JP4755863B2 - Interpretation support device, interpretation support method, and program thereof - Google Patents

Description

  The present invention relates to an interpretation report creation support apparatus that supports creation of an image interpretation report.

  Recently, it is generated based on three-dimensional image data configured as an aggregate of a plurality of slice image data acquired by CT (Computed Tomography), MRI (Magnetic Resonance Imaging), etc. A cross-sectional image or a three-dimensional display image is used to discover a lesion and observe the state of the lesion to diagnose the presence or absence of a disease and the progress of the disease. Compared with the conventional two-dimensional radiographic image data, since it has more information about the subject, a highly accurate diagnosis is possible.

  In addition, image diagnosis performed by observing a photographed image is usually performed in the following procedure. First, when an examination order is issued from an examination requesting department (for example, internal medicine) that requests image diagnosis to the radiology department, imaging is performed according to the examination order. This test order includes the patient ID number, patient name, date of birth, gender, name of the department requesting the examination, name of the doctor requesting the examination, examination modality (CR equipment, CT equipment, MRI equipment, etc.), examination site, examination purpose In addition, clinical information is included, and an inspection engineer performs imaging according to an inspection order. Next, an image interpretation doctor interprets the captured image and creates an image interpretation report according to the image interpretation result. This interpretation report is sent to the doctor in charge and used to determine the treatment policy of the doctor in charge.

However, it takes a lot of work and time to manually input such an interpretation report. In view of this, a method has been proposed for reducing the burden by using the fixed portions of CAD and interpretation reports (for example, Patent Document 1).
JP 7-31591 A

  As described above, the diagnosis result is also input as an interpretation report for the three-dimensional image data. However, once a tomogram is read and a report on abnormal shadows is created, moving to the next tomogram and interpreting it may make it difficult to determine whether the lesion has already been reported in another slice. It was bad.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide an image interpretation support apparatus, an image interpretation support method, and a program for improving diagnostic efficiency during image interpretation.

The image interpretation support apparatus of the present invention includes a three-dimensional medical image storage means for storing a three-dimensional medical image obtained by photographing a subject,
Cross-sectional display means for displaying a plurality of cross-sectional images of the three-dimensional medical image;
Instruction means for indicating a region of interest existing on any one of the plurality of cross-sectional images displayed by the cross-section display means;
Position recording means for recording the position of the region of interest instructed by the instruction means;
An attention area position display means for recognizing and displaying the position of the attention area on a cross-sectional image other than the cross-sectional image indicating the attention area is provided.

Further, the interpretation support method of the present invention includes a cross-sectional display step for displaying a plurality of cross-sectional images of a three-dimensional medical image and a three-dimensional medical image obtained by photographing a subject,
An instruction step for indicating a region of interest existing on any one of the plurality of cross-sectional images displayed by the cross-sectional display step;
A position recording step for recording the position of the region of interest instructed by the instructing step;
An attention area position displaying step of recognizing and displaying the position of the attention area on a cross-sectional image other than the cross-sectional image indicating the attention area is provided.

Further, the program of the present invention provides a computer,
Cross-sectional display means for displaying a plurality of cross-sectional images of a three-dimensional medical image obtained by photographing a subject;
Instruction means for indicating a region of interest existing on any one of the plurality of cross-sectional images displayed by the cross-section display means;
Position recording means for recording the position of the region of interest instructed by the instruction means;
It functions as an attention area position display means for recognizing and displaying the position of the attention area on a cross-sectional image other than the cross-sectional image indicating the attention area.

  The “three-dimensional medical image” is volume data composed of a large number of voxel data, and refers to an image that retains density value information for each voxel data and represents anatomical structures such as bones and organs. For example, it is an image taken by a tomography apparatus such as CT or MRI.

  Further, the “cross-sectional image of a three-dimensional medical image” is not limited to a cross-sectional image created from a three-dimensional medical image captured with one modality, but may be a cross-sectional image of a three-dimensional medical image captured with a different modality. . For example, it may be a cross-sectional image of both CT and MRI, or a tomographic image obtained by superimposing CT and MRI cross-sectional images.

Further, diagnostic information input means for inputting diagnostic information of the instructed attention area to the interpretation support device;
Determination means for determining whether a cross-sectional image other than the cross-sectional image indicating the region of interest based on the diagnostic information is a cross-sectional image related to the region of interest;
The region-of-interest position display means may display the position of the region of interest in a recognizable manner on the cross-sectional image determined by the determining unit as a cross-sectional image related to the region of interest.

Furthermore, the diagnostic information includes the size of the region of interest,
The determination unit may determine that a cross-sectional image within a predetermined range from a cross-sectional image indicating the region of interest corresponding to the size is a cross-sectional image related to the region of interest.

  The region of interest may be a candidate region for a lesion.

  According to the present invention, by indicating the attention area existing on the displayed cross-sectional image, and displaying on the cross-sectional image other than the cross-sectional image indicating the attention area so that the position of the attention area can be recognized, It becomes easy to determine whether or not the shadow has been checked once by the doctor.

  Further, by inputting diagnostic information of the indicated attention area, and determining whether a cross-sectional image other than the cross-sectional image indicating the attention area based on the diagnostic information is a cross-sectional image related to the attention area, Since the position of the attention area can be displayed only on the relevant cross section, it is easier to determine whether or not the attention area has already been checked.

  If the size of the attention area is included in the diagnostic information, it is possible to determine that the cross-sectional image within the range corresponding to the size is a cross-sectional image related to the attention area. It becomes possible to accurately determine the area.

  Furthermore, if the attention area is a candidate area for a lesion, the disease can be accurately determined.

  Hereinafter, an interpretation support apparatus of the present invention will be described with reference to the drawings. As shown in FIG. 1, an interpretation support apparatus 1 of the present invention includes a three-dimensional medical image storage unit 10 that stores a three-dimensional medical image 100 obtained by photographing a subject, and a plurality of cross-sectional images of the three-dimensional medical image. Cross-section display means 20 for displaying, instruction means 30 for indicating the attention area existing on any one of the displayed cross-sectional images, and position recording means 40 for recording the position of the indicated attention area. And attention area position display means 50 for displaying the position of the designated attention area on a cross-sectional image other than the cross-sectional image instructing the attention area, and diagnostic information for inputting diagnostic information of the designated attention area Input means 60 and determination means 70 for determining whether or not a cross-sectional image other than the cross-sectional image indicating the attention area based on the diagnostic information is a cross-sectional image related to the attention area.

  The three-dimensional medical image storage means 10 is a large-capacity storage device such as an image server. A CT image or MRI image obtained by imaging a subject to be examined using a CT imaging device, an MRI imaging device, or the like is stored in the image server as a three-dimensional medical image 100, and if necessary, the three-dimensional medical image 100 of each subject. Is stored for retrieval.

  The cross-sectional display means 20 displays the cross-sectional image of the three-dimensional medical image 100 stored in the three-dimensional medical image storage means 10 on the display device. A plurality of cross-sectional images having different depths in the same direction are usually generated and displayed on the screen on the display device. When displaying, a plurality of generated cross-sectional images are displayed side by side, or cross-sectional images having different depths are switched and displayed. Further, not only a cross-sectional image of a three-dimensional medical image photographed with one modality but also a cross-sectional image of a three-dimensional medical image photographed with a plurality of modalities may be displayed on the surface image. Good. For example, when a three-dimensional medical image photographed by an MRI photographing apparatus and a CT photographing apparatus is stored in the three-dimensional medical image storage means 10, both sectional images are displayed side by side, or both sectional images are overlapped. You may make it display.

  In this way, since images taken with different modalities have different forms appearing on the images, it is possible to arrange the tomographic images taken with a plurality of modalities or to observe and superimpose the images displayed in a superimposed manner, thereby diagnosing the lesion. , Observation of the state of the lesion, diagnosis of progress, etc. can be performed accurately. For example, since a bone is imaged in the CT image but no bone is imaged in the MRI image, the position and state of the disease can be accurately determined by observing both images.

  The instruction unit 30 instructs a region of interest existing on any one of the plurality of cross-sectional images displayed on the display device by the cross-sectional display unit 20 (region surrounded by a circle in FIG. 2). Specifically, a shadow or the like on the cross-sectional image is marked as a region of interest ROI using a pointing device such as a mouse.

  The position recording means 40 records the position of the attention area ROI instructed by the interpreter with the instruction means 30. Specifically, the cross-sectional image's cross-sectional direction and depth position are recorded so that it can be seen which cross-sectional image is a plurality of cross-sectional images. Record the position of.

  The attention area position display means 50 displays a mark so that the position corresponding to the attention area ROI can be seen on the cross-sectional image other than the cross-sectional image instructing the attention area ROI.

  The diagnosis information input means 60 inputs the contents diagnosed by observing the attention area ROI instructed by the instruction means 30 as diagnosis information. The diagnosis information is input as a report for each abnormal shadow such as a found cancer and stored in association with the region of interest ROI designated by the instruction means 30. The report is usually input as document data from the keyboard (see REPORT in FIG. 3). Findings such as the size and characteristics of abnormal shadows are entered in the report.

  The determination unit 70 determines whether or not a cross-sectional image other than the cross-sectional image indicating the region of interest ROI is a cross-sectional image related to the region of interest ROI based on the diagnosis information (report). Since the report is stored as document data, it is preferable that the report has a function of interpreting the input contents by lexical analysis of a character string in the document data. For example, when the content “There is an abnormal shadow candidate having a radius of about 5 mm” is recorded, there is a high possibility that the same abnormal shadow appears in the cross-sectional image of about 5 mm above and below from the cross-sectional image marked with the region of interest ROI. From the cross-sectional image marked with the attention area ROI, a cross-sectional image of about 5 mm above and below is determined to be a cross-sectional image related to the attention area ROI.

  Here, the operation of the image interpretation support apparatus 1 will be specifically described with reference to the flowchart of FIG. 4 when an interpreter such as a doctor interprets an abnormal shadow such as cancer by observing a chest CT image.

  First, the doctor observes the tomographic image in the Axial direction as shown in FIG. 2 while sequentially displaying the tomographic image as shown in FIG. 2 with the cross-section display unit 20 of the image interpretation support apparatus 1 (S100). When the doctor finds a shadow that appears to be an abnormal shadow on a certain cross-sectional image, the pointing means 30 marks the shadow as a region of interest ROI using a mouse or the like (◯ in FIG. 2) (S104). The position of the marked region of interest ROI is stored in the memory of the image interpretation support apparatus 1 by the position recording means 40 (S105). Further, the doctor inputs the findings of the region of interest ROI as a report using the diagnostic information input means 60 (S106). This report is stored in association with the position of the attention area ROI (S107).

  Further, the doctor displays the next cross-sectional image on the screen by the cross-section display means 20 (S100). At this time, when there is a record of the position of the attention area ROI (S101), the report of the attention area ROI recorded by the determination means 70 is analyzed, and the sectional screen currently displayed on the screen is noticed. It is determined whether or not the cross-sectional image is related to the region ROI (S102). For example, if a cross-sectional image is created with a slice interval of about 2 mm, and a report of this region of interest ROI has been entered as “There is an abnormal shadow candidate with a radius of about 5 mm and the medical condition is unknown at this stage”, As shown in FIG. 5, there is a high possibility that shadows related to the shadow marked as the attention area ROI appear in the upper and lower two cross-sectional images marked with the attention area ROI. Therefore, as shown in FIG. 3, the abnormal shadow under 2 mm is shown on the two upper and lower cross-sectional images of the specified cross-sectional image, as shown in FIG. Is already reported ", and the previous report is also displayed (S103). When an abnormal shadow exists at a position different from the position where the attention area ROI is marked on the sectional image next to the sectional image where the attention area ROI is marked, it is understood that the abnormal shadow is different from the marked shadow.

  The above operation is repeated to observe the tomographic image of the three-dimensional medical image.

  In the above description, the case where the fact that the region of interest has already been marked is displayed only on the cross-sectional image determined to be related to the marked region of interest ROI has been described, but it may be displayed on all cross-sectional images. Good.

  Alternatively, not only the cross-sectional image of the same modality but also the position similar to the above may be displayed at the position corresponding to the cross-sectional image of another modality or the position for the image obtained by superimposing the cross-sectional images of different modalities. Good.

  In the above description, the case where the cross-sectional images are displayed while being switched has been described. However, even when the cross-sectional images are displayed side by side, the location of the abnormal shadow instructed on one cross-sectional image can be found on the other cross-sectional images. You may make it display as follows.

  As described above in detail, since a doctor observes a cross-sectional image and can confirm whether it is a shadow once checked on other cross-sectional images, many abnormal shadows like multiple cancers. Even when the same shadow is observed, it is easy to distinguish whether the same shadow is observed or another shadow is observed.

  In addition, the shadow is judged by displaying the cross-sectional images taken with different modalities or the same position on the image where the cross-sectional images taken with different modalities are overlapped. Detailed information can be obtained.

  In addition, the region to be marked as the attention region is usually where an abnormal shadow such as a cancer appears, but the part where the normal structure appears is marked as the attention region on a cross-sectional image, The position of the attention area may be confirmed on the cross-sectional image.

  Moreover, it can be made to operate | move as an image interpretation assistance apparatus by installing in a computer using the medium which recorded the program which functions each said means on a computer.

Configuration diagram showing the configuration of the interpretation support device An example of a cross-sectional image with a marked region of interest An example of displaying a region of interest on a cross-sectional image other than the cross-sectional image where the region of interest is marked Flow chart for explaining the operation of the image interpretation support device The figure for demonstrating the determination of the cross-sectional image relevant to an attention area

Explanation of symbols

1 Interpretation support device
10 Three-dimensional medical image storage means
20 Section display means
30 Instruction means
40 Position recording means
50 Attention area position display means
60 Diagnostic information input means
70 Judgment means
100 3D medical images

Claims (6)

Three-dimensional medical image storage means for storing a three-dimensional medical image obtained by photographing a subject;
Cross-sectional display means for displaying a plurality of cross-sectional images of the three-dimensional medical image;
Instruction input receiving means for receiving an instruction input of a candidate region of a lesion existing on one of the plurality of cross-sectional images displayed by the cross-sectional display means;
Position recording means for recording the position of the candidate area input by the instruction input receiving means;
Diagnostic information input receiving means for receiving an input of diagnostic information of the candidate area including the size of the candidate area inputted by the instruction;
Corresponding to the size, a cross-sectional image other than a cross-sectional image instructing and inputting the candidate region, and a cross-sectional image within a range where the candidate region may exist is a cross-sectional image related to the candidate region. Determining means for determining
The position of the candidate area is displayed in a recognizable manner on the cross-sectional image determined to be a cross-sectional image related to the candidate area by the determining means, and the diagnostic information of the candidate area is already input. An interpretation support apparatus comprising a display means for displaying.   The image interpretation support apparatus according to claim 1, wherein the display unit further displays the diagnosis information.   The image interpretation support apparatus according to claim 1, wherein the display unit further displays which position of the plurality of cross-sectional images the diagnostic information has been input. The diagnostic information is input as document data consisting of character strings,
The image interpretation support apparatus according to claim 1, wherein the determination unit obtains the size of the candidate area by performing lexical analysis on a character string of document data of the diagnostic information. Three-dimensional medical image storage means for storing a three-dimensional medical image obtained by photographing a subject;
Cross-section display means for displaying a cross-sectional image;
Instruction input receiving means for receiving an instruction input of a candidate area of a lesion existing on the cross-sectional image;
Position recording means for recording the position of the candidate area;
Diagnostic information input receiving means for receiving input of diagnostic information of the candidate area;
Determining means for determining a cross-sectional image associated with the candidate region;
In the operation method of the image interpretation support apparatus comprising display means for displaying that the diagnostic information of the candidate area has already been input,
A cross-section display step of displaying a plurality of cross- sectional images of a three-dimensional medical image obtained by photographing a subject by the cross- section display means ;
An instruction input receiving step of receiving an instruction input of a candidate area of a lesion portion existing on one of the plurality of cross-sectional images displayed by the cross-sectional display step by the instruction input receiving unit ;
A position recording step of recording the position of the candidate area input by the instruction input receiving step by the position recording means ;
A diagnostic information input receiving step for receiving input of diagnostic information of the candidate area including the size of the candidate area inputted by the instruction by the diagnostic information input receiving means ;
Corresponding to the size, a cross-sectional image other than a cross-sectional image instructing and inputting the candidate region, and a cross-sectional image within a range where the candidate region may exist is a cross-sectional image related to the candidate region. And a determination step of determining by the determination means ,
On the cross-sectional image determined to be a cross-sectional image related to the candidate area by the determination step, the position of the candidate area is displayed in a recognizable manner, and diagnostic information of the candidate area has already been input. An operation method of an image interpretation support apparatus for executing a display step of displaying by the display means . Computer
Cross-sectional display means for displaying a plurality of cross-sectional images of a three-dimensional medical image obtained by photographing a subject;
Instruction input receiving means for receiving an instruction input of a candidate region of a lesion existing on one of the plurality of cross-sectional images displayed by the cross-sectional display means;
Position recording means for recording the position of the candidate area input by the instruction input receiving means;
Diagnostic information input receiving means for receiving an input of diagnostic information of the candidate area including the size of the candidate area inputted by the instruction;
Corresponding to the size, a cross-sectional image other than a cross-sectional image instructing and inputting the candidate region, and a cross-sectional image within a range where the candidate region may exist is a cross-sectional image related to the candidate region. Determining means for determining
The position of the candidate area is displayed in a recognizable manner on the cross-sectional image determined to be a cross-sectional image related to the candidate area by the determining means, and the diagnostic information of the candidate area is already input. A program that functions as a display means for displaying.

Images