JP2012230528A - Image processor and control method of image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique which reduces a burden on a user when the user refers to an image to which attention is paid at the past reference time.SOLUTION: A display device of the present invention comprises: an acquisition means for acquiring reference history information which indicates past reference states of a plurality of tomographic images whose imaging positions are different from each other, and includes at least information indicating a cumulative reference count or a cumulative reference time of each tomographic image; an extraction means for extracting, using the reference history information, a tomographic image whose cumulative reference count or cumulative reference time is more than a predetermined threshold from among the plurality of tomographic images as a tomographic image to which attention is paid at the past reference time; and a classification means for, when a plurality of tomographic images are extracted by the extraction means, grouping the plurality of tomographic images according to groups of tomographic images having an interval of imaging positions equal to or less than a predetermined value.

Description

  The present invention relates to an image processing apparatus and a control method for the image processing apparatus.

  At the time of interpretation, a doctor may refer to an image that has been noticed in past interpretation in order to know the state of interpretation performed by himself or another doctor in the past. For example, in double interpretation in lung cancer screening, for the purpose of reducing the risk of oversight at the time of interpretation, a point (image) noted by another doctor may be referred to. In addition, for the purpose of improving the accuracy of interpretation, the location that the user has noticed at the time of past interpretation may be referred to. Therefore, if a past attention image (an image noticed in past interpretation) can be easily confirmed, it is possible to reduce the burden on the doctor at the time of interpretation. In particular, in Computed Tomography (CT), an image exceeding 1000 sheets may be generated in one inspection. Therefore, in lung cancer screening using CT, it is particularly important that past attention images can be efficiently confirmed from a number of tomographic images.

  Doctors who interpret images when diagnosing lung cancer (hereinafter referred to as interpreting physicians), for example, the size / shape of the nodule, the positional relationship between the nodule and the bronchus / mediastinum / chest wall, etc. Interpret the tomographic image for the purpose of confirming the presence or absence. The interpreting doctor selects an interpretation image (interpretation location) from the tomographic image, and performs operations such as length measurement, enlargement / reduction, and comparison with the previous and subsequent images (images adjacent to the imaging position). Interpretation while performing (interpretation operation). In other words, the interpretation doctor performs interpretation while switching the interpretation location and interpretation operation depending on the purpose of interpretation.

Patent Document 1 discloses a technique for displaying only an image focused on in the past and an image having a spatial distance close to the focused image when interpreting a tomographic image or the like.
In Patent Document 2, the date and time of digital image shooting, the date and time when the digital image was stored, printed and processed, the type of operation performed on the digital image, and the time interval between each operation were combined. A technique for grouping is disclosed.

JP 2006-271541 A JP 2007-323416 A

  According to the technique of Patent Document 1, priority is set for each image based on the number of times the image is displayed, the display time, the spatial distance from the image of interest, etc. Only the image of the degree can be displayed. However, the technique of Patent Document 1 cannot distinguish and display images with different interpretation purposes.

  Further, according to the technique disclosed in Patent Document 2, images can be grouped according to combinations of types of operations performed on images and time intervals between the operations. However, in the technique of Patent Document 2, images that have been subjected to the same operation at the time of past interpretation are classified into the same group even if the interpretation purposes are different. For example, an image that has undergone an enlargement operation to interpret the lung in detail and an image that has undergone an enlargement operation to interpret the stomach in detail are classified into the same group.

  As described above, when the conventional technique is used, grouping (selection) is insufficient, and thus the burden on the interpreting doctor cannot be reduced sufficiently.

  An object of this invention is to provide the technique which can reduce the burden of the user at the time of referring the image noted at the time of the past reference.

The image processing apparatus of the present invention
Acquisition means for acquiring reference history information that represents past reference statuses of a plurality of tomographic images having different imaging positions, and includes at least information indicating a cumulative reference count or a cumulative reference time of each tomographic image; ,
Using the reference history information, an extraction unit that extracts, among the plurality of tomographic images, a tomographic image whose cumulative reference count or cumulative reference time is equal to or greater than a predetermined threshold as a tomographic image noted at the time of past reference;
When there are a plurality of tomographic images extracted by the extracting unit, a classifying unit for grouping for each collection of tomographic images whose imaging position intervals are equal to or smaller than a predetermined value;
It is characterized by having.

The control method of the image processing apparatus of the present invention includes:
Obtaining reference history information including reference history information indicating past reference situations of a plurality of tomographic images having different imaging positions, and including at least information indicating a cumulative reference count or a cumulative reference time of each tomographic image;
Using the reference history information, extracting, among the plurality of tomographic images, a tomographic image having a cumulative reference count or a cumulative reference time of a predetermined threshold or more as a tomographic image noted at the time of past reference;
When there are a plurality of the extracted tomographic images, the step of grouping for each collection of tomographic images whose imaging position interval is a predetermined value or less;
It is characterized by having.

  ADVANTAGE OF THE INVENTION According to this invention, the user's burden at the time of referring the image noticed at the time of the past reference can be reduced.

1 is a functional block diagram of an image processing apparatus according to Embodiment 1 of the present invention. A diagram for explaining the relationship between a subject and a tomographic image The figure for demonstrating the interpretation history information used in Example 1 of this invention The figure for demonstrating the grouping process of Example 1 of this invention The figure for demonstrating the group information of Example 1 of this invention The figure for demonstrating the example of grouping display of Example 1 of this invention The figure for demonstrating the interpretation history information used in Example 2 of this invention. The figure for demonstrating the grouping process of Example 2 of this invention The figure for demonstrating the grouping display example of Example 2 of this invention The figure for demonstrating the classification | category of interpretation operation of Example 3 of this invention The figure for demonstrating the grouping process of Example 3 of this invention

<Example 1>
Hereinafter, an image processing apparatus and a control method for the image processing apparatus according to Embodiment 1 of the present invention will be described. In this embodiment, an example will be described in which a user refers to a medical image (for example, a tomographic image of a human body) that has been noted at the time of past interpretation using an image processing apparatus.
FIG. 1 is a block diagram illustrating an example of a functional configuration of the image processing apparatus according to the first embodiment. The image processing apparatus 100 includes an input unit 101, a communication unit 102, a storage unit 103, a display control unit 104, a group generation unit 105, a display unit 106, and the like. The image processing apparatus 100 is connected to the server 107 via the communication unit 102 so as to be communicable. The display unit 106 may be a part of the image processing apparatus 100 or may be a separate device from the image processing apparatus.

The input unit 101 receives an input related to image display from the user and outputs it to the communication unit 102 and the display control unit 104.
The communication unit 102 acquires medical images to be interpreted (a plurality of tomographic images having different imaging positions) from the server 107 based on a request from the user, and outputs them to the storage unit 103. Further, the communication unit 102 receives interpretation history information (reference history) when a confirmation request (a attention image confirmation request) of a tomographic image (a attention image) that has been noticed at the time of past interpretation is input from the user via the input unit 101. Information) is acquired from the server 107 and stored in the storage unit 103. The interpretation history information is information representing the past interpretation situation (reference situation) of a plurality of acquired tomographic images. The interpretation history information is information recorded at the time of interpretation by an interpretation apparatus (not shown), and is information representing the interpretation history according to the operation of the interpretation apparatus by the user. Note that the image processing apparatus 100 may have a function of recording the interpretation history information.

The storage unit 103 stores the tomographic image acquired via the communication unit 102 and the interpretation history information.
The display control unit 104 reads a tomographic image from the storage unit 103 in response to an input from the user, generates a display image, and outputs the display image to the display unit 106. When group information is generated by the group generation unit 105 described later, a display image of the attention image grouped based on the group information is generated and output to the display unit 106. The group generation unit 105 extracts and groups an attention image from the acquired tomographic image based on the interpretation history information stored in the storage unit 103 when an attention image confirmation request is input by the user. Generate group information. The display unit 106 displays a display image in response to an input from the display control unit 104.

Here, an example of the flow of processing when the image interpretation device generates image interpretation history information will be described.
First, when an interpretation start is detected by an input of an interpretation start request by a user (for example, an interpreting physician), the interpretation apparatus newly creates interpretation history information including patient information, examination information, imaging date, interpretation date, and the like.
Next, each time the user performs an operation such as switching the display image or enlarging / reducing the user, the image interpretation apparatus displays the newly created image interpretation history indicating the time when the operation was performed, the tomographic image where the operation was performed, the type of operation, and the like. Add to the information.
Finally, when the interpretation end is detected by the input of the interpretation termination request by the user, the interpretation apparatus records the interpretation history information in the server 107 in association with the interpreted image group.

  Next, in the image processing apparatus according to the first embodiment, an example of processing in which the group generation unit 105 generates group information from the interpretation history information, and the display control unit 104 and the display unit 106 display past attention images in groups. Will be described in detail with reference to FIGS.

  FIG. 2 is a diagram schematically illustrating a relationship between a subject (human chest) and a captured CT image (a tomographic image captured by CT). As shown in FIG. 2, each tomographic image shows a different part. Here, slice numbers (0000, 0001, 0002,...) Are assigned to each tomographic image in order from the tomographic image on the head side of the human body.

FIG. 3 is a diagram schematically showing the interpretation history information used in the present embodiment. As shown in FIG.
The interpretation history information includes patient information, examination information, imaging date, and interpretation date. When a plurality of interpretation history information is recorded in the server 107, the communication unit 102 specifies and acquires past interpretation history information corresponding to the current interpretation based on patient information, examination information, imaging date, interpretation date, and the like. To do. The interpretation history information also includes information on interpretation work (past interpretation status). Here, the slice number of the displayed tomographic image, the display start time, and the type of operation performed are recorded in chronological order as information relating to the interpretation work.
For each tomographic image, the difference between the display start time and the display end time of the tomographic image (that is, the display start time of the next tomographic image) is calculated for each tomographic image. The interpretation time (reference time) for each time can be calculated. Then, the cumulative display time (cumulative reference time) of each tomographic image can be calculated by accumulatively adding each interpretation time of the tomographic images having the same slice number. Further, by counting the number of appearances of tomographic images having the same slice number, the cumulative display number (cumulative reference number) of each tomographic image can be calculated. Further, by referring to the slice number corresponding to the type of operation performed, it is possible to identify a tomographic image for which a predetermined operation has been performed at the time of past reference.
In the present embodiment, the cumulative display count of each tomographic image, the cumulative display time of each tomographic image, and the interpretation time for each tomographic image are calculated from the interpretation history information. May contain such information. In this embodiment, the slice number corresponding to the type of operation that has been performed is referred to identify a tomographic image that has been subjected to a predetermined operation at the time of past reference. Information that directly indicates a tomographic image that has been subjected to a predetermined operation at the time of past reference may be included.
In addition, although a present Example demonstrates the structure by which the image interpretation log | history information like FIG. 3 is matched for every image group (for example, every patient, imaging | photography site | part, each imaging | photography device, etc.) containing a some tomographic image, The tomographic image interpretation history information may be associated with each tomographic image.

  FIG. 4 is a diagram schematically showing a past interpretation situation for the tomographic image shown in FIG. 2 and a flow of grouping (grouping) of the tomographic images. In the example of FIG. 4, as indicated by 4 a, in the past interpretation, the interpretation is performed on the tomographic images of slice numbers 0000 to 1000 while switching the tomographic images. In particular, it is assumed that the tomographic images of slice numbers 0100 to 0140 and 0680 to 0700 have been repeatedly displayed in order to focus interpretation (for example, the tomographic images of slice numbers 0100 to 0140 and 0680 to 0700 are each five times or more. Suppose it was displayed). Furthermore, a length measurement operation is performed on the tomographic images of slice numbers 0130 and 0131 in order to measure the size of an abnormal shadow or the like. Note that the operation history information in FIG. 3 represents a partial situation of the past interpretation. From FIG. 3, the interpretation time of the tomographic images of slice numbers 0100 to 0140 (each 5 seconds or longer) is longer than the interpretation time of the tomographic images of slice numbers 0000 to 0099 (each 2 seconds), and slice numbers 0130 and 0131. It can be read that the measurement operation is performed on the tomographic image of the image.

  At this time, the group generation unit 105 extracts a tomographic image (attention image) that has been noted at the time of past interpretation from a plurality of tomographic images acquired by the communication unit 102 using the interpretation history information. In the present embodiment, a tomographic image whose cumulative display count at the time of past interpretation is a predetermined number (for example, 5 times) or more and a tomographic image whose cumulative display time is a predetermined time (for example, 20 seconds) or more are extracted as attention images. To do. The accumulated display count and accumulated display time of each tomographic image are acquired from the interpretation history information stored in the storage unit 103. Specifically, the cumulative display count is acquired for each tomographic image by counting the number of slice numbers of the tomographic image included in the interpretation history information. The cumulative display time is acquired by calculating the difference between the display start time of the tomographic image and the display start time of the next tomographic image for each tomographic image and cumulatively adding the calculation results. As a result, in the present embodiment, as indicated by 4b in FIG. 4, tomographic images having slice numbers 0100 to 0140 and slice numbers 0680 to 0700 are extracted as attention images.

In addition, in order to simplify the processing, instead of the above processing, a tomographic image that has been subjected to a specific operation (predetermined operation) such as enlargement at the time of past interpretation may be extracted as an attention image. In this case, operations such as enlargement and measurement, which are highly likely to be performed on the target image, are recorded in advance as specific operations, and the tomographic image subjected to the specific processing is extracted as the target image. .

  Next, the group generation unit 105 groups the extracted attention images based on the interpretation points. In the present embodiment, the group generation unit 105 performs grouping based on the distance between the images of interest in the direction perpendicular to the tomographic image (difference in imaging position). The direction perpendicular to the tomographic image is the z-axis direction (the direction of the arrow 201) with respect to the tomographic image capturing plane (xy plane, plane parallel to the arrow 202) shown in FIG.

  Specifically, when there are a plurality of extracted tomographic images (target images), the group generation unit 105 has tomographic images whose interval between imaging positions (difference between imaging positions between adjacent tomographic images) is equal to or less than a predetermined value. Group by image collection. In this embodiment, the group generation unit 105 sets the group boundary in a portion where the slice number of the target image is discontinuous, that is, the slice number of the target image is separated by a predetermined number (for example, 2) or more. , Group the images of interest. As a result, the target images can be grouped for each spatially continuous tomographic image group (spatial discontinuous tomographic images can be grouped into different groups). As a result of the above, in the example of FIG. 4, the tomographic images with slice numbers 0100 to 0140 are set to group 1 and the tomographic images with slice numbers 0680 to 0700 are set to group 2 as indicated by 4 c. Note that whether or not the slice number of the target image is discontinuous is related to the threshold value in the above-described condition for extracting the target image. Therefore, in order to perform grouping with higher accuracy, the threshold value for setting the group boundary may be determined empirically in relation to the conditions for extracting the target image. For example, if the target image extraction conditions are loose (the cumulative display time and the cumulative display count threshold are small), group boundaries may be set at portions where the slice number is N or more (N is an integer of 1 or more). Good. If the target image extraction conditions are strict (the cumulative display time and the cumulative display count threshold are large), the group boundary may be set at a portion where the slice number is M or more (M is an integer greater than N).

  Next, the group generation unit 105 groups the grouped attention images in more detail based on the interpretation operation. Specifically, using the interpretation history information, the group generation unit 105 sets, for each group, a group of tomographic images that have been subjected to a predetermined operation at the time of past interpretation among the tomographic images belonging to the group as another group. Set.

In this embodiment, the group generation unit 105 extracts a tomographic image in which a predetermined operation has been performed for each of the groups 1 and 2 if it exists in the group, and sets this as another group. Here, the predetermined operation is an operation performed in accordance with the purpose of interpretation, for example, an enlargement operation, length measurement, WW / WL (Window Width / Window LEVEL) change, and the like. As a result, in the example of FIG. 4, the length measurement tool is used for the tomographic images of slice numbers 0130 and 0131 as indicated by 4 d, so that the tomographic images of slice numbers 0130 and 0131 are group 3. Set to
When a plurality of types of processing are set as the predetermined processing, the above processing (grouping based on the interpretation operation) may be performed for each processing type, or the plurality of types of processing may be performed as one predetermined processing. The above process may be performed as the process.
Also in the grouping based on the interpretation operation, the tomographic images on which the predetermined operation has been performed may be grouped so that the intervals of the photographing positions become a group having a predetermined value or less.

  As described above, the group generation unit 105 groups the images of interest based on the site of interest and the interpretation operation. In this embodiment, after grouping based on the interpretation location, further grouping is performed based on the interpretation operation. However, the order of grouping may be reversed.

  Finally, the group generation unit 105 outputs the group information illustrated in FIG. 5 to the display control unit 104 as information regarding grouping. As shown in FIG. 5, in this embodiment, the group information includes, for each group, the group ID, the slice number of the tomographic image belonging to the group, the number of tomographic images belonging to the group, and the type of operation performed on the group. . These pieces of information are acquired by grouping processing by the group generation unit 105.

Next, the display control unit 104 configures a display screen for grouping and displaying past attention images based on the group information, and the display unit 106 displays the configured display screen. FIG. 6 shows a display example in which tomographic images are grouped and displayed.
The group display area 601 is an area for displaying a group included in the tomographic image and for allowing the user to select a group to be displayed in detail. In the present embodiment, the group ID is displayed in the area, but more detailed information may be displayed. For example, information to be displayed in the group information display area 602 described later may be written together.

  The group information display area 602 is an area for displaying information regarding the group selected by the user on the group display area 601. In the present embodiment, the number of image-cutting images belonging to the selected group, the type of operation performed on the group, and the time required for group interpretation are displayed. The type of operation is determined by preparing a table in which the type is registered in advance and performing a comparison process with the table during the grouping process shown in FIG. Similarly, the time required for group interpretation is also acquired during the group generation process shown in FIG. In addition, when displaying the information regarding a group in detail, you may display the image display frequency, an image display time, etc. in the group information display area 602. FIG.

  The image display area 603 is an area for displaying a tomographic image belonging to the group selected by the user on the group display area 601. The tomographic image displayed in the area can be switched by an operation bar described later. Note that the tomographic image that is displayed first when the group is selected is the tomographic image at the top of the group, the tomographic image with the largest number of past display times in the group, the tomographic image with the longest past display time, and the like.

  The operation bar display area 604 is an area for displaying an operation bar for allowing the user to switch the tomographic image displayed in the image display area 603. In this embodiment, a button for displaying the first / last tomographic image of the group is displayed in the operation bar display area 604, and a tomographic image before / after the tomographic image being displayed in the image display area 603 is displayed. A button is displayed. In addition, a button for reproducing cine from the top of the tomographic image belonging to the group and a button for reproducing cine from all of the tomographic images (all of the attention images and all of the tomographic images acquired by the communication unit 102) are displayed. The

As described above, according to the present embodiment, it is possible to group the images of interest based on the interpretation points and display them in units of groups. This facilitates confirmation of an image that has been noted at the time of past reference, and can reduce the burden on the user (doctor) when referring to an image that has been noted at the time of past reference.
Further, in this embodiment, since grouping based on the interpretation operation is further performed, it is easy to confirm the tomographic image group on which the predetermined operation has been performed, and the burden on the user can be further reduced.

In the present embodiment, the case where the tomographic image is a medical image has been described, but the tomographic image may not be a medical image. For example, a tomographic image representing a cross section of members constituting the apparatus may be used.
In this embodiment, the images of interest grouped based on the reference location are grouped in more detail based on the interpretation operation. However, the grouping based on the interpretation operation may not be performed.
Depending on the purpose, it is sufficient that only tomographic images with a high number of references or tomographic images with a long reference time can be confirmed. Therefore, even if the grouping based on the interpretation operation is not performed, the effect equivalent to the above effect can be obtained. Can do. When the grouping based on the interpretation operation is not performed, the interpretation history information may not include information indicating a tomographic image on which a predetermined operation has been performed at the time of past reference.
In this embodiment, the interpretation history information includes information indicating the cumulative reference count and information indicating the cumulative reference time. However, the interpretation history information is not limited to this. It is only necessary to include at least either information indicating the cumulative reference count or information indicating the cumulative reference time. Then, a tomographic image whose cumulative reference count or cumulative reference time is equal to or greater than a predetermined threshold may be extracted as a target image. Even if it is such a structure, the effect according to the said effect can be acquired.
In this embodiment, for each tomographic image, the reference time for each tomographic image is calculated and the accumulated time is calculated. However, the present invention is not limited to this configuration. For example, the interpretation history information may include not the display start time of each tomographic image but the accumulated reference time of each tomographic image itself.

<Example 2>
In the first embodiment, the images of interest are grouped based on the distance in the direction perpendicular to the tomographic image, but in this embodiment, the images of interest are based on the distance in the direction perpendicular to the tomographic image and in the horizontal direction. Group. The horizontal direction with respect to the tomographic image is the direction in the xy plane (arrow 202) shown in FIG.
The block diagram of the image processing apparatus in this embodiment is the same as that shown in FIG. The processing contents of the input unit 101, the communication unit 102, the storage unit 103, the display control unit 104, and the display unit 106 are the same as those in the first embodiment. Hereinafter, processing of the group generation unit 105 according to the present embodiment will be described with reference to FIGS.

  FIG. 7 is a diagram schematically showing the interpretation history information used in this embodiment. As shown in FIG. 7, the interpretation history information used in this embodiment includes patient information, examination information, imaging date, interpretation date, and information related to interpretation work. Operations such as enlargement operation and length measurement are performed on a partial region of the tomographic image. In this embodiment, the slice number of the displayed tomographic image, the display start time, the type of operation performed, and the position of the area where the operation was performed are recorded in time-series order as information about the interpretation work. . The position of the region where the operation is performed is represented by coordinates (x, y), for example, with the top left vertex of the tomographic image as the origin (0, 0).

  FIG. 8 is a diagram schematically illustrating a past interpretation situation for the tomographic image illustrated in FIG. 2 and a flow of grouping the tomographic images. In the example of FIG. 8, as shown by 8a, in the past interpretation, the interpretation of the tomographic images of slice numbers 0000 to 1000 is performed while switching the tomographic images. In particular, it is assumed that the tomographic images having slice numbers 0200 to 0400 have been repeatedly displayed in order to focus interpretation. Furthermore, for the tomographic images of slice numbers 0230 to 0235, since there was an abnormal shadow near the coordinates (400, 100) in the tomographic image, the coordinates (400, 100) are centered in order to confirm the details of the abnormal shadow. An enlargement operation has been performed. Similarly, for the tomographic images of slice numbers 0270 to 0275, an enlargement operation is performed around the coordinates (80, 300) in the tomographic image. As described above, the history regarding the type and coordinates of the operation is recorded as the interpretation history information shown in FIG.

At this time, the group generation unit 105 extracts a target image from a plurality of tomographic images acquired by the communication unit 102 as in the first embodiment. As a result, in this embodiment, as shown by 8b in FIG. 8, tomographic images having slice numbers 0200 to 0400 are extracted.
Next, as in the first embodiment, the group generation unit 105 performs grouping based on the distance between the images of interest in the direction perpendicular to the tomographic image. As a result, in the example of FIG. 8, as indicated by 8 c, the tomographic images (all images of interest) with slice numbers 0200 to 0400 are classified into group 1.

  Then, the group generation unit 105 groups the tomographic images based on the horizontal distance. Specifically, the group generation unit 105 uses, for each group, the reference history information, and among the tomographic images belonging to the group, the difference in the position of the region where the predetermined operation between the tomographic images is performed is a predetermined value. A group of tomographic images below (for example, 50 pixels or less) is set as another group.

In the present embodiment, first, a location (a focused location) that is particularly noticed by the interpretation doctor is identified in each tomographic image. Specifically, the group generation unit 105 refers to the interpretation history information shown in FIG. 7 and extracts the coordinate information of the place where the operation is performed on the tomographic image. As a result, in the present embodiment, it is specified that the point of interest of the tomographic images with slice numbers 0230 to 0235 is the coordinates (400, 100) and the point of interest of the tomographic images with the slice numbers 0270 to 0275 is the coordinates (80, 300). Is done.
Next, the group generation unit 105 calculates a distance between points of interest from the coordinate information, and sets a group in which the difference in distance is equal to or less than a predetermined value as another group. The predetermined value is set in accordance with the number of pixels in the tomographic image in consideration of the correspondence between the coordinate distance and the actual length of the subject depending on the number of pixels in the tomographic image. As a result, in the example of FIG. 8, as shown by 8d, the tomographic images with slice numbers 0230 to 0235 are set to group 2, and the tomographic images with slice numbers 0270 to 0275 are set to group 3.

  Then, the group generation unit 105 groups the tomographic images in more detail based on the interpretation operation. In the present embodiment, the enlargement operation is performed on the tomographic images of slice numbers 0230 to 0235 and slice numbers 0270 to 0275. However, these tomographic images are already grouped as group 2 and group 3, and no interpretation operation other than the enlargement operation is performed. For this reason, in the example of FIG. 8, as shown by 8e, the result of the grouping does not change even if the detailed grouping by the interpretation operation is performed. In the grouping based on the interpretation operation, when the horizontal distance of the tomographic image is not considered, the tomographic images having the slice numbers 0230 to 0235 and the slice numbers 0270 to 0275 may be set as one group.

  Based on the grouping made as described above, the display control unit 104 configures a display screen, and the display unit 106 displays the configured display screen. FIG. 9 shows a display example in which tomographic images are grouped and displayed. In the present embodiment, the coordinate information of the attention location is further displayed in the group information display area 602, and the marker 901 indicating the attention location is superimposed and displayed in the image display area 603.

As described above, according to the present embodiment, in addition to the configuration of the first embodiment, grouping based on the distance in the horizontal direction of the tomographic image (the distance between the points of interest) is further performed. Thereby, the tomographic images in which the distances between the points of interest are separated can be divided into different groups. And it becomes easy to confirm the tomographic image near the point of interest, and the burden on the user can be further reduced.
The order of the above three grouping processes (grouping based on the distance in the direction perpendicular to the tomographic image, grouping based on the distance in the direction parallel to the tomographic image, and grouping based on the interpretation operation) is as follows. It may be performed in any order.
In the present embodiment, an example in which one attention location exists in one tomographic image has been described. However, a plurality of attention locations may exist in one tomographic image. When a plurality of points of interest exist in one tomographic image, for example, the following processing may be performed. First, when determining whether or not the distance in the horizontal direction exceeds a predetermined value, the distances between all the points of interest are calculated. When there are combinations of points of interest that do not exceed the predetermined value, the tomographic images are set in the same group. As a result, it is possible to group tomographic images having a plurality of points of interest so as to belong to a plurality of groups.

<Example 3>
Interpretation includes, for example, confirming the presence / absence of cancer metastasis and confirming the extent of spread as confirming cancer invasion. In the former case, it is expected that the same image is displayed for a relatively long time, and in the latter case, the tomographic image is repeatedly displayed for a short time while switching between adjacent tomographic images. That is, if the purpose at the time of image interpretation is different, the relationship between the number of display times of the image and the display time is also different. Therefore, in this embodiment, the tomographic image groups having a specific interpretation pattern such as a tomographic image group with a small display number but a long display time, or a tomographic image group with a large display number but a short display time are different from each other. As shown in FIG. According to the present embodiment, it is possible to realize grouping according to an interpretation pattern at the time of interpretation even for a tomographic image for which a specific operation such as enlargement / measurement has not been performed.

  The block diagram of the image display apparatus in the present embodiment is the same as FIG. The processing contents of the input unit 101, the communication unit 102, the storage unit 103, the display control unit 104, and the display unit 106 are the same as those in the first embodiment. The interpretation history information used in this embodiment is the same as that in the first embodiment. Hereinafter, processing of the group generation unit 105 will be described with reference to FIGS.

  The group generation unit 105 according to the present exemplary embodiment sets, for each group, tomographic image groups having different interpretation patterns among the tomographic images belonging to the group as separate groups using the interpretation history information. Specifically, a group of tomographic images in which the maximum reference time is equal to or greater than the first threshold and the cumulative reference count is equal to or less than the second threshold is set as one group. Also, a set of tomographic images in which the maximum reference time is equal to or smaller than a third threshold smaller than the first threshold and the cumulative reference count is equal to or larger than a fourth threshold larger than the second threshold is set as one group. To do. In the present embodiment, the group generation unit 105 performs such grouping using the table shown in FIG.

  FIGS. 10A and 10B are diagrams for explaining tables used in pattern classification of interpretation operations used in this embodiment. As shown in FIG. 10A, in this embodiment, for example, the maximum value of reference time (maximum display time) is a predetermined value (for example, 20 seconds) or more, and the cumulative reference count (cumulative display count) is a predetermined value. A tomographic image that is less than (for example, 5 times) is classified as A. A tomographic image whose maximum display time is a predetermined value (for example, 5 seconds) or less and whose cumulative display count is a predetermined value (for example, 10 times) or more is set as a classification B. The group generation unit 105 holds information related to the classification in advance as a table illustrated in FIG. Note that the threshold for the maximum display time and the threshold for the total number of display times may be determined experimentally based on the examination subject, the age of the interpretation doctor, and the like. For example, in order to confirm the presence or absence of a nodule or the like with the table in FIG. 10B, a tomographic image that has been displayed for a long time with a small number of displays is confirmed in the category A to confirm the extent of the nodule or the like. In addition, a tomographic image with a short display time but a large number of display times is classified as classification B.

  FIG. 11 is a diagram schematically illustrating a past interpretation state of the tomographic image illustrated in FIG. 2 and a flow of grouping the tomographic images. In the example of FIG. 11, as indicated by 11 a, in the past interpretation, the interpretation is performed while switching the tomographic images for the tomographic images of slice numbers 0000 to 1000. In particular, it is assumed that the tomographic images having slice numbers 0200 to 0400 have been repeatedly displayed in order to focus interpretation. Specifically, in order to confirm the presence or absence of nodules, the tomographic images of slice numbers 0220 to 0225 are each displayed twice, and are displayed for 22 seconds or more at a time. In addition, in order to confirm the spread of the nodule, the tomographic images of slice numbers 0300 to 0320 are displayed 12 times each and are displayed for 4 seconds or less at a time.

At this time, the group generation unit 105 extracts a target image from a plurality of tomographic images acquired by the communication unit 102 as in the first embodiment. As a result, in this embodiment, as shown by 11b in FIG. 11, tomographic images having slice numbers 0200 to 0400 are extracted.
Next, as in the first embodiment, the group generation unit 105 performs grouping based on the distance between the images of interest in the direction perpendicular to the images. As a result, in the example of FIG. 11, the tomographic images (all images of interest) with slice numbers 0200 to 0400 are classified into group 1 as indicated by 11c.

  Then, the group generation unit 105 calculates the maximum display time and the total number of times of display for the extracted attention images, and classifies the attention images into classifications A and B according to the calculation result. The above calculation is performed by referring to the interpretation history information. The classification is performed by referring to the table shown in FIG. As a result, in the example of FIG. 11, the tomographic images with slice numbers 0220 to 0225 are classified into classification A, and the tomographic images with slice numbers 0300 to 0320 are classified into classification B.

  Next, the group generation unit 105 further groups the attention images grouped based on the distance in the direction perpendicular to the tomographic image using the classification result. Specifically, the tomographic image groups classified into the classifications A and B are set as separate groups. As a result, in the example of FIG. 11, the tomographic images with slice numbers 0220 to 0225 are set to group 2 and the tomographic images with slice numbers 0300 to 0320 are set to group 3 as indicated by 11 d.

  Based on the grouping made as described above, the display control unit 104 configures a display screen, and the display unit 106 displays the configured display screen. Note that when displaying information related to a group in more detail, the maximum display time and the cumulative display count of each tomographic image of the group may be displayed in the group information display area 602 in the display example of FIG.

  As described above, according to the present embodiment, a specific interpretation pattern is determined from the relationship between the number of display times and the display time, and grouping is performed based on the pattern. Thereby, confirmation of a tomographic image of a specific interpretation pattern is facilitated, and the burden on the user can be further reduced.

DESCRIPTION OF SYMBOLS 100 Image processing apparatus 102 Communication part 105 Group generation part

Claims (5)

Acquisition means for acquiring reference history information that represents past reference statuses of a plurality of tomographic images having different imaging positions, and includes at least information indicating a cumulative reference count or a cumulative reference time of each tomographic image; ,
Using the reference history information, an extraction unit that extracts, among the plurality of tomographic images, a tomographic image whose cumulative reference count or cumulative reference time is equal to or greater than a predetermined threshold as a tomographic image noted at the time of past reference;
When there are a plurality of tomographic images extracted by the extracting unit, a classifying unit for grouping for each collection of tomographic images whose imaging position intervals are equal to or smaller than a predetermined value;
An image processing apparatus comprising: The reference history information includes information indicating a tomographic image in which a predetermined operation has been performed at the time of past reference,
The classification means uses the reference history information to set, for each group, a group of tomographic images in which the predetermined operation has been performed at the time of past reference among the tomographic images belonging to the group as another group. The image processing apparatus according to claim 1, wherein: The reference history information includes information indicating a tomographic image in which a predetermined operation has been performed on a partial region of the tomographic image at the time of past reference, and the position of the region in the tomographic image in which the predetermined operation has been performed. Including information representing
The classification means uses, for each group, the difference between the positions of the areas in the tomographic image in which the predetermined operation is performed between the tomographic images among the tomographic images belonging to the group using the reference history information. The image processing apparatus according to claim 1, wherein a group of the following tomographic images is set as another group. The reference history information includes information representing a cumulative reference count of each tomographic image and information representing a reference time for each tomographic image at the time of past reference,
The classification means uses the reference history information, and for each group, among the tomographic images belonging to the group, the maximum value of the reference time is not less than the first threshold value, and the cumulative reference count is the second threshold value. The collection of tomographic images and the maximum reference time are equal to or smaller than a third threshold smaller than the first threshold, and the cumulative reference count is equal to or larger than a fourth threshold larger than the second threshold. The image processing apparatus according to claim 1, wherein a set of tomographic images is set as a separate group. Obtaining reference history information including reference history information indicating past reference situations of a plurality of tomographic images having different imaging positions, and including at least information indicating a cumulative reference count or a cumulative reference time of each tomographic image;
Using the reference history information, extracting, among the plurality of tomographic images, a tomographic image having a cumulative reference count or a cumulative reference time of a predetermined threshold or more as a tomographic image noted at the time of past reference;
When there are a plurality of the extracted tomographic images, the step of grouping for each collection of tomographic images whose imaging position interval is a predetermined value or less;
A control method for an image processing apparatus, comprising:

Images