CN113995388B

CN113995388B - Fractional flow reserve calculation method and device, electronic equipment and readable storage medium

Info

Publication number: CN113995388B
Application number: CN202111522999.2A
Authority: CN
Inventors: 赵宏凯; 白彬
Original assignee: Shukun Beijing Network Technology Co Ltd
Current assignee: Shukun Technology Co ltd; Yukun Beijing Network Technology Co ltd
Priority date: 2021-12-13
Filing date: 2021-12-13
Publication date: 2022-06-03
Anticipated expiration: 2041-12-13
Also published as: CN113995388A

Abstract

The application discloses a fractional flow reserve calculation method, a fractional flow reserve calculation device, electronic equipment and a readable storage medium. According to the method, the pressure drop predicted value is determined through the artery characteristic parameters after dimensionality reduction, so that compared with a three-dimensional fluid mechanics calculation method, the calculation speed is higher, the pressure drop predicted value is subjected to pressure drop correction value prediction processing, the pressure drop predicted value is corrected through the obtained pressure drop correction value, the defect that the pressure drop predicted value is not accurate is overcome, and therefore the whole process of calculating the blood flow reserve fraction not only ensures that the calculation speed is superior to that of a traditional method, but also ensures that the accuracy of the blood flow reserve fraction obtained through final calculation is high.

Description

Fractional flow reserve calculation method and device, electronic equipment and readable storage medium

Technical Field

The application relates to the technical field of medical treatment, in particular to a fractional flow reserve calculation method, a fractional flow reserve calculation device, electronic equipment and a readable storage medium.

Background

Fractional Flow Reserve (FFR) is defined as the ratio of the maximum blood Flow that a diseased vessel can provide to the maximum blood Flow that the vessel can provide when it is fully normal, and is commonly measured by measuring the ratio of the pressure at the distal end of a diseased stenosis to the pressure at the proximal end of an arterial artery over a pressure-interventional guidewire or microcatheter at maximum hyperemia. This index is considered to be the "gold standard" for determining the degree of ischemic lesion.

However, in the currently adopted measurement method, the invasive measurement method may cause damage to the human body, while the current non-invasive measurement method cannot simultaneously guarantee the calculation speed and the calculation accuracy, and the practicability is not strong, so that a blood flow reserve fraction calculation method capable of guaranteeing both the calculation speed and the calculation accuracy is required.

Disclosure of Invention

The application provides a fractional flow reserve calculation method, a device, an electronic device and a readable storage medium, and aims to solve the problem that a fractional flow reserve calculation method which can not only ensure the calculation speed but also ensure the calculation accuracy is needed.

In a first aspect, the present application provides a fractional flow reserve calculation method, including:

acquiring artery characteristic parameters of a target lesion artery;

determining a pressure drop prediction value of a target area in the target focus artery according to the artery characteristic parameters subjected to dimensionality reduction through a preset pressure drop prediction model;

performing pressure drop loss prediction processing on the pressure drop prediction value to obtain a pressure drop correction value;

and determining the fractional flow reserve of the target area according to the predicted pressure drop value and the corrected pressure drop value.

In a possible implementation manner of the present application, the preset pressure drop prediction model is one of a machine learning model, a deep learning model, and a dimension reduction model.

In a possible implementation manner of the present application, the performing pressure drop loss prediction processing on the pressure drop prediction value to obtain a pressure drop correction value includes:

inputting the artery characteristic parameters subjected to the dimension reduction into a trained pressure drop correction model to obtain a pressure drop correction value of the pressure drop prediction value, wherein the pressure drop correction model is a machine learning model.

In a possible implementation manner of the present application, before the inputting the artery characteristic parameter into a trained pressure drop correction model to obtain a pressure drop correction value of the pressure drop prediction value, the method further includes:

acquiring a preset first training parameter, wherein the training parameter comprises a first artery characteristic training parameter of a first training artery and a real pressure drop correction value of a lesion area in the first training artery, the real pressure drop correction value is obtained by calculation according to a first pressure drop real value and a first pressure drop predicted value, and the first pressure drop predicted value is obtained through a preset pressure drop prediction model according to the first artery characteristic training parameter after dimensionality reduction;

inputting the first artery characteristic training parameters subjected to dimensionality reduction into a preset pressure drop correction model to obtain a sample pressure drop correction value of a lesion area in the first training artery;

and adjusting parameters in the preset pressure drop correction model according to the real pressure drop correction value and the sample pressure drop correction value to obtain the trained pressure drop correction model.

In a possible implementation manner of the present application, the determining, by using a preset pressure drop prediction model, a predicted pressure drop value of a target region in an artery of a target lesion according to the artery feature parameter after dimension reduction includes:

inputting the artery characteristic parameters subjected to dimension reduction into a preset pressure drop prediction model to obtain a predicted value of pressure drop to be processed in a target region in the target lesion artery;

and inputting the predicted value of the pressure drop to be processed, the artery characteristic parameters after dimensionality reduction and the artery characteristic parameters into a trained pressure drop correction dimensionality increasing model to obtain a predicted value of the pressure drop of the target area.

In a possible implementation manner of the present application, before the inputting the predicted value of the pressure drop to be processed, the artery characteristic parameter after the dimensionality reduction, and the artery characteristic parameter into a trained pressure drop correction dimension-increasing model to obtain a predicted value of the pressure drop of a target area, the method further includes:

acquiring a preset second training parameter, wherein the second training parameter comprises a second artery feature training parameter of a second training artery, a second pressure drop true value and a second pressure drop predicted value of a lesion region in the second training artery, and the second pressure drop predicted value is obtained according to the second artery feature training parameter after dimension reduction through the pressure drop prediction model;

inputting the second artery characteristic training parameter, the second artery characteristic training parameter and the second pressure drop predicted value after dimensionality reduction into a preset pressure drop correction dimension increasing model to obtain a dimension increasing pressure drop predicted value of the second training artery;

and adjusting parameters in the preset pressure drop correction dimensionality model according to the predicted pressure drop dimensionality value and the second pressure drop dimensionality value to obtain the trained pressure drop correction dimensionality model.

In one possible implementation of the present application, the pressure drop correction dimensionality reduction model is one of a machine learning model and a deep learning model.

In one possible implementation manner of the present application, the determining the fractional flow reserve of the target region according to the predicted pressure drop value and the corrected pressure drop value includes:

determining a second inlet blood pressure of the target area according to a first inlet blood pressure of a target lesion artery inlet in the artery characteristic parameters;

calculating to obtain outlet blood pressure of the target area according to the second inlet blood pressure, the pressure drop predicted value and the pressure drop corrected value;

and calculating the fractional flow reserve of the target area according to the first inlet blood pressure and the outlet blood pressure.

In a possible implementation manner of the present application, the determining a second entry blood pressure of the target region according to a first entry blood pressure of an artery entry of a target lesion in the artery feature parameter includes:

determining a blood pressure attenuation value between the target region and the entrance of the target lesion artery according to the artery characteristic parameter and a preset attenuation value calculation relation;

and determining a second inlet blood pressure of the target region according to the blood pressure attenuation value and the first inlet blood pressure of the target lesion artery inlet in the artery characteristic parameters.

In a possible implementation manner of the present application, the acquiring the artery characteristic parameters of the target focal artery includes:

acquiring a medical image of a target lesion artery;

performing segmentation processing on the medical image to obtain a target area image and an artery geometric parameter of a target focus artery in the target area image;

and acquiring the artery physiological parameters of the target lesion artery through a preset physiological detection component.

In a possible implementation manner of the present application, after determining the fractional flow reserve of the target region according to the predicted pressure drop value and the corrected pressure drop value, the method further includes:

inquiring a preset database to obtain target patient information corresponding to the target lesion artery;

and displaying the fractional flow reserve and the target patient information in a preset display terminal.

In a second aspect, the present application provides a fractional flow reserve calculation apparatus, comprising:

the acquisition unit is used for acquiring the artery characteristic parameters of the target lesion artery;

the determining unit is used for determining a pressure drop prediction value of a target area in the target focus artery according to the artery characteristic parameters subjected to dimension reduction through a preset pressure drop prediction model;

the deviation prediction unit is used for carrying out pressure drop loss prediction processing on the pressure drop prediction value to obtain a pressure drop correction value;

and the fraction determining unit is used for determining the fractional flow reserve of the target area according to the predicted pressure drop value and the pressure drop correction value.

In one possible implementation manner of the present application, the deviation prediction unit is further configured to:

inputting the first artery characteristic training parameters subjected to the dimensionality reduction into a preset pressure drop correction model to obtain a sample pressure drop correction value of a lesion area in the first training artery;

In one possible implementation manner of the present application, the determining unit is further configured to:

acquiring a preset second training parameter, wherein the second training parameter comprises a second artery feature training parameter of a second training artery, a second pressure drop true value and a second pressure drop predicted value of a lesion region in the second training artery, and the second pressure drop predicted value is obtained according to the second artery feature training parameter after dimensionality reduction through the pressure drop prediction model;

In one possible implementation manner of the present application, the score determining unit is further configured to:

determining a second entrance blood pressure of the target area according to a first entrance blood pressure of a target lesion artery entrance in the artery characteristic parameters;

In a possible implementation manner of the present application, the obtaining unit is further configured to:

acquiring a medical image of a target lesion artery;

In one possible implementation manner of the present application, the fractional flow reserve calculation apparatus further includes a display unit, and the display unit is configured to:

In a third aspect, the present application further provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and when the processor calls the computer program in the memory, the processor performs any of the steps in the fractional flow reserve calculation methods provided in the present application.

In a fourth aspect, the present application further provides a readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps in any of the fractional flow reserve calculation methods provided in the present application.

To sum up, the fractional flow reserve calculation method provided by the embodiment of the present application includes: acquiring artery characteristic parameters of a target lesion artery; determining a pressure drop prediction value of a target area in the target focus artery according to the artery characteristic parameters subjected to dimensionality reduction through a preset pressure drop prediction model; performing pressure drop loss prediction processing on the pressure drop prediction value to obtain a pressure drop correction value; and determining the fractional flow reserve of the target area according to the predicted pressure drop value and the corrected pressure drop value. Therefore, the blood flow reserve fraction calculating method provided by the embodiment of the application determines the pressure drop predicted value through the artery characteristic parameters after dimensionality reduction, so that the calculating speed is higher than that of a three-dimensional fluid mechanics calculating method, the pressure drop predicted value is subjected to pressure drop correction value prediction processing to obtain a pressure drop correction value, the pressure drop predicted value is corrected through the obtained pressure drop correction value, and the defect that the pressure drop predicted value is inaccurate is overcome, so that the whole blood flow reserve fraction calculating process not only ensures that the calculating speed is better than that of a traditional method, but also ensures that the accuracy of the finally calculated blood flow reserve fraction is high.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic view of an application scenario of a fractional flow reserve calculation method provided in an embodiment of the present application;

fig. 2 is a schematic flow chart of a fractional flow reserve calculation method provided in an embodiment of the present application;

FIG. 3 is a schematic diagram of a dimension reduction model provided in an embodiment of the present application;

FIG. 4 is a schematic flow chart of calculating fractional flow reserve from inlet blood pressure provided in an embodiment of the present application;

FIG. 5 is a schematic flow chart illustrating a process of calculating a second inlet blood pressure from a blood pressure attenuation value according to an embodiment of the present application;

fig. 6 is a schematic flow chart of displaying fractional flow reserve provided in an embodiment of the present application;

fig. 7 is a schematic structural diagram of an embodiment of a fractional flow reserve calculation apparatus provided in an embodiment of the present application;

fig. 8 is a schematic structural diagram of an embodiment of an electronic device provided in the embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

In the description of the embodiments of the present application, it should be understood that the terms "first", "second", and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, features defined as "first", "second", may explicitly or implicitly include one or more of the described features. In the description of the embodiments of the present application, "a plurality" means two or more unless specifically defined otherwise.

The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for the purpose of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known processes have not been described in detail so as not to obscure the description of the embodiments of the present application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed in the embodiments herein.

The embodiment of the application provides a fractional flow reserve calculation method, a fractional flow reserve calculation device, electronic equipment and a readable storage medium. The fractional flow reserve calculation apparatus may be integrated in an electronic device, and the electronic device may be a server or a terminal.

An execution main body of the blood flow reserve fraction calculation method according to the embodiment of the present application may be the blood flow reserve fraction calculation apparatus provided in the embodiment of the present application, or different types of electronic devices such as a server device, a physical host, or a User Equipment (UE) integrated with the blood flow reserve fraction calculation apparatus, where the blood flow reserve fraction calculation apparatus may be implemented in a hardware or software manner, and the UE may specifically be a terminal device such as a smart phone, a tablet computer, a notebook computer, a palm computer, a desktop computer, or a Personal Digital Assistant (PDA).

The electronic device may adopt a working mode of independent operation, or may also adopt a working mode of a device cluster.

Before introducing the fractional flow reserve calculation method, apparatus, electronic device, and readable storage medium provided in the embodiments of the present application, first, the technical background of the embodiments of the present application is introduced:

for calculating the fractional flow reserve, the current mainstream methods include two methods: invasive and non-invasive.

The invasive measurement method obtains information inside the human body mainly by placing a specific measurement tool inside the human body. However, this measurement method has certain limitations. On one hand, the invasive measuring tool is expensive, which increases the extra cost of the patient and prolongs the time of the interventional operation; on the other hand, when measurement is carried out, vasodilatation medicines such as adenosine and the like are possibly injected into a patient body, and the medicines have certain damage to the human body and are not suitable for certain patient groups (such as liver and kidney insufficiency, medicine allergy and the like).

Current non-invasive measurements include coronary-based methods of calculating FFR. After reconstructing the vessel based on medical images of coronary CT angiography (CCTA) or Digital Subtraction Angiography (DSA), the FFR can be evaluated based on a method of calculating three-dimensional hydrodynamics. However, the method based on computational fluid dynamics alone consumes large computational resources and has long operation time, and thus, the method faces the problem of long data discharge and report time, and the clinical practicability of the method is reduced.

To solve this problem, a dimension reduction model and machine learning for calculating FFR are proposed for reducing the time required for calculation. However, both methods have their limitations, in that the dimension reduction model loses high-dimensional information, and the machine learning model requires a large amount of clinical data, which both image the accuracy of its results.

The method provided by the embodiment of the application calculates the initial FFR by using the dimensionality reduction model/machine learning model with high calculation efficiency, then performs deviation correction on the initial FFR to obtain the accurate FFR, overcomes the defects of the prior art, and provides a new blood flow reserve fraction calculation method based on coronary angiography by combining the advantages of the dimensionality reduction model and the machine learning model, so that the pressure drop distribution and the FFR of the whole blood vessel can be quickly and accurately obtained.

Referring to fig. 1, fig. 1 is a schematic view of a fractional flow reserve calculation system according to an embodiment of the present disclosure. The fractional flow reserve calculation system may include an electronic device 100, and the fractional flow reserve calculation apparatus is integrated in the electronic device 100.

In addition, as shown in fig. 1, the fractional flow reserve calculation system may further include a memory 200 for storing data, such as text data.

It should be noted that the scenario diagram of the fractional flow reserve calculation system shown in fig. 1 is merely an example, and the fractional flow reserve calculation system and the scenario described in the embodiment of the present application are for more clearly illustrating the technical solution of the embodiment of the present application, and do not form a limitation on the technical solution provided in the embodiment of the present application, and it is known by those skilled in the art that as the fractional flow reserve calculation system evolves and new business scenarios appear, the technical solution provided in the embodiment of the present invention is also applicable to similar technical problems.

In the following, a method for calculating fractional flow reserve provided in an embodiment of the present application is described, where an electronic device is used as an execution subject, and the execution subject will be omitted in subsequent embodiments of the method for simplifying and facilitating description.

Referring to fig. 2, fig. 2 is a schematic flow chart of a fractional flow reserve calculation method according to an embodiment of the present application. It should be noted that, although a logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different than that shown or described herein. The fractional flow reserve calculation method may specifically include the following steps 201 to 204, where:

201. and acquiring the artery characteristic parameters of the target lesion artery.

The target focal artery refers to a diseased coronary artery of the patient. The fractional flow reserve may be used to evaluate the lesion status of the coronary artery when there is a stenosis in the coronary artery, and thus, in the embodiment of the present application, the target lesion artery may refer to the coronary artery having a stenosis region.

The artery characteristic parameter refers to the parameter of the target lesion artery. The artery characteristic parameters may include both the geometric parameters of the artery vessel and the operating parameters of the artery vessel. For example, the artery characteristic parameters may include the cross-sectional dimension of each point in the artery vessel, the length between each point in the artery vessel, and parameters related to the spatial distribution and size of each blood vessel branch in the target lesion artery, such as the number of blood vessel branches in the artery vessel and the position of the blood vessel branch, and the artery characteristic parameters may also include working parameters related to the operation of the artery vessel, such as cardiac output, systolic pressure, and the like, which may also be referred to as physiological parameters of the artery vessel, and the working parameters may characterize the working capacity of the target lesion artery, for example, when the cardiac output is low, it may indicate that the capacity of the target lesion artery for pumping blood is insufficient.

The method for acquiring the artery characteristic parameters of the target lesion artery is not limited in the embodiments of the present application. Illustratively, the artery characteristic parameters of the target lesion artery can be characterized by medical images, and the method for obtaining the artery characteristic parameters is specifically exemplified as follows:

and (1.1) acquiring a 2D or 3D medical image through a preset image acquisition device. For example, the medical image may be a tissue image of a location such as the heart, head and neck.

And (1.2) segmenting the medical image to obtain a target area image. For example, when the medical image is a physiological tissue image of a heart position, a coronary vessel region in the medical image may be segmented by using a segmentation model constructed by a deep learning neural network model (U-net network) to obtain a 2D or 3D target region image, where the target region image includes artery characteristic parameters of a target lesion artery.

And (1.3) carrying out dimensionality reduction on the target region image to obtain the geometric parameters of the artery vessel. For example, the geometric information of the arterial vessel may be acquired manually or automatically.

(1.3) the operating parameters of the arterial vessel are obtained by biomedical tests, for example, the blood flow in the arterial vessel can be measured by a Color Velocity Image Quality (CVIQ) method.

It should be noted that, in the embodiment of the present application, the order of acquiring each parameter in the artery characteristic parameters is not limited, and the step (1.1) -the step (1.2) may be executed first, and then the step (1.3) is executed, or the step (1.3) may be executed first, then the step (1.1) -the step (1.2) is executed, or the step (1.1) -the step (1.2) and the step (1.3) may be executed at the same time.

Or, the patient can be detected by a preset physiological detection component, so that the working parameters of the artery vessel in the artery characteristic parameters can be obtained. At this time, the acquiring of the artery characteristic parameter of the target lesion artery includes:

and (2.1) acquiring a medical image of the target lesion artery.

And (2.2) carrying out segmentation processing on the medical image to obtain a target region image and an artery geometric parameter of a target focus artery in the target region image.

And (2.3) acquiring the artery physiological parameters of the target lesion artery through a preset physiological detection component.

The steps (2.1) to (2.2) correspond to the steps (1.1) to (1.2), and in the step (2.3), the preset physiological detection component may be a component capable of measuring a working parameter of a target focal artery, such as a blood pressure meter, a heart rate detector, or the like, or also referred to as an artery physiological parameter.

In some embodiments, the artery characteristic parameters may also include parameters that cannot be directly measured. Illustratively, the artery feature parameters may include boundary parameters of the target lesion artery. For example, the artery characteristic parameters may include the inlet blood flow of the target lesion artery. Specifically, the inlet blood flow of the target lesion artery can be calculated by equation (1):

Q_in0.045CO formula (1)

Wherein Q is_inFor inlet blood flow, CO is cardiac output.

For another example, the artery characteristic parameter may further include an entry blood pressure of the target lesion artery. Specifically, the inlet blood pressure of the target lesion artery can be calculated by equation (2):

where MAP is inlet pressure, DBP is diastolic pressure, HR is heart rate, and SBP is systolic pressure.

For another example, the artery characteristic parameter may further include an outlet blood flow of the target lesion artery. Specifically, the outlet blood flow of each blood vessel branch of the artery of the target lesion can be calculated by the following formula (3):

wherein Q is_outiThe blood flow of the outlet of the i-th section of the blood vessel branch in the target focal artery, d_iIs the diameter, k, of the i-th vessel branch in the target focal artery₁Is a preset coefficient.

The purpose of calculating the boundary parameters is to provide boundary conditions for constraints in the subsequent prediction of pressure drop correction values, and to improve the accuracy of the calculation in the subsequent calculation of fractional flow reserve.

202. And determining a pressure drop prediction value of a target area in the target focus artery according to the artery characteristic parameters subjected to dimension reduction through a preset pressure drop prediction model.

And the dimension reduction of the artery characteristic parameters refers to the dimension reduction of the artery geometric parameters in the artery characteristic parameters. As illustrated below, when the artery geometric parameters are obtained by the methods from step (1.1) to step (1.2), the artery geometric parameters correspond to geometric parameters of the target focal artery in 2D or 3D, and if the geometric parameters in 2D or 3D are used for calculation, the calculation amount is large, which affects the calculation speed of the fractional flow reserve, so that the artery geometric parameters in 2D or 3D can be reduced to one dimension to calculate the fractional flow reserve.

The pressure drop prediction model is used for predicting the pressure drop value of the narrow area, the artery characteristic parameters are input into the pressure drop prediction model, and the pressure drop prediction model can predict and output the pressure drop value according to the artery characteristic parameters. In the embodiment of the application, in order to overcome the defect of slow calculation speed in the prior art, one of a machine learning model, a deep learning model or a dimension reduction model is adopted as a pressure drop prediction model.

The dimension reduction model is used for simplifying each arterial vessel in the target lesion artery into a one-dimensional linear structure according to the geometric parameters after dimension reduction in the artery characteristic parameters, but still keeping the model of the three-dimensional geometric information of the target lesion artery, and obtaining a pressure reduction value according to the one-dimensional linear structure, the three-dimensional geometric information and other parameters in the artery characteristic parameters by a computational method of hydrodynamics. Referring to fig. 3, fig. 3 shows a schematic diagram obtained by simplifying a target lesion artery through a dimension reduction model, in which a three-dimensional blood vessel is reduced into a blood vessel center line formed by connecting blood vessel center line points, including the distribution of each artery blood vessel in the target lesion artery, the size of each artery blood vessel and the position of a target region, so that the geometric parameters of the artery blood vessels in the artery characteristic parameters can be obtained through fig. 3.

The machine learning model is a model for predicting FFR from artery characteristic parameters by a machine learning method. Specifically, the machine learning model may be one of Support Vector Machine (SVM), multi-layer perceptron (MLP), multiple linear regression Model (MVR), neural network (neural network), decision tree (tree-based classifier), weighted linear regression (weighted linear), and logistic regression (logistic regression). For example, the artery characteristic parameters may be input into a machine learning model formed by a Convolutional Neural Network (CNN), and subjected to processing at a Convolutional layer, a pooling layer, and the like in the Convolutional Neural network, and then the obtained characteristics may be predicted to obtain a predicted pressure drop value.

In some embodiments, the predicted pressure drop value may be calculated by equation (4) -equation (8):

where Δ P is a predicted value of pressure drop, μ is a kinetic viscosity coefficient of blood, ρ is a density of blood, and is an average blood flow in one cardiac cycle, D0 is a normal diameter of a blood vessel, and Q is a blood flow in an artery of a target lesion, which may be represented by Q in equation (1) in the embodiment of the present application_inThe blood flow at other positions in the target lesion artery, or the blood flow in the target lesion artery obtained by calculation methods such as average and variance may be used. K_vTo express the coefficient of viscous action, K_tTo express the coefficient by which stenosis leads to a change in the flow structure, K_uTo express the coefficient of the periodic pressure change caused by unsteady flow, K_cTo represent the additional pressure reduction factor due to unsteady flow, L_sIs the length of the target area, A₀Is the cross-sectional area of a normal blood vessel in the target focal artery, A_sIs the minimum cross-sectional area in the target region, r (x) is the radius of the vessels in the target region, and f is the heart rate in the arterial characteristic parameter. A. the₀Can be determined by the area of the blood vessel in the target region, e.g., the area of the upstream start and downstream end of the blood vessel in the target region can be summed and averaged to obtain A₀The diameter of the normal blood vessel is gradually reduced from the upstream to the downstream, so that the area of the upstream starting part and the area of the downstream ending part of the blood vessel in the target area are summed and averaged to be closer to the real human condition.

It can be seen that, in the embodiment of the present application, since the statistical model is used as the pressure drop prediction model, the fluid equation is solved for the infinitesimal volume in the entire region without using a three-dimensional fluid mechanics calculation method, and after the solution is converged, the required information such as pressure distribution is integrated and extracted, so that the consumption of calculation resources is low and the calculation time is short.

The target region is a stenotic region in a target focal artery, and there may be only one or a plurality of target regions.

The pressure drop prediction value is a pressure drop value output by a pressure drop prediction model and represents the pressure drop condition between an inlet and an outlet of the target area. In some embodiments, the predicted pressure drop prediction values obtained through prediction may be multiple, that is, the preset pressure drop prediction model outputs a sequence of pressure drop prediction values, where the sequence may be one or multiple, one sequence includes pressure drop prediction values of multiple blood vessel center points in one blood vessel in the target focal artery, and when the target focal artery includes m blood vessels, the output sequence includes m sequences.

In some embodiments, the output value of the preset pressure drop prediction model may be subjected to dimension increase to obtain a more accurate pressure drop prediction value. At this time, the step of "determining a predicted value of pressure drop of a target region in the target lesion artery according to the artery characteristic parameter after dimension reduction through a preset pressure drop prediction model" includes:

(1) inputting the artery characteristic parameters subjected to the dimension reduction into a preset pressure drop prediction model to obtain a predicted value of the pressure drop to be processed in a target region in the target lesion artery.

(2) And inputting the predicted value of the pressure drop to be processed, the artery characteristic parameters after dimensionality reduction and the artery characteristic parameters into a trained pressure drop correction dimensionality increasing model to obtain a predicted value of the pressure drop of the target area.

The pressure drop correction dimension increasing model refers to a model for increasing the dimension of data after dimension reduction. Illustratively, the pressure drop correction dimension-increasing model may obtain a mapping relationship between a geometric parameter of the target focal artery in a low-dimensional space and a geometric parameter in a high-dimensional space according to the reduced-dimension artery characteristic parameter and the unreduced artery characteristic parameter, and increase a predicted value of the pressure drop to be processed to a dimension corresponding to the artery characteristic parameter according to the mapping relationship. For example, when the artery characteristic parameter is a parameter of a target focus artery in a three-dimensional space, and the artery characteristic parameter after dimensionality reduction is a parameter of the target focus artery in a one-dimensional space, the dimension-increasing model can be corrected through pressure drop, and the predicted value of the pressure drop to be processed in the one-dimensional space is increased to the predicted value of the pressure drop in the three-dimensional space.

The reason for increasing the dimension is that a large amount of high-dimensional information is lost during dimension reduction, and if only the output of the pressure drop prediction model is adopted as the pressure drop prediction value, the calculated fractional flow reserve is inaccurate, so that the result after dimension increase can be used as the pressure drop prediction value, and the calculation accuracy of the fractional flow reserve is improved.

The trained pressure drop correction dimensionality-increase model can be obtained by training in the following mode:

(1) and acquiring a preset second training parameter, wherein the second training parameter comprises a second artery feature training parameter of a second training artery, a second pressure drop true value and a second pressure drop predicted value of a lesion region in the second training artery, and the second pressure drop predicted value is obtained according to the second artery feature training parameter after dimensionality reduction through the pressure drop prediction model.

(2) Inputting the second artery characteristic training parameter, the second artery characteristic training parameter and the second pressure drop predicted value after dimensionality reduction into a preset pressure drop correction dimension increasing model to obtain a dimension increasing pressure drop predicted value of the second training artery.

(3) And adjusting parameters in the preset pressure drop correction dimensionality model according to the predicted pressure drop dimensionality value and the second pressure drop dimensionality value to obtain the trained pressure drop correction dimensionality model.

203. And carrying out pressure drop correction value prediction processing on the pressure drop prediction value to obtain a pressure drop correction value.

The pressure drop correction value prediction processing refers to the deviation between a pressure drop prediction value output by the prediction pressure drop prediction model and an actual true value, and the pressure drop correction value is a result obtained through the pressure drop correction value prediction processing.

In the embodiment of the application, in order to ensure the calculation speed of the fractional flow reserve, a trained machine learning model may be used to perform pressure drop correction value prediction processing on the pressure drop prediction value. Before step 203 is performed, the preset machine learning model may be trained according to preset training parameters, and the training parameters provided to the preset machine learning model may be based on experimental results of real coronary arteries or artificially constructed coronary arteries. For example, a plurality of artificial coronary arteries including a stenosis region may be obtained in advance through a three-dimensional modeling method, then a pressure drop simulation value, i.e., an actual real value, of the stenosis region in each artificial coronary artery is obtained through simulation, and a parameter of each artificial coronary artery is input into the pressure drop prediction model, so as to obtain a predicted value output by the pressure drop prediction model. After the predicted value is obtained, the predicted value, the parameter corresponding to the artificial coronary artery and the pressure drop simulation value are input into a preset machine learning model, and the machine learning model is trained so that the pressure drop correction value can be predicted according to the input coronary artery parameter and the pressure drop prediction value.

Specifically, the machine learning model may be trained by:

and (2.1) obtaining a preset first training parameter, wherein the training parameter comprises a first artery characteristic training parameter of a first training artery and a real pressure drop correction value of a lesion area in the first training artery, the real pressure drop correction value is obtained by calculation according to a first pressure drop real value and a first pressure drop predicted value, and the first pressure drop predicted value is obtained through a preset pressure drop prediction model according to the first artery characteristic training parameter after dimension reduction.

The meaning of the artery feature training parameter may refer to the description of the artery feature parameter in step 201, where the artery feature training parameter is a feature parameter of a training artery used for training, and the training artery may be an artificial coronary artery as described above or a real coronary artery. In the artery feature training parameters, parameters which cannot be directly measured, such as geometric parameters of the artery vessel, working parameters of the artery vessel, boundary parameters of the artery vessel and the like, can also be included.

The actual value of the pressure drop is obtained by a three-dimensional Computational Fluid Dynamics (CFD) method or a method capable of accurately obtaining the pressure drop value by an invasive method or the like when the training artery is a real coronary artery, and it can be considered that the real pressure drop value represents a real situation in a stenosis region of the training artery. Illustratively, the true pressure drop value can be obtained by performing a simulation by a three-dimensional computational fluid dynamics method such as a finite volume method, a finite element method, a finite difference method, an immersion boundary method, a lattice boltzmann method, and a smooth particle dynamics method. For example, the training artery may be calculated by simulation using a finite volume method. The boundary condition of the inlet can be selected from a speed boundary condition, the flow of the blood vessel section is converted into the flow velocity through the cross-sectional area of the inlet, and the boundary condition of the outlet can be set as a zero gradient boundary condition. The fluid mechanics control equation, Navier-Stokes equation, can be solved by a Semi-Implicit Method (SIMPLE) algorithm of a steady-state Pressure coupled equation set. And acquiring the pressures of the upstream of the narrow region and the downstream of the narrow region in the training artery after simulation, wherein the difference value is a real pressure drop value of the fluid mechanics simulation by utilizing three-dimensional calculation.

The real pressure drop correction value may be a difference between the first real pressure drop value and the first predicted pressure drop value.

And (2.2) inputting the first artery characteristic training parameters subjected to the dimension reduction into a preset pressure drop correction model to obtain a sample pressure drop correction value of a lesion area in the first training artery.

And (2.3) adjusting parameters in the preset pressure drop correction model according to the real pressure drop correction value and the sample pressure drop correction value to obtain the trained pressure drop correction model.

204. And determining the fractional flow reserve of the target area according to the predicted pressure drop value and the corrected pressure drop value.

According to the predicted pressure drop value and the corrected pressure drop value, the real pressure drop value of the target area can be calculated, for example, the real pressure drop value of the target area can be calculated by adopting an equation (9):

ΔP＝ΔP_1D+Δ_ΦP_mlequation (9)

Wherein, the delta P is the real pressure drop value of the target area, and the delta P_1DFor the predicted value of pressure drop, Δ_ΦP_mlIs a pressure drop correction value.

According to the real pressure drop value of the target area, the fractional flow reserve can be obtained according to various methods. Illustratively, fractional flow reserve may be obtained by a table lookup method. If the pressure drop correction value is predicted by the pressure drop correction value prediction model, the pressure drop correction value is determined by considering the spatial distribution condition of each blood vessel branch in the target lesion artery and the size of each blood vessel branch, the size of the pressure drop correction value is also related to the spatial distribution condition of each blood vessel branch in the target lesion artery and the size of each blood vessel branch, the spatial distribution condition of each blood vessel branch in the target lesion artery and the size of each blood vessel branch are determined, and the change condition of the blood pressure in the target lesion artery is also determined, so that the pressure drop correction value and the real pressure drop value of a target area, namely delta P and delta P in a formula (9) can be used in a preset blood flow reserve fraction table according to the pressure drop correction value and the real pressure drop value of the target area_ΦP_mlAnd inquiring a corresponding blood flow reserve fraction table.

In some embodiments, the fractional flow reserve may be calculated by calculating the blood pressure at the entrance of the artery of the target lesion and the blood pressure at the exit of the target region according to the predicted pressure drop value and the corrected pressure drop value. Referring to fig. 4, at this time, the determining the fractional flow reserve of the target region according to the predicted pressure drop value and the corrected pressure drop value includes:

301. and determining a second entrance blood pressure of the target area according to the first entrance blood pressure of the target lesion artery entrance in the artery characteristic parameters.

The target lesion artery entrance refers to an entrance where blood flows into the target lesion artery. The first inlet blood pressure is referred to herein as the blood pressure. Specifically, the first inlet blood pressure may be calculated according to the cardiac output when the artery characteristic parameter is obtained, for example, the first inlet blood pressure may be calculated by referring to the calculation manner of equation (2).

The second inlet blood pressure is the blood pressure at the inlet where blood flows into the target region, i.e. the blood pressure at the inlet of the stenotic region in the target lesion artery. In some embodiments, the bifurcation of each vessel branch in the target focal artery, as well as the blood pressure decay in the target focal artery, may not be considered. Thus, the second inlet blood pressure can be equated to the first inlet blood pressure in the calculation if the first inlet blood pressure is P₀The second inlet blood pressure is P₁Then, it can be equivalent to: p₀＝P₁Therefore, the calculation process is further simplified, and the calculation speed is improved.

302. And calculating to obtain the outlet blood pressure of the target area according to the second inlet blood pressure, the predicted pressure drop value and the corrected pressure drop value.

The outlet blood pressure refers to the blood pressure at which blood flows out of a target region, i.e., the blood pressure at the outlet of a stenotic region in a target focal artery.

Specifically, the outlet blood pressure of the target region can be calculated by equation (10):

P₂＝P₁-ΔP_1D-Δ_ΦP_mlformula (10)

Wherein, P₂For outlet blood pressure, P₁Is the second inlet blood pressure,. DELTA.P_1DFor the predicted value of pressure drop, Δ_ΦP_mlIs a pressure drop correction value. It can be understood that Δ P_1D+Δ_ΦP_mlIs the actual pressure drop value of the target region, i.e., Δ P in equation (9), and therefore, the essence of equation (10) is to subtract the actual pressure drop value of the stenosis region from the blood pressure at the entrance of the stenosis region to obtain the blood pressure at the exit of the stenosis region.

303. And calculating the fractional flow reserve of the target area according to the first inlet blood pressure and the outlet blood pressure.

If a plurality of blood vessel branches are included in the target lesion artery, a partial pressure phenomenon may occur in the target lesion artery if blood flows into the plurality of blood vessel branches when the blood flows in the target lesion artery, and thus the first inlet blood pressure may represent a blood pressure when the inlet of the target lesion artery is not partially pressurized, that is, a maximum blood pressure in the target lesion artery. Furthermore, the blood pressure in the stenosis may gradually decrease in the direction of blood flow, and thus the outlet blood pressure may be indicative of the blood pressure in the stenosis that is not attenuated, i.e. the maximum blood pressure in the stenosis. The fractional flow reserve is defined as the ratio of the maximum blood flow provided by the diseased blood vessel to the maximum blood flow provided by the blood vessel when the blood vessel is completely normal, and the blood pressure and the blood flow are in positive correlation, so that the more the blood flow, the larger the blood pressure, and the fractional flow reserve can be calculated according to the outlet blood pressure and the first inlet blood pressure. Specifically, fractional flow reserve can be calculated by equation (11):

FFR＝P₂/P₀equation (11)

Wherein FFR is fractional flow reserve, P₂For outlet blood pressure, P₀The first inlet blood pressure. Combining equation (10) and equation (11), equation (12) can be obtained:

FFR＝(P₁-ΔP_1D-Δ_ΦP_ml)/P₀equation (12)

Therefore, the fractional flow reserve can be calculated by the first inlet blood pressure, the second inlet blood pressure, the pressure drop predicted value and the pressure drop corrected value.

In some embodiments, factors such as bifurcation of each blood vessel branch in the target focal artery, blood pressure attenuation in the target focal artery, etc., which may cause the first inlet blood pressure and the second inlet blood pressure to be different, may be considered, further improving the accuracy of the fractional flow reserve. Referring to fig. 5, at this time, the determining a second entry blood pressure of the target region according to the first entry blood pressure of the target lesion artery entry in the artery feature parameter includes:

401. and determining a blood pressure attenuation value between the target region and the entrance of the target lesion artery according to the artery characteristic parameter and a preset attenuation value calculation relation.

The blood pressure attenuation values are generated for different reasons for different scenarios. The following three scenarios may be specifically included:

scene one: only the blood pressure attenuation effect caused by the partial pressure effect of the target blood vessel branch in the target area is considered, the blood pressure attenuation effect caused by the partial pressure effect of the blood vessel branches except the target blood vessel branch in the target focal artery is not considered, and the blood pressure attenuation effect between the entrance of the target focal artery and the entrance of the target area is not considered. Therefore, in the calculation, only the partial pressure effect of the target blood vessel branch is considered, and the influence of the cross-sectional area of other blood vessel branches on the partial pressure is not required to be considered, so that the calculation can be performed according to the number of the blood vessel branches and the first inlet blood pressure. For example, the blood pressure attenuation value can be obtained by calculating the relationship of the attenuation values in equation (13):

wherein, P_{Fork with fork arm}Is the blood pressure attenuation value, n is the number of vessel branches in the artery characteristic parameter, P₀Is the first entry blood pressure in the artery characteristic parameter.

Scene two: simultaneously considering the blood pressure attenuation effect caused by the partial pressure effect of the target blood vessel branch in the target area and the blood pressure attenuation effect caused by the partial pressure effect of the blood vessel branch except the target blood vessel branch in the target focal arteryThe attenuation effect is not considered, and the blood pressure attenuation effect between the entrance of the target focal artery and the entrance of the target area is not considered. Therefore, in the calculation, the partial pressure effect of all the blood vessel branches in the target focal artery needs to be considered at the same time, except for P of the formula (13)_{Fork with fork arm}In addition, the effect of the cross-sectional area of all the vessel branches on the partial pressure needs to be taken into account. For example, the blood pressure attenuation value can be calculated by the sifting value calculation relationship in equation (14) and equation (15):

wherein, P_{Fork with fork arm}To take into account the blood pressure attenuation value, P, calculated when the blood pressure attenuation effect caused by the blood pressure partial pressure effect of the target blood vessel branch in the target region is taken into account_i ^{Attenuation 1}Calculating the blood pressure attenuation value d when considering the blood pressure attenuation effect caused by the blood pressure partial pressure effect of the ith segment of blood vessel branch in the target focal artery_iIs the cross-sectional area of the ith blood vessel branch in the artery characteristic parameter, n is the number of the blood vessel branches in the artery characteristic parameter, P₀Is the first entry blood pressure in the artery characteristic parameter.

Scene three: meanwhile, the blood pressure attenuation effect caused by the partial pressure effect of the target blood vessel branch in the target area, the blood pressure attenuation effect caused by the partial pressure effect of the blood vessel branch except the target blood vessel branch in the target focal artery and the blood pressure attenuation effect between the target focal artery inlet and the target area inlet are considered. For example, the blood pressure attenuation value can be obtained by calculating a relationship from the sifting values in equation (16) to equation (18):

wherein, P_{Fork with fork arm}To take into account the blood pressure attenuation value, P, calculated when the blood pressure attenuation effect caused by the blood pressure partial pressure effect of the target blood vessel branch in the target region is taken into account_i ^{Attenuation 1}Calculating the blood pressure attenuation value d when considering the blood pressure attenuation effect caused by the blood pressure partial pressure effect of the ith segment of blood vessel branch in the target focal artery_iIs the cross-sectional area of the ith blood vessel branch in the artery characteristic parameter, n is the number of the blood vessel branches in the artery characteristic parameter, Q₀Is the blood flow at the entrance of the artery of the target lesion in the artery characteristic parameters, P₀Is the first inlet blood pressure, P, in the artery characteristic parameter^{Attenuation 2}Calculating the blood pressure attenuation value when considering the blood pressure attenuation effect between the entrance of the target focus artery and the entrance of the target area, wherein L is the distance between the entrance of the target focus artery and the entrance of the target area in the artery characteristic parameter, and k₂Is a preset coefficient. Alternatively, the blood pressure attenuation value in consideration of the blood pressure attenuation effect between the entrance of the target lesion artery and the entrance of the target region may also be calculated by the methods of equation (4) -equation (8), and the specific process is not described herein again.

402. And determining a second inlet blood pressure of the target region according to the blood pressure attenuation value and the first inlet blood pressure of the target lesion artery inlet in the artery characteristic parameters.

In some embodiments, the blood pressure decay value may be subtracted from the first inlet blood pressure to simply obtain the second inlet blood pressure. For example, the blood pressure attenuation value includes only P_{Fork with fork arm}The first inlet blood pressure is P₀When the second inlet blood pressure is P1, the second inlet blood pressure can be calculated by equation (19):

P₁＝P₀-P_{fork with fork arm}Equation (19)

In other embodiments, the second inlet blood pressure may be calculated by a preset functional relationship, and the blood flow in the artery of the target lesion may be integrated to obtain the second inlet blood pressure. Specifically, for the above-mentioned scene one, scene two and scene three, the second inlet blood pressure can be calculated by the following equation (20) -equation (22):

P₁＝f₁(P₀，P_{fork with fork arm}Q) type (20)

P₁＝f₂(P₀，P_{Fork with fork arm}，P_i ^{Attenuation 1}Q) formula (21)

P₁＝f₃(P₀，P_{Fork with fork arm}，P_i ^{Attenuation 1}，P^{Attenuation 2}Q) type (22)

Wherein, P₀Is the first inlet blood pressure, P₁The second inlet blood pressure, P_{Fork with fork arm}To take into account the blood pressure attenuation value, P, calculated when the blood pressure attenuation effect caused by the blood pressure partial pressure effect of the target blood vessel branch in the target region is taken into account_i ^{Attenuation 1}The blood pressure attenuation value P is calculated when the blood pressure attenuation effect caused by the blood pressure partial pressure effect of the ith segment of blood vessel branch in the target lesion artery is considered^{Attenuation 2}To take into account the effect of blood pressure attenuation between the entrance of the target focal artery and the entrance of the target region, the calculated blood pressure attenuation value, f₁(x)，f₂(x)，f₃(x) Respectively corresponding to a scene one, a scene two and a scene three, and Q is the blood flow in the artery of the target lesion, which may be represented by Q in the equation (1) in the embodiment of the present application_inThe blood flow at other positions in the target lesion artery, or the blood flow in the target lesion artery obtained by calculation methods such as average and variance may be used.

It can be seen that through steps 401-402, the second inlet blood pressure can be calculated more accurately, so as to improve the accuracy of calculating the fractional flow reserve.

In some embodiments, the calculated fractional flow reserve and the corresponding patient information can be displayed on the display terminal at the same time, and since the fractional flow reserve calculation method provided in the embodiment of the present application has a high calculation speed, the fractional flow reserve of the patient can be provided in real time, so that medical staff can judge the condition of the patient in time to improve the immediacy of medical care. Referring to fig. 6, at this time, after determining the fractional flow reserve of the target region according to the predicted pressure drop value and the corrected pressure drop value, the method further includes:

501. and inquiring a preset database to obtain target patient information corresponding to the target lesion artery.

The preset database is used for storing patient information, and can be a database of a hospital or a database provided by a third-party supplier. The database can store patient information such as the name, past medical history, genetic medical history and the like of the patient, and during query, the target lesion artery can be matched with the patient in the database, a target patient corresponding to the target lesion artery is determined, and the information of the target patient is obtained. For example, when the medical image is acquired, medical staff can scan registration two-dimensional code information of a target patient, so that the medical image can carry identity information of the target patient in the registration two-dimensional code information, the identity information can be matched with information such as name and identity card number in a database to obtain target patient information, the registration two-dimensional code information of the target patient can also be scanned when the target patient is detected through the preset physiological detection component, and subsequent matching processes are similar to the medical image and are not repeated specifically.

502. And displaying the fractional flow reserve and the target patient information in a preset display terminal.

The display terminal can be any medical instrument or a display terminal carried by a communication device. For example, the display terminal may be a screen of a mobile phone, or may be a screen of a personal computer. When the target patient information is matched, part of the target patient information or all the target patient information and the fractional flow reserve can be displayed in the display terminal at the same time, so that medical staff can judge the disease condition of the target patient.

The following provides a specific procedure for calculating fractional flow reserve for ease of understanding:

(a) and acquiring a 2D or 3D medical image through a preset image acquisition device. For example, the medical image may be a tissue image of a location such as the heart, head and neck.

(b) And segmenting the medical image to obtain a target area image. For example, when the medical image is a physiological tissue image of a cardiac site, a coronary vessel region in the medical image may be segmented using a segmentation model constructed using a deep learning neural network model (U-net network) to obtain a 2D or 3D target region image, and the target region image includes information of the coronary vessel region.

(c) And performing dimensionality reduction on the target area image to obtain the geometric parameters of the arterial vessel. For example, the geometric information of the artery vessel in the artery characteristic parameters can be acquired manually or automatically.

(d) The operating parameters of the artery in the artery characteristic parameters are obtained through biomedical tests, and for example, the blood flow in the artery can be measured by a Color Velocity Image Quality (CVIQ) method.

(e) According to the geometric parameters of the artery and the working parameters of the artery, the boundary parameters of the target lesion artery in the artery characteristic parameters are calculated, so that boundary conditions for constraint are provided when pressure drop correction value prediction is carried out subsequently, and the calculation accuracy is improved when the blood flow reserve fraction is calculated subsequently.

(f) And training the preset pressure drop correction value prediction model through the preset training parameters to obtain the trained pressure drop correction value prediction model.

(g) And inputting the artery characteristic parameters into a preset pressure drop prediction model, and obtaining a pressure drop prediction value of a target area in the target lesion artery through a formula (4) to a formula (8). Alternatively, the pressure drop prediction model may be a dimension reduction model or a machine learning model.

(h) And inputting the pressure drop predicted value and the artery characteristic parameter into the trained pressure drop correction value prediction model to obtain the pressure drop correction value of the pressure drop predicted value.

(i) According to actual conditions, an appropriate calculation mode is selected from the formula (19) to the formula (22), and the second inlet blood pressure is calculated according to parameters such as boundary parameters, the number of blood vessel branches and the like in the artery characteristic parameters.

(j) And calculating the fractional flow reserve of the target area according to the first inlet blood pressure, the second inlet blood pressure, the pressure drop predicted value and the pressure drop corrected value by an equation (12).

In order to better implement the fractional flow reserve calculation method in the embodiment of the present application, on the basis of the fractional flow reserve calculation method, an embodiment of the present application further provides a fractional flow reserve calculation apparatus, as shown in fig. 7, which is a schematic structural diagram of an embodiment of the fractional flow reserve calculation apparatus in the embodiment of the present application, and the fractional flow reserve calculation apparatus 600 includes:

an obtaining unit 601, configured to obtain an artery characteristic parameter of a target lesion artery;

a determining unit 602, configured to determine, through a preset pressure drop prediction model, a pressure drop prediction value of a target region in the target focal artery according to the artery characteristic parameter after dimension reduction;

a deviation prediction unit 603, configured to perform pressure drop loss prediction processing on the pressure drop prediction value to obtain a pressure drop correction value;

a fraction determining unit 604, configured to determine a fractional flow reserve of the target region according to the predicted pressure drop value and the pressure drop correction value.

In a possible implementation manner of the present application, the deviation prediction unit 603 is further configured to:

In a possible implementation manner of the present application, the determining unit 602 is further configured to:

and adjusting parameters in the preset pressure drop correction dimensionality-increasing model according to the predicted pressure drop dimensionality-increasing value and the second pressure drop dimensionality-increasing value to obtain the trained pressure drop correction dimensionality-increasing model.

In a possible implementation manner of the present application, the score determining unit 604 is further configured to:

In a possible implementation manner of the present application, the obtaining unit 601 is further configured to:

acquiring a medical image of a target lesion artery;

In one possible implementation manner of the present application, the fractional flow reserve calculation apparatus further includes a display unit 605, and the display unit 605 is configured to:

In a specific implementation, the above units may be implemented as independent entities, or may be combined arbitrarily to be implemented as the same or several entities, and the specific implementation of the above units may refer to the foregoing method embodiments, which are not described herein again.

Since the fractional flow reserve calculation apparatus can execute the steps in the fractional flow reserve calculation method in any embodiment, the beneficial effects that can be achieved by the fractional flow reserve calculation method in any embodiment of the present application can be achieved, which are described in detail in the foregoing description and will not be repeated herein.

In addition, in order to better implement the fractional flow reserve calculation method in the embodiment of the present application, based on the fractional flow reserve calculation method, an electronic device is further provided in the embodiment of the present application, referring to fig. 8, fig. 8 shows a schematic structural diagram of the electronic device in the embodiment of the present application, specifically, the electronic device provided in the embodiment of the present application includes a processor 701, and when the processor 701 is used for executing a computer program stored in a memory 702, each step of the fractional flow reserve calculation method in any embodiment is implemented; alternatively, the processor 701 is configured to implement the functions of the units in the corresponding embodiment of fig. 7 when executing the computer program stored in the memory 702.

Illustratively, a computer program may be partitioned into one or more modules/units, which are stored in memory 702 and executed by processor 701 to implement embodiments of the present application. One or more modules/units may be a series of computer program instruction segments capable of performing certain functions, the instruction segments being used to describe the execution of a computer program in a computer device.

The electronic device may include, but is not limited to, a processor 701, a memory 702. Those skilled in the art will appreciate that the illustrations are merely examples of electronic devices and are not meant to be limiting, and may include more or fewer components than those shown, or some components may be combined, or different components.

The Processor 701 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being the control center for the electronic device and the various interfaces and lines connecting the various parts of the overall electronic device.

The memory 702 may be used to store computer programs and/or modules, and the processor 701 may implement various functions of the computer apparatus by running or executing the computer programs and/or modules stored in the memory 702 and invoking data stored in the memory 702. The memory 702 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, video data, etc.) created according to the use of the electronic device, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.

It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the fractional flow reserve calculation apparatus, the electronic device and the corresponding units thereof described above may refer to the description of the fractional flow reserve calculation method in any embodiment, and are not described herein again.

It will be understood by those skilled in the art that all or part of the steps in the methods of the above embodiments may be performed by instructions or by related hardware controlled by the instructions, which may be stored in a readable storage medium and loaded and executed by a processor.

For this reason, the embodiments of the present application provide a readable storage medium, where a computer program is stored on the readable storage medium, and the computer program is executed by a processor to perform the steps in the fractional flow reserve calculation method in any embodiment of the present application, and specific operations may refer to descriptions of the fractional flow reserve calculation method in any embodiment, and are not repeated herein.

Wherein the readable storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.

Since the instructions stored in the readable storage medium can execute the steps in the fractional flow reserve calculation method in any embodiment of the present application, the beneficial effects that can be achieved by the fractional flow reserve calculation method in any embodiment of the present application can be achieved, which are described in detail in the foregoing description and will not be described again here.

The method, the apparatus, the storage medium, and the electronic device for calculating fractional flow reserve provided in the embodiments of the present application are described in detail above, and specific examples are applied herein to explain the principle and the implementation of the present application, and the description of the embodiments above is only used to help understand the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A fractional flow reserve calculation method, comprising:

acquiring artery characteristic parameters of a target lesion artery;

inputting the artery characteristic parameters subjected to dimensionality reduction into a trained pressure drop correction model to obtain a pressure drop correction value of the pressure drop prediction value, wherein the pressure drop correction model is a machine learning model;

2. The fractional flow reserve calculation method according to claim 1, wherein the preset pressure drop prediction model is one of a machine learning model, a deep learning model and a dimension reduction model.

3. The fractional flow reserve calculation method according to claim 1, wherein before inputting the artery feature parameters after dimensionality reduction into a trained pressure drop correction model to obtain a pressure drop correction value of the pressure drop prediction value, the method further comprises:

4. The fractional flow reserve calculation method according to claim 1, wherein the determining, by a preset pressure drop prediction model, a predicted pressure drop value of a target region in the target focal artery according to the artery characteristic parameters after dimension reduction comprises:

5. The fractional flow reserve calculation method according to claim 4,

before the pressure drop predicted value to be processed, the artery characteristic parameter after dimensionality reduction and the artery characteristic parameter are input into a trained pressure drop correction dimensionality increasing model to obtain a pressure drop predicted value of a target area, the method further comprises the following steps:

6. The fractional flow reserve calculation method of claim 4, wherein the pressure drop modified upscaling model is one of a machine learning model and a deep learning model.

7. The fractional flow reserve calculation method of claim 1, wherein the determining the fractional flow reserve of the target region based on the predicted pressure drop value and the corrected pressure drop value comprises:

8. The fractional flow reserve calculation method of claim 7, wherein the determining a second entry blood pressure of the target region according to a first entry blood pressure of a target focal artery entry in the artery characteristic parameters comprises:

9. The fractional flow reserve calculation method according to claim 1, wherein the artery characteristic parameters include an artery geometric parameter and an artery physiological parameter, and the obtaining the artery characteristic parameters of the target focal artery includes:

acquiring a medical image of a target focus artery;

and acquiring the artery physiological parameters of the target focal artery through a preset physiological detection component.

10. The fractional flow reserve calculation method of any one of claims 1-9, wherein after determining the fractional flow reserve of the target region based on the predicted pressure drop value and the corrected pressure drop value, the method further comprises:

11. A fractional flow reserve calculation apparatus, comprising:

the deviation prediction unit is used for inputting the artery characteristic parameters subjected to the dimensionality reduction into a trained pressure drop correction model to obtain a pressure drop correction value of the pressure drop prediction value, wherein the pressure drop correction model is a machine learning model;

12. An electronic device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the fractional flow reserve calculation method according to any one of claims 1 to 10 when executing the computer program.

13. A readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the fractional flow reserve calculation method according to any one of claims 1 to 10.