JP2008006187A - Medical image display processing aparatus and medical image display processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an apparatus and the like capable of easily displaying the axial image of a part desired by a user among a series of axial images acquired by imaging and inspecting a subject by modality. <P>SOLUTION: An image server 2 for displaying the axial image at an image display terminal 3 on the basis of image data indicating a series of axial images obtained by imaging and inspecting the subject comprises: a part recognition part 12 for recognizing the part of a body part indicated in each of a series of axial images; and a display processing part 14 for displaying one axial image included in a series of axial images at the image display terminal 3 and displaying the axial image indicating the part different from the part of the axial image displayed at the image display terminal 3 at the image display terminal 3 on the basis of a recognized result by the part recognition part 12 in the case of receiving a part change instruction. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a medical image display processing device for displaying a medical image based on image data acquired by a medical imaging modality, and a medical image display processing program used in such a device.

  In recent years, medical images displaying the inside of a living body are often used in medical diagnosis, and in order to acquire such medical images, X-ray imaging or X-ray CT (computed tomography) is used. Various technologies and devices (modalities) such as devices, ultrasound (US) diagnostic devices, MRI (magnetic resonance imaging) devices, and PET (positron emission tomography) are widely used. Yes. Many of these devices have been digitized, and construction of diagnostic information processing systems and the like in hospitals is underway. Among these imaging technologies, CT and MRI can obtain and display biological dislocation images at relatively narrow intervals, and thus have achieved great results in the detection and evaluation of lesion sites in the living body. Here, the axial discontinuity image refers to a tomographic image in which a plane orthogonal to or approximately orthogonal to the body axis of the subject (so-called a circular cut surface) is represented. Hereinafter, the axial dislocation image is also simply referred to as a slice image.

  By the way, when performing tomographic imaging such as CT examination, only one part (for example, only the chest, only the abdomen, etc.) is not necessarily imaged in one examination, but over a plurality of parts (for example, (Chest-abdomen, head-chest, etc.) imaging is often performed. Since a series (one series) of slice images obtained by imaging usually has several hundred to several thousand images, the labor and burden for displaying the slice images of the part desired by the interpreting physician is considerable. large.

  As a related technique, the present applicant has proposed an image display device capable of reproducing and displaying a cross-sectional image designated by a user (see Patent Document 1 below). This image display device, based on three-dimensional medical image data representing a subject, moves a cross-section of a cross-sectional image to be displayed around a direction perpendicular to the cross-section, and generates a plurality of cross-sectional images continuous in the direction of the axis. In an image display device that sequentially displays, a designation unit that designates a cross-sectional image desired by a user among a plurality of cross-sectional images that are sequentially displayed, and a direction perpendicular to the cross-section of the cross-sectional image designated by the designation unit A first storage means for storing a first display parameter for displaying a cross-sectional image including the position, and a first display parameter stored in the first storage means in response to a user request; And a first image display means for displaying a specified cross-sectional image (second page, FIG. 2). That is, in Patent Document 1, a user specifies a desired cross-sectional image, stores a first display parameter for displaying the specified cross-sectional image, and is specified based on the stored first display parameter. A cross-sectional image is displayed.

  In Patent Document 1, it is necessary to specify a cross-sectional image desired by the user. However, since one series of slice images usually has several thousand to several thousand images, the slice image desired by the user is selected. The effort and burden to specify is large.

  Further, as a related technique, Patent Document 2 discloses that a huge amount of medical images are extracted from medical images by automatically extracting imaging attribute information such as an imaging device, a part, and an imaging direction from management images and adding them to management information. A system for enabling efficient management of a system is disclosed. This medical image identification system is a system for storing and managing medical images, and classifies input images, template image storage means for storing template images relating to categories to be identified for each classification, and the template images A means for selecting a plurality of categories of template images that are candidates for identification using the classification result of the input image from the storage means, and comparing the selected template image with the input image, and the category of the template image that best matches Image identification means for determining the management information, and means for adding the management information of the template image to the management information of the input image (second page, FIG. 1). That is, in Patent Document 2, an input image is classified based on the size of the input image, the number of rectangular areas in the input image, and the like, and a plurality of categories of template images are selected according to the classification, and the selected template image is selected. The input information is extracted by extracting the gray level information of the input image or the one matching the mosaic image, and the management information (for example, the imaging part) given to the template image is given as the management information of the input image. The imaging part of the image is determined.

However, only one part is not always represented in one medical image. For example, a plurality of parts are often displayed such that a chest is displayed in a partial area of the image and an abdomen is displayed in another partial area. However, Patent Document 2 does not describe any part recognition for an image displaying a plurality of parts.
Japanese Patent Laying-Open No. 2004-167042 (second page, FIG. 2) JP 2002-253539 A (2nd page, FIG. 1)

  Therefore, in view of the above points, the present invention provides an axial discontinuity image of a part desired by a user in one series of axial discontinuity images obtained by imaging examination of a subject using a modality such as CT or MRI. It is an object of the present invention to provide a device capable of easily displaying a message and a program used in such a device.

  In order to solve the above problems, a medical image display processing device according to one aspect of the present invention is based on image data representing a series of axial discontinuity images obtained by imaging and examining a subject. An apparatus for displaying an image on a display unit, part recognition means for recognizing a part of a body part represented in each of a series of axial discontinuity images, and 1 included in one series of axial discontinuity images When one axial disconnection image is displayed on the display unit and a part change instruction is received, a part different from the part of the axial disconnection image displayed on the display part is represented based on the recognition result by the part recognition means Display processing means for displaying an axial dislocation image on the display unit.

  In addition, a medical image display processing program according to one aspect of the present invention provides a display unit with an axial dislocation image based on image data representing a series of axial discontinuity images obtained by imaging and examining a subject. A program for displaying, a procedure (a) for recognizing a part of a body part represented in each of a series of axial position disconnection images, and one axis included in the one series of axial position disconnection images The procedure (b) for displaying the dislocation image on the display unit and the part of the axial dislocation image displayed on the display unit are different based on the recognition result in the procedure (a) when receiving the region change instruction. And causing the CPU to execute a procedure (c) of displaying an axial discontinuity image representing a part on the display unit.

  According to the present invention, it is possible to perform site recognition on an axial discontinuity image and display an axial discontinuity image of a region desired by the user based on the recognition result. Thereby, it becomes possible to reduce a user's effort and a burden.

Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings. In addition, the same reference number is attached | subjected to the same component and description is abbreviate | omitted.
FIG. 1 is a block diagram showing a configuration of a medical image photographing system including a medical image display processing apparatus according to the first embodiment of the present invention. This medical imaging system includes a modality 1 for performing a medical image imaging examination on a subject, an image server 2 as a medical image display processing device according to the first embodiment of the present invention, an image display terminal 3, and an image interpretation. Terminal 4 for use. These devices 1 to 4 comply with the DICOM (Digital Imaging and Communications in Medicine) standard.

  The modality 1 includes a medical imaging apparatus such as a CT apparatus 1a, a CR (computed radiography) apparatus 1b, an MRI apparatus 1c, a PET apparatus 1d, and an ultrasonic diagnostic apparatus (US) 1e. These modalities 1a to 1e generate image data by performing an imaging inspection, and output the image data to the image server 2 together with the image supplementary information.

  The image server 2 is a server for PACS (Picture Archiving and Communication System) that stores and manages the image data acquired by the modality 1. The image server 2 outputs the image data to the image display terminal 3 in accordance with a request from the interpretation terminal 4 described later.

  As illustrated in FIG. 1, the image server 2 includes a control unit 11, a part recognition unit 12, a storage unit 13, a display processing unit 14, and an initial display slice information storage unit 15. The control unit 11, the part recognition unit 12, and the display processing unit 14 are configured by, for example, a CPU (Central Processing Unit) and a medical image display processing program according to the present embodiment.

  The control unit 11 stores the image data output from the modality 1 in the storage unit 13. Further, the control unit 11 confirms the direction of the image (axial, coronal, sagittal, etc.) represented by the input image data, and the axial direction. If it is, the image data is also output to the part recognition unit 12. Note that the orientation of the image is acquired by, for example, image supplementary information with a DICOM tag (0020,0037): Image Orientation (Patient) or (0020,0020): Patient Orientation.

  The site recognizing unit 12 indicates which site of the subject is represented in each axial location image based on a plurality of axial location images (hereinafter also referred to as “slice images”) represented by one series of image data. Recognize Then, information (part information) including the recognition result (part) is generated and stored in the storage unit 13 in association with the image data. The part may be indicated by a character string such as “Head (head)”, “Neck (neck)”, “Chest (chest)”, or the like: 1: head, 2: neck, 3: It may be indicated by a pre-coded integer value, such as chest.

  The storage unit 13 is, for example, a hard disk drive built in the image server 2, and under the control of the control unit 11, image data and its image supplementary information, part information generated by the part recognition part 12, A control program or the like for operating the part recognition unit 12 is stored. In addition to the hard disk, MO, MT, RAM, CD-ROM, DVD-ROM, or the like may be used as the recording medium. In that case, a drive device that drives the recording medium includes: It is built in the image server 2 or connected to the outside of the image server 2.

  The display processing unit 14 receives display part instruction information for instructing a part desired by the user and display slice image instruction information for instructing a slice image desired by the user from the interpretation terminal 4, and one series of images. The display part instruction information in the plurality of slice images represented by the data and the slice image of the part corresponding to the display slice image instruction information are displayed on the image display terminal 3.

  The initial display slice information storage unit 15 displays slice images to be displayed on the image display terminal 3 by the display processing unit 14 in the initial stage (for example, when the user (interpretation physician) operates the interpretation terminal 4 to start interpretation). Stores initial display slice information for determination. The initial display slice information may be a slice number (for example, “1” (represents the first slice image in one series of axial dislocation images)) or a part (for example, “the upper end of the abdomen”). Etc.). When the initial display slice information is not stored in the initial display slice information storage unit 15, the display processing unit 14 displays any one of the axial position-disrupted images in one series. Also good. The initial display slice information may be writable or updateable from the interpretation terminal 4. Further, the storage unit 13 and the initial display slice information storage unit 15 may be realized by a single recording medium.

The image display terminal 3 is a terminal device that displays an inspection image, and includes a high-definition display. Note that a plurality of axial dislocation images are schematically shown on the screen 3a of the image display terminal 3 shown in FIG.
The interpretation terminal 4 is a device used by a user (interpretation doctor) to create an interpretation report while referring to the examination image displayed on the image display terminal 3, and includes a screen 4a for displaying the interpretation report and the like. And an input device 4b such as a keyboard.

Next, the configuration and operation of the part recognition unit 12 shown in FIG. 1 will be described with reference to FIGS.
FIG. 2 is a block diagram illustrating functions of the part recognition unit 12 illustrated in FIG. As shown in FIG. 2, the part recognition unit 12 includes a feature amount calculation unit 21, a part certainty factor calculation unit 22, a score table storage unit 22a, a part determination processing unit 23, a part information storage unit 24, and a part. And a correction processing unit 25. Among these, the feature amount calculation unit 21 to the region determination processing unit 23 operate so as to tentatively determine the region represented there for each slice image, and the region correction processing unit 25 performs the determination of a plurality of slice images. Based on the region information, the region tentatively determined for each slice image is corrected. The part information of the slice image will be described later.

FIG. 3 is a flowchart showing the operation of the part recognition unit 12. When image data determined to represent an axial dislocation image by the control unit 11 (FIG. 1) is input to the region recognition unit 12, a region recognition operation described below is started.
In step S <b> 1, the image data and the image supplementary information are input to the feature amount calculation unit 21 for each slice. Here, the image supplementary information includes information ((0020, 0037): Image Orientation (Patient) or (0020, 0020): Patient Orientation) representing information indicating the direction of the image and information representing the pitch of one pixel (( 0028, 0030): Pixel Spacing), information indicating slice thickness ((0018, 0050): Slice Thickness), information indicating the number of pixels included in one row and one column ((0028, 0010): Raws , And (0028, 0011): Columns) and information ((0020, 0032): Image Position (Patient)) representing the three-dimensional coordinates of the upper left position of the image. Here, the parentheses indicate the DICOM tag and attribute name of each information.

  In step S2, the feature amount calculation unit 21 calculates a feature amount for one slice image. Here, the feature amount is a numerical value of the feature of the body part represented in the slice image. The feature amount is calculated based on the shape of the body part represented in the slice image, for example, as shown in (a) below. In addition, when the value of each pixel data (that is, the luminance of the pixel) corresponds to the characteristics of the body part (organizational properties, etc.), the value is set as shown in the following (b) and (c). The feature amount may be calculated accordingly. For example, the value of the pixel data in the CT image is determined by the CT value, and this value is a physical quantity representing the amount of radiation that has passed through the body part. The CT value of water is 0 HU, the CT value of the air region is about −1000 HU, and the CT value of the bone region is about 250 HU to 3000 HU.

(A) Circularity ρ of the whole body part The circularity ρ is calculated by the following expression (1) using the area S of the target region and the surrounding length L.
ρ = 4πS / L ² (1)
The degree of circularity ρ approaches 1.0 as the shape of the target region approaches a perfect circle, and decreases as the shape moves away from the perfect circle (for example, as the ellipticity moves away from 1). For example, when the target region is the head, the circularity is relatively high. On the other hand, when the target region is the chest or abdomen, the circularity is relatively low.
(B) Air region feature amount: (number of pixels of CT value indicating air region) / (number of pixels of the whole body)
For example, when the target region is the chest, the air region is relatively wide because the lungs are present. On the other hand, when the target area is the head, the air area is almost zero.
(C) Bone region feature amount: (number of pixels of CT value indicating bone region) / (number of pixels of entire body part)
For example, when the target region is the abdomen, the region of the bone portion relative to the entire body portion is a relatively narrow range. On the other hand, when the target region is the leg, the bone portion occupies a large proportion of the whole body portion.

  Next, in step S <b> 3, the part certainty factor calculation unit 22 calculates the part certainty factor based on the feature amount calculated by the feature amount calculation unit 21. Here, the part certainty factor is a numerical value of the possibility that the target part is “a part” (like “head” or “chest”). In this embodiment, the part certainty factor is calculated using a score table prepared in advance.

  FIG. 4 shows an example of a score table used for calculating the part certainty factor. This score table is used when obtaining the “head” -likeness, “chest” -likeness, “abdomen-like” -likeness, and “leg-likeness” -likeness based on the value of the air region feature quantity (air quantity). In addition to this, “pelvic part” may be cited as an item of the part. In addition, an item (for example, “a boundary region) between two adjacent parts or a region where a plurality of parts are mixed (a mixed region, for example, the head and neck, or the chest and abdomen). Head and neck ”or“ chest and abdomen ”).

For example, when the air region feature amount of a body part represented in a certain CT image is 60%, referring to the “40-80%” column containing 60% in the score table, the “head” of the body part The score for “part” is −1.0, the score for “chest” is 0.8, the score for “abdomen” is −0.2, and the score for “leg” is −1. It turns out that it is 0.0.
Such a score table is created for each feature amount and stored in the score table storage unit 22a. The score table may be created based on statistics, or may be intentionally created based on the experience of a user (doctor or the like).

  The part certainty factor calculation unit 22 obtains a score of each “part" likeness for each feature amount by referring to such a score table, and adds the scores for each part. The sum of the scores for each part obtained in this way is the part certainty factor.

  Referring to FIG. 3 again, in step S4, the part determination processing unit 23 temporarily sets the part confidence factor obtained in step S3 having the largest value as the part of the body part represented in the slice image. decide. In that case, when there are a plurality of parts having a large part certainty value and the difference between them is within a predetermined range (for example, within 10%), both parts may be adopted. For example, when the certainty factor of the chest and abdomen is large, the slice part is set to “chest” or “abdomen”. Alternatively, if there is an item (for example, “chest and abdomen”) indicating a boundary region or a mixed region, it may be adopted.

In step S5, the part information storage unit 24 displays the feature amount obtained in step S2, the part certainty obtained in step S3, the part provisionally determined in step S4, and part information (parts) of the slice image. Information). Note that it is not always necessary to store all the feature amounts, and only predetermined feature amounts (for example, feature amounts used in step S8 described later) may be stored.
Such operations in steps S1 to S5 are performed for all slice images included in one series (step S6).

  When the part information about all slice images is obtained, the part correction processing unit 25 arranges the part information stored in the part information storage unit 24 in the order of slices in step S7. This is because the image data generated in the modality 1 (FIG. 1) is not always transmitted to the image server 2 in the slice order. Note that the slice order is determined based on image position information ((0020, 0032): Image Position (Patient)) in the image supplementary information. Alternatively, instead of step S7, the part information storage unit 24 may store the part information in step S5 while arranging the part information in the order of slices based on the image position information.

Next, in step S8, the part correction processing unit 25 corrects the part temporarily determined for each slice using the part information of the plurality of slice images. Examples of the correction method include the following methods (1) to (3).
(1) Method Using Region Information of Adjacent Slice This method is a method of correcting a region provisionally determined for a certain slice image based on the positional relationship between adjacent slices.
The tentatively determined parts are, for example, the neck (neck) in the first to fifth slices, the head (head) in the sixth slice, the neck (neck) in the seventh to tenth slices, and the first to fifteenth. Consider the case where the slice is the chest, the 16th slice is the leg, the 17th to 20th slices are the chest, and the 21st to 30th slices are the abdomen. In this case, the slices before and after the sixth slice are the cervical part, so that the sixth slice is the head is a recognition error, and correctly the cervical part. Further, since the slices before and after the 16th slice are the chest, it is a recognition error that the 16th slice is the leg, and correctly the chest. As described above, when the part temporarily determined for a certain slice image is different from the parts of the preceding and succeeding slice images, the part of the slice image is corrected by referring to the preceding and succeeding slice images.

(2) Method Using Feature Amount This method is a method of correcting a part temporarily determined for a certain slice image based on a change in the feature amount in the body axis direction.
(A) of FIG. 5 is a figure which shows an air area | region feature-value in order of a slice position (body-axis direction), and (b) of FIG. 5 is a figure which shows the differential value of an air area | region feature-value. As shown in FIG. 5B, when the change of the air region feature amount is observed from the upper part (head side) to the lower part (leg side) of the subject, the air region feature amount starts to suddenly increase. Exists. This is the starting position of the chest. Further, when observed further toward the leg side, there is a portion where the air region feature amount starts to increase from the decrease. This is the border between the chest and abdomen. If there is a slice image in which a part other than the chest is provisionally determined between the start position of the chest and the boundary between the chest and the abdomen, the part of the slice image is corrected to the chest.

(3) Method Using Matching Curve This method is a method of correcting the part temporarily determined for each slice image by referring to the normal arrangement of the part in the subject (for example, human body).
First, as shown in FIG. 6, the part temporarily determined for each slice image is arranged in the order of slices from the upper part (head side) to the lower part (leg part side) of the subject. As shown in FIG. 6, the region recognition result shows an area where the head (Head) and the neck (Neck) appear alternately, and an area where the neck appears between the chest (Chest). Therefore, it is considered that many recognition errors are included in the tentatively determined part.

  Next, as shown in FIG. 7, a matching curve between the part recognition result shown in FIG. 6 and a reference part created in advance is searched. Here, since the parts of the human body are arranged in the order of the head, neck, chest, and abdomen, as shown in the vertical axis of FIG. Create in advance.

  When searching for a matching curve, a cost is calculated when the part recognition result and the reference part do not coincide with each other and the cost is minimized. As the search method, various methods for solving the optimization problem can be applied. The matching curve search method using dynamic programming, which is well known as one of them, will be described below.

  First, a weight map as shown in FIG. 8 is created. In FIG. 8, columns correspond to slice numbers, and rows correspond to parts. In this weight map, the tentatively determined part is set so that the weight becomes zero (bold frame region). For example, referring to FIG. 6, since the first slice is provisionally determined as the head, the value of the cell of the head (Head) of slice number 1 in the weight map is “0.0”. For other cells, a value greater than zero is set. Specifically, when the certainty factor is calculated for each slice image, a difference value between the certainty factor and the certainty factor of the provisionally determined part may be set, or a different predetermined value may be set. A value may be set.

Next, a cost map as shown in FIG. 9 is created. In FIG. 9, the cost of each cell (n, m) is set as follows. Here, n indicates a slice number, and m indicates a part number (1: Head, 2: Neck, 3: Chest, 4: Abomen).
(1,1): Value of (1,1) in the weight map (see FIG. 8)
(N, 1): (n−1, 1) value in weight map + predetermined value (1, m): (1, m−1) value in weight map + predetermined value (n, m): The minimum value among the following (i) to (iii)
(I) Value of (n-1, m-1) in the cost map
+ Value of (n, m) in weight map
(Ii) Value of (n, m-1) in the cost map
+ (N, m) value in weight map + predetermined value
(Iii) Value of (n-1, m) in the cost map
+ (N, m) value in weight map + predetermined value

Next, the minimum value around the cost map is sequentially traced from the right side to the left side. Thereby, a correspondence map between slice numbers and parts is created.
As shown in FIG. 7, the part is corrected by replacing the provisionally determined part with the corresponding part in the reference part based on the matching curve thus obtained.

  Referring to FIG. 3 again, in step S8, the part correction processing unit 25 outputs the corrected part information as image supplementary information to the storage unit 13 for storage. The part information output from the part recognition unit 12 may be managed by an image information database, or may be written as a tag in the image data already stored in the storage unit 13.

  As described above, in the present embodiment, after performing site recognition for each of a plurality of slice images included in one series, the site information of each slice is obtained using the correlation of the site information regarding the multiple slices. Correct it. Thus, the advantage of performing site recognition through two stages is as follows. That is, since the part recognition process can be started in the order of the input slices without waiting for all the image data of one series to be input to the server 2, a part recognition result can be obtained at a relatively high speed. Further, since the part recognition result obtained for each slice is corrected based on the three-dimensional information of the subject represented by a set of a plurality of slice images, a large part recognition error can be reduced. Therefore, it is possible to perform efficient and accurate site recognition.

  Here, in this embodiment, the part recognition part 12 (FIG. 2) is performing the part recognition process about all the input slice images. However, before starting the part recognition process, the DICOM tag may be referred to and the part recognition process may be performed only on the slice image in which the information ((0018, 0015): Body Part) indicating the imaging part does not exist. . This is because a part may be attached at the imaging stage.

  Alternatively, the part recognizing unit 12 may perform the part recognizing process on the continuously input slice images while thinning out at a predetermined slice interval. In that case, it becomes possible to speed up the processing as a whole. Furthermore, the part recognizing unit 12 may perform the part recognizing process only for a predetermined range of slice images input continuously. For example, when the diagnosis target is the abdomen of the subject, it is only necessary to recognize the start region of the abdomen (or the mixed region of the chest and abdomen). In that case, for example, a region recognition process is performed for a range that is clearly considered to be a leg as determined from information (DICOM tag (0020, 0032): Image Position (Patient)) representing the three-dimensional coordinates of the slice image. May be omitted.

In the present embodiment, a score table is used when the body part represented in the slice in steps S3 and S4 is provisionally determined. Instead, a machine learning method such as a neural network is used. Then, the part may be recognized.
Here, a method for provisionally determining a site using a neural network will be described.
As shown in FIG. 10, the feature quantities of the body part represented in the slice image (for example, the feature quantities (a) to (c) described in step S2) are input to the neural network. Then, the neural network is trained so that 1 is output for a portion matching the portion represented in the slice image and zero is output for other portions. For example, when the head is represented in the slice image, the output of “Head” is set to 1, and the outputs of “Neck”, “Chest”, and “Abdomen” are set to zero. By using the neural network learned in this way, a part corresponding to the input feature amount is acquired.

  Moreover, in this embodiment, after the site | part recognition part 12 performs site | part recognition about each of the several slice image contained in 1 series, the site | part information of each slice is used using the correlation of the site | part information regarding several slices. Although it is supposed to be corrected, the part may be recognized by comparing the template image and the slice image (see Patent Document 2 described above).

Next, the operation of the display processing unit 14 shown in FIG. 1 will be described with reference to FIG.
FIG. 11 is a flowchart showing the operation of the display processing unit 14 shown in FIG.
In step S11, the display processing unit 14 reads the initial display slice information from the initial display slice information storage unit 15, reads the image data for displaying the slice image determined by the initial display slice information, and reads the image data. A slice image based on the obtained image data is displayed on the image display terminal 3.

  In step S12, the display processing unit 14 checks whether the display part instruction information is received from the interpretation terminal 4, and if the display part instruction information is received from the interpretation terminal 4, the process proceeds to step S13. If the display part instruction information has not been received from the interpretation terminal 4, the process proceeds to step S14.

  In step S <b> 13, the display processing unit 14 reads image data for displaying a slice image of a part corresponding to the display part instruction information from the storage unit 13, and displays a slice image based on the read image data on the image display terminal 3. Let

  On the other hand, in step S14, the display processing unit 14 checks whether or not the display slice instruction information is received from the interpretation terminal 4, and if the display slice instruction information is received from the interpretation terminal 4, the process is stepped. The process moves to S15, and if the display slice instruction information has not been received from the interpretation terminal 4, the process returns to Step S12.

  In step S15, the display processing unit 14 reads image data for displaying a slice image corresponding to the display slice instruction information from the storage unit 13, and causes the image display terminal 3 to display a slice image based on the read image data.

  Next, the operation of the display processing unit 14 will be described with a specific example. Here, it is assumed that one series of image data stored in the storage unit 13 includes first to twentieth image data. Then, the part recognition unit 12 recognizes that the first to tenth image data is image data representing a slice image of the chest, and the first to twentieth image data is image data representing a slice image of the abdomen. Shall be. Assume that the initial display slice information storage unit 15 stores the slice number “15” as the initial display slice information.

In this case, the display processing unit 14 initially reads the fifteenth image data (representing an abdominal slice image) from the storage unit 13 and causes the image display terminal 3 to display a slice image based on the read image data ( (See step S11).
Here, when the user (interpretation doctor) presses a predetermined first key (for example, a cursor upward movement key (“↑” key) or the like) (see step S12), the display processing unit 14 is currently displayed. The image data (in this case, the tenth image data) for displaying the slice image of the lower end of the chest, which is an adjacent part at the upper part of the abdomen, which is the part of the slice image being read, is read from the storage unit 13 and read. A slice image (chest) based on the image data is displayed on the image display terminal 3 (see step S13).

  Next, when the user presses a predetermined second key (for example, a cursor leftward movement key (“←” key) or the like) (see step S14), the display processing unit 14 displays the slice image currently displayed. Image data (in this case, ninth image data) for displaying the slice image one level above (based on the tenth image data) is read from the storage unit 13, and a slice image (chest) based on the read image data ) Is displayed on the image display terminal 3 (see step S15).

  Next, when the user presses a predetermined third key (for example, a cursor downward movement key (“↓” key)) (see step S12), the display processing unit 14 displays the slice image currently displayed. The image data for displaying the slice image of the upper end of the abdomen, which is an adjacent part in the lower part of the chest, which is the part of this part, is read from the storage unit 13 and is based on the read image data The slice image (abdomen) is displayed on the image display terminal 3 (see step S13).

  Next, when the user presses a predetermined fourth key (for example, a cursor right movement key (“→” key)) (see step S14), the display processing unit 14 displays the slice image currently displayed. Image data (in this case, twelfth image data) for displaying a slice image immediately below (based on the eleventh image data) is read from the storage unit 13, and a slice image (abdomen) based on the read image data ) Is displayed on the image display terminal 3 (see step S15).

  As described above, according to the present embodiment, site recognition is performed on each of a plurality of slice images included in one series, and a slice image of a site desired by the user (interpretation physician) based on the site recognition result. Can be displayed on the image display terminal 3. Thereby, it becomes possible to reduce a burden of a user (interpretation doctor).

  Note that the part recognition unit 12 may create identification information for identifying an axial discontinuity image representing the upper end part and / or the lower end part of each part. For example, one series of image data stored in the storage unit 13 includes first to twentieth image data, and the first to tenth image data are image data representing slice images of the chest, When the 11th to 20th image data is image data representing an abdominal slice image, the region recognition unit 12 is the image data in which the 10th image data represents the lower end of the chest, and the 11th image data is Identification information for identifying image data representing the upper end of the abdomen may be created. This identification information may be added to the tenth and eleventh image data as image supplementary information or a tag, or may be stored in the storage unit 13 as data independent of the image data. If the identification information is created in this way, the tenth and eleventh image data representing the slice images of the lower end of the chest and the upper end of the abdomen can be read out at high speed.

  In the present embodiment, the part recognition unit 12 performs a part recognition process on the image data directly input from the modality 1 to the image server 2. However, the part recognition process may be performed by reading the image data once generated in the modality 1 and temporarily stored in the recording medium into the image server 2. Further, when the display processing unit 14 displays an image on the image display terminal 3, a part recognition process may be performed.

Next, a medical image display processing apparatus according to the second embodiment of the present invention will be described.
FIG. 12 is a block diagram illustrating a configuration of a medical image photographing system including a medical image display processing apparatus according to the present embodiment. This medical image photographing system includes a modality 1, an image server 60 as a medical image display processing apparatus according to the second embodiment of the present invention, an image display terminal 3, and an interpretation terminal 4. These devices 1, 3, 4, and 60 are compliant with the DICOM standard.

  The image server 60 is a PACS server that stores and manages image data acquired by the modality 1. The image server 60 outputs the image data to the image display terminal 3 in accordance with a request from the interpretation terminal 4 described later.

  As shown in FIG. 12, the image server 60 includes a schema (Schema) in addition to the control unit 11, the part recognition unit 12, the storage unit 13, the display processing unit 14, and the initial display slice information storage unit 15 described above. ) A display processing unit 16 is further provided. The schema display processing unit 16 includes, for example, a CPU and a control program.

  The schema display processing unit 16 causes the image display terminal 3 to display a schema that schematically represents a human body. The user (interpretation doctor) can select a desired part in the schema displayed on the image display terminal 3.

Next, operations of the display processing unit 14 and the schema display processing unit 16 shown in FIG. 12 will be described with reference to FIG.
FIG. 13 is a flowchart showing operations of the display processing unit 14 and the schema display processing unit 16 shown in FIG.

In step S <b> 21, the schema display processing unit 16 displays the schema on the image display terminal 3.
In step S22, the display processing unit 14 reads the initial display slice information from the initial display slice information storage unit 15, reads the image data for displaying the slice image determined by the initial display slice information, and reads the image data. A slice image based on the obtained image data is displayed on the image display terminal 3.

  In step S23, the display processing unit 14 checks whether the display part instruction information is received from the interpretation terminal 4, and if the display part instruction information is received from the interpretation terminal 4, the process proceeds to step S24. If the display part instruction information has not been received from the interpretation terminal 4, the process proceeds to step S25.

  In step S24, the display processing unit 14 reads image data for displaying a slice image of a part corresponding to the display part instruction information from the storage unit 13, and displays the slice image based on the read image data on the image display terminal 3. Let

  On the other hand, in step S25, the display processing unit 14 checks whether or not the display slice instruction information is received from the interpretation terminal 4, and if the display slice instruction information is received from the interpretation terminal 4, the process is performed as a step. The process moves to S26, and if the display slice instruction information is not received from the interpretation terminal 4, the process returns to Step S23.

  In step S <b> 26, the display processing unit 14 reads image data for displaying a slice image corresponding to the display slice instruction information from the storage unit 13, and causes the image display terminal 3 to display a slice image based on the read image data.

  Next, operations of the display processing unit 14 and the schema display processing unit 16 will be described with specific examples. Here, it is assumed that one series of image data stored in the storage unit 13 includes first to 45th image data. The site recognition unit 12 includes image data in which the first to fifth image data represent cervical slice images, the sixth to twenty-fifth image data represent image slice images of the chest, Assume that the region recognition is that the 26th to 35th image data is image data representing a thoracoabdominal slice image and the 36th to 45th image data is image data representing an abdominal slice image. The initial display slice information storage unit 15 stores “the last slice image in one series of slice images” as initial display slice information (here, corresponding to a slice image (abdominal part) based on the 45th image data). Is stored.

  FIG. 14 is a diagram illustrating an example of a display screen of the image display terminal 3. As shown in FIG. 14, a schema 72 is displayed in the area 71 of the display screen 3a of the image display terminal 3 (see step S21). In the present embodiment, the region is recognized that one series of image data stored in the storage unit 13 is obtained by imaging the neck to the abdomen of the subject. ~ The abdomen is highlighted (here, hatched). Further, the boundary between the head and neck of the schema 72, the boundary between the neck and the chest, the boundary between the chest and the abdomen, the boundary between the chest abdomen and the abdomen, the boundary between the abdomen and the pelvis, and the pelvis Buttons 73a to 73f are respectively displayed on the right side of the boundary between the leg and the leg. In the present embodiment, the boundary between the neck and the chest of the subject, the boundary between the chest and the abdomen, and the boundary between the chest and the abdomen are imaged, so that they are located in the right direction in these drawings. The buttons 73b to 73d are highlighted (in this case, black). On the other hand, in the present embodiment, since the boundary between the head and neck of the subject, the boundary between the abdomen and the pelvis, and the boundary between the pelvis and the leg are not imaged, the buttons 73a, 73e, 73f is not highlighted.

  In the area 74 of the display screen 3a of the image display terminal 3, the initial display slice information is stored in the initial display slice information storage unit 15 as “the last slice image of one series of slice images” (here, the 45th slice image). Are stored), the slice image (abdominal part) based on the 45th image data is displayed (see step S22).

  Here, when the user (reading doctor) clicks the button 73c (position in the right direction in the figure of the boundary between the chest and the chest abdomen) (see step S23), the display processing unit 14 slices the lower end of the chest. Is read from the storage unit 13 and a slice image (chest) based on the read image data is displayed in the area 74 of the display screen 3a of the image display terminal 3. (See step S24).

  FIG. 15 is a diagram showing the display screen 3a of the image display terminal 3 at this time. As shown in FIG. 15, a slice image (chest) based on the 25th image data is displayed in the area 74 of the display screen 3 a of the image display terminal 3. When the button 73c is clicked, the display processing unit 14 reads out the image data (here, the 26th image data) for displaying the slice image of the upper end of the thoracoabdominal region from the storage unit 13, and reads out the image data. A slice image (chest and abdomen) based on the image data may be displayed in the area 74 of the display screen 3a of the image display terminal 3.

  In addition, when the user clicks the button 73d (positioned in the right direction in the figure of the boundary between the chest and abdomen) (see step S23), the display processing unit 14 displays a slice image of the lower end of the chest and abdomen. Image data (here, the 35th image data) is read from the storage unit 13, and a slice image (chest and abdomen) based on the read image data is displayed in the region 74 of the display screen 3a of the image display terminal 3 (step (See S24).

  FIG. 16 is a diagram showing the display screen 3a of the image display terminal 3 at this time. As shown in FIG. 16, a slice image (abdomen) based on the 35th image data is displayed in the area 74 of the display screen 3a of the image display terminal 3. When the button 73d is clicked, the display processing unit 14 reads out the image data (here, the 36th image data) for displaying the slice image of the upper end of the abdomen from the storage unit 13 and reads it out. A slice image (abdomen) based on the image data may be displayed in the area 74 of the display screen 3a of the image display terminal 3.

  The operation when a predetermined key (for example, a cursor movement key) is pressed is the same as that of the first embodiment described above.

  As described above, according to the present embodiment, the schema 72 is displayed in the area 71 of the display screen 3 a of the image display terminal 3, and the slice image representing the part selected by the user in the schema 72 is displayed on the image display terminal 3. It can be displayed in the area 74 of the display screen 3a. Thereby, it becomes possible to reduce a burden of a user (interpretation doctor).

  The present invention is used in a medical image display processing device for displaying an axial dislocation image based on image data acquired by a medical imaging modality, and a medical image display processing program used in such a device. Is possible.

1 is a diagram illustrating a configuration of a medical image photographing system including a medical image display processing device according to a first embodiment of the present invention. It is a block diagram which shows the function of the site | part recognition part shown in FIG. It is a flowchart which shows operation | movement of the site | part recognition part shown in FIG. It is a figure which shows the site | part certainty factor score table using the air area | region feature-value. It is a figure which shows an air area | region feature-value and its differential value. It is the figure which has arrange | positioned the part (part recognition result) provisionally determined in order of a slice. It is a figure which shows the matching curve between a site | part recognition result and a reference site | part. It is a weight map used for the search of the matching curve by dynamic programming. It is a cost map used for the search of the matching curve by dynamic programming. It is a figure for demonstrating the temporary determination method of the site | part using a neural network. It is a flowchart which shows operation | movement of the display process part shown in FIG. It is a figure which shows the structure of the medical image imaging system containing the medical image display processing apparatus which concerns on the 2nd Embodiment of this invention. 13 is a flowchart showing operations of a display processing unit and a schema display processing unit 16 shown in FIG. It is a figure which shows the example of the image displayed on the image display terminal shown in FIG. It is a figure which shows the example of the image displayed on the image display terminal shown in FIG. It is a figure which shows the example of the image displayed on the image display terminal shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Modality 1a Computer tomography (CT) apparatus 1b Computer radiography (CR) apparatus 1c Magnetic resonance imaging (MRI) apparatus 1d Positron tomography (PET) apparatus 1e Ultrasound imaging (US) apparatus 2, 60 Image server 3 Image display Terminal 3a, 4a Screen 4 Interpretation terminal 4b Input device 11 Control unit 12 Site recognition unit 13 Storage unit 14 Display processing unit 15 Initial display slice information storage unit 16 Schema display processing unit 21 Feature amount calculation unit 22 Site certainty calculation unit 22a Score table storage unit 23 Site determination processing unit 24 Site information storage unit 25 Site correction processing unit 71, 74 Area 72 Schema 73a-73f Button

Claims (12)

An apparatus for displaying an axial dislocation image on a display unit based on image data representing a series of axial discontinuity images obtained by imaging and inspecting a subject,
A part recognizing means for recognizing the part of the body part represented in each of the series of axial discontinuity images;
The display unit displays one axial discontinuity image included in one series of axial discontinuity images on the display unit, and when receiving a region change instruction, the display unit Display processing means for displaying on the display unit an axial discontinuity image representing a region different from the region of the axial discontinuity image displayed on
A medical image display processing apparatus. The part recognizing means creates identification information for identifying two axial dislocation images located at the boundary between two adjacent parts,
When the display processing unit receives a part change instruction, by using the identification information, the display unit displays an axial dislocation image located at a boundary part of a part desired by the user.
The medical image display processing apparatus according to claim 1.   The axial discontinuity image according to preset initial display information in one series of axial discontinuity images is displayed on the display unit when the display processing means has not received a part change instruction. The medical image display processing apparatus according to 1 or 2. It further comprises a schema display processing means for displaying a schema representing a human body for selecting a part on the display unit,
The display processing means causes the display unit to display an axial discontinuity image representing a part selected in the schema;
The medical image display processing apparatus according to any one of claims 1 to 3. The part recognition means
A part determining means for tentatively determining a part of the body part represented in each of the series of axial position discontinuity images;
A part correcting means for correcting a part tentatively determined for at least one axial discontinuity image by the part determining means based on information on one series of axial discontinuity images;
The medical image display processing apparatus of any one of Claims 1-4 containing these.   The region recognition means includes a head, a neck, a chest, an abdomen, a pelvis, a leg, and two adjacent boundary regions among them or a mixed region including a plurality of regions therein The medical image display processing apparatus according to claim 1, wherein at least two of the two are recognized as parts. A program for displaying an axial dislocation image on a display unit based on image data representing a series of axial discontinuity images obtained by imaging and inspecting a subject,
A procedure (a) for recognizing a part of a body part represented in each of a series of axial dislocation images;
A step (b) of displaying one axial position discontinuity image included in one series of axial position discontinuity images on the display unit;
When receiving a part change instruction, based on the recognition result in step (a), the display unit displays an axial discontinuity image representing a part different from the part of the axial discontinuity image displayed on the display unit. Step (c);
A medical image display processing program for causing the CPU to execute The step (a) includes creating identification information for identifying two axial position dislocation images located at a boundary portion between two adjacent parts,
When the step (c) receives a part change instruction, the display part displays an axial position dislocation image located at a boundary part of a part desired by the user by using the identification information.
The medical image display processing program according to claim 7.   The medical image according to claim 7 or 8, wherein the step (b) includes causing the display unit to display an axial discontinuity image corresponding to preset initial display information in one series of axial discontinuity images. Display processing program. The CPU further executes a procedure (b ′) for displaying a schema representing a human body on the display unit for site selection,
The medical image display processing program according to any one of claims 7 to 9, wherein the procedure (c) includes a procedure for causing the display unit to display an axial discontinuity image representing a part selected in the schema. Step (a)
A procedure (a1) for tentatively determining a part of the body part represented in each of the series of axial discontinuity images;
A procedure (a2) for correcting a portion tentatively determined for at least one axial dislocation image in the procedure (a1) based on information on one series of axial discontinuity images;
The medical image display processing program according to claim 7, comprising: Step (a) is a mixed region including a head, neck, chest, abdomen, pelvis, leg, and two adjacent boundary regions among them or a plurality of regions therein. The medical image display processing program of any one of Claims 7-11 including recognizing at least 2 of these as a site | part.