CN111297388B

CN111297388B - Fractional flow reserve measuring device

Info

Publication number: CN111297388B
Application number: CN202010260601.1A
Authority: CN
Inventors: 张贺晔; 郭赛迪; 张冬
Original assignee: Sun Yat Sen University
Current assignee: Sun Yat Sen University
Priority date: 2020-04-03
Filing date: 2020-04-03
Publication date: 2023-06-20
Anticipated expiration: 2040-04-03
Also published as: CN111297388A

Abstract

The invention discloses a fractional flow reserve measuring device, comprising: establishing a corresponding relation between a blood vessel characteristic vector of a coronary artery and a blood flow reserve fraction through an artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve; acquiring a current blood vessel feature vector of a patient; and determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation. Can accurately predict fractional flow reserve and has wide application range.

Description

Fractional flow reserve measuring device

Technical Field

The invention relates to the field of medical detection, in particular to a fractional flow reserve measuring device.

Background

Coronary angiography and intravascular ultrasound are considered to be "gold standards" for diagnosing coronary heart disease, but they can only perform imaging evaluation on the stenosis degree, and how much the stenosis affects the far-end blood flow is unknown; fractional Flow Reserve (FFR) has now become a well-established indicator of functional assessment of coronary stenosis, the most important function of which is the accurate assessment of the functional consequences of an unknown affected coronary stenosis.

Fractional Flow Reserve (FFR) refers to the ratio of the maximum blood flow available to a region of the myocardium supplied by a target measurement vessel in the presence of stenotic lesions in the coronary arteries to the maximum blood flow theoretically available to the same region in a normal situation. FFR is obtained mainly by calculating the ratio of the coronary stenosis distal pressure to the aortic root pressure. The stenotic distal pressure may be measured by the pressure guidewire at maximum perfusion blood flow (by intracoronary or intravenous injection of papaverine or adenosine or ATP).

FFR＝P _d /P _a (P _d To guide the coronary stenosis distal pressure measured by the catheter, P _a Arterial pressure measured for a pressure guidewire) FFR refers to the maximum hyperemic state where the concept of "resting FFR" does not exist.

The normal epicardial coronary artery has little resistance to blood flow, and the normal value of FFR is 1.0; the FFR value will be less than 1.0 indicating the presence of a stenosis in the current epicardial coronary artery.

In the case of FFR < 0.75, almost all of the cases of stenosis will cause myocardial ischemia, and in the case of FFR.gtoreq.0.75, the probability of myocardial ischemia due to the stenosis will be very small.

The coronary artery CTA can accurately evaluate the coronary artery stenosis degree, can distinguish the plaque property of the tube wall, is a noninvasive and simple-to-operate method for diagnosing coronary artery lesions, and can be used as a first-choice method for screening high-risk groups. Thus, if an intervention is performed with respect to a vessel of a coronary heart disease patient, a CTA evaluation of the patient's coronary artery should be performed at a previous stage. Chronic total occlusion lesions (CTOs) of the coronary arteries, if assessed using CTA, must have some valuable information.

The FFR (CTFFR) obtained in a noninvasive way through coronary artery CT angiography CCTA calculation does not need additional image examination or medicines, has good correlation with FFR measured during radiography, and can fundamentally avoid unnecessary coronary angiography and blood transportation reconstruction treatment. The defactor test results also clearly demonstrate that in coronary CT, analysis of CTFFR results provides physiological information of lesions that truly restrict blood flow and increase patient risk. CTFFR combines the advantages of coronary CTA and FFR, and can evaluate coronary stenosis from both structural and functional aspects, becoming a novel non-invasive detection system providing anatomical and functional information of coronary lesions.

However, the existing detection systems generally have the following disadvantages for blood flow storage fraction measurement systems: disadvantages of invasive FFR techniques: there is a risk of surgery, and the use of vasodilators is toxic and may cause allergy to the patient, which is expensive.

Drawbacks of non-invasive FFR techniques based on hemodynamic simulation: the calculated amount is large, and a high-performance computer is needed; time consuming and unable to meet real-time requirements _。

Disclosure of Invention

In view of the problems, the present application has been made to provide fractional flow reserve measurement methods and apparatus that overcome or at least partially address the problems, including:

A fractional flow reserve measurement method comprising the steps of:

establishing a corresponding relation between a blood vessel characteristic vector of a coronary artery and a blood flow reserve fraction through an artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve;

acquiring a current blood vessel feature vector of a patient, and specifically acquiring the current blood vessel feature vector according to a coronary artery computed tomography image of the patient;

and determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation.

Further, the vessel feature vector includes:

the local geometrical characteristics of the blood vessel are as follows: radius of each cross section of the vessel;

the upstream and downstream geometric features of the blood vessel are specifically: stenosis proximal radius, stenosis entrance length, minimum radius zone length, stenosis exit length, radius reduction ratio:

Wherein r is _r Representing a radius reduction ratio; r is (r) _s Represents the minimum radius of the stenosis; r is (r) _p Representing the normal radius of the segment approaching stenosis; r is (r) _d Indicating the normal radius of the distal end of the stenosis.

Further, the first generating network and the first judging network are trained through sample data to obtain network parameters of a second generating network, and the second generating network and a second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vascular feature vector and the fractional flow reserve, comprising:

independently performing iterative training on the first generation network and the first discrimination network by maximizing the difference capability of the discrimination network and minimizing the distribution loss function of the generation network until the discrimination output probability value of the blood flow reserve fraction generated by the first generation network in the first discrimination network is close to 0.5;

according to the artificial neuron parameters of the first three network layers in the first judging network and even the artificial neuron parameters of the first three network layers in the second generating network;

and independently performing iterative training on the second generation network and the second discrimination network through maximizing the difference capability of the discrimination network and minimizing the distribution loss function of the generation network until the discrimination output probability value of the blood flow reserve fraction generated by the second generation network in the second discrimination network is close to 0.5.

Further, the first generation network includes: first input layer, first deconvolution layer G ₁ Second deconvolution layer G ₁ Third deconvolution layer G ₁ Fourth deconvolution layer G ₁ And a first output layer, wherein the number of artificial neurons of the first input layer is the same as the number of blood vessel feature vectors in the sample data, a first deconvolution layer G ₁ 512 artificial neurons, a second deconvolution layer G ₁ 256, third deconvolution layer G ₁ 128 artificial neurons, a fourth deconvolution layer G ₁ Is 64 and the number of artificial neurons of the first output layer is 1.

Further, the first discrimination network includes: second input layer, first convolution layer D ₁ Second convolution layer D ₁ Third convolution layer D ₁ Fourth convolution layer D ₁ And a second output layer, wherein the number of artificial neurons of the second input layer is the same as the total number of fractional flow reserve generated by the first generation network through sample data, the first convolution layer D ₁ Is 32, the second convolution layer D ₁ Is 64, the third convolution layer D ₁ 256, the fourth convolution layer D ₁ 512 artificial neurons, and 1 artificial neurons of the second output layer.

Further, the second generation network includes: third input layer, first deconvolution layer G ₂ Second deconvolution layer G ₂ Third deconvolution layer G ₂ Fourth deconvolution layer G ₂ Fifth deconvolution layer G ₂ And a third output layer, wherein the number of artificial neurons of the third input layer is the same as the number of blood vessel feature vectors in the sample data, the first deconvolution layer G ₂ Is 32, the second deconvolution layer G ₂ Is 64, the third deconvolution layer G ₂ 256, the fourth deconvolution layer G ₂ Is 64, the fifth deconvolution layer G ₂ The number of artificial neurons of the third output layer is 32, and the number of artificial neurons of the third output layer is 1.

Further, the second discrimination network includes: fourth input layer, first convolution layer D ₂ Second convolution layer D ₂ Third convolution layer D ₂ Fourth convolution layer D ₂ And a fourth output layer, wherein the number of artificial neurons of the fourth input layer is the same as the total number of the fractional flow reserve generated by the second generation network through the sample data and the actual fractional flow reserve corresponding to the sample data, the first convolution layer D ₂ Is 32, the second convolution layer D ₂ Is 64, the third convolution layer D ₂ 256, the fourth convolution layer D ₂ 512 artificial neurons, and 1 artificial neurons of the fourth output layer.

Further, the iterative training step of the first discrimination network or the second discrimination network includes:

setting a fractional flow reserve calculation value calculated by a blood vessel feature vector as a false sample set, and setting all class labels of the false sample set as 0;

setting a blood flow reserve fraction invasive measurement value obtained through invasive detection as a true sample set, and setting all class labels of the true sample set as 1;

and (3) inputting a blood vessel characteristic vector and a blood flow reserve fraction invasive measurement value, and adjusting the weight by comparing the value output by the discrimination network with a direct difference value of 1 to enable the value output by the discrimination network to be close to 1.

And (3) inputting a blood vessel characteristic vector and a blood flow reserve fraction calculation value, and comparing a value output by the discrimination network with a direct difference value of 0 to adjust the weight so that the value output by the discrimination network is close to 1.

Further, the iterative training steps of the first generation network and the second generation network include:

fixing the calculation parameters of the discrimination network;

inputting a blood vessel feature vector and performing initial calculation to obtain an initial fractional flow reserve calculation value, and setting a label of the initial fractional flow reserve calculation value as 1;

and adjusting the weight by comparing the difference value between the calculated value of the fractional flow reserve output by the generating network and the invasive measurement value of the fractional flow reserve, so that the calculated value of the fractional flow reserve output by the generating network is close to the invasive measurement value of the fractional flow reserve.

fixing the calculation parameters of the discrimination network;

and inputting the blood vessel characteristic vector and the calculated value of the fractional flow reserve output by the generation network into a discrimination network, and adjusting the weight of the generation network according to the direct difference value between the value output by the discrimination network and 1, so that the calculated value of the fractional flow reserve output by the generation network is close to 1 in the discrimination result of the discrimination network.

A fractional flow reserve measurement device comprising:

the establishing module is used for establishing a corresponding relation between the blood vessel characteristic vector of the coronary artery and the blood flow reserve fraction through the artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve;

the acquisition module is used for acquiring the current blood vessel feature vector of the patient, and particularly acquiring the current blood vessel feature vector according to the coronary artery computed tomography image of the patient;

and the measurement module is used for determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation.

An apparatus comprising a processor, a memory and a computer program stored on the memory and capable of running on the processor, which when executed by the processor performs the steps of a fractional flow reserve measurement method as described above.

A computer readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of a fractional flow reserve measurement method as described above.

The application has the following advantages:

in the embodiment of the application, a corresponding relation between a blood vessel characteristic vector of a coronary artery and a blood flow reserve fraction is established through an artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve; acquiring a current blood vessel feature vector of a patient, and specifically acquiring the current blood vessel feature vector according to a coronary artery computed tomography image of the patient; and determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation. The method can accurately predict the fractional flow reserve, has wide application range and larger degree of freedom, generates the training method of the generation network (G network) and the discrimination network (D network) of the antagonism network through the vascular characteristics extracted from the coronary CT image, avoids the operation risk, does not need to use a vasodilator, is safer for patients, has low calculation complexity and high calculation speed compared with the noninvasive FFR technology based on the hemodynamic simulation, and can meet the real-time requirement.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the description of the present application will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort to a person skilled in the art.

FIG. 1 is a flow chart of steps of a fractional flow reserve measurement method according to an embodiment of the present application;

fig. 2 is a schematic diagram of a network structure of a first generation network according to an embodiment of the present application;

fig. 3 is a network structure schematic diagram of a first discrimination network according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a network structure of a second generation network according to an embodiment of the present application;

fig. 5 is a schematic network structure of a second discrimination network according to an embodiment of the present application;

FIG. 6 is a schematic flow chart of iterative training of a discrimination network according to an embodiment of the present application;

FIG. 7 is a flow chart of generating iterative training of a network according to an embodiment of the present application;

FIG. 8 is a schematic block diagram of a fractional flow reserve measurement device according to an embodiment of the present application;

Fig. 9 is a schematic structural diagram of a computer device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, features and advantages of the present application more comprehensible, the present application is described in further detail below with reference to the accompanying drawings and detailed description. It will be apparent that the embodiments described are some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

It should be noted that in any embodiment of the present invention, concatenation is an efficient structure that can be used to automatically correlate tasks and can improve the performance of a multi-task model. For the multitasking problem, the cascading operation designs the tasks into a cascading mode, the former task transmits effective information to the latter task, and the latter task models according to the transmitted effective information.

FFR (fractional flow reserve) of coronary artery is defined as the maximum blood flow QS after myocardial induced hyperemia in the stenotic coronary artery innervation area _max Compared with the maximum blood flow QN at the position when no stenosis is assumed _max I.e.

P in the formula _d Is the pressure at the distal end of coronary stenosis, P _a Is the pressure of the main pulse, R _s And R is _N For microcirculation resistance, P _V Is the central venous pressure. In general, P _V Nearly close to zero and constant. When a vasodilator such as intravenous or intra-arterial adenosine is used, the coronary artery is brought to a maximum hyperemic state, i.e. resistance to microcirculation is minimised, at which point R is considered to be _s ＝R _N ，P _V Relative to P _a And P _d Negligible in doubt, i.e. the above formula can be reduced to:

wherein P is _d And P _a All can be measured by pressure guide wires.

Referring to fig. 1, the present invention provides a fractional flow reserve measurement method, comprising the steps of:

s110, establishing a corresponding relation between a blood vessel characteristic vector of a coronary artery and a blood flow reserve fraction through an artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve;

S120, acquiring a current blood vessel feature vector of a patient, and specifically acquiring the current blood vessel feature vector according to a coronary artery computed tomography image of the patient;

s130, determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation.

Next, the fractional flow reserve measurement method in the present exemplary embodiment will be further described.

As described in the above step S110, a correspondence between the vascular feature vector of the coronary artery and the fractional flow reserve is established through the artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve.

For example: the state rule of the blood vessel feature vector corresponding to the blood flow reserve fraction is analyzed by utilizing an artificial neural network algorithm, and the mapping rule among the blood vessel feature vectors corresponding to the blood flow reserve fraction of the patient is found by the self-learning and self-adapting characteristics of the artificial neural network.

For example: the artificial neural network algorithm can be utilized, the blood vessel characteristic vectors corresponding to the blood flow reserve fractions of a plurality of volunteers (including but not limited to one or more of age, weight, sex, condition and the like) are collected together, the blood vessel characteristic vectors corresponding to the blood flow reserve fractions of a plurality of volunteers are selected as sample data, the neural network is learned and trained, the relation between the blood vessel characteristic vectors corresponding to the blood flow reserve fractions fitted by the neural network is made through adjusting the weight between the network structure and the network nodes, and finally the neural network can accurately fit the corresponding relation between the blood vessel characteristic vectors corresponding to the blood flow reserve fractions of different patients.

It should be noted that, the blood vessel feature vector is generally obtained by integrating multiple feature values in a target specified area, where the target specified area may be a coronary artery, but is not limited to the coronary artery, and the target specified area may be correspondingly changed according to the change of the test purpose; the fractional flow reserve is a fractional flow reserve measurement obtained by invasive measurements corresponding to the vascular feature vector.

In an embodiment, the vessel feature vector comprises:

wherein r is _r Representing a radius reduction ratio; r is (r) _s Represents the minimum radius of the stenosis; r is (r) _p Representing the normal radius of the segment approaching stenosis; r is (r) _d Indicating the normal radius of the distal end of the stenosis。

Specifically, first, the upstream and downstream stenosis regions (including all the stenosis with a radius reduction of more than 10%) are identified by an automatic detection algorithm. The stenosis is then ranked based on the extent to which the radius decreases. Finally, the four most important stenosis upstream and the four most important stenosis downstream along the main branch path downstream are selected, for each of these stenosis, the extracted features are: 1) A stenosis proximal radius, i.e., a stenosis minimum radius and a stenosis distal radius; 2) The stenosis entrance length, i.e. the length along the centre line between the start of the stenosis and the start of the segment with the smallest radius; 3) The minimum radius region length, i.e., the length along the centerline between the beginning and end of the segment having the minimum radius; 4) A narrow outlet length, i.e. a length along the centre line between the end of the segment with the smallest radius and the end of the narrow; 5) Radius reduction ratio:

All the eigenvalues are combined to form an eigenvector.

In an embodiment, the correspondence relationship includes: functional relationship.

Preferably, the vascular feature vector of the coronary artery is an input parameter of the functional relationship, and the fractional flow reserve is an output parameter of the functional relationship;

determining a current fractional flow reserve corresponding to the current vessel feature vector, further comprising:

when the corresponding relation comprises a functional relation, inputting the current blood vessel feature vector into the functional relation, and determining the output parameter of the functional relation as the current blood flow reserve fraction.

Therefore, through the corresponding relation of various forms, the flexibility and convenience of determining the current fractional flow reserve can be improved.

Referring to fig. 2-5, as an example, the network structure is a CGAN network structure, and includes a first generating network, a second generating network, a first discriminating network, and a second discriminating network, where the first generating network, the second generating network, the first discriminating network, and the second discriminating network are respectively formed by deep convolutional neural networks.

Referring to fig. 2, in an embodiment, the first generation network includes: first input layer, first deconvolution layer G ₁ Second deconvolution layer G ₁ Third deconvolution layer G ₁ Fourth deconvolution layer G ₁ And a first output layer, wherein the number of artificial neurons of the first input layer is the same as the number of blood vessel feature vectors in the sample data, a first deconvolution layer G ₁ 512 artificial neurons, a second deconvolution layer G ₁ 256, third deconvolution layer G ₁ 128 artificial neurons, a fourth deconvolution layer G ₁ The number of the artificial neurons of the first output layer is 64, the number of the artificial neurons of the first output layer is 1, and the output value is the blood flow reserve fraction value obtained by initial calculation and is used as the input of the first discrimination network.

Referring to fig. 3, in an embodiment, the first discrimination network includes: second input layer, first convolution layer D ₁ Second convolution layer D ₁ Third convolution layer D ₁ Fourth convolution layer D ₁ And a second output layer, wherein the number of artificial neurons of the second input layer is the same as the total number of fractional flow reserve generated by the first generation network through sample data, the first convolution layer D ₁ Is 32, the second convolution layer D ₁ Is 64, the third convolution layer D ₁ 256, the fourth convolution layer D ₁ The number of the artificial neurons of the second output layer is 512, and the number of the artificial neurons of the second output layer is 1, wherein the output layer contains a single node of a linear activation function, the activation function is a Sigmoid function, the obtained value is a judgment probability value of input, the Sigmoid function is used for solving binary classification, and a value from 0 to 1 is output.

Referring to fig. 4, in an embodiment, the second generation network includes: third input layer, first deconvolution layer G ₂ Second deconvolution layer G ₂ Third deconvolution layer G ₂ Fourth deconvolution layer G ₂ Fifth deconvolution layer G ₂ And a third output layer, wherein the number of artificial neurons of the third input layer is the same as the number of blood vessel feature vectors in the sample data, the first deconvolution layer G ₂ Is 32, the second deconvolution layer G ₂ Is 64, the third deconvolution layer G ₂ 256, the fourth deconvolution layer G ₂ Is 64, the fifth deconvolution layer G ₂ The number of artificial neurons of the third output layer is 32, and the number of artificial neurons of the third output layer is 1.

Referring to fig. 5, in an embodiment, the second discrimination network includes: fourth input layer, first convolution layer D ₂ Second convolution layer D ₂ Third convolution layer D ₂ Fourth convolution layer D ₂ And a fourth output layer, wherein the number of artificial neurons of the fourth input layer is the same as the total number of the fractional flow reserve generated by the second generation network through the sample data and the actual fractional flow reserve corresponding to the sample data, the first convolution layer D ₂ Is 32, the second convolution layer D ₂ Is 64, the third convolution layer D ₂ 256, the fourth convolution layer D ₂ The number of the artificial neurons of the fourth output layer is 512, and the number of the artificial neurons of the fourth output layer is 1, wherein the output layer contains a single node of a linear activation function, the activation function is a Sigmoid function, the obtained value is a judgment probability value of input, the Sigmoid function is used for solving binary classification, and a value from 0 to 1 is output.

In an embodiment, the first generating network and the first judging network are trained through sample data to obtain network parameters of the second generating network, and the second generating network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vascular feature vector and the fractional flow reserve, comprising:

it should be noted that, since the first discrimination network has a high discrimination probability value for a real sample during initial operation, the discrimination probability value for a sample output generated by the first generation network is low. The output discrimination probability is fed back to the first generation network through mutual loop iterative learning, and the first generation network continuously optimizes network parameters, so that the first discrimination network cannot judge true or false; the first discrimination network is also continuously optimizing network parameters, so that the discrimination capability is improved, and the discrimination probability values of true and false samples are different. Finally, training is iterated for a plurality of times until the discrimination output probability value from the sample generated by the first generation network to the first discrimination network is close to 0.5, namely the true and false samples are difficult to distinguish, and the training is completed.

The first three-layer neuron parameters of the second generation network are transmitted and obtained by the first discrimination network after training, and the output value is the initial calculated fractional value of the blood flow reserve and is used as the input of the second discrimination network.

The first three-layer neuron parameters of the second generation network are transmitted by the first discrimination network, so that the second discrimination network outputs a discrimination probability value to the samples generated by the second generation network in the initial operation. The output discrimination probability is fed back to the second generation network through mutual loop iterative learning, and the second generation network continuously optimizes network parameters, so that the second discrimination network cannot judge true or false; the second discrimination network is also continuously optimizing network parameters, improving the discrimination capability and making the discrimination probability values of the true and false samples have a gap. Finally, training is iterated for a plurality of times until the discrimination output probability value from the sample generated by the second generation network to the second discrimination network is close to 0.5, namely the true and false samples are difficult to distinguish, and the training is completed.

In one embodiment, the specific procedure of "establishing correspondence between the vascular feature vector of the coronary artery and the fractional flow reserve" in step S110 may be further described in conjunction with the following description.

As described in the following steps: acquiring sample data for establishing a correspondence between the vessel feature vector and the fractional flow reserve;

in a further embodiment, a specific procedure of "acquiring sample data for establishing a correspondence between a vascular feature vector of the coronary artery and a fractional flow reserve" may be further described in connection with the following description.

As described in the following steps: collecting the vascular feature vector and the fractional flow reserve of patients with different cardiac conditions;

for example: data collection: collecting blood vessel feature vectors and corresponding fractional flow reserve of patients with different health conditions; collecting blood vessel characteristic vectors and corresponding fractional flow reserve of patients of different ages; and collecting blood vessel characteristic vectors and corresponding fractional flow reserve of patients with different sexes.

Therefore, the operation data are collected through various ways, the quantity of the operation data is increased, the learning capacity of the artificial neural network is improved, and the accuracy and the reliability of the determined corresponding relation are further improved.

As described in the following steps: analyzing the blood vessel feature vector, and combining pre-stored expert experience information, and selecting data related to the blood flow reserve score as the blood vessel feature vector (for example, selecting the blood vessel feature vector with influence on the blood flow reserve score as an input parameter and a specified parameter as an output parameter);

for example: the fractional flow reserve in the relevant data of the volunteers with the blood vessel feature vector in the relevant data of the volunteers with the diagnosis as the input parameter and the fractional flow reserve in the relevant data as the output parameter.

As described in the following steps: and taking the data pair consisting of the blood flow reserve fraction and the selected blood vessel characteristic vector as sample data.

For example: the obtained input and output parameter pairs are used as training sample data, and are used as test sample data.

Therefore, the collected blood vessel feature vectors are analyzed and processed, so that sample data are obtained, the operation process is simple, and the reliability of an operation result is high.

As described in the following steps: analyzing the characteristics and the rules of the blood vessel feature vector, and determining the network structure and the network parameters of the artificial neural network according to the characteristics and the rules;

For example: the characteristics of multi-view coronary angiography data and morphological parameter characteristics of coronary stenosis are analyzed, and the basic structure of the network, the number of input nodes and output nodes of the network, the number of hidden nodes of the network, the initial weight of the network and the like can be preliminarily determined.

Optionally, the specific process of training the network structure and the network parameters in the step of training and testing the network structure and the network parameters using the sample data, determining the correspondence of the blood vessel feature vector and the fractional flow reserve may be further described in connection with the following description.

Selecting a part of data in the sample data as a training sample, inputting the blood vessel feature vector in the training sample into the network structure, and training through a loss function, an activation function and the network parameters of the network structure to obtain an actual training result;

specifically, minimizing a loss function through a gradient descent algorithm, updating network parameters, and training a current multi-view neural network to obtain an actual training result;

determining whether an actual training error between the actual training result and a corresponding fractional flow reserve in the training sample meets a preset training error; when the actual training error meets the preset training error, determining that the training of the network structure and the network parameters is completed;

Specifically, when the actual training error meets the preset training error, and the current training model converges, the training of the network structure and the network parameters is determined to be completed.

More optionally, training the network structure and the network parameters further includes:

when the actual training error does not meet the set training error, updating the network parameters through an error loss function of the network structure; retraining by using the loss function of the network structure, an activation function and the updated network parameters until the retrained actual training error meets the set training error;

for example: if the test error meets the requirement, the network training test is completed.

Therefore, the test samples are used for testing the network structure and the network parameters obtained through training, so that the reliability of the network structure and the network parameters is further verified.

Optionally, the specific procedure of testing the network structure and the network parameters in the step of training and testing the network structure and the network parameters using the sample data, determining the correspondence of the vascular feature vector and the fractional flow reserve may be further described in connection with the following description.

Selecting another part of the sample data as a test sample, inputting the blood vessel feature vector in the test sample into the trained network structure, and testing by using the loss function, the activation function and the trained network parameter to obtain an actual test result; determining whether an actual test error between the actual test result and a corresponding fractional flow reserve in the test sample meets a set test error; and when the actual test error meets the set test error, determining that the test on the network structure and the network parameters is completed.

In the CGAN network framework, the first generation network, the second generation network, the first discrimination network, and the second discrimination network perform joint training using a small-batch gradient descent method, and both the generation network frameworks have own resistance loss. Therefore, the calculation formula of using the feature matching loss function for the first discrimination network is as follows:

wherein D is _1-feature Representing the output characteristics of the first discriminating network, L _1-feature Is the feature matching loss function of the first discrimination network, x and

Real data and dummy data generated by the first generation network, P _ri And P _fi The characteristic distribution of real data and dummy data, respectively.

The countering loss function calculation formula of the first generation network is as follows:

in the method, in the process of the invention,

is the countering loss of the first generation network, +.>

Is the countermeasures loss of the first discriminant network, λ is the weight of the penalty term. Let λ=10.

Feature matching function L of second discrimination network _2－feature The calculation formula of (2) is as follows:

in the method, in the process of the invention,

and->

The generated real data and the generated dummy data of the second generation network, P _ls And P _us Characteristic distribution respectively representing real data and false data, D _2－feature And represents the feature output of the second discrimination network.

Similar to the countering loss of the first generation network and the first discrimination network, L _2－feature Introducing the countermeasures loss of the second generation network and the second discrimination network, wherein the calculation formula is as follows:

the countering loss described above can be regarded as an unsupervised loss because it uses only the generated data of the second generation network, not the real data.

Finally, using the real data s _l Supervision loss function L is defined based on Dice loss _G2sup The calculation formula is as follows:

updating supervision loss simultaneously when algorithm runs

And counter-loss- >

Since the training flows of the first and second discrimination networks are similar, only the training flow of the first discrimination network will be briefly described here.

Referring to fig. 6, in an embodiment, the iterative training step of the first discrimination network or the second discrimination network includes:

it should be noted that, parameters of the generation network need to be fixed before calculating the fractional flow reserve calculation value, so as to avoid that when the generation network generates the fractional flow reserve calculation value, an unnecessary uncertain condition is introduced due to the change of the network parameters, so that errors occur, and further, the result of iterative training is inclined;

the test condition (blood vessel characteristic value) of the above-mentioned blood flow reserve score invasive measurement value corresponds to the calculation condition (blood vessel characteristic value) of each blood flow reserve score calculation value in the above-mentioned step;

It should be noted that, since there are only 0 or 1 labels of the input samples, a value between 0 and 1 is obtained according to the discrimination results of the true sample set and the false sample set, and since the data genuineness of the true sample set and the false sample set in the training parameters is known, the discrimination network can be completed by manually performing the deviation of the weights according to the discrimination results to make the discrimination network record and correct the results, and in this step, the weights are adjusted by calculating the direct difference between the discrimination results output by the discrimination network and 1, and when the direct difference between the discrimination results and 1 is close to 0, the discrimination network training is completed.

It should be noted that, when the direct difference between the above-mentioned determination result and 0 is close to 1, that is, the network training is determined to be completed, the step may be used alternately with the corresponding step in the previous embodiment in the iterative training process or training is performed by using any step as the checking step.

It should be noted that the training flows of the first generating network and the second generating network are similar, so only the training flow of the first generating network is briefly described here.

Referring to fig. 7, in an embodiment, the iterative training steps of the first generation network and the second generation network include:

fixing the calculation parameters of the discrimination network;

The calculation parameters of the discrimination network are fixed, the discrimination model is defined as a quantitative condition, and the calculation values of the fractional flow reserve generated by the generation network are discriminated, so that the generation weight of the generation network is adjusted, and the calculation parameters generally include, but are not limited to, the discrimination weight of the discrimination network;

the label of the fractional flow reserve calculation value is set to 1, that is, the fractional flow reserve invasive measurement value under the condition that the fractional flow reserve calculation value is regarded as the current feature vector when the judgment is performed is indicated, the true and false judgment is performed through the judgment network, and the judgment result of the judgment network is recorded, wherein the judgment result is a numerical value between 0 and 1, that is, the judgment weight;

The method includes the steps of obtaining a blood flow reserve score invasive measurement value of the blood vessel feature vector which is the same as the blood flow reserve score calculation value, calculating a difference value between the blood flow reserve score calculation value and the blood flow reserve score invasive measurement value, adjusting generation weights in a generation network according to the difference value, and perfecting the generation network.

fixing the calculation parameters of the discrimination network;

The deviation ratio of the fractional flow reserve calculated by the generation network is obtained by calculating the direct difference between the discrimination result of the discrimination network and 1, and the examination weight of the generation network is adjusted.

The current vessel feature vector of the patient is acquired as described in the above step S120, in particular, the current vessel feature vector is acquired from the coronary artery computed tomography image of the patient, and the pressure P at the distal end of the coronary stenosis is measured by the pressure guide wire in an invasive manner _d Pressure P of the main pulse _a And pass through the formula

And (5) calculating to obtain the product.

As described in step S130 above, the current fractional flow reserve corresponding to the current blood vessel feature vector is determined from the correspondence.

For example: vascular feature vectors of coronary arteries of a patient are identified in real-time.

Therefore, the current blood flow reserve fraction of the coronary artery is effectively identified according to the current blood vessel feature vector based on the corresponding relation, so that an accurate judgment basis is provided for the diagnosis of doctors, and the judgment result is accurate.

In an alternative example, determining the current fractional flow reserve corresponding to the vessel feature vector in step S130 may include: and determining the blood flow reserve score corresponding to the blood vessel feature vector which is the same as the current blood vessel feature vector in the corresponding relation as the current blood flow reserve score.

In an alternative example, determining the current fractional flow reserve corresponding to the blood vessel feature vector in step S130 may further include: when the correspondence may include a functional relationship, the current vessel feature vector is input into the functional relationship, and an output parameter of the functional relationship is determined to be a current fractional flow reserve.

Therefore, the current blood flow reserve fraction is determined according to the current blood vessel feature vector based on the corresponding relation or the functional relation, the determination mode is simple and convenient, and the reliability of the determination result is high.

For example, the trained artificial neural network is used to detect fractional flow reserve for each sample in the test set.

In an alternative embodiment, the method may further include: a process of verifying whether the current fractional flow reserve matches the actual fractional flow reserve.

Optionally, when a verification result that the current fractional flow reserve does not match the actual fractional flow reserve is received and/or it is determined that the corresponding relationship does not have the same vascular feature vector as the current vascular feature vector, at least one maintenance operation of updating, correcting and relearning the corresponding relationship may be performed.

For example: the device itself cannot learn the actual fractional flow reserve, and needs to have feedback operation of a doctor, namely if the device intelligently judges the fractional flow reserve, the doctor can learn through operation feedback that the fractional flow reserve does not accord with the actual state.

Verifying that the current fractional flow reserve matches the actual fractional flow reserve (e.g., the actual fractional flow reserve may be displayed by an AR display module to verify that the determined current fractional flow reserve matches the actual fractional flow reserve).

And when the current blood flow reserve score is not consistent with the actual blood flow reserve score and/or the corresponding relation does not have the blood vessel characteristic vector which is the same as the current blood vessel characteristic vector, at least one maintenance operation of updating, correcting and relearning is carried out on the corresponding relation.

For example: and determining the current blood flow reserve fraction according to the current blood vessel characteristic vector according to the maintained corresponding relation. For example: and determining the blood flow reserve score corresponding to the blood vessel feature vector which is the same as the current blood vessel feature vector in the maintained corresponding relation as the current blood flow reserve score.

Therefore, the maintenance of the corresponding relation between the determined blood vessel characteristic vector and the blood flow reserve fraction is beneficial to improving the accuracy and the reliability of blood flow reserve fraction determination.

Referring to fig. 8, the present invention proposes a fractional flow reserve measurement device comprising:

the establishing module 810 is configured to establish, through an artificial neural network, a correspondence between a vascular feature vector of a coronary artery and a fractional flow reserve; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve;

an obtaining module 820, configured to obtain a current blood vessel feature vector of a patient, specifically, obtain the current blood vessel feature vector according to a coronary artery computed tomography image of the patient;

a measurement module 830, configured to determine, according to the correspondence, a current fractional flow reserve corresponding to the current vascular feature vector.

In an embodiment, the vessel feature vector comprises:

In one embodiment, the establishing module 810 includes:

the first iterative training sub-module is used for independently performing iterative training on the first generation network and the first discrimination network by maximizing the difference capacity of the discrimination network and minimizing the distribution loss function of the generation network until the discrimination output probability value of the fractional flow reserve generated by the first generation network in the first discrimination network is close to 0.5;

the neuron parameter generation sub-module is used for generating the artificial neuron parameters of the first three network layers in the first discrimination network and even the artificial neuron parameters of the first three network layers in the second generation network according to the artificial neuron parameters of the first three network layers in the first discrimination network;

a second iterative training sub-module, configured to perform iterative training on the second generation network and the second discrimination network independently by maximizing a differential capability of the discrimination network and minimizing a distribution loss function of the generation network, until a discrimination output probability value of a fractional flow reserve generated by the second generation network in the second discrimination network is close to 0.5

In an embodiment, the first generation network comprises: first input layer, first deconvolution layer G ₁ Second deconvolution layer G ₁ Third deconvolution layer G ₁ Fourth deconvolution layer G ₁ And a first output layer, wherein the number of artificial neurons of the first input layer is the same as the number of blood vessel feature vectors in the sample data, a first deconvolution layer G ₁ 512 artificial neurons, a second deconvolution layer G ₁ 256, third deconvolution layer G ₁ 128 artificial neurons, a fourth deconvolution layer G ₁ Is 64 and the number of artificial neurons of the first output layer is 1.

In an embodiment, the first discrimination network includes: second input layer, first convolution layer D ₁ Second convolution layer D ₁ Third convolution layer D ₁ Fourth convolution layer D ₁ And a second output layer, wherein the number of artificial neurons of the second input layer is the same as the total number of fractional flow reserve generated by the first generation network through sample data, the first convolution layer D ₁ Is 32, the second convolution layer D ₁ Is 64, the third convolution layer D ₁ 256, the fourth convolution layer D ₁ 512 artificial neurons, and 1 artificial neurons of the second output layer.

In an embodiment, the second generation network comprises: third input layer, first deconvolution layer G ₂ Second deconvolution layer G ₂ Third deconvolution layer G ₂ Fourth deconvolution layer G ₂ Fifth deconvolution layer G ₂ And a third output layer, wherein the number of artificial neurons in the third input layer and blood in the sample dataThe number of the tube eigenvectors is the same, the first deconvolution layer G ₂ Is 32, the second deconvolution layer G ₂ Is 64, the third deconvolution layer G ₂ 256, the fourth deconvolution layer G ₂ Is 64, the fifth deconvolution layer G ₂ The number of artificial neurons of the third output layer is 32, and the number of artificial neurons of the third output layer is 1.

In an embodiment, the second discrimination network includes: fourth input layer, first convolution layer D ₂ Second convolution layer D ₂ Third convolution layer D ₂ Fourth convolution layer D ₂ And a fourth output layer, wherein the number of artificial neurons of the fourth input layer is the same as the total number of the fractional flow reserve generated by the second generation network through the sample data and the actual fractional flow reserve corresponding to the sample data, the first convolution layer D ₂ Is 32, the second convolution layer D ₂ Is 64, the third convolution layer D ₂ 256, the fourth convolution layer D ₂ 512 artificial neurons, and 1 artificial neurons of the fourth output layer.

In an embodiment, the first iterative training sub-module or the second iterative training sub-module comprises:

a first false sample set generation sub-module for setting a fractional flow reserve calculation value calculated by the blood vessel feature vector as a false sample set and setting all class labels of the false sample set as 0;

a first true sample set generating sub-module, configured to set a fractional flow reserve invasive measurement value obtained by invasive detection as a true sample set, and set all class labels of the true sample set as 1;

the first weight adjustment sub-module is used for inputting blood vessel characteristic vectors and blood flow reserve fraction invasive measurement values, and adjusting weights by comparing the direct difference value of the values output by the discrimination network and 1, so that the values output by the discrimination network are close to 1.

a second false sample set generation sub-module for setting a fractional flow reserve calculation value calculated by the blood vessel feature vector as a false sample set and setting all class labels of the false sample set as 0;

A second true sample set generating sub-module, configured to set a fractional flow reserve invasive measurement value obtained by invasive detection as a true sample set, and set all class labels of the true sample set as 1;

and the second weight adjustment sub-module is used for inputting the blood vessel characteristic vector and the blood flow reserve fraction calculation value, and comparing the value output by the discrimination network with the direct difference value adjustment weight of 0 so as to enable the value output by the discrimination network to be close to 1.

the first fixed calculation parameter sub-module is used for fixing the calculation parameters of the discrimination network;

the first input submodule is used for inputting the blood vessel feature vector and carrying out initial calculation to obtain an initial fractional flow reserve calculation value, and the label of the initial fractional flow reserve calculation value is set to be 1;

and the third weight adjustment sub-module is used for adjusting the weight by comparing the difference value between the calculated value of the fractional flow reserve output by the generating network and the fractional flow reserve invasive measurement value, so that the calculated value of the fractional flow reserve output by the generating network is close to the fractional flow reserve invasive measurement value.

The second fixed calculation parameter sub-module is used for fixing the calculation parameters of the discrimination network;

the second input submodule is used for inputting the blood vessel feature vector and carrying out initial calculation to obtain an initial fractional flow reserve calculation value, and the label of the initial fractional flow reserve calculation value is set to be 1;

and the fourth weight adjustment sub-module is used for inputting the blood vessel characteristic vector and the calculated value of the fractional flow reserve output by the generation network into the discrimination network, and adjusting the weight of the generation network according to the direct difference value between the value output by the discrimination network and 1 so that the calculated value of the fractional flow reserve output by the generation network is close to 1 in the discrimination result of the discrimination network.

Referring to fig. 9, a computer device for a fractional flow reserve measurement method of the present invention may specifically include the following:

the computer device 12 described above is embodied in the form of a general purpose computing device, and the components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, a bus 18 that connects the various system components, including the system memory 28 and the processing units 16.

Bus 18 represents one or more of several types of bus 18 structures, including a memory bus 18 or memory controller, a peripheral bus 18, an accelerated graphics port, a processor, or a local bus 18 using any of a variety of bus 18 architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus 18, micro channel architecture (MAC) bus 18, enhanced ISA bus 18, video Electronics Standards Association (VESA) local bus 18, and Peripheral Component Interconnect (PCI) bus 18.

Computer device 12 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 30 and/or cache memory 32. The computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer storage media. By way of example only, storage system 34 may be used to read from or write to non-removable, nonvolatile magnetic media (commonly referred to as a "hard disk drive"). Although not shown in fig. 9, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk such as a CD-ROM, DVD-ROM, or other optical media may be provided. In such cases, each drive may be coupled to bus 18 through one or more data medium interfaces. The memory may include at least one program product having a set (e.g., at least one) of program modules 42, the program modules 42 being configured to carry out the functions of embodiments of the invention.

A program/utility 40 having a set (at least one) of program modules 42 may be stored in, for example, a memory, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules 42, and program data, each or some combination of which may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods of the embodiments described herein.

The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, camera, etc.), one or more devices that enable a user to interact with the computer device 12, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 12 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 22. Moreover, computer device 12 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet, through network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be appreciated that although not shown in fig. 9, other hardware and/or software modules may be used in connection with computer device 12, including, but not limited to: microcode, device drivers, redundant processing units 16, external disk drive arrays, RAID systems, tape drives, data backup storage systems 34, and the like.

The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, for example, implementing fractional flow reserve measurement methods provided by embodiments of the present invention.

That is, the processing unit 16 realizes when executing the program: establishing a corresponding relation between a blood vessel characteristic vector of a coronary artery and a blood flow reserve fraction through an artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve; acquiring a current blood vessel feature vector of a patient, and specifically acquiring the current blood vessel feature vector according to a coronary artery computed tomography image of the patient; and determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation.

In an embodiment of the present invention, the present invention further provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a fractional flow reserve measurement method as provided in all embodiments of the present application:

that is, the program is implemented when executed by a processor: establishing a corresponding relation between a blood vessel characteristic vector of a coronary artery and a blood flow reserve fraction through an artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; specifically, the first generation network and the first discrimination network are trained through sample data to obtain network parameters of the second generation network, and the second generation network and the second other network are trained through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve; acquiring a current blood vessel feature vector of a patient, and specifically acquiring the current blood vessel feature vector according to a coronary artery computed tomography image of the patient; and determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation.

Any combination of one or more computer readable media may be employed. The computer readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having 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 (EPOM 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. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Computer program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

While preferred embodiments of the present embodiments have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the present application.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.

The above description has been made in detail for a fractional flow reserve measurement device provided in the present application, and specific examples are applied herein to illustrate the principles and embodiments of the present application, the above examples are only for helping to understand the method and core ideas of the present application; meanwhile, as those skilled in the art will have modifications in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims

1. A fractional flow reserve measurement device, comprising:

the establishing module is used for establishing a corresponding relation between the blood vessel characteristic vector of the coronary artery and the blood flow reserve fraction through the artificial neural network; the artificial neural network comprises a first generation network, a second generation network, a first discrimination network and a second discrimination network; training a first generating network and a first judging network through sample data to obtain network parameters of a second generating network, and training the second generating network and the second judging network through the sample data to obtain the corresponding relation; wherein the sample data is the vessel feature vector and the fractional flow reserve;

The acquisition module is used for acquiring the current blood vessel feature vector of the patient and acquiring the current blood vessel feature vector according to the coronary artery computed tomography image of the patient;

the measuring module is used for determining the current blood flow reserve fraction corresponding to the current blood vessel feature vector according to the corresponding relation;

wherein, the establishment module includes:

the neuron parameter generation sub-module is used for generating the artificial neuron parameters of the first three-layer network layer of the second generation network according to the artificial neuron parameters of the first three-layer network layer in the first discrimination network;

and the second iterative training sub-module is used for independently carrying out iterative training on the second generation network and the second discrimination network by maximizing the difference capability of the discrimination network and minimizing the distribution loss function of the generation network until the discrimination output probability value of the fractional flow reserve generated by the second generation network in the second discrimination network is close to 0.5.

2. The apparatus of claim 1, wherein the vessel feature vector comprises:

local geometry of the vessel: radius of each cross section of the vessel;

upstream and downstream geometry of the vessel: stenosis proximal radius, stenosis entrance length, minimum radius zone length, stenosis exit length, radius reduction ratio:

in the method, in the process of the invention,

representing a radius reduction ratio; />

Represents the minimum radius of the stenosis; />

Representing the normal radius of the segment approaching stenosis; />

Indicating the normal radius of the distal end of the stenosis.

3. The apparatus of claim 1, wherein the device comprises a plurality of sensors,

the first generation network includes: the system comprises a first input layer, a first deconvolution layer, a second deconvolution layer, a third deconvolution layer, a fourth deconvolution layer and a first output layer, wherein the number of the artificial neurons of the first input layer is the same as the number of the blood vessel feature vectors in sample data, the number of the artificial neurons of the first deconvolution layer is 512, the number of the artificial neurons of the second deconvolution layer is 256, the number of the artificial neurons of the third deconvolution layer is 128, the number of the artificial neurons of the fourth deconvolution layer is 64, and the number of the artificial neurons of the first output layer is 1.

4. The apparatus of claim 1, wherein the device comprises a plurality of sensors,

the first discrimination network includes: the system comprises a second input layer, a first convolution layer, a second convolution layer, a third convolution layer, a fourth convolution layer and a second output layer, wherein the number of the artificial neurons of the second input layer is the same as the total number of fractional flow reserve generated by the first generation network through sample data, the number of the artificial neurons of the first convolution layer is 32, the number of the artificial neurons of the second convolution layer is 64, the number of the artificial neurons of the third convolution layer is 256, the number of the artificial neurons of the fourth convolution layer is 512, and the number of the artificial neurons of the second output layer is 1.

5. The apparatus of claim 1, wherein the device comprises a plurality of sensors,

the second generation network includes: the system comprises a third input layer, a first deconvolution layer, a second deconvolution layer, a third deconvolution layer, a fourth deconvolution layer, a fifth deconvolution layer and a third output layer, wherein the number of the artificial neurons of the third input layer is the same as the number of the vascular eigenvectors in sample data, the number of the artificial neurons of the first deconvolution layer is 32, the number of the artificial neurons of the second deconvolution layer is 64, the number of the artificial neurons of the third deconvolution layer is 256, the number of the artificial neurons of the fourth deconvolution layer is 64, the number of the artificial neurons of the fifth deconvolution layer is 32, and the number of the artificial neurons of the third output layer is 1.

6. The apparatus of claim 1, wherein the device comprises a plurality of sensors,

the second discrimination network includes: the system comprises a fourth input layer, a first convolution layer, a second convolution layer, a third convolution layer, a fourth convolution layer and a fourth output layer, wherein the sum of the fractional flow reserve generated by the second generation network through sample data and the actual fractional flow reserve corresponding to the sample data is the same as the number of the artificial neurons of the fourth input layer, the number of the artificial neurons of the first convolution layer is 32, the number of the artificial neurons of the second convolution layer is 64, the number of the artificial neurons of the third convolution layer is 256, the number of the artificial neurons of the fourth convolution layer is 512, and the number of the artificial neurons of the fourth output layer is 1.

7. The apparatus of claim 1, wherein the iterative training sub-module of the first discrimination network or the second discrimination network comprises:

8. The apparatus of claim 1, wherein the iterative training sub-module of the first discrimination network or the second discrimination network comprises:

9. The apparatus of claim 1, wherein the iterative training sub-module of the first generation network and the second generation network comprises:

10. The apparatus of claim 1, wherein the iterative training sub-module of the first generation network and the second generation network comprises: