WO2014163334A1 - Method for modeling and analyzing computational fluid dynamics on basis of material properties - Google Patents

Method for modeling and analyzing computational fluid dynamics on basis of material properties Download PDF

Info

Publication number
WO2014163334A1
WO2014163334A1 PCT/KR2014/002654 KR2014002654W WO2014163334A1 WO 2014163334 A1 WO2014163334 A1 WO 2014163334A1 KR 2014002654 W KR2014002654 W KR 2014002654W WO 2014163334 A1 WO2014163334 A1 WO 2014163334A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
medical image
region
mesh
interest
Prior art date
Application number
PCT/KR2014/002654
Other languages
French (fr)
Korean (ko)
Inventor
김영학
양동현
김남국
Original Assignee
재단법인 아산사회복지재단
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from KR1020130080429A external-priority patent/KR101530352B1/en
Application filed by 재단법인 아산사회복지재단 filed Critical 재단법인 아산사회복지재단
Publication of WO2014163334A1 publication Critical patent/WO2014163334A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
    • G06T2207/30104—Vascular flow; Blood flow; Perfusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2210/00—Indexing scheme for image generation or computer graphics
    • G06T2210/24—Fluid dynamics

Definitions

  • the present disclosure relates to a computational fluid dynamics modeling and analysis method based on material properties as a whole, and to a computational fluid dynamics modeling and analysis method based on material properties for analyzing perfusion based on material properties of blood vessel walls and plaques.
  • vascular stenosis due to plaque formed in blood vessels such as the carotid and coronary arteries is an important risk factor such as stroke and myocardial ischemia.
  • the severity of the stenosis determines the method of treatment, for example, intervention, stent placement, or drug treatment.
  • An indicator called Myocardial Fractional Flow Reserve (FFR) is used to assess the severity of stenosis or the likelihood that plaque will break from the vessel.
  • FFR refers to the ratio of blood pressure at a particular location in the coronary artery to blood pressure in the aorta in myocardial perfusion analysis.
  • a method is used in which a catheter is inserted into a vessel and moved to the FFR measurement location.
  • this method is invasive, making the patient uncomfortable and at risk of damaging the body. Therefore, recently, a method of diagnosing and evaluating lesions of blood vessels by a non-invasive method has attracted attention.
  • 3D modeling of cardiovascular vessels by means of computational fluid dynamics (CFD), and analysis of myocardial perfusion are used to evaluate FFR.
  • FFR_CT the FFR calculated by the CFD model
  • U.S. Patent No. 8,315,812 reflects input blood pressure, blood flow rate, mass of myocardium fed by blood vessels, and material properties of plaque.
  • a region of interest included in a medical image is modeled by a finite element method to determine the region of interest.
  • Generating a 3D model Mapping an intensity of a medical image corresponding to each finite element of the 3D model of the ROI to material properties of each finite element; And analyzing the flow associated with the 3D model of the region of interest by means of Computational Fluid Dynamics (CFD) means. do.
  • CFD Computational Fluid Dynamics
  • 1 is a view showing an example of a 3D cardiac image generated by the heart CT
  • FIG. 2 is a diagram illustrating an example of coronary artery and plaque divided based on a heart image
  • FIG. 3 shows an example of lumens of blood vessels in segmented coronary arteries
  • FIG. 4 is a view showing that the CT density is different depending on the type of plaque in the medical image separated lumen
  • FIG. 5 is a diagram illustrating an example of a blood vessel wall model and a plaque model modeled by a finite element method
  • FIG. 6 is a view showing the vessel wall and the plaque modeled by the tetrahedral volume mesh
  • FIG. 7 is a view for explaining a method of mapping intensity values of a heart image to each mesh constituting a blood vessel wall model and a plaque model;
  • FIG. 8 is a view for explaining an example of a method for setting boundary conditions of a computational fluid dynamics modeling and analysis method based on material properties according to the present disclosure
  • FIG. 9 is a diagram illustrating that a computer-related flow of a 3D model of a region of interest is computed by computational fluid dynamics means
  • FIG. 1 is a diagram illustrating an example of a 3D heart image generated by a heart CT.
  • a region of interest included in a medical image as shown in FIG. 1 is modeled by a finite element method and thus a region of interest.
  • a 3D model of is created.
  • the intensity of the medical image corresponding to each finite element of the 3D model of the ROI is mapped to the material properties of each finite element.
  • the flow associated with the 3D model of the region of interest is analyzed by means of Computational Fluid Dynamics (CFD).
  • CFD Computational Fluid Dynamics
  • a process of generating a 3D model of the ROI is described.
  • a cardiac image as shown in FIG. 1 is generated by 64 slice coronary CT angiography.
  • Cardiac images include coronary artery (region of interest). Coronary artery imaging alone does not provide information about the hemodynamic significance of coronary lesions (flags). Therefore, coronary arteries and plaques are modeled to enable the analysis of computational fluid dynamics.
  • FIG. 2 is a diagram illustrating an example of a coronary artery 10 and a plaque 11 segmented based on a heart image.
  • the heart image is a collection of voxels with gray scale.
  • the cardiac image is binarized to segment a region of interest or other region for computational purposes.
  • the cardiovascular can be segmented using an adaptive threshold method, and a coronary tree can be obtained.
  • FIG. 3 shows an example of lumens of blood vessels in segmented coronary arteries.
  • the walls of blood vessels are very thin, there is no noticeable difference between the surrounding tissues and the intensity, and due to the partial volume effect, they are hardly visible on the cardiac image.
  • the blood vessel wall is a boundary portion in contact with the lumen, and since the plaque also contacts the lumen, as shown in FIG. 3, the surface of the lumen needs to be clearly distinguished first.
  • FIG. 4 is a view showing that the CT density is different according to the type of plaque in the medical image divided lumen.
  • Atherosclerotic plaques can be classified into three types in CT. Pointed arrows indicate calcium, non-calcified, and mixed plaques, respectively. Small boxes each represent a cross section of the plaque (orthogonal plane to the vessel axis). Calcium-type plaques are hard, and non-calcium-type plaques are softer than calcium-type plaques, but are believed to be somewhat hard. The plaque is in contact with the lumen and has a different intensity than the vessel wall, for example a different value of Hounsfield unit (HU). For example, the intensities of calcium-type plaques, fibrous tissue plaques and lipid plaques are on the order of 657-416 HU, 88-18 HU and 25-19 HU, respectively.
  • HU Hounsfield unit
  • FIG. 5 is a diagram illustrating an example of a blood vessel wall model and a plaque model modeled by a finite element method.
  • the boundary that contacts the lumen is modeled as a blood vessel wall by a 3D triangular mesh. Plaques are also modeled by 3D triangular mesh.
  • FIG. 5 (a) shows the fluid mesh
  • FIG. 5 (b) shows the solid mesh
  • FIG. 5 (c) shows the vessel wall 20 and the plaque 21 modeled by the large coarse mesh
  • FIG. 5 (d) shows that the number of meshes of the plaque 21 is significantly increased by FIG. 5 (c) by the fine mesh.
  • FIG. 6 is a diagram showing blood vessel walls and plaques modeled by a tetrahedral volume mesh.
  • FIG. 6 shows modeled vessel wall 33, lumen 30, lipid plaque 31, fibrous plaque 32.
  • FIG. 6 (d) shows that the lumen 30 is surrounded by the vessel wall 33.
  • Various types of meshes can be mixed and used to model complex vascular walls and plaques.
  • the meshes that make up the plaques may be generated more finely.
  • the fluid domain is modeled by the fluid mesh, and hundreds of thousands of tetrahedral volume meshes can be used.
  • FIG. 7 is a view for explaining a method of mapping intensity values of a heart image to each mesh constituting the blood vessel wall model and the plaque model.
  • the vessel wall model and the plaque model were generated by modeling a region of interest.
  • the intensity of the heart image corresponding to each mesh of the vessel wall model and the plaque model is mapped to the material properties on each mesh.
  • the cardiac image is a collection of voxels, each voxel having an intensity, for example a CT value. As shown in the upper part of FIG. 7, the intensity of the voxel may be obtained as a pixel value of the heart image.
  • the pixels 40 corresponding to the plaques are displayed in a different color from the surroundings.
  • each mesh constituting the plaque model does not match one-to-one with the voxels of the heart image
  • the voxels closest to each mesh may be found.
  • the distance from the node of each mesh to the voxel can be obtained by computer calculation.
  • the CT values of the voxels closest to the nodes of each mesh constituting the plaque are mapped to the properties of the meshes.
  • the intensity of the voxels corresponding to the vessel wall can be obtained. Since the vessel wall is so thin, the CT values of the voxels closest to the nodes of the mesh at the interface with the lumens are mapped to the material properties of the mesh. For example, because the vessel wall is in contact with the lumen, the CT value of the voxel closest to the node outside of the lumen for the node that is in contact with or closest to the lumen is mapped to the mesh as a material property.
  • the material property may be density. Since the CT value reflects the density of the material, it may be used as it is or as a value indicating the density of the plaque through a separate conversion.
  • the vessel wall model and the plaque model are based on the coronary artery image included in the heart image, the shape is close to reality, and the intensity of the coronary image is mapped to the density of each mesh constituting the vessel wall model and the plaque model. .
  • blood vessels and plaques are not homogenized, but are modeled by reflecting density, which may be different depending on the position of one vessel or plaque.
  • density mapped to each mesh is based on the intensity of the cardiac image, which is very close to the physical reality of blood vessels and plaques.
  • FIG. 8 is a view for explaining an example of a method of setting the boundary conditions of the computational fluid dynamics modeling and analysis method based on the material properties according to the present disclosure.
  • the reliability of the results of analysis or interpretation by means of computational fluid dynamics for the flow or perfusion associated with the vessel wall model and the plaque model requires the accuracy of the boundary conditions as well as the modeling reflecting the material properties as described above. It is desirable to be a patient specipic boundary condition.
  • Boundary conditions include input conditions, output conditions, and so on.
  • the input condition is preferably patient-specific blood flow input (blood pressure, blood flow rate, etc.), and the output condition may be calculated in consideration of other conditions, for example, the mass of the myocardium of the patient, based on the input condition.
  • FIG. 8 shows the blood flow rate that changes with time at the input boundary measured by MRI venc.
  • Patient-specific blood flow input can be obtained based on patient-specific blood input measurements using clinical data, MRI venc and cardiac muscle segmentation, and Left Ventricle Volume.
  • venc velocity encoding
  • the myocardial splitting method can be used to measure the amount of myocardium that the coronary arteries feed and, as a result, obtain the output boundary condition of the CFD model.
  • boundary conditions of the CFD model can be obtained from clinical data.
  • boundary conditions such as sex, age, pulse rate, blood pressure, hematocrit values can be obtained from clinical data.
  • the CFD model based on the setting of patient-specific boundary conditions and the material properties improves the reliability of FFR_CT.
  • FIG. 9 is a diagram illustrating the computation of the flow associated with the 3D model of the ROI by computational fluid dynamics means.
  • FIG. 10 shows FFR_CT of coronary arteries obtained by CFD means.
  • vessel walls and plaques were modeled using meshes, with CT values mapped to material properties on each mesh.
  • patient-specific boundary conditions were set using MRI venc. The flow associated with the 3D model of the region of interest is then analyzed by CFD means.
  • the FFR (CT Fractional Flow Reserve) is computerized for blood flow before and after plaque at a specific location in the vessel wall model to obtain FFR_CT.
  • the stability of the plaque is evaluated based on FFR_CT.
  • the computational fluid dynamics model including the vascular wall model and the plaque model is trimmed and the boundary conditions are defined as described above.
  • FFR is Equation (1) Is defined as:
  • Pd distal coronary pressure
  • distal blood pressure from the center of the body Pa is distal coronary pressure
  • Pa is the central coronary pressure
  • arterial blood pressure may be used.
  • Pd is the pressure of the blood flow through the plaque.
  • CFD hemodynamic analysis
  • Equation (1) Even if only a few percent of Pd is changed, the FFR_CT can cross the boundary between normal and abnormal. It is therefore important that the model more accurately reflect the physical reality.
  • the blood vessel wall and the plaque may vary in density depending on the position of the mesh, not a homogeneous material. Morphological features of blood vessel walls and plaques are also modeled closer to reality based on heart images. In addition, the outer boundary condition was found to be patient-specific boundary condition, especially using MRI venc. Therefore, in the hemodynamic analysis, the analysis is closer to the physical reality of the patient.
  • mapping of material properties, eg, density, to each mesh constituting the vessel wall model and the plaque model affects the solution of several flow equations used in hemodynamic analysis. For example, there may be a difference in the solution of the flow equations when modeling blood vessel walls and plaques as homogeneous and when material properties are mapped to each mesh according to the present disclosure.
  • ⁇ f and ⁇ s are fluid density and solid density, respectively, p is fluid pressure and ⁇ is Newtonian fluid viscosity.
  • ⁇ is the Cauchy stress tensor
  • f B is the body force the solid experiences.
  • the equations include solid density, mesh motion and displacement. Therefore, according to the present disclosure, when density is mapped to the blood vessel wall and the plaque mesh, the stress-strain analysis may be more accurate than the case where the material properties of the blood vessel wall and the plaque are uniformly modeled. As a result, the FFR_CT becomes more accurate, and the stability of the plaque or the possibility of falling from the blood vessel, etc. can be more accurately evaluated.
  • Calcium-type plaques may have a very small or negligible strain due to stress.
  • fibrous and lipidic plaques must take into account the magnitude of strain due to blood pressure, in particular lipidic plaques are relatively softer. Therefore, simply considering the morphological features of the plaque or modeling it as a homogeneous material is less reliable in stress-strain analysis for the plaque.
  • the stress-strain CFD calculation results for the vessel wall and the plaque are very accurate.
  • the calculated blood pressure values before and after the plaque are also very close to the measured value, and the difference between the FFR_CT and the measured FFR is within the tolerance range.
  • the stress applied to the lipidic plaque is calculated by reflecting the morphological deformation of the lipidic plaque, so that more accurate information can be obtained in evaluating the possibility of dropping from the blood vessel of the lipidic plaque by FFR_CT.
  • Computational fluid dynamics modeling and analysis methods based on the material properties described in FIGS. 1-10 may be performed automatically by one or more software or in combination with a user interface.
  • (1) generating the 3D model of the region of interest includes a process of modeling the region of interest by a finite element mesh; wherein the intensity of the medical image is mapped to the material properties of each finite element. Mapping the intensity of a medical image corresponding to each mesh to material properties of each mesh; and analyzing a flow associated with a 3D model of a region of interest may include identifying a 3D model of the region of interest.
  • Computational fluid dynamics modeling and analysis method characterized in that it comprises a; computer-calculated FFR (Fractional Flow Reserve) at the location.
  • (2) generating the 3D model of the ROI may include generating a segmented medical image by segmenting the ROI or another region into a set of voxels based on the medical image; A region of interest based on the image is modeled by a finite element method to generate a 3D model of the region of interest, and the intensity of the medical image is mapped to the material properties of each finite element. Comprising the step of mapping the intensity of the voxel closest to each finite element to the material properties of each finite element; Computational fluid dynamics modeling and analysis method, characterized in that it comprises a.
  • (3) generating the 3D model of the ROI may include generating a segmented medical image by segmenting the ROI or another region into a set of voxels based on the medical image;
  • the region of interest is modeled by a finite element mesh based on the image.
  • the step of mapping the intensity of the medical image to the material properties of each finite element includes: Computational fluid dynamics modeling and analysis method, characterized in that it comprises a; process of mapping the intensity of the voxel closest to the node of the material to each mesh.
  • (4) generating the 3D model of the ROI may include generating a medical image including blood vessels; A process of segmenting lumens of blood vessels based on a medical image; A region of interest in contact with the lumen is modeled by a finite element mesh to generate a blood vessel wall model; Computational fluid dynamics modeling based on material properties, comprising: a region of interest in contact with the lumen and having an intensity different from that of the vessel wall is modeled by a finite element mesh to generate a lesion model. And analytical method.
  • the step of mapping the intensity of the medical image to the material properties of each finite element is such that the pixel value of the medical image corresponding to each mesh of the blood vessel wall model is a material property of each mesh of the blood vessel wall model.
  • (6) generating the 3D model of the ROI may include generating a medical image by contrast-enhanced heart CT; A process in which a cardiovascular is segmented into a set of voxels based on a medical image; A process of generating a blood vessel wall model by modeling a region of interest of the soft tissue in contact with the lumen of the cardiovascular vessel by a 3D triangular mesh; And a process of generating a plaque model by modeling a region of interest in contact with the cardiovascular lumen and having different intensities from the soft tissues by a 3D triangular mesh.
  • the step of mapping the intensity of the medical image to the material properties of each finite element is such that the CT density of the voxel closest to each 3D triangular mesh of the vessel wall model outside the lumen is determined by the CT density of the 3D triangular mesh.
  • Computational fluid dynamics modeling and analysis method based on the material properties characterized in that it comprises a; mapping to density (density).
  • the step of mapping the intensity of the medical image to the material properties of each finite element is such that the CT values of the voxels closest to each 3D triangular mesh of the plaque model outside the lumen are mapped to the density of the corresponding 3D triangular mesh.
  • Computational fluid dynamics modeling and analysis method based on the material characteristics, characterized in that it comprises a; process.
  • the analysis of the flow associated with the 3D model of the region of interest may include computerized calculation of the Fractional Flow Reserve (FFR) for perfusion before and after the passage of plaque in the cardiovascular system; And Computational fluid dynamics modeling and analysis method based on the material characteristics, characterized in that it comprises a; process based on the calculated FFR stability (stability) is evaluated.
  • FFR Fractional Flow Reserve
  • the setting of the condition may include taking a blood flow input of the cardiovascular vessel using an MRI Venc image; And setting a patient-specific input boundary condition in the CFD means based on the photographed blood flow input.
  • (11) establishing patient-specific boundary conditions prior to the step of analyzing the flow associated with the 3D model of the region of interest by means of Computational Fluid Dynamics (CFD) means;
  • the step of setting the condition is a process in which the amount of myocardium that is fed by the cardiovascular system is measured using a myocardial splitting method;
  • setting a patient-specific output boundary condition on the CFD means based on the measured amount of myocardium.
  • a computer-readable recording medium having recorded thereon a computer program for performing computational fluid dynamics modeling and analysis method based on material properties.
  • the reliability of the method of evaluating the severity of vascular lesions in a non-invasive manner is improved.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Computer Graphics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

Disclosed is a method for modeling and analyzing computational fluid dynamics (CFD) on the basis of material properties, comprising the steps for: generating a 3D model of a region of interest included in a medical image by modeling the region of interest by using a finite element method; mapping the intensity of the medical image, which corresponds to each finite element of the 3D model of the region of interest, to material properties of each finite element; and analyzing a flow related to the 3D model of the region of interest by a CFD means.

Description

물질특성에 기반한 전산유체역학 모델링 및 분석 방법Computational Fluid Dynamics Modeling and Analysis Method Based on Material Properties
본 개시(Disclosure)는 전체적으로 물질특성에 기반한 전산유체역학 모델링 및 분석 방법에 관한 것으로, 혈관벽 및 플라그의 물질특성에 기반하여 관류를 해석하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법에 관한 것이다.The present disclosure relates to a computational fluid dynamics modeling and analysis method based on material properties as a whole, and to a computational fluid dynamics modeling and analysis method based on material properties for analyzing perfusion based on material properties of blood vessel walls and plaques.
여기서는, 본 개시에 관한 배경기술이 제공되며, 이들이 반드시 공지기술을 의미하는 것은 아니다(This section provides background information related to the present disclosure which is not necessarily prior art).This section provides background information related to the present disclosure which is not necessarily prior art.
경동맥, 관상동맥 등 혈관에 형성된 플라그(plaque)에 의한 혈관 협착(stenosis)은 뇌졸중, 심근 허혈(myocardial ischemia) 등의 중요 위험 요소이다. 협착의 심각성에 따라 치료 방법, 예를 들어, 중재수술(intervention), 스탠트 삽입(stent placement) 또는 약물 치료등의 방법이 결정된다. 협착의 심각성 또는 플라그가 혈관으로부터 떨어질(rupture) 가능성을 평가하기 위해 심근분획혈류예비력(Fractional Flow Reserve; FFR)이라는 지표가 사용된다. 예를 들어, FFR은 심근관류 분석에서는 대동맥의 혈압에 대한 관상동맥의 특정 위치에서 혈압의 비율을 의미한다.Vascular stenosis due to plaque formed in blood vessels such as the carotid and coronary arteries is an important risk factor such as stroke and myocardial ischemia. The severity of the stenosis determines the method of treatment, for example, intervention, stent placement, or drug treatment. An indicator called Myocardial Fractional Flow Reserve (FFR) is used to assess the severity of stenosis or the likelihood that plaque will break from the vessel. For example, FFR refers to the ratio of blood pressure at a particular location in the coronary artery to blood pressure in the aorta in myocardial perfusion analysis.
카데터(catheter)가 혈관에 삽입되어 FFR 측정 위치까지 이동되는 방법이 사용된다. 그러나 이 방법은 침습적(invasive)이어서 환자가 불편하고, 신체에 손상을 줄 위험이 있다. 따라서 최근에는 비침습적(non-invasive) 방법으로 혈관의 병변(lesions)을 진단 및 평가하는 방법이 주목 받고 있다. 예를 들어, 전산유체역학(computational fluid dynamics; CFD)을 수단으로 심혈관을 3D 모델링하고, 심근관류(myocardial perfusion)를 해석해서 FFR을 평가하는 방법이 사용된다.A method is used in which a catheter is inserted into a vessel and moved to the FFR measurement location. However, this method is invasive, making the patient uncomfortable and at risk of damaging the body. Therefore, recently, a method of diagnosing and evaluating lesions of blood vessels by a non-invasive method has attracted attention. For example, 3D modeling of cardiovascular vessels by means of computational fluid dynamics (CFD), and analysis of myocardial perfusion are used to evaluate FFR.
FFR 값은 전술한 것과 같이 병변에 대한 대처 방법의 결정에 중요하므로 매우 신뢰성 있고 정확한 평가가 필요하다. 카데터를 이용한 침습적 방법은 FFR을 실측한 것이지만, CFD 모델에 의해 계산된 FFR(FFR_CT)은 모델에 의한 관류를 해석하고 평가한 것이다. 따라서 FFR_CT가 실측값에 근접하기 위해서는 모델링의 신뢰성이 중요하다.FFR values are important for the determination of how to cope with lesions, as described above, and therefore require very reliable and accurate evaluation. The invasive method using the catheter is the FFR measurement, but the FFR calculated by the CFD model (FFR_CT) is the analysis and evaluation of perfusion by the model. Therefore, the reliability of modeling is important for FFR_CT to be close to the measured value.
미국특허 US 8,315,812에는 심혈관 및 심근관류를 모델링 하는데 있어서, 입력 혈압, 혈류속도, 혈관에 의해 먹여 살려지는 심근의 질량, 플라그의 물질특성 등을 반영하고 있다.In modeling cardiovascular and myocardial perfusion, U.S. Patent No. 8,315,812 reflects input blood pressure, blood flow rate, mass of myocardium fed by blood vessels, and material properties of plaque.
그러나, 이 특허 문헌에서는 플라그를 CT 상에서 찾고, 플라그의 CT 이미지로부터 하드(hard) 플라그인지 소프트(soft) 플라그 인지 여부를 구분하고 있지만, 그 다음 과정이 플라그에 대한 CFD 모델링이 아니라, 플라그가 혈관으로부터 떨어질 위험성을 테스트하는 과정을 개시하고 있어서, 신뢰성 높은 FFR_CT를 구하는 데에는 해결책을 제시하지 못한다.However, while the patent literature finds plaques on the CT and distinguishes between hard plaques and soft plaques from the CT images of the plaques, the next step is not CFD modeling of the plaques, but the plaques are vascular The process of testing the risk of falling off does not provide a solution for obtaining a reliable FFR_CT.
이에 대하여 '발명의 실시를 위한 형태'의 후단에 기술한다.This will be described later in the section on Embodiments of the Invention.
여기서는, 본 개시의 전체적인 요약(Summary)이 제공되며, 이것이 본 개시의 외연을 제한하는 것으로 이해되어서는 아니된다(This section provides a general summary of the disclosure and is not a comprehensive disclosure of its full scope or all of its features).This section provides a general summary of the disclosure and is not a comprehensive disclosure of its full scope or all, provided that this is a summary of the disclosure. of its features).
본 개시에 따른 일 태양에 의하면(According to one aspect of the present disclosure), 의료영상(medical image)에 포함된 관심영역(region of interest)이 유한요소법(finite element method)에 의해 모델링되어 관심영역의 3D 모델이 생성되는 단계; 관심영역의 3D 모델의 각 유한요소(finite element)에 대응하는 의료영상의 인텐시티(intensity)가 각 유한요소의 물질특성(material properties)으로 매핑(mapping)되는 단계; 그리고 전산유체역학(Computational Fluid Dynamics; CFD) 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계;를 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법이 제공된다.According to one aspect of the present disclosure (According to one aspect of the present disclosure), a region of interest included in a medical image is modeled by a finite element method to determine the region of interest. Generating a 3D model; Mapping an intensity of a medical image corresponding to each finite element of the 3D model of the ROI to material properties of each finite element; And analyzing the flow associated with the 3D model of the region of interest by means of Computational Fluid Dynamics (CFD) means. do.
이에 대하여 '발명의 실시를 위한 형태'의 후단에 기술한다.This will be described later in the section on Embodiments of the Invention.
도 1은 심장 CT에 의해 생성된 3D 심장 영상의 일 예를 보여주는 도면,1 is a view showing an example of a 3D cardiac image generated by the heart CT,
도 2는 심장 영상을 기초로 분할된 관상동맥 및 플라그의 일 예를 나타내는 도면,2 is a diagram illustrating an example of coronary artery and plaque divided based on a heart image;
도 3은 분할된 관상동맥 중 혈관의 루멘의 일 예를 보여주는 도면,3 shows an example of lumens of blood vessels in segmented coronary arteries,
도 4는 루멘이 구분된 의료영상에서 플라그 종류에 따라 CT density가 다른 것을 나타내는 도면,4 is a view showing that the CT density is different depending on the type of plaque in the medical image separated lumen,
도 5는 유한요소법(finite element method)에 의해 모델링된 혈관벽 모델 및 플라그 모델의 일 예를 나타내는 도면,5 is a diagram illustrating an example of a blood vessel wall model and a plaque model modeled by a finite element method;
도 6은 tetrahedral volume mesh에 의해 모델링된 혈관벽 및 플라그를 나타내는 도면,6 is a view showing the vessel wall and the plaque modeled by the tetrahedral volume mesh,
도 7은 혈관벽 모델 및 플라그 모델을 구성하는 각 메시에 심장 영상의 인텐시티 값을 매핑하는 방법을 설명하는 도면,FIG. 7 is a view for explaining a method of mapping intensity values of a heart image to each mesh constituting a blood vessel wall model and a plaque model; FIG.
도 8은 본 개시에 따른 물질특성에 기반한 전산유체역학 모델링 및 분석 방법의 경계조건을 설정하는 방법의 일 예를 설명하는 도면,8 is a view for explaining an example of a method for setting boundary conditions of a computational fluid dynamics modeling and analysis method based on material properties according to the present disclosure;
도 9는 전산유체역학 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 컴퓨터 계산되는 것을 설명하는 도면,9 is a diagram illustrating that a computer-related flow of a 3D model of a region of interest is computed by computational fluid dynamics means;
도 10은 CFD 수단에 의해 구해진 관상동맥의 FFR_CT를 보여주는 도면.10 shows FFR_CT of coronary arteries obtained by CFD means.
이하, 본 개시를 첨부된 도면을 참고로 하여 자세하게 설명한다(The present disclosure will now be described in detail with reference to the accompanying drawing(s)).The present disclosure will now be described in detail with reference to the accompanying drawing (s).
도 1은 심장 CT에 의해 생성된 3D 심장 영상의 일 예를 보여주는 도면이다.1 is a diagram illustrating an example of a 3D heart image generated by a heart CT.
물질특성에 기반한 전산유체역학 모델링 및 분석 방법에서, 도 1에 도시된 것과 같은 의료영상(medical image)에 포함된 관심영역(region of interest)이 유한요소법(finite element method)에 의해 모델링되어 관심영역의 3D 모델이 생성된다. 이후, 관심영역의 3D 모델의 각 유한요소(finite element)에 대응하는 의료영상의 인텐시티(intensity)가 각 유한요소의 물질특성(material properties)으로 매핑된다. 다음으로, 전산유체역학(Computational Fluid Dynamics; CFD) 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석된다. In the computational fluid dynamics modeling and analysis method based on material properties, a region of interest included in a medical image as shown in FIG. 1 is modeled by a finite element method and thus a region of interest. A 3D model of is created. Thereafter, the intensity of the medical image corresponding to each finite element of the 3D model of the ROI is mapped to the material properties of each finite element. Next, the flow associated with the 3D model of the region of interest is analyzed by means of Computational Fluid Dynamics (CFD).
먼저, 관심영역의 3D 모델이 생성되는 과정이 설명된다. 예를 들어, 64 slice coronary CT angiography에 의해 도 1에 도시된 것과 같은 심장 영상이 생성된다. 심장 영상에는 관상동맥 영상(관심영역)이 포함된다. 관상동맥 영상 자체로는 관상동맥 병변(플라그)의 혈류역학적(hemodynamic) 중요성에 대한 정보를 주지 못한다. 따라서 전산유체역학의 분석수단을 사용할 수 있도록 관상동맥 및 플라그가 모델링된다. First, a process of generating a 3D model of the ROI is described. For example, a cardiac image as shown in FIG. 1 is generated by 64 slice coronary CT angiography. Cardiac images include coronary artery (region of interest). Coronary artery imaging alone does not provide information about the hemodynamic significance of coronary lesions (flags). Therefore, coronary arteries and plaques are modeled to enable the analysis of computational fluid dynamics.
도 2는 심장 영상을 기초로 분할된 관상동맥(10) 및 플라그(11)의 일 예를 나타내는 도면이다.2 is a diagram illustrating an example of a coronary artery 10 and a plaque 11 segmented based on a heart image.
심장 영상은 그레이 스케일(gray scale)을 가지는 복셀들의 집합이다. 컴퓨터 계산을 하기 위해 심장 영상을 이진화(image binarization)하여 관심영역 또는 다른 영역을 분할(segmentation)한다. 예를 들어, 어뎁티브 쓰레쉬홀드(adaptive threshold) 방법을 사용하여 심혈관을 분할하고, 관상동맥 트리(tree)가 구해질 수 있다.The heart image is a collection of voxels with gray scale. The cardiac image is binarized to segment a region of interest or other region for computational purposes. For example, the cardiovascular can be segmented using an adaptive threshold method, and a coronary tree can be obtained.
도 3은 분할된 관상동맥 중 혈관의 루멘의 일 예를 보여주는 도면이다.3 shows an example of lumens of blood vessels in segmented coronary arteries.
혈관벽은 매우 얇고, 주변의 조직과 인텐시티 차이가 뚜렷하지 않으며, 부분볼륨효과(partial volume effect)로 인해서 심장 영상에서는 잘 보이지 않는다. 다만 혈관벽은 루멘(lumen; 내강)과 접하는 경계부이고, 플라그도 루멘과 접하기 때문에 도 3에 도시된 것과 같이, 루멘의 표면을 먼저 명확히 구분할 필요가 있다.The walls of blood vessels are very thin, there is no noticeable difference between the surrounding tissues and the intensity, and due to the partial volume effect, they are hardly visible on the cardiac image. However, the blood vessel wall is a boundary portion in contact with the lumen, and since the plaque also contacts the lumen, as shown in FIG. 3, the surface of the lumen needs to be clearly distinguished first.
도 4는 루멘이 구분된 의료영상에서 플라그 종류에 따라 CT density가 다른 것을 나타내는 도면이다.4 is a view showing that the CT density is different according to the type of plaque in the medical image divided lumen.
동맥경화반의 특성은 CT에서 아래 세 가지로 구분될 수 있다. 화살표가 가르키는 것이 칼슘형(calcified), 비칼슘형(non-calcified), 혼합형(mixed) 플라그를 각각 나타낸다. 작은 박스(box)들은 플라그의 단면(혈관 축(axis)에 수직면(orthogonal plane))을 각각 나타낸다. 칼슘형 플라그는 단단하고, 비칼슘형 플라그는 칼슘형 플라그보다는 소프트하지만, 어느 정도 단단할 것으로 추정된다. 플라그는 루멘과 접하며, 혈관벽과는 다른 인텐시티, 예를 들어, 다른 값의 HU(Hounsfield unit)를 가진다. 예를 들어, 칼슘형 플라그, 섬유 조직(fibrous tissue) 플라그 및 지질(lipid) 플라그의 인텐시티는 각각 657-416 HU, 88-18 HU 및 25-19 HU 정도이다.Atherosclerotic plaques can be classified into three types in CT. Pointed arrows indicate calcium, non-calcified, and mixed plaques, respectively. Small boxes each represent a cross section of the plaque (orthogonal plane to the vessel axis). Calcium-type plaques are hard, and non-calcium-type plaques are softer than calcium-type plaques, but are believed to be somewhat hard. The plaque is in contact with the lumen and has a different intensity than the vessel wall, for example a different value of Hounsfield unit (HU). For example, the intensities of calcium-type plaques, fibrous tissue plaques and lipid plaques are on the order of 657-416 HU, 88-18 HU and 25-19 HU, respectively.
도 5는 유한요소법(finite element method)에 의해 모델링된 혈관벽 모델 및 플라그 모델의 일 예를 나타내는 도면이다.FIG. 5 is a diagram illustrating an example of a blood vessel wall model and a plaque model modeled by a finite element method.
예를 들어, 분할된 심장 영상에서 루멘과 접하는 경계부가 3D triangular mesh에 의해 혈관벽으로 모델링된다. 또한, 플라그가 3D triangular mesh에 의해 모델링된다. For example, in the segmented cardiac image, the boundary that contacts the lumen is modeled as a blood vessel wall by a 3D triangular mesh. Plaques are also modeled by 3D triangular mesh.
도 5(a)은 fluid mesh을 나타내고, 도 5(b)는 solid mesh를 나타내며, 도 5(c)는 크고 거친(coarse) mesh에 의해 모델링된 혈관벽(20) 및 플라그(21)를 나타내고, 도 5(d)는 fine mesh에 의해 플라그(21)의 메시 개수가 도 5(c)보다 현저히 증가한 것을 나타낸다.FIG. 5 (a) shows the fluid mesh, FIG. 5 (b) shows the solid mesh, and FIG. 5 (c) shows the vessel wall 20 and the plaque 21 modeled by the large coarse mesh, FIG. 5 (d) shows that the number of meshes of the plaque 21 is significantly increased by FIG. 5 (c) by the fine mesh.
도 6는 tetrahedral volume mesh에 의해 모델링된 혈관벽 및 플라그를 나타내는 도면이다.6 is a diagram showing blood vessel walls and plaques modeled by a tetrahedral volume mesh.
적절한 표면 메시(surface mesh)를 사용하여 혈관벽 및 플라그를 구성하는 4-node tetrahedral volume mesh가 생성될 수 있다. 도 6은 모델링된 혈관벽(33), 루멘(30), 지질 플라그(31), 섬유성 플라그(32)를 도시하고 있다. 예를 들어 도 6(d)는 루멘(30)이 혈관벽(33)에 의해 둘러싸여 있는 것을 도시하고 있다.Using a suitable surface mesh, a 4-node tetrahedral volume mesh can be created that constitutes the vessel wall and plaque. FIG. 6 shows modeled vessel wall 33, lumen 30, lipid plaque 31, fibrous plaque 32. For example, FIG. 6 (d) shows that the lumen 30 is surrounded by the vessel wall 33.
복잡한 형상의 혈관벽 및 플라그를 모델링하기 위해 다양한 타입의 메시를 혼합(mixed finite element)하여 사용할 수 있다. 혈류역학에서 플라그가 받는 스트레스 필드(stress field)를 더 정확히 분석하기 위해 플라그를 구성하는 메시가 더 세밀하게(fine) 생성될 수도 있다. fluid mesh에 의해 fluid domain이 모델링되며, 수십만 개의 tetrahedral volume mesh가 사용될 수 있다.Various types of meshes can be mixed and used to model complex vascular walls and plaques. In order to more accurately analyze the stress field that plaques receive in hemodynamics, the meshes that make up the plaques may be generated more finely. The fluid domain is modeled by the fluid mesh, and hundreds of thousands of tetrahedral volume meshes can be used.
도 7은 혈관벽 모델 및 플라그 모델을 구성하는 각 메시에 심장 영상의 인텐시티 값을 매핑하는 방법을 설명하는 도면이다.FIG. 7 is a view for explaining a method of mapping intensity values of a heart image to each mesh constituting the blood vessel wall model and the plaque model.
도 1 내지 도 6에서 설명된 것과 같이, 관심영역을 모델링하여 혈관벽 모델 및 플라그 모델이 생성되었다.As illustrated in FIGS. 1 to 6, the vessel wall model and the plaque model were generated by modeling a region of interest.
이후, 혈관벽 모델 및 플라그 모델의 각 메시에 대응하는 심장 영상의 인텐시티(intensity)가 각 메시에 물질특성(material properties)으로 매핑된다.Then, the intensity of the heart image corresponding to each mesh of the vessel wall model and the plaque model is mapped to the material properties on each mesh.
심장 영상은 복셀의 집합이며, 각 복셀은 인텐시티, 예를 들어, CT 값(CT density)을 가진다. 도 7의 상단에 도시된 것과 같이 복셀의 인텐시티는 심장 영상의 픽셀(pixel) 값으로 구해질 수 있다. 플라그에 대응하는 픽셀(40)들이 주변과 다른 색으로 표시되어 있다.The cardiac image is a collection of voxels, each voxel having an intensity, for example a CT value. As shown in the upper part of FIG. 7, the intensity of the voxel may be obtained as a pixel value of the heart image. The pixels 40 corresponding to the plaques are displayed in a different color from the surroundings.
플라그 모델을 구성하는 각 메시는 심장 영상의 복셀들과 일대일로 매칭되지는 않을 지라도 각 메시에 가장 근접한 복셀이 찾아질 수 있다. 예를 들어, 각 메시의 노드(node)로부터 복셀까지의 거리가 컴퓨터 계산에 의해 구해질 수 있다. 플라그를 구성하는 각 메시의 노드에 가장 근접한 복셀의 CT 값이 해당 메시에 물질특성으로 매핑된다.Although each mesh constituting the plaque model does not match one-to-one with the voxels of the heart image, the voxels closest to each mesh may be found. For example, the distance from the node of each mesh to the voxel can be obtained by computer calculation. The CT values of the voxels closest to the nodes of each mesh constituting the plaque are mapped to the properties of the meshes.
이와 마찬가지로, 혈관벽에 대응하는 복셀의 인텐시티가 구해질 수 있다. 혈관벽은 매우 얇기 때문에 루멘과 접하는 경계부에서 메시의 노드에 가장 근접한 복셀의 CT 값이 해당 메시에 물질특성으로 매핑된다. 예를 들어, 혈관벽은 루멘과 접하므로, 루멘에 접하는 또는 루멘에 가장 근접한 노드에 대해 루멘의 바깥에서 노드에 가장 근접한 복셀의 CT 값이 해당 메시에 물질특성으로 매핑된다.Similarly, the intensity of the voxels corresponding to the vessel wall can be obtained. Since the vessel wall is so thin, the CT values of the voxels closest to the nodes of the mesh at the interface with the lumens are mapped to the material properties of the mesh. For example, because the vessel wall is in contact with the lumen, the CT value of the voxel closest to the node outside of the lumen for the node that is in contact with or closest to the lumen is mapped to the mesh as a material property.
여기서 물질특성은 밀도(density)가 될 수 있다. CT 값은 물질의 밀도를 반영하는 값이므로 그대로 또는 별도의 변환을 통해 플라그의 밀도를 나타내는 값으로 사용될 수 있다.In this case, the material property may be density. Since the CT value reflects the density of the material, it may be used as it is or as a value indicating the density of the plaque through a separate conversion.
이와 같이, 혈관벽 모델 및 플라그 모델은 심장 영상에 포함된 관상동맥 영상을 기초로 하므로 그 형상이 실재에 가깝고, 관상동맥 영상의 인텐시티가 혈관벽 모델 및 플라그 모델을 구성하는 각 메시에 밀도로 매핑되어 있다. As described above, since the vessel wall model and the plaque model are based on the coronary artery image included in the heart image, the shape is close to reality, and the intensity of the coronary image is mapped to the density of each mesh constituting the vessel wall model and the plaque model. .
따라서, 본 개시에 따른 물질특성에 기반한 전산유체역학 모델링 및 분석 방법에서는 혈관 및 플라그가 균질화된 것이 아니라, 하나의 혈관 또는 플라그라 할지라도 위치에 따라 다를 수 있는 밀도를 반영하여 모델링된다. 즉, 각 메시에 매핑된 밀도는 심장 영상의 인텐시티를 기초로 하므로, 혈관 및 플라그의 물리적 실재에 매우 가깝다. Therefore, in the computational fluid dynamics modeling and analysis method based on the material properties according to the present disclosure, blood vessels and plaques are not homogenized, but are modeled by reflecting density, which may be different depending on the position of one vessel or plaque. In other words, the density mapped to each mesh is based on the intensity of the cardiac image, which is very close to the physical reality of blood vessels and plaques.
도 8은 본 개시에 따른 물질특성에 기반한 전산유체역학 모델링 및 분석 방법의 경계조건을 설정하는 방법의 일 예를 설명하는 도면이다. 8 is a view for explaining an example of a method of setting the boundary conditions of the computational fluid dynamics modeling and analysis method based on the material properties according to the present disclosure.
혈관벽 모델 및 플라그 모델에 관련된 흐름 또는 관류에 대해 전산유체역학 수단으로 분석 또는 해석한 결과의 신뢰성은 전술된 것과 같이 물질특성을 반영하는 모델링뿐만 아니라 경계조건의 정확성이 필요하며, 경계조건의 정확성을 위해서는 환자맞춤(patient specipic) 경계조건이 되는 것이 바람직하다.The reliability of the results of analysis or interpretation by means of computational fluid dynamics for the flow or perfusion associated with the vessel wall model and the plaque model requires the accuracy of the boundary conditions as well as the modeling reflecting the material properties as described above. It is desirable to be a patient specipic boundary condition.
경계조건(boundary condition)은 입력조건, 출력조건 등을 포함한다. 입력조건은 환자맞춤 혈류입력(혈압, 혈류속도 등)이 바람직하고, 출력조건은 입력조건을 기초로 다른 조건, 예를 들어, 환자의 심근의 질량 등을 고려하여 계산될 수 있다.Boundary conditions include input conditions, output conditions, and so on. The input condition is preferably patient-specific blood flow input (blood pressure, blood flow rate, etc.), and the output condition may be calculated in consideration of other conditions, for example, the mass of the myocardium of the patient, based on the input condition.
예를 들어, 도 8에는 MRI venc로 측정된 입력경계에서 시간에 따라 변화하는 혈류속도가 나타나 있다. 환자맞춤 혈류입력은 임상데이터, MRI venc 및 심장근육 분할, Left Ventricle Volume 등을 이용한 환자맞춤 혈류입력(Input) 측정을 기반으로 구해질 수 있다.For example, FIG. 8 shows the blood flow rate that changes with time at the input boundary measured by MRI venc. Patient-specific blood flow input can be obtained based on patient-specific blood input measurements using clinical data, MRI venc and cardiac muscle segmentation, and Left Ventricle Volume.
MRI를 이용하면 몸속의 혈류속도를 in-vivo로 촬영할 수 있는 venc(velocity encoding) 영상을 사용할 수 있다. Venc 영상을 이용하여 관상동맥의 혈류입력을 촬영하고, 이러한 환자맞춤 혈류입력이 CFD 모델의 경계조건으로 입력된다. 또한, 심근분할 방법을 사용하여 관상동맥이 먹여 살리는 심근의 양을 측정할 수 있고, 그 결과 CFD 모델의 출력(output) 경계조건을 구할 수 있다.Using MRI, you can use venc (velocity encoding) images to capture blood flow in your body in-vivo. Venc images are taken of coronary blood flow inputs, and these customized blood flow inputs are input as boundary conditions of the CFD model. In addition, the myocardial splitting method can be used to measure the amount of myocardium that the coronary arteries feed and, as a result, obtain the output boundary condition of the CFD model.
또한, 임상데이터로부터 CFD 모델의 다른 경계조건을 구할 수 있다. 예를 들어, 임상데이터로부터 성별, 나이, 맥박수, 혈압, 적혈구용적률(hematocrit) 수치 등의 경계조건을 구할 수 있다.In addition, other boundary conditions of the CFD model can be obtained from clinical data. For example, boundary conditions such as sex, age, pulse rate, blood pressure, hematocrit values can be obtained from clinical data.
이러한 환자맞춤 경계조건의 설정과 물질특성에 기반한 CFD 모델은 FFR_CT의 신뢰성을 향상시킨다.The CFD model based on the setting of patient-specific boundary conditions and the material properties improves the reliability of FFR_CT.
도 9는 전산유체역학 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 컴퓨터 계산되는 것을 설명하는 도면이다. 도 10은 CFD 수단에 의해 구해진 관상동맥의 FFR_CT를 보여주는 도면이다.FIG. 9 is a diagram illustrating the computation of the flow associated with the 3D model of the ROI by computational fluid dynamics means. FIG. 10 shows FFR_CT of coronary arteries obtained by CFD means. FIG.
전술된 것과 같이, 혈관벽 및 플라그가 메시를 사용하여 모델링되며, 각 메시에 CT 값이 물질특성으로 매핑되었다. 또한, MRI venc 등을 이용하여 환자맞춤 경계조건을 설정하였다. 이후, CFD 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석된다.As described above, vessel walls and plaques were modeled using meshes, with CT values mapped to material properties on each mesh. In addition, patient-specific boundary conditions were set using MRI venc. The flow associated with the 3D model of the region of interest is then analyzed by CFD means.
예를 들어, 혈관벽 모델의 특정 위치에서 플라그 전후의 혈류에 대해 FFR(Fractional Flow Reserve)이 컴퓨터 계산되어 FFR_CT가 획득된다. FFR_CT를 기초로 플라그의 안정성(stability)이 평가된다.For example, the FFR (CT Fractional Flow Reserve) is computerized for blood flow before and after plaque at a specific location in the vessel wall model to obtain FFR_CT. The stability of the plaque is evaluated based on FFR_CT.
CFD 수단에 의한 분석을 위해, 먼저, 혈관벽 모델 및 플라그 모델을 포함하는 전산유체역학 모델이 다듬어지고(trimed), 전술된 것과 같이 경계조건이 정의된다.For analysis by CFD means, first, the computational fluid dynamics model including the vascular wall model and the plaque model is trimmed and the boundary conditions are defined as described above.
이후, 예를 들어, 도 9에 도시된 것과 같이, 혈류, 혈압, 혈류속도 등이 최대충혈(maximum hyperemia) 상태에서 컴퓨터 계산된다. 그 결과, 도 10에 도시된 것과 같이, FFR_CT 값이 모델의 특정 위치마다 구해진다. Then, for example, as shown in FIG. 9, blood flow, blood pressure, blood flow rate, and the like are computed in the state of maximum hyperemia. As a result, as shown in FIG. 10, an FFR_CT value is obtained for each specific position of the model.
FFR은 수학식(1)
Figure PCTKR2014002654-appb-I000001
와 같이 정의된다.FFR is Equation (1)
Figure PCTKR2014002654-appb-I000001
Is defined as:
여기서, Pd는 원위 관상동맥 혈압(Distal coronary Pressure)으로서 몸 중심으로부터 말단측 혈압이고, Pa는 근위 관상동맥 압력(Proximal coronary Pressure)으로서 몸 중심측 혈압이며, 동맥혈압이 사용될 수 있다.Here, Pd is distal coronary pressure, distal blood pressure from the center of the body, Pa is distal coronary pressure, Pa is the central coronary pressure, and arterial blood pressure may be used.
Pd는 플라그를 통과한 혈류의 압력이다. 주어진 관심영역의 모델이 CFD 수단에 의해 혈류역학적 해석이 이루어져서 Pd가 구해진다. 따라서 Pd 값이 실측값에 근접할 때, 실측된 FFR과 계산된 FFR(FFR_CT)의 차이가 허용오차 범위 내에 들어올 수 있다.Pd is the pressure of the blood flow through the plaque. Given the model of the region of interest, hemodynamic analysis is performed by CFD means to obtain Pd. Therefore, when the Pd value approaches the measured value, the difference between the measured FFR and the calculated FFR (FFR_CT) may fall within the tolerance range.
상기 수학식(1)에서 알 수 있듯이 Pd가 몇 퍼센트만 달라져도 FFR_CT가 정상과 비정상의 경계를 넘어갈 수 있다. 따라서 모델이 물리적 실재를 더욱 정확하게 반영하는 것이 중요하다.As can be seen from Equation (1), even if only a few percent of Pd is changed, the FFR_CT can cross the boundary between normal and abnormal. It is therefore important that the model more accurately reflect the physical reality.
본 개시에서는 전술된 것과 같이 혈관벽 및 플라그가 균질의 물질이 아니라 메시의 위치에 따라 밀도가 달라질 수 있는 것으로 모델링되어 있다. 또한 혈관벽 및 플라그의 형태적 특징도 심장 영상에 기초하여 실재에 가깝게 모델링되어 있다. 또한, 외부 경계조건은 특히 MRI venc를 이용하여 환자맞춤 경계조건이 찾아졌다. 따라서 혈류역학적 해석에서 더욱 환자의 물리적 실재에 근접한 해석이 가능하다.In the present disclosure, as described above, it is modeled that the blood vessel wall and the plaque may vary in density depending on the position of the mesh, not a homogeneous material. Morphological features of blood vessel walls and plaques are also modeled closer to reality based on heart images. In addition, the outer boundary condition was found to be patient-specific boundary condition, especially using MRI venc. Therefore, in the hemodynamic analysis, the analysis is closer to the physical reality of the patient.
혈관벽 모델 및 플라그 모델을 구성하는 각 메시에 물질특성, 예를 들어, 밀도가 매핑된 점은 혈류역학 해석에 사용되는 여러 유동 방정식들의 해(solution)에 영향을 미친다. 예를 들어, 혈관벽 및 플라그를 균질한 것으로 모델링하는 경우와 본 개시에 따라 각 메시에 물질특성을 매핑한 경우 상기 유동 방정식들의 해에 차이가 있을 수 있다.The mapping of material properties, eg, density, to each mesh constituting the vessel wall model and the plaque model affects the solution of several flow equations used in hemodynamic analysis. For example, there may be a difference in the solution of the flow equations when modeling blood vessel walls and plaques as homogeneous and when material properties are mapped to each mesh according to the present disclosure.
예를 들어, CFD가 계산하는 유동방정식들은 아래의 질량보존방정식(Mass Conservationequation)과For example, the flow equations that CFD calculates are the following mass conservation equations:
Figure PCTKR2014002654-appb-I000002
Figure PCTKR2014002654-appb-I000002
아래의 모멘트평형방정식(Momentum Balance equation)에 기초하여 유도된다. It is derived based on the Momentum Balance equation below.
Figure PCTKR2014002654-appb-I000003
Figure PCTKR2014002654-appb-I000003
incompressible Newtonian fluid and solid domains에 관련된 운동방정식은 아래와 같이 표시될 수 있다.The equation of motion associated with incompressible Newtonian fluid and solid domains can be expressed as
Figure PCTKR2014002654-appb-I000004
Figure PCTKR2014002654-appb-I000004
Figure PCTKR2014002654-appb-I000005
Figure PCTKR2014002654-appb-I000005
여기서, ρf및 ρs는 각각 fluid density 및 solid density이고, p는 fluid pressure, μ는 Newtonian fluid viscosity이다. Where ρ f and ρ s are fluid density and solid density, respectively, p is fluid pressure and μ is Newtonian fluid viscosity.
Figure PCTKR2014002654-appb-I000006
,
Figure PCTKR2014002654-appb-I000007
,
Figure PCTKR2014002654-appb-I000008
는 각각 각각 fluid velocity, mesh velocity 및 solid displacement vector를 나타낸다. τ는 Cauchy stress tensor, 그리고f B는 body force the solid experiences를 나타낸다.
Figure PCTKR2014002654-appb-I000006
,
Figure PCTKR2014002654-appb-I000007
,
Figure PCTKR2014002654-appb-I000008
Represent the fluid velocity, mesh velocity and solid displacement vector, respectively. τ is the Cauchy stress tensor, and f B is the body force the solid experiences.
상기 유동방정식에 나타난 것과 같이, solid의 밀도, 메시의 운동 및 변위(displacement) 등이 방정식에 포함됨을 알 수 있다. 따라서 본 개시에 따라 혈관벽 및 플라그의 메시에 각각 밀도가 매핑되어 있는 경우 혈관벽 및 플라그의 물질특성을 균일하게 모델링하는 경우에 비하여 스트레스-스트레인(stress-strain) 해석이 더욱 정확하게 될 수 있다. 그 결과, FFR_CT가 더 정확해지고, 플라그의 안정성(stability) 또는 혈관으로부터 떨어질 가능성 등이 보다 정확하게 평가될 수 있다.As shown in the flow equation, it can be seen that the equations include solid density, mesh motion and displacement. Therefore, according to the present disclosure, when density is mapped to the blood vessel wall and the plaque mesh, the stress-strain analysis may be more accurate than the case where the material properties of the blood vessel wall and the plaque are uniformly modeled. As a result, the FFR_CT becomes more accurate, and the stability of the plaque or the possibility of falling from the blood vessel, etc. can be more accurately evaluated.
칼슘형 플라그는 스트레스에 의한 스트레인의 크기가 매우 작거나 무시할 수 있는 경우가 있다. 그러나, 섬유성 플라그 및 지질성 플라그는 혈류압력에 의한 변형(strain)의 크기를 고려해야 하고, 특히 지질성 플라그는 상대적으로 더 소프트하다. 따라서, 단순히 플라그의 형태적 특징만을 고려하거나 또는 균질한 물질로 모델링하는 것으로는 플라그에 대한 스트레스-스트레인 해석의 신뢰성이 떨어진다.Calcium-type plaques may have a very small or negligible strain due to stress. However, fibrous and lipidic plaques must take into account the magnitude of strain due to blood pressure, in particular lipidic plaques are relatively softer. Therefore, simply considering the morphological features of the plaque or modeling it as a homogeneous material is less reliable in stress-strain analysis for the plaque.
본 개시에서는 플라그의 밀도가 각 메시마다 매핑되어 있으므로, 혈관벽 및 플라그에 대한 스트레스-스트레인 CFD 계산 결과가 매우 정확하다. 그 결과 플라그 전후의 혈압의 계산값도 실측값에 매우 근접하게 되고, FFR_CT와 실측된 FFR의 차이가 허용오차 범위 내로 있게 된다. 특히 지질성 플라그에 가해지는 스트레스는 지질성 플라그의 형태적인 변형을 반영하여 계산된 것이어서 FFR_CT에 의한 지질성 플라그의 혈관으로부터 떨어질 가능성을 평가하는 데 더 정확한 정보를 얻을 수 있다.In the present disclosure, since the density of the plaque is mapped for each mesh, the stress-strain CFD calculation results for the vessel wall and the plaque are very accurate. As a result, the calculated blood pressure values before and after the plaque are also very close to the measured value, and the difference between the FFR_CT and the measured FFR is within the tolerance range. In particular, the stress applied to the lipidic plaque is calculated by reflecting the morphological deformation of the lipidic plaque, so that more accurate information can be obtained in evaluating the possibility of dropping from the blood vessel of the lipidic plaque by FFR_CT.
도 1 내지 도 10에서 설명된 물질특성에 기반한 전산유체역학 모델링 및 분석 방법은 하나 이상의 소프트웨어에 의해 자동으로 또는 사용자 인터페이스와 결합하여 수행될 수 있다.Computational fluid dynamics modeling and analysis methods based on the material properties described in FIGS. 1-10 may be performed automatically by one or more software or in combination with a user interface.
이하 본 개시의 다양한 실시 형태에 대하여 설명한다.Hereinafter, various embodiments of the present disclosure will be described.
(1) 관심영역의 3D 모델이 생성되는 단계는 유한요소망(finite element mesh)에 의해 관심영역이 모델링되는 과정;을 포함하며, 의료영상의 인텐시티가 각 유한요소의 물질특성으로 매핑되는 단계는 각 메시(mesh)에 대응하는 의료영상의 인텐시티가 각 메시에 물질특성으로 매핑되는 과정;을 포함하며, 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계는 관심영역의 3D 모델의 특정 위치에서 FFR(Fractional Flow Reserve)이 컴퓨터 계산되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(1) generating the 3D model of the region of interest includes a process of modeling the region of interest by a finite element mesh; wherein the intensity of the medical image is mapped to the material properties of each finite element. Mapping the intensity of a medical image corresponding to each mesh to material properties of each mesh; and analyzing a flow associated with a 3D model of a region of interest may include identifying a 3D model of the region of interest. Computational fluid dynamics modeling and analysis method, characterized in that it comprises a; computer-calculated FFR (Fractional Flow Reserve) at the location.
(2) 관심영역의 3D 모델이 생성되는 단계는 의료영상을 기초로 관심영역 또는 다른 영역이 복셀(Boxel)의 집합으로 분할(segmentation)되어 분할된 의료영상이 생성되는 과정;그리고, 분할된 의료영상을 기초로 관심영역이 유한요소법(finite element method)에 의해 모델링되어 관심영역의 3D 모델이 생성되는 과정;을 포함하며, 의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는 각 유한요소에 가장 근접한 복셀의 인텐시티가 각 유한요소의 물질특성으로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(2) generating the 3D model of the ROI may include generating a segmented medical image by segmenting the ROI or another region into a set of voxels based on the medical image; A region of interest based on the image is modeled by a finite element method to generate a 3D model of the region of interest, and the intensity of the medical image is mapped to the material properties of each finite element. Comprising the step of mapping the intensity of the voxel closest to each finite element to the material properties of each finite element; Computational fluid dynamics modeling and analysis method, characterized in that it comprises a.
(3) 관심영역의 3D 모델이 생성되는 단계는 의료영상을 기초로 관심영역 또는 다른 영역이 복셀(Boxel)의 집합으로 분할(segmentation)되어 분할된 의료영상이 생성되는 과정;그리고, 분할된 의료영상을 기초로 관심영역이 유한요소망(finite element mesh)에 의해 모델링되는 과정;을 포함하며, 의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는 각 메시(mesh)의 노드(node)에 가장 근접한 복셀의 인텐시티가 각 메시에 물질특성으로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(3) generating the 3D model of the ROI may include generating a segmented medical image by segmenting the ROI or another region into a set of voxels based on the medical image; The region of interest is modeled by a finite element mesh based on the image. The step of mapping the intensity of the medical image to the material properties of each finite element includes: Computational fluid dynamics modeling and analysis method, characterized in that it comprises a; process of mapping the intensity of the voxel closest to the node of the material to each mesh.
(4) 관심영역의 3D 모델이 생성되는 단계는 혈관을 포함하는 의료영상이 생성되는 과정; 의료영상을 기초로 혈관의 루멘(Lumen)이 분할(segmentation)되는 과정; 루멘과 접하는 관심영역이 유한요소망(finite element mesh)에 의해 모델링되어 혈관벽(blood vessel wall) 모델이 생성되는 과정; 그리고 루멘과 접하며 혈관벽과 다른 인텐시티를 가지는 관심영역이 유한요소망(finite element mesh)에 의해 모델링되어 병변(lesions) 모델이 생성되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(4) generating the 3D model of the ROI may include generating a medical image including blood vessels; A process of segmenting lumens of blood vessels based on a medical image; A region of interest in contact with the lumen is modeled by a finite element mesh to generate a blood vessel wall model; Computational fluid dynamics modeling based on material properties, comprising: a region of interest in contact with the lumen and having an intensity different from that of the vessel wall is modeled by a finite element mesh to generate a lesion model. And analytical method.
(5) 의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는 혈관벽 모델의 각 mesh에 대응하는 의료영상의 픽셀값(pixel value)이 혈관벽 모델의 각 메시에 물질특성으로 매핑되는 과정; 그리고 병변 모델의 각 mesh에 대응하는 의료영상의 픽셀값이 병변 모델의 각 메시에 물질특성으로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(5) The step of mapping the intensity of the medical image to the material properties of each finite element is such that the pixel value of the medical image corresponding to each mesh of the blood vessel wall model is a material property of each mesh of the blood vessel wall model. The process of mapping; And mapping the pixel values of the medical image corresponding to each mesh of the lesion model to the material characteristics of each mesh of the lesion model.
(6) 관심영역의 3D 모델이 생성되는 단계는 조영증강 심장 CT에 의해 의료영상이 생성되는 과정; 의료영상을 기초로 심혈관이 복셀(Boxel)의 집합으로 분할(segmentation)되는 과정; 심혈관의 루멘과 접하는 연조직(soft tissue)의 관심영역이 3D triangular mesh에 의해 모델링되어 혈관벽 모델이 생성되는 과정; 그리고 심혈관의 루멘과 접하며 연조직과 다른 인텐시티를 가지는 관심영역이 3D triangular mesh에 의해 모델링되어 플라그(plaque) 모델이 생성되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(6) generating the 3D model of the ROI may include generating a medical image by contrast-enhanced heart CT; A process in which a cardiovascular is segmented into a set of voxels based on a medical image; A process of generating a blood vessel wall model by modeling a region of interest of the soft tissue in contact with the lumen of the cardiovascular vessel by a 3D triangular mesh; And a process of generating a plaque model by modeling a region of interest in contact with the cardiovascular lumen and having different intensities from the soft tissues by a 3D triangular mesh. .
(7) 의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는 루멘 바깥에서 혈관벽 모델의 각 3D triangular mesh에 가장 근접한 복셀의 CT 값(CT density)이 해당 3D triangular mesh의 밀도(density)로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(7) The step of mapping the intensity of the medical image to the material properties of each finite element is such that the CT density of the voxel closest to each 3D triangular mesh of the vessel wall model outside the lumen is determined by the CT density of the 3D triangular mesh. Computational fluid dynamics modeling and analysis method based on the material properties, characterized in that it comprises a; mapping to density (density).
(8) 의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는 루멘 바깥에서 플라그 모델의 각 3D triangular mesh에 가장 근접한 복셀의 CT 값이 해당 3D triangular mesh의 밀도로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(8) The step of mapping the intensity of the medical image to the material properties of each finite element is such that the CT values of the voxels closest to each 3D triangular mesh of the plaque model outside the lumen are mapped to the density of the corresponding 3D triangular mesh. Computational fluid dynamics modeling and analysis method based on the material characteristics, characterized in that it comprises a; process.
(9) 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계는 심혈관 내에서 플라그 통과 전후의 관류(perfusion)에 대해 FFR(Fractional Flow Reserve)이 컴퓨터 계산되는 과정; 그리고 계산된 FFR을 기초로 플라그의 안정성(stability)이 평가되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(9) The analysis of the flow associated with the 3D model of the region of interest may include computerized calculation of the Fractional Flow Reserve (FFR) for perfusion before and after the passage of plaque in the cardiovascular system; And Computational fluid dynamics modeling and analysis method based on the material characteristics, characterized in that it comprises a; process based on the calculated FFR stability (stability) is evaluated.
(10) 전산유체역학(Computational Fluid Dynamics; CFD) 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계 이전에, 환자맞춤 경계조건을 설정하는 단계;를 포함하며, 환자맞춤 경계조건을 설정하는 단계는 MRI Venc 영상을 이용하여 심혈관의 혈류입력을 촬영하는 과정; 그리고 촬영된 혈류입력을 기초로 CFD 수단에 환자맞춤 입력 경계조건이 설정되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(10) prior to the step of analyzing the flow associated with the 3D model of the region of interest by means of Computational Fluid Dynamics (CFD) means, establishing patient-specific boundary conditions; The setting of the condition may include taking a blood flow input of the cardiovascular vessel using an MRI Venc image; And setting a patient-specific input boundary condition in the CFD means based on the photographed blood flow input.
(11) 전산유체역학(Computational Fluid Dynamics; CFD) 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계 이전에, 환자맞춤 경계조건을 설정하는 단계;를 포함하며, 환자맞춤 경계조건을 설정하는 단계는 심근분할 방법을 사용하여 심혈관이 먹여 살리는 심근의 양이 측정되는 과정; 그리고 측정된 심근의 양을 기초로 CFD 수단에 환자맞춤 출력 경계조건이 설정되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.(11) establishing patient-specific boundary conditions prior to the step of analyzing the flow associated with the 3D model of the region of interest by means of Computational Fluid Dynamics (CFD) means; The step of setting the condition is a process in which the amount of myocardium that is fed by the cardiovascular system is measured using a myocardial splitting method; And setting a patient-specific output boundary condition on the CFD means based on the measured amount of myocardium.
(12) 물질특성에 기반한 전산유체역학 모델링 및 분석 방법을 컴퓨터에서 실행시키기 위한 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록 매체.(12) A computer-readable recording medium having recorded thereon a computer program for performing computational fluid dynamics modeling and analysis method based on material properties.
본 개시에 따른 하나의 물질특성에 기반한 전산유체역학 모델링 및 분석 방법에 의하면, 관심영역의 3D 모델을 CFD 모델로 해석한 결과와 실재 측정된 결과의 차이를 허용오차 범위 내로 줄일 수 있다.According to the computational fluid dynamics modeling and analysis method based on one material property according to the present disclosure, it is possible to reduce the difference between the result of analyzing the 3D model of the region of interest into the CFD model and the actual measured result within the tolerance range.
또한, 본 개시에 따른 다른 하나의 물질특성에 기반한 전산유체역학 모델링 및 분석 방법에 의하면, CFD에 의해 계산된 FFR_CT와 실측된 FFR와의 차이를 허용오차 범위 내로 줄일 수 있다.In addition, according to another method of computational fluid dynamics modeling and analysis based on the material properties according to the present disclosure, it is possible to reduce the difference between the FFR_CT calculated by the CFD and the measured FFR within the tolerance range.
또한, 본 개시에 따른 또 다른 하나의 물질특성에 기반한 전산유체역학 모델링 및 분석 방법에 의하면, 비침습적으로 혈관의 병변의 심각성 정도를 평가하는 방법의 신뢰성이 향상된다.In addition, according to the computational fluid dynamics modeling and analysis method based on another material property according to the present disclosure, the reliability of the method of evaluating the severity of vascular lesions in a non-invasive manner is improved.

Claims (13)

  1. 의료영상(medical image)에 포함된 관심영역(region of interest)이 유한요소법(finite element method)에 의해 모델링되어 관심영역의 3D 모델이 생성되는 단계;A region of interest included in the medical image is modeled by a finite element method to generate a 3D model of the region of interest;
    관심영역의 3D 모델의 각 유한요소(finite element)에 대응하는 의료영상의 인텐시티(intensity)가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계; 그리고Mapping an intensity of a medical image corresponding to each finite element of the 3D model of the ROI to material properties of each finite element; And
    전산유체역학(Computational Fluid Dynamics; CFD) 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계;를 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method comprising the step of analyzing the flow associated with the 3D model of the region of interest by means of Computational Fluid Dynamics (CFD) means.
  2. 청구항 1에 있어서,The method according to claim 1,
    관심영역의 3D 모델이 생성되는 단계는 유한요소망(finite element mesh)에 의해 관심영역이 모델링되는 과정;을 포함하며, The generating of the 3D model of the ROI may include a process of modeling the ROI by a finite element mesh.
    의료영상의 인텐시티가 각 유한요소의 물질특성으로 매핑되는 단계는 각 메시(mesh)에 대응하는 의료영상의 인텐시티가 각 메시에 물질특성으로 매핑되는 과정;을 포함하며,The step of mapping the intensity of the medical image to the material properties of each finite element includes a process of mapping the intensity of the medical image corresponding to each mesh to the material properties of each mesh.
    관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계는 CFD 수단에 의해 관심영역의 3D 모델의 특정 위치에서 FFR(Fractional Flow Reserve)가 컴퓨터 계산되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.The step of analyzing a flow related to the 3D model of the ROI may include a process of computing a FFR (Fractional Flow Reserve) at a specific position of the 3D model of the ROI by means of CFD. Computational fluid dynamics modeling and analysis method based on.
  3. 청구항 1에 있어서,The method according to claim 1,
    관심영역의 3D 모델이 생성되는 단계는 의료영상을 기초로 관심영역 또는 다른 영역이 복셀(Boxel)의 집합으로 분할(segmentation)되어 분할된 의료영상이 생성되는 과정; 그리고, 분할된 의료영상을 기초로 관심영역이 유한요소법(finite element method)에 의해 모델링되어 관심영역의 3D 모델이 생성되는 과정;을 포함하며,Generating the 3D model of the ROI may include generating a segmented medical image by segmenting the ROI or another region into a set of voxels based on the medical image; And a process of generating a 3D model of the region of interest by modeling the region of interest based on the segmented medical image by a finite element method.
    의료영상의 인텐시티가 각 유한요소의 물질특성으로 매핑되는 단계는 각 유한요소에 가장 근접한 복셀의 인텐시티가 각 유한요소의 물질특성으로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.The step of mapping the intensity of the medical image to the material properties of each finite element comprises the step of mapping the intensity of the voxel closest to each finite element to the material properties of each finite element. Mechanics modeling and analysis method.
  4. 청구항 1에 있어서,The method according to claim 1,
    관심영역의 3D 모델이 생성되는 단계는 의료영상을 기초로 관심영역 또는 다른 영역이 복셀(Boxel)의 집합으로 분할(segmentation)되어 분할된 의료영상이 생성되는 과정;그리고, 분할된 의료영상을 기초로 관심영역이 유한요소망(finite element mesh)에 의해 모델링되는 과정;을 포함하며,Generating the 3D model of the ROI may include generating a segmented medical image by segmenting the ROI or another region into a set of voxels based on the medical image; and based on the segmented medical image. The RO region is modeled by a finite element mesh.
    의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는 각 메시(mesh)의 노드(node)에 가장 근접한 복셀의 인텐시티가 각 메시에 물질특성으로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.The step of mapping the intensity of the medical image to the material properties of each finite element includes the step of mapping the intensity of the voxel closest to the node of each mesh to the material properties of each mesh. Computational fluid dynamics modeling and analysis method based on the material properties, characterized in that.
  5. 청구항 1에 있어서,The method according to claim 1,
    관심영역의 3D 모델이 생성되는 단계는:The steps for generating a 3D model of the region of interest are:
    혈관을 포함하는 의료영상이 생성되는 과정;Generating a medical image including blood vessels;
    의료영상을 기초로 혈관의 루멘(Lumen)이 분할(segmentation)되는 과정;A process of segmenting lumens of blood vessels based on a medical image;
    루멘과 접하는 관심영역이 유한요소망(finite element mesh)에 의해 모델링되어 혈관벽(blood vessel wall) 모델이 생성되는 과정; 그리고A region of interest in contact with the lumen is modeled by a finite element mesh to generate a blood vessel wall model; And
    루멘과 접하며 혈관벽과 다른 인텐시티를 가지는 관심영역이 유한요소망(finite element mesh)에 의해 모델링되어 병변(lesions) 모델이 생성되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling based on material properties, including the process of generating a lesion model by modeling a region of interest having lumens and different intensities from the vessel wall by a finite element mesh. Analytical Method.
  6. 청구항 5에 있어서,The method according to claim 5,
    의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는: The steps in which the intensity of the medical image is mapped to the material properties of each finite element are:
    혈관벽 모델의 각 mesh에 가장 가까운 루멘에 접하는 영역의 픽셀값(pixel value)이 혈관벽 모델의 각 메시에 물질특성으로 매핑되는 과정; 그리고A process in which a pixel value of an area in contact with the lumen closest to each mesh of the blood vessel wall model is mapped to a material property on each mesh of the blood vessel wall model; And
    병변 모델의 각 mesh에 대응하는 의료영상의 픽셀값이 병변 모델의 각 메시에 물질특성으로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method, characterized in that it comprises a; mapping the pixel values of the medical image corresponding to each mesh of the lesion model to the material properties of each mesh of the lesion model.
  7. 청구항 1에 있어서,The method according to claim 1,
    관심영역의 3D 모델이 생성되는 단계는:The steps for generating a 3D model of the region of interest are:
    심장 CT에 의해 의료영상이 생성되는 과정;A process in which a medical image is generated by a heart CT;
    의료영상을 기초로 심혈관이 복셀(Boxel)의 집합으로 분할(segmentation)되는 과정;A process in which a cardiovascular is segmented into a set of voxels based on a medical image;
    심혈관의 루멘과 접하는 연조직(soft tissue)의 관심영역이 3D triangular mesh에 의해 모델링되어 혈관벽 모델이 생성되는 과정; 그리고A process of generating a blood vessel wall model by modeling a region of interest of the soft tissue in contact with the lumen of the cardiovascular vessel by a 3D triangular mesh; And
    심혈관의 루멘과 접하며 연조직과 다른 인텐시티를 가지는 관심영역이 3D triangular mesh에 의해 모델링되어 플라그(plaque) 모델이 생성되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method characterized in that it comprises a process of generating a plaque model of the region of interest that is in contact with the lumen of the cardiovascular system and has different intensities from the soft tissue is generated by the 3D triangular mesh.
  8. 청구항 7에 있어서,The method according to claim 7,
    의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는:The steps in which the intensity of the medical image is mapped to the material properties of each finite element are:
    루멘에 접하는 영역에서 혈관벽 모델의 각 3D triangular mesh에 가장 근접한 복셀의 CT 값(CT density)이 해당 3D triangular mesh의 밀도(density)로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computing CT values of the voxels closest to each 3D triangular mesh of the vascular wall model in the region adjacent to the lumen is mapped to the densities of the 3D triangular mesh. Hydrodynamic Modeling and Analysis Methods.
  9. 청구항 7에 있어서,The method according to claim 7,
    의료영상의 인텐시티가 각 유한요소의 물질특성(material properties)으로 매핑되는 단계는:The steps in which the intensity of the medical image is mapped to the material properties of each finite element are:
    루멘 바깥에서 플라그 모델의 각 3D triangular mesh에 가장 근접한 복셀의 CT 값이 해당 3D triangular mesh의 밀도로 매핑되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method, characterized in that it comprises the step of mapping the CT value of the voxel closest to each 3D triangular mesh of the plaque model outside the lumen to the density of the 3D triangular mesh.
  10. 청구항 7에 있어서,The method according to claim 7,
    관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계는: The steps involved in analyzing the flow associated with the 3D model of the region of interest are:
    심혈관 내에서 플라그 통과 전후의 혈류에 대해 FFR(Fractional Flow Reserve)이 컴퓨터 계산되는 과정; 그리고Computer-calculated Fractional Flow Reserve (FFR) for blood flow before and after plaque passage in the cardiovascular system; And
    계산된 FFR을 기초로 플라그의 안정성(stability)이 평가되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method based on the material properties, characterized in that it comprises a; process to evaluate the stability of the plaque based on the calculated FFR.
  11. 청구항 7에 있어서,The method according to claim 7,
    CFD 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계 이전에, 환자맞춤 경계조건을 설정하는 단계;를 포함하며,Prior to the step of analyzing the flow associated with the 3D model of the region of interest by means of the CFD means, setting patient-specific boundary conditions;
    환자맞춤 경계조건을 설정하는 단계는:The steps for setting patient-specific boundary conditions are:
    MRI Venc 영상을 이용하여 심혈관의 혈류입력을 촬영하는 과정; 그리고Photographing blood flow input of the cardiovascular system using an MRI Venc image; And
    촬영된 혈류입력을 기초로 CFD 수단에 환자맞춤 입력 경계조건이 설정되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method, characterized in that it comprises a; step of setting the patient-specific input boundary conditions on the CFD means based on the photographed blood flow input.
  12. 청구항 7에 있어서,The method according to claim 7,
    CFD 수단에 의해 관심영역의 3D 모델과 관련된 흐름(flow)이 분석되는 단계 이전에, 환자맞춤 경계조건을 설정하는 단계;를 포함하며,Prior to the step of analyzing the flow associated with the 3D model of the region of interest by means of the CFD means, setting patient-specific boundary conditions;
    환자맞춤 경계조건을 설정하는 단계는:The steps for setting patient-specific boundary conditions are:
    심근분할 방법을 사용하여 심혈관이 먹여 살리는 심근의 양이 측정되는 과정; 그리고A process in which the amount of myocardium that is fed by the cardiovascular system is measured using a myocardial division method; And
    측정된 심근의 양을 기초로 CFD 수단에 환자맞춤 출력 경계조건이 설정되는 과정;을 포함하는 것을 특징으로 하는 물질특성에 기반한 전산유체역학 모델링 및 분석 방법.Computational fluid dynamics modeling and analysis method, characterized in that it comprises a; step of setting the patient-specific output boundary conditions on the CFD means based on the measured amount of myocardium.
  13. 청구항 1 내지 청구항 12 중 어느 한 항에 기재된 방법을 컴퓨터에서 실행시키기 위한 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록 매체.A computer-readable recording medium having recorded thereon a program for executing the method according to any one of claims 1 to 12 on a computer.
PCT/KR2014/002654 2013-04-02 2014-03-28 Method for modeling and analyzing computational fluid dynamics on basis of material properties WO2014163334A1 (en)

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
KR20130035761 2013-04-02
KR10-2013-0035761 2013-04-02
KR20130035763 2013-04-02
KR10-2013-0035763 2013-04-02
KR10-2013-0080429 2013-07-09
KR1020130080429A KR101530352B1 (en) 2013-04-02 2013-07-09 Cfd modeling and analysis method based on material properties

Publications (1)

Publication Number Publication Date
WO2014163334A1 true WO2014163334A1 (en) 2014-10-09

Family

ID=51658575

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2014/002654 WO2014163334A1 (en) 2013-04-02 2014-03-28 Method for modeling and analyzing computational fluid dynamics on basis of material properties

Country Status (1)

Country Link
WO (1) WO2014163334A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108846192A (en) * 2018-06-08 2018-11-20 中国船舶科学研究中心（中国船舶重工集团公司第七0二研究所） A kind of ship three dimensional sound flexibility analysis method of any impedance bundary of structure
CN109363661A (en) * 2018-09-25 2019-02-22 杭州晟视科技有限公司 Blood flow reserve score determines system, method, terminal and storage medium
EP3525744A4 (en) * 2016-10-14 2020-03-25 Di Martino, Elena Methods, systems, and computer readable media for evaluating risks associated with vascular pathologies
US11395597B2 (en) 2018-06-26 2022-07-26 General Electric Company System and method for evaluating blood flow in a vessel

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006517822A (en) * 2003-02-12 2006-08-03 コーニンクレッカ　フィリップス　エレクトロニクス　エヌ　ヴィ Method for three-dimensional modeling of tubular tissue
US20090318802A1 (en) * 2007-12-18 2009-12-24 Searete Llc, A Limited Liability Corporation Of The State Of Delaware System, devices, and methods for detecting occlusions in a biological subject
WO2012021307A2 (en) * 2010-08-12 2012-02-16 Heartflow, Inc. Method and system for patient-specific modeling of blood flow

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006517822A (en) * 2003-02-12 2006-08-03 コーニンクレッカ　フィリップス　エレクトロニクス　エヌ　ヴィ Method for three-dimensional modeling of tubular tissue
US20090318802A1 (en) * 2007-12-18 2009-12-24 Searete Llc, A Limited Liability Corporation Of The State Of Delaware System, devices, and methods for detecting occlusions in a biological subject
WO2012021307A2 (en) * 2010-08-12 2012-02-16 Heartflow, Inc. Method and system for patient-specific modeling of blood flow

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
DAVID A. STEINMAN: "Image-Based Computational Fluid Dynamics Modeling in Realistic Arterial Geometries", ANNALS OF BIOMEDICAL ENGINEERING, vol. 30, 2002, pages 483 - 497 *
SAMUEL A. KOCK ET AL.: "Mechanical stresses in carotid plaques using MRI-based fluid- structure interaction models", JOURNAL OF BIOMECHANICS, vol. 41, 2008, pages 1651 - 1658 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3525744A4 (en) * 2016-10-14 2020-03-25 Di Martino, Elena Methods, systems, and computer readable media for evaluating risks associated with vascular pathologies
EP4094742A1 (en) * 2016-10-14 2022-11-30 Di Martino, Elena Method for evaluating risks associated with vascular pathologies
US11521741B2 (en) 2016-10-14 2022-12-06 Elena Di Martino Methods, systems, and computer readable media for evaluating risks associated with vascular pathologies
CN108846192A (en) * 2018-06-08 2018-11-20 中国船舶科学研究中心（中国船舶重工集团公司第七0二研究所） A kind of ship three dimensional sound flexibility analysis method of any impedance bundary of structure
US11395597B2 (en) 2018-06-26 2022-07-26 General Electric Company System and method for evaluating blood flow in a vessel
CN109363661A (en) * 2018-09-25 2019-02-22 杭州晟视科技有限公司 Blood flow reserve score determines system, method, terminal and storage medium

Similar Documents

Publication Publication Date Title
US11896416B2 (en) Method for calculating coronary artery fractional flow reserve on basis of myocardial blood flow and CT images
John et al. Magnetic resonance to assess the aortic valve area in aortic stenosis: how does it compare to current diagnostic standards?
CN111134651B (en) Method, device and system for calculating fractional flow reserve based on intracavity images and computer storage medium
CN104736046B (en) System and method for number evaluation vascular system
CN108511075B (en) Method and system for non-invasively acquiring fractional flow reserve
CN102551780B (en) Device and computed tomography scanner for determining and visualizing the perfusion of the myocardial muscle
JP2021106887A (en) Medical image display device, method and program
CN108294735A (en) The method and system of non-invasive functional assessment for coronary artery stenosis
CN107257655A (en) Method, system and computer program product for determining haemodynamic condition parameter using the signal obtained from many spectral coverage blood flows and Perfusion Imaging
CN106659399A (en) Method and system for non-invasive functional assessment of coronary artery stenosis using flow computations in diseased and hypothetical normal anatomical models
CN106659400A (en) Apparatus for determining a fractional flow reserve value
CN106537392A (en) Method and system for hemodynamic computation in coronary arteries
CN106803251B (en) The apparatus and method of aortic coaractation pressure difference are determined by CT images
WO2018155765A1 (en) Method and device analyzing plaque from computed tomography image
JP2019512131A (en) System and method for identifying and modeling unresolved blood vessels in an image based patient specific hemodynamic model
JP2018509971A (en) Method and apparatus for quantitative flow analysis
WO2014163334A1 (en) Method for modeling and analyzing computational fluid dynamics on basis of material properties
CN112749521A (en) Processing method and system of hemodynamic index data
JP2021183165A (en) Method for estimating flow rates and pressure gradients in arterial networks from patient-specific computed tomography angiogram-based contrast distribution
Marlevi et al. Non-invasive estimation of relative pressure for intracardiac flows using virtual work-energy
WO2015147595A1 (en) Method for distinguishing artifact and diseased area in medical images
KR101530352B1 (en) Cfd modeling and analysis method based on material properties
Liu et al. Consistency in geometry among coronary atherosclerotic plaques extracted from computed tomography angiography
Sarrami-Foroushani et al. Velocity measurement in carotid artery: quantitative comparison of time-resolved 3D phase-contrast MRI and image-based computational fluid dynamics
US10631811B2 (en) Method and system for processing of medical images for generating a prognosis of cardiac function

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14779683

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14779683

Country of ref document: EP

Kind code of ref document: A1