CN103300820A - Method and system for non-invasive functional assessment of coronary artery stenosis - Google Patents

Abstract

A method and system for non-invasive assessment of coronary artery stenosis is disclosed. Patient-specific anatomical measurements of the coronary arteries are extracted from medical image data of a patient acquired during rest state. Patient-specific rest state boundary conditions of a model of coronary circulation representing the coronary arteries are calculated based on the patient-specific anatomical measurements and non-invasive clinical measurements of the patient at rest. Patient-specific rest state boundary conditions of the model of coronary circulation representing the coronary arteries are calculated based on the patient-specific anatomical measurements and non-invasive clinical measurements of the patient at rest. Hyperemic blood flow and pressure across at least one stenosis region of the coronary arteries are simulated using the model of coronary circulation and the patient-specific hyperemic boundary conditions. Fractional flow reserve (FFR) is calculated for the at least one stenosis region based on the simulated hyperemic blood flow and pressure.

Description

The method and system that is used for the Noninvasive functional assessment of coronary stricture

The application requires the U.S. Provisional Application No.61/610 of submission on March 13rd, 2012,134 priority, and it discloses incorporated herein by reference.

Technical field

The present invention relates to the Noninvasive functional assessment of coronary stricture, and more specifically, relate to the Noninvasive functional assessment according to the coronary stricture of medical image and Simulation of Blood.

Background technology

Cardiovascular disease (CVD) is global main causes of death.Among various CVD, coronary artery disease (CAD) almost accounts for 50 percent of those death toll.Although the remarkable improvement of medical imaging and other diagnostic modes (modality), it is very high that the increase of CAD patient's too early M ﹠ M remains.The current clinical practice that is used for the diagnosis of coronary stricture and management relates to visually or the assessment of the lesion vessels by quantitative coronary visualization (QCA).The anatomy summary that such assessment provides narrow section and carried the tumor blood vessel to the clinician, described anatomy summary comprise area reducing, damaged length and minimum cavity footpath, but the functional assessment about the damage influence of the blood flow by this blood vessel is not provided.Compare with invasive angiogram, measuring the better option that blood flow reserve mark (FFR) has been proved to be to be used in reference to emissary vein regeneration decision in the blood vessel of suffering from stenosis by Pressure wire is inserted into, is more effective because FFR causes in the damage in the identification ischemia.If narrow QCA only assesses morphological meaning and has many other limitation.Relate to and Pressure wire need to be inserted into the risk that the intervention in the blood vessel is associated based on the FFR measurement of Pressure wire, and for very narrow narrow, Pressure wire can be induced additional pressure drop.

Summary of the invention

The invention provides the method and system for the Noninvasive functional assessment of coronary stricture.Embodiments of the invention are by calculating the functional assessment that blood flow reserve mark (FFR) and/or other functional measurements provide the order of severity of coronary stricture according to medical image and Simulation of Blood.Embodiments of the invention have utilized bottom depression of order patient-specific (patient-specific) Studies on Hemodynamic Changes of use computational fluid dynamics (CFD) simulation.This makes FFR or other hematodinamics amounts of the order of severity of can be during the image acquisition process near in real time computational representation damage of the present invention.Embodiments of the invention have also utilized other to calculate the boundary condition of simulating for patient-specific CFD based on non-image Noninvasive patient information.

In one embodiment of the invention, patient-specific anatomic measurement coronarius extracts from the patient's that gathers during static (rest) state medical image.Calculate the patient-specific resting state boundary condition of the model of expression coronary circulation coronarius based on the patient-specific anatomic measurement that is in static patient and Noninvasive clinical measurement.Based on the silent boundary condition be used for the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation.Simulate congested blood flow and pressure across at least one at least one stenosis area coronarius with the model of described coronary circulation and the congested boundary condition of described patient-specific.The blood flow reserve mark (FFR) that calculates described at least one stenosis area based on the congested blood flow of simulating and pressure.

In another embodiment of the present invention, during congestive state, from patient's medical image, gather patient-specific anatomic measurement coronarius.Calculate the congested boundary condition of patient-specific of the model of expression coronary circulation coronarius based on the described patient-specific anatomic measurement that is in congested patient and Noninvasive clinical measurement.Simulate congested blood flow and pressure across at least one at least one stenosis area coronarius with the model of described coronary circulation and the congested boundary condition of described patient-specific.The blood flow reserve mark (FFR) that calculates described at least one stenosis area based on the congested blood flow of simulating and pressure.

The following specifically describes and accompanying drawing by reference, these and other features of the present invention will be apparent for the ordinary skill in the art.

Description of drawings

Fig. 1 illustrates the framework that is used for according to an embodiment of the invention the Noninvasive functional assessment of coronary stricture;

Fig. 2 illustrates the method that is used for according to an embodiment of the invention the Noninvasive functional assessment of coronary stricture;

Fig. 3 illustrates the example results for the patient-specific anatomic model that generates coronary vessel tree;

Fig. 4 illustrates the reduced-order model that is used for according to an embodiment of the invention the simulation coronary circulation;

Fig. 5 illustrates the method that is used for estimating according to an embodiment of the invention resting state blood capillary resistance;

Fig. 6 illustrates the calculating of the FFR that uses according to an embodiment of the invention the Extraordinary reduced-order model; And

Fig. 7 is the high level block diagram that can realize computer of the present invention.

The specific embodiment

The present invention relates to use the method and system of the Noninvasive functional assessment that is used for coronary stricture of medical image and Simulation of Blood.In this article embodiments of the invention are described to provide for the visual understanding of simulation blood flow with the method for assessment coronary stricture.Digital picture usually is made of the numeral of one or more objects (or shape).Usually in this article the numeral of object is being described aspect identification and the manipulating objects.This type of manipulation is the virtual manipulation that realizes in the memorizer of computer system or other circuit/hardware.Therefore, should be understood that, can in computer system, carry out embodiments of the invention with the data that are stored in this computer system.

Fig. 1 illustrates the framework that is used for according to an embodiment of the invention the Noninvasive functional assessment of coronary stricture.As illustrated among Fig. 1, described framework comprises image acquisition phase 102, anatomy modelling phase 104, Simulation of Blood stage 106 and blood flow reserve mark (FFR) calculation stages 108.In image acquisition phase 102, the medical image such as coronary artery computer tomography (CT) and other Noninvasive clinical measurements such as heart rate, blood pressure etc. of patient are gathered.In the anatomy modelling phase 104, image segmentation and central line pick-up algorithm are used to generate patient's patient-specific anatomic model coronarius.Can be based on adjusting the patient-specific anatomic model from clinician 110 feedback.In the Simulation of Blood stage 106, computational fluid dynamics is used to simulation by blood flow coronarius.In one embodiment, the depression of order circulation model can be used to the patient-specific Simulation of Blood in independent model that vascular tree adds that each is narrow and the potential boundary condition.The patient-specific boundary condition uses the patient-specific modeling of maximum congested condition and (auto-regulation) mechanism of automatically regulating is calculated.Clinician 110 can provide the feedback about Simulation of Blood, for example with the various parameters that change circulation model or with the rank of the modeling that changes circulation model.In FFR calculation stages 108, based on the simulated pressure that is produced by Simulation of Blood for each narrow calculating FFR.When the method for reference Fig. 2, image acquisition phase 102, anatomy are built the stage of touching 104, Simulation of Blood stage 106 and FFR calculation stages 108 and are described in more detail.

Fig. 2 illustrates the method that is used for according to an embodiment of the invention the Noninvasive functional assessment of coronary stricture.With reference to figure 2, at step 202 place, patient's medical image and Noninvasive clinical measurement are received.Can receive the medical image from one or more imaging patterns.For example, this medical image can comprise the medical imaging pattern of computer tomography (CT), Dyna CT, magnetic resonance (MR), angiography, ultrasound wave, single photon emission computed tomography and any other type.Medical image can be 2D, 3D or 4D (3D+ time) medical image.Can directly receive medical image from the one or more image capture devices such as CT Scanner, MR scanner unit, angiography scan machine, ultrasonic equipment etc., perhaps can receive medical image by loading the pre-stored medical image that is used for the patient.

In favourable embodiment, 3D coronary artery CT angiography (CTA) image is gathered on CT Scanner.The CTA image guarantees to comprise that the crown vascular system that comprises narrow (one or more) blood vessel comes suitably imaging with the contrast agent that is injected among the patient.In this stage, the clinician can provide by interactively on image and check that they identify the option of interested damage (narrow).Can also carry out this step (step 204) to the anatomic model that from view data, extracts.Replacedly, narrowly can use the algorithm for the automatic detection of coronary stricture automatically to be detected in view data, described algorithm is such as the method for the automatic detection that is used for coronary stricture described in U.S. publication application No.2011/0224542, and this patent application is incorporated herein by reference.Except medical image, also gathered such as patient's heart rate and other Noninvasive clinical measurements heart contraction blood pressure and the diastole blood pressure.

At step 204 place, measurement coronarius is extracted from patient's medical image.In the exemplary embodiment, this medical image is under static state gathered, and measurement coronarius is extracted from the view data that under static state gathers.In favourable embodiment, measurement coronarius is extracted by the patient-specific anatomic model that generates the coronary vessel tree that is generated by this medical image, but the invention is not restricted to this.In order to generate patient-specific anatomic model coronarius, coronary artery uses the coronary artery central line pick-up algorithm of automatization to cut apart in the 3D medical image.Coronary artery can be cut apart with any coronary artery dividing method.For example, coronary artery can use in the U.S.'s described method of publication application No.2010/0067760 divided in the CT volume, and this patent application is incorporated herein by reference.In case coronary artery centrage tree has been extracted, just can generate the cross-sectional profiles line at each some place of this centrage tree.Cross-sectional profiles line at each centerline points place has provided the cross section taken in correspondence area measurement at this some place in coronary artery.Then generate the geometric jacquard patterning unit surface model for the coronary artery of cutting apart.For example, in U.S. Patent No. 7,860,290 and U.S. Patent No. 7,953,266 in the method that is used for anatomy modeling coronarius has been described, described patent is both incorporated herein by reference.Except coronary artery, the patient-specific anatomic model can comprise aortic root and aortal proximal part.Each narrow detailed 3D model also extracts with similar algorithm, and described detailed 3D model comprises the quantification of near-end blood vessel diameter and area, distal vessels diameter and area, minimum cavity footpath and area and narrow length.Fig. 3 illustrates the example results for the patient-specific anatomic model that generates coronary vessel tree.The image 300 of Fig. 3 shows coronary artery CTA data.Image 310 shows the centrage tree 312 from the CTA extracting data.Image 320 shows the cross-sectional profiles line 322 that extracts at each some place of centrage tree 312.Image 330 shows the 2D surface mesh 332 of coronary artery, aortic root and aortal proximal part.

Above-mentioned anatomy modeling task can automatically be carried out or can be driven by the user, thereby allows user (clinician) interactively that anatomic model is changed to analyze such change to the impact of the subsequent calculations of FFR.Except coronary vessel tree, cardiac muscle also divided in medical image (automatically or manually) to determine the estimation of left ventricular mass, described estimation is used to estimate this patient's absolute rest flow according to embodiments of the invention.In the exemplary embodiment, the patient-specific anatomic model of heart is automatically generated by view data.The anatomy heart model is the multi-part model with a plurality of heart components, and described a plurality of heart components comprise four chambers (left ventricle, left atrium, right ventricle and right atrium).The anatomy heart model can also comprise the parts such as cardiac valve (aortic valve, Bicuspid valve, Tricuspid valve and valve of pulmonary trunk) and aorta.Such aggregative model of heart is used to catch various form, function and pathological change.The modularization and administrative levels method can be used to reduce the anatomy complexity and help effectively and flexibly estimating of each dissection.4D anatomy heart model can generate the individual body Model of each heart component by for example using marginal space learning (MSL), and then generates by setting up the corresponding integrated heart component of mesh point.Described the additional detail about the generation of such 4D patient-specific heart model in U.S. publication application No.2012/0022843, this patent application is incorporated herein by reference.

Turn back to Fig. 2, at step 206 place, the patient-specific Simulation of Blood uses the boundary condition that calculates based on Noninvasive patient-specific clinical measurement to carry out.The interested hematodinamics amount of coronary circulation such as FFR is based on flow or the meansigma methods of pressure during cardiac cycle.For the effective clinical workflow that is used for via the assessment of the FFR of simulation, model complexity and being equilibrated in the situation about need not compromise to result's degree of accuracy between computation time enjoy expectation.In favourable embodiment of the present invention, reduced-order model is used to the patient-specific Simulation of Blood, and this is so that can realize the assessment of the functional meaning of coronary stricture.This reduced-order model provides flow in the vascular tree and the accurate estimation of pressure distribution, and is that computational efficiency is high, therefore so that can realize and clinical workflow seamless integrated.Although described reduced-order model for the coronary circulation simulation in this article, the invention is not restricted to this, but also can use full Scale Model or multiple dimensioned model.

Fig. 4 illustrates the reduced-order model that is used for according to an embodiment of the invention the simulation coronary circulation.As shown in Figure 4, heart model 402 is coupling in aortal root place.Heart model 402 may be implemented as full 3D heart model, perhaps may be implemented as by the parameterized lumped model of patient-specific data.Aorta and large artery trunks are (for example, left coronary artery (LCA), right coronary artery (RCA) etc.) be represented as 1D flow model 404,406,408,410,412,414,416,418 and 420, because these 1D flow model 404-418 is producing reliable result and is considering the ripple propagation phenomenon aspect pressure and the flow rate value.All capillary beds will be simulated by lumped parameter model 422,424,426,428 and 430, and described lumped parameter model has illustrated the resistance that is applied to blood flow, and the compliance of distal vessels has been described.For coronary arterial tree, the flow in large (visceral pericardium) blood vessel calculates by the 1D model in the phylogenetic tree model 421.Narrow section 432 and 434 (namely, the zone that detects in the blood vessel is narrow or narrows) can not simulate with the 1D flow model, because in cross-sectional area, exist high variability and narrow shape to affect the blood flow behavior, and affect particularly the trans narrow pressure drop that in the assessment of so narrow functional importance, plays a major role.Crown vascular bed comes modeling by lumped parameter model 424,426,428 and 430, and described lumped parameter model 424,426,428 and 430 has been considered at them that scheming is shunk under the meaning of impact of flow waveform and has been suitable for coronary circulation.

The reduced-order model of coronary circulation

As shown in Figure 4, aorta (404), large artery trunks (406,408,410,412,414,416,418 and 420) and the crown visceral pericardium blood vessel (421) supplied by this aorta are modeled as axisymmetric 1-D blood vessel section, wherein blood flow satisfies following characteristic: the conservation of mass, the conservation of momentum, and for the state equation of wall distortion.Blood vessel wall can be modeled as purely elastic or viscoelasticity.The entrance boundary condition can be by stipulating with the implicit expression coupling of heart model 402 or by the data on flows of measuring.Export boundary condition is coupled to provide by the implicit expression with the model (424,426,428 and 430) of crown vascular bed, and cross point (bifurcation) solves by the seriality of considering gross pressure and flow.In addition, can introduce following loss coefficient, described loss coefficient has illustrated that described loss coefficient depends on the angle between the blood vessel section in the energy loss at place, cross point:

$\frac{\∂A\left(t\right)}{\∂t}+\frac{\∂q\left(t\right)}{\∂x}=0---\left(1\right)$

$\frac{\∂q\left(t\right)}{\∂t}+\frac{\∂}{\∂t}\left(\mathrm{\α}\frac{q^2\left(t\right)}{A\left(t\right)}\right)+\frac{A\left(t\right)}{\mathrm{\ρ}}\frac{\∂p\left(t\right)}{\∂x}=K_R\frac{q\left(t\right)}{A\left(t\right)}---\left(2\right)$

$p\left(t\right)=\frac{4}{3}\frac{\mathrm{Eh}}{r_0}(1-\sqrt{\frac{A_0}{A\left(t\right)}}),---\left(3\right)$

Wherein q is flow rate, and A is cross-sectional area, and p is pressure, and α is the momentum flux correction coefficient, K _RBe the mark parameter, ρ is density, and E is Young (Young) modulus, and h is wall thickness and r ₀It is initial radium.The wall attribute can be by estimating to determine for the empirical relation match of the measured data in the patient-specific anatomic model that extracts or based on the patient-specific of wall compliance.Can also use other interchangeable formulism of accurate 1-D flow equation, modeling is carried out in the impact of viscoelasticity, non-newtonian behaviour etc.

Stenosis models

Above-mentioned accurate 1-D equation (equation 1-3) is derived by consider the hypothesis that a series of simplification are all set up for normal, healthy blood vessel.In the described hypothesis one is that axial velocity is preponderated and radial component is negligible.For example, for narrow, this hypothesis is no longer set up in the flip-flop situation in footpath, chamber, and radial component no longer can be excluded.Thereby accurate 1-D equation is not correctly caught across narrow pressure drop.

Research activities aspect formerly, the most concern pointed to the local velocity field, but for the FFR assessment only trans narrow pressure drop be important.In favourable embodiment, the semiempirical stenosis models can be included in the 1-D flow model, compares this with full Scale Model and has obtained accurate result.For example, in drag, pressure drop be represented as three (viscosity term, turbulent flow item or Bernoulli Jacob (Bernoulli) and Inertias) and:

$\mathrm{\Δ}{P}_s=\frac{\mathrm{\μ}{K}_v}{2\mathrm{\π}{r}_0^3}q+\frac{\mathrm{\ρ}{K}_t}{2A_0^2}{(\frac{A_0}{A_s}-1)}^2\left|q\right|q+\frac{\mathrm{\ρ}{K}_u{L}_s}{A_0}\frac{\∂q}{\∂t},---\left(4\right)$

Wherein μ is blood viscosity, L _sNarrow length, K _v, K _tAnd K _uRespectively viscosity, turbulent flow and inertia coeffeicent (refer to normal dimension with 0 time all amount of target, and refer to narrow value with target amount under the s).In favourable embodiment, combine with vascular tree (with bottom heart and crown bed model) for such semiempirical model of each narrow section (432 and 434), with calculate during the resting state and under maximum hyperemia both across narrow physiology pressure drop.Should be understood that, the invention is not restricted to the semiempirical stenosis models of equation (4), and replacedly can use other narrow these class models with a plurality of Pressure drop factors.In addition, in interchangeable embodiment, each narrow full rank 3D model can combine to simulate the pressure drop across narrow with remaining vascular tree.In this case, this the narrow patient-specific 3D geometric model that extracts from medical image (for example, CTA data) and the measure of similar quantification coronary angiography (QCA) use to be used in the stenosis models personalization of individual patient in combination.

About the coupling to remaining coronary vessel tree of depression of order or full rank stenosis models, in the first possible embodiment, momentum equation is fit to, and is used as additional loss item by the determined additional pressure drop of turbulent flow item and is included in equational right-hand side.In the second possible embodiment, regular momentum equation is fully ignored and is replaced by equation (2).The section that is regarded as narrow section is coupled to the regular section of coronary vessel tree by the seriality of considering gross pressure and flow rate.

The patient-specific modeling of crown bedside circle condition

An importance of flow simulation is represented by the boundary condition (outflow boundary condition) at the termination of coronary vessel tree.Usually, can exert pressure at the end website place of arterial trees, the relation between flow or flow and the pressure.If measured data for example time-varying speed, flow rate or even pressure be available, then they can easily be used.Do not have (it is this situation typically) under such information state, embodiments of the invention calculate the special boundary condition to the behavior modeling of distal artery section.Thereby, capillary bed by lump or the 0-D model come modeling: system's bed can be by the element that comprises different numbers (for example, between two and four elements) regular elastic cavity unit usually represent, and the coronary artery bed is represented by the certain moduli type, and described certain moduli type has illustrated the impact (during systole low and high during the diastole in early days) of myocardial contraction flow waveform.Fig. 4 has shown the example for this type of special purpose model of coronary circulation, and has presented the concrete element of such boundary condition.

Depend on the position of coronary arterial tree on heart, the key property of this type of lumped model is to consider myocardial contraction by introducing left ventricular pressure or right ventricular pressure.The model that shows among Fig. 4 is considered as individual unit with capillary bed, considers individually under the visceral pericardium and more special purpose model subendocardial capillary bed but can also utilize.Usually, blood vessel is less under the visceral pericardium is subject to heart contraction impact (skin of their expression cardiac muscles), and blood vessel is more vulnerable to outside action impact (their expressions are closer to the internal layer of ventricle) under the endocardium.This is the subendocardial reason that is easier to ischemia and is easier to myocardial infarction why.

Because it is very little that the Resistance Value of trunk is compared with resistance capillaceous with arteriole, so total stress level is almost determined by capillary bed separately.Under the background of Noninvasive FFR assessment, capillary bed usually and the coronary artery bed especially play another Main Function.Because FFR is based on congested lower determined value, so in order non-invasively to determine the value of this diagnostic indicators, Simulation of Blood must be to the congestive state modeling.In clinical practice, FFR is measured after the dispensing in the dispensing of the intravenous of vasodilation or coronary artery.In many angiopathys or the narrow situation of serial, the persistent period that importantly has the congestive state of raising is fallen curve after rise reliably in order to assess all narrow functional meanings and generate.Thereby, the usually intravenous of preferred vasodilation dispensing.This causes the slight raising of heart rate and the reduction of blood pressure.Because for simulation, the impact of vasodilation can ad infinitum be prolonged in the coronary artery, and obtains this congested replacement scheme and do not affect heart rate and blood pressure, therefore is easier to modeling, so the method enjoys expectation.Yet although the intravenous dispensing can simulated, all capillary beds must be by correspondingly adaptive.

The dispensing of congested induced drug (adenylic acid, papaverine etc.) causes the vasodilation effect of capillary bed, and this represents reducing in a large number of Resistance Value.The Resistance Value (for normal condition) of system or coronary artery lumped model inside can be from patient-specific be measured, obtain from data in literature or the non-linear relation between resistance and lumen size.Compliance plays a secondary role, and does not affect the interested average pressure of estimation of FFR because they only affect the transient state value.The coronary artery congestive state by as reduce to come modeling (having proved that visceral pericardium is that large artery trunks is not affected by vasodilation) by the correspondence of the caused blood capillary resistance of the dispensing of IC adenylic acid, and cause three to the five times raisings of normal coronary flow in healthy blood vessel.Coronary artery makes cardiac muscle avoid ischemia during automatically being adjusted in resting state, and cause the resistance that reduces of diseased vessel, reference value be must with without the identical flow of disease situation.Therefore normal condition can be by easily modeling, but does not represent very interested for the assessment of FFR.

Must estimative major parameter be mean arterial pressure (MAP) and coronary microvascular resistance (it is negligible that the resistance of near-end visceral pericardium tremulous pulse is compared with the blood capillary resistance).Because FFR uses the only average measurement of pressure (narrow far-end and near-end), so do not need accurately to estimate compliance, because they only print the waveform that affects pressure and flow, but do not affect only by the determined meansigma methods of resistance.MAP can non-invasively easily be measured, and as stated previously, MAP is similar under resting state and congestive state.The coronary microvascular resistance is low-down under hyperemia in order to allow the flow rate of raising on the other hand.In order to determine the Resistance Value under hyperemia, at first, can estimate repose resistance and then can estimate vasodilator drug dispensing impact and can estimate congested resistance.

According to favourable embodiment of the present invention, calculating for the patient-specific boundary condition of coronary artery bed was implemented in two stages: in the phase I, mean arterial pressure (MAP) and coronary microvascular resistance in each exit of patient-specific vascular tree during the resting state of simulation are estimated, and in second stage, be in congested blood capillary resistance and estimated.

Fig. 5 illustrates the method that is used for estimating according to an embodiment of the invention resting state blood capillary resistance.As illustrating among Fig. 5, at step 502 place, mean arterial pressure (MAP) is estimated based on patient's heart rate, heart contraction blood pressure, diastole blood pressure.Especially, MAP is calculated as

$\mathrm{MAP}=\mathrm{DBP}+[\frac{1}{3}+(\mathrm{HR}\·0.0012)]\·(\mathrm{SBP}-\mathrm{DBP}),---\left(5\right)$

Wherein HR, SBP and DBP represent respectively patient's heart rate, heart contraction blood pressure, diastole blood pressure, and it is by non-invasively measuring.

At step 504 place, total heart muscle perfusion q _RestUse speed-pressure product (RPP) relation to estimate.RPP is the product of heart rate and heart contraction blood pressure.From RPP, static perfusion q _RestCan be estimated as:

q _rest＝8·{[0.7·(HR·SBP)·10 ^-3]-0.4}[ml/min/100g]， (6)

Wherein, HR is heart rate and SBP is the heart contraction blood pressure.Can notice, this relation is effective in the oxygen demand situation that satisfies the subject that flows only.

At step 506 place, total static coronary flow is based on static perfusion q _RestEstimated with the quality of patient's left ventricle (LV).The quality of left ventricle is estimated based on the amount of cutting apart that derives from medical image.In a possible embodiment, cardiac muscle is cut apart the method for for example using based on the MSL machine learning with automatic ventricle and is cut apart.Volume can for example use in the U.S. Patent No. 8 of title for " Method and System for Measuring Left Ventricle Volume ", 098, method described in 918 is automatically calculated according to the cardiac muscle of cutting apart, the by reference combination of this patent.Then the LV volume is multiplied by density so that the quality (M of LV to be provided _LV).

In another possible embodiment, the volume of LV chamber can be confirmed as:

$V=\frac{4}{3}\·\mathrm{\π}\·\frac{d_a}{2}\·\frac{d_l}{2}\·\frac{l}{2}\left[\mathrm{ml}\right],---\left(7\right)$

D wherein _aAnd d _lBe two transverse diameters that calculate from two vertical planes, and l is a maximum chamber length of measuring in described vertical plane.Then the volume that calculates is proofreaied and correct by known regression equation:

V′＝0.928·V-3.8[ml] (8)

After having measured wall thickness, the volume of LV chamber and muscular wall can be calculated as follows:

$V_{c+w}=\frac{4}{3}\·\mathrm{\π}\·(\frac{d_a}{2}+h)\·(\frac{d_l}{2}+h)\·(\frac{l}{2}+h)\left[\mathrm{ml}\right]---\left(9\right)$

Then the LV quality can be calculated as:

M _LV＝(V _c+w-V′)·1.050[kg]， (10)

Wherein 1.050 represent myocardium proportion.

In another possible embodiment, the LV quality can be calculated as:

M _LV＝1.04·[(LVEDD+IVSEDD+PWEDD) ³-LVEDD ³]-13.6， (11)

Wherein LVEDD is left ventricle heart volume in easypro latter stage, and IVSEDD is ventricle inner septum heart diameter in easypro latter stage, and PWEDD is rear wall heart diameter in easypro latter stage.

In order to determine the absolute value of static flow, static perfusion must be multiply by myocardial mass.In normal heart, suppose that usually left ventricle represents 2/3rds of gross mass, and right ventricle and atrium represent to be left 1/3rd.Therefore, in case left ventricular mass M _LVBe determined, the absolute rest flow just can be confirmed as:

Q _rest＝q _rest·1.5·M _LV[ml/min]. (12).

If determined the cube proportional of flow rate and radius, then for all outflow flows of coronary vasodilator and the absolute rest flow can be represented as:

$Q_{\mathrm{rest}}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}k\·r_i^3=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{Q}_i---\left(13\right)$

At step 508 place, calculated the terminal resistance of each blood vessel.Especially, terminal resistance calculates with following relation:

$R_i=\frac{\mathrm{MAP}}{Q_i}.---\left(14\right)$

Q _iDetermined by following equation:

$\frac{Q_i}{Q_{\mathrm{rest}}}=\frac{k\·r_i^3}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}k\·r_j^3}=\frac{r_i^3}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{r}_j^3},---\left(15\right)$

Thus:

$Q_i=\frac{Q_{\mathrm{rest}}\·r_i^3}{\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{r}_j^3},---\left(16\right)$

R wherein _iThe tip radius that is blood vessel (equals end diameter d _iHalf) and n is the coefficient of impact.Therefore, the terminal resistance at each blood vessel place can be calculated as:

$R_i=\frac{\mathrm{MAP}}{Q_i}=\frac{\mathrm{MAP}\·\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{r}_j^3}{Q_{\mathrm{rest}}\·r_i^3}.\left(17\right)$

Estimate the second stage calculating congestive state blood capillary resistance of patient-specific coronary artery bedside circle condition.The static blood capillary resistance that the input of second stage is calculated by user's formula (5)-(17) represents, as described in the method for Fig. 5.Crown congestive state can be by reducing to come modeling by the caused blood capillary resistance of the dispensing of adenylic acid in the coronary artery.Proved that the visceral pericardium tremulous pulse is not affected by vasodilation, therefore only the blood capillary resistance need to be changed.This finally causes three to the five times raisings of coronary flow in healthy blood vessel.For normal, healthy subject (not having coronary artery disease), adenylic acid causes approximately the raising of 4.5 coronary artery flowing velocity.The CFVR value in coronary artery in pill, the coronary artery infusion or intravenous infusion be determined after giving a series of subjects' dispensing.In all three groups experiments, confirmed 4.5 value.The increase of coronary artery speed equals the increase of flow, because can suppose similar velocity profile (profile) for resting state and congestive state.Because between the stage of congestion, blood pressure reduces slightly, so 4.5 times of increases in the flow do not mean that 4.5 times in the coronary resistance reduce.Total coronary resistance index (TCRI) can calculate as described below.

Congested blood capillary resistance can be calculated as follows.At first, static average peak speed is calculated as based on patient's heart rate and heart contraction blood pressure:

rAPV＝0.0009·SBP·HR+5.925[cm/s]， (18)

Its medium-rate-pressure product represents with [mmHg* heart beating/min].Then can come to calculate the CFVR value for each medial fascicle in the coronary vessel tree (mean branch) with following equation:

LAD：CFVR＝10 ^{1.16-0.48·log(rAPV)-0.0025·age}， (19)

LCX：CFVR＝10 ^{1.14-0.45·log(rAPV)-0.0031·age}， (20)

RCA：CFVR＝10 ^{1.15-0.50·log(rAPV)-0.0021·age}. (21)

Then can calculate TCRI with the value of 5mmHg for Δ MAP:

$\frac{1}{\mathrm{TCRI}}=\frac{\frac{{\mathrm{MAP}}_{\mathrm{rest}}}{Q_{\mathrm{rest}}}}{\frac{{\mathrm{MAP}}_{\mathrm{hyper}}}{Q_{\mathrm{hyper}}}}=\frac{{\mathrm{MAP}}_{\mathrm{rest}}}{{\mathrm{MAP}}_{\mathrm{hyper}}}\·\frac{Q_{\mathrm{hyper}}}{Q_{\mathrm{rest}}}=\frac{{\mathrm{MAP}}_{\mathrm{erst}}}{{\mathrm{MAP}}_{\mathrm{hyper}}}\·\mathrm{CFVR}=\frac{{\mathrm{MAP}}_{\mathrm{rest}}}{{\mathrm{MAP}}_{\mathrm{rest}-\mathrm{\ΔMAP}}}\·\mathrm{CFVR}---\left(22\right)$

Replacedly, replace above-mentioned steps, below relation can be used to determine the TCRI value based on having the result of the test that low-down standard deviation has high reliability thus:

Then calculate congested blood capillary resistance with following equation based on resting state blood capillary resistance:

(R _i) _hyper＝(R _i) _rest·TCRI (24)

(R wherein _i) _RestIt is the value for the determined resting state blood capillary of the method resistance that uses above-mentioned Fig. 6.Should be understood that, be used for estimating that the interchangeable method of the blood capillary resistance under hyperemia can also easily be merged to be used to the method for calculating FFR as herein described.

Heart model

IMP power is an important element of coronary artery modeling.Thereby the critical piece of Reduced Order Modeling is heart model.Fig. 4 has shown lump heart model 402, but can also use more complicated and complete model.There are several lumped models, such as different elastic models and ultimate fibre model.These can determine pressure and flow in the different ventricles in the situation of the spatial model that need not consider heart.Several parameters for example contractility, stroke volume, to peaked time, dead volume (dead volume) (V ₀), heart rate can be adapted in order to the different conditions of health is described, and in order to makes model personalized.The simplest model is represented by different elastic models, described different elastic model can easily be coupled to the aorta input by lump aortic valve model 402, and indirectly is coupled to the special-purpose blood capillary model of coronary arterial tree by left ventricular pressure.Different elastic models can be represented as:

$E\left(t\right)=\frac{P_{\mathrm{LV}}\left(t\right)}{V_{\mathrm{LV}}\left(t\right)-V_0}.---\left(25\right)$

Several consider that items have caused phylogenetic tree but not the only modeling of all major arteries of coronary arterial tree.Like this, heart can directly be coupled to aorta, and flow can be determined by the interaction between left ventricle and the system impedance.Similarly, total stress level is mainly determined by large artery trunks, and coronary resistance (based on blood capillary and narrow) has negligible impact, and trans narrow pressure drop can be by more accurately modeling thus.Depend on obtainable additional data (such as ultrasoundcardiogram, cardiac MRI), can make further heart model personalized for individual patient.These patterns allow easily to be used to make heart model Extraordinary information, such as stroke volume, ejection fraction etc.

Turn back to Fig. 2, at step 208 place, based on Simulation of Blood for each narrow FFR that calculates.In case under maximum vasodilation according to the analog computation of patient-specific depression of order the time pressure changeable and flow rate, the FFR value just the average pressure (Pd) by getting simply narrow far-end with respect to mean aortic pressure power (P _a) ratio during cardiac cycle determines:

$\mathrm{FFR}=\frac{P_d}{P_a}.---\left(26\right)$

Can during the anatomy modeling procedure, all damages for user's appointment automatically carry out this calculating.In addition, the user can also specify any position in the vascular tree during post-processing step, and then will be as calculating as described above corresponding FFR value.

Fig. 6 illustrates the calculating of the FFR that uses according to an embodiment of the invention the Extraordinary reduced-order model.As shown in Figure 6, at step 602 place, adenylic acid is estimated the impact of terminal resting state blood capillary resistance, produces terminal congested blood capillary resistance.At 604 places, depression of order is simulated and is used heart model, coronary vasodilator geometry (geometry), is in estimated resistance and the stenosis models of hyperemia and carries out.The congested blood flow of described analogsimulation and congested pressure.At 606 places, FFR is calculated as the average congested pressure (P that simulates of narrow far-end _d) and average congested aortic pressure (P _a) ratio in cardiac cycle.Except FFR, can also calculate according to the result of Simulation of Blood other hematodinamics amounts based on flow rate and pressure.

Except coronary artery CT data, can also use said method to other view data, described other view data are such as 3-D angiography (Angio), rotational angiography, Dyna-CT.For angiographic data, static and maximum congested during both, the time and space of propagating via radiography represents that the image-based analysis of the propagation of contrast agent can be used to As time goes on regain flow rate robust.Simultaneously, such image acquisition can also be performed during the congested condition of maximum, and is used in combination determining the FFR value with the experience stenosis models.Therefore, in interchangeable embodiment of the present invention, medical image can be collected under hyperemia, and detected congested boundary condition is directly based on the view data that gathers under hyperemia and the non-imaging measurement of Noninvasive (for example, heart rate, heart contraction blood pressure and diastole blood pressure).Use the simulation of congested boundary condition then can be used to calculate FFR.

Other sources (when existing) of patient information can be used to the further personalization of model.For example, angiography can provide the 3-D strain figure, its can be used on heart contraction in the epicardial coronary arteries blood vessel each affect modeling.Between the right side of heart and left side, there is Main Differences, but can considers equally local more detailed variability.The 3-D strain figure that extracts from view data can be used to apply additional patient-specific boundary condition.Along identical line, 3-D colorful blood or phase-contrast MRI measure and can also be utilized for the coronary vessel tree blood flow boundary condition that provides access.The availability of the 3-D+t anatomic model of heart (from CT, MR or ultrasound data) makes it simulate blood flow in the ventricle by CFD.This can also be used to further make the model personalization and apply boundary condition.

Because the FFR value is calculated closely in real time with the depression of order analogue model, so the clinician can provide feedback observing the effect of various changes, the variability that described various changes cause such as the change owing to segmentation result, owing to being used for the change of seed points of central point extraction and the variability that lacks variability that centrage causes, the variability that causes owing to the change of the position of branches end and variability that the impact on the boundary condition causes, owing to the elimination of vascellum laterale branch and cause owing to total picture quality.

Replace using as described above speed-pressure product in order to determine coronary perfusion, in interchangeable embodiment, can applied stress-quality-speed product in order to directly determine overall coronary flow:

$\mathrm{Stress}=\frac{0.334\·\mathrm{SBP}\·\mathrm{LVDd}}{\mathrm{LVPWT}}\·(1+\frac{\mathrm{LVPWT}}{\mathrm{LVDd}}),---\left(27\right)$

Wherein SBP is cardiac systolic pressure, and LVDd is at ED left ventricle diameter, and LVPWT is left ventricular posterior wall thickness.Then the static flow of overall situation coronary artery can be determined as follows:

Q _Rest=0.0218StressM _LVHR10 ^-3+ 120.11, (28) wherein stress-quality-speed product represent with [g*k dyne (dyne)/cm2* heart beating/min], and MLV represents left ventricular mass.

The method of the non-invasive evaluation that is used for coronary stricture mentioned above can realize with well-known computer processor, memory cell, memory device, computer software and other parts on computers.The high level block diagram of such computer is shown among Fig. 7.Computer 702 comprises processor 704, and described processor 704 is controlled the overall operation of computer 702 by carrying out the computer program instructions that limits such operation.Computer program instructions can be stored in the memory device 712 (for example, disk), and is loaded in the memorizer 710 when the instruction of needs computer program.Therefore, the step of the method for Fig. 2 can be defined by the computer program instructions that is stored in memorizer 710 and/or the bin 712, and is controlled by the processor 704 of computer program instruction.Image capture device 720 such as CT scan equipment, MR scanning device, ultrasonic equipment etc. can be connected to computer 702 view data is input to computer 702.Image capture device 720 and computer 702 are realized it being possible as an equipment.Image capture device 720 and computer 702 by network wireless to communicate by letter also be possible.Computer 702 also comprises the one or more network interfaces 706 that communicate for via network and other equipment.Computer 702 also comprises so that can realize other mutual input-output apparatus 708 (for example, display, keyboard, mouse, speaker, button etc.) of user and computer 702.This type of input-output apparatus 708 can combine with one group of computer program and use the volume that receives from image capture device 720 to explain as annotation tool (annotation tool).Person of skill in the art will appreciate that, the embodiment of actual computer can also comprise miscellaneous part, and Fig. 7 is the senior expression for some parts of such computer of illustrative purpose.

Aforesaid specific descriptions should be interpreted as it is being illustrative and exemplary aspect each, but not restrictive, and scope of the present invention disclosed herein be can't help to specifically describe institute and is determined, and is determined by the claim that the complete width of permitting according to Patent Law is explained.Should be understood that, the embodiment that illustrates and describe at this paper only illustrates principle of the present invention, and those skilled in the art can realize various modifications in the situation that do not deviate from scope and spirit of the present invention.Those skilled in the art can realize various other characteristics combination in the situation that do not deviate from scope and spirit of the present invention.

Claims (30)

1. method that is used for the non-invasive evaluation of coronary stricture comprises:

From the patient's that during resting state, gathers medical image, extract described patient-specific anatomic measurement coronarius;

Calculate the patient-specific resting state boundary condition of the model of the described coronary circulation coronarius of expression based on the described patient-specific anatomic measurement that is in static described patient and Noninvasive clinical measurement;

Based on described silent boundary condition be used for the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation;

Simulate congested blood flow and pressure across at least one at least one stenosis area coronarius with the model of described coronary circulation and the congested boundary condition of described patient-specific; And

The blood flow reserve mark (FFR) that calculates described at least one stenosis area based on congested blood flow and the pressure of described simulation.

2. method according to claim 1, wherein, described coronary circulation model comprises described coronary artery and the aortal one dimension computation model that represents described patient.

3. method according to claim 2, wherein, described coronary circulation model further comprises the depression of order semiempirical stenosis models that represents described at least one stenosis area, and described depression of order semiempirical stenosis models is coupled to described at least one the described one dimension computation model coronarius of expression.

4. method according to claim 3, wherein, described semiempirical stenosis models will across the pressure drop of described at least one stenosis area as viscosity term, turbulent flow item and Inertia and calculate.

5. method according to claim 2, wherein, described coronary circulation model further comprises the full rank three-dimensional computations model of described at least one stenosis area, and described full rank three-dimensional computations model is coupled to described at least one the described one dimension computation model coronarius of expression.

6. method according to claim 2, wherein, described coronary circulation model further comprises the lumped model of expression coronary microvascular bed, each coronary microvascular bed is coupled to the end of the one dimension computation model of expression coronary arterial tree.

7. method according to claim 1, wherein, the patient-specific resting state boundary condition that calculates the model of the described coronary circulation coronarius of expression based on the described patient-specific anatomic measurement that is in static described patient and Noninvasive clinical measurement comprises:

Calculate each the static blood capillary resistance of end in described a plurality of branches based on the described patient-specific anatomic measurement that is in static described patient and described Noninvasive clinical measurement.

8. method according to claim 7, wherein, based on described silent boundary condition be used for the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation and comprise:

Calculate each the described congested blood capillary resistance of end in described a plurality of branches based on the static blood capillary resistance of described calculating.

9. method according to claim 8, wherein, each the static blood capillary resistance of end that calculates in described a plurality of branches based on the described patient-specific anatomic measurement that is in static described patient and described Noninvasive clinical measurement comprises:

Heart rate, heart contraction blood pressure and diastole blood pressure based on described patient calculate mean arterial pressure (MAP);

Described heart rate and described heart contraction blood pressure based on described patient calculate static heart muscle perfusion;

Calculate total static coronary flow based on described patient's described static heart muscle perfusion and the quality of described left ventricle; And

Calculate each the described static blood capillary resistance of end in described a plurality of branches based on described MAP and described total static coronary flow.

10. method according to claim 9, wherein, the described quality of described left ventricle is estimated from described patient's described medical image.

11. method according to claim 8, wherein, each the described congested blood capillary resistance of end that calculates in described a plurality of branches based on the static blood capillary resistance of described calculating comprises:

Described Noninvasive clinical measurement based on described patient is determined the total coronary resistance index of patient-specific (TCRI); And

Calculate each the described congested blood capillary resistance of end in described a plurality of branches based on the congested blood capillary resistance of described calculating and described patient-specific TCRI.

12. method according to claim 11 wherein, determines that based on described patient's described Noninvasive clinical measurement the total coronary resistance index of patient-specific (TCRI) comprising:

Heart rate and heart contraction blood pressure based on described patient calculate static average peak speed;

Calculate the CFVR value of described each branch coronarius based on described patient's age; And

Calculate the described TCRI of each branch based on the static mean arterial pressure (MAP) of described CFVR value and estimation.

13. method according to claim 11 wherein, determines that based on described patient's described Noninvasive clinical measurement the total coronary resistance index of patient-specific (TCRI) comprising:

Heart rate based on described patient calculates described TCRI.

14. method according to claim 1, wherein, the blood flow reserve mark (FFR) that calculates described at least one stenosis area based on congested blood flow and the pressure of described simulation comprising:

Ratio in cardiac cycle calculates as the congested pressure of the averaging analog of described at least one narrow far-end and the congested aortic pressure of averaging analog with the described FFR of described at least one stenosis area.

15. an equipment that is used for the non-invasive evaluation of coronary stricture comprises:

Be used for extracting from the patient's that during resting state, gathers medical image the device of described patient-specific anatomic measurement coronarius;

Be used for calculating based on the described patient-specific anatomic measurement that is in static described patient and Noninvasive clinical measurement the device of patient-specific resting state boundary condition of the model of the described coronary circulation coronarius of expression;

Be used for based on described silent boundary condition and be used for the device of the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation;

Be used for simulating across the congested blood flow of at least one at least one stenosis area coronarius and the device of pressure with model and the congested boundary condition of described patient-specific of described coronary circulation; And

The device that is used for the blood flow reserve mark (FFR) that congested blood flow and pressure based on described simulation calculates described at least one stenosis area.

16. equipment according to claim 15, wherein, described coronary circulation model comprises described coronary artery and the aortal one dimension computation model that represents described patient.

17. equipment according to claim 15, wherein, the described device of patient-specific resting state boundary condition that is used for calculating based on the described patient-specific anatomic measurement that is in static described patient and Noninvasive clinical measurement the model of the described coronary circulation coronarius of expression comprises:

Be used for calculating the device of static blood capillary resistance of end in each of described a plurality of branches based on the described patient-specific anatomic measurement that is in static described patient and described Noninvasive clinical measurement.

18. equipment according to claim 17 wherein, is used for based on described silent boundary condition and is used for the described device of the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation comprising:

Be used for static blood capillary resistance based on described calculating and calculate the device of described congested blood capillary resistance of end in each of described a plurality of branches.

19. equipment according to claim 18, wherein, be used for comprising based on each the described device of static blood capillary resistance of end that the described patient-specific anatomic measurement that is in static described patient and described Noninvasive clinical measurement calculate in described a plurality of branches:

Be used for calculating based on described patient's heart rate, heart contraction blood pressure and diastole blood pressure the device of mean arterial pressure (MAP);

Be used for the device that described heart rate and described heart contraction blood pressure based on described patient calculate static heart muscle perfusion;

Be used for calculating based on the quality of described patient's described static heart muscle perfusion and described left ventricle the device of total static coronary flow; And

Be used for calculating the device of described static blood capillary resistance of end in each of described a plurality of branches based on described MAP and described total static coronary flow.

20. equipment according to claim 18 wherein, is used for each the described device of described congested blood capillary resistance of end that static blood capillary resistance based on described calculating calculates in described a plurality of branches and comprises:

Be used for determining based on described patient's described Noninvasive clinical measurement the device of the total coronary resistance index of patient-specific (TCRI); And

Be used for calculating the device of described congested blood capillary resistance of end in each of described a plurality of branches based on the congested blood capillary resistance of described calculating and described patient-specific TCRI.

21. a storage is used for the nonvolatile computer-readable medium of computer program instructions of the non-invasive evaluation of coronary stricture, described computer program instructions makes described processor executable operations when being carried out by processor, and described operation comprises:

From the patient's that during resting state, gathers medical image, extract described patient-specific anatomic measurement coronarius;

Calculate the patient-specific resting state boundary condition of the model of the described coronary circulation coronarius of expression based on the described patient-specific anatomic measurement that is in static described patient and Noninvasive clinical measurement;

Based on described silent boundary condition be used for the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation;

Simulate congested blood flow and pressure across at least one at least one stenosis area coronarius with the model of described coronary circulation and the congested boundary condition of described patient-specific; And

The blood flow reserve mark (FFR) that calculates described at least one stenosis area based on congested blood flow and the pressure of described simulation.

22. nonvolatile computer-readable medium according to claim 21, wherein, described coronary circulation module comprises described coronary artery and the aortal one dimension computation model that represents described patient.

23. nonvolatile computer-readable medium according to claim 21, wherein, the patient-specific resting state boundary condition that calculates the model of the described coronary circulation coronarius of expression based on the described described patient-specific anatomic measurement that is in static described patient and Noninvasive clinical measurement comprises:

Calculate each the static blood capillary resistance of end in described a plurality of branches based on the described patient-specific anatomic measurement that is in static described patient and described Noninvasive clinical measurement.

24. nonvolatile computer-readable medium according to claim 23, wherein, based on described silent boundary condition be used for the congested boundary condition of patient-specific that the congested model of simulation calculates the model of described coronary circulation and comprise:

Calculate each the described congested blood capillary resistance of end in described a plurality of branches based on the static blood capillary resistance of described calculating.

25. nonvolatile computer-readable medium according to claim 24, wherein, each the static blood capillary resistance of end that calculates in described a plurality of branches based on the described patient-specific anatomic measurement that is in static described patient and described Noninvasive clinical measurement comprises:

Heart rate, heart contraction blood pressure and diastole blood pressure based on described patient calculate mean arterial pressure (MAP);

Described heart rate and described heart contraction blood pressure based on described patient calculate static heart muscle perfusion;

Calculate total static coronary flow based on described patient's described static heart muscle perfusion and the quality of described left ventricle; And

Calculate each the described static blood capillary resistance of end in described a plurality of branches based on described MAP and described total static coronary flow.

26. nonvolatile computer-readable medium according to claim 24, wherein, each the described congested blood capillary resistance of end that calculates in described a plurality of branches based on the static blood capillary resistance of described calculating comprises:

Described Noninvasive clinical measurement based on described patient is determined the total coronary resistance index of patient-specific (TCRI); And

Calculate each the described congested blood capillary resistance of end in described a plurality of branches based on the congested blood capillary resistance of described calculating and described patient-specific TCRI.

27. nonvolatile computer-readable medium according to claim 26 wherein, determines that based on described patient's described Noninvasive clinical measurement the total coronary resistance index of patient-specific (TCRI) comprising:

Heart rate and heart contraction blood pressure based on described patient calculate static average peak speed;

Calculate the CFVR value of described each branch coronarius based on described patient's age; And

Calculate the described TCRI of each branch based on the static mean arterial pressure (MAP) of described CFVR value and estimation.

28. nonvolatile computer-readable medium according to claim 26 wherein, determines that based on described patient's described Noninvasive clinical measurement the total coronary resistance index of patient-specific (TCRI) comprising:

Heart rate based on described patient calculates described TCRI.

29. nonvolatile computer-readable medium according to claim 21, wherein, the blood flow reserve mark (FFR) that calculates described at least one stenosis area based on congested blood flow and the pressure of described simulation comprising:

Ratio in cardiac cycle calculates as the congested pressure of the averaging analog of described at least one narrow far-end and the congested aortic pressure of averaging analog with the described FFR of described at least one stenosis area.

30. a method that is used for the non-invasive evaluation of coronary stricture comprises:

From the patient's that during congestive state, gathers medical image, extract described patient-specific anatomic measurement coronarius;

Calculate the congested boundary condition of patient-specific of the model of the described coronary circulation coronarius of expression based on the described patient-specific anatomic measurement that is in congested described patient and Noninvasive clinical measurement;

Simulate congested blood flow and pressure across at least one at least one stenosis area coronarius with the model of described coronary circulation and the congested boundary condition of described patient-specific; And

The blood flow reserve mark (FFR) that calculates described at least one stenosis area based on congested blood flow and the pressure of described simulation.

Images