JP7045256B2 - Heart disease diagnosis support program, heart disease diagnosis support method and heart disease diagnosis support device - Google Patents

Description

The present invention relates to a heart disease diagnosis support program, a heart disease diagnosis support method, and a heart disease diagnosis support device.

Traditionally, myocardial scintigraphy involves injecting a subject with a radioisotope-labeled drug (radiopharmaceutical), photographing the radiation emitted by the injected drug, and computer-processing the radiation dose into an image of the myocardium. This is a test that images blood flow and energy metabolism. This test is widely used clinically because it is less invasive and has the excellent characteristics of being able to obtain various physiological and biochemical information about the heart by selecting appropriate radiopharmaceuticals. .. For example, the so-called Coronary Flow Reserve (CFR), which is determined by the ratio to the myocardial blood flow during drug or exercise load, is used as an index showing how much the coronary artery is damaged by heart disease. ing.

As mentioned above, CFR is used as an index showing the damage to the coronary arteries due to the disease, but it may be different from the partial coronary flow reserve (FFR). It has been pointed out (see Non-Patent Documents 1 and 2). Then, it is proposed to utilize this mode of divergence in determining the treatment policy.

On the other hand, 17 segments recommended by the AHA (American Heart Association) and 3 segments representing a roughly controlled region of coronary arteries have been conventionally used as regional division criteria for displaying polar coordinates of myocardial scintigram.

Nils P. Johnson et al., Is Discordance of Coronary Flow Reserve and Fractional Flow Reserve Due to Methodology or Clinically Relevant Coronary Pathophysiology ?, 2012, p.193-202 Tim P. van de Hoef et al., Physiological Basis and Long-Term Clinical Outcome of Discordance Between Fractional Flow Reserve and Coronary Flow Velocity Reserve in Coronary Stenoses of Intermediate Severity., 2014, p.301-11

In the commonly used diagnosis of myocardial scintigram, evaluation is performed in the above-mentioned uniformly classified segments. However, since these division methods are set by a uniform standard, they may not match the individual coronary artery running.

That is, since there are individual differences in the running of the coronary arteries of the heart, even if divisions such as 17 segments and 3 segments are performed, each division region may not reflect the region controlled by the individual's coronary arteries. For example, there is a divergence such that the place represented as the right coronary artery (hereinafter referred to as RCA) region in 3 Segment was actually the left anterior descending artery (hereinafter referred to as LAD) region. Sometimes. Therefore, with such a method, it is difficult to identify the diseased part of the coronary artery while considering individual differences in the running of the coronary artery of the heart.

Also, if one of the coronary arteries is damaged due to some disease, in the conventional division method, the damaged coronary arteries may span a plurality of division regions. Therefore, there is also a problem that it is difficult to identify the site where the disorder occurs only by the conventional region division standard.

The present invention has been made in view of the above problems, and is a heart disease diagnosis support program, a heart disease diagnosis support program, which can more easily support correct diagnosis of the heart reflecting coronary artery running regardless of individual differences of subjects. It is an object of the present invention to provide a diagnosis support method and a heart disease diagnosis support device.

In the heart disease diagnosis support program according to one aspect of the present invention, a computer is used to obtain a three-dimensional coronary artery image in a subject, at least a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medical image at rest and under load, and the subject. A process of acquiring an index value indicating a local stenosis state of each coronary artery, a process of dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image, and a process of dividing the left ventricle into a plurality of regions. A process of extracting a boundary line partitioning a plurality of the regions obtained by dividing the three-dimensional morphological image and dividing the three-dimensional cardiac dynamic nuclear medicine image using the boundary line and a division of the three-dimensional cardiac dynamic nuclear medicine image. A process of calculating the CFR of the subject for each of the regions, a process of obtaining an index for determining a treatment policy based on a comparison between the CFR in the region and the index value corresponding to the region, and the index. Is executed on the display unit.

The method for supporting the diagnosis of heart disease according to another aspect of the present invention includes a three-dimensional coronary artery image in a subject, a three-dimensional morphological image of at least the left ventricle, a three-dimensional cardiac dynamic nuclear medical image at rest and under load, and each of the above-mentioned subjects. An information acquisition step for acquiring an index value indicating a local stenosis state of the coronary artery, and a first division step for dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image. In the first division step, a boundary line for partitioning a plurality of the divided regions of the three-dimensional morphological image of the left ventricle is extracted, and the three-dimensional cardiac dynamic nuclear medicine image is divided using the boundary line. The CFR calculation step for calculating the CFR of the subject for each of the two division steps, the three-dimensional cardiac dynamic nuclear medicine image divided in the second division step, and the region calculated in the CFR calculation step. In the index calculation step for obtaining an index for determining a treatment policy based on the comparison between the CFR in the above region and the index value corresponding to the region, and a display for displaying the index calculated in the index calculation step on the display unit. Including steps.

The heart disease diagnosis support device according to another aspect of the present invention includes a three-dimensional coronary artery image in a subject, a three-dimensional morphological image of at least the left ventricle, a three-dimensional cardiac dynamic nuclear medical image at rest and under load, and the subject. A first division process for dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on an information acquisition unit that acquires an index value indicating a local stenosis state of each coronary artery and the three-dimensional coronary artery image. A boundary line partitioning a plurality of the divided regions of the three-dimensional morphological image of the left ventricle is extracted by the unit and the first division processing unit, and the three-dimensional cardiac dynamic nuclear medicine image is obtained using the boundary line. The second division processing unit to be divided, the CFR calculation unit that calculates the CFR of the subject for each of the regions in which the three-dimensional cardiac dynamic nuclear medicine image is divided by the second division processing unit, and the CFR calculation unit. Based on the comparison between the calculated CFR in the region and the index value corresponding to the region, the index calculation unit for obtaining an index for determining the treatment policy and the index calculated by the index calculation unit are displayed. It is characterized by including a display processing unit for displaying on the unit.

According to the heart disease diagnosis support program, the heart disease diagnosis support method, and the heart disease diagnosis support device according to the present invention configured as described above, the diagnosis of heart disease can be supported more accurately.

It is a functional block diagram which shows the myocardial image processing apparatus and its peripheral apparatus which concerns on embodiment of this invention. It is explanatory drawing explaining the method of Voronoi division of the 3D morphological image of the left ventricle (the 1). It is explanatory drawing explaining the method of Voronoi division of the 3D morphological image of the left ventricle (the 2). It is explanatory drawing explaining the method of Voronoi division of the 3D morphological image of the left ventricle (the 3). It is explanatory drawing explaining the method of Voronoi division of the 3D morphological image of the left ventricle (the 4). It is a flowchart of the whole processing performed by the myocardial image processing apparatus and its peripheral apparatus. It is explanatory drawing which shows the result of the Voronoi division processing by the 1st division processing unit. It is explanatory drawing which shows the result of the polar coordinate expansion processing of a boundary line by the boundary line polar coordinate expansion part. It is a screen block diagram which shows the structure of the screen which is used for the operation of extracting the coronary artery and the left ventricular myocardium. It is a screen block diagram which shows the structure of the screen which is subjected to the operation of selecting the site which becomes the starting point of the region division from the extracted coronary arteries. It is a screen block diagram which shows the structure of the screen which is used for the calculation operation of the control area. It is explanatory drawing which shows the comparison result with this method and the method of determining a dominant area by a conventional 3 segment. It is explanatory drawing which shows the comparison result with this method and the method of determining a dominant area by a conventional 17 segment. It is an example of an image obtained by superimposing a three-dimensional coronary artery image and a three-dimensional left ventricle image obtained from a three-dimensional CT image which is an example of a morphological image.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are designated by the same reference numerals, and the description thereof will be omitted as appropriate. The following description is merely an example of a preferred embodiment, and is not intended to limit the scope of the present invention.

First, the outline of the present embodiment will be described. FIG. 1 is a functional configuration diagram showing a heart disease diagnosis support device 100 and its peripheral devices according to an embodiment of the present invention. Here, as peripheral devices, an image pickup device 110, an operation reception unit 120, a display device 130 (display unit), and a database 140 are exemplified, but the peripheral devices are not limited thereto.

The imaging device 110 captures a nuclear medicine image of a subject and provides the image to the heart disease diagnosis support device 100. Examples of such an imaging device 110 include a nuclear medicine imaging device such as a SPECT (Single Photon Emission Computed Tomography) device or a PET (Positron Emission Tomography) device. Nuclear medicine images taken by these devices are administered with a drug labeled with a specific radioisotope (hereinafter referred to as radiopharmaceutical drug), and gamma rays emitted directly or indirectly from the radiopharmaceutical drug are captured by a dedicated camera. Obtained by detecting and rebuilding.

As radiopharmaceuticals, Talium chloride ( ²⁰¹ TlCl) injection, Tetrophosmin technetium ( ^99m Tc Tetrofosmin) injection, Hexakis (2-methoxyisobutylisonitrile) technetium ( ^99m Tc MIBI) injection, 15 -(4-Iodophenyl) -3 (R, S) -methylpentadecanoic acid ( ¹²³ I) injection, and ¹³ N-ammonia and rubidium chloride ( ⁸² RbCl) injection as PET preparations. Also, in a single test, a radiopharmaceutical consisting of two different nuclides may be administered.

The heart disease diagnosis support device 100 includes an information acquisition unit 102, a first division processing unit 103, a boundary pole coordinate expansion unit 104, a cardiac dynamic nuclear medicine image normalization unit 105, a second division processing unit 106, and a CFR calculation unit 107. , An index calculation unit 108 and a display processing unit 109.

The information acquisition unit 102 is a three-dimensional coronary artery image in the subject, at least a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medicine image at rest and during loading, and an index value representing a local stenosis state of each coronary artery in the subject. To get. The acquisition here includes reading out from the database 140 an image or an index value that is generated in advance and is stored in the database 140. The three-dimensional coronary artery image is preferably a coronary artery image obtained by imaging the coronary artery by an inspection method using a modality (not shown) different from that of the imaging device 110, such as CT and contrasted or non-contrast MRI. On the other hand, it refers to an image that has undergone three-dimensional image processing, but is not limited to this. That is, any image may be used as long as the coronary artery of the subject is depicted. The three-dimensional coronary artery image may be taken separately from the three-dimensional morphological image described later, but the coronary artery is extracted on the three-dimensional morphological image in which the coronary artery is depicted and subjected to three-dimensional image processing. It may be a thing. In the latter case, the three-dimensional coronary artery image is generated as one image together with the three-dimensional morphological image. Examples of the three-dimensional image processing include MPR (Multi Planar Reconstruction) processing, three-dimensional coronary artery extraction processing, and three-dimensional left ventricle extraction processing.

"At least the left ventricle" includes a combination of the left ventricle and the right ventricle in addition to the left ventricle. The "three-dimensional morphological image of at least the left ventricle" is an image obtained by performing three-dimensional image processing on a morphological image obtained by imaging only the left ventricle, and an image obtained by imaging both the left ventricle and the right ventricle. An image obtained by performing three-dimensional image processing on the obtained morphological image is included. The morphological image is not particularly limited as long as it is an image that can recognize the morphology of the heart. The morphological image includes, for example, an image obtained by three-dimensional image processing of an image captured by an inspection method using a modality (not shown) different from that of the image pickup device 110, such as CT and contrasted or non-contrast MRI. An image obtained by subjecting the nuclear medicine image obtained by the image pickup apparatus 110 to three-dimensional image processing can be used. FIG. 14 shows an image obtained from a three-dimensional CT image which is an example of a morphological image. In this image, the three-dimensional coronary artery image of the left ventricle also extracted from the CT image is superimposed and displayed on the three-dimensional morphological image of the left ventricle extracted from the CT image.

The cardiac dynamic nuclear medicine image is an image obtained by a nuclear medicine measurement performed on the heart as an examination target, and is an image obtained by using SPECT or PET as a method of the nuclear medicine measurement. The three-dimensional cardiac dynamic nuclear medicine image is an image obtained by performing three-dimensional image processing on the cardiac dynamic nuclear medicine image data. The three-dimensional coronary artery image, three-dimensional morphological image, and three-dimensional cardiac dynamic nuclear medicine image may be stored in advance in the database 140 so that they can be read out from the database 140 as needed, and support for diagnosis of heart disease. It may be made possible to read from the outside of the device 100 (for example, the image pickup device 110) via a network or the like.

The index value representing the local stenosis state of each coronary artery in the subject can be, for example, an index value selected from FFR (Fractional Flow Reserve) and FFRCT. FFR is an index showing the ratio of the contribution of coronary artery stenosis to the total impaired blood flow. The FFR indicates what percentage of the maximum coronary flow supply capacity would be in an ideal vessel with no vascular resistance between the sensor and the aortic pressure. From its extensive evidence, FFR has been shown to be a useful indicator of myocardial ischemia. FFR can be said to be the simplest and most reliable index that can be evaluated in a catheterization laboratory.

The first division processing unit 103 divides at least the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image. In a preferred embodiment, this division can be performed by the Voronoi division method (see http://gihyo.jp/dev/serial/01/geometry/0011 for details). Voronoi division is a division method that defines the neighborhood of a given point as the sphere of influence of that point, and is a major concept used in various fields such as physics, ecology, and regional problems. For example, when n points are given in the space, a method is used in which the sphere of influence of each point is defined and this is represented by a Voronoi polygon.

The points to be divided are called mother points. The vertices of a Voronoi polygon are called Voronoi points, and the sides are called Voronoi sides. The Boronoi side is a locus of points equidistant from the mother points on both sides, that is, a part of the perpendicular bisector of the line segment connecting the mother points. The Voronoi point is the circumscribed circle of a triangle whose vertices are three mother points. In the examples shown in FIGS. 2 to 5, the straight lines A and B that are the targets of Voronoi division schematically show the coronary arteries (FIG. 2).

In the straight line A, four mother points A1, A2, A3, and A4 are specified at predetermined intervals (FIGS. 3 to 5). Similarly, in the straight line B, four mother points B1, B2, B3 and B4 are specified at predetermined intervals (FIGS. 3 to 5). Hereinafter, an example of the procedure of the first division processing in the image processing performed by the present embodiment will be listed with reference to FIGS. 3 to 5.

First, as shown in FIG. 3, a triangle having the mother points A1, B1 and B2 as vertices is formed. Then, the midpoint of the side A1B1 is defined as the point M1, and the midpoint of the side A1B2 is defined as the point M2.
Next, a triangle having the mother points A2, B1 and B2 as vertices is formed. Then, the midpoint of the side A2B1 is defined as the point M3, and the midpoint of the side A2B2 is defined as the point M4.
Next, a triangle having the mother points A2, B2, and B3 as vertices is formed. Then, the midpoint of the side A2B2 is defined as the point M4, and the midpoint of the side A2B3 is defined as the point M5.
Next, a triangle having the mother points A3, B2, and B3 as vertices is formed. Then, the midpoint of the side A3B2 is defined as the point M6, and the midpoint of the side A3B3 is defined as the point M7.
Next, a triangle having the mother points A3, B3, and B4 as vertices is formed. Then, the midpoint of the side A3B3 is defined as the point M7, and the midpoint of the side A3B4 is defined as the point M8.
Next, a triangle having the mother points A4, B3, and B4 as vertices is formed. Then, the midpoint of the side A4B3 is defined as the point M9, and the midpoint of the side A4B4 is defined as the point M10.

Subsequently, as shown in FIG. 4, a triangle having the mother points B1, A1 and B2 as vertices is formed. Then, the midpoint of the side B1A1 is defined as the point N1, and the midpoint of the side B2A1 is defined as the point N2.
Next, a triangle having the mother points B1, A2, and B2 as vertices is formed. Then, the midpoint of the side B1A2 is defined as the point N3, and the midpoint of the side B2A2 is defined as the point N4.
Next, a triangle having the mother points B2, A2, and B3 as vertices is formed. Then, the midpoint of the side B2A2 is defined as the point N4, and the midpoint of the side B3A2 is defined as the point N5.
Next, a triangle having the mother points B2, A3, and B3 as vertices is formed. Then, the midpoint of the side B2A3 is defined as the point N6, and the midpoint of the side B3A3 is defined as the point N7.
Next, a triangle having the mother points B3, A3, and B4 as vertices is formed. Then, the midpoint of the side B3A3 is defined as the point N7, and the midpoint of the side B4A3 is defined as the point N8.
Next, a triangle having the mother points B3, A4, and B4 as vertices is formed. Then, the midpoint of the side B3A4 is defined as the point N9, and the midpoint of the side B4A4 is defined as the point N10.
FIG. 5 is a straight line obtained based on a polygonal line (Voronoi side) connecting the midpoint groups M1 to M10 and N1 to N10, and shows an approximate straight line L virtually drawn between the straight lines A and B. .. The approximate straight line L is derived, for example, by a known least squares method. When a boundary line is drawn between meandering lines such as coronary arteries, the boundary line also has a meandering shape. In this case, as a preferred embodiment, it can be obtained by creating a line connecting each midpoint (corresponding to M1 to M10 and N1 to N10 in this example) and performing a smoothing process as necessary. A known method can be used for smoothing, and for example, a three-point smoothing method or a five-point smoothing method can be used.
Further, the position of the boundary line may be corrected in consideration of the thickness of the coronary artery. For example, when the diameters of the left and right coronary arteries are in a 6: 4 relationship, the division point obtained by the above method is not the midpoint between the mother points, but the line segment connecting the mother points is 6: 4. It can be defined as a point that divides into.

The boundary line polar coordinate expansion unit 104 extracts a boundary line that divides a plurality of regions obtained by dividing the three-dimensional morphological image of the left ventricle by the first division processing unit 103, and expands the boundary line in polar coordinates. The cardiac dynamic nuclear medicine image normalization unit 105 develops a three-dimensional cardiac dynamic nuclear medicine image in polar coordinates. Since such a method for expanding polar coordinates is already well known, for example, disclosed in Japanese Patent Application Laid-Open No. 2017-62213, detailed description thereof will be omitted, but for example, the heart is imaged by the image pickup apparatus 110. There is a method of radially expanding the maximum count of pixels obtained from the tomographic image on a concentric circle from the apex of the heart toward the base of the heart based on the tomographic image obtained.

The second division processing unit 106 generates a composite image in which a boundary line similarly expanded in polar coordinates is superimposed on a three-dimensional cardiac dynamic nuclear medicine image expanded in polar coordinates. The second division processing unit 106 divides the three-dimensional cardiac dynamic nuclear medicine image developed in polar coordinates based on the generated composite image into a plurality of regions by Voronoi division in the same manner as the first division processing unit 103. Each region divided here is defined as the region controlled by the coronary arteries running in the region.

The CFR calculation unit 107 calculates the CFR of the subject for each region in which the three-dimensional cardiac dynamic nuclear medicine image is divided by the second division processing unit 106. CFR is expressed by the following equation (1).
[Number 1]
CFR = myocardial blood flow during drug or exercise load / myocardial blood flow at rest ... (1)

In this example, the myocardial blood flow is calculated using a three-dimensional cardiac dynamic nuclear medicine image. As a method for quantifying myocardial blood flow, various known methods can be used. As a known method, described in the literature (Masao Miyagawa et al., Estimation of myocardial flow reserve utilizing an ultrafast cardiac SPECT: Comparison with coronary angiography, fractional flow reserve, and the SYNTAX score, 2017, p.347-353). In addition to the method based on the two-compartment model of, for example, the method disclosed in Japanese Patent No. 5695003 or Japanese Patent No. 5700712 can be used.

The index calculation unit 108 is based on a comparison between the CFR in the region calculated by the CFR calculation unit 107 and the index value indicating the local stenosis state of each coronary artery in the subject corresponding to the region acquired by the information acquisition unit 102. , Ask for indicators to determine treatment policy.
This index can roughly divide the state of the subject into four (indexes A to D), for example, depending on the mode of disagreement between CFR and FFR. The treatment policy for each index A to D is determined as follows, for example.
A: A state in which both the CFR value and the FFR value are maintained (for example, CFR> 2.0, FFR> 0.8) is determined to be a normal region.
B: In the state where the CFR value is maintained but the FFR value is low (for example, CFR> 2.0, FFR <0.8), vascular lesions are suspected, so PCI (Percutaneous Coronary) Intervention) treatment is judged to be effective.
C: In a state where both the CFR value and the FFR value are decreased (for example, CFR <2.0, FFR <0.8), both vascular lesions and peripheral lesions are suspected, so bypass surgery or the like is selected. Will be considered.
D: In the state where the CFR value is decreased but the FFR value is maintained (for example, CFR <2.0, FFR> 0.8), peripheral lesions are suspected, so treatment with PCI is performed. Drug therapy is selected, not targeted.

The display processing unit 109 displays the index calculated by the index calculation unit 108 on the display device 130. For example, the display processing unit 109 may display the index calculated by the index calculation unit 108 on the display device 130 in association with the composite image generated by the second division processing unit 106.

(Flow of overall processing performed by the heart disease diagnosis support device 100 and its peripheral devices)
FIG. 6 is a flowchart of the entire processing performed by the myocardial image processing device and its peripheral devices.
First, the information acquisition unit 102 (FIG. 1) acquires a three-dimensional coronary artery image, a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medicine image, and an FFR in the subject (step S101). The three-dimensional coronary artery image, the three-dimensional morphological image of the left ventricle, and the three-dimensional cardiac dynamic nuclear medicine image are images generated using the corresponding instruments. Each image or FFR may be read out from the one stored in the database 140 in advance, or may be directly taken in from each device through the network. Further, these pieces of information may be those stored in a storage medium such as a hard disk or a semiconductor memory and taken in via a peripheral device interface (not shown) or the like.

Next, the first division processing unit 103 (FIG. 1) divides the three-dimensional morphological image of the left ventricle into a plurality of regions (FIG. 7) based on the three-dimensional coronary artery image (step S102). A series of flows performed in the Voronoi division process are shown in FIGS. 9, 10 and 11. FIG. 9 is a screen configuration diagram showing a screen configuration used for an operation of extracting the coronary artery and the left ventricular myocardium. FIG. 10 is a screen configuration diagram showing a screen configuration used for an operation of selecting a site to be a starting point of region division from a coronary artery. FIG. 11 is a screen configuration diagram showing a screen configuration used for a calculation operation of a boundary line of a controlled region on a three-dimensional morphological image. The method of creating the boundary line is as described above.

Next, the boundary line polar coordinate expansion unit 104 (FIG. 1) extracts a boundary line that divides a plurality of dominant regions obtained by dividing the three-dimensional morphological image of the left ventricle by the first division processing unit 103, and publicizes the boundary line. (Step S103), the polar coordinates are expanded by using the method of.

Next, the cardiac dynamic nuclear medicine image normalization unit 105 (FIG. 1) develops a three-dimensional cardiac dynamic nuclear medicine image in polar coordinates using a known method (step S104).

Next, the second division processing unit 106 superimposes the boundary line expanded in polar coordinates by the boundary line polar coordinate expansion unit 104 on the three-dimensional cardiac dynamic nuclear medicine image expanded in polar coordinates by the cardiac dynamic nuclear medicine image normalization unit 105. A composite image (FIG. 8) is generated (step S105). Here, Sep, Ant, and Lat shown in FIG. 8 indicate the septum, the anterior wall, and the side wall, respectively. The second division processing unit 106 divides the three-dimensional cardiac dynamic nuclear medicine image expanded in polar coordinates by the cardiac dynamic nuclear medicine image normalization unit 105 into a plurality of dominant regions (FIG. 8) based on the generated composite image. do.

The upper part of FIG. 8 shows an image in which polar coordinates are expanded radially on a concentric circle centering on the apex of the heart toward the base of the heart. The lower part of FIG. 8 shows an image in which a boundary line is superimposed on a three-dimensional cardiac dynamic nuclear medicine image so as to correspond to each part of the left ventricle divided into a plurality of dominant regions.

Next, the CFR calculation unit 107 calculates the CFR of the subject for each region in which the three-dimensional cardiac dynamic nuclear medicine image is divided by the second division processing unit 106 (step S106). Specific examples of the CFR calculated here are as described above.

Next, the index calculation unit 108 obtains an index for determining the treatment policy based on the comparison between the CFR in the region calculated by the CFR calculation unit 107 and the FFR corresponding to the region (step S107). Specific examples of the indicators calculated here are as described above.

The display processing unit 109 displays the index calculated by the index calculation unit 108 on the display device 130 (step S108).

Next, using FIGS. 12 and 13, the myocardial image display method (hereinafter, simply abbreviated as this method) according to the present embodiment is compared with the conventional method of determining the dominant region of the coronary artery in 3 segments or 17 segments. The experimental results that examined the usefulness will be explained. The present invention is not limited to the present embodiment.

First, with reference to FIG. 12, a comparison is made between the method of determining the controlled region by 3 segments and this method. Line BL1 is a boundary line used in 3 segments. Line BL2 is a boundary line obtained by Voronoi partitioning according to this method. As is clear from FIG. 12, it can be said that the controlled area defined by 3 segments is controlled by a plurality of coronary arteries unlike the controlled area defined by the boundary line (BL2) defined by this law. Therefore, it can be said that a plurality of dominant regions divided by this method can obtain a result that more reflects the patient's coronary artery running as compared with the method of determining the dominant region by 3 segments.

Next, with reference to FIG. 13, a comparison is made between the method of determining the dominant region in 17 segments and this method. Line BL3 is the boundary line used in the 17 segment. The line BL4 is a boundary line obtained by Voronoi partitioning according to this method. As is clear from FIG. 13, it can be said that the controlled area defined by the 17 segment is controlled by a plurality of coronary arteries unlike the controlled area defined by the boundary line (BL4) defined by this law. Therefore, it can be said that a plurality of dominant regions divided by this method can obtain results that more reflect the patient's coronary artery running than the method of determining the dominant region by 17 segments.

Therefore, according to the present embodiment, the treatment policy can be determined based on the mode of discrepancy between CFR and FFR for each controlled area based on coronary artery running.

As described above, according to the present embodiment, it is possible to provide a technique capable of more easily supporting the correct diagnosis of the heart regardless of the individual difference of the subject.

The present invention has been described by showing embodiments so far, but these are examples. Further, the various components of the present invention do not have to be individually independent, and a plurality of components are configured as a single component, and one component is divided into a plurality of components. It is allowed that a certain component is a part of another component, a part of a certain component overlaps with a part of another component, and the like.

Further, the various components described above are not necessarily essential components, and may be omitted to the extent that the effects of the present invention are not impaired, or may be replaced with other components having the same function or function.

Further, the flow shown in the above embodiment is also an example of the embodiment of the present invention, and is not limited to the procedure shown here. Therefore, it is permissible to allow one step illustrated in the above flowchart to be separated into a plurality of executions, a plurality of steps to be executed as one step, a plurality of steps to be executed in parallel, and the like.

In the above embodiment, the first division processing unit 103 shows an example in which the midpoint of each side of a plurality of triangles (FIG. 5) is detected and the midpoints are connected to form a Voronoi side. The present invention is not limited to this aspect, and for example, the first division processing unit 103 is the shortest straight line drawn from an arbitrary point on the straight line A (FIG. 2) toward the straight line B (FIG. 2). A boronoy side may be formed by connecting the midpoints of a straight line. Further, the midpoint obtained in this way may be corrected by the method described above to reflect the difference in the thickness of the blood vessels.

In the above embodiment, an example of deriving an approximate straight line L (FIG. 5) by a known least squares method is shown. The present invention is not limited to this aspect, and the approximate straight line L may be derived by statistical analysis by a method such as first-order correlation. In short, the approximate straight line L may be derived by any method commonly used these days.

In the above embodiment, before the second division processing unit 106 divides the three-dimensional cardiac dynamic nuclear medicine image into a plurality of regions, the boundary line polar coordinate expansion unit 104 and the cardiac dynamic nuclear medicine image normalization unit 105 have polar coordinates. An example of unfolding is shown. The present invention is not limited to this aspect, and it is not always necessary to perform polar coordinate expansion before the second division processing unit 106 divides the three-dimensional cardiac dynamic nuclear medicine image into a plurality of regions.

This embodiment includes the following technical ideas.
(1) On the computer
A process for acquiring a three-dimensional coronary artery image in a subject, at least a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medicine image at rest and under load, and an index value indicating a local stenosis state of each coronary artery in the subject. ,
A process of dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image,
A process of extracting a boundary line partitioning a plurality of the regions obtained by dividing the three-dimensional morphological image of the left ventricle and dividing the three-dimensional cardiac dynamic nuclear medicine image using the boundary line.
The process of calculating the CFR of the subject for each of the regions obtained by dividing the three-dimensional cardiac dynamic nuclear medicine image, and
A process of obtaining an index for determining a treatment policy based on a comparison between the CFR in the region and the index value corresponding to the region, and
A heart disease diagnosis support program for executing a process of displaying the index on the display unit.
(2) The heart disease diagnosis support program according to (1), wherein the index value is any one of FFR (Fractional Flow Reserve) and FFRCT.
(3) Obtain a three-dimensional coronary artery image in the subject, at least a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medicine image at rest and during loading, and an index value showing a local stenosis state of each coronary artery in the subject. Information acquisition steps to be performed and
A first division step of dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image, and
In the first division step, a boundary line for partitioning a plurality of the divided regions of the three-dimensional morphological image of the left ventricle is extracted, and the boundary line is used to divide the three-dimensional cardiac dynamic nuclear medicine image. Split step and
A CFR calculation step for calculating the CFR of the subject for each of the regions obtained by dividing the three-dimensional cardiac dynamic nuclear medicine image in the second division step.
An index calculation step for obtaining an index for determining a treatment policy based on a comparison between the CFR in the region calculated in the CFR calculation step and the index value corresponding to the region.
A heart disease diagnosis support method including a display step of displaying the index calculated in the index calculation step on a display unit.
(4) The heart disease diagnosis support method according to (3), wherein the index value is any one of FFR (Fractional Flow Reserve) and FFRCT.
(5) Obtain a three-dimensional coronary artery image in the subject, at least a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medicine image at rest and during loading, and an index value showing a local stenosis state of each coronary artery in the subject. Information acquisition department and
A first division processing unit that divides the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image, and
The first division processing unit extracts a boundary line that divides a plurality of the divided regions of the three-dimensional morphological image of the left ventricle, and divides the three-dimensional cardiac dynamic nuclear medicine image using the boundary line. 2 division processing units and
A CFR calculation unit that calculates the CFR of the subject for each region in which the three-dimensional cardiac dynamic nuclear medicine image is divided by the second division processing unit.
An index calculation unit that obtains an index for determining a treatment policy based on a comparison between the CFR in the region calculated by the CFR calculation unit and the index value corresponding to the region.
A heart disease diagnosis support device comprising a display processing unit that displays the index calculated by the index calculation unit on the display unit.
(6) The heart disease diagnosis support device according to (5), wherein the index value is any one of FFR (Fractional Flow Reserve) and FFRCT.

100 Heart disease diagnosis support device 102 Information acquisition unit 103 First division processing unit 106 Second division processing unit 107 CFR calculation unit 108 Index calculation unit 109 Display processing unit 130 Display device (display unit)

Claims (6)

On the computer
A process for acquiring a three-dimensional coronary artery image in a subject, at least a three-dimensional morphological image of the left ventricle, a three-dimensional cardiac dynamic nuclear medicine image at rest and under load, and an index value indicating a local stenosis state of each coronary artery in the subject. ,
A process of dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image,
A process of extracting a boundary line partitioning a plurality of the regions obtained by dividing the three-dimensional morphological image of the left ventricle and dividing the three-dimensional cardiac dynamic nuclear medicine image using the boundary line.
The process of calculating the CFR of the subject for each of the regions obtained by dividing the three-dimensional cardiac dynamic nuclear medicine image, and
A process of obtaining an index for determining a treatment policy based on a comparison between the CFR in the region and the index value corresponding to the region, and
A heart disease diagnosis support program for executing a process of displaying the index on the display unit. In the heart disease diagnosis support program according to claim 1,
The index value is one of FFR (Fractional Flow Reserve) and FFRCT, which is a heart disease diagnosis support program. Acquisition of information to acquire three-dimensional coronary artery images in the subject, at least three-dimensional morphological images of the left ventricle, three-dimensional cardiac dynamic nuclear medicine images at rest and under load, and index values representing the local stenosis state of each coronary artery in the subject. Steps and
A first division step of dividing the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image, and
In the first division step, a boundary line for partitioning a plurality of the divided regions of the three-dimensional morphological image of the left ventricle is extracted, and the boundary line is used to divide the three-dimensional cardiac dynamic nuclear medicine image. Split step and
A CFR calculation step for calculating the CFR of the subject for each of the regions obtained by dividing the three-dimensional cardiac dynamic nuclear medicine image in the second division step.
An index calculation step for obtaining an index for determining a treatment policy based on a comparison between the CFR in the region calculated in the CFR calculation step and the index value corresponding to the region.
A heart disease diagnosis support method including a display step of displaying the index calculated in the index calculation step on a display unit. In the heart disease diagnosis support method according to claim 3,
The index value is one of FFR (Fractional Flow Reserve) and FFRCT, which is a heart disease diagnosis support method. Acquisition of information to acquire three-dimensional coronary artery images in the subject, at least three-dimensional morphological images of the left ventricle, three-dimensional cardiac dynamic nuclear medicine images at rest and under load, and index values representing the local stenosis state of each coronary artery in the subject. Department and
A first division processing unit that divides the three-dimensional morphological image of the left ventricle into a plurality of regions based on the three-dimensional coronary artery image, and
The first division processing unit extracts a boundary line that divides a plurality of the divided regions of the three-dimensional morphological image of the left ventricle, and divides the three-dimensional cardiac dynamic nuclear medicine image using the boundary line. 2 division processing units and
A CFR calculation unit that calculates the CFR of the subject for each region in which the three-dimensional cardiac dynamic nuclear medicine image is divided by the second division processing unit.
An index calculation unit that obtains an index for determining a treatment policy based on a comparison between the CFR in the region calculated by the CFR calculation unit and the index value corresponding to the region.
A heart disease diagnosis support device comprising a display processing unit that displays the index calculated by the index calculation unit on the display unit. In the heart disease diagnosis support device according to claim 5.
The index value is one of FFR (Fractional Flow Reserve) and FFRCT, which is a heart disease diagnosis support device.

Images