The result of the determination is output to the presentation information creating unit 114. The presentation information creation unit 114 creates presentation information based on the result determined by the determination unit 112. The created presentation information is output to, for example, the display unit 200 and presented by the monitor 1005.

  Note that at least a part of each unit of the support device 100 illustrated in FIG. 1 may be realized as an independent device. Further, each may be realized as software for realizing the function. In the present embodiment, it is assumed that each unit is realized by executing predetermined software by a computer.

  FIG. 2 is a diagram illustrating a basic hardware configuration of a computer device for realizing each function of each unit illustrated in FIG. 1 by executing software. That is, the support device 100 of the present embodiment can be realized by a computer device. The CPU 1001 mainly controls the operation of each component. The main memory 1002 stores a control program executed by the CPU 1001 and provides a work area when the CPU 1001 executes the program. The magnetic disk 1003 stores an operating system (OS), a device drive of a peripheral device, various application software including a program for performing processing described later, and the like.

  The display memory 1004 temporarily stores display data. The monitor 1005 is, for example, a CRT monitor or a liquid crystal monitor, and displays an image, text, or the like based on data from the display memory 1004. A mouse 1006 and a keyboard 1007 are used by the user to perform pointing input and input of characters and the like. The interface 1008 connects the support device 100 to a network or the like. For example, the case information terminal 300 is connected to the support device 100 via the interface 1008. The above components are communicably connected to each other by a common bus 1009.

  Next, a process performed by the support apparatus 100 according to the present embodiment having the above-described configuration (a process for supporting creation of a report at the time of image interpretation) will be described with reference to a flowchart of FIG. In the present embodiment, this is realized by the CPU 1001 executing a program for realizing the function of each unit stored in the main memory 1002. In the following description, the position (pixel) on the screen of the monitor 1005 is represented by the MN coordinate system, and the position (voxel) in the medical image data is represented by the XYZ coordinate system. In the following description, it is assumed that the line-of-sight information is detected by the line-of-sight information obtaining unit 104 at a predetermined sampling rate (for example, 0.1 second). Then, the line-of-sight information at a certain point in time is set to Ot (otx, oty, otz) (t = 1, 2,...).

  In step S300, the report creation unit 101 displays an image interpretation report creation screen on the monitor 1005. As shown in FIG. 5, the image interpretation report creation screen includes a tomographic image display area 501 for displaying medical image data to be interpreted, a report sentence input area 502 for receiving a report (text data) input by a user, and an image interpretation end. Button 503 is included. In the tomographic image display area 501, for example, an axial tomographic image of medical image data is displayed. The two screens shown in FIG. 5 may be simultaneously displayed on one monitor screen, may be switched to one monitor screen and displayed, or may be separately displayed using two monitors. May be done. In the present embodiment, it is assumed that the images are displayed separately using two monitors, and that the line-of-sight information described later is obtained on the monitor displaying the medical image data.

  Next, in step S301, the area information acquisition unit 102 transmits the medical image data transmitted from the case information terminal 300, that is, the correspondence information (area) between the coordinate position of the medical image data to be interpreted and the anatomical structure of the subject. Information). This region information is acquired, for example, as a correspondence table in which a different label is assigned to each anatomical structure and a label is linked to each voxel. Alternatively, it may be obtained in the form of a conditional expression. FIGS. 4A and 4B show respective examples. FIG. 4A shows region information in which a label indicating an anatomical structure is linked to each voxel. Further, in FIG. 4B, region information defining the voxel coordinate range is shown in the label of each anatomical structure. Of course, the area information may be configured in other formats. Hereinafter, a case where the area information is acquired in the form of the correspondence table as shown in FIG.

  Next, in step S302, the line-of-sight information acquisition unit 104 acquires line-of-sight information indicating which position of the medical image data the user is looking at. The report sentence acquisition unit 106 acquires a report sentence in which the user interprets medical image data and summarizes the results. In the present embodiment, under the control of the report creation unit 101, the user can input a report sentence using the keyboard 1007 in the report sentence input area 502 shown in FIG.

  Here, the acquisition of the line of sight information by the line of sight information acquisition unit 104 will be described. FIG. 5 shows an example in which an image interpretation report creation screen and medical image data to be interpreted are displayed on the monitor 1005. In this example, an axial tomographic image composed of medical image data is displayed. The axial tomographic image is a tomographic image perpendicular to the top and bottom of the body (head and tail, body axis). In the coordinate system of the present embodiment, the axial tomographic image is a tomographic image on the XY plane at a certain Z coordinate (a certain position on the body axis). expressed. A gaze tracking device 1045 is provided below the monitor 1005, and detects which position (display pixel) of the monitor 1005 the user is looking at as coordinates in the MN coordinate system.

  Now, for the sake of simplicity, consider a case where the size of the display pixel of the monitor and the size of the pixel of the axial tomographic image (XY plane) completely match. Further, the correspondence between the coordinates (m, n) indicating the display pixel of the monitor and the coordinates (x, y) indicating the pixel of the axial tomographic image is represented by x = m-a, y = n-b. In other words, consider a case where the value of one voxel of medical image data is displayed on one pixel of a monitor. Focusing on one point (m, n) of the monitor in this state, the values of x (= m−a) and y (= n−b) in the medical image data are determined. Furthermore, since the position of z is determined according to the displayed axial tomographic image, when the coordinates (m, n) on the monitor are determined, the coordinates (x, y, z) on the medical image data are uniquely determined. By utilizing this, it is possible to obtain the coordinates of the position of the line of sight on the medical image data (that is, the line of sight information) from the coordinates of the line of sight on the monitor 1005 detected by the line of sight tracking device 1045.

  If the size of the display pixel of the monitor and the size of the pixel of the axial tomographic image (XY plane) do not match, the coordinates of the position on the medical image data can be calculated using the ratio of the size of the pixel. . Further, in consideration of the visual field, the coordinate position on the medical image data indicated by the pixels around the pixel of the axial tomographic image corresponding to the display pixel of the monitor 1005 may be obtained as the coordinates of the position to which the line of sight is directed. That is, coordinate positions on a plurality of medical image data may be acquired for one acquired line-of-sight position.

  The user presses the image reading end button 503 to notify the support apparatus 100 of the end of the input of the report sentence. When the input end of the report sentence is notified, the process proceeds from step S302 to step S303. If it is determined in step S303 that the image interpretation end button 503 has not been pressed, the process returns to step S302, and the processing of acquiring the line-of-sight information and inputting the report sentence is continued.

In step S304, the attention area acquisition unit 108 acquires a set of attention areas based on the line-of-sight position detected during image interpretation. In other words, based on the area information obtained in step S300 and the line-of-sight information obtained in step S301, the attention area obtaining unit 108 selects an area of the medically divided area on the medical image data that is noted by the user. Is obtained. As a method for determining whether or not the area A is the attention area, for example,
(1) In the area A displayed in the tomographic image display area 501 at the time of interpretation, the number NoA of the number of pixels (or voxels) at which the line-of-sight position is detected is acquired,
(2) Obtain the total number of pixels (the total number of voxels) NaA in the displayed area A, and determine the ratio (NoA / NaA) of the number of pixels whose gaze position is detected to the total number of pixels included in the area A. A method of setting a region exceeding the value as a region of interest can be cited. Note that a method may be considered in which a region in which the line-of-sight position is present for a predetermined time period or longer is set as a region of interest. It will be judged. Therefore, it is desirable to combine the determination using the time when the line-of-sight position exists with the determination using the ratio (NoA / NaA) described above. For example, when NoA / NaA exceeds a predetermined value and the total time during which the line-of-sight position exists in the area A exceeds a predetermined time, it is preferable to determine the area A as the attention area.

  In the present embodiment, the above procedures (1) and (2) are used. More specifically, a voxel observed even once is regarded as an observed region, and the number of voxels observed for each label to which an anatomical structure is assigned is counted. Then, based on the number of voxels observed for each label and the total number of voxels having the label, it is determined whether the label (that is, the anatomical structure) has been watched. For example, when more than a certain percentage of the total number of voxels having a certain label is observed, it is determined that the anatomical structure corresponding to the label has been watched. Therefore, in the following, the attention area determined in the procedure of (1) and (2) is also referred to as the attention area.

  This will be described with reference to FIG. In FIG. 6, reference numeral 601 denotes an axial tomographic image at a certain Z position (Z = 6) with area information and visual line information. Reference numeral 602 denotes the area information acquired by the area information acquisition unit 102 for the axial tomographic image 601. Reference numeral 603 denotes gaze information acquired by the gaze information acquisition unit 104.

  The number written on each pixel (x, y) of the axial tomographic image 601 indicates a label assigned to each voxel (x, y, 6) indicated by the area information 602. For example, pixel (3, 7) (ie, voxel (3, 7, 6)) has a label of “1”, indicating that it is the upper right leaf. Pixels painted in gray indicate pixels indicated by the line-of-sight information 603 (that is, observed voxels). In the line-of-sight information 603, the line-of-sight position (x, y, z) (coordinates of voxels) at time Ot is recorded in chronological order (in the order of Ot = O1, O2, O3...). In the example of this tomographic image, the number of observed voxels is “0 (outside the body surface)”, 14 “1 (upper right lobe)”, 9 “2 (right middle lobe)”, and “3 (right). 10), 6 "4 (upper left)", 19 "5 (lower left)", and 23 "9 (other)". By applying this to the entire medical image data, the number of observed voxels for each label in the entire medical image data can be counted.

  In the present embodiment, the observed number of voxels of label k is Nok, the total number of voxels belonging to label k is Nak, and when satisfying Nok / Nak> 0.8, it is assumed that the anatomical structure indicated by label k is watched. . In the following description, it is assumed that the “upper right lobe” and the “lower right lobe” are observed as the gaze areas.

  In step S305, the description area acquisition unit 110 acquires a description area that is an area described in a report sentence based on the area information acquired in step S301 and the report sentence acquired in step S303. In the present embodiment, it is assumed that the anatomical structure name indicated by each label of the area information is used as a keyword, and a description area is acquired using keyword matching for a report sentence. The matching may be performed by expanding the keyword using a synonym or ontology. For example, in the example of the report sentence illustrated in FIG. 5 (the example illustrated in the report sentence input area 502), “upper left” and “lower right” are acquired as the description area.

  In step S306, the determination unit 112 determines the consistency between the attention area and the description area based on the attention area (gaze area) acquired in step S303 and the description area acquired in step S305. In the present embodiment, it is determined whether there is information included in the attention area (gaze area) and not included in the description area. That is, it is determined whether or not there is a description omission that is not described in the report sentence despite watching. In the above example, since the gaze area “upper right lobe” is not included in the description area, it is determined that there is a description omission.

  In step S307, the presentation information creation unit 114 creates presentation information based on the result determined in step S306. In the present embodiment, a message box is created as presentation information. In the example described above, since “upper right lobe” is not included in the description area, a message box notifying the fact is created. Then, in step S308, the presentation information creation unit 114 displays the created message box on the monitor 1005.

  FIG. 7 shows an example of presentation information displayed on the monitor 1005 in the present embodiment. The presentation information is displayed in the form of a message box 704. Here, when the “Yes” button in the message box 704 is pressed, the report sentence can be corrected (re-input). That is, in step S309, it is determined that the support for correction has been input, and the process returns to step S302, where the report creation unit 101 can create (edit) a report sentence. On the other hand, if the user presses the "No" button, the process proceeds to step S310, the report creation unit 101 saves the report sentence on the magnetic disk 1003 or the like, and ends the report creation support process.

  If the consistency is passed in step S306 (if it is determined that there is no omission in description), for example, a message box for displaying the fact is generated and displayed, and whether to continue to create a report sentence. It may be inquired about whether or not it is not. Alternatively, when it is determined that there is no omission of the description, the report sentence may be saved and the process may be ended similarly to the case where the above-mentioned “No” button is pressed.

  As described above, according to the present embodiment, it is possible to alert the doctor to omission of description in an area that is not noted in the report sentence despite being watched at the time of interpretation. Therefore, the doctor can create a report sentence with no omission. In particular, it is expected that after reading the medical image data all at once, it will work effectively in the interpretation style in which report texts are collectively described. Thereby, the doctor can create an appropriate interpretation report.

[Second embodiment]
In the first embodiment, the description has been given of a configuration in which “missing description” is detected by determining the consistency between the set of attention areas and the set of description areas, and a warning is issued. The interpretation report creation support device according to the second embodiment alerts the doctor about an area that is not watched at the time of interpretation even though the doctor has described the report text. That is, in the second embodiment, a configuration in which a warning is given for omission of confirmation is described.

  The configuration of the support apparatus 100 for creating an interpretation report according to the second embodiment is the same as that of the first embodiment (FIG. 1). The basic configuration of a computer that realizes the interpretation report creation support processing of the support apparatus 100 by executing software is also the same as that of the first embodiment (FIG. 2). A flowchart for explaining the entire processing performed by the support apparatus 100 is the same as that in FIG. However, a part of the processing is different from the first embodiment. Hereinafter, the entire processing performed by the support device 100 according to the second embodiment will be described with reference to the flowchart in FIG.

  The processing from step S300 to step S305 and the processing from step S307 to step S310 are the same as the processing in the first embodiment. In step S306, the determination unit 112 determines consistency between the attention area and the description area based on the set of attention areas acquired in step S304 and the set of description areas acquired in step S305.

  In the present embodiment, it is determined whether there is information included in the description area and not included in the attention area. Note that, similarly to the first embodiment, the gaze area is used. That is, it is determined whether there is any omission in confirmation that the doctor did not pay attention during interpretation even though the description was made. For example, as in the example described in the first embodiment, consider the case where the gaze area is “upper right lobe” and “lower right lobe” and the description areas are “upper left lobe” and “lower right lobe”. In this case, since the description area “upper left lobe” is not included in the gaze area, it is determined that there is a confirmation omission.

  In the present embodiment, a message box including a schema image is created as presentation information presented in step S308. Specifically, it is assumed that a message box including an image in which the anatomical structure corresponding to the unchecked region is emphasized is created. In the above-described example, since “upper left” is not included in the gaze area, a message box including a schema image in which “upper left” is emphasized is created. Then, the created message box is displayed on the monitor 1005.

  FIG. 8 shows an example of presentation information displayed on the monitor 1005 in the present embodiment. The presentation information is displayed in the form of a message box 805. Here, when the “Yes” button in the message box 805 is pressed, the image can be read again (that is, the process returns to step S301). On the other hand, if “No” is pressed, the support device 100 saves the report sentence and ends.

  If it is determined in step S306 that there is no omission in confirmation, the report sentence is saved and the process ends, similarly to the case where "NO" is pressed as described above. If it is determined in step S306 that there is no omission, a message box may be generated and displayed to indicate that, and an inquiry may be made as to whether or not to continue creating a report sentence.

  As described above, according to the second embodiment, it is possible to alert the doctor to omission of confirmation in an area that is not watched at the time of interpretation even though the doctor describes the report sentence. Therefore, it is possible to prevent a doctor from describing a region that is not watched.

  Further, as another effect, it can be expected that in the interpretation of a follow-up case in which a past interpretation report exists, the effect is effectively exerted when a report sentence is created by diverting the past interpretation report. In other words, if a previous report is diverted and a description that does not fit the current state is left as it is, a warning of omission of confirmation will be issued, so that the doctor will notice the existence of unnecessary description in the report for the current state. Can be expected. With the above configuration, the doctor can create an appropriate interpretation report.

(Modification 1)
In the first and second embodiments, an example has been described in which, in step S <b> 304, a gaze area in which line-of-sight information is obtained in an area indicated by an axial tomographic image displayed in the tomographic image display area 501 is used as the attention area. . However, it is not always necessary to use whether or not the line-of-sight position has been detected as a method for determining whether or not an area has received attention. In the first modification, the region of interest is determined by treating the displayed region of the axial tomographic image in the same manner as the region where the line-of-sight position is detected. That is, as described in the first embodiment, the number of pixels (voxels) of the area A in which the line-of-sight position has been detected is the number of pixels of the part of the area A displayed by the monitor 1005 as the number NoA.

  FIG. 9 is a diagram illustrating an example in which a region displayed on a monitor is set as a region of interest. FIG. 9A shows an example of a case in which the interpretation of the axial tomographic image 906 is performed with the general medical image viewer function being translated. In this case, only the portion displayed in the tomographic image display area 501 is set as the attention area. In this example, a hatched portion of the axial tomographic image 906 is a portion of the medical image data displayed at the time of interpretation. Therefore, for each of the medical regions that at least partially belong to the hatched portion of the medical image data (axial tomographic image 906), the ratio between the number of pixels displayed at the time of interpretation and the total number of pixels in that region is determined. It is determined whether or not the region is a region of interest. FIG. 9B is an example showing a case in which the interpretation is performed in a state where the axial tomographic image 906 is enlarged and displayed by a general medical image viewer function. In this example, a hatched portion of the axial tomographic image 906 is a region displayed during image reading. As described above, an area where at least a part of a medically divided area of the medical image data (axial tomographic image 906) belongs to a hatched area is a target to be determined as to whether or not it is an attention area. . In a state where the axial tomographic image 906 is displayed in a reduced size (that is, when the entire axial tomographic image 906 is contained in the tomographic image display area 501), naturally, the entire area indicated by the axial tomographic image 906 becomes the attention area.

(Modification 2)
In the first and second embodiments, in step S304, a region of interest is acquired as a specific anatomical structure. Based on the description region indicating the anatomical structure acquired in step S305, consistency is determined in step S306. Had been determined. However, the determination of consistency is not limited to this method.

  For example, in step S304, the attention area acquisition unit 108 acquires all of the observed voxels (the voxels where the line of sight exists) as attention areas, and uses the area information acquired in step S301 to generate a voxel group for each anatomical structure. Convert to Then, in step S306, the determination unit 112 may determine the consistency based on how many voxels acquired as the attention area are included in the voxel group for each anatomical structure indicated by the description area.

(Modification 3)
In the first and second embodiments, the omission of description is determined separately by the method of the first embodiment, and the omission of confirmation is determined separately by the method of the second embodiment. However, it is not always necessary to perform the determination separately, and the determination of omission of description and the determination of omission of confirmation may be performed simultaneously. That is, in the determination of consistency, both the omission of description described in the first embodiment and the omission of confirmation described in the second embodiment may be determined. In such a case, the method of acquiring the set of attention areas may be different between the detection of the omission of description and the detection of the omission of confirmation. For example, a set of attention areas for detecting omission of description may be a set of areas where a line of sight has existed even once, and a set of attention areas for detecting omission of confirmation may be a set of attention areas.

  Further, in the first and second embodiments, the fact that the region of interest is displayed on the image interpretation report creation screen and that the line of sight is present is used, but the present invention is not limited to these. For example, a touch input unit such as a touch panel may be provided on the monitor 1005, and the user may touch and input a position of interest during interpretation. In this case, the attention area acquisition unit 108 acquires an attention area based on the touch input position.

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

100: interpretation report creation support device, 102: area information acquisition unit, 104: gaze information acquisition unit, 106: report sentence acquisition unit, 108: attention area acquisition unit, 110: description area acquisition unit, 112: determination unit, 114: Presentation unit, 200: display unit

Claims (12)

A support device for supporting creation of an interpretation report,
Display means for displaying medical image data to be interpreted,
A first acquisition means for acquiring a target area that is determined as the target area during the interpretation in the medical image data displayed by the display means,
A second acquisition means for acquiring the description area is a medical area described in the report statement corresponding to the medical image data,
Determining means for determining consistency of the description area and the target area,
And a presentation unit for presenting information based on a result of the determination by the determination unit. Further comprising detection means for detecting the line of sight of the radiogram interpreter,
It said first acquisition means support device according to claim 1, characterized in that obtaining the target area based on the detected line-of-sight position during interpretation.   The first acquisition unit is configured to set, in a medically divided region of the medical image data, a region in which the ratio of the number of pixels whose line-of-sight positions are detected to the total number of pixels included in the region exceeds a predetermined value. The support device according to claim 2, wherein the support device acquires the region of interest.   The said 1st acquisition means acquires the area | region in which the line-of-sight position was detected in the area | region exceeds predetermined time among the area | regions divided | segmented of the said medical image data as said attention area | region. 4. The support device according to 2 or 3.   The first acquisition unit is a region in which at least a part of the medically divided region of the medical image data is displayed in a display region provided in the display unit at the time of interpretation, and The support apparatus according to claim 1, wherein an area in which a ratio of the number of displayed pixels to a total number of included pixels exceeds a predetermined value is acquired as the attention area. The display means has a touch input means,
It said first acquisition means support device according to claim 1, characterized in that obtaining the target area based on the touch input position to the display means for displaying the medical image data. The medically divided region of the medical image data is a region that is divided based on an anatomical structure,
7. The support apparatus according to claim 1, wherein the second acquisition unit acquires an anatomical structure name described in the report sentence as the description area. 8. Said second acquisition means, as a keyword anatomy names of all the areas obtained by dividing based on the anatomy of the medical image data, the description area by the keyword matching from the report text support device according to claim 7, characterized in that to obtain the. The determination unit, when the included noting area is not included in the description area regions are present, one of the claims 1 to 8, characterized in that determining that there is omissions the report text The support device according to claim 1. Said determining means, when the area not included in the target area is included in the description area is present, according to claim 1 to 9, characterized in that determining that there is a check omission for the medical image data The support device according to claim 1. A control method of a support device that supports creation of an image interpretation report,
A display step of displaying medical image data to be interpreted on display means,
Among medically divided region of the medical image data displayed on said display means, a first acquisition step for acquiring target area that is determined to be focused on during image interpretation area,
A second acquisition step of acquiring a description area is a medical area described in the report statement corresponding to the medical image data,
A determination step of determining consistency of the description area and the target area,
A presentation step of presenting information based on a result of the determination in the determination step.   A program for causing a computer to execute each step of the control method of the support device according to claim 11.

Images