Method and system for calculating FFR based on blood flow ratio before and after stenosis
By establishing a coronary tree model and an unstenosis model, combining computational fluid mechanics and microcirculation blood vessel model, calculating blood flow ratio, the problems of FFR calculation error and clinical application limitation in the existing technology are solved, and higher accuracy and Security.
Patent Information
- Application Number
- CN202110216206.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-02-26
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-02-26
AI Technical Summary
In the prior art, invasive measurement is risky and expensive when calculating blood flow ratio (FFR). The non-invasive measurement method has a large calculation error due to model differences and pressure ratio error, which limits clinical applications.
Establish a coronary tree model and an unstenosis model, combine calculation fluid mechanics to solve the average blood flow of the two models, calculate the blood flow ratio to reduce the FFR calculation error, increase the collateral blood vessels to better fit the actual physiological conditions, and generate a microcirculation blood vessel model to Accurately calculate the microcycle resistance value.
It improves the reliability and accuracy of FFR calculations, reduces clinical costs, avoids invasive detection risks, and accurately estimates the effect of coronary artery treatment, which helps in clinical treatment plan planning.
Smart Images

Figure CN114947909B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of human physiology and hemodynamics, and in particular to a method and system for calculating FFR based on the blood flow ratio before and after a stenotic lesion. Background Art
[0002] The main cause of coronary heart disease is coronary artery stenosis caused by arteriosclerosis. The fractional flow reserve (FFR) refers to the ratio of the maximum blood flow that can be obtained by the myocardial area supplied by the blood vessel in the presence of coronary artery stenosis to the maximum blood flow that can be obtained by the same area under normal conditions in theory. FFR can indicate the impact of coronary artery stenosis on distal blood flow and is used to assess whether the myocardium is ischemic. FFR has become a recognized indicator for the functional evaluation of coronary artery stenosis.
[0003] When determining FFR, since it is difficult to directly obtain the value of blood flow, the calculation of FFR is often simplified to the ratio of the mean pressure in the coronary artery distal to the stenosis (Pd) to the mean pressure in the aorta at the coronary artery orifice (Pa) under the state of maximum myocardial hyperemia. Currently, FFR is mostly obtained invasively, using a pressure guidewire to obtain blood pressure upstream and downstream of the stenosis to invasively measure the FFR value, which is high-risk and expensive.
[0004] Another way to obtain FFR is a non-invasive measurement method that combines medical images and computational fluid dynamics (CFD). The main idea is to construct a coronary hemodynamic model based on medical images and computational fluid dynamics (CFD), simulate and introduce physiological parameters such as cardiac output, aortic pressure, and microcirculatory resistance, and then calculate FFR based on the flow parameters of the model.
[0005] Chinese patent CN201611234903.1 discloses a system for simulating and calculating blood flow reserve fraction using computational fluid dynamics (CFD) and a method for setting corresponding inlet boundary conditions and outlet boundary conditions for the computational fluid dynamics (CFD) model of the vascular tree. However, the established vascular tree model inevitably differs from the actual physiological condition of the human body, and the use of the pressure ratio instead of the blood flow ratio to calculate FFR brings certain errors, resulting in a certain difference between the calculated FFR and the actual FFR, which greatly limits the clinical use of the FFR non-invasive measurement method. Summary of the invention
[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method and system for calculating FFR based on the blood flow ratio before and after stenosis lesions, establish a coronary tree model and a non-stenosis model, combine computational fluid dynamics to solve the average blood flow of the two models, calculate the ratio of the blood flow of corresponding sections on the two models to obtain the blood flow reserve fraction, return to the original definition of the blood flow reserve fraction, reduce the calculation error of FFR, and obtain a more reliable FFR.
[0007] The purpose of the present invention can be achieved by the following technical solutions:
[0008] A method for calculating FFR based on the blood flow ratio before and after a stenotic lesion, comprising the following steps:
[0009] S1: acquiring routine blood data and medical images of the heart, performing segmentation and reconstruction processing on the medical images, establishing a coronary tree model, wherein the coronary tree model includes the aorta root, adding multiple side branch vessels to the coronary tree model, that is, adding multiple new coronary outlets, and determining the volume of the myocardium and the radius of each coronary outlet based on the coronary tree model;
[0010] S2: determine the stenosis segment on the coronary artery tree model, determine the theoretical normal radius of the stenosis segment according to the vessel radius upstream of the stenosis segment and the vessel radius downstream of the stenosis segment, and use the theoretical normal radius to correct the coronary artery tree model to obtain a non-stenosis model, which is used to represent the coronary artery tree model without stenosis lesions;
[0011] S3: Based on the volume of the myocardium and the radius of each coronary artery outlet, the volume of the myocardium supplying each coronary artery outlet is determined respectively, a microcirculatory vascular model is generated, and the microcirculatory resistance value of each coronary artery outlet is calculated based on the blood routine data and the microcirculatory vascular model;
[0012] S4: Generate computational fluid dynamics calculation grid, establish a three-dimensional CFD model of the coronary tree model, a three-dimensional CFD model of the non-stenotic model, and a one-dimensional CFD model of the microcirculation vascular model, and obtain the physical properties and flow equations of the blood;
[0013] S5: setting boundary conditions of the three-dimensional CFD model of the coronary tree model, and setting boundary conditions of the three-dimensional CFD model of the non-stenosis model, wherein the boundary conditions include an inlet boundary condition and an outlet boundary condition;
[0014] S6: Based on the boundary conditions of the three-dimensional CFD model of the coronary tree model, the physical properties of the blood and the flow equations, the fluid parameters at each position of the coronary tree model are solved; based on the boundary conditions of the three-dimensional CFD model of the non-stenotic model, the physical properties of the blood and the flow equations, the fluid parameters at each position of the non-stenotic model are solved; the fluid parameters include pressure, blood flow and velocity.
[0015] S7: Cut at least one section on the coronary tree model and the non-stenotic model respectively and obtain the average blood flow of each section. The sections of the coronary tree model correspond to the sections of the non-stenotic model one by one and the cutting positions are the same. The average blood flow of a section is the average blood flow passing through the section in one cardiac cycle. The blood flow reserve fraction of each section is calculated respectively. The blood flow reserve fraction of a section is equal to the ratio of the average blood flow of the section on the coronary tree model to the average blood flow of the corresponding section of the section on the non-stenotic model.
[0016] Preferably, the selected position of the cross section includes between the upper and lower bifurcations of the coronary artery.
[0017] Preferably, the medical image of the heart in step S1 is a cardiac CT image, the cardiac CT image is segmented, a myocardial image is obtained through morphological operations and grayscale extraction, and the volume of the myocardium is determined; the cardiac CT image is segmented, a coronary tree including the aortic root is obtained through morphological operations and grayscale extraction, the area of each coronary outlet is determined, and the equivalent radius of each coronary outlet is used as the radius of each coronary outlet; the center line and vessel wall of each blood vessel are calculated and reconstructed, and a coronary tree model is established.
[0018] Preferably, in step S1, p (1≤p≤7) side branch vessels are added to the coronary artery tree model, and the side branch vessels are added at a side of the coronary artery close to the heart and between the upper and lower bifurcations of the existing coronary artery. The length of the side branch vessel is 3-5 mm, and the radius of the side branch vessel is the minimum radius value of the existing blood vessels on the coronary artery tree model.
[0019] Preferably, the value of p is determined by the change in the blood vessel area between the upper and lower bifurcations. Assuming that the following relationship exists between the blood vessel radius before and after the coronary artery bifurcation: the nth power of the blood vessel radius before the bifurcation = the sum of the nth power of the coronary artery radius after the bifurcation, the value of p satisfies the following formula:
[0020] p*r 侧枝血管 ) n =(r 上级分叉 ) n -(r 下级分叉 ) n
[0021] Among them, r 侧枝血管 represents the radius of the collateral vessel, r 上级分叉 represents the radius of the vessel downstream of the superior bifurcation, r 下级分叉 It represents the radius of the upstream blood vessel of the lower bifurcation, and the value range of n is 2.3-2.44.
[0022] Preferably, in step S2, the theoretical normal radius of the stenosis segment is determined by an interpolation method.
[0023] Preferably, the volume of the myocardium supplied by the coronary artery outlet in step S3 is determined in detail by: determining the volume of the myocardium supplied by each coronary artery outlet in turn, wherein the calculation formula for the volume of the myocardium supplied by the i-th (1≤i≤N) coronary artery outlet is as follows:
[0024]
[0025] Among them, V i represents the volume of the myocardium supplied by the i-th coronary artery outlet, V represents the volume of the myocardium, r i represents the radius of the i-th coronary artery outlet, k represents the radius of the kth coronary artery outlet, N represents the number of coronary artery outlets on the coronary tree model, and the value range of n is 2.3-2.44.
[0026] Preferably, the specific process of generating the microcirculation blood vessel model and calculating the microcirculation resistance in step S3 is as follows:
[0027] A1: Initialize and generate a microcirculation blood vessel model, wherein the microcirculation blood vessel model includes N coronary artery outlets, and each coronary artery outlet is used as a parent generation, and step A2 is executed;
[0028] A2: Each parent generation performs a bifurcation operation, and one parent generation obtains multiple children, each of which is a blood vessel. The blood supply myocardium of the parent generation is allocated to each child generation, and all children generations are added to the microcirculation vascular model according to the corresponding relationship between the children and the parent generation;
[0029] A3: Traverse all the children obtained in the previous step. If there are children with a radius greater than r0, take all the children with a radius greater than r0 as the parent and execute step A2. Otherwise, the generation of the small blood vessel tree model is completed; wherein r0 is the preset minimum blood vessel radius of the microcirculation blood vessel model;
[0030] A4: Based on the microcirculatory vascular model and routine blood data, the microcirculatory resistance value of each coronary artery outlet is calculated using Poiseuille's theorem and Fallin effect.
[0031] Preferably, in step A2, a parent generation performs a fork operation as follows:
[0032] The blood supply myocardium of the parent generation is equivalent to an ellipsoid, and a point on the surface of the ellipsoid is taken as the root node. The root node represents the location of the parent generation's exit. The ellipsoid is evenly divided into m (2≤m≤5) sub-parts through the root node. The center of mass of each sub-part is found respectively, and the center of mass is connected with the root node. The m line segments obtained are m offspring. The volume of the blood supply myocardium of each offspring is the volume of one sub-part. The radius of the m offspring is the same, and the radius of the offspring is r 子代 The radius r of the parent 父代Satisfy the relationship: m*(r 子代 ) n =(r 父代 ) n .
[0033] Preferably, when performing the bifurcation operation, the blood-supplying myocardium of the parent generation is equivalent to a sphere.
[0034] Preferably, when performing the bifurcation operation, the ellipsoid is evenly divided into 2 sub-parts through the root node.
[0035] Preferably, the size of the minimum blood vessel radius r0 of the pre-set microcirculation blood vessel model is 0.005 mm.
[0036] Preferably, in step A4, the resistance values of each section of blood vessel are calculated using Poiseuille's theorem, where the viscosity coefficient is obtained using the Fahraeus effect, the hematocrit in the Fahraeus effect is determined from blood routine data, and the microcirculation resistance values at each coronary artery outlet are calculated using the series and parallel theorem of resistance.
[0037] Preferably, in step S5, the boundary conditions of the three-dimensional CFD model of the coronary artery tree model are set in the same way as those of the three-dimensional CFD model of the non-stenosed model, and both include the following steps:
[0038] S51: Set each inlet boundary condition of the three-dimensional CFD model to the blood pressure P0 of the corresponding coronary artery orifice respectively, and set each outlet boundary condition of the three-dimensional CFD model to a times (0 < a < 1) of the blood pressure of the corresponding coronary artery orifice, that is, the pressure P at each outlet of the three-dimensional CFD model is P = a*P0;
[0039] S52: Obtain the current boundary conditions of the three-dimensional CFD model. For each outlet of the three-dimensional CFD model, solve the three-dimensional CFD model based on the physical properties of the blood and the flow equation to obtain the blood flow Q at each outlet of the three-dimensional CFD model;
[0040] S53: The inlets of the one-dimensional CFD model correspond one-to-one with the outlets of the three-dimensional CFD model. Set each inlet boundary condition of the one-dimensional CFD model to the blood flow Q at the corresponding outlet of the three-dimensional CFD model respectively, and set each outlet boundary condition of the one-dimensional CFD model to the venous pressure Pout;
[0041] S54: Calculate the blood pressure Pin at each inlet of the one-dimensional CFD model according to the formula Pin-Pout=R*Q. If Pin and P are not equal and the difference is greater than the preset convergence accuracy, correct each outlet boundary condition of the three-dimensional CFD model respectively. The correction formula is P`=(P*4+Pin) / 5. Repeat step S52. Otherwise, use the current boundary conditions of the three-dimensional CFD model as the final boundary conditions; wherein R represents the microcirculation resistance value, and P` represents the outlet boundary conditions of the corrected three-dimensional CFD model.
[0042] A system for calculating FFR based on the blood flow ratio before and after a stenotic lesion, based on the method for calculating FFR as described above, comprises:
[0043] A coronary tree model generation module is used to obtain a medical image of the heart, segment and reconstruct the medical image, obtain a coronary tree model, and add multiple side branch vessels to the coronary tree model;
[0044] A non-stenosis model generation module, used for establishing a non-stenosis model according to a coronary tree model;
[0045] Microcirculatory vascular module, used to establish a microcirculatory vascular model and calculate the microcirculatory resistance value corresponding to each coronary artery outlet on the coronary tree model;
[0046] A computational grid generation module, used to generate computational fluid dynamics computational grids for the coronary tree model, the non-stenotic model and the microcirculatory vascular model, so as to establish a three-dimensional CFD model of the coronary tree model, a three-dimensional CFD model of the non-stenotic model and a one-dimensional CFD model of the microcirculatory vascular model;
[0047] A boundary condition setting module, used to set boundary conditions of the three-dimensional CFD model of the coronary tree model and boundary conditions of the three-dimensional CFD model of the non-stenosis model;
[0048] Property setting module, used to set the physical properties and flow equations of blood;
[0049] A solver, based on the boundary conditions of the three-dimensional CFD model of the coronary tree model, the physical properties of the blood and the flow equation, solves the fluid parameters of each location of the coronary tree model, and based on the boundary conditions of the three-dimensional CFD model of the non-stenotic model, the physical properties of the blood and the flow equation, solves the fluid parameters of each location of the non-stenotic model;
[0050] The post-processing module calculates the blood flow reserve fraction of each section based on the average blood flow of each section on the coronary tree model and the average blood flow of each section on the non-stenotic model.
[0051] Preferably, in the property setting module, based on routine blood data and the Fallin effect formula, it is assumed that the blood is a Newtonian fluid or a generalized Newtonian fluid and adopts laminar flow, individual-specific blood density and blood viscosity; the flow equation includes a steady flow equation or an unsteady flow equation, and a Navier-Stokes equation group based on incompressible flow.
[0052] Compared with the prior art, the present invention has the following beneficial effects:
[0053] (1) A coronary artery tree model and a non-stenotic model were established. The average blood flow of the two models was obtained by combining computational fluid dynamics. The blood flow ratio of the corresponding cross-sections on the two models was calculated to obtain the blood flow reserve fraction. The original definition of the blood flow reserve fraction was regressed, the calculation error of FFR was reduced, and the obtained FFR was more reliable.
[0054] (2) By adding collateral vessels to the coronary artery tree model, the resulting coronary artery tree model is more consistent with the actual physiological conditions. Then, by continuously performing bifurcation operations, a microcirculatory vascular model is established to accurately obtain the microcirculatory resistance value at the coronary artery outlet. The boundary outlet conditions are set according to the microcirculatory resistance value, thereby reducing the difference between the coronary artery tree model and the non-stenosis model and the actual physiological conditions of the human body, thereby improving the accuracy of the blood flow reserve fraction calculation.
[0055] (3) When adding collateral vessels, it is assumed that the nth power of the vessel radius before the bifurcation is equal to the sum of the nth power of the coronary artery radius after the bifurcation. The minimum radius of the vessel on the current coronary tree model is used as the radius of the collateral vessel, and then the number of collateral vessels is determined. This eliminates the error caused by the generation of the coronary tree model from medical images to a certain extent, and the final coronary tree model is more in line with the actual physiological condition of the human body.
[0056] (4) When generating the microcirculatory vascular model, the blood supply myocardium is equivalent to a sphere, and then a bifurcation operation is performed. The parent generation is subjected to the bifurcation operation to obtain two offspring generations, and then the blood supply myocardium of the parent generation is evenly divided into the two offspring generations until the radius of the offspring generation is less than 0.005 mm, which is more in line with the physiological condition of the human body, and the obtained microcirculatory resistance is more accurate.
[0057] (5) The calculation of FFR in this application does not require the support of invasive medical data, which reduces medical costs and avoids the risks of invasive testing. At the same time, the coronary artery tree model constructed in this application can accurately predict the effect of coronary treatment, which is helpful for the planning of clinical treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 A flowchart of a method for calculating FFR based on the ratio of blood flow before and after a stenotic lesion;
[0059] Figure 2A schematic diagram of the architecture of a system for calculating FFR based on the ratio of blood flow before and after a stenotic lesion;
[0060] Figure 3 The flowchart is a method for calculating the terminal coronary microvascular resistance;
[0061] Figure 4 Flowchart for generating microcirculatory vascular models. DETAILED DESCRIPTION
[0062] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.
[0063] Embodiment 1:
[0064] A method for calculating FFR based on the ratio of blood flow before and after the stenotic lesion, such as Figure 1 As shown, the following steps are included:
[0065] S1: Obtain routine blood data and medical images of the heart, segment and reconstruct the medical images, and establish a coronary tree model. The coronary tree model contains the aortic root, and multiple side branches are added to the coronary tree model, that is, multiple new coronary outlets are added. The volume of the myocardium and the radius of each coronary outlet are determined based on the coronary tree model;
[0066] In this embodiment, the medical image of the heart in step S1 is a cardiac CT image, the cardiac CT image is segmented, and a myocardial image is obtained through morphological operations and grayscale extraction to determine the volume of the myocardium; the cardiac CT image is segmented, and a coronary tree including the aortic root is obtained through morphological operations and grayscale extraction to determine the area of each coronary outlet, and the equivalent radius of each coronary outlet is used as the radius of each coronary outlet; the center line and the vessel wall of each blood vessel are calculated and reconstructed to establish a coronary tree model.
[0067] In step S1, p (1≤p≤7) side branches are added to the coronary artery tree model. The side branches are added on the side of the coronary artery close to the heart and between the upper and lower bifurcations of the existing coronary artery. The length of the side branches is 3-5 mm, and the radius of the side branches is the minimum radius of the existing blood vessels on the coronary artery tree model.
[0068] The value of p is determined by the change in the blood vessel area between the upper and lower bifurcations. Assuming that the following relationship exists between the blood vessel radius before and after the coronary artery bifurcation: the nth power of the blood vessel radius before the bifurcation = the sum of the nth power of the coronary artery radius after the bifurcation, the value of p satisfies the following formula:
[0069] p*(r 侧枝血管 )n =(r 上级分叉 ) n -(r 下级分叉 ) n
[0070] Among them, r 侧枝血管 represents the radius of the collateral vessel, r 上级分叉 represents the radius of the vessel downstream of the superior bifurcation, r 下级分叉 It represents the radius of the upstream blood vessel of the lower bifurcation, and the value range of n is 2.3-2.44.
[0071] S2: Determine the stenosis segment on the coronary artery tree model, determine the theoretical normal radius of the stenosis segment according to the vascular radius upstream of the stenosis segment and the vascular radius downstream of the stenosis segment, use the theoretical normal radius to correct the coronary artery tree model to obtain a non-stenosis model, which is used to represent the coronary artery tree model without stenosis lesions; the theoretical normal radius of the stenosis segment can be determined by interpolation method.
[0072] S3: Based on the volume of the myocardium and the radius of each coronary artery outlet, the volume of the myocardium supplying each coronary artery outlet is determined respectively, a microcirculatory vascular model is generated, and the microcirculatory resistance value of each coronary artery outlet is calculated based on the blood routine data and the microcirculatory vascular model;
[0073] The method of determining the volume of the myocardium supplied by the coronary artery outlet in step S3 is as follows: the volume of the myocardium supplied by each coronary artery outlet is determined in sequence, wherein the calculation formula of the volume of the myocardium supplied by the i-th (1≤i≤N) coronary artery outlet is as follows:
[0074]
[0075] Among them, V i represents the volume of the myocardium supplied by the i-th coronary artery outlet, V represents the volume of the myocardium, r i represents the radius of the i-th coronary artery outlet, r k represents the radius of the kth coronary artery outlet, N represents the number of coronary artery outlets on the coronary tree model, and the value range of n is 2.3-2.44.
[0076] The calculation method of terminal coronary microvascular resistance is as follows Figure 3 As shown in Figure 2, the process of generating a microcirculatory vascular model is as follows Figure 4 As shown, the specific process of generating the microcirculation blood vessel model and calculating the microcirculation resistance in step S3 is as follows:
[0077] A1: Initialize and generate a microcirculatory vascular model. The microcirculatory vascular model includes N coronary artery outlets. Each coronary artery outlet is used as a parent generation, and step A2 is executed;
[0078] A2: Each parent generation performs a bifurcation operation, and one parent generation obtains multiple children, each of which is a blood vessel. The blood supply myocardium of the parent generation is allocated to each child generation, and all children generations are added to the microcirculation vascular model according to the corresponding relationship between the children and the parent generation;
[0079] In step A2, a parent performs a fork operation as follows:
[0080] The blood supply myocardium of the parent generation is equivalent to an ellipsoid, and a point on the surface of the ellipsoid is taken as the root node. The root node represents the location of the parent generation's exit. The ellipsoid is evenly divided into m (2≤m≤5) sub-parts through the root node. The center of mass of each sub-part is found respectively, and the center of mass is connected with the root node. The m line segments obtained are m offspring. The volume of the blood supply myocardium of each offspring is the volume of one sub-part. The radius of the m offspring is the same, and the radius of the offspring is r 子代 The radius r of the parent 父代 Satisfies the relationship: m*(r 子代 ) n =(r 父代 ) n .
[0081] In this embodiment, when performing a bifurcation operation, the blood supply myocardium of the parent generation is equivalent to a sphere, and the ellipsoid is evenly divided into two sub-parts through the root node. A bifurcation operation is as follows: first, the blood supply myocardium of the parent generation is equivalent to a sphere, a point on the surface of the sphere is taken as the root node, and the ellipsoid is evenly divided into two hemispheres (sub-parts) through the root node, and the center of mass of each hemisphere is found respectively, and the center of mass is connected to the root node. The two line segments obtained are two offspring (two blood vessels), and the blood supply myocardium of each offspring is a hemisphere. The radius of the two offspring is the same, and the radius r of the offspring is the same. 子代 The radius r of the parent 父代 Satisfies the relationship: 2*(r 子代 ) n =(r 父代 ) n .
[0082] A3: Traverse all the children obtained in the previous step. If there are children with a radius greater than r0, take all the children with a radius greater than r0 as parents and execute step A2. Otherwise, complete the generation of the small blood vessel tree model; wherein r0 is the preset minimum blood vessel radius of the microcirculation blood vessel model; in this embodiment, the radius r0 is 0.005 mm.
[0083] A4: Based on the microcirculatory vascular model and routine blood data, the microcirculatory resistance value of each coronary artery outlet is calculated using Poiseuille's theorem and the Fallin effect (including the reversal of the Fallin effect).
[0084] In step A4, the resistance values of each section of blood vessels are calculated using Poiseuille's theorem, where the viscosity coefficient is obtained using the Fahraeus effect, the hematocrit in the Fahraeus effect is determined from blood routine data, and the microcirculation resistance values at each coronary artery outlet are calculated using the series-parallel theorem of resistors.
[0085] S4: Generate a computational fluid dynamics (CFD) calculation grid, establish a three-dimensional CFD model of the coronary artery tree model, a three-dimensional CFD model of the unstenosed model, and a one-dimensional CFD model of the microcirculation blood vessel model, and obtain the physical properties of the blood and the flow equation.
[0086] S5: Set the boundary conditions for the three-dimensional CFD model of the coronary artery tree model and the boundary conditions for the three-dimensional CFD model of the unstenosed model. The boundary conditions include inlet boundary conditions and outlet boundary conditions.
[0087] In step S5, the method for setting the boundary conditions for the three-dimensional CFD model of the coronary artery tree model and the boundary conditions for the three-dimensional CFD model of the unstenosed model is the same, and both include the following steps:
[0088] S51: Set each inlet boundary condition of the three-dimensional CFD model to the blood pressure P0 at the corresponding coronary artery orifice, and set each outlet boundary condition of the three-dimensional CFD model to a times (0 < a < 1) the blood pressure at the corresponding coronary artery orifice, that is, the pressure P at each outlet of the three-dimensional CFD model is P = a * P0.
[0089] S52: Obtain the current boundary conditions of the three-dimensional CFD model. For each outlet of the three-dimensional CFD model, solve the three-dimensional CFD model based on the physical properties of the blood and the flow equation to obtain the blood flow Q at each outlet of the three-dimensional CFD model.
[0090] S53: The inlets of the one-dimensional CFD model correspond one-to-one with the outlets of the three-dimensional CFD model. Set each inlet boundary condition of the one-dimensional CFD model to the blood flow Q at the corresponding outlet of the three-dimensional CFD model, and set each outlet boundary condition of the one-dimensional CFD model to the venous pressure Pout.
[0091] S54: Calculate the blood pressure Pin at each inlet of the one-dimensional CFD model according to the formula Pin - Pout = R * Q. If Pin is not equal to P and the difference is greater than the preset convergence accuracy, then correct each outlet boundary condition of the three-dimensional CFD model respectively. The correction formula is P` = (P * 4 + Pin) / 5, and repeat step S52. Otherwise, take the current boundary conditions of the three-dimensional CFD model as the finally set boundary conditions; where R represents the microcirculation resistance value, and P` represents the corrected outlet boundary condition of the three-dimensional CFD model.
[0092] In this embodiment, the value of a is 0.7. In other implementations, the value of a can be increased or decreased as needed.
[0093] S6: Based on the boundary conditions of the three-dimensional CFD model of the coronary tree model, the physical properties of the blood and the flow equations, the fluid parameters at each position of the coronary tree model are solved; based on the boundary conditions of the three-dimensional CFD model of the non-stenotic model, the physical properties of the blood and the flow equations, the fluid parameters at each position of the non-stenotic model are solved;
[0094] S7: Cut at least one section on the coronary tree model and the non-stenotic model respectively and obtain the average blood flow of each section. The sections of the coronary tree model correspond to the sections of the non-stenotic model one by one and the cutting positions are the same. The average blood flow of a section is the average blood flow passing through the section in one cardiac cycle. The blood flow reserve fraction of each section is calculated respectively. The blood flow reserve fraction of a section is equal to the ratio of the average blood flow of the section on the coronary tree model to the average blood flow of the corresponding section of the section on the non-stenotic model.
[0095] The selected section location includes between the upper and lower bifurcations of the coronary artery.
[0096] A system that calculates FFR based on the ratio of blood flow before and after a stenotic lesion, such as Figure 2 As shown, including:
[0097] A coronary tree model generation module is used to obtain a medical image of the heart, segment and reconstruct the medical image, obtain a coronary tree model, and add multiple side branch vessels to the coronary tree model;
[0098] A non-stenosis model generation module, used for establishing a non-stenosis model according to a coronary tree model;
[0099] Microcirculatory vascular module, used to establish a microcirculatory vascular model and calculate the microcirculatory resistance value corresponding to each coronary artery outlet on the coronary tree model;
[0100] A computational grid generation module is used to generate computational fluid dynamics computational grids for the coronary tree model, the non-stenotic model and the microcirculatory vascular model, so as to establish a three-dimensional CFD model of the coronary tree model, a three-dimensional CFD model of the non-stenotic model and a one-dimensional CFD model of the microcirculatory vascular model;
[0101] A boundary condition setting module, used to set boundary conditions of the three-dimensional CFD model of the coronary tree model and boundary conditions of the three-dimensional CFD model of the non-stenosis model;
[0102] The property setting module is used to set the physical properties of blood and the flow equation. In the property setting module, based on the routine blood data and the Farin effect formula, it is assumed that the blood is a Newtonian fluid or a generalized Newtonian fluid and adopts laminar flow, individual-specific blood density and blood viscosity. The flow equation includes a steady flow equation or an unsteady flow equation, and a Navier-Stokes equation group based on incompressible flow.
[0103] The solver obtains the fluid parameters at various locations of the coronary tree model based on the boundary conditions of the three-dimensional CFD model of the coronary tree model, the physical properties of the blood and the flow equations. The solver obtains the fluid parameters at various locations of the non-stenotic model based on the boundary conditions of the three-dimensional CFD model of the non-stenotic model, the physical properties of the blood and the flow equations. The fluid parameters include pressure, flow and velocity.
[0104] The post-processing module calculates the blood flow reserve fraction of each section based on the average blood flow of each section on the coronary tree model and the average blood flow of each section on the non-stenotic model.
[0105] In the property setting module, based on routine blood data and the Fallin effect formula, it is assumed that the blood is a Newtonian fluid or a generalized Newtonian fluid and adopts laminar flow, individual-specific blood density and blood viscosity; the flow equation includes a steady flow equation or an unsteady flow equation, and a Navier-Stokes equation group based on incompressible flow.
[0106] In actual application, the coronary tree model generation module analyzes and segments the blood vessels in the patient's medical images (such as various medical images that comply with DICOM specifications). Various open source software such as vmtk (The Vascular Modeling Toolkit) can be used to calculate and reconstruct the blood vessel centerline and blood vessel wall based on the results of analyzing and segmenting the blood vessels in the medical images, thereby forming a coronary tree model.
[0107] The computational grid generation module can utilize various open source software such as vmtk to generate high-quality computational grids that meet CFD calculation requirements for the coronary tree model and the non-stenotic model, thereby establishing a three-dimensional CFD model of the coronary tree model and a three-dimensional CFD model of the non-stenotic model.
[0108] In the prior art, it is difficult to give the microcirculation resistance at the end of the coronary artery based on individual characteristics, and the established coronary tree model cannot accurately reflect the individual's physiological characteristics. When establishing the coronary tree model, the present application adds side branches, and the obtained coronary tree model is more in line with the actual physiological conditions. Then, the microcirculation vascular model is established by continuously performing bifurcation operations, so as to accurately obtain the microcirculation resistance value of the coronary artery outlet, and then the boundary conditions are set according to the microcirculation resistance value, which can reflect the individual specificity.
[0109] A coronary artery tree model and a non-stenosis model were established, and the fluid parameters of the two models were obtained by computational fluid dynamics. The blood flow of the non-stenosis model is the blood flow of the coronary artery when there is no stenosis lesion, and the blood flow of the coronary artery tree model is the current blood flow of the coronary artery. The ratio of the blood flow in the corresponding sections of the two models was calculated to obtain the blood flow reserve fraction, which regressed to the original definition of the blood flow reserve fraction and avoided the problem that various assumptions in the clinical measurement of FFR are difficult to achieve in numerical simulation, making the calculated FFR more reliable.
[0110] In addition, this application does not require the support of invasive medical data, which reduces medical costs and avoids the risks of invasive testing. At the same time, the coronary artery tree model constructed by this application can accurately predict the effect of coronary artery treatment, which is helpful for the planning of clinical treatment plans.
[0111] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A method for calculating FFR based on the blood flow ratio before and after a stenotic lesion, characterized in that: The following steps are involved: S1: Obtain routine blood data and medical images of the heart, establish a coronary tree model based on the medical images, add multiple collateral vessels to the coronary tree model, that is, add multiple new coronary outlets, and determine the volume of the myocardium and the radius of each coronary outlet based on the coronary tree model; S2: determine the stenosis segment on the coronary artery tree model, determine the theoretical normal radius of the stenosis segment according to the vessel radius upstream of the stenosis segment and the vessel radius downstream of the stenosis segment, and use the theoretical normal radius to correct the coronary artery tree model to obtain a non-stenosis model, which is used to represent the coronary artery tree model without stenosis lesions; S3: Based on the volume of the myocardium and the radius of each coronary artery outlet, the volume of the myocardium supplying each coronary artery outlet is determined respectively, a microcirculatory vascular model is generated, and the microcirculatory resistance value of each coronary artery outlet is calculated based on the blood routine data and the microcirculatory vascular model; S4: Generate computational fluid dynamics calculation grid, establish a three-dimensional CFD model of the coronary tree model, a three-dimensional CFD model of the non-stenotic model, and a one-dimensional CFD model of the microcirculation vascular model, and obtain the physical properties and flow equations of the blood; S5: setting boundary conditions of the three-dimensional CFD model of the coronary tree model, and setting boundary conditions of the three-dimensional CFD model of the non-stenosis model, wherein the boundary conditions include an inlet boundary condition and an outlet boundary condition; S6: Based on the boundary conditions of the three-dimensional CFD model of the coronary tree model, the physical properties of blood and the flow equation, the fluid parameters at each position of the coronary tree model are solved; Based on the boundary conditions of the three-dimensional CFD model of the non-stenotic model, the physical properties of the blood and the flow equation, the fluid parameters at each position of the non-stenotic model are solved; the fluid parameters include the blood flow; S7: intercepting at least one cross section on the coronary tree model and the non-stenotic model respectively and obtaining the average blood flow of each cross section, the cross section of the coronary tree model corresponds to the cross section of the non-stenotic model one by one and the interception position is the same, the average blood flow of a cross section is the average blood flow passing through the cross section in one cardiac cycle, and the blood flow reserve fraction of each cross section is calculated respectively, the blood flow reserve fraction of a cross section is equal to the ratio of the average blood flow of the cross section on the coronary tree model to the average blood flow of the corresponding cross section of the cross section on the non-stenotic model; In the step S1, p side branches are added to the coronary artery tree model, 1≤p≤7, the side branches are added at a side of the coronary artery close to the heart, and between the upper and lower bifurcations of the existing coronary artery, the length of the side branches is 3-5 mm, and the radius of the side branches is the minimum radius of the existing blood vessels on the coronary artery tree model; The value of p is determined by the change in the blood vessel area between the upper and lower bifurcations, specifically: p*(r 侧枝血管 ) n =(r 上级分叉 ) n -(r 下级分叉 ) n Among them, r 侧枝血管 represents the radius of the collateral vessel, r 上级分叉 represents the radius of the vessel downstream of the superior bifurcation, r 下级分叉 represents the radius of the upstream vessel of the lower bifurcation, and the value range of n is 2.3-2.44; The step S3 of determining the volume of the myocardium supplied by the coronary artery outlet is specifically as follows: determining the volume of the myocardium supplied by each coronary artery outlet in turn, wherein the calculation formula for the volume of the myocardium supplied by the i-th coronary artery outlet is as follows: Among them, V i represents the volume of the myocardium supplied by the i-th coronary artery outlet, V represents the volume of the myocardium, r i represents the radius of the i-th coronary artery outlet, r k represents the radius of the kth coronary artery outlet, N represents the number of coronary artery outlets on the coronary tree model, the value range of n is 2.3-2.44, 1≤i≤N; The specific process of generating the microcirculation blood vessel model and calculating the microcirculation resistance in step S3 is as follows: A1: Initialize and generate a microcirculation blood vessel model. The microcirculation blood vessel model includes N coronary artery outlets. Take each coronary artery outlet as a parent and perform step A2; A2: Each parent performs a bifurcation operation. One parent gets multiple offspring. Each offspring is a blood vessel. Allocate the blood-supplying myocardium of the parent to each offspring, and add all the offspring to the microcirculation blood vessel model according to the corresponding relationship between the offspring and the parent; A3: Traverse all the offspring obtained in the previous step. If there are offspring with a radius greater than r0, take all the offspring with a radius greater than r0 as parents and perform step A2. Otherwise, complete the generation of the small blood vessel tree model; where r0 is the preset minimum blood vessel radius of the microcirculation blood vessel model; A4: Based on the microcirculation blood vessel model and blood routine data, calculate the microcirculation resistance values of each coronary artery outlet using Poiseuille's law and Fahraeus effect; In step A2, the specific operation of a parent performing a bifurcation operation is as follows: The blood supply myocardium of the parent generation is equivalent to an ellipsoid, and a point is taken on the surface of the ellipsoid as the root node. The root node represents the location of the parent generation's exit. The ellipsoid is evenly divided into m sub-parts through the root node, 2≤m≤5, and the center of mass of each sub-part is found respectively. The center of mass is connected with the root node, and the m line segments obtained are the m offspring. The volume of the blood supply myocardium of each offspring is the volume of one sub-part. The radius of the m offspring is the same, and the radius of the offspring r 子代 The radius r of the parent 父代 Satisfies the relationship: m*(r 子代 ) n =(r 父代 ) n ; In step S5, the boundary conditions of the three-dimensional CFD model of the coronary artery tree model are set in the same way as those of the three-dimensional CFD model of the unstenosed model, and both include the following steps: S51: Set the boundary condition of each inlet of the three-dimensional CFD model to the blood pressure P0 of the corresponding coronary artery orifice, and set the boundary condition of each outlet of the three-dimensional CFD model to a times the blood pressure of the corresponding coronary artery orifice, where 0 < a < 1, that is, the pressure P at each outlet of the three-dimensional CFD model is P = a * P0; S52: Obtain the current boundary conditions of the three-dimensional CFD model. For each outlet of the three-dimensional CFD model, solve the three-dimensional CFD model based on the physical properties of the blood and the flow equation to obtain the blood flow Q at each outlet of the three-dimensional CFD model; S53: The inlets of the one-dimensional CFD model correspond one by one to the outlets of the three-dimensional CFD model. Set the boundary condition of each inlet of the one-dimensional CFD model to the blood flow Q at the corresponding outlet of the three-dimensional CFD model, and set the boundary condition of each outlet of the one-dimensional CFD model to the venous pressure Pout; S54: Calculate the blood pressure Pin at each inlet of the one-dimensional CFD model according to the formula Pin - Pout = R * Q. If Pin is not equal to P and the difference is greater than the preset convergence accuracy, respectively correct the boundary condition of each outlet of the three-dimensional CFD model. The correction formula is P` = (P * 4 + Pin) / 5, and repeat step S52. Otherwise, take the current boundary conditions of the three-dimensional CFD model as the finally set boundary conditions; where R represents the microcirculation resistance value, and P` represents the corrected boundary condition of the outlet of the three-dimensional CFD model.
2. The method for calculating FFR based on the blood flow ratio before and after the stenosis according to claim 1, characterized in that: In step S2, the theoretical normal radius of the stenosis section is determined by the interpolation method.
3. The method for calculating FFR based on the blood flow ratio before and after the stenosis according to claim 1, characterized in that: The size of the preset minimum blood vessel radius r0 of the microcirculation blood vessel model is 0.005 mm.
4. A system for calculating FFR based on the ratio of blood flow before and after a stenotic lesion, characterized in that: Based on the method for calculating FFR as described in any one of claims 1 - 3, it includes: A coronary tree model generation module is used to obtain a medical image of the heart, segment and reconstruct the medical image, obtain a coronary tree model, and add multiple side branch vessels to the coronary tree model; A non-stenosis model generation module, used for establishing a non-stenosis model according to a coronary tree model; Microcirculatory vascular module, used to establish a microcirculatory vascular model and calculate the microcirculatory resistance value corresponding to each coronary artery outlet on the coronary tree model; A computational grid generation module, used to generate computational fluid dynamics computational grids for the coronary tree model, the non-stenotic model and the microcirculatory vascular model, so as to establish a three-dimensional CFD model of the coronary tree model, a three-dimensional CFD model of the non-stenotic model and a one-dimensional CFD model of the microcirculatory vascular model; A boundary condition setting module, used to set boundary conditions of the three-dimensional CFD model of the coronary tree model and boundary conditions of the three-dimensional CFD model of the non-stenosis model; Property setting module, used to set the physical properties and flow equations of blood; A solver, based on the boundary conditions of the three-dimensional CFD model of the coronary tree model, the physical properties of the blood and the flow equation, solves the fluid parameters of each location of the coronary tree model, and based on the boundary conditions of the three-dimensional CFD model of the non-stenotic model, the physical properties of the blood and the flow equation, solves the fluid parameters of each location of the non-stenotic model; The post-processing module calculates the blood flow reserve fraction of each section based on the average blood flow of each section on the coronary tree model and the average blood flow of each section on the non-stenotic model.
Citation Information
Patent Citations
System and method for using computational fluid mechanics to simulate and calculate fractional blood flow reserve
CN106650267A
Medical image processing methods and systems
CN107615335A
Coronary flow reserve score calculating method based on ultrasonic and CT imaging technology
CN109686450A
Coronary flow reserve calculation method based on myocardial blood flow and CT (Computed Tomography) image
CN111227822A
Method, device and system for acquiring blood flow of large artery on heart surface, and storage medium
CN112384138A
