CN118172349A - Vascular quantization parameter prediction method and device, electronic equipment and storage medium - Google Patents
Vascular quantization parameter prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN118172349A CN118172349A CN202410394420.6A CN202410394420A CN118172349A CN 118172349 A CN118172349 A CN 118172349A CN 202410394420 A CN202410394420 A CN 202410394420A CN 118172349 A CN118172349 A CN 118172349A
- Authority
- CN
- China
- Prior art keywords
- predicted
- node
- blood vessel
- nodes
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000013139 quantization Methods 0.000 title claims description 27
- 230000002792 vascular Effects 0.000 title description 14
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 148
- 238000013528 artificial neural network Methods 0.000 claims abstract description 41
- 238000011002 quantification Methods 0.000 claims abstract description 9
- 230000008859 change Effects 0.000 claims description 22
- 238000005070 sampling Methods 0.000 claims description 22
- 230000000747 cardiac effect Effects 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 16
- 230000017531 blood circulation Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 description 26
- 230000008569 process Effects 0.000 description 10
- 238000012545 processing Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 239000012530 fluid Substances 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000001052 transient effect Effects 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 201000000057 Coronary Stenosis Diseases 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 210000000709 aorta Anatomy 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002526 effect on cardiovascular system Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002107 myocardial effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000000748 cardiovascular system Anatomy 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 208000006331 coronary aneurysm Diseases 0.000 description 1
- 238000002586 coronary angiography Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000002608 intravascular ultrasound Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 208000031225 myocardial ischemia Diseases 0.000 description 1
- 210000004165 myocardium Anatomy 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000011477 surgical intervention Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses a blood vessel quantification parameter prediction method, a blood vessel quantification parameter prediction device, electronic equipment and a storage medium. Comprising the following steps: acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line; determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted; inputting initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period; and determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted. The parameter prediction results of a plurality of times to be predicted corresponding to the medical image data are determined through the graph neural network prediction model, and the prediction speed and the prediction precision are improved.
Description
Technical Field
The present invention relates to the field of cardiovascular prediction technologies, and in particular, to a method and apparatus for predicting a vascular quantization parameter, an electronic device, and a storage medium.
Background
Fractional flow reserve (Fractional Flow Reserve, FFR for short) in the coronary is a common measurement in the cardiovascular field and its main purpose is to assess the effect of coronary stenosis on myocardial blood flow. The FFR can be quickly and accurately determined, which facilitates accurate assessment of whether coronary stenosis is sufficient to affect the blood flow supply to the myocardium.
Computational fluid dynamics (CFD, computational Fluid Dynamics) has become an essential tool in cardiovascular system research. For example, CFD simulation has been used to non-invasively assess the severity of coronary aneurysms, assess the extent of myocardial ischemia and whether surgical intervention is required, and assist physicians in planning and predicting the effectiveness of stenting procedures prior to stenting. The complete three-dimensional coronary blood flow simulation is carried out, a blood vessel model of a patient is usually obtained by using imaging methods such as CTA, DSA and IVUS, a Navier-Stokes equation is solved by adopting a finite element, finite difference or finite volume method, and fluid mechanical parameters such as pressure, speed, wall shear stress and the like of intravascular flow can be obtained after the solution. The three-dimensional equation to be solved can be simplified into a zero-dimensional or one-dimensional equation by a reduced-order simulation method, and the fluid mechanics parameters can be solved.
Based on the prior art scheme, computational Fluid Dynamics (CFD) based computer blood flow simulation modeling is adopted to perform three-dimensional or reduced-order computational fluid dynamics simulation, and the pressure value in the blood vessel obtained by the simulation is used, so that the requirement on high precision requirement is met, the complex setting during calculation exists, the configuration requirement on computer hardware equipment is higher, and the problems of high calculation cost and complex calculation exist. Although the complexity of the problem can be reduced and the requirement on hardware configuration and the calculation cost can be reduced by the order reduction method, the accuracy of the order reduction method is poor, and the clinical accuracy requirement cannot be well met.
Disclosure of Invention
The invention provides a blood vessel quantification parameter prediction method, a blood vessel quantification parameter prediction device, electronic equipment and a storage medium, which are used for solving the problems of complex calculation, high calculation cost and low calculation precision in the prior art.
According to an aspect of the present invention, there is provided a blood vessel quantization parameter prediction method, including:
acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line;
determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted;
Inputting initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period;
And determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
Optionally, in the case that the node to be predicted includes a blood vessel inlet node type and a blood vessel outlet node type, the initialization data corresponding to the node to be predicted includes boundary condition data of each time to be predicted; the boundary condition data includes vessel inlet parameters, vessel outlet parameters.
Optionally, in the case that the node to be predicted further includes one or more of a vessel bifurcation node and a vessel single branch node, the initialization data corresponding to the node to be predicted includes one or more of initial parameter data and attribute information of the node to be predicted; the initial parameter data comprise a pressure value and a flow value which correspond to the time to be predicted; the attribute information of the node to be predicted includes geometric information of the node to be predicted.
Optionally, the determining manner of the single branch node of the blood vessel includes:
for a single blood vessel central line on a blood vessel central line, gradually sampling on the single blood vessel central line according to a preset node sampling interval to obtain a single blood vessel node on the single blood vessel central line; and
And acquiring a blood vessel diameter change curve and a curvature change curve of each node along the central line, determining nodes corresponding to inflection points on the blood vessel diameter change curve and the curvature change curve, and taking the nodes corresponding to the inflection points as nodes to be predicted, wherein the nodes corresponding to the inflection points belong to single branch nodes of the blood vessel.
Optionally, the determining manner of the bifurcation node includes:
determining a vessel centerline having a superior-inferior relationship based on the vessel hierarchical relationship;
And determining the closest distance point on the upper-level blood vessel central line based on the head node coordinate position on the lower-level blood vessel central line, and determining the closest distance point on the upper-level blood vessel central line as a blood vessel bifurcation node.
Optionally, the preset prediction time period includes a cardiac cycle, and the time to be predicted is at least one time in the cardiac cycle;
Or the preset prediction time period comprises a plurality of cardiac cycles, and the time to be predicted is the critical time of the cardiac cycles;
Inputting initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period, wherein the parameter prediction result comprises the following steps:
inputting the initialization data into a pre-trained graph neural network prediction model to obtain a parameter prediction result corresponding to a first time to be predicted;
updating initialization data based on a parameter prediction result corresponding to the first time to be predicted to obtain input parameter data of the next time to be predicted, inputting the input parameter data of the next time to be predicted to a pre-trained graph neural network prediction model, and obtaining a parameter prediction result corresponding to the second time to be predicted until parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained.
Optionally, the parameter prediction result includes a pressure value;
Determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted, including:
And calculating the ratio of the pressure value of the node to be predicted in the prediction result to the preset pressure value to obtain the fractional flow reserve of the node to be predicted.
According to another aspect of the present invention, there is provided a blood vessel quantization parameter prediction apparatus, comprising:
the data acquisition module is used for acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line;
The initialization data determining module is used for determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted;
The prediction result determining module is used for inputting the initialization data of the nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset time period;
and the blood flow reserve score determining module is used for determining the blood flow reserve score of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
According to another aspect of the present invention, there is provided an electronic apparatus including:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the vessel quantization parameter prediction method of any one of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute a vessel quantization parameter prediction method according to any one of the embodiments of the present invention.
According to the technical scheme, medical image data are acquired, a blood vessel center line is extracted based on the medical image data, and nodes to be predicted are determined based on the blood vessel center line; determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted; inputting initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period; and determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted. The method solves the problems of complex calculation process, high hardware equipment configuration and high calculation cost in the prior art, realizes the determination of the parameter prediction results at a plurality of to-be-predicted moments corresponding to medical image data through a graph neural network prediction model, obtains a transient prediction result, is beneficial to grasping the transient result of a vessel quantization parameter, can improve the prediction speed while meeting the clinical precision requirement, reduces the configuration requirement on the hardware equipment, and reduces the calculation cost.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for predicting a vessel quantization parameter according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for predicting a vessel quantization parameter according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a predicting device for a blood vessel quantification parameter according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device implementing a blood vessel quantization parameter prediction method according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a flowchart of a method for predicting a blood vessel quantization parameter according to an embodiment of the present invention, where the method may be performed by a blood vessel quantization parameter prediction device, which may be implemented in hardware and/or software, and the blood vessel quantization parameter prediction device may be configured in a computer, where the method is applicable to predicting a blood vessel quantization parameter based on medical image data. As shown in fig. 1, the method includes:
s110, acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line.
The medical image data refers to image data obtained through shooting by an image acquisition device, wherein the image scanning device can be a CT scanning device, and the medical image data comprises but is not limited to coronary angiography image data and CTA image data.
Specifically, medical image data are processed through a blood vessel central line extraction method to obtain a blood vessel central line corresponding to a blood vessel in the medical image data, and then sampling is carried out on the blood vessel central line through a preset sampling method to obtain at least one node on the blood vessel central line, and the sampled node is used as a node to be predicted. The blood vessel centerline extraction method may be a pre-trained blood vessel centerline extraction model, or may be a centerline extraction algorithm, which is not limited herein.
In one embodiment, the medical image data may be segmented by using a deep learning method to obtain a preliminary segmented image corresponding to the medical image data, then a vessel center line in the preliminary segmented image is extracted, node division is performed along the vessel center line, and nodes on the vessel center line are selected as nodes to be predicted, and it is to be noted that the number of the selected nodes to be predicted may be determined according to the requirement, and in general, the more the number of the selected nodes to be predicted is, the more accurate the prediction result will be.
S120, initializing data corresponding to different types of nodes to be predicted are determined, wherein the initializing data comprise one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted.
It should be noted that, according to the difference of the positions of the nodes to be predicted on the blood vessel center line, the nodes to be predicted may be divided into different types of nodes, where the node types include a blood vessel inlet node type, a blood vessel outlet node type, a blood vessel bifurcation node, and a blood vessel single branch node. The initialization data may be specifically understood as initial data of associated parameters of different types of nodes on a blood vessel centerline, where the initialization data includes one or more of boundary condition data of a plurality of moments to be predicted, initial parameter data, and attribute information of the nodes to be predicted. The plurality of time points to be predicted can be specifically understood as time points to be predicted, and can be set at any plurality of time points in one cardiac cycle or any plurality of time points in a plurality of continuous cardiac cycles, the setting of the predicted time points is determined according to actual requirements, and it is to be noted that boundary condition data corresponding to each predicted time point needs to be determined while determining the plurality of predicted time points.
Optionally, in the case that the node to be predicted includes a blood vessel inlet node type and a blood vessel outlet node type, the initialization data corresponding to the node to be predicted includes boundary condition data of each time to be predicted; the boundary condition data includes vessel inlet parameters, vessel outlet parameters.
Each vessel center line comprises a vessel inlet node and at least one vessel outlet node, and the vessel inlet node type and the vessel center line node corresponding to the vessel outlet node type can be used as nodes to be predicted. Under the condition that the to-be-predicted nodes of the blood vessel inlet node type and the blood vessel outlet node type are subjected to prediction processing, initializing data corresponding to the to-be-predicted nodes are required to be acquired, and the required initializing data comprise boundary condition data of each to-be-predicted moment; the boundary condition data includes vessel inlet parameters, vessel outlet parameters. Wherein, the vascular inlet parameters include, but are not limited to, vascular inlet flow, vascular inlet pressure, and the vascular outlet parameters include, but are not limited to, vascular outlet flow, vascular outlet resistance. It should be noted that, the number of parameters included in the boundary condition may be set according to actual requirements, and the parameters of the blood vessel inlet and the parameters of the blood vessel outlet may be adjusted in the actual application process, so as to meet the prediction result of accurately predicting the nodes corresponding to the type of the blood vessel inlet node and the type of the blood vessel outlet node.
Specifically, when the node to be predicted includes a blood vessel inlet node type and a blood vessel outlet node type, determining the prediction results of a plurality of prediction moments requires determining initialization data corresponding to the node to be predicted required by each prediction moment, wherein the initialization includes boundary condition data of each prediction moment; the boundary condition data includes vessel inlet parameters, vessel outlet parameters.
In a specific embodiment, parameters such as the vascular inlet flow and the vascular outlet resistance of the patient can be calculated according to the myocardial volume of the patient, the CTA/DSA image or the height and weight of the patient corresponding to the medical image data, so as to obtain boundary conditions, and further, parameters in the blood vessel are initialized according to the obtained boundary conditions of the blood vessel, so that the vascular outlet pressure and the vascular outlet flow are obtained as vascular outlet parameters. The Murray Law algorithm may be used for traffic distribution. The calculation formula for each vessel outlet flow q i is as follows:
wherein i represents the i-th vessel outlet node number, j represents the j-th vessel outlet node number, q represents the total flow of the vessel, a represents the area of the vessel outlet, and α represents the empirical coefficient. After the flow of each vessel outlet node is obtained, the flow of all nodes can be obtained through mass conservation reasoning.
The calculation formula for each vessel outlet pressure p is as follows:
p＝qiRi
Where R i is the resistance of the ith vessel outlet (given by the boundary condition), and q i is the flow of the ith vessel outlet. For each vessel, the pressure difference Δp between every two nodes can be calculated using the following formula:
Where μ is the viscosity coefficient of blood, l is the vessel length, d is the vessel diameter, q i is the flow at the i-th vessel outlet, and according to the above formula, the initial pressure at each node can be calculated.
Determining boundary conditions of a vessel inlet node and a vessel outlet node by the calculation formula
Optionally, in the case that the node to be predicted further includes one or more of a vessel bifurcation node and a vessel single branch node, the initialization data corresponding to the node to be predicted includes one or more of initial parameter data and attribute information of the node to be predicted; the initial parameter data comprise a pressure value and a flow value which correspond to the time to be predicted; the attribute information of the node to be predicted includes geometric information of the node to be predicted.
It will be appreciated that the vessel centerline may include only nodes of the vessel single-branch node type, or may include both nodes of the vessel bifurcation node type and nodes of the vessel single-branch node type. Under the condition that the node to be predicted comprises one or more of a blood vessel bifurcation node and a blood vessel single branch node, initializing parameter data corresponding to the node to be predicted needs to be determined, wherein the initializing parameter data comprises one or more of the initializing parameter data and attribute information of the node to be predicted, and the initializing parameter data is set according to actual requirements. The setting of the time to be predicted specifically refers to any one of a plurality of prediction time points, and preferably, a second prediction time point of the plurality of prediction time points is taken as the setting time to be predicted. The initial parameter data refers to parameter data set for the nodes corresponding to the single node type and the bifurcation node type at the set time to be predicted, and pressure values and flow values of the nodes to be predicted are set, and certain connection relations exist between the nodes to be predicted, so that geometric information of the nodes to be predicted can be input, wherein the geometric information comprises, but is not limited to, distance data between two adjacent nodes to be predicted, cross-sectional areas of blood vessel contours corresponding to the nodes to be predicted, and tangential directions of the nodes to be predicted along a blood vessel central line.
Optionally, the determining manner of the single branch node of the blood vessel includes: for a single blood vessel central line on a blood vessel central line, gradually sampling on the single blood vessel central line according to a preset node sampling interval to obtain a single blood vessel node on the single blood vessel central line; and acquiring a blood vessel diameter change curve and a curvature change curve of each node along the central line, determining nodes corresponding to inflection points on the blood vessel diameter change curve and the curvature change curve, and taking the nodes corresponding to the inflection points as nodes to be predicted, wherein the nodes corresponding to the inflection points belong to single branch nodes of the blood vessel.
Specifically, a single blood vessel central line on the blood vessel central line is obtained according to the classification label on the blood vessel central line, sampling is sequentially carried out along the inlet of the blood vessel central line according to the sampling distance of the preset node, at least one sampling point on the single blood vessel central line is obtained, and preferably, the number of the sampling points on the single blood vessel central line is not less than 2. The preset node sampling distance can be set to a fixed value according to actual needs, and can also be set according to the cross-sectional area of a blood vessel contour corresponding to a node on a single blood vessel central line, and an inlet endpoint of a single blood vessel is taken as a first node to be predicted on the single blood vessel central line, the node sampling distance of a next node is set based on the diameter of the blood vessel contour corresponding to the first node to be predicted, preferably, the node sampling distance is set to be no more than five times the diameter of the blood vessel contour corresponding to the first node to be predicted, a second node to be predicted is obtained based on the obtained node sampling distance, and then the node sampling distance of a next node is set according to the diameter of the blood vessel contour corresponding to the second node to be predicted until sampling processing is completed on the current single blood vessel central line. It can be understood that a node with curvature larger than a preset curvature value exists on a blood vessel central line, so that a blood vessel diameter change curve and a curvature change curve of each node on the central line can be obtained, a node with the blood vessel diameter change curve and the curvature change curve meeting preset thresholds is selected, a node corresponding to an inflection point on the blood vessel diameter change curve and the curvature change curve is obtained, and the node corresponding to the inflection point is taken as a node to be predicted, wherein the node corresponding to the inflection point belongs to a single branch node of the blood vessel. In this embodiment, in addition to obtaining the node to be predicted in a sampling manner according to the preset node sampling interval, a node corresponding to an inflection point on a blood vessel centerline is selected as the node to be predicted, so that the obtained prediction result corresponding to the node to be predicted can better provide a reliable reference value for subsequent medical evaluation.
Optionally, the determining manner of the bifurcation node includes: determining a vessel centerline having a superior-inferior relationship based on the vessel hierarchical relationship; and determining the closest distance point on the upper-level blood vessel central line based on the head node coordinate position on the lower-level blood vessel central line, and determining the closest distance point on the upper-level blood vessel central line as a blood vessel bifurcation node.
Specifically, in the process of extracting the central line of the blood vessel, a corresponding label is set for the corresponding hierarchical relation of each single blood vessel in the central line of the blood vessel, the hierarchical relation between the single blood vessels is used for representing the hierarchical relation between the single blood vessels, in the process of determining the bifurcation node of the blood vessel, the hierarchical relation data of the blood vessels corresponding to the central line of the blood vessel is obtained, the upper and lower level relations between the single blood vessels are identified, the coordinate position of the head node belonging to the central line of the lower level blood vessel is obtained, the distance between the head node and each node on the central line of the upper level blood vessel is calculated, and the nearest distance point between the head node and the central line of the upper level blood vessel is screened out, and is taken as the bifurcation node of the blood vessel. The first-stage blood vessel and the second-stage blood vessel are determined according to the blood vessel hierarchical relationship, the distance between the head node of each second-stage blood vessel and the node on the first-stage blood vessel is calculated, the node with the smallest distance with the first-stage blood vessel is determined, the node corresponding to the smallest distance is taken as a blood vessel bifurcation node, and correspondingly, the node on the second-stage blood vessel corresponding to the smallest distance is also a blood vessel bifurcation node.
In this embodiment, by determining corresponding initialization data for different types of nodes to be predicted, where the initialization data includes one or more of boundary condition data, initial parameter data, and attribute information of the nodes to be predicted at multiple times, a prediction result of the nodes to be predicted at multiple times to be predicted may be achieved, a transient prediction result of the nodes to be predicted may be obtained at the same time, and an influence of an access boundary condition and a bifurcation node on a hydrodynamic parameter in a blood vessel may be better predicted without multiple predictions, thereby avoiding repetitive work and improving prediction efficiency.
S130, inputting initialization data of the nodes to be predicted into a pre-trained graph neural network prediction model to obtain parameter prediction results of at least one moment to be predicted of each node to be predicted in a preset prediction time period.
The pre-trained graph neural network prediction model is obtained by training based on the graph neural network by taking the result of the three-dimensional simulation model as a training label, and can rapidly and accurately predict pressure data, flow data and the like in a blood vessel.
Specifically, initializing data corresponding to the determined nodes to be predicted of different types are used as input parameters, the input parameters are input into a pre-trained graph neural network prediction model, and parameter prediction results of the nodes to be predicted at corresponding moments to be predicted are obtained after model processing. Optionally, the preset prediction time period includes a cardiac cycle, and the time to be predicted is at least one time in the cardiac cycle; or the preset prediction time period comprises a plurality of cardiac cycles, and the time to be predicted is the critical time of the cardiac cycles. That is, at least one time in the same cardiac cycle may be selected, or a plurality of times in a plurality of consecutive cardiac cycles may be selected, without paying attention to whether or not the time to be predicted belongs to the same cardiac cycle.
Preferably, the initialization data of the nodes to be predicted can be input into the pre-trained graphic neural network prediction model in a vector form, namely, the initialization parameters of each node to be predicted are converted into corresponding parameter vectors, and the parameter vectors corresponding to the nodes to be predicted are sequentially input into the pre-trained graphic neural network prediction model to obtain the prediction results of the nodes to be predicted at all the moments to be predicted.
In the embodiment, the initialization parameters of the nodes to be predicted are input into the pre-trained graph neural network prediction model to obtain the prediction results of each node to be predicted at a plurality of times to be predicted, so that the rapidity, the flexibility and the accuracy of determining the prediction results are improved.
And S140, determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
Specifically, after obtaining the parameter prediction result of the node to be predicted at the moment to be predicted, the parameter prediction result may be post-processed according to a preset calculation method to obtain a blood vessel quantization parameter value to be focused, thereby obtaining the fractional flow reserve of the node to be predicted at the moment to be predicted. For example, in the case where the coronary stenosis needs to be evaluated, the parameter prediction result may be post-processed to obtain the fractional flow reserve of the node to be predicted, and since the obtained parameter prediction result of the node to be predicted at the moment to be predicted may include multiple moments, the fractional flow reserve of the node to be predicted at the multiple moments may be determined at the same time, so that the coronary stenosis condition may be better evaluated.
Optionally, the parameter prediction result includes a pressure value; determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted, including: and calculating the ratio of the pressure value of the node to be predicted in the prediction result to the preset pressure value to obtain the fractional flow reserve of the node to be predicted.
Specifically, the fractional flow reserve FFR of the node to be predicted is calculated by the following formula:
Wherein, P d is the pressure value of each node to be predicted, P a is the aortic pressure, and if the calculation includes the aorta, the aortic head node pressure is used; if the aorta is not included, a constant value may be used, wherein the constant value may be determined from a priori experience.
On the basis of the above embodiment, after the parameter prediction result is determined, in addition to calculating the corresponding result according to the fractional flow reserve calculation formula, other vessel quantization parameters may be calculated according to a preset calculation method according to the prediction result.
According to the technical scheme, medical image data are acquired, a blood vessel center line is extracted based on the medical image data, and nodes to be predicted are determined based on the blood vessel center line; determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted; inputting initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period; and determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted. The method solves the problems of complex calculation process, high hardware equipment configuration and high calculation cost in the prior art, and realizes the determination of the parameter prediction results of a plurality of to-be-predicted moments corresponding to medical image data through a graph neural network prediction model, so that the prediction speed and the transient prediction results of to-be-predicted nodes can be improved while the clinical precision requirement is met, the configuration requirement on hardware equipment is reduced, and the calculation cost is reduced.
Example two
Fig. 2 is a flowchart of a method for predicting a vessel quantization parameter according to a second embodiment of the present invention, where the method according to the foregoing embodiment is further optimized, and optionally, initialization data is input into a pre-trained neural network prediction model to obtain a parameter prediction result corresponding to a first time to be predicted; updating initialization data based on a parameter prediction result corresponding to the first time to be predicted to obtain input parameter data of the next time to be predicted, inputting the input parameter data of the next time to be predicted to a pre-trained graph neural network prediction model, and obtaining a parameter prediction result corresponding to the second time to be predicted until parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained. As shown in fig. 2, the method includes:
s210, acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line.
S220, initializing data corresponding to different types of nodes to be predicted are determined, wherein the initializing data comprise one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted.
S230, inputting the initialization data into a pre-trained graph neural network prediction model to obtain a parameter prediction result corresponding to the first time to be predicted.
Specifically, the initialization data is input into a pre-trained graphic neural network prediction model according to a preset format, namely, the boundary condition of the 0 th moment to be predicted in the initialization data, the initial parameter data and the attribute information of the nodes to be predicted are input into the pre-trained graphic neural network prediction model, and a parameter prediction result corresponding to the first moment to be predicted is obtained. Alternatively, the initialization data may be input to the pre-trained neural network prediction model according to a vector format corresponding to each node, or may be converted into a vector matrix and input to the pre-trained neural network prediction model. And are not limited herein.
S240, updating initialization data based on a parameter prediction result corresponding to the first time to be predicted to obtain input parameter data of the next time to be predicted, inputting the input parameter data of the next time to be predicted to a pre-trained graph neural network prediction model and obtaining a parameter prediction result corresponding to the second time to be predicted until parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained.
Specifically, parameters matched with the prediction parameters in the initialization parameters are obtained, the parameter prediction results corresponding to the first time to be predicted are updated to the parameters in the matched initialization data, the updated initialization parameter data are combined with the boundary conditions of the next time to be predicted to obtain the input parameter data of the next time to be predicted, the input parameter data of the next time to be predicted are input into a pre-trained graph neural network prediction model, the parameter prediction results corresponding to the second time to be predicted are obtained after model processing, and the like until the parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained.
S250, determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
According to the technical scheme of the embodiment, the medical image data are acquired, the blood vessel center line is extracted based on the medical image data, and the node to be predicted is determined based on the blood vessel center line. And determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted. Inputting the initialization data into a pre-trained graph neural network prediction model to obtain a parameter prediction result corresponding to the first time to be predicted. Updating initialization data based on a parameter prediction result corresponding to the first time to be predicted to obtain input parameter data of the next time to be predicted, inputting the input parameter data of the next time to be predicted to a pre-trained graph neural network prediction model, and obtaining a parameter prediction result corresponding to the second time to be predicted until parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained. And determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted. The method solves the problems of complex calculation process, high hardware equipment configuration and high calculation cost in the prior art, and realizes the determination of the parameter prediction results of a plurality of to-be-predicted moments corresponding to medical image data through a graph neural network prediction model, so that the prediction speed and the transient prediction results of to-be-predicted nodes can be improved while the clinical precision requirement is met, the configuration requirement on hardware equipment is reduced, and the calculation cost is reduced.
Example III
Fig. 3 is a schematic structural diagram of a predicting device for a blood vessel quantization parameter according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes:
a data acquisition module 310, configured to acquire medical image data, extract a vessel centerline based on the medical image data, and determine a node to be predicted based on the vessel centerline;
The initialization data determining module 320 is configured to determine initialization data corresponding to different types of nodes to be predicted, where the initialization data includes one or more of boundary condition data, initial parameter data, and attribute information of the nodes to be predicted at a plurality of moments to be predicted;
The prediction result determining module 330 is configured to input initialization data of the nodes to be predicted into a pre-trained neural network prediction model to obtain a parameter prediction result of at least one time to be predicted of each node to be predicted within a preset time period;
The fractional flow reserve determining module 340 is configured to determine the fractional flow reserve of the node to be predicted at the time to be predicted based on the parameter prediction result of the node to be predicted at the time to be predicted.
According to the technical scheme, medical image data are acquired through a data acquisition module, a blood vessel center line is extracted based on the medical image data, and nodes to be predicted are determined based on the blood vessel center line; the initialization data determining module determines initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted; the method comprises the steps that a prediction result determining module inputs initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain parameter prediction results of at least one moment to be predicted of each node to be predicted in a preset time period; the fractional flow reserve determining module determines fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted. The method solves the problems of complex calculation process, high hardware equipment configuration and high calculation cost in the prior art, and realizes the determination of the parameter prediction results of a plurality of to-be-predicted moments corresponding to medical image data through a graph neural network prediction model, so that the prediction speed can be improved while the clinical precision requirement is met, the configuration requirement on the hardware equipment is reduced, and the calculation cost is reduced.
Based on the above embodiment, optionally, the initialization data determining module 320 includes: a vessel single-branch node determining unit and a vessel bifurcation node determining unit. The single-branch blood vessel node determining unit is specifically used for sequentially sampling a single-branch blood vessel central line on the blood vessel central line according to a preset node sampling interval to obtain a single-branch blood vessel node on the single-branch blood vessel central line; and acquiring a blood vessel diameter change curve and a curvature change curve of each node along the central line, determining nodes corresponding to inflection points on the blood vessel diameter change curve and the curvature change curve, and taking the nodes corresponding to the inflection points as nodes to be predicted, wherein the nodes corresponding to the inflection points belong to single branch nodes of the blood vessel. The blood vessel bifurcation node determination unit is specifically configured to determine a blood vessel centerline having a superior-inferior relationship based on a blood vessel hierarchical relationship; and determining the closest distance point on the upper-level blood vessel central line based on the head node coordinate position on the lower-level blood vessel central line, and determining the closest distance point on the upper-level blood vessel central line as a blood vessel bifurcation node.
Optionally, the prediction result determining module 330 is specifically configured to:
inputting the initialization data into a pre-trained graph neural network prediction model to obtain a parameter prediction result corresponding to a first time to be predicted;
updating initialization data based on a parameter prediction result corresponding to the first time to be predicted to obtain input parameter data of the next time to be predicted, inputting the input parameter data of the next time to be predicted to a pre-trained graph neural network prediction model, and obtaining a parameter prediction result corresponding to the second time to be predicted until parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained.
Optionally, the fractional flow reserve determination module 340 is specifically configured to:
And calculating the ratio of the pressure value of the node to be predicted in the prediction result to the preset pressure value to obtain the fractional flow reserve of the node to be predicted.
The blood vessel quantization parameter prediction device provided by the embodiment of the invention can execute the blood vessel quantization parameter prediction method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 4 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 4, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the vessel quantization parameter prediction method.
In some embodiments, the vessel quantization parameter prediction method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the vessel quantification parameter prediction method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the vessel quantification parameter prediction method by any other suitable means (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
The computer program for implementing the vessel quantification parameter prediction method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
Example five
The fifth embodiment of the present invention also provides a computer readable storage medium storing computer instructions for causing a processor to execute a blood vessel quantization parameter prediction method, the method comprising:
acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line;
determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted;
Inputting initialization data of nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period;
And determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method for predicting a vessel quantization parameter, comprising:
Acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line;
determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted;
Inputting the initialization data of the nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period;
And determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
2. The method according to claim 1, wherein, in the case that the node to be predicted includes a blood vessel inlet node type and a blood vessel outlet node type, the initialization data corresponding to the node to be predicted includes boundary condition data of each time to be predicted; the boundary condition data includes vessel inlet parameters, vessel outlet parameters.
3. The method according to claim 2, wherein, in the case that the node to be predicted further includes one or more of a vessel bifurcation node and a vessel single branch node, the initialization data corresponding to the node to be predicted includes one or more of the initial parameter data and attribute information of the node to be predicted; the initial parameter data comprise a pressure value and a flow value which correspond to the set moment to be predicted; the attribute information of the node to be predicted comprises the geometric information of the node to be predicted.
4. A method according to claim 3, wherein the manner in which the single vessel node is determined comprises:
For a single blood vessel central line on the blood vessel central line, gradually sampling on the single blood vessel central line according to a preset node sampling interval to obtain a single blood vessel node on the single blood vessel central line; and
And acquiring a blood vessel diameter change curve and a curvature change curve of each node along a central line, determining nodes corresponding to inflection points on the blood vessel diameter change curve and the curvature change curve, and taking the nodes corresponding to the inflection points as the nodes to be predicted, wherein the nodes corresponding to the inflection points belong to the single branch nodes of the blood vessel.
5. A method according to claim 3, wherein the manner of determining the bifurcation node comprises:
determining a vessel centerline having a superior-inferior relationship based on the vessel hierarchical relationship;
And determining the nearest distance point on the upper-level blood vessel central line based on the head node coordinate position on the lower-level blood vessel central line, and determining the nearest distance point on the upper-level blood vessel central line as a blood vessel bifurcation node.
6. The method of claim 1, wherein the preset prediction period comprises a cardiac cycle, and the time to be predicted is at least one time within the cardiac cycle;
or the preset prediction time period comprises a plurality of cardiac cycles, and the moment to be predicted is critical moment of the cardiac cycles;
Inputting the initialization data of the nodes to be predicted into a pre-trained neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset prediction time period, wherein the parameter prediction result comprises:
Inputting the initialization data into the pre-trained graph neural network prediction model to obtain a parameter prediction result corresponding to a first time to be predicted;
updating the initialization data based on the parameter prediction result corresponding to the first time to be predicted to obtain input parameter data of the next time to be predicted, inputting the input parameter data of the next time to be predicted to the pre-trained graph neural network prediction model and obtaining the parameter prediction result corresponding to the second time to be predicted until the parameter prediction results corresponding to the nodes to be predicted at a plurality of times to be predicted are obtained.
7. The method of claim 1, wherein the parameter prediction result comprises a pressure value;
The determining the fractional flow reserve of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted comprises the following steps:
And calculating the ratio of the pressure value of the node to be predicted in the prediction result to a preset pressure value to obtain the fractional flow reserve of the node to be predicted.
8. A vessel quantization parameter prediction apparatus, comprising:
The data acquisition module is used for acquiring medical image data, extracting a blood vessel center line based on the medical image data, and determining a node to be predicted based on the blood vessel center line;
the initialization data determining module is used for determining initialization data corresponding to different types of nodes to be predicted, wherein the initialization data comprises one or more of boundary condition data, initial parameter data and attribute information of the nodes to be predicted at a plurality of moments to be predicted;
The prediction result determining module is used for inputting the initialization data of the nodes to be predicted into a pre-trained graph neural network prediction model to obtain a parameter prediction result of at least one moment to be predicted of each node to be predicted in a preset time period;
and the blood flow reserve score determining module is used for determining the blood flow reserve score of the node to be predicted at the moment to be predicted based on the parameter prediction result of the node to be predicted at the moment to be predicted.
9. An electronic device, the electronic device comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the vessel quantification parameter prediction method of any of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to perform the vessel quantization parameter prediction method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410394420.6A CN118172349A (en) | 2024-04-02 | 2024-04-02 | Vascular quantization parameter prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410394420.6A CN118172349A (en) | 2024-04-02 | 2024-04-02 | Vascular quantization parameter prediction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118172349A true CN118172349A (en) | 2024-06-11 |
Family
ID=91358106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410394420.6A Pending CN118172349A (en) | 2024-04-02 | 2024-04-02 | Vascular quantization parameter prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118172349A (en) |
-
2024
- 2024-04-02 CN CN202410394420.6A patent/CN118172349A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113705628B (en) | Determination method and device of pre-training model, electronic equipment and storage medium | |
| CN117274266B (en) | Method, device, equipment and storage medium for grading acne severity | |
| CN116629620A (en) | Risk level determining method and device, electronic equipment and storage medium | |
| US20150278976A1 (en) | Systems and methods for using geometry sensitivity information for guiding workflow | |
| CN117351307A (en) | Model training method, device, equipment and storage medium | |
| CN118172349A (en) | Vascular quantization parameter prediction method and device, electronic equipment and storage medium | |
| CN115482358B (en) | Triangular mesh curved surface generation method, device, equipment and storage medium | |
| CN115762798A (en) | Method and device for determining hemodynamic parameters, storage medium and electronic equipment | |
| CN114972220B (en) | Image processing method and device, electronic equipment and readable storage medium | |
| CN116172598A (en) | Fractional flow reserve determination method, fractional flow reserve determination device, electronic equipment and storage medium | |
| CN115861189A (en) | Image registration method and device, electronic equipment and storage medium | |
| CN115482261A (en) | Blood vessel registration method, device, electronic equipment and storage medium | |
| CN114972165A (en) | Method and device for measuring time-average shearing force | |
| CN116245853A (en) | Fractional flow reserve determination method, fractional flow reserve determination device, electronic equipment and storage medium | |
| CN114972242B (en) | Training method and device for myocardial bridge detection model and electronic equipment | |
| CN117688865A (en) | Microcirculation resistance index determination method, device, equipment and medium | |
| CN118172348A (en) | Method and device for predicting shearing stress of wall surface of blood vessel, electronic equipment and storage medium | |
| CN117688866A (en) | Microcirculation resistance reserve fraction obtaining method, device, equipment and medium | |
| CN117688867A (en) | Method, device, equipment and medium for determining microcirculation resistance reserve fraction | |
| CN117372399A (en) | Postoperative detection method, device and equipment for aortic dissection and storage medium | |
| CN110742688A (en) | Blood vessel model establishing method and device and readable storage medium | |
| CN116721148A (en) | Center line generation method and device, electronic equipment and storage medium | |
| CN115496007A (en) | Method and device for determining blood flow mechanical energy loss, electronic equipment and medium | |
| CN116933896B (en) | Super-parameter determination and semantic conversion method, device, equipment and medium | |
| CN115690143B (en) | Image segmentation method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |