JP5595750B2 - Medical image processing apparatus and medical image processing program - Google Patents

Description

  The present invention relates to a medical image processing apparatus that generates medical image data representing a lesioned part, and a medical image processing program.

  A subject is imaged using a medical image imaging apparatus such as an X-ray CT apparatus or an MRI apparatus, and diagnosis using medical image data obtained by the imaging is performed. For example, in preoperative diagnosis including screening examination, CT colonography is performed using medical image data.

  As an image display method in CT colonography, there is a developed image display (Virtual Grossology: VGP) or a virtual endoscope display (Virtual Endoscope: VE).

  The developed image represents a state in which luminal bodies such as the large intestine, blood vessels, and stomach are cut open. That is, the developed image is an image representing a state in which the inner surface of the luminal body is developed, and represents the inner surface of the luminal body.

  The virtual endoscopic image is a virtual endoscopic image that three-dimensionally represents the inner surface of a lumen body such as a large intestine or a blood vessel. A method for displaying an image using a virtual endoscopic image is called fly-through. Fly-through is a technique for observing the inner surface of a luminal body such as the large intestine or blood vessel with the same line of sight as an endoscope.

  As an example using a fly-through, there is a method described in Patent Document 1. The method described in Patent Literature 1 obtains a normal vector for each voxel on the surface of the observation target region. Then, an angle formed by the normal vector serving as a reference and each normal vector is obtained, and the dispersion of the angle is obtained as the degree of surface unevenness. In this way, a lesion candidate is detected by obtaining the surface roughness.

JP 2007-14483 A

  Conventionally, a surgeon such as a doctor observes the surface of the inner wall of the large intestine represented in a developed image or a virtual endoscopic image, and determines the presence or absence of a tumor based on the shape of the surface. In the case of a bulging type tumor, the operator can determine the presence or absence of the tumor relatively easily by observing the shape of the surface. Raised tumors include pedunculated tumors, subcarpal tumors, and sessile tumors.

  However, when the tumor is a non-protruding type tumor (surface type tumor), it is difficult for the operator to determine the presence or absence of the tumor only by observing the surface shape. Non-protruding type tumors (surface type tumors) include surface type tumors, surface type tumors, and surface depression type tumors. When a surface-bulged tumor is formed, a low bump is formed on the surface of the inner wall, but the shape of the tumor is almost flat. Even if a flat surface type tumor is formed, no bulge is formed on the surface of the inner wall, and the shape of the tumor is flat. When a surface-recessed tumor is formed, a recess is formed on the surface of the inner wall. Thus, the non-bulged tumor does not have a shape protruding from the surface of the inner wall of the large intestine. Even a surface bulging tumor has a low degree of bulging and the shape of the tumor is almost flat. Therefore, it is difficult to determine the presence or absence of a tumor simply by observing the shape of the surface. In general, non-protruding tumors are less likely to be found in screening tests than bulging tumors.

  Non-bulged tumors, unlike bulged tumors, often become cancerous without going through a multi-stage carcinogenic process. Therefore, it is important to detect a non-lifting tumor early and remove it early.

  The present invention solves the above problem, and provides a medical image processing apparatus and a medical image processing program capable of generating information that makes it easy to grasp the shape of a non-raised lesion. To do.

According to the first aspect of the present invention, there is provided contour extracting means for receiving volume data representing a luminal body and obtaining a first contour of an inner wall of the luminal body based on the volume data, and the first contour. Based on a pixel value of a pixel in a region outside the inner wall with respect to a reference, a lesion part extraction means for obtaining a position of a lesion part, and based on the position of the lesion part and the first contour, the lesion part Estimating means for estimating the second contour of the inner wall in the case where there is not, and a ridge level calculating means for determining the degree of unevenness of the lesion based on the first contour and the second contour; Image generating means for generating image data representing the luminal body based on the volume data, and display control means for displaying on the display means information indicating the degree of unevenness and an image based on the image data , Above The transformation part extracting unit translates the first contour from the position of the inner wall toward the outer region by a predetermined distance, and based on the pixel value of the pixel on the first contour after the translation. A medical image processing apparatus for obtaining a location where the lesioned part exists .
According to the seventh aspect of the present invention, the computer receives volume data representing a luminal body, and obtains a first contour of an inner wall of the luminal body based on the volume data, and the first Based on the pixel value of the pixel in the area outside the inner wall with reference to the contour of the lesion, based on the lesion extraction function to determine the position of the lesion, the position of the lesion and the first contour, Based on the estimation function for estimating the second contour of the inner wall when the lesion is not present, and the first contour and the second contour, the ridge level calculation for determining the degree of unevenness of the lesion A display controller for displaying a function, an image generation function for generating image data representing the lumen body based on the volume data, and information indicating the degree of unevenness and an image based on the image data When, by the execution, predetermined distance from the position of the inner wall toward the outer region, the first by translating the contour, based on the pixel values of the pixels on the first contour after the translation A medical image processing program characterized by obtaining a location where the lesion is present .

  According to the present invention, the degree of unevenness of the lesioned part is obtained based on the first contour of the inner wall and the estimated second contour, and the degree of unevenness of the lesioned part is displayed together with the image. Thus, by obtaining the degree of unevenness of the lesioned part, it is possible to provide information that makes it easy to grasp the shape of the non-raised lesioned part.

1 is a block diagram illustrating a medical image processing apparatus according to an embodiment of the present invention. It is sectional drawing which shows the cross section along the longitudinal direction of large intestine. It is sectional drawing which shows the cross section along the longitudinal direction of large intestine. It is sectional drawing which shows the cross section along the longitudinal direction of large intestine. It is a figure which shows an example of the correspondence of a protruding level and a color. It is a figure which shows an expansion | deployment image and a virtual endoscopic image. It is sectional drawing which shows the cross section along the longitudinal direction of large intestine. It is a flowchart which shows a series of operation | movement by the medical image processing apparatus which concerns on this embodiment.

  A medical image processing apparatus according to an embodiment of the present invention will be described with reference to FIG. For example, a medical image photographing apparatus 90 is connected to the medical image processing apparatus 1 according to this embodiment.

(Medical image capturing device 90)
As the medical image capturing apparatus 90, an image capturing apparatus such as an X-ray CT apparatus or an MRI apparatus is used. The medical image photographing device 90 generates medical image data by photographing a photographing region including an observation target. For example, the medical image photographing apparatus 90 generates volume data by photographing a three-dimensional photographing region. One example of an observation object is a luminal body. The lumen body includes a tubular tissue and a bag-shaped tissue. One example of tubular tissue is the large intestine or blood vessels. One example of a bag-like tissue is the stomach. For example, when the tubular tissue is an observation target, the medical image capturing apparatus 90 generates volume data representing the imaging region including the tubular tissue by capturing the imaging region including the tubular tissue.

  For example, an X-ray CT apparatus as the medical image imaging apparatus 90 generates CT image data in a plurality of cross sections having different positions by imaging a three-dimensional imaging region. The X-ray CT apparatus generates volume data using a plurality of CT image data. When the large intestine is an observation target as an example of the tubular tissue, the X-ray CT apparatus generates volume data representing an imaging region including the large intestine by imaging a three-dimensional imaging region including the large intestine. The medical image photographing device 90 outputs the volume data to the medical image processing device 1. Hereinafter, the large intestine will be described as an example of an observation target, and the tumor will be described as an example of a lesion.

(Medical image processing apparatus 1)
The medical image processing apparatus 1 includes an image storage unit 2, an image processing unit 3, an image generation unit 4, a display control unit 5, a user interface (UI) 6, and a condition storage unit 7.

(Image storage unit 2)
The image storage unit 2 stores medical image data output from the medical image photographing device 90. For example, the image storage unit 2 stores volume data representing an imaging region including the large intestine.

  The medical image processing apparatus 1 may generate volume data without the medical image capturing apparatus 90 generating volume data. In this case, the medical image photographing apparatus 90 outputs a plurality of medical image data (for example, CT image data) to the medical image processing apparatus 1. The medical image processing apparatus 1 receives a plurality of medical image data generated by the medical image photographing apparatus 90. The image storage unit 2 stores a plurality of medical image data. The image generation unit 4 reads a plurality of medical image data from the image storage unit 2 and generates volume data based on the plurality of medical image data. The image storage unit 2 stores the volume data generated by the image generation unit 4.

(Image processing unit 3)
The image processing unit 3 includes an observation target extraction unit 31, a contour extraction unit 32, a lesion extraction unit 33, an estimation unit 34, a bulge level calculation unit 35, and a conversion unit 36. The image processing unit 3 reads the volume data from the image storage unit 2 and obtains the degree of unevenness (bump level) of the lesioned part based on the volume data. A tumor is taken as an example of a lesion. In the case where the large intestine is the observation target, the image processing unit 3 obtains the degree of unevenness (protrusion level) of the tumor present in the large intestine. Specifically, the image processing unit 3 extracts the contour of the inner wall of the large intestine from the volume data, obtains the position of the tumor, and estimates the normal contour of the inner wall. The image processing unit 3 obtains the degree of unevenness of the tumor (bump level) based on the extracted inner wall contour and the estimated normal inner wall contour. A normal inner wall is an inner wall in a state where no tumor is formed in the large intestine.

  Processing performed by the image processing unit 3 will be described with reference to FIGS. 2 to 4 are sectional views showing a section along the longitudinal direction of the large intestine. FIG. 5 is a diagram illustrating an example of a correspondence relationship between the bulge level and the color.

(Observation target extraction unit 31)
The observation target extraction unit 31 reads volume data from the image storage unit 2 and extracts volume data representing the large intestine from the volume data. A known method can be used for this extraction. For example, the observation target extraction unit 31 extracts volume data representing the large intestine by executing a known segmentation process. As an example, the observation target extraction unit 31 extracts volume data representing the large intestine based on pixel values such as CT values. The observation target extraction unit 31 outputs volume data representing the large intestine to the contour extraction unit 32 and the lesion extraction unit 33.

(Outline extraction unit 32)
The contour extraction unit 32 obtains the contour of the inner wall (intestinal wall) of the large intestine in the three-dimensional space based on the volume data representing the large intestine. A known method can be used as a method for obtaining the contour. For example, the contour extraction unit 32 obtains the contour of the inner wall of the large intestine based on pixel values such as CT values. Hereinafter, the contour of the inner wall obtained by the contour extraction unit 32 may be referred to as a first contour. The contour extraction unit 32 may obtain the contour of the outer wall of the large intestine in the three-dimensional space based on the volume data representing the large intestine. The contour extraction unit 32 may obtain the position of the heart line of the large intestine in the three-dimensional space based on the volume data representing the large intestine. For example, the contour extracting unit 32 may obtain a line passing through the center of the large intestine along the longitudinal direction of the large intestine as a heart line. The contour extracting unit 32 includes first inner wall information (coordinate information) indicating the position of the inner wall contour (first contour), outer wall information indicating the position of the outer wall contour, and core line information indicating the position of the core wire. Are output to the lesioned part extracting unit 33 and the raised level calculating unit 35.

  FIG. 2 shows an example of the contour extracted by the contour extraction unit 32. For example, FIG. 2A shows an inner wall contour 10 (first contour) and an outer wall contour 11. As an example, the tumor 12 exists in a region between the inner wall and the outer wall (region inside the colon wall). The thickness T corresponds to the length between the contour 10 of the inner wall and the contour 11 of the outer wall. As an example, the thickness T is about 2 mm.

(Lesion extraction unit 33)
The lesion extraction unit 33 obtains pixel values (for example, CT values) in a region outside the inner wall of the large intestine. That is, the lesion part extraction part 33 calculates | requires the pixel value in the structure | tissue inside a colon wall. The lesion extraction unit 33 obtains the tumor position based on the pixel value. Specifically, the lesion extraction unit 33 obtains a pixel value at a position away from the inner wall of the large intestine by a predetermined distance in the direction of the outer wall. The lesion extraction unit 33 obtains the tumor position based on the pixel value.

  Tumors are not formed on the surface of the inner wall of the large intestine, but in the mucosal layer closest to the surface of the inner wall. Therefore, it is possible to determine whether or not a tumor is formed in the large intestine by obtaining the pixel value in the mucosal layer. However, depending on the spatial resolution of the medical image capturing apparatus 90, it is difficult to obtain the position of the adhesive layer based on the pixel value.

  Therefore, the lesion extraction unit 33 defines a position that is separated from the inner wall contour (first contour) indicated by the first inner wall information by a predetermined distance in the direction toward the outer wall as a position to search for a tumor. For example, as shown in FIG. 2A, the lesion part extraction unit 33 is configured to move a predetermined moving distance in a direction (X direction in the drawing) from the inner wall contour 10 (first contour) to the outer wall contour 11; The contour 10 is translated. The lesion extraction unit 33 defines the moved contour 10 as a search line 13 (a line indicated by a broken line). An example of the predetermined movement distance is a distance corresponding to one pixel (one pixel). That is, the lesion extraction unit 33 moves all points on the contour 10 in the direction (X direction) toward the contour 11 by the same distance (a distance corresponding to one pixel), and uses the destination contour 10 as the search line 13. Define. The search line 13 is a virtual line obtained by translating the contour 10 of the inner wall. Therefore, the shape of the search line 13 is the same as the shape of the inner wall contour 10.

  As an example, the length of one side of one pixel is 0.25 mm. When an X-ray CT apparatus is used, 0.25 mm corresponds to the minimum slice thickness. The lesion extraction unit 33 translates the contour 10 by 0.25 mm in the direction from the inner wall contour 10 toward the outer wall contour 11, and defines the moved position as the search line 13. It is presumed that a mucosal layer is present at a position 0.25 mm away from the contour 10 of the inner wall. Therefore, it is estimated that a mucous membrane layer exists at the position of the search line 13. The moving distance is not limited to 0.25 mm, and may be a distance corresponding to several pixels.

  The lesion extraction unit 33 obtains a pixel value (CT value) of each pixel on the search line 13. The lesion extraction unit 33 obtains the tumor position based on the pixel value. For example, the lesion extraction unit 33 determines whether or not the pixel value of each pixel on the search line 13 is included in a value within a predetermined range set in advance. The value in the predetermined range is a reference for determining whether or not the tissue represented by the pixel is a lesion (tumor). The value in the predetermined range is a value corresponding to the pixel value (CT value) of the lesioned part (tumor). The value in the predetermined range is stored in advance in a storage unit (not shown). The operator can arbitrarily change the value in the predetermined range using the operation unit 62.

  The lesion extraction unit 33 determines that a pixel whose pixel value is included in a value within a predetermined range is a pixel included in the tumor. The lesion extraction unit 33 determines whether or not a pixel value is included in a value within a predetermined range for each pixel on the search line 13. Thereby, the lesion part extraction unit 33 obtains the position of the tumor on the search line 13. The lesion part extraction unit 33 outputs search line information (position information) indicating the position of the search line 13 and lesion part information (coordinate information) indicating the position of the lesion part (tumor) to the estimation unit 34.

  As described above, the pixel value in the region close to the surface of the inner wall of the large intestine can be obtained to obtain the position of the tumor. That is, it is estimated that a mucous membrane layer exists at the position of the search line 13. Therefore, the pixel value of each pixel on the search line 13 is estimated to be the mucosal layer pixel value. As a result, based on the pixel value of each pixel on the search line 13, it can be determined whether or not a tumor is formed in the adhesive layer.

(Estimation unit 34)
The estimation unit 34 obtains the outline of the normal inner wall of the large intestine. The estimation unit 34 receives the search line information and the lesion part information, and excludes the range where the tumor exists in the search line 13. The estimation unit 34 is a search line 13 where there is no tumor, and connects adjacent search lines 13. For example, as illustrated in FIG. 2A, the estimation unit 34 excludes a range where the tumor 12 exists in the search line 13. The estimation unit 34 connects the search lines 13 extending on both sides of the tumor 12. As shown in FIGS. 2A and 2B, the estimation unit 34 calculates an end point A of the search line 13 extending to one side of the tumor 12 and an end point B of the search line 13 extending to the other side of the tumor 12. The connecting line 13A is obtained by connecting with a straight line.

  As shown in FIG. 2 (b), the estimation unit 34 performs the predetermined movement distance (for example, a distance of one pixel) in the direction from the outer wall contour 11 toward the inner wall contour 10 (Y direction in the drawing). Then, the connecting line 13A is moved in parallel. The Y direction is the opposite direction to the X direction. As illustrated in FIG. 2C, the estimation unit 34 defines the contour after movement as the contour 13B. That is, the estimating unit 34 translates the connecting line 13A in the direction opposite to the direction in which the contour 10 is moved, by the same distance as the distance in which the contour 10 is moved. In this way, the estimation unit 34 obtains the outline 13B of the normal inner wall of the large intestine.

  As described above, the estimation unit 34 generates the connection line 13A by connecting the search line 13 from which the tumor range is removed, except the range where the tumor 12 exists from the search line 13. The estimation unit 34 obtains the normal outline 13B of the large intestine by returning the connecting line 13A to the position before the parallel movement.

  Hereinafter, the outline 13B of the inner wall estimated by the estimation unit 34 may be referred to as a second outline. The estimation unit 34 outputs the second inner wall information (coordinate information) indicating the position of the inner wall contour 13 </ b> B (second contour) to the uplift level calculation unit 35.

  FIG. 3 shows an example of the estimated contour 13B. FIG. 3 shows an inner wall contour 10 (first contour), an outer wall contour 11, and an estimated inner wall contour 13B (second contour). The contour 10 of the inner wall is a contour obtained by the contour extracting unit 32. The contour 13 </ b> B is a contour obtained by the estimation unit 34. For example, the tumor 12 and the tumor 14 exist in a region between the inner wall and the outer wall (region inside the large intestine wall). Due to the presence of the tumor 12, the contour 10 around the tumor 12 protrudes in a direction opposite to the direction toward the contour 11 of the outer wall. Due to the presence of the tumor 14, the contour 10 around the tumor 14 is recessed in a direction toward the contour 11 of the outer wall. If it is a normal large intestine without a tumor, the inner wall around the location where the tumor 12 and the tumor 14 are present is estimated to be flat. The estimated contour 13B is flat where the tumor 12 and the tumor 14 exist. In order to simplify the description, the tumor 12 and the tumor 14 are schematically shown.

(Uplift level calculator 35)
Based on the inner wall contour 10 (first contour) obtained by the contour extracting unit 32 and the inner wall contour 13B (second contour) estimated by the estimating unit 34, the ridge level calculating unit 35 The degree of unevenness (protrusion level) of the part (tumor) is obtained. For example, the size of the tumor and the direction of the tumor correspond to the degree of unevenness (bump level) of the tumor.

  First, the ridge level calculation unit 35 calculates the difference between the position of the contour 10 (first contour) obtained by the contour extraction unit 32 and the position of the contour 13B (second contour) estimated by the estimation unit 34. Ask. This difference corresponds to the degree of unevenness of the tumor. The bulge level calculation unit 35 obtains a region where the contour is different as a lesion (tumor).

  In the vicinity of the tumor, there is a difference between the position of the contour 10 (first contour) obtained by the contour extraction unit 32 and the position of the contour 13B (second contour) estimated by the estimation unit 34. That is, when a tumor is present in the large intestine, the contour 10 obtained by the contour extraction unit 32 is raised or depressed depending on the shape and size of the tumor. On the other hand, the outline 13B estimated by the estimation unit 34 is an outline from which a range in which a tumor exists is removed. Therefore, the outline 13B represents the outline of the normal inner wall when it is assumed that there is no tumor. Therefore, by obtaining the difference between the position of the contour 10 obtained by the contour extraction unit 32 and the position of the contour 13B estimated by the estimation unit 34, the position of the tumor and the degree of unevenness can be obtained.

  Next, the ridge level calculation unit 35 obtains the size of the tumor for a region (tumor) having a difference in contour. Specifically, the bulge level calculation unit 35 obtains the distance from the estimated contour 13B to the actual contour 10 for each tumor for a region (tumor) having a difference in contour. That is, the bulge level calculation unit 35 obtains the distance from the normal inner wall to the actual inner wall for each tumor. The ridge level calculation unit 35 defines the distance as the size of the tumor.

  With reference to FIG. 4, a method for determining the degree of unevenness of the tumor will be described. The depression 15 is an area where there is a difference between the estimated position of the contour 13 </ b> B and the actual position of the contour 10. In the recess 15, the contour 10 is recessed in the direction toward the contour 11 of the outer wall. The dent of the dent 15 is caused by the tumor. The ridge level calculation unit 35 sets an orthogonal line orthogonal to the contour 13B in the region where the tumor exists (the depression 15). The ridge level calculation unit 35 obtains an intersection where the orthogonal line and the actual contour 10 intersect. The ridge level calculation unit 35 calculates the distance from the contour 13B to the intersection along the orthogonal line. The ridge level calculation unit 35 changes the position in the area where the tumor exists (the depression 15), and sets orthogonal lines orthogonal to the contour 13B at each position. The ridge level calculation unit 35 obtains an intersection where each orthogonal line and the actual contour 10 intersect. The ridge level calculation unit 35 calculates the distance from the contour 13B to the intersection for each orthogonal line. The ridge level calculation unit 35 obtains the maximum distance (length L1) among the distances obtained for each orthogonal line. That is, the ridge level calculation unit 35 obtains the maximum distance (length L1) between the contour 13B and the contour 10. The maximum distance (length L1) corresponds to the size of the tumor in the depression 15. The ridge level calculation unit 35 defines the maximum distance (length L1) as the size of the tumor in the depression 15.

  The protrusion 16 is an area where there is a difference between the estimated position of the contour 13 </ b> B and the actual contour 10. In the protrusion 16, the contour 10 protrudes in a direction toward the large intestine core line CL. The protrusion of the protrusion 16 is caused by a tumor. The bulge level calculation unit 35 changes the position in the region where the tumor exists (projection 16), and sets an orthogonal line orthogonal to the contour 13B at each position. The ridge level calculation unit 35 obtains an intersection where each orthogonal line intersects the contour 10 for each orthogonal line. The ridge level calculation unit 35 calculates the distance from the contour 13B to the intersection for each orthogonal line. The ridge level calculation unit 35 calculates a maximum distance (length L2) among the distances calculated for each orthogonal line. The maximum distance (length L2) corresponds to the size of the tumor in the protrusion 16. The bulge level calculation unit 35 defines the maximum distance (length L2) as the size of the tumor in the protrusion 16.

  Further, the bulge level calculation unit 35 obtains the direction of the tumor for each tumor by targeting a region (tumor) having a difference in contour. Specifically, the bulge level calculation unit 35 obtains, for each tumor, the direction in which the actual contour 10 of the inner wall exists with reference to the estimated position of the contour 13B.

  For example, as shown in FIG. 4, the direction from the estimated contour 13B toward the contour 11 of the outer wall is defined as a minus direction (− direction). The direction opposite to the minus direction (− direction) is defined as the plus direction (+ direction). In other words, the direction from the contour 13B to the large intestine core line CL is defined as a plus direction (+ direction). The direction of the contour 10 based on the position of the contour 13B corresponds to the type of tumor. A tumor in which the contour 10 is located in the minus direction with respect to the contour 13B corresponds to a surface depression type tumor as an example. The region where the contour 10 is located in the plus direction with respect to the contour 13B corresponds to a surface-bulged tumor as an example.

  The ridge level calculation unit 35 obtains the direction of the contour 10 in the region where the tumor exists (the hollow portion 15) with reference to the position of the contour 13B. In the hollow portion 15, the contour 10 is located closer to the contour 11 side of the outer wall than the contour 13B. That is, in the hollow portion 15, the contour 10 is positioned in the minus direction with respect to the contour 13B. The ridge level calculation unit 35 determines the direction of the tumor in the depression 15 as the minus direction. Therefore, it can be determined that the tumor in the depression 15 is a surface-recessed tumor.

  Similarly, the ridge level calculation unit 35 obtains the direction of the contour 10 in the region where the tumor exists (projection 16) with reference to the position of the contour 13B. In the protrusion 16, the contour 10 is located closer to the core line CL than the contour 13 </ b> B. That is, in the protrusion 16, the contour 10 is positioned in the plus direction with respect to the contour 13 </ b> B. The ridge level calculation unit 35 determines the direction of the tumor in the protrusion 16 as a plus direction. Therefore, it can be determined that the tumor in the protrusion 16 is a surface-bulged tumor.

  As described above, the bulge level calculation unit 35 determines the tumor size and the tumor direction for each tumor. As described above, the size of the tumor and the direction of the tumor correspond to the degree of unevenness (bump level) of the tumor. For example, for the tumor in the depression 15, the length L1 corresponds to the size of the tumor, and the minus direction corresponds to the direction of the tumor. For the tumor in the protrusion 16, the length L2 corresponds to the size of the tumor, and the plus direction corresponds to the direction of the tumor.

  The bulge level calculation unit 35 includes bulge level information including position information (coordinate information) indicating the position of the tumor, length information indicating the size (length) of the tumor, and direction information indicating the direction of the tumor. Generated for each tumor. The bump level calculation unit 35 outputs the bump level information to the conversion unit 36.

  In the example shown in FIG. 4, for the tumor in the depression 15, the bulge level calculation unit 35 includes position information indicating the position of the tumor, the value of the tumor size (length L1), and the direction of the tumor (minus direction). The ridge level information including the direction information indicating is generated. In addition, for the tumor in the protrusion 16, the bulge level calculation unit 35 includes position information indicating the position of the tumor, a value of the tumor size (length L2), and direction information indicating the direction of the tumor (plus direction). Protrusion level information including The uplift level calculation unit 35 outputs each uplift level information to the conversion unit 36.

  The bulge level calculator 35 may determine the type of tumor based on the direction of the tumor. Specifically, the bulge level calculation unit 35 determines that a tumor having a negative direction is a surface depression type tumor. In addition, the ridge level calculation unit 35 determines that a tumor whose direction is a plus direction is a surface bulge type tumor. The uplift level calculation unit 35 may include the determination result in the uplift level information and output the uplift level information to the conversion unit 36.

(Conversion unit 36)
The conversion unit 36 receives the elevation level information from the elevation level calculation unit 35. The conversion unit 36 determines a color corresponding to the degree of unevenness of the tumor based on the size of the tumor and the direction of the tumor indicated by the bump level information. For example, a look-up table (LUT) for converting the degree of unevenness of the tumor into a color is stored in the condition storage unit 7. The look-up table is a table in which the degree of unevenness of the tumor and the color are associated with each other. The conversion unit 36 refers to the look-up table stored in the condition storage unit 7 and converts the degree of tumor unevenness into a color.

  FIG. 5 shows an example of a lookup table. The lookup table 20 is a table in which a tumor size (length), a tumor direction, and a color are associated with each other. A point of 0 mm is a reference point. With reference to the reference point (0 mm), the + direction and the − direction correspond to the direction of the tumor. The + direction corresponds to a surface raised tumor. The direction corresponds to a tumor with a surface depression. The reference point (0 mm) corresponds to a flat surface tumor. The distance from the reference point corresponds to the size (length) of the tumor. The farther away from the reference point, the larger the tumor size (length). In FIG. 5, the color of the lookup table 20 is schematically represented by hatching for the sake of simplicity. The color in the look-up table 20 changes stepwise according to the size (length) of the tumor.

  The color corresponding to the degree of unevenness of the tumor may be a different color depending on the degree of unevenness, or may be a black and white shade. That is, the color may be changed according to the degree of unevenness of the tumor, or the degree of black and white shading may be changed.

  The operator may be able to change the lookup table 20 using the operation unit 62. That is, the operator may use the operation unit 62 to change the correspondence between the degree of tumor unevenness and the color.

  The conversion unit 36 selects a color corresponding to the size and direction of the tumor indicated by the bump level information from the lookup table 20. For example, the conversion unit 36 selects a color corresponding to the size (length L1) of the tumor and the direction (− direction) of the tumor from the lookup table 20 for the tumor in the depression 15. Further, the conversion unit 36 selects a color corresponding to the size (length L2) of the tumor and the direction (+ direction) of the tumor from the look-up table 20 for the tumor in the protrusion 16.

  The conversion unit 36 generates color data including tumor position information included in the bulge level information and color information indicating a color corresponding to the degree of unevenness of the tumor. The conversion unit 36 outputs the color data to the display control unit 5.

(Image generation unit 4)
The image generation unit 4 reads volume data from the image storage unit 2 and generates image data such as three-dimensional image data based on the volume data. For example, the image generation unit 4 expands the inside of the tissue represented by the volume data into a planar shape, and generates expanded image data that represents the internal surface in a planar manner. The image generation unit 4 may generate virtual endoscopic image data that three-dimensionally represents the internal surface of the tubular tissue by performing volume rendering on the volume data. For example, when the operator designates the position of the viewpoint and the line-of-sight direction using the operation unit 62, the viewpoint information indicating the position of the viewpoint and the line-of-sight direction information indicating the line-of-sight direction are transmitted from the user interface (UI) 6 to the image generation unit 4. Is output. The image generation unit 4 generates virtual endoscope image data by performing volume rendering from the position of the viewpoint toward the viewing direction. Further, the image generation unit 4 may generate MPR image data in an arbitrary cross section by performing MPR (Multi Planar Reconstruction) processing on the volume data. For example, when the operator specifies the position of the cross section using the operation unit 62, position information indicating the position of the cross section is output from the user interface (UI) 6 to the image generation unit 4. The image generation unit 4 generates MPR image data in the designated cross section. The image generation unit 4 outputs image data such as developed image data, virtual endoscopic image data, or MPR image data to the display control unit 5.

  The image generation unit 4 may generate only one of the developed image data and the virtual endoscopic image data, or may generate both image data. When the operator uses the operation unit 62 to specify the type of image data, the image generation unit 4 may generate image data specified by the operator.

(Display control unit 5)
The display control unit 5 receives the image data from the image generation unit 4 and causes the display unit 61 to display an image based on the image data. For example, the display control unit 5 receives the developed image data from the image generation unit 4 and causes the display unit 61 to display a developed image based on the developed image data. The display control unit 5 receives virtual endoscopic image data from the image generation unit 4 and causes the display unit 61 to display a virtual endoscopic image based on the virtual endoscopic image data.

  The display control unit 5 receives the color data from the conversion unit 36 and colors the color indicated by the color data at the position indicated by the color data in the developed image. Further, the display control unit 5 may color the color indicated by the color data at the position indicated by the color data in the virtual endoscopic image. The color indicated by the color data indicates the degree of unevenness of the tumor. That is, the color indicated by the color data indicates the size of the tumor and the direction of the tumor. The direction of the tumor corresponds to the type of tumor. Examples of tumor types include a surface raised type, a surface flat type, and a surface recessed type.

  The display control unit 5 causes the display unit 61 to display a developed image colored according to the degree of unevenness of the tumor. The display control unit 5 may cause the display unit 61 to display a virtual endoscopic image colored according to the degree of unevenness of the tumor.

  The display control unit 5 may display only one of the developed image and the virtual endoscopic image on the display unit 61, or may display both images side by side on the display unit 61. When the operator designates the type of image using the operation unit 62, the display control unit 5 may cause the display unit 61 to display the image designated by the operator.

  FIG. 6 shows an example of a colored image. FIG. 6A shows a developed image 100. In the developed image 100, the inner wall of the large intestine is planarly represented. The display control unit 5 causes the display unit 61 to display the developed image 100 by adding colors indicated by the respective color data to the region 110 and the region 120 where the tumor (lesioned portion) exists. The color indicated by the color data indicates the degree of unevenness of the tumor. Therefore, the color given to the area | region 110 and the area | region 120 represents the grade of the unevenness | corrugation of the tumor in each position. For example, the color of region 110 indicates that the tumor in region 110 is a surface-bulged tumor. Further, the color of the region 110 indicates the degree of bulge of the surface bulge type tumor. The color of region 120 indicates that the tumor in region 120 is a surface-recessed tumor. The color of the region 120 indicates the degree of depression of the surface-recessed tumor.

  FIG. 6B shows a virtual endoscopic image 200. In FIG. 6C, a virtual endoscopic image 300 is shown. In the virtual endoscopic image 200 and the virtual endoscopic image 300, the inner wall of the large intestine is three-dimensionally represented. As illustrated in FIG. 6B, the display control unit 5 causes the display unit 61 to display the virtual endoscopic image 200 by adding a color indicated by the color data to the region 210 where the tumor exists. Further, as shown in FIG. 6C, the display control unit 5 causes the display unit 61 to display the virtual endoscopic image 300 with the color indicated by the color data added to the region 310 where the tumor is present. For example, the color of region 210 indicates that the tumor in region 210 is a surface-bulged tumor. The color of the region 210 indicates the degree of bulge of the surface bulge type tumor. The color of region 310 indicates that the tumor in region 310 is a surface-recessed tumor. The color of the region 310 indicates the degree of depression of the surface-recessed tumor.

  For example, the area 110 represented in the developed image 100 and the area 210 represented in the virtual endoscopic image 200 correspond to each other. The area 120 shown in the developed image 100 and the area 310 shown in the virtual endoscopic image 300 correspond to each other.

  The display control unit 5 may display the developed image 100 and the virtual endoscopic image 200 (or the virtual endoscopic image 300) side by side on the display unit 61. Alternatively, the display control unit 5 may cause the display unit 61 to display either one of the developed image 100 and the virtual endoscopic image 200 (or the virtual endoscopic image 300).

  By referring to the color displayed in the developed image or the virtual endoscopic image, the operator can grasp the degree of unevenness (bump level) of the tumor. That is, the operator can determine whether the tumor is a surface-bulged type tumor, a surface-flat type tumor, or a surface-concave type tumor by the color displayed in the image. Become. Thereby, even if the kind of tumor is a non-lifting type, the operator can grasp the presence or absence of the tumor and the kind of tumor. For example, when the medical image processing apparatus 1 is applied to a preoperative diagnosis including a screening test, the preoperative diagnosis can be performed efficiently.

(Modification)
A modification will be described with reference to FIG. The ridge level calculation unit 35 may obtain the degree of unevenness at a plurality of locations in a region (tumor) having a difference in outline. The depression 15 is an area where there is a difference between the estimated position of the contour 13 </ b> B and the actual position of the contour 10. The ridge level calculation unit 35 changes the position in the area where the tumor exists (the depression 15), and sets orthogonal lines orthogonal to the contour 13B at each position. The ridge level calculation unit 35 obtains an intersection where each orthogonal line intersects the contour 10 for each orthogonal line. The ridge level calculation unit 35 calculates the distance from the contour 13B to the intersection for each orthogonal line. Further, the bulge level calculation unit 35 obtains the direction of the tumor at each position.

  For example, the ridge level calculation unit 35 sets points 41, 42, 43, 44, and 45 at different positions on the contour 13B. The ridge level calculation unit 35 sets an orthogonal line orthogonal to the contour 13 </ b> B at each of the points 41 to 45. The ridge level calculation unit 35 obtains an intersection where each orthogonal line intersects the contour 10 for each orthogonal line. The ridge level calculation unit 35 calculates the distance from the contour 13B to the intersection for each orthogonal line. In the example illustrated in FIG. 7, the ridge level calculation unit 35 includes a distance at the point 41 (length L3), a distance at the point 42 (length L4), a distance at the point 43 (length L5), and a distance at the point 44 (long). L6) and the distance (length L7) at the point 45 are obtained.

  The ridge level calculation unit 35 obtains the direction of the tumor at each point set on the contour 13B. In the example illustrated in FIG. 7, the ridge level calculation unit 35 obtains the direction of the tumor at each of the points 41 to 45.

  As described above, the bulge level calculation unit 35 obtains the lengths and directions of a plurality of locations. The ridge level calculation unit 35 includes position information (coordinate information) indicating the position of a point set on the outline 13B, length information indicating the size (length) of the tumor at the point, and the tumor at the point. Protrusion level information including direction information indicating the direction is generated for each point set on the contour 13B. In the example illustrated in FIG. 7, the uplift level calculation unit 35 generates uplift level information for each of the points 41 to 45. The ridge level calculation unit 35 outputs the ridge level information for each point to the conversion unit 36.

  The conversion unit 36 determines a color for each point according to the degree of unevenness of the tumor according to the ridge level information for each point. In the example illustrated in FIG. 7, the conversion unit 36 determines a color for each of the points 41 to 45. The conversion unit 36 outputs color data including position information indicating the position of a point set on the contour 13B and color information indicating a color according to the degree of unevenness of the tumor to each point set on the contour 13B. Generate about. The conversion unit 36 outputs the color data to the display control unit 5.

  The display control unit 5 colors the color indicated by the color data for each point at the position indicated by the color data in the developed image or virtual endoscopic image. The display control unit 5 causes the display unit 61 to display a developed image or a virtual endoscopic image colored according to the degree of unevenness at each point in the tumor.

  As described above, it is possible to display the unevenness distribution in the tumor by obtaining the degree of unevenness in a plurality of locations in the tumor. As a result, the operator can grasp the degree of unevenness of the tumor in more detail.

  The observation target may be a site other than the large intestine. For example, even when a blood vessel is an observation target, the same operations and effects as when the large intestine is the observation target can be achieved by performing the processing by the image processing unit 3 described above. In addition, even when a bag-like tissue such as the stomach is an observation target, the same actions and effects as when the large intestine is the observation target can be achieved.

(User interface (UI) 6)
The user interface (UI) 6 includes a display unit 61 and an operation unit 62. The display unit 61 includes a monitor such as a CRT or a liquid crystal display. The operation unit 62 includes an input device such as a keyboard and a mouse.

  Each of the image processing unit 3, the image generation unit 4, and the display control unit 5 includes a processing device (not shown) such as a CPU, GPU, or ASIC, and a storage device (not shown) such as a ROM, RAM, or HDD. Also good. The storage device stores an image processing program for executing the functions of the image processing unit 3. The image processing program executes the observation target extraction program for executing the function of the observation target extraction unit 31, the contour extraction program for executing the function of the contour extraction unit 32, and the function of the lesion extraction unit 33. For executing the function of the estimation unit 34, the estimation program for executing the function of the estimation unit 34, the elevation level calculation program for executing the function of the elevation level calculation unit 35, and the function of the conversion unit 36 A conversion program is included. The storage device stores an image generation program for executing the function of the image generation unit 4. The storage device stores a display control program for executing the functions of the display control unit 5. A processing device such as a CPU executes each program stored in the storage device, so that the function of each unit is executed.

(Operation)
A series of operations performed by the medical image processing apparatus 1 will be described with reference to FIG. The observation object is the large intestine. For example, a contrast medium is administered to a subject, and carbon dioxide is injected into the large intestine. The X-ray CT apparatus as the medical image capturing apparatus 90 captures the subject, thereby generating volume data representing an imaging region including the large intestine.

(Step S01)
The medical image processing apparatus 1 acquires volume data representing an imaging region including the large intestine from the medical image imaging apparatus 90. The image storage unit 2 stores volume data.

(Step S02)
The observation target extraction unit 31 reads volume data from the image storage unit 2 and executes segmentation processing to extract volume data representing the large intestine from the volume data.

(Step S03)
The contour extraction unit 32 obtains the contour of the inner wall of the large intestine (first contour) based on the volume data representing the large intestine.

(Step S04)
The lesion part extraction part 33 calculates | requires the pixel value inside a colon wall. For example, as illustrated in FIG. 2A, the lesion extraction unit 33 translates the contour 10 by a distance of one pixel in a direction from the inner wall contour 10 (first contour) toward the outer wall contour 11. . The lesion extraction unit 33 defines the moved contour 10 as the search line 13. The lesion extraction unit 33 obtains a pixel value (CT value) of each pixel on the search line 13.

(Step S05)
The lesion extraction unit 33 determines that a pixel whose pixel value is included in a value within a predetermined range is a pixel included in the tumor. Thereby, the lesion part extraction part 33 calculates | requires the position of the tumor on the search line 13. FIG.

(Step S06)
The estimation unit 34 obtains the outline of the normal inner wall of the large intestine. For example, as illustrated in FIGS. 2A and 2B, the estimation unit 34 excludes a range where the tumor 12 exists in the search line 13. The estimation unit 34 obtains the connecting line 13A by connecting the end point A of the search line 13 extending to one side of the tumor 12 and the end point B of the search line 13 extending to the other side of the tumor 12 by a straight line. 2B, the estimation unit 34 translates the connecting line 13A by a distance of one pixel in the direction from the outer wall contour 11 toward the inner wall contour 10. As shown in FIG. 2C, the estimation unit 34 defines the contour after movement as a normal intimal contour 13B.

(Step S07)
The bulge level calculation unit 35 obtains the degree of tumor unevenness (bulge level) based on the inner wall contour 10 (first contour) and the estimated contour 13B (second contour). For example, as shown in FIG. 4, the ridge level calculation unit 35 obtains the size (length L1) of the tumor in the depression 15 and the direction of the tumor (minus direction). Further, the bulge level calculation unit 35 obtains the size (length L2) of the tumor in the protrusion 16 and the direction of the tumor (plus direction). The ridge level calculation unit 35 generates ridge level information including position information indicating the position of the tumor, length information indicating the size (length) of the tumor, and direction information indicating the direction of the tumor for each tumor. .

(Step S08)
The conversion unit 36 refers to the look-up table stored in the condition storage unit 7, and based on the tumor size and the direction of the tumor indicated by the uplift level information, the color corresponding to the degree of the unevenness of the tumor, Determine for each tumor. In this way, the conversion unit 36 converts the degree of unevenness of the tumor into a color. The conversion unit 36 generates color data including tumor position information and color information indicating a color corresponding to the degree of unevenness of the tumor.

(Step S09)
The image generation unit 4 reads volume data from the image storage unit 2 and generates developed image data or virtual endoscopic image data based on the volume data.

(Step S10)
The display control unit 5 colors the color indicated by the color data at a position indicated by the color data in the developed image or virtual endoscopic image. The display control unit 5 causes the display unit 61 to display a developed image or a virtual endoscopic image colored according to the degree of unevenness of the tumor.

  According to the above processing, a color corresponding to the degree of unevenness (protrusion level) of the tumor is represented in an image such as a developed image or a virtual endoscopic image. The operator can grasp the degree of unevenness of the tumor by referring to the color displayed in the image. Thereby, even if the type of tumor is a non-protruding type, the operator can grasp the presence or absence of the tumor and the type of tumor.

  The order of the process from step S02 to step S08 and the process of step S09 may be reversed. That is, the process of step S09 may be executed before the process of step S02, or may be executed in the middle of the process from step S02 to step S08.

  In the above-described embodiments and modifications, CT colonography using an X-ray CT apparatus is described as an example. As another example, MR colonography using an MRI apparatus may be used.

DESCRIPTION OF SYMBOLS 1 Medical image processing apparatus 2 Image storage part 3 Image processing part 4 Image generation part 5 Display control part 6 User interface (UI)
7 Condition Storage Unit 31 Observation Object Extraction Unit 32 Contour Extraction Unit 33 Lesion Part Extraction Unit 34 Estimation Unit 35 Uplift Level Calculation Unit 36 Conversion Unit

Claims (7)

Contour extracting means for receiving volume data representing a luminal body and obtaining a first contour of the inner wall of the luminal body based on the volume data;
A lesion portion extraction means for obtaining a position of a lesion portion based on a pixel value of a pixel in a region outside the inner wall with respect to the first contour;
Estimating means for estimating a second contour of the inner wall when the lesion does not exist based on the position of the lesion and the first contour;
Based on the first contour and the second contour, a bulge level calculating means for determining the degree of unevenness of the lesioned part,
Image generating means for generating image data representing the lumen body based on the volume data;
Display control means for causing the display means to display information indicating the degree of unevenness and an image based on the image data ;
The lesion portion extraction means translates the first contour from the position of the inner wall toward the outer region by a predetermined distance, and is based on pixel values of pixels on the first contour after the translation. A medical image processing apparatus for obtaining a location where the lesion is present . The estimation means connects the first contours from which the lesioned part has been removed except for the part where the lesioned part exists from the first contour after the parallel movement, and connects the first contours. contour, medical image processing apparatus according to claim 1 for obtaining a second contour by returning to the position before the translation. Said raised level calculating means, the difference between seeking a place where there is the first contour and the second contour, medical according to claim 1 or claim 2 obtains the difference as the degree of unevenness of the lesion Image processing device. The ridge level calculating means obtains a distance between the first contour and the second contour, obtains a direction in which the first contour exists with reference to the second contour, The medical image processing apparatus according to claim 3 , wherein the direction is a degree of unevenness of the lesioned part. Further comprising conversion means for converting the degree of unevenness into a color;
Wherein the display control unit, the converted color as the information indicating the degree of the unevenness, the medical image processing apparatus according to any one of claims 1 to 4 to be displayed on the display unit superposed on the image. Based on the distance and the direction, the image processing apparatus further includes conversion means for converting the degree of the unevenness into a color corresponding to any one of a raised type, a planar type, and a recessed type that is a type of the shape of the lesioned part. ,
The medical image processing apparatus according to claim 4 , wherein the display control unit causes the display unit to display the converted color as information indicating the degree of unevenness on the display unit. On the computer,
A contour extraction function that receives volume data representing a lumen body and obtains a first contour of the inner wall of the lumen body based on the volume data;
A lesion extraction function for determining a position of a lesion based on a pixel value of a pixel in a region outside the inner wall with respect to the first contour;
Based on the position of the lesioned part and the first contour, an estimation function for estimating the second contour of the inner wall when the lesioned part does not exist;
Based on the first contour and the second contour, a raised level calculation function for determining the degree of unevenness of the lesion,
An image generation function for generating image data representing the lumen body based on the volume data;
A display control function for causing the display device to display information indicating the degree of unevenness and an image based on the image data;
Was executed,
The first contour is translated by a predetermined distance from the position of the inner wall toward the outer region, and the lesioned part is present based on the pixel value of the pixel on the first contour after the translation A medical image processing program characterized by obtaining a location to be performed .

Images