METHOD FOR CALCULATING INDEX OF MICROCIRCULATORY RESISTANCE BASED ON MYOCARDIAL BLOOD FLOW AND CT IMAGE

Abstract

A method for calculating an index of microcirculatory resistance includes determining myocardial volume by extracting myocardial images; locating a coronary artery inlet and accurately segmenting coronary arteries; generating a grid model required for calculation; determining myocardial blood flow in a rest state and CFR; calculating total flow at the coronary artery inlet in a maximum hyperemia state; determining flow in different blood vessels in a coronary artery tree in the maximum hyperemia state and then determining a flow velocity V.sub.1 in the maximum hyperemia state; obtaining the average conduction time in the maximum hyperemia state Tmn, and calculating a pressure drop ΔP from the coronary artery inlet to a distal end of a coronary artery stenosis, and a mean intracoronary pressure P.sub.d at the distal end of the stenosis P.sub.d=P.sub.a−ΔP, and calculating the index of microcirculatory resistance.

Claims

1. A method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image, comprising the following steps: S01: segmenting the CT image of heart, obtaining an image of heart via a morphological operation, subjecting the image of heart to a histogram analysis to obtain an image contains ventricular and atrial, obtaining a myocardial image by making a difference between the image of heart and the image contains ventricular and atrial, determining a myocardial volume by the myocardial image; S02: obtaining a full aortic complementary image by processing an aortic image, obtaining the aortic image containing a coronary artery inlet through regional growth of the full aortic complementary image, and obtaining an image containing the coronary artery inlet according to the aortic image containing the coronary artery inlet and the full aortic complementary image, to determine the coronary artery inlet by the image containing the coronary artery inlet; S03: extracting a coronary artery through regional growth by taking the coronary artery inlet as a seed point on the myocardial image, calculating an average gray and average variance of the coronary artery, and along a direction of the coronary artery, extracting a coronary artery tree according to a gray distribution of the coronary artery; S04: binarizing the coronary artery image, drawing an isosurface image to obtain a three-dimensional grid image of the coronary artery; S05: calculating a total flow at the coronary artery inlet in a maximum hyperemia state, Q.sub.total=myocardial volume×myocardial blood flow×CFR, and CFR being the coronary flow reserve; S06: calculating a blood flow velocity V.sub.1 in a hyperemia state and an average conduction time Tmn in the maximum hyperemia state; S07: calculating a pressure drop ΔP from the coronary artery inlet to a distal end of a coronary artery stenosis using V.sub.1 as an inlet flow velocity of coronary artery stenosis blood vessel, and using P.sub.d=P.sub.a−ΔP calculating a mean intracoronary pressure P.sub.d at the distal end of the coronary artery stenosis, wherein P.sub.a is a mean aortic pressure, and obtaining the index of microcirculatory resistance using IMR=P.sub.d*Tmn.

2. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein after obtaining the image containing the coronary artery inlet in step S02, the image containing the coronary artery inlet is subjected to a connected domain analysis, and each connected domain is identified by different gray labels to determine the coronary artery inlet.

3. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein in step S02, based on a feature that an aortic cross-section is circular, an ascending aortic and a center line are extracted from the image of heart to obtain the aortic image.

4. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein binarizing the coronary artery image in step S04 comprises: going through voxels in the coronary artery image, and remaining a pixel value unchanged if a voxel pixel is equal to 0; and setting the pixel value to 1 to obtain new data if the voxel pixel is not equal to 0.

5. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein in step S05, myocardial blood flow in a rest state and coronary flow reserve (CFR) are determined by cardiac ultrasound (MCE) or single photon emission computed tomography (SPECT) or positron emission tomography (PET) or cardiac nuclear magnetic (MRI) or CT perfusion.

6. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein step S06 comprises: S61: determining a blood flow Q in any one of blood vessels in the tree using Q=Q.sub.total×(V/V.sub.total).sup.3/4 based on a flow volume scale and a cardiac surface coronary artery tree obtained using a three-dimensional reconstruction of a cardiac CT, wherein V.sub.total is the a volume of blood in all cardiac surface coronary artery obtained using the three-dimensional reconstruction of the cardiac CT, V is a sum volume of blood in any one of blood vessels and its downstream blood vessel in the cardiac surface coronary artery tree; S62: determining the blood flow velocity in any one of blood vessel in the tree using V.sub.1=Q/D based on the flow volume scale and the cardiac surface coronary artery tree obtained using the three-dimensional reconstruction of the cardiac CT, wherein D is an average diameter of the blood vessel; S63: based on a length L of the coronary blood vessel and the flow velocity V.sub.1, obtaining the average conduction time in the maximum hyperemia state Tmn=L/V.sub.1.

7. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein step S07 specifically comprises: solving three-dimensional grids of blood vessels, and solving a continuity and Navier-Stokes equations using numerical methods: $\begin{array}{cc}\nabla .Math.\stackrel{.fwdarw.}{V}=0& \left[\mathrm{A1}\right]\\ \rho \frac{\partial \stackrel{.fwdarw.}{V}}{\partial t}+\rho \stackrel{.fwdarw.}{V}.Math.\nabla \stackrel{.fwdarw.}{V}=-\nabla P+\nabla .Math.\mu \left(\nabla \stackrel{.fwdarw.}{V}+{\left(\nabla \stackrel{.fwdarw.}{V}\right)}^T\right)& \left[\mathrm{A2}\right]\end{array}$ wherein, V, P, ρ, μ are flow velocity, pressure, blood flow density, blood flow viscosity, respectively; an inlet boundary condition being: the inlet flow velocity V.sub.1 of the coronary artery stenosis vessel in the maximum hyperemia state; calculating a pressure drop of each coronary artery stenosis through three-dimensional computational fluid dynamics to obtain ΔP1, ΔP2, ΔP3 . . . , calculating the pressure drop from the coronary artery inlet to the distal end of the coronary artery stenosis using ΔP=ΣΔP.sub.i (i=1,2,3 . . . ), and obtaining the mean intracoronary pressure P.sub.d at the distal end of the stenosis using P.sub.d=P.sub.a−ΔP, wherein P.sub.a is the mean aortic pressure.

8. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 1, wherein step S07 comprises: based on the geometric structure reconstructed by CT, straightening blood vessel with stenosis, constructing a two-dimensional axisymmetric model, dividing two-dimensional grids, and solving a continuity and Navier-Stokes equation by numerical methods: $\begin{array}{cc}\frac{1}{r}\frac{\partial }{\partial r}\left({\mathrm{ru}}_r\right)+\frac{\partial u_z}{\partial z}=0& \left[\mathrm{A3}\right]\\ \rho \left(\frac{\partial u_r}{\partial t}+u_r\frac{\partial u_r}{\partial r}+u_z\frac{\partial u_r}{\partial z}\right)=-\frac{\partial p}{\partial r}+\mu \left[\frac{1}{r}\frac{\partial }{\partial r}\left(r\frac{\partial u_r}{\partial r}\right)+\frac{{\partial }^2{u}_r}{\partial z^2}-\frac{u_r}{r^2}\right]& \left[\mathrm{A4}\right]\\ \rho \left(\frac{\partial u_z}{\partial t}+u_r\frac{\partial u_z}{\partial r}+u_z\frac{\partial u_z}{\partial z}\right)=-\frac{\partial p}{\partial z}+\mu \left[\frac{1}{r}\frac{\partial }{\partial r}\left(r\frac{\partial u_z}{\partial r}\right)+\frac{{\partial }^2{u}_z}{\partial z^2}\right]& \left[\mathrm{A5}\right]\end{array}$ wherein, ρ represents the density of blood, u.sub.z and u.sub.r represent the flow velocity in the z direction and r direction, respectively, μ represents the dynamic viscosity of the blood, and p represents the pressure of the blood; an inlet boundary condition being: the inlet flow velocity V.sub.1 of the coronary artery stenosis vessel in the maximum hyperemia state; calculating a pressure drop of each coronary artery stenosis through two-dimensional computational fluid dynamics to obtain ΔP.sub.1, ΔP.sub.2, ΔP.sub.3 . . . , calculating the pressure drop from the coronary artery inlet to the distal end of the coronary artery stenosis using ΔP=ΣΔP.sub.i (i=1,2,3 . . . ), and obtaining the mean intracoronary pressure P.sub.d at the distal end of the stenosis using P.sub.d=P.sub.a−ΔP, wherein P.sub.a is the mean aortic pressure.

9. The method for calculating an index of microcirculatory resistance based on myocardial blood flow and CT image according to claim 8, wherein the step S07 further comprises: for different types of bends of the blood vessels, calculating a pressure difference from the inlet to an outlet using a three-dimensional model; in comparison with a calculation of two-dimensional axisymmetric model, establishing a database for storing correction coefficients of various types of bending relative to two-dimensional axisymmetric results; after obtaining the pressure, obtaining a corrected pressure difference from the inlet to the outlet with reference to the correction coefficients in the database, and then calculating IMR.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0038] The present disclosure will be further described below with reference to the drawings and embodiments:

[0039] FIG. 1 is a flow chart of a method of the present disclosure.

[0040] FIG. 2 shows results of myocardial segmentation for a CT image of heart.

[0041] FIG. 3 shows results of aorta segmentation with a coronary artery inlet.

[0042] FIG. 4 shows results of coronary artery inlet segmentation.

[0043] FIG. 5 shows results of coronary artery segmentation.

[0044] FIG. 6 is a grid model for the results of the coronary artery segmentation.

[0045] FIG. 7 is a schematic diagram of blood flow of the heart and coronary artery.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0046] In order to make purposes, technical solutions and advantages of the disclosure dearer, the present disclosure will be described in detail below in conjunction with the specific embodiments and the corresponding accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present disclosure. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0047] Given a CT image of heart, the image of the heart is extracted and obtained by extracting the heart according to a reverse method, and processing descending aorta, spine, and ribs in non-target areas as objects, and gradually removing non-heart tissues, such as chest walls, lungs, vertebrae, and descending aorta. Based on a feature of cross section of the aorta to be circular, ascending aorta and center line are extracted from the obtained image of heart to obtain the aortic image.

[0048] As shown in FIG. 1, a method for calculating an index of microcirculatory resistance IMR based on myocardial blood flow and CT image of the present disclosure comprises extracting a myocardial image, extracting a coronary artery inlet, extracting coronary artery, generating a grid model for the coronary artery, and determining myocardial blood flow in a rest state and coronary flow reserve (CFR), calculating total flow at the coronary artery inlet in a maximum hyperemia state, calculating a blood flow velocity V.sub.1 in the hyperemia state, and determining the index of microcirculatory resistance. It specifically comprises the following steps:

[0049] 1: Extracting the Myocardial Image:

[0050] The CT image of heart is segmented, the image of heart is obtained through a morphological operation, a histogram analysis is subjected to the image of heart to obtain a ventricular and atrial image, and a myocardial image is obtained by making the difference between the image of heart and the ventricular and atrial image, as shown in FIG. 2.

[0051] 2: Extracting the Coronary Artery Inlet:

[0052] A morphological expansion is subjected to the binary image of the aortic image to obtain the binary image of the whole aorta, and a complementary image of the whole aorta is obtained by pixel inversion.

[0053] Regional growth is carried out according to average gray level of points on the center line of the aorta to obtain the aortic image with the coronary artery inlet, as shown in FIG. 3.

[0054] The graphic multiplication method is carried out using the aortic image with the coronary artery inlet and the complementary image of the whole aorta to obtain an image with the coronary artery inlet. Connected domain analysis is subjected to the image with the coronary artery inlet, and each connected domain is identified by using different gray labels to determine the coronary artery inlet, as shown in FIG. 4.

[0055] 3: Extracting the Coronary Artery:

[0056] The coronary artery inlet is taken as a seed point on the myocardial image, the coronary artery is extracted through regional growth, an average gray and average variance of the coronary artery are calculated, and a coronary artery tree is extracted along a direction of the coronary artery according to a gray distribution of the coronary artery, as shown in the FIG. 5.

[0057] 4: Generating the Grid Model for the Coronary Artery:

[0058] Image data V.sub.1 of the coronary artery is obtained through step 3. The voxels in the data spatially form a cube. A pixel value of the voxel belonging to the coronary artery is not 0 (the pixel value is approximately between −3000 and 3000), and pixel values of the remaining voxels are all 0.

[0059] In this step, the data needs to be transformed into spatial three-dimensional grid data V3 to facilitate calculation in step 5.

[0060] (1) Binarization of Coronary Artery Data

[0061] The voxels in the image data V1 of the coronary artery are went through and a simple judgment for pixel values is made. If pixel A1 is equal to 0, the pixel value remains unchanged; if A1 is not equal to 0, the pixel value of A1 is set to 1.

[0062] Finally, a new image data V2 will be obtained. In this image, the pixel value of voxels belonging to part of the coronary artery is 1, and the rest is 0.

[0063] (2) Generation of Isosurface

[0064] A voxel is defined as a tiny hexahedron with eight vertices on a cube composed of four pixels between adjacent upper and lower layers. The isosurface is a collection of points with the same attribute value in space. It can be expressed as:

{(x,y,z)|f(x,y,z)=c}, c is a constant.

[0065] In this method, c is the pixel value 1 given in the three-dimensional reconstruction process.

[0066] The process of extracting the isosurface is as follows:

[0067] (1) reading the preprocessed original data into a specific array;

[0068] (2) extracting a unit body from the grid data body to become the current unit body, and at the same time obtaining all information of the unit body;

[0069] (3) comparing the function values of the 8 vertices of the current unit body with the given isosurface value C to obtain a state table of the unit body;

[0070] (4) according to a state table index of the current unit body, finding out the edge of the unit body intersecting with the isosurface, and using linear interpolation method to calculate the position coordinates of each intersection;

[0071] (5) using the center difference method to determine the normal vectors of the 8 vertices of the current unit body, and then using the linear interpolation method to obtain the normal directions of each vertex of the triangle patch;

[0072] (6) drawing an image of the isosurface according to the coordinates of the vertices of each triangle patch and the normal vector of the vertices.

[0073] Finally, three-dimensional grid image data V3 of the coronary artery is obtained, as shown in FIG. 6.

[0074] 5: Calculating the Blood Flow Velocity V.sub.1 in the Hyperemia State:

[0075] The myocardial blood flow in a rest state and coronary flow reserve (CFR) are determined by non-invasive measurements such as cardiac ultrasound (MCE) or single photon emission computed tomography (SPECT) or positron emission tomography (PET) or cardiac nuclear magnetic (MRI) or CT perfusion; the total flow at the coronary artery inlet (including the sum of the left coronary artery tree and the right coronary artery tree) in the maximum hyperemia state is calculated: Q.sub.total=myocardial volume×myocardial blood flow×CFR;

[0076] Based on a flow volume scale and a cardiac surface coronary artery tree obtained using a three-dimensional reconstruction of a cardiac CT, the blood flow Q in any one of blood vessels in the tree is determined: Q=Q.sub.total×(V/V.sub.total).sup.3/4, wherein V.sub.total is a sum volume of blood in all cardiac surface coronary artery obtained using the three-dimensional reconstruction of the cardiac CT, V is a sum volume of blood in any one of blood vessels and its downstream vessel in the cardiac surface coronary artery tree, as shown in FIG. 7; based on the flow volume scale and the cardiac surface coronary artery tree obtained using the three-dimensional reconstruction of cardiac CT, a blood flow velocity in any one of blood vessels in the tree is determined: V.sub.1=Q/D, wherein D is average diameter of the blood vessel (the blood volume of the blood vessel divided by the length of the blood vessel); based on a length L of the coronary blood vessel and the flow velocity V.sub.1, obtaining the average conduction time in the maximum hyperemia state Tmn=L/V.sub.1.

[0077] 6: Calculating the Index of Microcirculatory Resistance:

[0078] A pressure drop of each coronary artery stenosis is calculated through computational fluid dynamics (CFD) using V.sub.1 as the inlet flow velocity of the coronary artery stenosis vessel to obtain ΔP.sub.1, ΔP.sub.2, ΔP.sub.3 . . . , the pressure drop from the coronary artery inlet to a distal end of the coronary artery stenosis is calculated using ΔP=ΣΔP.sub.i(i=1,2,3 . . . ), and the mean intracoronary pressure P.sub.dat the distal end of the stenosis is obtained using P.sub.d=P.sub.a−ΔP, wherein P.sub.a is a mean aortic pressure, and finally the index of microcirculatory resistance is calculated by the formula IMR=P.sub.d*Tmn.

[0079] The processing steps for the 3D model comprise:

[0080] Based on the geometric structure reconstructed by CT, dividing three-dimensional grids, and using numerical methods (such as: finite difference, finite element, finite volume method, etc.) to solve the continuity and Navier-Stokes equations:

[00003]$\begin{array}{cc}\nabla .Math.\stackrel{.fwdarw.}{V}=0& \left[\mathrm{A1}\right]\\ \rho \frac{\partial \stackrel{.fwdarw.}{V}}{\partial t}+\rho \stackrel{.fwdarw.}{V}.Math.\nabla \stackrel{.fwdarw.}{V}=-\nabla P+\nabla .Math.\mu \left(\nabla \stackrel{.fwdarw.}{V}+{\left(\nabla \stackrel{.fwdarw.}{V}\right)}^T\right)& \left[\mathrm{A2}\right]\end{array}$

[0081] Wherein, V, P, ρ, μ are flow velocity, pressure, blood flow density, blood flow viscosity, respectively.

[0082] An inlet boundary condition is: the inlet flow velocity V.sub.1 of the coronary artery stenosis vessel in the maximum hyperemia state.

[0083] Based on formula [A1] and [A2], a pressure drop of each coronary artery stenosis is calculated by carrying out three-dimensional CFD to obtain ΔP.sub.1, ΔP.sub.2, ΔP.sub.3 . . . , the pressure drop from the coronary artery inlet to a distal end of the coronary artery stenosis is calculated using ΔP=ΣΔP.sub.i (i=1,2,3 . . . ), and the mean intracoronary pressure P.sub.d at the distal end of the stenosis is obtained using P.sub.d=P.sub.a−ΔP, wherein P.sub.a is the mean aortic pressure.

[0084] For the two-dimensional model, the following steps are included:

[0085] Based on the geometric structure reconstructed by CT, straightening the blood vessel with stenosis (two-dimensional axisymmetric model), dividing two-dimensional grids, and using numerical methods (such as finite difference, finite element, finite volume method, etc.) to solve the continuity and Navier-Stokes equation:

[00004]$\begin{array}{cc}\frac{1}{r}\frac{\partial }{\partial r}\left({\mathrm{ru}}_r\right)+\frac{\partial u_z}{\partial z}=0& \left[\mathrm{A3}\right]\\ \rho \left(\frac{\partial u_r}{\partial t}+u_r\frac{\partial u_r}{\partial r}+u_z\frac{\partial u_r}{\partial z}\right)=-\frac{\partial p}{\partial r}+\mu \left[\frac{1}{r}\frac{\partial }{\partial r}\left(r\frac{\partial u_r}{\partial r}\right)+\frac{{\partial }^2{u}_r}{\partial z^2}-\frac{u_r}{r^2}\right]& \left[\mathrm{A4}\right]\\ \rho \left(\frac{\partial u_z}{\partial t}+u_r\frac{\partial u_z}{\partial r}+u_z\frac{\partial u_z}{\partial z}\right)=-\frac{\partial p}{\partial z}+\mu \left[\frac{1}{r}\frac{\partial }{\partial r}\left(r\frac{\partial u_z}{\partial r}\right)+\frac{{\partial }^2{u}_z}{\partial z^2}\right]& \left[\mathrm{A5}\right]\end{array}$

[0086] Wherein, ρ represents the density of blood, u.sub.z and u.sub.r represent the flow velocity in the z direction and r direction, respectively, μ represents the dynamic viscosity of the blood, and p represents the pressure of the blood.

[0087] An inlet boundary condition is: the inlet flow velocity V.sub.1 of the coronary artery stenosis vessel in the maximum hyperemia state.

[0088] Based on the formula [A3]-[A5], a pressure drop of each coronary artery stenosis is calculated by carrying out two-dimensional CFD to obtain ΔP.sub.1, ΔP.sub.2, ΔP.sub.3 . . . , the pressure drop from the coronary artery inlet to a distal end of the coronary artery stenosis is calculated using ΔP=ΣΔP.sub.i (i=1,2,3 . . . ), and the mean intracoronary pressure Pd at the distal end of the stenosis is obtained using P.sub.d=P.sub.a−ΔP, wherein P.sub.a is the mean aortic pressure.

[0089] As for different types of bending of the blood vessels, pressure difference from the inlet to the outlet is calculated by using the three-dimensional model, and the calculation is made by comparing with the two-dimensional axisymmetric model to establish a database for storing correction coefficients of various types of bending relative to the two-dimensional axisymmetric results; the pressure is calculated and then the corrected pressure difference from the inlet to the outlet is obtained by comparing with the correction coefficients in the database, and the index of microcirculatory resistance is finally calculated by the formula IMR=P.sub.d*Tmn.

[0090] It should be understood that the specific embodiments mentioned above are merely intended to exemplify or explain of the principles of the present disclosure and not to be limitations to the present disclosure. Therefore, any modifications, equivalent substitutions, improvements and the like made without departing from the spirit and scope of the present disclosure should be included in the protection scope of the present disclosure. In addition, the appended claims of the present disclosure are intended to cover all changes and modifications falling within the scope and boundary of the appended claims, or equivalents of such scope and boundary.