CN109919916A

CN109919916A - A kind of wall shear stress optimization method and device, storage medium

Info

Publication number: CN109919916A
Application number: CN201910127951.8A
Authority: CN
Inventors: 魏润杰; 王洪平; 高琪; 吴鹏
Original assignee: Hangzhou Sheng Shi Technology Co Ltd
Current assignee: Hangzhou Sheng Shi Technology Co Ltd
Priority date: 2019-02-20
Filing date: 2019-02-20
Publication date: 2019-06-21
Anticipated expiration: 2039-02-20
Also published as: CN109919916B

Abstract

The embodiment of the invention discloses a kind of wall shear stress optimization method and devices, storage medium, this method comprises:, according to blood vessel nuclear magnetic resonance data, constructing blood vessel three-dimensional model when getting blood vessel nuclear magnetic resonance data；According to blood vessel nuclear magnetic resonance data and blood vessel three-dimensional model, initial vessel velocity field is obtained；Based on conservation of mass condition, initial vessel velocity field is optimized, obtains optimization vessel velocity field；According to blood vessel three-dimensional model and optimization vessel velocity field computation wall shear stress, blood vessel wall surface shear Stress Distribution information is obtained.

Description

A kind of wall shear stress optimization method and device, storage medium

Technical field

The present invention relates to the shearing stress computing technique in biomedical engineering technology field more particularly to a kind of wall shear stress Optimization method and device, storage medium.

Background technique

Clinical treatment diagnosis in, color ultrasound imaging, magnetic resonance imaging (Magnetic Resonance Imaging, MRI) and Digital subtraction angiography etc. is all the medical imaging techniques for being commonly used in measuring blood flow rate, these medical imaging techniques are logical It crosses different principle and perspective imaging is carried out to human body, obtain the gray level image information of characterization human body inner tissue organ, while can capture The measurement of blood of human body flowing information realization blood flow velocity.Doctor is by gray level image information and blood flow information to patient's Case is analyzed to provide diagnostic comments.Therefore, blood flow quantitative target pair is accurately obtained by medical imaging techniques Medical diagnosis is very important.And in numerous blood flow quantitative targets, wall shear stress (wall shear Stress, WSS) in the development and formation of vascular diseases play important role.

In the prior art, the acquisition of wall shear stress is usually to obtain blood flow information first with medical imaging techniques, Wall shear stress is obtained by analysis blood flow information again.However, during above-mentioned acquisition wall shear stress, due to By various restrictions such as image-forming principle, shadowgraph technique and hardware device performances, the blood flow information got, which exists, makes an uproar The problems such as sound is big, velocity field bad point is more, spatial and temporal resolution is low causes the wall surface got by analysis blood flow information to be cut and answers Power inaccuracy.

Summary of the invention

In order to solve the above technical problems, an embodiment of the present invention is intended to provide a kind of wall shear stress optimization method and device, Storage medium can be improved the accuracy of wall shear stress.

The technical scheme of the present invention is realized as follows:

In a first aspect, the embodiment of the invention provides a kind of wall shear stress optimization methods, which comprises

When getting blood vessel nuclear magnetic resonance data, according to the blood vessel nuclear magnetic resonance data, blood vessel three-dimensional model is constructed；

According to the blood vessel nuclear magnetic resonance data and the blood vessel three-dimensional model, initial vessel velocity field is obtained；

Based on conservation of mass condition, the initial vessel velocity field is optimized, obtains optimization vessel velocity field；

According to the blood vessel three-dimensional model and the optimization vessel velocity field computation wall shear stress, obtains blood vessel wall surface and cut Stress distribution information.

In the above scheme, described according to the blood vessel nuclear magnetic resonance data, construct blood vessel three-dimensional model, comprising:

Obtain the initial blood vessel 3-D image in the blood vessel nuclear magnetic resonance data；

Denoising is carried out to the initial blood vessel 3-D image, the initial blood vessel 3-D image after being denoised；

Vessel information extraction is carried out to the initial blood vessel 3-D image after the denoising using default partitioning algorithm, is obtained a little Cloud blood vessel 3-D image；

Described cloud blood vessel 3-D image is reconstructed using default restructing algorithm, obtains the blood vessel three-dimensional model.

In the above scheme, described according to the blood vessel nuclear magnetic resonance data and the blood vessel three-dimensional model, it obtains initial Vessel velocity field, comprising:

Obtain the vessel velocity field in the blood vessel nuclear magnetic resonance data；

Interpolation is carried out to the vessel velocity field according to default spacing, the vessel velocity field after obtaining interpolation；

On the blood vessel three-dimensional model, according to vessel borders to the velocity node in the vessel velocity field after the interpolation It is marked, obtains the initial vessel velocity field.

In the above scheme, described to be based on conservation of mass condition, the initial vessel velocity field is optimized, is obtained excellent Change vessel velocity field, comprising:

According to wall surface non-slip condition, the corresponding derivative information in the initial vessel velocity field is determined；

Based on the conservation of mass condition and the derivative information, the corresponding blood vessel speed in the initial vessel velocity field is constructed Spend field optimization method；

Based on vessel velocity field optimization method, default cross validation algorithm and default optimization algorithm, determine that optimization is flat Sliding parameter；

The optimization smoothing parameter is input to vessel velocity field optimization method, obtains the optimization vessel velocity ?.

In the above scheme, described according to the wall surface non-slip condition, determine that the initial vessel velocity field is corresponding Derivative information, comprising:

Calculate the wall surface distance of the velocity node in the initial vessel velocity field；

According to default spacing and the wall surface distance, the velocity node in the initial vessel velocity field is divided into nearly wall Millet cake and remote wall surface point；

By wall surface distance, the wall surface non-slip condition and the first default derivative algorithm to the near wall point based on Derivative is calculated, corresponding first derivative information is obtained；

By wall surface distance, the wall surface non-slip condition and the second default derivative algorithm to the remote wall surface point based on Derivative is calculated, corresponding flection information is obtained；

Using first derivative information and the flection information as the derivative information.

In the above scheme, described to be calculated based on the wall surface distance, the wall surface non-slip condition and the first default derivative Method calculates derivative to the near wall point, obtains corresponding first derivative information, comprising:

According to the wall surface distance, the vascular wall point and nearest velocity node of the near wall point are determined；

According to the wall surface distance, vascular wall point and the nearest velocity node and the wall surface without sliding item Part and the first default derivative algorithm obtain the corresponding first derivative of the near wall point and second dervative, and described first is pre- If derivative algorithm is used to determine the derivative information of the near wall point；

Using the first derivative and the second level derivative as first derivative information.

In the above scheme, described to be based on the conservation of mass condition and the derivative information, construct the initial blood vessel The corresponding vessel velocity field optimization method of velocity field, comprising:

According to the derivative information, the corresponding smoothness in initial vessel velocity field and discrete divergence are determined；

According to the smoothness, the discrete divergence and the conservation of mass condition, the initial vessel velocity is constructed The corresponding initial vessel velocity field optimization object function in field；

According to Lagrange multiplier algorithm, the initial vessel velocity field optimization object function is converted, is turned Initial vessel velocity field optimization object function after changing；

Minimum processing is carried out to the initial vessel velocity field optimization object function after the conversion, obtains the blood vessel speed Spend field optimization method.

In the above scheme, described to be based on vessel velocity field optimization method, default cross validation algorithm and preset excellent Change algorithm, determine optimization smoothing parameter, comprising:

According to the default cross validation algorithm, smooth verification information is constructed；

Based on the default optimization algorithm, at least one initial smoothing parameter is determined；

At least one described initial smoothing parameter is separately input into vessel velocity field optimization method, is obtained corresponding At least one initial optimization vessel velocity field；

According at least one described initial optimization vessel velocity field, determine the smooth verification information it is corresponding at least one Smooth validation value；

By the corresponding initial smoothing parameter of minimum smooth validation value at least one described smooth validation value, as described excellent Change smoothing parameter.

In the above scheme, described at least one initial optimization vessel velocity field according to, determines the smooth verifying At least one corresponding smooth validation value of information, comprising:

According to the derivative information, the coefficient information in the smooth verification information is determined；

According to default fitting algorithm, the corresponding characteristic value of the coefficient information is determined；

The characteristic value is input to the smooth verification information, obtains the corresponding smooth verification information of simplification；

At least one described initial optimization vessel velocity field is separately input into the smooth verification information of the simplification, obtains institute State at least one smooth validation value.

In the above scheme, described to cut and answer according to the blood vessel three-dimensional model and the optimization vessel velocity field computation wall surface Power obtains blood vessel wall surface shear Stress Distribution information, comprising:

It is enterprising in the wall surface normal orientation of the blood vessel three-dimensional model according to the optimization vessel velocity field and default spacing Row interpolation obtains interpolation speed；

According to the interpolation speed and the wall surface normal orientation, determine that blood flows to；

It is flowed to according to the interpolation speed and the blood, obtains corresponding matched curve；

According to blood flow flow direction, the matched curve and default hemodynamics viscosity, obtain the wall surface and cut to answer Power distributed intelligence.

In the above scheme, described to be flowed to according to the interpolation speed and the blood, corresponding matched curve is obtained, is wrapped It includes:

The interpolation speed projection information upward in the blood stream is calculated, corresponding velocity profile information is obtained；

The velocity profile information is fitted, the matched curve is obtained.

In the above scheme, described that the velocity profile information is fitted, obtain the matched curve, comprising:

Obtain blood attribute information；

According to the blood attribute information, at least one default blood vessel wall surface friction velocity and default fitting coefficient, and Default fitting algorithm, is fitted the velocity profile information, obtains corresponding at least one initial matched curve, described default Fitting algorithm is for being fitted the velocity profile information；

Determine at least one described initial matched curve at least one difference corresponding with the velocity profile information respectively Information；

The corresponding initial matched curve of the smallest distinctive information will be distinguished at least one described distinctive information as described in Matched curve.

In the above scheme, the default fitting algorithm are as follows:

Wherein, V ' is the corresponding continuous dependent variable value of the initial matched curve, u_τIt rubs for the default blood vessel wall surface Speed is wiped, W ' is the height of the corresponding wall surface normal direction of speed each in the velocity profile information, and κ and s are the default fitting Coefficient, ρ are the density of the blood in the blood attribute information, and μ is the dynamic viscosity of the blood in the blood attribute information Coefficient.

Second aspect, the embodiment of the invention provides a kind of wall shear stress to optimize device, and described device includes: processing Device, memory and communication bus, the memory are communicated by the communication bus with the processor, the memory The executable program of the processor is stored, when described program is performed, is executed by the processor as described above Wall shear stress optimization method.

The third aspect, the embodiment of the invention provides a kind of computer readable storage mediums, are stored thereon with program, described Wall shear stress optimization method as described above is realized when program is executed by processor.

The embodiment of the invention provides a kind of wall shear stress optimization method and devices, storage medium, firstly, when getting When blood vessel nuclear magnetic resonance data, according to blood vessel nuclear magnetic resonance data, blood vessel three-dimensional model is constructed: secondly, total according to blood vessel nuclear-magnetism Data of shaking and blood vessel three-dimensional model, obtain initial vessel velocity field；Then, it is based on conservation of mass condition, to initial vessel velocity Field optimizes, and obtains optimization vessel velocity field；Finally, being cut according to blood vessel three-dimensional model and optimization vessel velocity field computation wall surface Stress obtains blood vessel wall surface shear Stress Distribution information.Using above-mentioned technic relization scheme, since optimization vessel velocity field is wall surface Shearing stress optimization device optimizes initial vessel velocity field by conservation of mass condition, and because conservation of mass item Part can ensure to optimize the flatness of vessel velocity field and without scattered property, and therefore, the blood flow informations such as optimization vessel velocity field are made an uproar Sound is small, velocity field bad point is few, spatial and temporal resolution is high, thus when wall shear stress optimization device is according to optimization vessel velocity field computation When wall shear stress, the accuracy of wall shear stress is improved.

Detailed description of the invention

Fig. 1 is a kind of wall shear stress optimization method implementation flow chart provided in an embodiment of the present invention；

Fig. 2 is a kind of illustrative blood vessel wall surface shear Stress Distribution information provided in an embodiment of the present invention；

Fig. 3 is a kind of schematic diagram of illustrative speed grid provided in an embodiment of the present invention；

Fig. 4 is a kind of schematic diagram of illustrative determining matched curve provided in an embodiment of the present invention；

Fig. 5 is a kind of blood vessel wall surface shear Stress Distribution illustratively calculated based on formula (9) provided in an embodiment of the present invention Information；

Fig. 6 is that a kind of blood vessel wall surface illustratively calculated based on wall surface second order polynomial provided in an embodiment of the present invention is cut Stress distribution information；

Fig. 7 is the structural schematic diagram one that a kind of wall shear stress provided in an embodiment of the present invention optimizes device；

Fig. 8 is the structural schematic diagram two that a kind of wall shear stress provided in an embodiment of the present invention optimizes device.

Specific embodiment

Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description.

Embodiment one

The embodiment of the invention provides a kind of wall shear stress optimization method, Fig. 1 is one kind provided in an embodiment of the present invention Wall shear stress optimization method implementation flow chart, as shown in Figure 1, the wall shear stress optimization method includes:

S101, when getting blood vessel nuclear magnetic resonance data, according to blood vessel nuclear magnetic resonance data, construct blood vessel three-dimensional mould Type.

In embodiments of the present invention, when carrying out the measuring blood flow rate of blood vessel, by nuclear magnetic resonance equipment to blood vessel institute Nuclear magnetic resonance is carried out at position, obtains corresponding blood vessel nuclear magnetic resonance data；And works as and blood vessel nuclear magnetic resonance data is input to wall When face shearing stress optimizes device, wall shear stress device has got blood vessel nuclear magnetic resonance data；At this point, due to blood vessel nuclear-magnetism It include initial blood vessel 3-D image in resonance data, therefore, wall shear stress device is isolated from blood vessel nuclear magnetic resonance data Initial blood vessel 3-D image, and image procossing is carried out to the initial blood vessel 3-D image, to construct blood only comprising blood vessel Pipe threedimensional model.

It should be noted that blood vessel nuclear magnetic resonance data characterization carries out the primary data obtained after nuclear magnetic resonance to blood vessel, Such as: the temporal resolution of nuclear magnetic resonance image is 49.2ms, the pixel size of initial blood vessel 3-D image be 256*25*80 and Single voxel size is 0.8*0.8*1.5mm³, nuclear magnetic resonance technique is 4D (Dimensional, dimension) flow MRI technique, 3D velocity field (vessel velocity field), etc..In addition, here, blood vessel three-dimensional model characterizes the 3-D geometric model of blood vessel, than Such as, the 3-D geometric model of aorta.

S102, according to blood vessel nuclear magnetic resonance data and blood vessel three-dimensional model, obtain initial vessel velocity field.

In embodiments of the present invention, due to including vessel velocity field in blood vessel nuclear magnetic resonance data, wall shear stress Device isolates vessel velocity field from blood vessel nuclear magnetic resonance data, and carries out on blood vessel three-dimensional model to the vessel velocity field Pretreatment, to obtain initial vessel velocity field.

It should be noted that initial vessel velocity field characterization is by pretreatment and can be used in carrying out the velocity field of blood vessel The velocity field of the blood vessel of optimization.

S103, it is based on conservation of mass condition, initial vessel velocity field is optimized, obtain optimization vessel velocity field.

In embodiments of the present invention, wall shear stress measures the flowing of vessel inner blood by nuclear magnetic resonance technique, belongs to Mobile category can not be pressed, so that the velocity field in blood vessel meets conservation of mass condition (without the condition of dissipating)；Therefore, work as wall shear stress After obtaining initial vessel velocity field, the upper conservation of mass condition that is based on is to the initial blood vessel in terms of without property and slickness two is dissipated Velocity field optimizes, the velocity field of the blood vessel after just having obtained the corresponding optimization in initial vessel velocity field, i.e. optimization blood vessel speed Spend field.

It should be noted that the difference of optimization vessel velocity field and initial vessel velocity field meets preset condition, here in advance It is small as far as possible that if condition characterizes difference, meanwhile, optimization vessel velocity field has without property is dissipated, i.e. divergence is zero.In this way, optimization blood vessel speed Field is spent compared to initial vessel velocity field or compared to the vessel velocity field in blood vessel nuclear magnetic resonance data, noise is small, process bad point is few, And when space division ratio it is high.

In addition, optimization vessel velocity field is corresponding with blood vessel three-dimensional model.That is, optimization vessel velocity field refers to blood The velocity field of blood vessel in pipe threedimensional model.

S104, according to blood vessel three-dimensional model and optimization vessel velocity field computation wall shear stress, obtain blood vessel wall surface and cut to answer Power distributed intelligence.

In embodiments of the present invention, the wall shear stress characterization wall shear stress of blood vessel optimizes device according to the speed of blood vessel The shearing stress that field is calculated on blood vessel three-dimensional model, therefore, when wall shear stress optimization device is getting blood vessel three After dimension module and optimization vessel velocity field (velocity field of blood vessel), it will be able to according to blood vessel three-dimensional model and optimization vessel velocity Field carries out the calculating of the wall shear stress of blood vessel, to obtain the wall shear stress of each region on blood vessel three-dimensional model, i.e. blood Tube wall face shear Stress Distribution information.

It should be noted that distribution of the wall shear stress of blood vessel wall surface shear Stress Distribution information representation blood vessel on blood vessel Information.

Fig. 2 is a kind of illustrative blood vessel wall surface shear Stress Distribution information provided in an embodiment of the present invention, as shown in Fig. 2, The unit of blood vessel wall shear stress WSS is Pa, and corresponding value range is 0.0 to 1.0, and is worth the wall shear stress of bigger characterization Bigger, region shown in graticule is high wall shear stress region.

It is understood that wall shear stress optimization device is obtained using conservation of mass condition to by nuclear magnetic resonance data Initial vessel velocity field optimize, reached while the velocity field for ensuring blood vessel is without scattered property and slickness optimization after The velocity field of blood vessel have the effect of that noise is small, process bad point is few and when space division ratio it is high, after the velocity field for improving blood vessel Treatment effect.In addition, when speed field computation wall shear stress of the wall shear stress optimization device according to the blood vessel after optimization, energy Enough so that obtained wall shear stress accuracy is high, to improve the distributed intelligence of wall shear stress in blood vessel three-dimensional model Precision.

Further, in embodiments of the present invention, wall shear stress optimizes device according to blood vessel nuclear magnetic resonance number in S101 According to building blood vessel three-dimensional model specifically includes S101a-S101d, in which:

Initial blood vessel 3-D image in S101a, acquisition blood vessel nuclear magnetic resonance data.

In embodiments of the present invention, after carrying out nuclear magnetic resonance to blood vessel due to nuclear magnetic resonance equipment, obtained blood vessel core It include the initial three-dimensional image of blood vessel in MR data, i.e., initial blood vessel 3-D image, wall shear stress optimization device is to blood Initial blood vessel 3-D image in tube nucleus MR data extracts, that is, has got initial in blood vessel nuclear magnetic resonance data Blood vessel 3-D image.

S101b, denoising is carried out to initial blood vessel 3-D image, the initial blood vessel 3-D image after being denoised.

In embodiments of the present invention, due to including noise in initial blood vessel 3-D image, wall shear stress optimization dress It sets after getting initial blood vessel 3-D image, denoising is carried out to the initial blood vessel 3-D image, has just obtained noise Initial blood vessel 3-D image after smaller or muting denoising.

It should be noted that the initial blood vessel 3-D image characterization after denoising meets the three-dimensional of the blood vessel of default noise conditions Image, and default noise conditions are such as are as follows: it is less than default noise threshold.

Illustratively, when initial blood vessel 3-D image is aorta 3-D image, wall shear stress optimizes device and uses Neighborhood variance method and neighborhood method of positive and negative cases carry out denoising to aorta 3-D image, obtain the general profile of aorta；And Median filtering is carried out in the general profile of aorta, so that remaining noise all be rejected, has just obtained muting active Arteries and veins 3-D image (the initial blood vessel 3-D image after denoising).

It is understood that wall shear stress optimizes device by carrying out denoising to initial blood vessel 3-D image, obtain Initial blood vessel 3-D image after to denoising provides convenience for the subsequent processing to the initial blood vessel 3-D image after denoising And accuracy.

S101c, vessel information extraction is carried out to the initial blood vessel 3-D image after denoising using default partitioning algorithm, obtained Point cloud blood vessel 3-D image.

In embodiments of the present invention, when wall shear stress optimization device obtain denoising after initial blood vessel 3-D image it Afterwards, the extraction of vessel information is carried out on the initial blood vessel 3-D image after denoising using default partitioning algorithm, what this was extracted Vessel information is point cloud blood vessel 3-D image.

It should be noted that the 3-D image of the blood vessel of point cloud blood vessel 3-D image characterization point cloud representation, and put cloud It only include vessel information in blood vessel 3-D image, not comprising the other information other than vessel information.And default partitioning algorithm is used for The extraction of vessel information, such as maximum variance between clusters.

Illustratively, wall shear stress optimization device is using maximum variance between clusters to muting aorta 3-D image (the initial blood vessel 3-D image after denoising) carries out automatic threshold segmentation, completes the extraction of aorta, obtains aorta dot matrix cloud The image (point cloud blood vessel 3-D image) of form.

S101d, a cloud blood vessel 3-D image is reconstructed using default restructing algorithm, obtains blood vessel three-dimensional model.

In embodiments of the present invention, after wall shear stress optimization device obtains point cloud blood vessel 3-D image, it will be able to Surface reconstruction is carried out to cloud blood vessel 3-D image using default restructing algorithm, to obtain blood vessel three-dimensional model.

It should be noted that default restructing algorithm is used for the surface reconstruction of blood-vessel image, such as Poisson surface reconstruction.

Illustratively, wall shear stress optimization device is using Poisson surface reconstruction to the image of aorta dot matrix cloud form (point cloud blood vessel 3-D image) is handled, and aorta geometrical model (blood vessel three-dimensional model) is obtained.

It is understood that wall shear stress optimization device passes through to the initial blood vessel 3-D image in nuclear magnetic resonance data It is handled, reconstructs blood vessel three-dimensional model, blood vessel three-dimensional model can be retouched more accurately relative to initial blood vessel 3-D image State the geological information of blood vessel.

Further, in embodiments of the present invention, wall shear stress optimizes device according to blood vessel nuclear magnetic resonance number in S102 According to and blood vessel three-dimensional model, obtain initial vessel velocity field, specifically include S102a-S102b, in which:

Vessel velocity field in S102a, acquisition blood vessel nuclear magnetic resonance data.

In embodiments of the present invention, after carrying out nuclear magnetic resonance to blood vessel due to nuclear magnetic resonance equipment, obtained blood vessel core It include vessel velocity field in MR data, wall shear stress optimizes device to the vessel velocity field in blood vessel nuclear magnetic resonance data It extracts, that is, has got the vessel velocity field in blood vessel nuclear magnetic resonance data.

S102b, interpolation is carried out to vessel velocity field according to default spacing, the vessel velocity field after obtaining interpolation.

In embodiments of the present invention, since vessel velocity field is made of the corresponding velocity vector of a large amount of velocity node Physical field, meanwhile, the spacing between velocity node is unequal, leads to not complete to carry out wall shear stress according to vessel velocity field It calculates；Here, wall shear stress optimization device is according to default spacing (for example, default spacing h=1.57mm) to vessel velocity field Interpolation (for example, equidistant batten difference) is carried out, all speed sections that the spacing between each velocity node is default spacing are chosen Point, the vessel velocity field after constituting interpolation.

It should be noted that each speed in vessel velocity field when vessel velocity field is 3D velocity field, after interpolation Spacing of the node on three-dimensional is default spacing.In addition, the vessel velocity field after interpolation is three-dimensional, for example, after interpolation Vessel velocity field dimension be 186*122*57.

S102c, on blood vessel three-dimensional model, according to vessel borders to the velocity node in the vessel velocity field after interpolation into Line flag obtains initial vessel velocity field.

In embodiments of the present invention, vessel velocity field after wall shear stress optimization device obtains interpolation and blood vessel are three-dimensional After model, vessel borders are determined on blood vessel three-dimensional model, and according to vessel borders in the vessel velocity field after interpolation Velocity node is marked, specially according to vessel borders by endovascular velocity node and extravascular velocity node respectively into Line flag, and the endovascular velocity node of label is constituted into initial vessel velocity field.

Illustratively, when blood vessel is that aorta and wall shear stress optimize device and determine blood on blood vessel three-dimensional model After tube edge circle, 1 is set by the corresponding token variable of endaortic velocity node, extravascular velocity node is corresponding Token variable is set as 0.All velocity nodes that wherein token variable is 1 just constitute initial vessel velocity field for totally 86379.

It is understood that wall shear stress optimization device is fast by the initial blood vessel for constituting endovascular velocity node Data of the field as measurement blood flow velocity are spent, the accuracy of measurement is further improved.

Further, in embodiments of the present invention, in S103 wall shear stress optimization device be based on conservation of mass condition and Wall surface non-slip condition optimizes initial vessel velocity field, obtains optimization vessel velocity field, specifically includes S103a- S103d, in which:

S103a, according to wall surface non-slip condition, determine the corresponding derivative information in initial vessel velocity field.

In embodiments of the present invention, after wall shear stress optimization device obtains initial vessel velocity field, it will be able to Initial vessel velocity field is optimized, it is zero that wall shear stress, which optimizes device according to speed of the blood on vascular wall, i.e. wall Face non-slip condition determines the corresponding derivative information of velocity node in initial vessel velocity field respectively.

It should be noted that derivative information characterizes the corresponding derivative difference information of velocity node in initial vessel velocity field, For example, derivative information characterizes the corresponding difference equation of first derivative and second dervative pair of velocity node in initial vessel velocity field The difference equation answered.

Further, in embodiments of the present invention, wall shear stress optimization device is determined according to wall surface non-slip condition The corresponding derivative information in initial vessel velocity field, specifically includes S103a1-S103a5, in which:

S103a1, the wall surface distance for calculating velocity node in initial vessel velocity field.

In embodiments of the present invention, spacing is equal in all directions for velocity node in initial vessel velocity field, is pre- If spacing, to constitute the corresponding speed grid in initial vessel velocity field, and each grid node in speed grid A corresponding velocity node.In the speed grid, it is also marked with blood vessel three-dimensional model, at this point, wall shear stress optimizes device Obtain the wall surface distance of vascular wall of the velocity node in initial vessel velocity field to blood vessel three-dimensional model.

It should be noted that wall surface distance is the shortest distance of the velocity node to vascular wall.

S103a2, basis preset spacing and wall surface distance, and the velocity node in initial vessel velocity field is divided into nearly wall Millet cake and remote wall surface point.

In embodiments of the present invention, after wall shear stress optimization device obtains wall surface distance, and by the wall surface distance Be compared with default spacing, wall surface distance be more than or equal to the velocity node of default spacing as remote wall surface point, by wall surface away from From the velocity node less than default spacing as near wall point.

S103a3, the calculating of near wall point is led based on wall surface distance, wall surface non-slip condition and the first default derivative algorithm Number, obtains corresponding first derivative information.

In embodiments of the present invention, wall shear stress optimizes device to the near wall point and remote wall in initial vessel velocity field Millet cake determines derivative information using different algorithms, and here, wall shear stress optimizes device according to wall surface distance, wall surface without sliding Condition and the first default derivative algorithm calculate the derivative information of near wall point, i.e. the first derivative information.

It should be noted that the first default derivative algorithm refers in speed grid, velocity node is determined according to wall surface distance Derivative information calculate direction (i.e. the distance of velocity node to wall surface most in short-term corresponding direction), using velocity node in derivative Information calculates the velocity node of specified quantity nearest from the velocity node on direction and velocity node is calculated in derivative information Point, that is, vascular wall point on direction on the nearest vascular wall of distance, determines the corresponding derivative information of the velocity node.

Specifically, wall shear stress optimization device determines the vascular wall point and speed recently of near wall point according to wall surface distance Spend node；And it is led according to wall surface distance, vascular wall point and nearest velocity node and wall surface non-slip condition and first are default Method is figured, obtains the corresponding first derivative of near wall point and second dervative, the first default derivative algorithm is for determining near wall point Derivative information；And using first derivative and second level derivative as the first derivative information.

Illustratively, Fig. 3 is a kind of schematic diagram of illustrative speed grid provided in an embodiment of the present invention, such as Fig. 3 institute Show, which is two-dimension speed field, and corresponding coordinate system includes x-axis and y-axis, and the illustrative speed grid is also Including vascular wall (speed on vascular wall is zero) and illustrative two near walls point A and B；Wherein, A is to vascular wall Wall surface distance be horizontal distance dx (the wall surface distance of dx>0, dx<default spacing h), B to vascular wall for vertical range dy (dy>0, dy<h)；For A, the point A on the vascular wall of A in the horizontal direction is chosen¹(vascular wall point) and A 1 in the horizontal direction Nearest velocity node A², and use u₀、u₁And u₂Respectively indicate A¹, A and A²Corresponding velocity amplitude is (wherein, according to wall surface without sliding Condition, u₀0), then to carry out Taylor expansion to speed and corresponding first can be obtained according to the arrangement of wall surface non-slip condition to lead in A point Number information: the corresponding difference equation of first derivativeDifference equation corresponding with second dervativeRespectively such as formula (1) With shown in formula (2):

Wherein, θ indicates the ratio of the wall surface distance and default spacing of A, i.e. θ=dx/h.

Similarly, the first derivative information of B: the corresponding difference equation of first derivativeDifference corresponding with second dervative EquationHere, v is used₀、v₁、v₂Difference equation corresponding with the dy/h progress first derivative of B and second dervative are corresponding The solution of difference equation, wherein v₀、v₁And v₂Respectively indicate B¹, B and B²Corresponding velocity amplitude is (wherein, according to wall surface without sliding item Part, v₀For 0), and B¹For the point on the vascular wall of B in vertical direction, that is, vascular wall point, B²Most for B in vertical direction 1 Nearly velocity node.

S103a4, the calculating of remote wall surface point is led based on wall surface distance, wall surface non-slip condition and the second default derivative algorithm Number, obtains corresponding flection information.

In embodiments of the present invention, wall shear stress optimization device according to based on wall surface apart from, wall surface non-slip condition and Second default derivative algorithm calculates the derivative information of remote wall surface point, i.e. flection information；Here, the second default derivative algorithm is Different from the algorithm of the first default derivative algorithm, for example, when the first default derivative algorithm be formula (1) and formula (2), second preset lead Figuring method is center difference scheme.

S103a5, using the first derivative information and flection information as derivative information.

Here, wall shear stress optimization device is by corresponding first derivative information of near wall point and remote wall surface point corresponding the Derivative information of two derivative informations collectively as velocity node in initial vessel velocity field.

S103b, it is based on conservation of mass condition and derivative information, it is excellent constructs the corresponding vessel velocity field in initial vessel velocity field Change equation.

In embodiments of the present invention, when wall shear stress optimization device obtains leading for velocity node in initial vessel velocity field After number information, initial vessel velocity field is optimized based on conservation of mass condition and derivative information, here, passes through building pair The vessel velocity field optimization method answered is realized.Here, vessel velocity field optimization method characterizes the optimization mesh of initial vessel velocity field Mark.

Further, in embodiments of the present invention, wall shear stress optimization device is based on conservation of mass condition and derivative Information constructs the corresponding vessel velocity field optimization method in initial vessel velocity field, specifically includes S103b1-S103b4, in which:

S103b1, according to derivative information, determine the corresponding smoothness in initial vessel velocity field and discrete divergence.

In embodiments of the present invention, when wall shear stress optimization device obtains velocity node in initial vessel velocity field After derivative information, it will be able to determine in initial vessel velocity field the smoothness of velocity node and discrete according to the derivative information Divergence.

Specifically, when the corresponding difference equation of derivative information characterization first derivative and the corresponding difference equation of second dervative When, smoothness is determined based on the corresponding difference equation of second dervative, and discrete divergence is based on the corresponding difference equation of first derivative It determines.

Illustratively, in initial vessel velocity field the corresponding difference equation of the second dervative of velocity node constitute one it is discrete Second derivative operator D (does not consider cross term), and here, D is matrix form, and the value of matrix is corresponding velocity node (three The velocity node of dimension) the corresponding difference equation of second dervative, it may be assumed thatAt this point, smoothness R (U) is formula (3) shown in:

R (U)=‖ DU ‖²=U^TD^TDU (3)

Wherein, U is optimization vessel velocity field to be determined, also known as initial optimization vessel velocity field.

In addition, the corresponding difference equation of first derivative of velocity node constitutes discrete divergence in initial vessel velocity field:Or CU.

S103b2, according to smoothness, discrete divergence and conservation of mass condition, it is corresponding just to construct initial vessel velocity field Beginning vessel velocity field optimization object function.

In embodiments of the present invention, after wall shear stress optimization device obtains smoothness and discrete divergence, root According to smoothness, discrete divergence and conservation of mass condition, the corresponding initial vessel velocity field optimization in initial vessel velocity field is constructed Objective function determines the corresponding optimization blood vessel speed in initial vessel velocity field by initial vessel velocity field optimization object function here Spend field.

Illustratively, when smoothness such as formula (3) is shown, discrete divergence isWhen, then initial vessel velocity field optimization Objective function J (U) is as shown in formula (4):

Wherein, s characterizes initial smoothing parameter, and the velocity field of the bigger characterization blood vessel of s > 0, s is more smooth.J (U) characterization In discrete divergenceIn the case where 0, and with initial vessel velocity field U_mInitial optimization blood vessel speed is carried out when difference minimum Spend the determination of field U.

S103b3, according to Lagrange multiplier algorithm, initial vessel velocity field optimization object function is converted, is obtained Initial vessel velocity field optimization object function after conversion.

In embodiments of the present invention, when wall shear stress optimization device obtains initial vessel velocity field optimization object function It later, need to be according to Lagrange in order to determine optimization vessel velocity field according to the initial vessel velocity field optimization object function Multiplier Algorithm converts initial vessel velocity field optimization object function, the initial blood vessel after being converted into the conversion convenient for calculating Velocity field optimization object function.

Illustratively, when Lagrange multiplier is undetermined coefficient λ, initial vessel velocity field optimization object function J (U) is formula (5) when, then the initial vessel velocity field optimization object function L (U, λ) after converting is as shown in formula (5):

S103b4, minimum processing is carried out to the initial vessel velocity field optimization object function after conversion, obtains blood vessel speed Spend field optimization method.

In embodiments of the present invention, initial vessel velocity field optimization after wall shear stress optimization device obtains conversion After objective function, minimum processing is carried out to the initial vessel velocity field optimization object function after conversion, can just access can The vessel velocity field optimization method of solution.

Illustratively, wall shear stress optimization device optimizes mesh to the initial vessel velocity field after converting shown in formula (5) Scalar functions carry out minimum processing, realize especially by setting L (U, λ) about the first derivative of U and λ is 0, to obtain formula (6) vessel velocity field optimization method shown in:

Here,For the coefficient matrix of vessel velocity field optimization method, the matrix being directed to It is sparse matrix, specific I is unit matrix, and C is that difference equation corresponding with first derivative in smoothing information is associated sparse Matrix.

S103c, it is based on vessel velocity field optimization method, default cross validation algorithm and default optimization algorithm, determines optimization Smoothing parameter.

In embodiments of the present invention, after wall shear stress optimization device obtains vessel velocity field optimization method, by There are initial smoothing parameters to be determined in the optimization method of vessel velocity field, and here, wall shear stress optimizes device using pre- If cross validation algorithm and default optimization algorithm simultaneously determine optimal smoothing parameter according to vessel velocity field optimization method, that is, optimize Smoothing parameter.

It is understood that since the noise level of the velocity field of blood vessel can not determine, initial smoothing parameter also without Method determines that wall shear stress optimizes device by the optimization to initial smoothing parameter, provides to optimize the determination of smoothing parameter Implementable solution.

Further, in embodiments of the present invention, wall shear stress optimization device be based on vessel velocity field optimization method, Default cross validation algorithm and default optimization algorithm determine optimization smoothing parameter, specifically include S103c1-S103c5, in which:

S103c1, basis preset cross validation algorithm, construct smooth verification information.

In embodiments of the present invention, wall shear stress optimization device is determining that basis is default first when optimizing smoothing parameter Cross validation algorithm constructs smooth verification information；Here, smooth verification information characterizes the optimization information of smoothing parameter undetermined.

It should be noted that default cross validation algorithm is Generalized Cross Validation function generalized cross Validation, GCV), it can determine optimization smoothing parameter by minimizing GCV function.Default cross validation algorithm can be with The algorithm of optimization smoothing parameter is determined for other, the present invention is not especially limit this.

Illustratively, smooth verification information GCV (s) is as shown in formula (7):

Wherein, n is the number of unknown quantity in formula (7), the mark of Tr representing matrix, the initial smoothing parameter of s expression, U_mIndicate s Corresponding initial optimization vessel velocity field.

S103c2, it is based on default optimization algorithm, determines at least one initial smoothing parameter.

In embodiments of the present invention, wall shear stress optimization device is when determining optimization smoothing parameter, it is also necessary to based on pre- If optimization algorithm determines at least one initial smoothing parameter, so that determining that optimization is flat from least one initial smoothing parameter Sliding parameter.Here, at least one initial smoothing parameter characterizes the corresponding multiple values of smoothing parameter to be determined.

S103c3, at least one initial smoothing parameter is separately input into vessel velocity field optimization method, obtained corresponding At least one initial optimization vessel velocity field.

In embodiments of the present invention, after wall shear stress optimization device obtains at least one initial smoothing parameter, Due to the correspondence that optimization method characterization in vessel velocity field is smoothing parameter, Lagrange multiplier and initial optimization vessel velocity field Relationship, wall shear stress optimizes device, and by this, at least one initial smoothing parameter is separately input into vessel velocity field optimization method, It can obtain at least one corresponding initial optimization vessel velocity field.

Here, wall shear stress optimization device determines optimization blood vessel speed from least one initial optimization vessel velocity field Spend field.

S103c4, according at least one initial optimization vessel velocity field, determine smooth verification information it is corresponding at least one Smooth validation value.

In embodiments of the present invention, when wall shear stress optimization device obtains at least one initial optimization vessel velocity field Later, which is separately input into smooth verification information, can be calculated corresponding At least one smooth validation value.

Specifically, in embodiments of the present invention, wall shear stress optimization device is according at least one initial optimization blood vessel speed Field is spent, determines at least one corresponding smooth validation value of smooth verification information, comprising: wall shear stress optimizes device according to smooth Information determines the coefficient information in smooth verification information；And according to default fitting algorithm, the corresponding feature of coefficient information is determined Value；And characteristic value is input to smooth verification information, obtain the corresponding smooth verification information of simplification；And it will be at the beginning of at least one Beginning optimization vessel velocity field, which is separately input into, simplifies smooth verification information, obtains at least one smooth validation value.

That is, here, coefficient information characterizes root since smooth verification information includes the corresponding characteristic value of coefficient information The information determined according to smoothing information is (for example, the D in formula (7)^TD), and coefficient information is sparse matrix form；Wall shear stress Optimization device accurately solves the characteristic value of the predetermined number (for example, preceding m) of coefficient information, and using default fitting algorithm according to The characteristic value (such as exponential function fitting algorithm) of predetermined number is fitted characteristic value, feature Distribution value is obtained, thus closely Seemingly calculate All Eigenvalues；Here characteristic value number is N, characteristic value λ_i, i be 1 to N integer.

It should be noted that the number of the quantity of the corresponding characteristic value of coefficient information and velocity node in initial vessel velocity field Amount is consistent, for example, when velocity node is 86379 in initial vessel velocity field, at this point, coefficient information D^TD is 86379* 259137 sparse matrix passes through fitting algorithm approximate solution D^TThe corresponding whole characteristic value of D.

At this point, simplifying shown in smooth verification information such as formula (8) when smooth verification information is as shown in formula (7):

Since the smooth verification information of simplification characterizes initial vessel velocity field, initial optimization vessel velocity field and initial smooth The corresponding relationship of parameter, therefore, when wall shear stress optimization device obtains initial vessel velocity field, at least one initial optimization After vessel velocity field and at least one initial smoothing parameter, it will be able to determine at least one smooth validation value.

S103c5, by the corresponding initial smoothing parameter of validation value minimum smooth at least one smooth validation value, as excellent Change smoothing parameter.

In embodiments of the present invention, after wall shear stress optimization device obtains at least one smooth validation value, by The optimization performance of initial smoothing parameter is characterized in smooth validation value, it is corresponding initial smooth specially when smooth validation value is smaller The optimization performance of parameter is higher, so that wall shear stress optimizes device for smooth validation value minimum at least one smooth validation value Corresponding initial smoothing parameter just obtains optimal smoothing parameter as optimization smoothing parameter.

S103d, optimization smoothing parameter is input to vessel velocity field optimization method, obtains optimization vessel velocity field.

In embodiments of the present invention, when wall shear stress optimization device is after obtaining optimization smoothing parameter, due to blood Pipe velocity field optimization method characterizes optimization smoothing parameter and optimizes the corresponding relationship of vessel velocity field, therefore, wall shear stress Optimization device is input to vessel velocity field optimization method for smoothing parameter is optimized, it will be able to obtain optimization vessel velocity field.

It is understood that wall shear stress optimizes device by determining blood according to derivative information and initial vessel velocity field Pipe velocity field optimization method, and vessel velocity field optimization method characterizes optimization smoothing parameter and optimizes the correspondence of vessel velocity field Relationship just can determine that out while meet without the velocity field constrained with slickness is dissipated, avoid after optimization smoothing parameter is determining Because speed field error greatly caused by wall shear stress inaccuracy problem.

Further, in embodiments of the present invention, in S104 wall shear stress processing unit according to blood vessel three-dimensional model and Optimize vessel velocity field computation wall shear stress, obtain blood vessel wall surface shear Stress Distribution information, specifically include S104a-S104d, Wherein:

S104a, it is carried out in the wall surface normal orientation of blood vessel three-dimensional model according to optimization vessel velocity field and default spacing Interpolation obtains interpolation speed.

In embodiments of the present invention, it after wall shear stress optimization device obtains optimization vessel velocity field, determines that The wall surface normal orientation of blood vessel three-dimensional model selects in the velocity vector in wall surface normal orientation from optimization vessel velocity field The velocity vector of preset quantity (for example, 7) out, has just obtained interpolation speed.Here, the first speed arrow in interpolation speed Amount is the velocity vector on vascular wall, and corresponding speed is zero.

S104b, according to interpolation speed and wall surface normal orientation, determine that blood flows to.

In embodiments of the present invention, when wall shear stress optimization device obtain interpolation speed and wall surface normal orientation it Afterwards, according to the corresponding direction of interpolation speed and wall surface normal orientation, it will be able to determine that blood flows to.Here, blood flow chart Levy the flow direction of blood in blood vessel.

Specifically, an intermediate direction, intermediate direction direction corresponding with interpolation speed and wall surface normal direction side are set To being vertical relation, thus, wall shear stress optimize device using the multiplication cross result of intermediate direction and wall surface normal orientation as Blood flow direction.

S104c, it is flowed to according to interpolation speed and blood, obtains corresponding matched curve.

In embodiments of the present invention, after wall shear stress optimization device obtains interpolation speed and blood flows to, energy Obtain the corresponding rate curve of interpolation speed, i.e. matched curve.

Specifically, in embodiments of the present invention, wall shear stress optimization device is flowed to according to interpolation speed and blood, is obtained Corresponding matched curve, comprising: wall shear stress optimizes device and calculates the interpolation speed projection information upward in blood stream, obtains Corresponding velocity profile information；And velocity profile information is fitted, obtain matched curve.

Preferably, wall shear stress optimization device is fitted velocity profile information, when obtaining matched curve, specifically adopts It is fitted with following steps: firstly, obtaining blood attribute information；Here, the blood flowed in blood attribute information characterization blood vessel The self attributes of liquid, for example, the density of blood and the dynamic viscosity coefficient of blood etc..Then, according to blood attribute information, at least One default blood vessel wall surface friction velocity and default fitting coefficient and default fitting algorithm, intend velocity profile information It closes, obtains corresponding at least one initial matched curve, default fitting algorithm is for being fitted velocity profile information；Here lead to It crosses at least one default blood vessel wall surface friction velocity and fits at least one corresponding initial matched curve.Finally, determining at least One initial matched curve at least one distinctive information corresponding with velocity profile information respectively；And by least one distinctive information It is middle to distinguish the corresponding initial matched curve of the smallest distinctive information as matched curve；Here, blood vessel is preset according at least one Wall friction speed determines at least one corresponding initial matched curve of velocity profile information by default fitting algorithm, and It is selected from least one initial matched curve with the smallest initial matched curve of velocity profile information difference as VELOCITY DISTRIBUTION The corresponding matched curve of information.

That is, in embodiments of the present invention, when wall shear stress optimization device is fitted velocity profile information, Using a kind of higher implicit vail function (default fitting algorithm) of precision, specifically as shown in formula (9):

Wherein, W ' is the height of the corresponding wall surface normal direction of each speed in velocity profile information；Parameter u_τ, ρ and μ be respectively The dynamic viscosity coefficient of blood vessel wall surface friction velocity, the density of blood and blood is (for example, ρ is 1060, μ 0.0035, wherein ρ It is respectively the density of the blood in blood attribute information and the dynamic viscosity coefficient of blood with μ)；κ and s is default fitting coefficient, It is constant (for example, empirically determined κ and s are respectively 0.41 and 0.001093).Here, unique unknown in formula (9) Number is blood vessel wall surface friction velocity u_τ, u can be obtained by curve matching_τBest value；And for u_τInitial value, be exactly in advance If blood vessel wall surface friction velocity, the initial matched curve of velocity profile information is determined by presetting blood vessel wall surface friction velocity (here, V ' is the corresponding continuous dependent variable value of the initial matched curve), and will be most matched first with velocity distribution curve Matched curve U ' of the beginning matched curve as velocity profile information.

It should be noted that formula (9) is one of fit approach, multinomial also can be used, to digit rate, other walls The forms such as function are fitted the velocity profile information of wall surface.By comparison, the wall shear stress that is obtained using formula (9) Error is minimum.

Illustratively, Fig. 4 is a kind of schematic diagram of illustrative determining matched curve provided in an embodiment of the present invention, is such as schemed It is the sectional view of velocity field shown in 4, which is the local coordinate in each vertex position foundation of blood vessel three-dimensional model System, wherein the preset quantity of interpolation is 7, and the distance between interpolation speed is identical as default spacing, and X ' is blood flow flow direction, and Y ' is Wall surface normal orientation, the expression of arrow graticule is the interpolation speed corresponding velocity profile information (speed on the wall surface of vascular wall For 0), U ' is matched curve.

S104d, according to blood flow flow direction, matched curve and default hemodynamics viscosity, obtain wall shear stress distribution Information.

In embodiments of the present invention, wall shear stress characterizes blood flow flow direction, matched curve and default hemodynamics viscosity The corresponding relationship of coefficient, therefore, when wall shear stress optimization device obtains blood flow flow direction, matched curve and default hemodynamics After viscosity, wall shear stress just can determine that out, so that all wall shear stress determined are constituted wall shear stress Distributed intelligence.

Illustratively, when the functional relation of matched curve characterization U ' and Y ', wall shear stress τ such as formula (10) or formula (11) It is shown:

Wherein, the result that formula (10) or formula (11) calculate is identical, equivalent；The direction of wall shear stress and blood flow to Direction it is consistent.

In addition, it is necessary to explanation, formula (11) is to determine that the corresponding default blood vessel wall surface of matched curve rubs in S104c Speed is wiped, wall surface is calculated according to the density of the blood in blood attribute information and default blood vessel wall surface friction velocity cuts and answer Power.

Illustratively, Fig. 5 illustrates a kind of blood vessel wall surface shear Stress Distribution information illustratively calculated based on formula (9), As shown in figure 5, identifying high wall shear stress region；Fig. 6 illustrates a kind of illustratively based on the calculating of wall surface second order polynomial Blood vessel wall surface shear Stress Distribution information, as shown in fig. 6, identifying high wall shear stress region；Practical knot corresponding with blood vessel Fruit comparison can obtain, and closer to exact value, accuracy is higher for the wall shear stress distributed intelligence that the former obtains.

It should be noted that after wall shear stress optimization device obtains wall shear stress distributed intelligence, according to the wall Face shear Stress Distribution information completes measuring blood flow rate as data based on diagnosis vascular diseases.For example, utilizing the wall surface Shear Stress Distribution information measurement entocranial artery blood flow and abdominal vascular blood flow.

It is understood that since optimization vessel velocity field is that wall shear stress optimization device passes through conservation of mass condition pair What initial vessel velocity field optimized, and because conservation of mass condition can ensure to optimize the flatness of vessel velocity field With without property is dissipated, therefore, blood flow informations noise is small, velocity field bad point is few, spatial and temporal resolution is high for optimization vessel velocity field etc., from And when wall shear stress optimization device is according to optimization vessel velocity field computation wall shear stress, improve the standard of wall shear stress True property.

Embodiment two

Based on the same inventive concept of embodiment one, the embodiment of the invention provides a kind of wall shear stress to optimize device 1, right It should be in a kind of wall shear stress optimization method；Fig. 7 is the knot that a kind of wall shear stress provided in an embodiment of the present invention optimizes device Structure schematic diagram one, as shown in fig. 7, wall shear stress optimization device 1 includes:

Construction unit 10, for when getting blood vessel nuclear magnetic resonance data, according to the blood vessel nuclear magnetic resonance data, structure Build blood vessel three-dimensional model；

Acquiring unit 11, for obtaining initial blood according to the blood vessel nuclear magnetic resonance data and the blood vessel three-dimensional model Pipe velocity field；

Optimize unit 12, for being based on conservation of mass condition, the initial vessel velocity field is optimized, is optimized Vessel velocity field；

Computing unit 13, for cutting and answering according to the blood vessel three-dimensional model and the optimization vessel velocity field computation wall surface Power obtains blood vessel wall surface shear Stress Distribution information.

Further, the construction unit 10, specifically for obtaining the initial blood vessel in the blood vessel nuclear magnetic resonance data 3-D image；Denoising is carried out to the initial blood vessel 3-D image, the initial blood vessel 3-D image after being denoised；Using Default partitioning algorithm carries out vessel information extraction to the initial blood vessel 3-D image after the denoising, obtains a cloud blood vessel three-dimensional figure Picture；Described cloud blood vessel 3-D image is reconstructed using default restructing algorithm, obtains the blood vessel three-dimensional model.

Further, the acquiring unit 11, specifically for obtaining the vessel velocity in the blood vessel nuclear magnetic resonance data ?；Interpolation is carried out to the vessel velocity field according to default spacing, the vessel velocity field after obtaining interpolation；It is three-dimensional in the blood vessel On model, the velocity node in the vessel velocity field after the interpolation is marked according to vessel borders, is obtained described initial Vessel velocity field.

Further, the optimization unit 12 includes the first determination unit 120, the determining list of target construction unit 121, second Member 122 and input unit 123；

First determination unit 120, for determining that the initial vessel velocity field is corresponding according to wall surface non-slip condition Derivative information；

The target construction unit 121 constructs described first for being based on the conservation of mass condition and the derivative information The corresponding vessel velocity field optimization method in beginning vessel velocity field；

Second determination unit 122, for based on vessel velocity field optimization method, default cross validation algorithm and Default optimization algorithm determines optimization smoothing parameter；

The input unit 123 is obtained for the optimization smoothing parameter to be input to vessel velocity field optimization method To the optimization vessel velocity field.

Further, first determination unit 120, specifically for calculating the speed section in the initial vessel velocity field The wall surface distance of point；According to default spacing and the wall surface distance, the velocity node in the initial vessel velocity field is divided For near wall point and remote wall surface point；Based on the wall surface distance, the wall surface non-slip condition and the first default derivative algorithm pair The near wall point calculates derivative, obtains corresponding first derivative information；Based on the wall surface distance, the wall surface without sliding item Part and the second default derivative algorithm calculate derivative to the remote wall surface point, obtain corresponding flection information；By described first Derivative information and the flection information are as the derivative information.

Further, first determination unit 120 is specifically used for determining the near wall according to the wall surface distance The vascular wall point and nearest velocity node of point；And according to wall surface distance, vascular wall point and the nearest velocity node, And the wall surface non-slip condition and the first default derivative algorithm, obtain the corresponding first derivative of the near wall point and Second dervative, the first default derivative algorithm are used to determine the derivative information of the near wall point；And the single order is led The several and second level derivative is as first derivative information.

Further, the target construction unit 121 is specifically used for determining the initial blood according to the derivative information The corresponding smoothness of pipe velocity field and discrete divergence；According to the smoothness, the discrete divergence and the conservation of mass Condition constructs the corresponding initial vessel velocity field optimization object function in the initial vessel velocity field；According to Lagrange multiplier Algorithm converts the initial vessel velocity field optimization object function, the initial vessel velocity field optimization after being converted Objective function；Minimum processing is carried out to the initial vessel velocity field optimization object function after the conversion, obtains the blood vessel Velocity field optimization method.

Further, second determination unit 122 is specifically used for according to the default cross validation algorithm, and building is flat Sliding verification information；Based on the default optimization algorithm, at least one initial smoothing parameter is determined；At least one is initial flat by described in Sliding parameter is separately input into vessel velocity field optimization method, obtains at least one corresponding initial optimization vessel velocity field； According at least one described initial optimization vessel velocity field, at least one corresponding smooth verifying of the smooth verification information is determined Value；By the corresponding initial smoothing parameter of minimum smooth validation value at least one described smooth validation value, put down as the optimization Sliding parameter.

Further, second determination unit 122 determines described smooth also particularly useful for according to the derivative information Coefficient information in verification information；According to default fitting algorithm, the corresponding characteristic value of the coefficient information is determined；By the feature Value is input to the smooth verification information, obtains the corresponding smooth verification information of simplification；It will at least one described initial optimization blood Pipe velocity field is separately input into the smooth verification information of the simplification, obtains at least one described smooth validation value.

Further, the computing unit 13 is specifically used for according to the optimization vessel velocity field and default spacing in institute The enterprising row interpolation of wall surface normal orientation for stating blood vessel three-dimensional model, obtains interpolation speed；According to the interpolation speed and the wall Face normal orientation determines that blood flows to；It is flowed to according to the interpolation speed and the blood, obtains corresponding matched curve；Root According to blood flow flow direction, the matched curve and default hemodynamics viscosity, the wall shear stress distributed intelligence is obtained.

Further, the computing unit 13, upward in the blood stream also particularly useful for the calculating interpolation speed Projection information obtains corresponding velocity profile information；The velocity profile information is fitted, the matched curve is obtained.

Further, the computing unit 13 is specifically used for obtaining blood attribute information；And according to the blood attribute Information, at least one default blood vessel wall surface friction velocity and default fitting coefficient and default fitting algorithm, to the speed point Cloth information is fitted, and obtains corresponding at least one initial matched curve, the default fitting algorithm is used for the speed point Cloth information is fitted；And determine at least one described initial matched curve respectively it is corresponding with the velocity profile information at least One distinctive information；And the corresponding initial matched curve of the smallest distinctive information will be distinguished at least one described distinctive information As the matched curve.

Further, the default fitting algorithm are as follows:

It should be noted that in practical applications, above-mentioned construction unit 10, optimization unit 12, calculates list at acquiring unit 11 First 13, first determination unit 120, target construction unit 121, the second determination unit 122 and input unit 123 can be by being located at wall surface Shearing stress optimizes the processor 14 on device 1 and realizes, specially CPU (Central Processing Unit, central processing Device), MPU (Microprocessor Unit, microprocessor), DSP (Digital Signal Processing, digital signal Processor) or field programmable gate array (FPGA, Field Programmable Gate Array) etc. realize.

The embodiment of the invention also provides a kind of wall shear stress to optimize device 1, as shown in figure 8, the wall shear stress Optimization device 1 includes: processor 14, memory 15 and communication bus 16, the memory 15 by the communication bus 16 with The processor 14 is communicated, and the memory 15 stores the executable program of the processor 14, when described program is held When row, the wall shear stress optimization method as described in embodiment one is executed by the processor 14.

In practical applications, above-mentioned memory 15 can be volatile memory (volatile memory), such as at random It accesses memory (Random-Access Memory, RAM)；Or nonvolatile memory (non-volatile memory), Such as read-only memory (Read-Only Memory, ROM), flash memory (flash memory), hard disk (Hard Disk Drive, HDD) or solid state hard disk (Solid-State Drive, SSD)；Or the combination of the memory of mentioned kind, and to place It manages device 14 and instruction and data is provided.

The embodiment of the invention provides a kind of computer readable storage mediums, are stored thereon with program, and described program is located Manage the wall shear stress optimization method realized as described in embodiment one when device 14 executes.

It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program Product.Therefore, the shape of hardware embodiment, software implementation or embodiment combining software and hardware aspects can be used in the present invention Formula.Moreover, the present invention, which can be used, can use storage in the computer that one or more wherein includes computer usable program code The form for the computer program product implemented on medium (including but not limited to magnetic disk storage and optical memory etc.).

The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates, Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or The function of being specified in multiple boxes.

These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one The step of function of being specified in a box or multiple boxes.

The foregoing is only a preferred embodiment of the present invention, is not intended to limit the scope of the present invention.

Claims

1. a kind of wall shear stress optimization method, which is characterized in that the described method includes:

According to the blood vessel three-dimensional model and the optimization vessel velocity field computation wall shear stress, blood vessel wall shear stress is obtained Distributed intelligence.

2. the method according to claim 1, wherein described according to the blood vessel nuclear magnetic resonance data, building blood Pipe threedimensional model, comprising:

Vessel information extraction is carried out to the initial blood vessel 3-D image after the denoising using default partitioning algorithm, obtains cloud blood Pipe 3-D image；

3. the method according to claim 1, wherein described according to the blood vessel nuclear magnetic resonance data and the blood Pipe threedimensional model obtains initial vessel velocity field, comprising:

On the blood vessel three-dimensional model, the velocity node in the vessel velocity field after the interpolation is carried out according to vessel borders Label, obtains the initial vessel velocity field.

4. the method according to claim 1, wherein described be based on conservation of mass condition, to the initial blood vessel Velocity field optimizes, and obtains optimization vessel velocity field, comprising:

Based on the conservation of mass condition and the derivative information, the corresponding vessel velocity field in the initial vessel velocity field is constructed Optimization method；

Based on vessel velocity field optimization method, default cross validation algorithm and default optimization algorithm, the smooth ginseng of optimization is determined Number；

The optimization smoothing parameter is input to vessel velocity field optimization method, obtains the optimization vessel velocity field.

5. according to the method described in claim 4, it is characterized in that, described according to the wall surface non-slip condition, determine described in The corresponding derivative information in initial vessel velocity field, comprising:

According to default spacing and the wall surface distance, the velocity node in the initial vessel velocity field is divided near wall point With remote wall surface point；

Near wall point calculating is led based on the wall surface distance, the wall surface non-slip condition and the first default derivative algorithm Number, obtains corresponding first derivative information；

The remote wall surface point calculating is led based on the wall surface distance, the wall surface non-slip condition and the second default derivative algorithm Number, obtains corresponding flection information；

6. according to the method described in claim 5, it is characterized in that, the wall surface distance, the wall surface of being based on is without sliding Condition and the first default derivative algorithm calculate derivative to the near wall point, obtain corresponding first derivative information, comprising:

According to wall surface distance, vascular wall point and the nearest velocity node and the wall surface non-slip condition and The first default derivative algorithm, obtains the corresponding first derivative of the near wall point and second dervative, and described first default leads Method is figured for determining the derivative information of the near wall point；

7. according to the method described in claim 4, it is characterized in that, described believed based on the conservation of mass condition and the derivative Breath constructs the corresponding vessel velocity field optimization method in the initial vessel velocity field, comprising:

According to the smoothness, the discrete divergence and the conservation of mass condition, it is right to construct the initial vessel velocity field The initial vessel velocity field optimization object function answered；

According to Lagrange multiplier algorithm, the initial vessel velocity field optimization object function is converted, after obtaining conversion Initial vessel velocity field optimization object function；

Minimum processing is carried out to the initial vessel velocity field optimization object function after the conversion, obtains the vessel velocity field Optimization method.

8. according to the method described in claim 4, it is characterized in that, described be based on vessel velocity field optimization method, preset Cross validation algorithm and default optimization algorithm determine optimization smoothing parameter, comprising:

At least one described initial smoothing parameter is separately input into vessel velocity field optimization method, obtain it is corresponding at least One initial optimization vessel velocity field；

According at least one described initial optimization vessel velocity field, determining the smooth verification information, corresponding at least one is smooth Validation value；

By the corresponding initial smoothing parameter of minimum smooth validation value at least one described smooth validation value, put down as the optimization Sliding parameter.

9. according to the method described in claim 8, it is characterized in that, described at least one initial optimization vessel velocity according to , determine at least one corresponding smooth validation value of the smooth verification information, comprising:

At least one described initial optimization vessel velocity field is separately input into the smooth verification information of the simplification, obtain it is described extremely A few smooth validation value.

10. the method according to claim 1, wherein described according to the blood vessel three-dimensional model and the optimization Vessel velocity field computation wall shear stress obtains blood vessel wall surface shear Stress Distribution information, comprising:

It is carried out in the wall surface normal orientation of the blood vessel three-dimensional model according to the optimization vessel velocity field and default spacing slotting Value, obtains interpolation speed；

According to blood flow flow direction, the matched curve and default hemodynamics viscosity, the wall shear stress point is obtained Cloth information.

11. according to the method described in claim 10, it is characterized in that, described according to the interpolation speed and the blood stream To obtaining corresponding matched curve, comprising:

The velocity profile information is fitted, the matched curve is obtained.

12. according to the method for claim 11, which is characterized in that it is described that the velocity profile information is fitted, it obtains To the matched curve, comprising:

Obtain blood attribute information；

According to the blood attribute information, at least one default blood vessel wall surface friction velocity and default fitting coefficient, and preset Fitting algorithm is fitted the velocity profile information, obtains corresponding at least one initial matched curve, the default fitting Algorithm is for being fitted the velocity profile information；

Determine at least one described initial matched curve at least one distinctive information corresponding with the velocity profile information respectively；

The corresponding initial matched curve of the smallest distinctive information will be distinguished at least one described distinctive information as the fitting Curve.

13. according to the method for claim 12, which is characterized in that the default fitting algorithm are as follows:

Wherein, V ' is the corresponding continuous dependent variable value of the initial matched curve, u_τFor the default blood vessel wall friction speed Degree, W ' are the height of the corresponding wall surface normal direction of speed each in the velocity profile information, and κ and s are the default fitting coefficient, ρ is the density of the blood in the blood attribute information, and μ is the dynamic viscosity coefficient of the blood in the blood attribute information.

14. a kind of wall shear stress optimizes device, which is characterized in that described device includes: that processor, memory and communication are total Line, the memory are communicated by the communication bus with the processor, and the memory stores the processor can The program of execution passes through the processor and executes such as the described in any item sides of claim 1-13 when described program is performed Method.

15. a kind of computer readable storage medium, is stored thereon with program, which is characterized in that described program is executed by processor The Shi Shixian such as described in any item methods of claim 1-13.