JP2011024826A - Medical image processor and medical image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine the risk of unstable plaque by obtaining the degree of contiguity between a deposit material represented by a lipid core and a blood vessel lumen. <P>SOLUTION: An extract part 104 extract a contour of the lumen of the blood vessel and a contour of deposit material adhering to the interior of the blood vessel from the section of the blood vessel obtained based on image data about a region including the vessel. An expansion part 105 expands at least part of one or both of the extracted contour of the lumen and contour of the deposit material by a predetermined amount. An overlapping region computing part 106 computes the size of an overlapping region when the contour of the inner lumen and the contour of deposit material expanded by the expansion part 105 overlap each other. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a medical image processing apparatus and a medical image processing program for obtaining a cross section of a blood vessel based on image data of a region including a blood vessel and executing image processing on the cross section.

  With the development of medical image collection technology, the intravascular echo technique represented by IVUS (intravascular ultrasound) has been used to analyze the properties of plaque in blood vessels. On the other hand, even in multi-slice CT, although the spatial resolution is not higher than that of IVUS at present, it is becoming possible to discriminate lipid cores attached to blood vessels by improving the spatial resolution. Since IVUS requires insertion into the body for examination, it is required to collect images of blood vessels from outside the body using CT (Computed Tomography). As disclosed in Patent Document 1, a technique for detecting unstable plaque by detecting light incident on a coronary artery via a probe is disclosed.

  An example of a target for collecting images is the coronary artery. A lipid core may accumulate in the coronary arteries and form unstable plaques. Plaques are generally classified into two types: unstable plaques with large lipid cores and thin fibrous coatings, and stable plaques with small lipid cores and thick fibrous coatings. The risk of unstable plaque is determined by the thickness of the fibrous cap, the size of the lipid core, the degree of macrophage infiltration, and the amount of smooth muscle cells.

Special table 2009-509694

  As the unstable plaque gradually thickens, the lumen of the coronary artery is narrowed, causing acute coronary syndrome and acute myocardial infarction. Although it is necessary to discriminate this unstable plaque by inspection, when an image is collected using CT, there is a problem that the collected image is less than the required high resolution. In CT, for example, bone can be clearly photographed, but it is difficult to photograph tissue in blood vessels with high resolution.

  In particular, the fibrous coating that separates the blood vessel lumen from the lipid core is generally very thin. It is necessary to determine the thickness of the fibrous coating and determine the risk of unstable plaque. However, when measuring the thickness of the fibrous coating, there is a problem that the spatial resolution of the obtained image is insufficient to determine the risk of unstable plaque because the fibrous coating is too thin. Therefore, there is a problem that the degree of proximity between the blood vessel lumen and the lipid core and the risk of unstable plaque due to the obtained image cannot be sufficiently confirmed.

  The present invention has been made in view of such circumstances, and an adhesion substance typified by a lipid core is obtained by obtaining a cross section of the blood vessel based on image data of a site including the blood vessel and subjecting the cross section to image processing. It is an object of the present invention to provide a medical image processing apparatus and a medical image processing program that can determine the degree of proximity between a blood vessel lumen and a blood vessel lumen and thereby determine the risk of unstable plaque.

  In order to achieve the above object, a medical image processing apparatus according to the first aspect of the present invention obtains a cross section of the blood vessel based on image data of a region including the blood vessel, and determines the contour of the lumen of the blood vessel from the cross section. Extraction means for extracting the outline of the adhering substance adhering in the blood vessel; expansion calculating means for expanding at least a part of one or both of the extracted outline of the lumen and the outline of the adhering substance by a predetermined amount; An area calculation unit that obtains the size of an overlapping area when the outline of the lumen expanded by the expansion calculation unit and the outline of the adhering substance overlap each cross section of the blood vessel; .

  In order to achieve the above object, a medical image processing apparatus according to the invention of claim 14 obtains a cross section of the blood vessel based on image data of a site including the blood vessel, and determines the contour of the lumen of the blood vessel from the cross section. An extraction step for extracting the outline of the adhering substance adhering to the blood vessel, and an expansion calculating step for expanding at least a part of one or both of the extracted outline of the lumen and the outline of the adhering substance by a predetermined amount; For each cross section of the blood vessel, the computer executes an area calculation step for obtaining a size of an overlapping area when the outline of the lumen expanded by the expansion calculation step and the outline of the attached substance overlap. Features.

  As described above, the vascular lumen and the lipid core can be expanded to obtain overlapping regions, and the degree of proximity between the adhered substance and the vascular lumen can be obtained from the size, so that an index instead of the thickness of the fibrous coating It can be. Thereby, even when the obtained image is obtained lower than the intended resolution, this overlapping region can be used as an index. Then, the risk level of the unstable plaque can be determined from the size of the overlapping region, and the user can easily check the state of the unstable plaque having a high risk level.

1 is a functional block diagram of a medical image diagnostic apparatus according to an embodiment. It is sectional drawing which cut | disconnected the coronary artery in the direction along the blood-vessel core line. It is sectional drawing which cut | disconnected the coronary artery in the direction perpendicular | vertical to the blood vessel core line. It is a conceptual diagram explaining the process which extracts the area | region of a lipid core and a blood vessel lumen. It is the schematic which shows the outline of the extracted blood vessel lumen | bore and a lipid core. It is the schematic which shows the case where the outline of a blood vessel lumen and a lipid core is expanded. It is the schematic which shows the blood vessel lumen after expansion | swelling, and the outline of a lipid core. It is the schematic explaining the overlapping part of the outline after expansion | swelling. It is the display screen which showed the change of the area of an overlap area as a graph. It is a cross-sectional image of a blood vessel in which the overlapping region is maximized. It is the schematic explaining the arrangement | positioning of the long axis and short axis inside an overlapping area | region. It is the display screen which arranged the change of the long axis and the short axis as a graph. It is the display screen which showed the change of the area of an overlap area on the graph as a set of the length direction of a blood vessel. It is sectional drawing which cut | disconnected the coronary artery in the direction along the blood-vessel core line in the case of having a plurality of lipid cores. It is a display example when displaying the information regarding the risk of each lipid core. It is a flowchart which shows the process which performs an analysis process about a series of coronary arteries. It is a flowchart which shows the calculation process of an overlap area | region. It is a flowchart explaining the process which adds the area of an overlapping area | region in the area of the length of a blood-vessel direction, and makes it a volume.

  FIG. 1 is a functional block diagram of the medical image diagnostic apparatus according to the present embodiment. The medical image diagnostic apparatus illustrated in FIG. 1 includes an X-ray CT apparatus 10 that acquires a medical image, a medical image processing apparatus 100 that processes the obtained medical image, a display control unit 20, and an output from the display control unit 20. And a display unit 30 for displaying an image.

  The medical image processing apparatus 100 includes an information processing device (CPU) and a storage device (not shown) such as a ROM, a RAM, and an HDD. The storage device stores a processing program for executing the function of each unit of the medical image processing apparatus 100. The processing program executes each process of the image data storage unit 101, the control unit 102, the blood vessel acquisition unit 103, the extraction unit 104, the expansion unit 105, the overlapping region calculation unit 106, and the image generation unit 107. A program is included. Each processing is executed by the information processing apparatus (CPU) executing each program.

  The X-ray CT apparatus 10 rotates an X-ray irradiation unit and an X-ray detection unit (both not shown) in a direction perpendicular to the body axis of the subject, and further irradiates the patient with X-rays while proceeding in the body axis direction. Then, X-rays that have passed through the region to be inspected are detected. The examination target region includes a blood vessel region of a patient suspected of having unstable plaque and a heart region around the blood vessel. Then, the X-ray CT apparatus 10 generates an X-ray CT image that is a three-dimensional image based on the detected X-rays. Then, the X-ray CT apparatus 10 outputs a signal to the control unit 102 and outputs the generated X-ray CT image to the medical image processing apparatus 100.

  The medical image processing apparatus 100 includes an image data storage unit 101, a control unit 102, a blood vessel acquisition unit 103, an extraction unit 104, an expansion unit 105, an overlapping area calculation unit 106, and an image generation unit 107. . The medical image processing apparatus 100 processes the image obtained from the X-ray CT apparatus 10 and outputs the processed image and an image obtained according to the resulting data to the display control unit 20.

  The image data storage unit 101 acquires and stores image data of a part including a blood vessel imaged by the X-ray CT apparatus 10. In particular, image data of the patient's heart region is read from the X-ray CT apparatus 10 as heart region data.

  The control unit 102 controls each unit of the medical image processing apparatus 100. As described above, when an X-ray CT image is input to the blood vessel acquisition unit 103, a signal is input from the X-ray CT apparatus 10, so that a series of processing is started. The control unit 102 executes a process for specifying the blood vessel cross section to be processed. When the image data is sent from the X-ray CT apparatus 10 to the image data storage unit 101, the control unit 102 starts processing for the first blood vessel cross section.

  The control unit 102 controls the routine work of the processing of the present embodiment by the blood vessel acquisition unit 103, the extraction unit 104, the expansion unit 105, and the overlap region calculation unit 106 in FIG. That is, the blood vessel acquisition unit 103 is instructed to process the first blood vessel cross section, and when a series of processing is completed for one blood vessel cross section, an end signal is input from the overlapping region calculation unit 106. Instructs processing of blood vessel cross section. When processing is completed for the entire target blood vessel, the end of a series of processing is instructed. The entire series of flows will be described later with reference to FIG.

  The blood vessel acquisition unit 103 extracts a blood vessel to be inspected from the image data of the entire part stored in the image data storage unit 101. As a blood vessel, a coronary artery will be described as an example, but is not limited thereto. In addition, relatively thick arteries such as the aorta and the cerebral artery can be targeted. Therefore, the blood vessel acquisition unit 103 extracts a blood vessel cross-sectional image from the coronary artery image data stored in the image data storage unit 101.

  More specifically, the blood vessel acquisition unit 103 first obtains a blood vessel core line for this coronary artery. The blood vessel core line is a line passing through the center of the blood vessel, and the direction of the blood vessel core line is the blood vessel progression direction and the blood vessel length direction. When a blood vessel core line is obtained, the blood vessel acquisition unit 103 generates a cross-sectional image of the coronary artery using a plane perpendicular to the direction of the blood vessel core line and a plane perpendicular to the direction of the blood vessel core line as a cross section of the coronary artery. This blood vessel cross section is obtained at predetermined intervals in the blood vessel core line direction. This predetermined interval is a distance that allows detection of the lipid core. If the interval is too close, it takes more time than necessary to acquire and process the blood vessel cross-sectional image, and if the interval is too wide, the state of the lipid core cannot be grasped. get.

  Since each blood vessel bends in accordance with the traveling direction, the direction of the cross section also bends accordingly. On the other hand, by obtaining a cross section perpendicular to the blood vessel core line, the direction of the cross section represents a constant thickness of the blood vessel, and variations in each cross section can be reduced to an extent suitable for comparison. In addition, since the contour of the blood vessel lumen and the lipid core are obtained from the cross section of the blood vessel as described later, the contour of the blood vessel lumen and the lipid core suitable for comparison can be obtained in the same manner by being perpendicular to the blood vessel core line.

  As described above, a cross section of the coronary artery is obtained at every predetermined interval of the blood vessel core line, but a cross-sectional image is finally obtained from the base portion of the coronary artery closer to the heart to the end portion. The region may be all from the base portion to the end portion, or may be only a necessary range by setting. In addition, the existing technique can be used for the method of extracting the blood vessel, the method of obtaining the blood vessel core line, and the method of obtaining the blood vessel cross section.

  The extraction unit 104 includes a lipid core extraction unit 120, a blood vessel lumen extraction unit 130, and a closed curve extraction unit 140. With these configurations, a closed curve indicating the contours of the coronary artery, lipid core, and blood vessel lumen is generated from the image input from the blood vessel acquisition unit 103.

  The lipid core extraction unit 120 extracts a region occupying a lipid core from the blood vessel cross-sectional image obtained by the blood vessel acquisition unit 103 based on the difference in pixel values. In general, pixel values estimated as images are different for blood vessel lumens, lipid cores, and other regions, so that a threshold value is provided for distinction and extraction. The lipid core is a calcified portion that accumulates in the blood vessel. An atheromatous plaque (atherosclerotic plaque) is formed by accumulating muddy substances such as cholesterol in the intima of arteries, and the portion rich in lipid components in atheromatous plaque is called a lipid core. Since this lipid core is formed in a bowl shape, it adheres to the inside of the blood vessel and becomes a factor for narrowing the lumen. In this example, the inside of a blood vessel is analyzed using a lipid core as a target, but it is considered that various substances such as calcium and the like adhere to the inside of the blood vessel. May be analyzed.

  The blood vessel lumen extraction unit 130 extracts a region occupying the lumen of the blood vessel from the blood vessel cross-sectional image obtained by the blood vessel acquisition unit 103 based on the pixel value. The inner cavity is an inner cavity, and is an inner side of a tubular or bag-like structure such as a blood vessel.

  The lipid core extraction unit 120 and the blood vessel lumen extraction unit 130 respectively extract a lipid core region and a blood vessel lumen region by segmentation processing using a region expansion method from the blood vessel cross-sectional image. The lipid core and the blood vessel lumen are arranged as shown in FIGS. 2 and 3 on a cross section along the blood vessel traveling direction. Extraction of a region using this region expansion method will be described later with reference to FIG.

  The closed curve extraction unit 140 draws a closed curve that draws the outside of the area that occupies the given image, and thereby extracts the outline of the area. Images are acquired from the blood vessel acquisition unit 103, the lipid core extraction unit 120, and the blood vessel lumen extraction unit 130, respectively, and the contour of the blood vessel cross section, the contour of the lipid core, and the contour of the blood vessel lumen are formed. An example of the contour thus formed and extracted is shown in FIG. Each contour data is output to the expansion unit 105.

  The expansion unit 105 expands the contour indicated by the given data. That is, one or both of the regions respectively indicated by the contour of the lipid core and the contour of the blood vessel lumen are expanded by a predetermined amount. In this embodiment, both are inflated and used for subsequent processing, but only one may be inflated. The dilation process of the morphological filter is applied to this expansion process. Thereby, each contour is spread outward. This expansion process will be described later with reference to FIG.

  In view of the performance of the X-ray CT apparatus 10, each contour output from the extraction unit 104 may be obtained as an image having a low resolution with respect to the actual thickness of the fibrous coating. Therefore, when the proximity of the lipid core and the blood vessel lumen is used as it is, there is a problem that the accuracy of the required proximity is low. In the present embodiment, by expanding each contour by the expansion unit 105, a range that can actually be approached from each contour image obtained at a low resolution can be obtained as an estimated image. As described later, the size of the overlapping region is obtained for the proximate range obtained as the estimated image, so that the possibility of danger for each cross section of the blood vessel can be scored.

  Moreover, although the case where the whole outline of a lipid core or a blood vessel lumen is expanded is demonstrated in the following Example, you may expand only a part of outline instead of the whole outline. For example, it is possible to divide the entire given contour in half on the side closer to the outer wall of the blood vessel and on the far side, and inflate only the far side. As will be described later with reference to FIG. 5, neither the blood vessel lumen nor the lipid core exists at the center of the blood vessel, both are shifted from the center, and the portions far from the respective blood vessels are close to each other. That is, the portion of the lipid core far from the blood vessel outer wall is close to the blood vessel lumen, and the portion of the blood vessel lumen far from the defective outer wall is close to the lipid core. Therefore, by expanding only the halves that are close to each other, it is possible to speed up the expansion process in half, while executing the necessary expansion process and performing calculations related to the subsequent overlapping area. Can do.

  In addition, although only the left / right half of each contour is shown here, various lengths such as 1/3 or 2/3 of the contour may be defined as a range. When the expansion is set to 1/3 of the contour, half of both contours may be obtained as described above, and 1/6 may be removed from the outside of the obtained half contour. Moreover, when setting it to 2/3 of an outline, you may obtain | require by calculating | requiring half each of both outlines as above-mentioned, respectively, and adding 1/6 from the outer side of the calculated | required half outline. In addition, various combinations are possible depending on the processing time to be shortened, such as one contour expanding all parts and the other contour expanding a part.

  The predetermined amount to be expanded is 1 pixel (1 pixel) in the embodiment of FIG. 6, but may be a larger number of pixels to expand a sufficient amount, for example, 2 pixels, 3 pixels, etc. be able to. Instead of expanding several pixels at a time, it is also possible to expand the remaining pixels for the expanded region after expanding one pixel. The setting of the number of pixels to be expanded and the expansion process of the pixels can be determined by setting.

  However, as will be described later, since this expansion process is a process for avoiding the disadvantage that the outline of the original area becomes rough due to the number of pixels, the size and shape of the outline of the original area are increased. It is desired to have an expansion that does not change. Therefore, the expansion range of the contour is a position that does not exceed the center of gravity of at least the other contour before expansion. Alternatively, the expansion width is based on the size before expansion, such as within a range of 1/3, 1/5, and 1/10 of the distance from the center of gravity of each region before expansion. You can also.

  By increasing the predetermined amount indicating the degree of expansion, an overlapping region described later becomes larger, so that it is possible to extract many portions where the lipid core and the blood vessel lumen are close to each other. As a result, even when the boundary region is unclear due to the roughness of the acquired image, the range that can be approached can be widened, and thus a wide range that can be dangerous can be obtained. In addition, by setting the predetermined amount to the minimum size, for example, when the acquired image is obtained with relatively high definition, the range that is predicted to be dangerous is limited to the range that is actually considered dangerous. Ranges that are not dangerous can be excluded. Thereby, it is possible to obtain data that is more suitable for the actual proximity situation. This selection can be set as necessary depending on the image quality of the obtained image.

  The overlap region calculation unit 106 obtains the size of the region formed by the contour of the lipid core and the contour of the blood vessel lumen that has been expanded by the expansion unit 105. That is, the size of the overlapping region when the contour of the inflated lumen and the contour of the adhered substance overlap is obtained. The size of this overlapping region is obtained for each cross section across the direction of the blood vessel core line. When only one of the inflated portions 105 is inflated, the overlap region is obtained using both the inflated contour and the non-expanded contour.

  Since the contour of the lipid core and the contour of the blood vessel lumen are both inflated, the two contours overlap. Therefore, the overlapping area calculation unit 106 determines whether or not both contours overlap each other, and when it is determined that they overlap each other, the overlapping area calculation unit 106 extracts an overlapping area due to the overlapping of both the contours. The extracted overlapping area data is stored in the memory of the overlapping area calculation unit 106 in correspondence with the position of each cross section along the length direction of the blood vessel.

  Then, the overlapping region calculation unit 106 calculates the area of the extracted overlapping region and outputs it to the image generation unit 107. The area of the overlapping region can be obtained by counting the number of dots inside the overlapping region. In addition, when both outlines do not overlap, the area of the overlapping region is zero. Also, even when the lipid core is not extracted by the lipid core extraction unit 120, the area of the overlapping region is zero. This overlapping area is shown in FIG. 8 as described later. The obtained area of the overlap region is obtained for each cross-sectional position along the length direction of the blood vessel, and is stored in the memory of the overlap region calculation unit 106 in correspondence with the position.

  The overlapping area calculation unit 106 calculates the length of the major axis and the length of the minor axis in the overlapping area together with or instead of the area of the overlapping area. In this embodiment, the case of obtaining both the long axis and the short axis will be described. However, only the long axis may be obtained. The calculation of the lengths of the major axis and the minor axis will be described later with reference to FIG. The long axis is a straight line connecting the two most distant points on the outline of the overlapping area, and the short axis is the longest line among the straight lines connecting the outline of the overlapping area perpendicular to the long axis. The overlapping area calculation unit 106 obtains the lengths of the long axis and the short axis, and stores them in the memory of the overlapping area calculation unit 106 in correspondence with the position of each cross section along the length direction of the blood vessel. Further, the overlapping area calculation unit 106 obtains the sum of the long axis length and the short axis length and the difference value between the long axis and the short axis and stores them in the memory of the overlapping area calculation unit 106.

  Further, the area of the overlap region cross section is obtained for each cross section of the blood vessel, and this is stored in the memory of the overlap region calculation unit 106, so the overlap region calculation unit 106 obtains the sum as a volume. Here, the overlapping area calculation unit 106 extracts sections in which the area of the overlapping area is continuously larger than 0 in the direction of the blood vessel core line, and adds up the areas in the extracted sections to total the sections. Find the value. Specifically, the overlapping area calculation unit 106 determines whether or not the overlapping area is larger than 0 at each position in the length direction of the blood vessel, and from the position determined to be larger than 0 to the position where it becomes 0 again. Is extracted as this section. Then, the overlapping areas in the extracted section are summed. The summed value becomes the volume of the overlapping area of each section, and is sent to the image generation unit 107. The process for obtaining the volume will be described later with reference to FIG.

  As described above, the processing of the extraction unit 104, the expansion unit 105, and the overlap region calculation unit 106 is applied to one blood vessel cross-sectional image. Since the blood vessels are connected from the base on the heart side to the terminal side, the processing is started from the cross section on the base side first, and then the processing is shifted to the cross section on the terminal side under the control of the control unit 102. When the process is executed for one cross section, the next blood vessel cross section is extracted as specified by the blood vessel acquisition unit 103, and a series of processes are repeated.

  The image generation unit 107 generates an image to be displayed on the display unit 30 based on data sent from the image data storage unit 101, the blood vessel acquisition unit 103, the extraction unit 104, and the overlapping area calculation unit 106, and performs display control. Send to part 20. The image generation unit 107 acquires image data including a blood vessel part from the image data storage unit 101 and generates a display image. In this case, not only the blood vessel to be examined but also the image data of the part including the blood vessel imaged by the X-ray CT apparatus 10 is acquired and sent to the display control unit 20. This image data may be three-dimensional image data of a part, or may be a Stretched MPR (Multi Planar Reformation) image.

  The image generation unit 107 acquires blood vessel data from the blood vessel acquisition unit 103 and generates a blood vessel image. Examples of the blood vessel image include an outline image, a cross-sectional image perpendicular to the core line, and a cross-sectional image including the core line along the blood vessel core line direction. The image generation unit 107 acquires data indicating the contour of the blood vessel, the contour of the lipid core, and the contour of the blood vessel lumen from the closed curve extraction unit 140, and generates an image indicating each contour from this data.

  Further, the image generation unit 107 acquires data related to the overlapping area from the overlapping area calculation unit 106 and generates an image indicating the overlapping area. The data relating to the overlapping region includes an image of the overlapping region, the area of the overlapping region, the length of the overlapping region, the length of the short axis, the total, the difference, and the volume of the overlapping region for each section. In addition, the image generation unit 107 acquires data indicating the size of the overlapping area, and generates graph information from this data.

  The display control unit 20 acquires an image from the image generation unit 107, converts it into a signal to be output to the display unit 30, and outputs this signal to the display unit 30. The display unit 30 is a display that displays an image based on a signal output from the display control unit 20.

[Extraction of lipid core and vascular lumen]
FIG. 2 is a cross-sectional view of the coronary artery cut in a direction along the blood vessel core line. Since it is a cross section along the blood vessel core line, it is actually connected, but for convenience, the upper side is the coronary artery 201 and the lower side is the coronary artery 202. A substance attached to the coronary arteries 201 and 202 and a region surrounded by them will be described. The lipid core 203 is attached to the coronary artery 202. Although it adheres to the coronary artery 202 in FIG. 2, the case where it adheres to the coronary artery 201 is also assumed. Macrophages 204 are also attached to the coronary artery 202.

  Next, processing of the lipid core extraction unit 120 and the blood vessel lumen extraction unit 130 will be described. The blood vessel lumen extraction unit 130 obtains a point corresponding to the blood vessel lumen from the cross-sectional view of the blood vessel shown in FIG. 2A, and obtains the blood vessel lumen region 205 by the region expansion method. The obtained blood vessel lumen region 205 is specified as the region shown in FIG. Further, the lipid core extraction unit 120 obtains a point corresponding to the lipid core from the cross-sectional view of the blood vessel shown in FIG. 2A, and obtains the lipid core 203 therefrom by the region expansion method. The obtained blood vessel lumen region 205 is specified as the region shown in FIG.

  In addition, although FIG. 2 demonstrated the cross section of the direction along a core wire, the cross section perpendicular | vertical to a core wire is also demonstrated below. Here, since a cross section including the lipid core 203 and a cross section not including the lipid core 203 will be described, a cross section including the lipid core 203 in FIG.

  FIG. 3 is a cross-sectional view of the coronary artery cut in a direction perpendicular to the blood vessel core line. FIG. 3A shows a cross section 210 that does not include the lipid core 203. In FIG. 2, the coronary arteries 201 and 202 have been described separately. However, when viewed from the cross-section 210 perpendicular to the core line, the shape is connected in a circumferential shape in this way. The cross section 210 does not include the lipid core 203 and the macrophages 204. FIG. 3B shows a cross section 220 including a lipid core 203. In the cross section 220, the lipid core 203 and the macrophages 204 are attached to the inside of the blood vessel. Furthermore, the lipid core 203 is separated from the blood vessel lumen by a fibrous coating 230.

  FIG. 4 is a conceptual diagram illustrating a process of extracting lipid core and vascular lumen regions. As described with reference to the lipid core extraction unit 120 and the blood vessel lumen extraction unit 130 in FIG. 1, the lipid core and the blood vessel lumen regions are respectively extracted by the segmentation process using the region expansion method. In addition, although the case of a lipid core and a blood vessel lumen is described together, this processing is executed separately, and the lipid core and the blood vessel lumen are extracted separately.

  First, the lipid core extraction unit 120 and the blood vessel lumen extraction unit 130 store in advance pixel values estimated and assumed to correspond to the lipid core or the blood vessel lumen, respectively, and correspond to the target blood vessel cross-sectional image 401. A pixel 402 serving as a pixel value is discriminated and extracted. This pixel may be limited to one specific pixel value or may have a certain width. Although the blood vessel cross-sectional image 401 is shown as a square for the sake of explanation, the blood vessel cross-sectional image 401 is actually discriminated and extracted over the entire inside of the circular shape of the blood vessel cross-section. The extracted pixel 402 is classified as a hypothetical region that is assumed to correspond to a lipid core or vessel lumen.

  In some cases, the pixel having the corresponding pixel value, that is, the pixel 402 is not detected. That is, the lipid core extraction unit 120 determines whether or not a lipid core exists, and in this case, determines that it does not exist. In this case, if the lipid core extraction process, it means that the lipid core was not detected from the blood vessel cross section, and the lipid core extraction process for the blood vessel cross section is terminated. Since the lipid core is not extracted, it is not necessary to extract or inflate the closed curve thereafter, and the processing for the blood vessel cross section is terminated. Since there is no lipid core, it can be seen that there are no unstable plaques at this point, so that subsequent processing can be omitted to reduce processing time.

  When the pixel 402 is detected, it is determined whether or not a condition is satisfied with respect to neighboring pixels and neighboring pixels adjacent to the pixel 402 classified as the hypothetical region. This condition is whether or not adjacent pixels to be determined have the same or close pixel values with respect to the pixel value of the center pixel 402. A threshold value is provided above and below centering on the matching pixel value, and it is determined whether or not the pixel value of the adjacent pixel to be determined is included in the range in which the pixel value is added or subtracted. Pixels that are determined to be included are further classified into hypothetical regions, and pixels that are not included are not classified into hypothetical regions. As a result, the hypothetical area is expanded as indicated by area 403.

  The hypothetical area is enlarged as indicated by the area 403, but it is determined whether or not the same condition is satisfied for the peripheral area after the enlargement. That is, for each pixel around each pixel in the region 403, it is determined whether or not the pixel value is within a certain threshold range. Alternatively, the determination may be made based on the pixel value of the pixel 402. As a result, the hypothetical region 403 further expands into the region 404 and becomes a blood vessel cross-sectional image 405.

[Expansion of each contour of blood vessel lumen and lipid core]
Next, the shape change of each area | region when the vascular lumen | bore extracted as mentioned above and the outline of a lipid core are each expanded and the overlapping area | region formed by it is calculated | required is demonstrated.

  FIG. 5 is a schematic view showing the contours of the extracted blood vessel lumen and lipid core. A blood vessel lumen 502 and a lipid core 503 are included inside the blood vessel 501. The blood vessel lumen 502 and the lipid core 503 may be adjacent to each other or may already overlap due to the resolution of the extracted image, but both are actually separated by the fibrous coating 230 shown in FIG. ing.

  FIG. 6 is a schematic view showing a case where the contours of the blood vessel lumen and the lipid core are expanded. Although the case where the blood vessel lumen 502 is expanded will be described here, the same processing is applied to the case where the lipid core 503 is expanded. A dilation process of morphological filtering is applied to the contour 600 of the blood vessel lumen 502 to be inflated by the inflating unit 105.

  In this dilation process, the contour of the processing target is expanded to the outer trajectory obtained by the moved component 601 by moving the determined component 601 along the outside of the processing target contour 600. It is processing to do. In FIG. 6, this component 601 is arranged along the contour 600. The components 601 are arranged outside the contour 600 so as to be in contact with the contour 600. The component 601 is assumed to be circular, but may be square, for example, and various shapes are conceivable. In addition, as described with reference to the expansion unit 105 in FIG. 1, it is assumed that the component 601 has a size of one pixel. However, depending on the size to be expanded, the component 601 may have various sizes of 2 pixels or more. it can.

  The arranged component 601 is moved along the contour 600 and returns to its original state after one round. The moving direction may be clockwise or counterclockwise. By moving the component 601 along the contour 600 in this way, the component 601 has a width instead of a point, so that a trajectory is formed along the outside of the component 601 and becomes a contour 602 after expansion. . This expanded contour 602 is a figure obtained by being expanded by the expanding portion 105.

  As described with reference to the inflating portion 105 of FIG. 1, it is not necessary to inflate all of the blood vessel lumen 502. For example, as shown in FIG. Since it is located, the inflating part 105 may inflate only the right side of the blood vessel lumen 502. Conversely, since the lipid core 503 is located on the right side, only the left side of the lipid core 503 may be expanded. Thus, even when only half of each is expanded, the time required for expansion can be shortened, but there is no problem in the calculation of the overlapping region. In addition, although only the left / right half of each contour is shown here, various lengths such as 1/3 or 2/3 of the contour may be defined as a range.

  FIG. 7 is a schematic view showing the contours of the blood vessel lumen and the lipid core after expansion. The blood vessel lumen 502 and the lipid core 503 shown in FIG. 5 are inflated as shown by the expansion contour 701 and the expansion contour 702 by the expansion process shown in FIG. In FIG. 7, the inside of the blood vessel 501 shown in FIG. 5 is divided into the case of the blood vessel lumen 502 and the case of the lipid core 502, and is shown separately in FIGS. 7A and 7B, respectively. Both are arranged in the same blood vessel 501.

[Calculation and display of overlapping part of vascular lumen region and lipid core region]
FIG. 8 is a schematic diagram for explaining the overlapping portion of the contour after expansion. Although the contours after expansion are shown in FIG. 7, all of them are in the blood vessel 501, so that the superimposed ones are shown in FIG. 8. Since the expansion profiles 701 and 702 are made larger than the original blood vessel lumen 502 and the lipid core 503, it is assumed that both overlap. Thereby, an overlapping region 801 is formed. In fact, since the lipid core 503 may not be large, the expansion contour 702 may not be so large and may not overlap with the expansion contour 701. In that case, the overlapping region 801 is not formed.

  The overlapping region calculation unit 106 obtains the shape of the overlapping region 801 from the expansion contours 701 and 702 formed by the expansion unit 105 as described above. Then, the internal area of the obtained overlapping region 801 is calculated. The area of the overlapping region can be obtained by counting the number of dots inside the overlapping region.

  FIG. 9 is a display screen showing the change in the area of the overlapping region as a graph. The display screen includes a graph 900 and an image 910. The graph 900 shows a change in the size of the area of the position along the length direction of the blood vessel. An image 910 shows a cross-sectional image at a position along the length direction of the blood vessel. As described above, the length direction of the blood vessel is arranged such that the upper side is the base, that is, the heart side, and the lower side is the end.

  In the graph 900, the upper side is the base direction and the lower side is the distal direction. However, not all are displayed, but a part of the blood vessel from the base to the distal end is displayed as the graph 900. The graph 900 displayed here can be scrolled in the designated direction by operating in the upward direction or the downward direction. As described above, the vertical direction is the blood vessel progression direction, while the horizontal direction indicates the area of the overlapping region 801 shown in FIG. When the waveform is located at the left end, this area is 0, and when the area is larger than 0, the waveform size corresponds to the size.

  The graph 900 is continuous from point 902 to point 903, where point 902 is distal as described above and point 903 is proximal as described above. Further, neither the terminal 902 nor the point 903 is displayed on the distal side or the base side by changing the display position instead of the terminal and base itself. Among them, for the position on the blood vessel where the overlapping region 801 shown in FIG. 8 is generated, the area is larger than 0, so that the waveform changes and a waveform 904 is obtained. On the other hand, when the overlapping region does not occur, that is, when the area is 0, the waveform change does not occur as indicated by the waveform 905.

  An image 910 is a cross-sectional image of the coronary artery when viewed by Stretched MPR. Similarly to the graph 900, the image 910 is also arranged along the length direction of the blood vessel, and each element in the length direction of the image 910 is a cross-sectional image of the blood vessel. The blood vessels do not travel in a straight line, but progress while winding in the body. When this is viewed by Stretched MPR, it is shown as an image 910.

  Each element in the longitudinal direction of the image 910 is corrected so as to have a shape seen from one side, so as to have a constant shape seen from one direction for each cross section. For this correction, the direction of the blood vessel is prescribed, and the outward direction of the blood vessel is based on the rule. Since the blood vessel travels in a winding direction, the directions are therefore different, but the cross-sections are unified according to the definition of this direction. As a result, an image obtained by connecting the unified cross-sectional images as elements for each length position is displayed as an image 910.

  The outward direction of the blood vessel is outward in the direction perpendicular to the core line penetrating the inside of the blood vessel. Then, when the advancing direction of the blood vessel is winding, the outside of the curve is the outward direction of the blood vessel, so the cross sections are changed and arranged so that the direction is unified. In the image 910 formed in this way, the left side shows a blood vessel cross-sectional film, and the effect of correction appears strongly every time it goes to the right side. In this embodiment, Stretched MPR is shown as an example, but other display forms such as Curved MPR for displaying a cross-sectional image in the length direction of the blood vessel may be used.

  Further, as a display form, the display unit 30 displays the blood vessel cross-sectional image at each position in the length direction of the blood vessel in association with information based on the size of the overlapping region of each cross section under the control of the display control unit 20. To do. That is, the position in the length direction of the blood vessel indicated by the graph 900 and the position of the cross-sectional image of the blood vessel indicated by one of the regions of the image 910 are displayed in association with each other.

  As described above, in this embodiment, the area of the overlapping region 801 is obtained along the blood vessel traveling direction, whereby the degree of proximity between the blood vessel lumen and the lipid core at each position of the blood vessel can be scored. Thus, by calculating | requiring a proximity degree from the area of the duplication area | region 801, since the area is large by this obtained proximity degree, it is judged that both are approaching. Thereby, even when images of the blood vessel lumen and the lipid core can be obtained only at a low resolution with respect to the thickness of the fibrous coating, the degree of proximity is determined, so that the risk of unstable plaque can be known.

  Note that the absolute risk level of each cross section may be obtained by the overlapping area calculation unit 106. In this case, the overlapping region calculation unit 106 stores in advance the critical point and threshold value of the overlapping region area, determines whether or not the obtained area exceeds the threshold value, and the position where the area exceeds the threshold value. It may be displayed that a certain degree of risk has been reached for the blood vessel cross section. Thus, it is determined whether or not a blood vessel cross section that reaches a certain risk level exists in the blood vessel, so that it is possible to know that the blood vessel to be examined is dangerous.

  FIG. 10 is a cross-sectional image of a blood vessel in which the overlapping region is maximized. 5 to 8, images obtained by extracting the blood vessel cross section and the inner blood vessel lumen and the outline of the lipid core are displayed. FIG. 10 shows an image of the actual cross section of the blood vessel itself. Here, a cross-sectional image of the blood vessel 501 is displayed at a position on the blood vessel where the area of the overlapping region is the maximum for the unstable plaque to be examined. When the cross-sectional image of the blood vessel 501 is displayed, the lipid core 503 therein is displayed. In addition, not only when an unstable plaque is displayed, but also when a point on the graph 900 is designated, a cross-sectional image of the blood vessel 501 can be displayed at that position. At this time, the lipid core 503 is similarly displayed, and when it does not exist, the lipid core 503 is not displayed.

  Up to this point, the size and the degree of proximity between the blood vessel lumen and the lipid core due to the area of the overlapping region 801 have been obtained. However, when obtaining the size, it is not necessarily limited to the area. It can also be determined by axis. Next, a process for obtaining the major axis and the minor axis inside the overlapping area 801 will be described.

  FIG. 11 is a schematic diagram for explaining the arrangement of the long axis and the short axis inside the overlapping region. In FIG. 11, only the overlapping area 801 shown in FIG. 8 is extracted and shown. Then, the major axis 1101 and the minor axis 1102 are obtained from the overlapping region 801.

  First, the overlapping area 801 is specified, and the two farthest points in the outline of the overlapping area are obtained. Then, the major axis is obtained by connecting the two points. Next, the longest line among the lines drawn in the overlapping region perpendicular to the long axis is obtained. Let the obtained line be the short axis. The lengths of the major axis and the minor axis obtained in this way are arranged in the same manner as the above-described area and plotted. The major axis and the minor axis may be respectively displayed, and the total value and the difference may be respectively displayed. Moreover, you may display side by side with the area of an overlapping area | region.

  By obtaining the major axis and the minor axis, the size of the overlapping region can be obtained, and the major axis and the minor axis can be used as an index indicating the size of the overlapping region. Further, by obtaining a difference value obtained by subtracting the minor axis from the major axis, the shape of the overlapping region can be obtained from the difference value.

  FIG. 12 is a display screen in which changes of the long axis and the short axis are arranged and shown as a graph. The display screen includes a graph 1201, a graph 1202, and an image 1203. The graph 1201 shows the change in the length of the long axis in the overlapping region in the blood vessel cross section at the position along the length direction of the blood vessel, and the graph 1202 is in the overlapping region in the blood vessel cross section at the position along the length direction of the blood vessel. The change of the length of a short axis is shown. An image 1203 shows a cross-sectional image at a position along the length direction of the blood vessel.

  As in the image 910 shown in FIG. 9, the blood vessel length direction is arranged such that the upper side is the base, that is, the heart side, and the lower side is the end. The graph 1202 and the image 1203 have a base direction on the top and a terminal direction on the bottom, but not all of them are displayed but a part of the blood vessel from the base to the end is displayed. The scrollable point is the same as in FIG. The left-right direction indicates the length of the major axis and the minor axis. When the waveform is located at the left end, this length is 0, and when the length is greater than 0, the waveform size corresponds to the size.

  Here, the long axis is a straight line connecting two farthest points on the outline of the overlapping region. The overlapping area calculation unit 106 obtains the long axis as follows. First, one point on the contour is determined. Next, the coordinates of the farthest point on the contour with respect to the coordinates of the point are obtained, and the lengths of both are obtained. In this way, the length is obtained for each point on the contour, and the largest value is obtained from them. The line connecting the two points having the largest length value is the major axis.

  Further, the overlapping area calculation unit 106 obtains the short axis as follows. The short axis is the longest of the straight lines connecting the outline of the overlapping region perpendicularly to the long axis. First, move the reference point from the end of the long axis. A straight line perpendicular to the long axis is extended so as to pass through this reference point, and a point where the straight line intersects the contour is obtained. Since it gathers at one point at the end of the long axis, the length of the vertical straight line is zero. Then, while advancing the reference point along the long axis into the contour, a straight line passing through the reference point perpendicular to the long axis is obtained, a point where the straight line intersects the contour is obtained, and a length connecting the contours is obtained. This length is determined over the point where the reference point is opposite the starting point. Of the lengths thus determined, the longest is the short axis.

  An image 1203 is a cross-sectional image of the coronary artery when viewed by Stretched MPR. The image 1203 is the same as the image 910 shown in FIG. Similar to the graphs 1201 and 1202, the image 1203 is arranged along the length direction of the blood vessel, and each element in the length direction of the image 1203 is a cross-sectional image of the blood vessel. The blood vessels do not travel in a straight line, but progress while winding in the body. When this is viewed by Stretched MPR, it is shown as an image 1203.

  As described above, the change in the length of the short axis of the long axis is indicated along the blood vessel traveling direction, so that the overlapping region is long when the difference between the long axis and the short axis is large, and the difference is small. Shows that the overlap area is round. When the overlapping region is long, the proximity of the lipid core and the blood vessel lumen is long, but the proximity is shallow. Conversely, when the overlapping area is round, the proximity degree between the two is short, but the proximity degree is deep. In general, the larger the overlapping area, the higher the risk of unstable plaque, but in the case of the same size, the longer the overlapping area, the longer the adjacent part will be, so unstable plaque It is a danger. By obtaining the length of the major axis and the minor axis in this way, the shape of the overlapping region can be obtained, and thereby information for determining the degree of risk assumed from the shape can be given.

  Furthermore, regarding the area of the overlapping region, when an area is generated for each position of a blood vessel, only one point has a specific large area, but has a certain area over a certain length before and after that. The reason is that when a lipid core that produces unstable plaque in a blood vessel occurs, it does not become unstable only at a certain cross section within the blood vessel, but it becomes unstable over a certain length of the region around it. is there. Then, the point which calculates | requires the risk of the unstable plaque of a lipid core unit by calculating | requiring the spatial magnitude | size of an overlap area | region by the following is demonstrated.

  FIG. 13 is a display screen showing a change in the area of the overlapping region as a set in the length direction of the blood vessel. The display screen includes a graph 1304 and an image 1307. A graph 1304 shows a change in the size of the area of the position along the length direction of the blood vessel. An image 1307 shows a cross-sectional image at a position along the length direction of the blood vessel. As described above, the length direction of the blood vessel is arranged such that the upper side is the base, that is, the heart side, and the lower side is the end. This graph 1304 is the same as the graph 900.

  Since this graph 1304 includes a region where the area larger than 0 is shown as a waveform 1306, a region 1305 including the waveform 1306 is extracted. When the extracted region 1305 is enlarged, a region 1301 is obtained.

  An image 1307 is a cross-sectional image of the coronary artery when viewed by Stretched MPR. The image 1307 is the same as the image 910 shown in FIG. Similar to the graph 1304, the image 1307 is arranged along the length direction of the blood vessel, and each element in the length direction of the image 1307 is a cross-sectional image of the blood vessel. The blood vessels do not progress in a straight line, but progress while winding in the body. When this is viewed by Stretched MPR, it is shown as an image 1307.

  A region 1301 includes a waveform 1303. Among the positions included in the waveform 1303, the position corresponding to the vertex 1302 has the largest area of the overlapping region. Although it is determined that the risk of unstable plaque is highest at the position of the vertex 1302, the area of the overlap region does not suddenly increase only at that position. The area of the overlapping region has a width as shown by the waveform 1303, gradually increasing from 0 at both ends of the width, and maximizing the area at the vertex 1302.

  First, a waveform 1303 including a certain region is generated by the lipid core. Therefore, the size of the overlap region obtained for each position is obtained as a volume by adding up over the section where the area is generated. First, since the area is obtained for each position as the number of dots inside the overlapping region, the total value of the areas at each position in this section is obtained as the volume. In FIG. 13, the area inside the region displayed as the waveform 1303 is obtained, and thereby the volume over the section where the overlapping region is generated is obtained. As indicated by the waveform 1303, unstable plaque is generated as a lump, and when the lump is passed, it becomes a stable plaque and the area of the overlapping region becomes zero. Therefore, the volume can be obtained in units of this mass. This cluster is generated in each lipid core unit, and the volume is determined for each lipid core.

  That is, the area obtained from the number of dots inside the overlapping region is added from one end of the waveform. Although the area is 0 at the extreme end, the area gradually increases, so add that value, add the maximum area at the vertex 1302, and add the decreasing area, Finally, the addition is repeated until the area becomes 0 at the other end.

  Thereby, the volume of the overlapping region formed from one lipid core specified by the waveform 1303 can be obtained. Thus, by obtaining the volume of the overlapping area, the spatial size of the overlapping area can be obtained. Although the risk of unstable plaque can also be determined by the area of the overlap region, it is a risk at a specific location, and of course it can be seen that the lipid core containing that specific location is dangerous, It cannot be said that it is a judgment for the entire lipid core. For example, even if the area of the overlap region at a specific point is the same, the risk of vulnerable plaque depends on whether it is a long or narrow range when viewed from the length of the blood vessel. Degree judgment is different.

  By calculating it as the volume of the overlapping region as described above, the risk of unstable plaque in one entire lipid core can be determined. In addition, since scoring can be performed as a volume, when there are a plurality of lipid cores in one blood vessel, the difference in the score among the plurality of lipid cores is compared to determine which lipid core is the most. It is possible to determine whether the degree of risk is high.

  FIGS. 14A and 14B are cross-sectional views of the coronary artery cut in a direction along the blood vessel core line when there are a plurality of lipid cores. The direction of the cross section is the same as in FIG. The lipid core 1401, lipid core 1402, and lipid core 1403 are included in this order from the left side of the blood vessel cross section. When any one of them is designated, the display unit 30 displays the graph 1304 shown in FIG. 13 according to the output of the image generation unit 107 and the control of the display control unit 20, and corresponds to the unstable plaque. An area 1305 and a waveform 1306 inside the area 1305 are displayed. Further, the display unit 30 displays a cross-sectional image of the blood vessel as shown in FIG. 10 for the position on the blood vessel where the area of the overlapping region is maximized.

  FIG. 15 is a display example when information on the risk level of each lipid core is displayed. As shown in FIG. 14, when a lipid core 1401, a lipid core 1402, and a lipid core 1403 are included, they are shown in the table as lipid cores A, B, and C, respectively. And next, the position in each blood vessel is shown. When the position is specified from the left side of FIG. 14, since the lipid core 1401 is closest to the starting point, the position 50 and the lipid core 1402 are next closest, so that the position 200 and the lipid core 1403 are the farthest, so that the position 350 is obtained. The position in the blood vessel is measured from the base, but may be measured from the end.

  Since the volume of the overlapping region is obtained for each lipid core, it is displayed as 30, 120, 90, respectively. Among them, the lipid core 1402 (B) has the largest overlap volume, so the rank is first. Since lipid core 1403 (C) is the next largest, the ranking is second. Since the lipid core 1401 (A) is the smallest, the rank is third. As described above, in this embodiment, a plurality of lipid cores included in the same blood vessel can be displayed side by side, and it is possible to indicate to the user which unstable plaque due to which lipid core is most dangerous.

  The display unit 30 displays the lipid core risk list shown in FIG. 15 described above by the output of the image generation unit 107 and the control of the display control unit 20. In this case, only FIG. 15 may be displayed alone, or may be displayed together with the blood vessel image of FIG. When both FIG. 14 and FIG. 15 are displayed, the degree of risk can be confirmed numerically while confirming the arrangement of the lipid core on the screen.

  As described above, the rank of the risk level is determined for each lipid core and the rank is displayed. However, the overlap area calculation unit 106 may calculate the absolute risk level rather than the relative level of each lipid core. Good. In this case, the overlapping region calculation unit 106 stores in advance the critical point and threshold value of the volume of the overlapping region, determines whether or not the obtained volume exceeds the threshold value, and determines the lipid whose volume exceeds the threshold value. The fact that a certain degree of risk has been reached may be displayed for the core. As a result, it is possible to determine whether or not there is a lipid core that has reached a certain degree of risk in the blood vessel, so it is possible to know whether or not a risk has occurred for each blood vessel.

[About a series of operations]
FIG. 16 is a flowchart showing processing for executing analysis processing for a series of coronary arteries. First, CT data of the heart region is acquired by the X-ray CT apparatus 10 (step S1001). Next, the blood vessel acquisition unit 103 extracts a coronary artery from the acquired CT data (step S1002). That is, the blood vessel acquisition unit 103 selects one coronary artery to be examined and acquires its three-dimensional image. Next, a blood vessel core line is extracted for the extracted coronary artery (step S1003). The blood vessel core line is a line connecting the centers of blood vessels, and is a line along the direction of the blood vessel. Then, the processing of each blood vessel cross section is executed for a series of coronary arteries.

  First, the control unit 102 specifies the first blood vessel cross section. Since this processing starts from the base, the control unit 102 specifies the first cross section by selecting the surface closest to the base from the surfaces orthogonal to the blood vessel core line (step S1004). However, the present invention is not limited to this, and conversely, the process may be started from the end of the blood vessel. In this case, the process is repeated from the end to the base.

  Next, the overlap region calculation process is executed for the identified blood vessel cross section (step S1005). This process will be described later with reference to FIG. Thereby, the overlapping region as a result of expanding the lipid core and the blood vessel lumen included in one blood vessel cross section is calculated. Here, the medical image processing apparatus 100 acquires a blood vessel cross-section for each position along the blood vessel core line, and obtains an overlapping region. The overlapping area calculation unit 106 obtains the area inside this overlapping area for each position along the blood vessel core line.

  After the calculation, the control unit 102 determines whether or not the processing has been completed for all cross sections (step S1006). When it is determined that the process has been completed (step S1006: Yes), the control unit 102 shifts to a display process (step S1008) and ends a series of processes. In other words, when it is determined that the processing of the last cross section corresponding to the end of the blood vessel has been completed and the processing has been performed for all the blood vessel inspection targets, the process proceeds to display processing.

  When it is determined that the process has not been completed (step S1006: No), that is, after processing other than the last section corresponding to the end of the blood vessel, the next section is specified (step 1007). That is, for the next position along the blood vessel core line, the blood vessel cross section orthogonal to the blood vessel core line is set as the next cross section. After specifying the next blood vessel cross section, the process returns to step S1005 to execute processing for the next blood vessel cross section.

  Next, the display process in step S1008 will be described. Since the area is obtained at each position of the blood vessel, as shown in FIG. 9, the obtained area value is sequentially plotted from the base at each position of the blood vessel in association with each position along the blood vessel core line. To do. A Stretched MPR image is also displayed side by side in association with each position along the blood vessel core line. Further, instead of displaying the area of the overlapping region, as described with reference to FIG. 11, the lengths of the major axis and the minor axis of the overlapping region can be obtained and displayed.

  FIG. 17 is a flowchart showing overlap area calculation processing. In the flowchart shown in FIG. 16, when the first cross section (step S1004) or the next cross section (step S1007) is specified and the overlap area calculation process (step S1005) is performed for the cross section, the process proceeds to step S1101 in FIG.

  First, the blood vessel acquisition unit 103 extracts a blood vessel cross section for the specified cross section (step S1101). This blood vessel cross section is as shown in cross sections 210 and 220 in FIGS. Next, it is determined whether or not the extracted blood vessel cross section includes a lipid core (step S1102). As described with reference to FIG. 4, the lipid core extraction unit 120 stores in advance pixel values estimated to correspond to lipid cores. Therefore, it is determined whether or not a pixel corresponding to this pixel value is included in the target blood vessel cross section. If it is determined that it is not included (step S1102: No), the overlapping area is set to 0 (step S1109), and the series of processes is terminated.

  When it is determined that it is included (step S1102: Yes), the lipid core extraction unit 120 extracts a lipid core region (step S1103). First, since a pixel having a pixel value corresponding to a lipid core is specified as described above, the region expansion method shown in FIG. 4 is applied around that pixel. Since the lipid core region is extracted by the region expansion method, the closed curve extraction unit 140 extracts the contour of the extracted lipid core region. The obtained contour is as shown in FIG. Next, the blood vessel lumen extraction unit 130 extracts a blood vessel lumen region (step S1104). Similar to the lipid core, the region expansion method is applied. Similar to the case of the lipid core, the closed curve extraction unit 140 extracts the contour of the extracted blood vessel lumen region.

  Next, the expansion unit 105 expands both regions (step S1105). That is, since the closed curved surface and the contour are obtained for both regions of the lipid core and the blood vessel lumen, the expanding portion 105 expands both the obtained contours as shown in FIGS. Go.

  Next, it is determined whether or not both expanded contours overlap (step S1106). Although FIG. 8 shows a case where the two contours overlap to form the overlapping region 801, there are cases where both the contours are not formed. At that time, the expansion contours 701 and 702 are separated from each other, and the overlapping region 801 is not formed. When it is determined that there is no overlap (step S1106: No), the overlap area is set to 0 and a series of processing is ended (step S1109).

  When it is determined that they overlap (step S1106: Yes), the overlapping area calculation unit 106 extracts the overlapping area (step S1107), calculates the area of the overlapping area (step S1108), and ends the series of processes. To do. In step S1108, instead of calculating the area of the overlapping region, the lengths of the major axis and the minor axis may be obtained as shown in FIG. In addition to obtaining the area of the overlapping region, the lengths of the major axis and the minor axis may be obtained.

  As described above, each of the blood vessel lumen and the lipid core is extracted and expanded, and the case where display processing related to this is performed by obtaining the size of the overlapping area, the major axis, the minor axis, etc. The risk of unstable plaque in lipid core units can also be obtained by adding the overlapping regions over the blood vessel direction in relation to FIG. 13 and obtaining the volume thereof. Next, the series of flows will be described with a flowchart.

  FIG. 18 is a flowchart for explaining the process of adding the areas of overlapping regions to the volume in the blood vessel direction to make the volume. Processing is started from the base of the blood vessel, but instead of starting processing from the base of the blood vessel, a specific lipid core is designated from FIG. 14, and the starting point of the lipid core, that is, the area of the overlapping region has a value from 0 The processing may be started from a position where the switching is made.

  First, it is determined whether or not the area of the overlapping region is area> 0 (step S1201). If the area is not> 0, that is, if the area is 0 (step S1201: No), the process proceeds to step S1208. That is, since the overlapping area has not been reached, the position is advanced further. If area> 0 (step S1201: Yes), that is, the area is switched to a position having a value from 0, the area and the position are stored (step S1202). This position is a position where an overlapping region due to the expansion of the lipid core and the blood vessel lumen has started to occur. Then, the position is moved (step S1203).

  Next, it is determined whether or not the area is 0 (step S1204). If the area is not 0 (step S1204: No), the process returns to step S1202 to move the position until the area is no longer 0. Repeat, save area and position. When the area becomes 0 (step S1204: Yes), the area and position are stored (step S1205). This position is the position where the overlapping region due to the expansion of the lipid core and the blood vessel lumen ends. As a result, it is possible to specify a section in which an overlapping area exists, with the position stored first in step S1202 as the start point and the position stored in step S1205 as the end point.

  Since the section where the overlapping area exists is obtained, the volume of the overlapping area in the designated section is obtained (step S1206). This designated section is an area surrounded by the waveform 1303 in FIG. 13, and the volume of the overlapping area of the designated section is obtained by obtaining the area in the area enclosed by the waveform 1303. Since the area on the graph in the specified section, that is, the volume of the overlapping region, has been obtained, this volume is assigned in accordance with the lipid core number (step S1207). That is, for the first designated section, the lipid core number is set to 1, and the volume of the overlapping region corresponding to the number is assigned. If the designated section exists first, the next number is obtained and the volume of the overlapping area corresponding to the number is assigned. When the lipid core number 2 has been previously assigned, the volume of the overlapping region obtained here is assigned as the lipid core number 3 for the designated section obtained next.

  The processing is repeated as described above to determine whether or not the end of the blood vessel has been reached (step S1208). When it is determined that the end of the blood vessel has been reached (step S1208: Yes), a series of processing ends. When it is determined that the end of the blood vessel has not been reached (step S1208: No), the position is moved (step S1209), and the process returns to step S1201 and is repeated.

  As described above, in this embodiment, by obtaining the areas of overlapping regions when the contours of the blood vessel lumen and the attached substance are expanded along the blood vessel traveling direction, the blood vessel lumen and the attached substance at each position of the blood vessel are obtained. The degree of proximity can be scored. By determining the degree of proximity in this way, the degree of proximity is determined even when images of the blood vessel lumen and lipid core are obtained only at a low resolution relative to the thickness of the fibrous coating. You can know the degree.

DESCRIPTION OF SYMBOLS 10 X-ray CT apparatus 20 Display control part 30 Display part 101 Image data memory | storage part 102 Control part 103 Blood vessel acquisition part 104 Extraction part 105 Expansion part 106 Overlapping area calculation part 107 Image generation part

Claims (14)

An extraction means for obtaining a cross-section of the blood vessel based on image data of a region including the blood vessel and extracting the outline of the lumen of the blood vessel and the outline of the adhered substance adhering to the blood vessel from the cross-section;
Expansion calculation means for expanding at least a portion of one or both of the extracted outline of the lumen and the outline of the attached substance by a predetermined amount;
For each cross section of the blood vessel, region calculation means for obtaining the size of the overlapping region when the contour of the lumen expanded by the expansion calculation means and the contour of the attached substance overlap,
A medical image processing apparatus comprising:   The medical image processing apparatus according to claim 1, wherein the extraction unit obtains a core line of the blood vessel, obtains a cross section perpendicular to the core line, and sets the cross section as the cross section of the blood vessel.   The medical image processing apparatus according to claim 1, wherein the predetermined amount expanded by the expansion calculation unit is one pixel.   The medical image processing apparatus according to claim 1, wherein the attached substance is a lipid core. Display means for displaying information based on the size of the overlapping area respectively obtained by the area calculation means at each position in the direction of the blood vessel core line;
The medical image processing apparatus according to claim 2, further comprising:   The medical image processing apparatus according to claim 1, wherein the size of the overlap region obtained by the region calculation means is an area of the overlap region.   The medical image processing apparatus according to claim 1, wherein the size of the overlapping area obtained by the area calculating unit is a length of a major axis and a minor axis in the overlapping area.   The medical image according to claim 5, wherein the display unit displays, for each cross section of the blood vessel, a graph in which a position in the length direction of the blood vessel and a size of the overlapping region are associated with each other. Processing equipment.   The said display means displays the blood vessel cross-sectional image of each position of the length direction of the said blood vessel corresponding to the information based on the magnitude | size of the overlap area | region of each cross section, The display means of Claim 5 characterized by the above-mentioned. Medical image processing apparatus.   2. The medical image processing apparatus according to claim 1, wherein the expansion calculation unit expands the extracted whole outline of the lumen and the whole outline of the attached substance by a predetermined amount.   The expansion calculation means moves a predetermined element along one or both of the outline of the lumen and the outline of the attached substance, and moves the outline of the lumen to an outer locus obtained by the moved element. The medical image processing apparatus according to claim 1, wherein one or both of contours of the adhering substance are expanded. Section extracting means for extracting a section in which the area of the overlapping region is continuously larger than 0 in the direction of the core of the blood vessel;
Summing means for summing each area in the section extracted by the section extracting means, and obtaining a total value for each section;
The medical image processing apparatus according to claim 1, further comprising:   The medical image processing apparatus according to claim 12, wherein the summing unit ranks the sections in the order of the total value of the sections. An extraction step for obtaining a cross section of the blood vessel based on image data of a site including the blood vessel and extracting the outline of the lumen of the blood vessel and the outline of the adhering substance attached to the blood vessel from the cross section;
An expansion calculation step of expanding at least a portion of one or both of the extracted outline of the lumen and the outline of the attached substance by a predetermined amount;
For each cross section of the blood vessel, an area calculation step for obtaining a size of an overlapping area when the outline of the lumen expanded by the expansion calculation step and the outline of the attached substance overlap;
A medical image processing program characterized by causing a computer to execute.