CN108109698B - System for calculating fractional flow reserve and method for setting boundary conditions - Google Patents
System for calculating fractional flow reserve and method for setting boundary conditions Download PDFInfo
- Publication number
- CN108109698B CN108109698B CN201711396887.0A CN201711396887A CN108109698B CN 108109698 B CN108109698 B CN 108109698B CN 201711396887 A CN201711396887 A CN 201711396887A CN 108109698 B CN108109698 B CN 108109698B
- Authority
- CN
- China
- Prior art keywords
- vessel tree
- computational fluid
- dynamics model
- boundary condition
- fluid dynamics
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
Landscapes
- Medical Informatics (AREA)
- Engineering & Computer Science (AREA)
- Public Health (AREA)
- Health & Medical Sciences (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention provides a system for calculating fractional flow reserve and a method for setting boundary conditions. The system comprises: the blood vessel tree model generation module is used for acquiring a medical image, executing segmentation and reconstructing a geometric model of an individual blood vessel tree; the computational grid generating module is used for generating computational grids for the geometric model and establishing a CFD model of the blood vessel tree; the boundary condition setting module is used for setting corresponding inlet and outlet boundary conditions for the CFD model of the blood vessel tree; the attribute setting module is used for setting the physical attribute and the flow equation of the blood; the solver is used for solving the CFD model of the blood vessel tree based on the inlet and outlet boundary conditions, the set physical properties and the set flow equation so as to obtain fluid parameters of all parts of the blood vessel tree; and a post-processing module that post-processes the fluid parameters to obtain a fractional flow reserve, the inlet and outlet boundary conditions being individual-specific. The system and method achieve clinically acceptable computation times while ensuring simulation result accuracy.
Description
The application is a divisional application of Chinese invention application, which has the application date of 2016, 12 and 28 and the application number of 201611234903.1 and is named as a system and a method for simulating and calculating the fractional flow reserve by applying computational fluid mechanics.
Technical Field
The present invention relates to a system and method for computational analysis of human physiology and hemodynamics, and more particularly to a system for applying Computational Fluid Dynamics (CFD) to simulate computational fractional flow reserve, and a method for setting respective entry and exit boundary conditions for a Computational Fluid Dynamics (CFD) model of a vessel tree.
Background
The main cause of coronary heart disease is coronary stenosis due to arteriosclerosis. Coronary stenosis can cause significant changes in coronary blood flow and other hemodynamics, causing relative and absolute ischemia of the myocardium. Coronary angiography and intravascular ultrasound are considered as "gold standards" for diagnosing coronary heart disease, but they only allow imaging evaluation of the degree of stenosis, which is not known how much influence the stenosis has on distal blood flow (functional assessment). From 1995 Pijs et al proposed a new indicator of coronary Flow by pressure measurement-Fractional Flow Reserve (FFR) (Pijs NH, Van Gelder B, Van der Voort P, Peels K, Bracke FA, Bonnier HJ, El Gamma MI Flow Reserve a user effective index to estimate the initial degree of an epidemic, of an anatomical coronary stenosis on myocardial regional Flow & circulation & 5Dec 1; 92(11):3183-93 is incorporated herein by reference), and for long-term basic and clinical studies, FFR has become the "gold standard" for functional assessment of coronary stenosis, and an indispensable guiding tool for coronary reconstruction.
FFR, as an invasive measure (i.e. an invasive measure), inevitably carries a certain risk and may cause some degree of trauma to the body. With the development of science and technology and research level, especially in the fields of medical image imaging and reconstruction technology, computational fluid mechanics, high-performance computation, etc., the numerical simulation FFR (virtual FFR) is becoming a leading direction of rapid development. Virtual FFR is a non-invasive measurement compared to invasive FFR. According to the clinical trial comparison, the virtual FFR has a very high ability to diagnose myocardial ischemia from the viewpoints of accuracy, sensitivity, specificity, positive predictive value, and negative predictive value on a scale detailed to each vascular grade of individual patients, and is superior in comparison with the method by the imaging evaluation.
Because of the superiority of the non-invasive, quantitative analysis and diagnostic capabilities of virtual FFR, more and more research and commercial institutions are conducting basic and clinical research on this technology. Since this is an emerging direction, various related research and improvement methods and ways are involved in hunting, but the overall can be classified into two major directions: 1. application and model of medical imageType reconstruction 2. as an important factor in the accuracy of Computational Fluid Dynamics (CFD): and (3) researching and processing boundary conditions according with human physiological characteristics. The current mainstream computing systems include FFR based on coronary computed tomography angiographyCTvFFR based on rotational angiography, FFR based on quantitative coronary angiography and myocardial infarction thrombolysisQCAAnd the like.
As a leading-edge technology, virtual FFR also faces various technical challenges and challenges. The calculation time is a key for clinical popularization and new 'gold standard' of the technology. For methods dominated by accuracy and model complexity, computing a case takes even over 100 hours; for fast calculation methods, the optimal time can be shortened to less than 10 minutes, but this requires an experienced analyst to operate, requires the help of invasive measurements, and has limited processing capability for complex vessel tree models. Even if the speed and accuracy are balanced, the time required for analog computation by applying the virtual FFR is estimated to be over 7 hours, which causes great limitation to the clinical popularization of the method. Furthermore, another major challenge in terms of accuracy in applying CFD to virtual FFR computation is how to reasonably obtain/set the boundary conditions of the CFD model. Whether the boundary conditions are suitable or not determines the accuracy of the simulation calculation to the maximum extent.
At present, almost all mainstream blood vessel calculation systems need to assume or count to obtain the flow distribution proportion of the human blood vessel tree when setting the boundary conditions of the CFD model, which is also a bottleneck of the accuracy of the existing calculation method in the physiological characteristics of individual patients, even refined to each blood vessel of the patient. The boundary conditions set based on such assumptions or statistics do not reflect the specificity of the individual, affecting the accuracy of the simulation calculations. Furthermore, existing vessel computing systems, in order to automatically and accurately predict the fluid parameters of the clinically interesting vessel branches, typically require CFD modeling and computation of the relevant entire vessel tree (e.g. of the coronary arteries), which is time-consuming and computationally expensive, thus preventing them from entering clinical popularization.
Therefore, there is a need for a system and method for fast and automatically simulating FFR calculation via CFD and setting corresponding entry and exit boundary conditions for a hydromechanical model of a vessel tree, which can automatically, quickly and accurately derive fluid parameters for each location of a vessel tree, such as the coronary artery, independently (without depending on the experience of the analyst and without invasive measurement intervention), and from this derive functional assessment thereof, including FFR and assessment of suspected lesion location, lesion effect on myocardial ischemia, complex vascular stenosis conditions (e.g., single branch, multiple vessel lesions, multiple lesions of the same vessel, moderate stenosis lesions, continuous lesions, diffuse lesions, in-stent restenosis, etc.), etc. Preferably, the system and the method can also perform the calculation resource inclination aiming at the clinically concerned blood vessel branch, thereby taking account of the calculation speed of the whole blood vessel tree and the calculation accuracy of the clinically concerned blood vessel branch under the premise of limited calculation resources.
Disclosure of Invention
In view of the above-identified technical problems, the present invention provides a system for simulating and calculating fractional flow reserve by applying computational fluid dynamics, and a method for setting corresponding inlet boundary conditions and outlet boundary conditions for a CFD model of a vessel tree, which do not make any assumption or statistics on specific flow distribution ratios of the vessel tree in setting of the boundary conditions, but actually predict the specific flow distribution ratios by calculation (e.g. calculation based on medical images of individuals or CFD calculation of the vessel tree), thereby ensuring that the boundary conditions are individual-specific, thereby improving the accuracy of the calculation results. Moreover, another important application of the system and method is to accurately predict the flow rate of each blood vessel outlet which is clinically concerned in a complex network, such as a brain blood vessel, and to take account of the total calculation time, so that the total calculation time can be controlled within a clinically applicable range. Moreover, the system and the method can be completely realized in a noninvasive and automatic manner, an analyst can use the system conveniently without abundant experience, and the support of invasive medical images is not needed, so that the burden of the analyst is reduced, and the pain of a patient is also reduced.
According to a first aspect of the present invention there is provided a system for applying computational fluid dynamics to simulate a computed fractional flow reserve, comprising:
a vessel tree model generation module for acquiring medical images, performing segmentation and reconstructing a geometric model of the vessel tree of an individual;
a computational mesh generation module for generating a computational mesh for a geometric model of the vessel tree, thereby building a CFD model of the vessel tree, including a three-dimensional (3D) CFD model;
a boundary condition setting module, configured to set corresponding entry boundary conditions and exit boundary conditions for the established CFD model of the vessel tree;
the attribute setting module is used for setting the physical attribute and the flow equation of the blood;
the solver is used for solving the established CFD model of the blood vessel tree based on the inlet boundary condition, the outlet boundary condition, the set physical property and the set flow equation so as to obtain fluid parameters of all parts of the blood vessel tree; and
a post-processing module for post-processing the fluid parameters obtained by solving the established 3D CFD model of the vessel tree by the solver to obtain fractional flow reserve at each location of the vessel tree,
the entry boundary conditions and the exit boundary conditions, at least those boundary conditions for a three-dimensional computational fluid dynamics model of the vessel tree to derive fluid parameters for post-processing, are individual-specific.
The CFD model of the vessel tree, together with the corresponding entry and exit boundary conditions, can sometimes be used to solve for fluid parameters throughout the vessel, but sometimes only for modeling to obtain intermediate data, e.g. for determining the entry and exit boundary conditions of the 3D CFD model, which are well suited to individual specificity. Regardless of the application, the accuracy of the simulation calculation can be improved by the entrance boundary condition and the exit boundary condition that reflect the individual specificity.
Preferably, the inlet boundary conditions for the 3D CFD model for solution to fluid parameters are the flow at each inlet and are determined from medical images of the individual.
Preferably, the CFD model of the vessel tree established by the computational mesh generation module further comprises a one-dimensional (1D) CFD model; the solver is also used for solving the 1D CFD model of the blood vessel tree to obtain fluid parameters of all parts of the blood vessel tree; the boundary condition setting module sets the inlet boundary condition at an inlet and/or the outlet boundary condition at an outlet of the 3D CFD model for the vessel tree for solving to obtain a fluid parameter as the fluid parameter corresponding to each position of the vessel tree obtained by the solver solving the 1D CFD model for the vessel tree. The fluid parameter may generally be any of the flow rate, velocity and pressure of the blood, and the inlet and outlet boundary conditions may set the same or different fluid parameters, as long as the convergence is better when solving.
Preferably, the exit boundary conditions used for the 3D CFD model for solving the fluid parameters are microvascular resistances at the respective exits of the individual's vessel tree. Or, the exit boundary condition is obtained by solving a 1D CFD model of the vessel tree, specifically, the CFD model of the vessel tree established by the computational mesh generation module further includes a 1D CFD model; the solver is also used for solving the 1D CFD model of the blood vessel tree to obtain the pressure of each part of the blood vessel tree; the boundary condition setting module sets the outlet boundary condition at the outlet of the 3D CFD model of the blood vessel tree as the pressure of each part corresponding to the blood vessel tree, which is obtained by solving the 1D CFD model of the blood vessel tree by the solver.
The system can calculate the microvascular resistance at each exit of the individual's vascular tree by a variety of methods. For example, in the system, the boundary condition setting module further sets an entrance boundary condition and an exit boundary condition in a resting state for the 3D CFD model of the vessel tree; the solver is also used for solving the 3D CFD model of the blood vessel tree based on the inlet boundary condition and the outlet boundary condition in the resting state so as to obtain the pressure and the flow at each outlet of the blood vessel tree; the boundary condition setting module calculates microvascular resistance at each outlet of the individual vascular tree according to pressure and flow at each outlet of the vascular tree obtained by solving the solver under the inlet boundary condition and the outlet boundary condition in the resting state, or according to flow at each outlet of the vascular tree obtained by solving the solver under the inlet boundary condition and the outlet boundary condition in the resting state and outlet pressure in the resting state.
Preferably, the inlet boundary condition and the outlet boundary condition set for the 3D CFD model of the vessel tree are not limited to the inlet boundary condition and the outlet boundary condition in a resting state, and an initial inlet boundary condition and an initial outlet boundary condition may also be set for the three-dimensional computational fluid dynamics model of the vessel tree according to an individual medical image and/or an empirical formula, and the solver further solves the three-dimensional computational fluid dynamics model of the vessel tree based on the initial inlet boundary condition and the initial outlet boundary condition to obtain the pressure and the flow at each outlet of the vessel tree. The fluid parameters obtained by solving the initial inlet boundary conditions and the initial outlet boundary conditions cannot be directly used for post-processing, but the accuracy is enough for calculating the microvascular resistance at the mouth, the obtained microvascular resistance at the outlet can reflect the specificity of an individual as the outlet boundary conditions, and the CFD model can be well converged. The boundary condition setting module calculates microvascular resistance at each exit of the individual vascular tree according to the pressure and flow at each exit of the vascular tree obtained by solving the solver under the initial entrance boundary condition and the initial exit boundary condition
For another example, in the system, the CFD model of the vessel tree established by the computational mesh generation module further comprises a 1D CFD model; the solver is also used for solving the 1D CFD model of the blood vessel tree to obtain the flow and pressure of each part of the blood vessel tree; and the boundary condition setting module calculates the microvascular resistance at each outlet of the individual blood vessel tree according to the flow and pressure at each position corresponding to the blood vessel tree, which are obtained by solving the 1D CFD model of the blood vessel tree by the solver.
Preferably, the exit boundary conditions for the 3D CFD model for solution of the fluid parameters may also be independent of the computation of the microvascular resistance of the vessel tree, but are implemented by 3D-1D coupled computation. In particular, the computational mesh generation module generates a computational mesh for the whole or branches of the geometric model of the vessel tree, thereby building a 1D CFD model of the whole of the vessel tree and building a 3D CFD model of the whole or branches of the vessel tree; the boundary condition setting module sets inlet boundary conditions of a 1D CFD model of the whole blood vessel tree as blood flow at an inlet in a hyperemic state, sets outlet boundary conditions of the blood vessel tree as venous pressure, and sets corresponding inlet boundary conditions as fluid parameters at the inlet in the hyperemic state when a 3D CFD model of the whole blood vessel tree is established for solution to obtain the fluid parameters; and when the 3D CFD model of the branch of the blood vessel tree is established for solving to obtain the fluid parameters, the boundary condition setting module sets the corresponding entrance boundary condition as the fluid parameters of each part corresponding to the blood vessel tree obtained by solving the whole 1D CFD model of the blood vessel tree by the solver.
According to a second aspect of the present invention, there is provided a method of setting respective entry and exit boundary conditions for a CFD model of a vessel tree, the CFD model being built by acquiring medical images, performing segmentation and reconstructing a geometric model of the vessel tree of an individual, and then generating a computational mesh for the geometric model of the vessel tree, and comprising a 3D CFD model, wherein at least the entry and exit boundary conditions for a three-dimensional computational fluid dynamics model of the vessel tree are each individual-specific.
The entry boundary conditions can be set to be individual-specific in a variety of ways. Preferably, the inlet boundary conditions for the 3D CFD model for solution to fluid parameters are the flow at each inlet and are determined from medical images of the individual.
More preferably, the established CFD model of the vessel tree further comprises a 1D CFD model, the method further comprising: solving the 1D CFD model of the vessel tree to obtain fluid parameters of all parts of the vessel tree, and setting the inlet boundary condition at an inlet and/or the outlet boundary condition at an outlet of the 3D CFD model for obtaining the fluid parameters by solution as the fluid parameters of all parts of the vessel tree corresponding to the vessel tree obtained by solving the 1D CFD model of the vessel tree. The fluid parameter may be any one of a flow rate, a velocity and a pressure of the blood. The inlet and outlet boundary conditions may use the same or different kinds of fluid parameters as long as a good convergence process is ensured.
As an alternative, the exit boundary conditions used for the 3D CFD model for solution to fluid parameters are microvascular resistances at the various exits of the individual's vessel tree. The microvascular resistance at each exit of the individual's vascular tree may be calculated in various ways. Preferably, the microvascular resistance at each exit of the individual's vascular tree is calculated as follows: setting an entrance boundary condition and an exit boundary condition in a resting state for the 3D CFD model of the vessel tree; solving the 3D CFD model of the blood vessel tree based on the entrance boundary condition and the exit boundary condition in the resting state to obtain the pressure and the flow at each exit of the blood vessel tree; and calculating the microvascular resistance at each outlet according to the flow at each outlet of the blood vessel tree and the outlet pressure in the resting state, which are obtained by solving under the inlet boundary condition and the outlet boundary condition in the resting state, or according to the pressure and the flow at each outlet of the blood vessel tree, which are obtained by solving under the inlet boundary condition and the outlet boundary condition in the resting state.
Preferably, the microvascular resistance at each exit of the individual's vascular tree can be calculated as follows: setting initial inlet boundary conditions and outlet boundary conditions for a three-dimensional computational fluid dynamics model of the vessel tree according to individual medical images and/or empirical formulas; solving the three-dimensional computational fluid mechanics model of the blood vessel tree based on the initial inlet boundary condition and the initial outlet boundary condition to obtain the pressure and the flow at each outlet of the blood vessel tree; and calculating the micro-vascular resistance of each outlet of the individual blood vessel tree according to the pressure and the flow at each outlet of the blood vessel tree obtained by solving under the initial inlet boundary condition and the initial outlet boundary condition.
In addition, preferably, the microvascular resistance at each exit of the individual's vascular tree is calculated as follows: establishing a 1D CFD model of the vessel tree; solving the 1D CFD model of the blood vessel tree to obtain the flow and pressure of each part of the blood vessel tree; and calculating the micro-vascular resistance at each outlet of the individual vascular tree according to the flow and pressure at each position corresponding to the vascular tree obtained by solving the 1D CFD model of the vascular tree.
Preferably, the method may further set the outlet boundary conditions for the 3D CFD model for solution to the fluid parameters by: the established CFD model of the vessel tree further comprises a 1D CFD model; solving the 1D CFD model of the blood vessel tree to obtain the pressure of each part of the blood vessel tree; and setting the outlet boundary condition at the outlet for the 3D CFD model of the vessel tree for solution to fluid parameters as the pressure at the location corresponding to the vessel tree obtained by solving the 1D CFD model of the vessel tree.
Preferably, the method may further set an entry boundary condition of the 3D CFD model by means of 3D-1D coupling calculation: the established CFD model of the vessel tree comprises a 1D CFD model of the whole vessel tree and a 3D CFD model of the whole or branches of the vessel tree; setting an inlet boundary condition of the overall 1D CFD model of the vessel tree as blood flow at an inlet under a hyperemic state, and setting an outlet boundary condition of the vessel tree as venous pressure; and setting corresponding inlet boundary conditions as blood flow at the inlet in a hyperemic state when establishing an overall 3D CFD model of the vessel tree for resolution to obtain fluid parameters; and when a 3D CFD model of the branches of the blood vessel tree is established for solution to obtain the fluid parameters, setting the corresponding entrance boundary conditions as the flow rates corresponding to the blood vessel tree obtained by solving the integral 1D CFD model of the blood vessel tree.
Drawings
Various embodiments of the present invention are described in detail below with reference to the attached drawing figures, wherein
FIG. 1 shows a block diagram of a system for applying computational fluid dynamics to simulate calculating fractional flow reserve, according to one embodiment of the invention;
FIG. 2 is a schematic diagram illustrating a process performed in sequence by various components of the system shown in FIG. 1;
fig. 3(a) -3 (c) show examples of CFD models and boundary conditions according to another embodiment of the present invention, wherein fig. 3(a) shows a relationship between a cross-sectional view of a medical image and a geometric model of a segmentation reconstruction, fig. 3(b) shows a three-dimensional geometric model of a vessel tree including an aorta, left and right arteries, and vessel branches resulting from the segmentation reconstruction, and fig. 3(c) shows a CFD model of a vessel tree of a left artery and corresponding boundary conditions, including a vessel wall and entrance and exit boundary conditions.
FIG. 4 shows a schematic diagram of a 3D-1D coupling network of a left artery and its branches resulting from a system and method for performing 3D-1D coupling calculations according to another embodiment of the present invention.
FIG. 5 shows a graphical representation of the FFR calculated everywhere in the left artery of an individual for a system and method for 3D-1D coupling calculation according to another embodiment of the present invention.
Detailed Description
As shown in fig. 1, a system 1 for applying computational fluid dynamics to simulate calculating fractional flow reserve according to an embodiment of the present invention, the system 1 comprising: a vessel tree model generation module 11, configured to acquire a medical image, perform segmentation, and reconstruct a geometric model of a vessel tree of an individual, for example, a cross-sectional view of the acquired medical image and a geometric model of the segmentation reconstruction are shown in fig. 3 (a); a computational mesh generation module 12 for generating a computational mesh for a geometric model of the vessel tree, thereby building a CFD model of the vessel tree, the CFD model comprising a 3D CFD model; a boundary condition setting module 13, configured to set corresponding entry boundary conditions and exit boundary conditions for the established CFD model of the blood vessel tree; an attribute setting module 14 for setting physical attributes of blood and a flow equation; a solver 15, configured to solve the established CFD model of the blood vessel tree based on the entry boundary condition and the exit boundary condition, the set physical attribute, and the set flow equation, so as to obtain fluid parameters at each location of the blood vessel tree; and a post-processing module 16, configured to perform post-processing on the fluid parameters obtained by solving the established 3D CFD model of the blood vessel tree by the solver to obtain a fractional flow reserve throughout the blood vessel tree, where the boundary condition module sets at least the entry boundary condition and the exit boundary condition that are individual-specific for the 3D CFD model, so as to calculate the fluid parameters throughout the blood vessel tree for post-processing. Therefore, the accuracy of the fluid parameters at all positions of the individual blood vessel tree is ensured, and the fluid parameters at each blood vessel branch of clinical interest in the blood vessel tree can be accurately predicted.
All the components of the system complete FFR calculation together, and the system relates to a plurality of important links including processing of medical images, segmentation and three-dimensional reconstruction of vessel trees, generation of geometric models and CFD calculation grids, CFD simulation calculation, post-processing and the like. The following describes a sequential process performed by each component of the system shown in fig. 1 by taking fig. 2 as an example.
The preprocessing of the FFR calculation is done by a vessel tree model generation module 11, which reconstructs a geometric model of the vessel tree of the individual based on the results of analyzing and segmenting the vessels of the medical images of the individual, e.g. various medical images complying with the DICOM specification. The vessel tree model generation module 11 can calculate and reconstruct a vessel centerline and a vessel wall of a generated vessel tree based on a result of analyzing and segmenting the vessel by using various open source software including vmtk (the vessel Modeling toolkit), for example, so as to construct a geometric model of the vessel tree.
The CFD calculation part of the FFR calculation is completed by the computational grid generation module 12, the boundary condition setting module 13, the attribute setting module 14, and the solver 15 in cooperation.
In particular, the computational mesh generation module 12 generates a high quality computational mesh that satisfies the CFD computational requirements for a geometric model of the vessel tree using various open source software including vmtk, to build a CFD model of the vessel tree, typically a 3D CFD model of the vessel tree, for computing fluid parameters throughout the vessel tree. If the 1D CFD model of the blood vessel tree is simply established and the fluid parameters at all positions of the blood vessel tree are calculated according to the model, the calculation is almost quick and can be calculated in real time, but the accuracy of the calculated fluid parameters at all positions of the blood vessel tree is not enough to be directly used for clinical diagnosis of blood vessel symptoms, but the boundary conditions of corresponding inlets and outlets of the 3D CFD model serving as the blood vessel tree are enough, the individual specificity is fully embodied, and the accuracy of simulation calculation of the 3D CFD model can be greatly improved.
The boundary condition setting module 13 sets boundary conditions conforming to human physiological characteristics for individual patients. A CFD computational mesh can be thought of as being composed of points that spatially represent the actual geometric model. Accurate CFD computation of the complete human vascular vein requires at least hundreds of thousands of meshes, involving millions of equations, matrix calculations, and therefore, engineering applications are almost impossible to simulate the complete human vascular vein. In practice, only partial regions of interest for the result, i.e. partial regions defined by boundaries, which we will refer to as vessel trees in the following, are generally considered, which may vary according to clinical requirements, e.g. a complete vessel tree of the coronary arteries, with the entry boundary (hereinafter also simply referred to as "entry") being defined as the bifurcation of the left and right aorta with the aorta and the exit boundary (hereinafter also simply referred to as "exit") being defined as the microvasculature of the vessel tree of the coronary arteries. Different vessel trees may also be defined by the setting of the boundary conditions, and the individual-specific setting of the boundary conditions is described below only with the complete vessel tree of the coronary arteries as an example, but the vessel tree is not limited thereto.
Fig. 3(a) -3 (c) show examples of CFD models and boundary conditions according to another embodiment of the present invention, wherein fig. 3(a) shows a cross-sectional view of a medical image, a geometric model of a segmentation reconstruction and a relationship between the two, fig. 3(b) shows a three-dimensional geometric model of a vessel tree resulting from the segmentation reconstruction, including an aorta, left and right aorta and major vessel branches, and fig. 3(c) shows a CFD model of a vessel tree of the left aorta and corresponding boundary conditions, including a vessel wall and inlet and outlet boundary conditions ( outlet 1, 2, …, 8).
In the prior art, the boundary condition setting module 13 can only set the entrance boundary condition of the blood vessel tree by counting or assuming the flow rate ratio at the entrance of the left and right aorta. The prior art also sets for example a universal inlet flow: the FFR measurement and simulation comparison of a large number of clinical cases are utilized to determine the optimized value of the inlet flow of the left aorta and the right aorta, and the value is applied to the calculation of all cases. This arrangement does not reflect the specificity of the individual, and the statistical or hypothetical entry boundary condition is not necessarily applicable to the individual, thereby reducing the accuracy of the CFD calculation result.
The present invention provides various examples of methods of setting respective entry and exit boundary conditions for a CFD model of a vessel tree, which achieve individual specificity of the entry and exit boundary conditions. Examples of these methods can be used by the above relevant modules in the system 1 (an explanation of how they are implemented by the relevant modules in the system 1 is given below in describing examples of these methods, but no limitation of the modules implemented is intended), or by additional modules in the system to determine individual specific entry and exit boundary conditions and to transmit to the boundary condition setting module 13 for use and setting.
The property setting module 14 is used for setting physical properties of blood, such as but not limited to, the blood being newtonian fluid and adopting laminar flow, the blood density and blood flow viscosity of an individual according to physiological characteristics of a human body, and flow equations, including, for example, the flow of the blood being unsteady flow, the Navier-Stokes equation set based on non-constant flow, and the like. The user can manually enter settings, or default settings, or select from a library of flow equations and physical properties of the blood carried by the system, as the case may be.
The solver 15 is configured to solve the established CFD model of the blood vessel tree based on the set inlet boundary condition and outlet boundary condition, the set physical property and the set flow equation to obtain an intermediate result or obtain fluid parameters at various positions of the blood vessel tree. The CFD model may include one-dimensional and three-dimensional, and as described above, in order to ensure the accuracy of the calculation results, the fluid parameters at various positions of the blood vessel tree obtained by solving the 1D CFD model are usually only used as intermediate results and are not directly used for clinical diagnosis of the pathological state of the blood vessel. The 3D CFD model can then be solved for the fluid parameters throughout the vessel tree under the corresponding entry and exit boundary conditions, which are sometimes set only for obtaining intermediate results and not for the fluid parameters throughout the vessel tree.
Post-processing of the FFR calculation is implemented using a post-processing module 16. In particular, the post-processing module 16 is configured to post-process the fluid parameters obtained by solving the established 3D CFD model of the vessel tree by the solver 15 to obtain Fractional Flow Reserve (FFR) throughout the vessel tree. For example, from the result of solving the 3D CFD model, the relationship FFR ═ P is usedd/Pa(PdPressure everywhere in the blood vessel, PaThe average pressure at the aorta) the FFR can be calculated throughout the vessel. Optionally, a case analysis report can be generated by post-processing the results of the 3D CFD model solution, preferably the FFR values at the distal end of the stenosis can be written into the case analysis report as final calculated values if FFR<0.75, this indicates that the stenosis may lead to functional ischemia of the vessel, requiring interventional therapy. The result of the solution to the 3D CFD model also includes other clinically interesting parameters, such as FFR pull-back (pull-back) curve, blood flow velocity and pressure distribution in various parts of the blood vessel, flow distribution in various branches of the blood vessel, shear stress in various parts of the blood vessel wall, etc., and these clinically interesting parameters can also be written into a case analysis report as needed for the doctor to use as a reference for diagnosis.
Various examples of the inventive method for setting respective entrance and exit boundary conditions for solving resulting fluid parameters throughout a vessel for a CFD model of a vessel tree, each example achieving individual specificity of the entrance and exit boundary conditions, are described in detail below. At least the entry boundary conditions and the exit boundary conditions for the 3D CFD model of the vessel tree may be derived from a medical image of the individual or via a solution of a computational fluid dynamics model of the vessel tree of the individual. In particular, in order to solve for post-processing fluid parameters, individual-specific inlet and outlet boundary conditions for the 3D CFD model may be derived from medical images or historical data of the individual, may be obtained accordingly via the solution of the 1D CFD model, or may be obtained by the solution of the 3D CFD model under the initial inlet and outlet boundary conditions.
Entry boundary condition
Taking the blood vessel tree of coronary artery as an example, the method calculates the left and right aorta separately, the entrance boundary is the bifurcation (branch) of the left and right aorta and aorta, and the blood flow of the left and right aorta at the bifurcation is set as the entrance boundary condition, which can be subdivided into two modes.
The first method is to calculate the myocardial mass of the case according to the individual medical image, obtain the aortic blood flow by using the relationship between the myocardial mass and the aortic blood flow, divide the left and right aorta in proportion (60% for left aorta and 40% for right aorta), thus obtain the blood flow of the left and right aorta at the intersection, and set it as the entrance boundary condition. Note that this is just one example of determining the flow at each entrance from a medical image of an individual, using the correspondence between myocardial mass and aortic blood flow, and that, depending on the cardiovascular tree to be modeled, it is also possible to use medical images of other individuals to calculate and set the blood flow at each entrance as an entrance boundary condition for the CFD model of that particular cardiovascular tree from other correspondences between information in the other medical images and the blood flow at the entrance.
The entry boundary condition of the 3D CFD model can also be set by building a complete 1D CFD model of the vessel tree (for example, using the vessel tree model generation module 11 and the computational mesh generation module 12 in the system 1), solving the model (for example, using the solver 15 in the system 1) to calculate the flow rate at each position in the complete vessel tree, and applying the calculated flow rate at each position in the complete vessel tree to the entry boundary of the 3D CFD model of the vessel tree (for example, using the boundary condition setting module 13 in the system 1). This second approach well remedies the drawbacks of the first approach: the blood flow of the blood vessels at the entrance sometimes cannot be determined from medical images of the individual. Although the second way is exemplified by a flow rate, any one or both of a flow rate, a pressure, and a velocity of blood may be used as the inlet and outlet boundary conditions, respectively. It is also possible to use the same fluid parameters for the inlet and outlet boundary conditions, as long as it is ensured that the convergence of the 3D CFD model meets the clinical requirements. Even if individual-specific entrance and exit boundary conditions are not set for the 1D CFD model, since the 1D CFD model itself reflects the individual blood vessel characteristics, the individual specificity is well reflected by the flow rates at various places corresponding to the entrance boundary of the 3D CFD model in the complete blood vessel tree obtained during convergence through the CFD iterative solution, and thus, the individual-specific entrance boundary conditions are set for the 3D CFD model. The 1D CFD calculation requires only a very small number of meshes and is very fast in calculation speed (which can be considered as real-time calculation), so that the entry boundary conditions of the 3D CFD model obtained by the 1D CFD calculation do not affect the calculation speed of the complete process of fluid parameters such as FFR while promoting the accuracy of the calculation results.
Outlet boundary condition
There are two types of setting of exit boundary conditions:
first, the exit boundary condition is set to the microvascular resistance throughout the vascular tree for the coronary arteries of the individual patient; in the second way, similar to the second setting of the entrance boundary condition, the 1D CFD calculation result is applied to the exit of the 3D CFD model, and the exit boundary condition of the 3D CFD model is set as the pressure at the corresponding position in the blood vessel obtained by the solution of the 1D CFD model of the blood vessel tree.
The first approach can be subdivided into two sub-approaches:
I. based on 3D CFD calculation
The inlet boundary Condition of the 3D CFD model is set as the average flow Q in the cardiac output period by utilizing a phenomenon of human body hemodynamics, namely the fact that the pressure change of a blood vessel section is very small (no matter whether the blood vessel section has stenosis) in a Resting state (stopping Condition)inSetting its outlet boundary condition to the corresponding mean pressure poutThe predicted flow Q at each exit boundary is calculated by solving the 3D CFD model under the entrance and exit boundary conditions in the resting state as abovei,out(i is the number of each outlet), using the relation Pout＝Qi,outRiCalculating microvascular resistance R at each exit boundaryi。
Of course, the pressure P at each exit of the vessel tree obtained by solving under the entrance boundary condition and the exit boundary condition in the resting state may also be usedi,outSum flow rate Qi,outUsing the relation Pi,out＝Qi,outRiCalculating microvascular resistance R at each exit boundaryi。
While it is convenient to use inlet and outlet boundary conditions at rest to calculate microvascular resistance at each outlet based on 3D CFD, there are other clinically feasible approaches. For example, the inlet and outlet boundary conditions may be initial inlet and outlet boundary conditions derived from individual medical images and/or empirical formulas (here "initial" is used to distinguish from those boundary conditions set for the 3D CFD model to directly calculate fluid parameters for post-processing). For example, the average cardiac output can be calculated from an ultrasound doppler image of, for example, the heart of an individual, and the flow at each location (e.g., left and right aorta, microvessels) can be obtained as the initial inlet and outlet boundary conditions from empirical formulas regarding the proportion of how the flow is distributed among the various levels of the vessel branches after the aortic valve blood is pumped out.
Based on 1D CFD calculation: the 1D CFD calculation requires only a very small number of meshes, and is very fast (considered to be real-time)Computation), with which a 1D CFD model of the complete coronary tree is built, the entry boundary conditions are set according to any of the above-described ways of setting the entry boundary conditions, and the exit boundary conditions can all be set to venous pressure (typically 0). Obtaining the flow and pressure of each outlet corresponding to the 3D CFD model in the 1D CFD model by using the calculation of the 1D CFD under the set inlet boundary condition and outlet boundary condition, and using a relational expression Pi,out＝Qi,outRiCalculating microvascular resistance R of each exit boundary of the 3D CFD modeli. Fig. 4 shows a schematic diagram of a 3D-1D coupling network. The application can greatly simplify the solving of the outlet boundary condition microvascular resistance in the 3D CFD calculation.
The second approach uses 1D CFD calculations to derive pressures throughout the vessel tree of the complete coronary artery and applies the corresponding pressures to the exit boundaries of the 3D CFD model. Of course, instead of pressure, other kinds of fluid parameters, including flow, velocity, etc. may be used as outlet boundary conditions for the 3D CFD model.
Both the 1D CFD calculation and the 3D CFD calculation involved in setting the exit boundary condition as described above can be cooperatively realized by the vessel tree model generation module 11, the computational mesh generation module 12, the boundary condition setting module 13, the attribute setting module 14, and the solver 15 in the system 1. For example, the functions of 1D and 3D model generation, mesh generation, boundary condition setting, attribute setting, and solution may be integrated into each of the above-described components in the system 1, or separate subcomponents for 1D and 3D may be included in each of the above-described components, 1D CFD calculation may be realized by each of the subcomponents for 1D, and 3D CFD calculation may be realized by each of the subcomponents for 3D.
For example, solver 15 may also include a 1D CFD solver and a 3D CFD solver, which may be used to calculate the entry boundary conditions and/or the exit boundary conditions of a 3D CFD model of the vessel tree (the 3D CFD model and the corresponding entry boundary conditions and exit boundary conditions are used to calculate fluid parameters everywhere in the vessel tree for clinical diagnosis) as well as to calculate fluid parameters everywhere in the vessel tree for clinical diagnosis.
Based on a high-performance computational CFD solver, FFR under Hyperemia status of the left and right aortic vascular trees (Hyperemia) and other relevant blood flow parameters of clinical interest can be simulated.
For example, the blood flow of the left and right aorta in the hyperemic state is applied as the inlet boundary condition by the boundary condition setting module 13, here, the blood flow of the left and right aorta in the hyperemic state may be determined according to the medical image and/or empirical formula of the individual, may be obtained by solving the 1D CFD model, may be obtained by solving the 3D CFD model under the initial import-export boundary conditions, examples of which are given above, and the microvascular resistance at each exit boundary obtained in the various ways described above is applied as an exit boundary condition, the attribute setting module 14 may set the vascular fluid to be newtonian fluid and laminar flow, set the blood density and blood flow viscosity according to the physiological characteristics of the human body, and calculate the unsteady flow until the result converges on the 3D CFD model of the vascular tree by the solver 15, so as to obtain the fluid parameters at each position of the vascular tree. The steps performed by the post-processing module 16 are not described in detail herein, and the FFR at each position in the blood vessel tree is calculated accordingly.
The above method calculates FFR, relies on the calculation of microvascular resistance of the coronary arteries as an exit boundary condition, and finally derives fluid parameters everywhere in the vessel tree by 3D CFD modeling and calculation of the complete vessel tree. In practice, there is a need to further simplify and speed up the setting of boundary conditions, as well as to further speed up the overall calculation of fluid parameters. Furthermore, physicians require that CFD modeling not only be able to accurately simulate and compute the complete coronary artery, but also to separately simulate and compute a specific vessel branch (such as a suspected lesion segment), and that the accuracy of the simulated computation is different from one coronary artery to another. For example, a physician may wish to quickly obtain a relatively accurate FFR distribution of an intact coronary artery by real-time 1D CFD modeling calculations, so that a particular vessel branch requiring more clinical attention can be manually or automatically selected therefrom, and a relatively time-consuming but more accurate 3D CFD modeling calculation can be performed on that particular vessel branch.
The present invention provides a system and boundary condition setting method for performing 3D-1D coupled calculations (which we refer to as virtual digital subtraction angiography) to meet the above upgrade requirements, which combines the features and respective advantages of 3D-1D CFD to greatly simplify the requirements for boundary conditions. The method can relatively accurately simulate and calculate the complete coronary artery, and can more accurately and specifically simulate and calculate the specific blood vessel branch (such as a suspected lesion segment) independently, so that the time required by calculation is further shortened greatly, and the accuracy of the simulated calculation of the specific blood vessel branch with more clinical attention is preferentially ensured.
The method utilizes an attribute setting module 14 to set the blood in the blood vessel to be Newtonian fluid and laminar flow, sets the blood density and blood flow viscosity which accord with the physiological characteristics of the human body, and sets the flow equation to be unsteady flow until the result is converged to a solver 15. Depending on the specific requirements for solving the fluid parameters, the 3D CFD mesh and model for the specific requirements may be generated for the vessel tree of the entire left and right aorta, or for a specific vessel branch of a certain segment; the 1D CFD mesh and model are generated for the vessel tree of the complete coronary artery, and this step can be implemented by the vessel tree model generation module 11 and the computational mesh generation module 12 in the system 1.
And setting corresponding boundary conditions for the 1D CFD model and the 3D CFD model of the blood vessel tree by using the boundary condition setting module 13. Here, the entrance boundary condition of the 1D CFD model is set to the blood flow of the left and right aorta in the hyperemic state, and the exit boundary condition of the 1D CFD model is set to the venous pressure (usually 0).
Different entry boundary conditions are set for whether the vessel trees of the entire left and right aorta or only for a specific vessel branch are simulated by the 3D CFD model. Specifically, if the vessel tree of the whole left and right aorta is simulated, the inlet boundary condition is set as the blood flow of the left and right aorta in the hyperemia state; if a specific blood vessel branch is simulated, the inlet boundary condition is set as the calculated flow rate at the corresponding position under the inlet boundary condition and the outlet boundary condition (venous pressure) set as above by using the 1D CFD model, and other fluid parameters besides the flow rate can also be used here.
The outlet boundary condition of the 3D CFD model is set to a pressure value calculated at the corresponding position using the 1D CFD model under the inlet boundary condition and the outlet boundary condition (venous pressure) set as above, and fluid parameters other than pressure may be used here.
Thus, the coupling calculation of 3D-1D CFD is realized. Specifically, based on the processing of the 1D and 3D meshes and boundary conditions, a 1D CFD solver is used to perform calculation, the calculation result of the 1D CFD is applied to the 3D boundary according to the processing of the 3D boundary conditions, and the 3D CFD solver is used to perform calculation to obtain the fluid parameters at each required position in the blood vessel tree.
In the 3D and 1D CFD calculation, the attribute setting module 14 is used for setting a Navier-Stokes (N-S) equation set (mass and momentum conservation equation) of the non-compressible flow:
▽·u＝0
u is the fluid velocity vector, p is the pressure, ρ is the fluid density, and ρ is the fluid' S kinematic viscosity, so that the 1D CFD solver and the 3D CFD solver (or solver 15 by integrating the functions of the 1D CFD and 3D CFD solver) solve based on the non-compressible flow N-S equation system.
And (3) CFD calculation post-processing: according to the calculated 3D CFD result, using the relation FFR ═ Pd/Pa(PdPressure everywhere in the blood vessel, PaThe average pressure at the aorta) the FFR can be calculated throughout the vessel. Optionally, a case analysis report can be generated by post-processing the result of the 3D CFD model solution, preferably, the FFR value of the distal segment of the stenosis can be written into the case analysis report as the final calculated value if the FFR is<0.75, this indicates that the stenosis may lead to functional ischemia of the vessel, requiring interventional therapy. The results of solving the 3D CFD model also include other parameters of clinical interest, such as FFR pull-back curves, vessel locationsThe blood flow velocity and pressure distribution, the flow rate distribution of each branch of the blood vessel, the shear stress at each part of the blood vessel wall, and the like, and these clinically significant parameters may be written into a case analysis report as necessary for the doctor to refer to the diagnosis.
Taking the example of the intention to model the left aorta in coronary arteries alone, fig. 4 shows a 1D CFD mesh generated for the complete coronary arteries including distal microvessels and a 3D CFD mesh generated for the left aorta of clinical interest only, with the FFR distribution everywhere in the left aorta obtained by the above-described 3D-1D coupling calculation, as shown in fig. 5, the 3D-1D coupling calculation results proved to provide sufficient accuracy and clinically acceptable time consumption, which can be controlled within 1 hour.
The invention highly integrates and automates the whole calculation process, including the processing of medical images, the reconstruction of calculation models, the calculation and the post-processing of results, and realizes the rapid and accurate calculation of large batches of cases. For example, and without limitation, the C/C + + language may be used to develop and optimize the above-mentioned 1D and/or 3D CFD simulation calculation flow and its coupled calculation flow, and each module in the system 1 and its implemented whole calculation flow (including pre-processing of FFR calculation, CFD calculation part of FFR calculation, and post-processing of FFR calculation, result analysis and report generation, etc.) may be automatically implemented using Python, Bash Script, etc., so as to improve user friendliness while ensuring accuracy of simulation results.
The above description of the embodiments is only intended to facilitate the understanding of the core ideas of the present invention. It should be noted that various changes and modifications could be made by those skilled in the art without departing from the principle of the invention, and these changes and modifications also fall into the scope of the invention as claimed.
Claims (18)
1. A system for calculating fractional flow reserve using computational fluid dynamics, comprising:
a vessel tree model generation module for acquiring medical images, performing segmentation and reconstructing a geometric model of the vessel tree of an individual;
a computational mesh generation module for generating a computational mesh for the geometric model of the vessel tree, thereby building a computational fluid dynamics model of the vessel tree, including a three-dimensional computational fluid dynamics model;
a boundary condition setting module for setting corresponding inlet boundary conditions and outlet boundary conditions for the established computational fluid dynamics model of the vessel tree;
the attribute setting module is used for setting the physical attribute and the flow equation of the blood;
the solver is used for solving the established computational fluid mechanics model of the blood vessel tree based on the inlet boundary condition, the outlet boundary condition, the set physical property and the set flow equation so as to obtain fluid parameters of all parts of the blood vessel tree; and
a post-processing module for post-processing the fluid parameters obtained by solving the established three-dimensional computational fluid dynamics model of the vessel tree by the solver to obtain the fractional flow reserve at each position of the vessel tree,
at least the entry boundary conditions and the exit boundary conditions for the three-dimensional computational fluid dynamics model of the vessel tree are individual-specific and are derived from a medical image of the individual or are derived accordingly via a solution of the computational fluid dynamics model of the vessel tree of the individual,
the computational fluid dynamics model of the vessel tree established by the computational mesh generation module further comprises a one-dimensional computational fluid dynamics model,
the computational mesh generation module generates a computational mesh for an entirety or branches of the geometric model of the vessel tree, thereby building a three-dimensional computational fluid dynamics model of the entirety or branches of the vessel tree,
the boundary condition setting module is configured to, when building a three-dimensional computational fluid dynamics model of the entirety of the vessel tree for solution to fluid parameters:
setting an inlet boundary condition of the one-dimensional computational fluid dynamics model of the vessel tree as blood flow at an inlet under a hyperemic state, and setting an outlet boundary condition thereof as venous pressure; and
setting the inlet boundary conditions of the integral three-dimensional computational fluid dynamics model of the blood vessel tree as fluid parameters at the inlet in a hyperemic state, and setting the corresponding outlet boundary conditions as the fluid parameters at each position corresponding to the blood vessel tree obtained by solving the one-dimensional computational fluid dynamics model of the blood vessel tree.
2. The system of claim 1, wherein the entry boundary conditions for the three-dimensional computational fluid dynamics model of the vessel tree as a whole are flow rates at each entry and are determined from medical images of individuals.
3. The system of claim 1, wherein when building a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters,
the solver also solves the one-dimensional computational fluid mechanics model of the vessel tree to obtain fluid parameters of all parts of the vessel tree,
the boundary condition setting module sets the inlet boundary condition at an inlet and/or the outlet boundary condition at an outlet of a three-dimensional computational fluid dynamics model for branches of the blood vessel tree for solution to obtain fluid parameters as the fluid parameters at each position corresponding to the blood vessel tree obtained by the solver performing solution on the one-dimensional computational fluid dynamics model of the blood vessel tree.
4. The system of claim 1, wherein the exit boundary conditions for a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters are microvascular resistances at respective exits of individual vessel tree branches.
5. The system of claim 3,
the fluid parameter is selected from the group consisting of flow rate, velocity, and pressure of blood.
6. The system of claim 4, wherein when building a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters,
the boundary condition setting module further sets an entrance boundary condition and an exit boundary condition in a resting state for a three-dimensional computational fluid dynamics model of branches of the vessel tree,
the solver also solves the three-dimensional computational fluid mechanics model of the branches of the blood vessel tree based on the inlet boundary condition and the outlet boundary condition in the resting state to obtain the pressure and the flow at each outlet of the branches of the blood vessel tree,
the boundary condition setting module is configured to: and calculating the microvascular resistance at each outlet of the individual vessel tree branches according to the pressure and flow at each outlet of the vessel tree branches obtained by solving the solver under the inlet boundary condition and the outlet boundary condition in the resting state.
7. The system of claim 4, wherein the system is a mobile phone
The boundary condition setting module further sets initial entry boundary conditions and exit boundary conditions for a three-dimensional computational fluid dynamics model of the branches of the vessel tree based on individual medical images and/or empirical formulas,
the solver further solves a three-dimensional computational fluid dynamics model of the branches of the vessel tree based on the initial inlet boundary conditions and outlet boundary conditions to obtain pressures and flows at the respective outlets of the branches of the vessel tree,
and the boundary condition setting module calculates micro-vessel resistance at each outlet of the individual vessel tree branches according to the pressure and flow at each outlet of the vessel tree branches obtained by solving the solver under the initial inlet boundary condition and the outlet boundary condition.
8. The system of claim 4, wherein when building a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters,
the solver also solves the one-dimensional computational fluid mechanics model of the vessel tree to obtain the flow and pressure at each location of the vessel tree,
the boundary condition setting module calculates micro-vessel resistance at each outlet of each branch of the individual vessel tree according to the flow and pressure at each position corresponding to the vessel tree, which are obtained by solving the one-dimensional computational fluid mechanics model of the vessel tree by the solver.
9. The system of claim 1, wherein the boundary condition setting module is configured to, when building a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters,
setting an inlet boundary condition of the one-dimensional computational fluid dynamics model of the vessel tree as blood flow at an inlet under a hyperemic state, and setting an outlet boundary condition thereof as venous pressure; and
setting the inlet boundary condition and the outlet boundary condition of the three-dimensional computational fluid dynamics model of the branches of the blood vessel tree as the fluid parameters of each part corresponding to the blood vessel tree, which are obtained by solving the one-dimensional computational fluid dynamics model of the blood vessel tree by the solver.
10. A method of setting boundary conditions for application to a computational fluid dynamics model of a vessel tree, the computational fluid dynamics model being built by taking a medical image, performing segmentation and reconstructing a geometric model of the vessel tree of an individual, and then generating a computational mesh for the geometric model of the vessel tree, and comprising a three-dimensional computational fluid dynamics model,
it is characterized in that the preparation method is characterized in that,
at least the entry boundary conditions and the exit boundary conditions for the three-dimensional computational fluid dynamics model of the vessel tree are individual-specific and derived from a medical image of the individual or via a solution of the computational fluid dynamics model of the vessel tree of the individual,
the established computational fluid dynamics model of the vessel tree further comprises a one-dimensional computational fluid dynamics model and a three-dimensional computational fluid dynamics model of the whole or branches of the vessel tree,
the method further comprises, when building a three-dimensional computational fluid dynamics model of the vessel tree as a whole for solution to fluid parameters:
setting the inlet boundary condition of the one-dimensional computational fluid dynamics model of the blood vessel tree as blood flow at an inlet under a hyperemic state, setting the outlet boundary condition of the one-dimensional computational fluid dynamics model of the blood vessel tree as venous pressure, and solving the one-dimensional computational fluid dynamics model of the blood vessel tree to obtain fluid parameters at each position of the blood vessel tree; and
setting the inlet boundary conditions of the integral three-dimensional computational fluid dynamics model of the blood vessel tree as fluid parameters at the inlet in a hyperemic state, and setting the corresponding outlet boundary conditions as the fluid parameters at each position corresponding to the blood vessel tree obtained by solving the one-dimensional computational fluid dynamics model of the blood vessel tree.
11. The method of claim 10, wherein the entry boundary conditions are flow rates at each entry when building a three-dimensional computational fluid dynamics model of the vessel tree as a whole for solution to fluid parameters, and are determined from medical images of individuals.
12. The method of claim 10,
the method further comprises, when building a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters:
solving the one-dimensional computational fluid dynamics model of the vessel tree to obtain fluid parameters throughout the vessel tree, an
Setting the inlet boundary conditions at the inlet and/or outlet boundary conditions at the outlet for solving the three-dimensional computational fluid dynamics model for the branches of the vessel tree to the fluid parameters at the places corresponding to the vessel tree obtained by solving the one-dimensional computational fluid dynamics model of the vessel tree.
13. The method according to claim 10, wherein the exit boundary condition for a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution of fluid parameters is microvascular resistance at each exit of an individual vessel tree branch.
14. The method of claim 12,
the fluid parameter is selected from the group consisting of flow rate, velocity, and pressure of blood.
15. The method of claim 13, wherein the microvascular resistance at each exit of an individual's vessel tree branch is calculated as follows:
setting an entrance boundary condition and an exit boundary condition in a resting state for the three-dimensional computational fluid dynamics model of the vessel tree branches,
solving the three-dimensional computational fluid dynamics model of the branches of the vessel tree based on the inlet boundary condition and the outlet boundary condition in the resting state to obtain the pressure and the flow at each outlet of the branches of the vessel tree,
and calculating the microvascular resistance at each outlet according to the pressure and flow at each outlet of the blood vessel tree branch obtained by solving under the entrance boundary condition and the exit boundary condition in the resting state.
16. The method of claim 13, wherein the microvascular resistance at each exit of an individual's vessel tree branch is calculated as follows:
setting initial inlet boundary conditions and outlet boundary conditions for a three-dimensional computational fluid dynamics model of the branches of the vessel tree according to individual medical images and/or empirical formulas,
solving a three-dimensional computational fluid dynamics model of the branches of the vessel tree based on the initial inlet and outlet boundary conditions to obtain pressures and flows at the respective outlets of the branches of the vessel tree,
and calculating the micro-vessel resistance of each outlet of the individual vessel tree branches according to the pressure and the flow of each outlet of the vessel tree branches obtained by solving under the initial inlet boundary condition and the initial outlet boundary condition.
17. The method of claim 13, wherein the microvascular resistance at each exit of an individual's vessel tree branch is calculated as follows:
establishing a one-dimensional computational fluid dynamics model of the vessel tree,
solving the one-dimensional computational fluid mechanics model of the vessel tree to obtain the flow and pressure at each position of the vessel tree,
and calculating the micro-vessel resistance at each outlet of each branch of the individual vessel tree according to the flow and pressure at each position corresponding to the vessel tree obtained by solving the one-dimensional computational fluid mechanics model of the vessel tree.
18. The method of claim 14, further comprising, when building a three-dimensional computational fluid dynamics model of the branches of the vessel tree for solution to fluid parameters:
setting the inlet boundary condition of the one-dimensional computational fluid dynamics model of the blood vessel tree as blood flow volume at an inlet in a hyperemic state, setting the outlet boundary condition of the one-dimensional computational fluid dynamics model of the blood vessel tree as venous pressure, and solving the one-dimensional computational fluid dynamics model of the blood vessel tree to obtain fluid parameters at each position of the blood vessel tree; and
setting the inlet boundary condition and the outlet boundary condition of the three-dimensional computational fluid dynamics model of the branches of the blood vessel tree as fluid parameters of each part corresponding to the blood vessel tree obtained by solving the one-dimensional computational fluid dynamics model of the blood vessel tree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711396887.0A CN108109698B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711396887.0A CN108109698B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
| CN201611234903.1A CN106650267B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611234903.1A Division CN106650267B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108109698A CN108109698A (en) | 2018-06-01 |
| CN108109698B true CN108109698B (en) | 2021-04-20 |
Family
ID=58832571
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611234903.1A Active CN106650267B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
| CN201711396887.0A Active CN108109698B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611234903.1A Active CN106650267B (en) | 2016-12-28 | 2016-12-28 | System for calculating fractional flow reserve and method for setting boundary conditions |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN106650267B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650267B (en) * | 2016-12-28 | 2020-03-17 | 北京昆仑医云科技有限公司 | System for calculating fractional flow reserve and method for setting boundary conditions |
| CN107491636B (en) * | 2017-07-26 | 2021-03-16 | 武汉大学 | Cerebrovascular reserve force simulation system and method based on computational fluid dynamics |
| CN108830848B (en) * | 2018-05-25 | 2022-07-05 | 深圳科亚医疗科技有限公司 | Device and system for determining a sequence of vessel condition parameters on a vessel using a computer |
| CN109036551B (en) * | 2018-07-10 | 2021-05-11 | 北京心世纪医疗科技有限公司 | Coronary artery physiological index relation establishing and applying method and device |
| CN110599444B (en) * | 2018-08-23 | 2022-04-19 | 深圳科亚医疗科技有限公司 | Device, system and non-transitory readable storage medium for predicting fractional flow reserve of a vessel tree |
| CN110367965B (en) * | 2018-09-19 | 2022-03-08 | 苏州润迈德医疗科技有限公司 | Method, device and system for conveniently measuring coronary artery blood vessel evaluation parameters |
| CN111227822B (en) * | 2018-11-28 | 2022-02-11 | 苏州润迈德医疗科技有限公司 | Coronary artery blood flow reserve fraction calculation method based on myocardial blood flow volume and CT image |
| CN109717953B (en) * | 2019-02-01 | 2021-01-08 | 杭州晟视科技有限公司 | Method and device for determining blood flow, electronic equipment and computer storage medium |
| KR102130254B1 (en) * | 2019-10-15 | 2020-07-03 | 주식회사 실리콘사피엔스 | Method and apparatus for simulating blood flow of a subject-specific blood vessel |
| CN111067494B (en) * | 2019-12-27 | 2022-04-26 | 西北工业大学 | Microcirculation resistance rapid calculation method based on blood flow reserve fraction and blood flow resistance model |
| CN112107362B (en) * | 2020-08-24 | 2022-05-20 | 江苏大学 | Computer-assisted surgery design system for coronary heart disease |
| CN114882977A (en) * | 2021-10-12 | 2022-08-09 | 青岛埃米博创医疗科技有限公司 | Method for evaluating radioactive embolism efficiency based on fluid dynamics |
| CN114664455A (en) * | 2022-03-14 | 2022-06-24 | 北京深睿博联科技有限责任公司 | Coronary artery blood flow reserve fraction calculation method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103300820A (en) * | 2012-03-13 | 2013-09-18 | 西门子公司 | Method and system for non-invasive functional assessment of coronary artery stenosis |
| CN105096388A (en) * | 2014-04-23 | 2015-11-25 | 北京冠生云医疗技术有限公司 | Computational Fluid Dynamics (CFD) based coronary artery blood flow simulating system and method |
| CN105764410A (en) * | 2013-10-22 | 2016-07-13 | 皇家飞利浦有限公司 | Fractional flow reserve (FFR) index with adaptive boundary condition parameters |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104768465B (en) * | 2012-11-06 | 2018-06-19 | 皇家飞利浦有限公司 | Blood flow reserve score (FFR) index |
| WO2015059706A2 (en) * | 2013-10-24 | 2015-04-30 | Cathworks Ltd. | Vascular characteristic determination with correspondence modeling of a vascular tree |
| US8977339B1 (en) * | 2013-12-05 | 2015-03-10 | Intrinsic Medical Imaging Llc | Method for assessing stenosis severity through stenosis mapping |
| CN103932694A (en) * | 2014-05-07 | 2014-07-23 | 霍云龙 | Method and device for accurately diagnosing FFR |
| US9754082B2 (en) * | 2014-05-30 | 2017-09-05 | Heartflow, Inc. | Systems and methods for reporting blood flow characteristics |
| CN106650267B (en) * | 2016-12-28 | 2020-03-17 | 北京昆仑医云科技有限公司 | System for calculating fractional flow reserve and method for setting boundary conditions |
-
2016
- 2016-12-28 CN CN201611234903.1A patent/CN106650267B/en active Active
- 2016-12-28 CN CN201711396887.0A patent/CN108109698B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103300820A (en) * | 2012-03-13 | 2013-09-18 | 西门子公司 | Method and system for non-invasive functional assessment of coronary artery stenosis |
| CN105764410A (en) * | 2013-10-22 | 2016-07-13 | 皇家飞利浦有限公司 | Fractional flow reserve (FFR) index with adaptive boundary condition parameters |
| CN105096388A (en) * | 2014-04-23 | 2015-11-25 | 北京冠生云医疗技术有限公司 | Computational Fluid Dynamics (CFD) based coronary artery blood flow simulating system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108109698A (en) | 2018-06-01 |
| CN106650267B (en) | 2020-03-17 |
| CN106650267A (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108109698B (en) | System for calculating fractional flow reserve and method for setting boundary conditions | |
| US20210358634A1 (en) | Systems and methods for image processing to determine blood flow | |
| JP7235710B2 (en) | Methods and systems for determining treatment by modifying patient-specific geometric models | |
| US20200202973A1 (en) | Method and system for facilitating physiological computations | |
| EP3127026B1 (en) | Systems and methods for determining blood flow characteristics using flow ratio | |
| CN102525443B (en) | For the method and apparatus to cardiovascular circulation modeling based on medical image | |
| CN106650272B (en) | The generation method and its application of the cardiovascular model of individual specificity | |
| JP2022500133A (en) | Calculation of boundary conditions for virtual FFR and iFR calculations based on myocardial staining characteristics | |
| JP2022517995A (en) | Patient-specific modeling of hemodynamic parameters in the coronary arteries | |
| CN114052764B (en) | Method, apparatus, system and computer storage medium for obtaining fractional flow reserve |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Block B, Mingyang International Center, 46 xizongbu Hutong, Dongcheng District, Beijing, 100005 Applicant after: Beijing Keya ark Medical Technology Co.,Ltd. Address before: Block B, Mingyang International Center, 46 xizongbu Hutong, Dongcheng District, Beijing, 100005 Applicant before: BEIJING CURACLOUD TECHNOLOGY Co.,Ltd. Address after: 3f301, East Tower, hadmen square, 8 Chongwenmenwai Street, Dongcheng District, Beijing 100062 Applicant after: Beijing Keya ark Medical Technology Co.,Ltd. Address before: Block B, Mingyang International Center, 46 xizongbu Hutong, Dongcheng District, Beijing, 100005 Applicant before: Beijing Keya ark Medical Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 3f301, East Tower, hadmen square, 8 Chongwenmenwai Street, Dongcheng District, Beijing 100062 Patentee after: Keya Medical Technology Co.,Ltd. Address before: 3f301, East Tower, hadmen square, 8 Chongwenmenwai Street, Dongcheng District, Beijing 100062 Patentee before: Beijing Keya ark Medical Technology Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |