Method, device, system and storage medium for determining fractional flow reserve

Abstract

The application discloses a blood flow reserve fraction determination method, device, system and storage medium. The method comprises the following steps: acquiring an initial blood vessel image, inputting the initial blood vessel image into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model; determining a stenosis region of the first blood vessel segmentation model, performing a de-stenosis processing on the stenosis region of the first blood vessel segmentation model to obtain a second blood vessel segmentation model; acquiring flow data of the stenosis region of the first blood vessel segmentation model and flow data of a de-stenosis region of the second blood vessel segmentation model; and determining a blood flow reserve fraction based on the flow data of the stenosis region of the first blood vessel segmentation model and the flow data of the de-stenosis region of the second blood vessel segmentation model. The technical solution can effectively reduce the error of the determined blood flow reserve fraction by determining the blood flow reserve fraction based on the flow data of the stenosis region of the first blood vessel segmentation model and the flow data of the de-stenosis region of the second blood vessel segmentation model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to a method, device and system for determining fractional flow reserve and a storage medium. BACKGROUND

[0002] Fractional flow reserve (FFR) is an important indicator for evaluating myocardial ischemia.

[0003] Currently, methods for obtaining fractional flow reserve include invasive and non-invasive methods. The invasive measurement method mainly measures the average pressure of the stenosis distal end and the coronary artery inlet position in the aorta under the condition of hyperemia by a pressure guide wire, and the ratio of the two average pressures is the value of FFR. The non-invasive measurement technique mainly has the ct-FFR technique, which also determines the value of FFR by pressure average.

[0004] In the process of implementing the present application, the inventors have found that the prior art method of determining FFR by pressure has the problem of large error in determining fractional flow reserve. SUMMARY

[0005] The present application provides a method, device and system for determining fractional flow reserve and a storage medium to solve the problem of large error in fractional flow reserve.

[0006] According to one aspect of the present application, a method for determining fractional flow reserve is provided, which is executed by a fractional flow reserve determination system and includes:

[0007] An initial blood vessel image is obtained, and the initial blood vessel image is input into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model;

[0008] A stenosis region of the first blood vessel segmentation model is determined, and the stenosis region of the first blood vessel segmentation model is processed to obtain a second blood vessel segmentation model;

[0009] Flow data of the stenosis region of the first blood vessel segmentation model and flow data of the de-stenosis region of the second blood vessel segmentation model are obtained;

[0010] Fractional flow reserve is determined based on the flow data of the stenosis region of the first blood vessel segmentation model and the flow data of the de-stenosis region of the second blood vessel segmentation model.

[0011] According to another aspect of the present application, a device for determining fractional flow reserve is provided, which includes:

[0012] The first blood vessel segmentation model determination module is configured to obtain an initial blood vessel image, input the initial blood vessel image into a pre-trained blood vessel segmentation model, and obtain a first blood vessel segmentation model.

[0013] The second blood vessel segmentation model determination module is configured to determine a stenosis region of the first blood vessel segmentation model, perform a de-stenosis process on the stenosis region of the first blood vessel segmentation model, and obtain a second blood vessel segmentation model.

[0014] The model flow data acquisition module is configured to acquire flow data of the stenosis region of the first blood vessel segmentation model and flow data of a de-stenosis region of the second blood vessel segmentation model.

[0015] The blood flow reserve fraction determination module is configured to determine a blood flow reserve fraction based on the flow data of the stenosis region of the first blood vessel segmentation model and the flow data of the de-stenosis region of the second blood vessel segmentation model.

[0016] According to another aspect of the present application, a system for determining a blood flow reserve fraction is provided, which comprises a processor and a memory, the memory storing a computer program, and the processor executing the following steps when running the computer program, comprising:

[0017] obtaining an initial blood vessel image, inputting the initial blood vessel image into a pre-trained blood vessel segmentation model, and obtaining a first blood vessel segmentation model;

[0018] determining a stenosis region of the first blood vessel segmentation model, performing a de-stenosis process on the stenosis region of the first blood vessel segmentation model, and obtaining a second blood vessel segmentation model;

[0019] acquiring flow data of the stenosis region of the first blood vessel segmentation model and flow data of a de-stenosis region of the second blood vessel segmentation model;

[0020] determining a blood flow reserve fraction based on the flow data of the stenosis region of the first blood vessel segmentation model and the flow data of the de-stenosis region of the second blood vessel segmentation model.

[0021] According to another aspect of the present application, a computer readable storage medium is provided, which stores computer instructions for causing a processor to execute a method for determining a blood flow reserve fraction according to any of the embodiments of the present application.

[0022] The technical solution of this invention involves acquiring an initial vascular image, inputting it into a pre-trained vascular segmentation model to obtain a first vascular segmentation model, identifying the stenotic region of the first vascular segmentation model, performing destenosis processing on the stenotic region of the first vascular segmentation model to obtain a second vascular segmentation model, acquiring flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model, and determining the fractional flow reserve (FLR) based on the flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model. Compared with the prior art, this invention determines the FLR without introducing pressure information with large errors, making the obtained FLR more accurate, thereby solving the problem of large errors in the FLR.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of a method for determining fractional blood flow reserve according to Embodiment 1 of the present invention;

[0026] Figure 2(a) is a schematic diagram of the first blood vessel segmentation model provided according to Embodiment 1 of the present invention;

[0027] Figure 2(b) is a schematic diagram of the second blood vessel segmentation model provided in Embodiment 1 of the present invention;

[0028] Figure 3 This is a flowchart of a method for determining fractional blood flow reserve according to Embodiment 2 of the present invention;

[0029] Figure 4 This is a flowchart of a method for determining fractional blood flow reserve according to Embodiment 3 of the present invention;

[0030] Figure 5 This is a schematic diagram of a device for determining fractional blood flow reserve according to Embodiment 4 of the present invention;

[0031] Figure 6 This is a schematic diagram of a system for determining fractional blood flow reserve according to Embodiment 5 of the present invention. Detailed Implementation

[0032] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0034] To clearly understand the technical solution of this invention, a detailed description of the prior art is provided. The definition of fractional flow reserve (FFR) is:

[0035]

[0036] Among them, Q s and Q n These represent the flow rate through the branch when the vessel is narrowed and the flow rate through the branch when there is no narrowing, respectively. In the non-invasive FFR determination process, the FFR formula is as follows:

[0037]

[0038] Where P0 represents venous blood pressure, R s and R n These represent the vascular resistance at the distal end of the stenosis in the stenotic state and the normal state, respectively. The above method requires two important assumptions:

[0039] Firstly, under congested conditions, it is assumed that the distal vascular resistance is the same regardless of whether stenosis is present or not, i.e., R... s =R n ;

[0040] Secondly, the venous blood pressure is 0, that is, P0 = 0 in the formula. In reality, the venous blood pressure is about 10-20 mmHg, and the venous blood pressure in the capillary network and arterioles is even higher.

[0041] These two assumptions are also the main problems with the current technology, which introduces a certain error into the entire calculation, resulting in a large error in the currently determined fractional blood flow reserve.

[0042] To address the technical problems in existing technologies and reduce the error in fractional flow reserve (FRR), this invention acquires an initial vascular image and inputs it into a pre-trained vascular segmentation model to obtain a first vascular segmentation model. The stenotic regions of the first vascular segmentation model are then identified, and destenosis processing is performed on these regions to obtain a second vascular segmentation model. Flow data from the stenotic regions of the first and second stenotic regions of the second vascular segmentation model are then acquired. Based on the flow data from the stenotic regions of the first and second stenotic regions, the FRR is determined. Compared to existing technologies, this invention determines the FRR without introducing pressure information, which has significant errors, resulting in a more accurate FRR and thus solving the problem of large FRR errors.

[0043] Example 1

[0044] Figure 1 This is a flowchart of a method for determining fractional flow reserve (FRR) according to Embodiment 1 of the present invention. This embodiment is applicable to cases where FRR is automatically determined. The method can be executed by a FRR determination device, which can be implemented in hardware and / or software and can be configured within a FRR determination system. Specifically, the FRR determination system includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it performs the following FRR determination method.

[0045] like Figure 1 As shown, the method includes:

[0046] S110. Obtain an initial blood vessel image and input the initial blood vessel image into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model.

[0047] In this embodiment, the initial vascular image refers to the image to be segmented. For example, the initial vascular image can be medical imaging data, such as computed tomographic arteriography (CTA) or magnetic resonance imaging (MRI) data. Optionally, the initial vascular image can be an image containing coronary arteries.

[0048] Specifically, one or more initial vascular images can be obtained from the preset storage location of the electronic device, or one or more initial vascular images can be obtained from other devices connected to the electronic device or the cloud, without limitation.

[0049] Furthermore, the initial vascular image can be used as input data to a pre-trained vascular segmentation model. The vascular segmentation model can predict and output a first vascular segmentation model based on the initial vascular image. This first vascular segmentation model can be a 3D vascular segmentation image; in other words, the 3D distribution of blood vessels can be viewed through the first vascular segmentation model. For example, the first vascular segmentation model can be a 3D coronary artery tree model, etc.

[0050] Optionally, the training steps of the blood vessel segmentation model include: acquiring an initial blood vessel sample image and a blood vessel segmentation annotation image corresponding to the initial blood vessel sample image; training an initial neural network model based on the initial blood vessel sample image and the blood vessel segmentation annotation image corresponding to the initial blood vessel sample image to obtain a blood vessel segmentation model.

[0051] For example, a blood vessel segmentation model can be pre-trained using a large number of initial blood vessel sample images. In the trained neural network model, features are extracted from the initial blood vessel sample images beforehand, and the model parameters in the neural network model are trained based on the extracted feature information. By continuously adjusting the model parameters, the distance deviation between the model's output and the blood vessel segmentation annotation image gradually decreases and tends to stabilize.

[0052] S120. Determine the narrow region of the first blood vessel segmentation model, and perform destenosis processing on the narrow region of the first blood vessel segmentation model to obtain the second blood vessel segmentation model.

[0053] In this embodiment, the narrow region refers to the narrow region of the blood vessel in the first blood vessel segmentation model, as shown by the arrow in Figure 2(a).

[0054] Specifically, stenosis detection methods can be used to detect stenosis in the first blood vessel segmentation model to obtain the stenosis region of the first blood vessel segmentation model. The stenosis detection method may include, but is not limited to: inputting the first blood vessel segmentation model into a pre-trained stenosis detection model to obtain the stenosis region of the first blood vessel segmentation model. The stenosis detection model can be trained using a large number of blood vessel segmentation sample images and corresponding stenosis annotation images. Alternatively, the stenosis region of the first blood vessel segmentation model can be determined based on the diameter changes of each blood vessel branch in the first blood vessel segmentation model.

[0055] After determining the narrow region of the first blood vessel segmentation model, the narrow region of the first blood vessel segmentation model can be de-stenotic to obtain the second blood vessel segmentation model under normal conditions. It can be understood that, compared with the first blood vessel segmentation model, the second blood vessel segmentation model eliminates the narrow region of the blood vessel, as shown by the arrow in Figure 2(b).

[0056] S130. Obtain the flow data of the stenotic region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model.

[0057] In this embodiment, the flow data of the stenotic region in the first vessel segmentation model refers to the flow rate of the branch vessel cross-section where the stenotic region is located in the first vessel segmentation model; similarly, the flow data of the destenotic region in the second vessel segmentation model refers to the flow rate of the branch vessel cross-section where the destenotic region is located in the second vessel segmentation model. It can be understood that the stenotic region and the destenotic region are the same region in both models.

[0058] Specifically, three-dimensional simulation or reduction model solving can be performed on the first and second vessel segmentation models respectively to obtain the hemodynamic parameters corresponding to the first and second vessel segmentation models. Then, the narrow region of the first vessel segmentation model is truncated to obtain the flow data of the narrow region of the first vessel segmentation model, and the de-narrowing region of the second vessel segmentation model is truncated to obtain the flow data of the narrow region of the second vessel segmentation model.

[0059] S140. Determine the fractional flow reserve based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model.

[0060] Specifically, the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model can be input into the fractional flow reserve determination model. The fractional flow reserve determination model determines the fractional flow reserve based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model, and outputs it.

[0061] Based on the above embodiments, the fractional flow reserve is determined based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model, including: dividing the flow data of the stenotic region of the first vessel segmentation model by the flow data of the destenotic region of the second vessel segmentation model to obtain the fractional flow reserve.

[0062] For example, a model for determining fractional blood flow reserve can be:

[0063] FFR = Qs / Qn;

[0064] Where Qs represents the flow data of the stenotic region in the first vessel segmentation model, and Qn represents the flow data of the destenotic region in the second vessel segmentation model.

[0065] The technical solution of this invention involves acquiring an initial vascular image, inputting it into a pre-trained vascular segmentation model to obtain a first vascular segmentation model, identifying the stenotic region of the first vascular segmentation model, performing destenosis processing on the stenotic region of the first vascular segmentation model to obtain a second vascular segmentation model, acquiring flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model, and determining the fractional flow reserve (FLR) based on the flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model. Compared with the prior art, this invention determines the FLR without introducing pressure information with large errors, making the obtained FLR more accurate, thereby solving the problem of large errors in the FLR.

[0066] Example 2

[0067] Figure 3This is a flowchart of a method for determining the fractional flow reserve (FVR) according to Embodiment 2 of the present invention. The method in this embodiment can be combined with various optional schemes in the methods for determining the FVR provided in the above embodiments. The method for determining the FVR provided in this embodiment has been further optimized. Optionally, determining the narrow region of the first vessel segmentation model includes: obtaining the equivalent vessel diameter and reference vessel diameter corresponding to the centerline points of each vessel branch in the first vessel segmentation model; and determining the narrow region of the first vessel segmentation model based on the equivalent vessel diameter and reference vessel diameter corresponding to the centerline points of each vessel branch in the first vessel segmentation model.

[0068] like Figure 3 As shown, the method includes:

[0069] S210. Obtain an initial blood vessel image and input the initial blood vessel image into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model.

[0070] S220. Obtain the equivalent diameter and reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model.

[0071] In this embodiment, the equivalent vessel diameter refers to the actual diameter of the vessel, which can be calculated based on the cross-sectional area of ​​the vessel and the formula for the area of ​​a circle. The reference vessel diameter refers to the predicted diameter of the vessel, which is the diameter of the vessel under conditions without stenosis, and can be predicted through constraints or a vessel diameter prediction model. It is understood that in non-stenotic regions, the equivalent vessel diameter and the reference vessel diameter are the same, while in stenotic regions, there is a significant difference between the two.

[0072] S230. Based on the equivalent diameter of the blood vessel and the reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, determine the narrow region of the first blood vessel segmentation model.

[0073] In some embodiments, if the difference between the equivalent diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model and the reference diameter of the blood vessel corresponding to the centerline point is greater than a preset difference threshold, and the length of the centerline that meets this condition is greater than a preset length threshold, then the blood vessel position corresponding to the centerline is determined as a narrow region of the first blood vessel segmentation model.

[0074] In some embodiments, if the stenosis rate between the equivalent diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model and the reference diameter of the blood vessel corresponding to the centerline point is greater than a preset stenosis rate threshold, then the region where the centerline point corresponding to the stenosis rate is located is determined as the stenosis region of the first blood vessel segmentation model.

[0075] S240. Perform destenosis processing on the narrow region of the first blood vessel segmentation model to obtain the second blood vessel segmentation model.

[0076] S250: Obtain the flow data of the stenotic region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model.

[0077] S260. Determine the fractional flow reserve based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model.

[0078] In this embodiment, by obtaining the equivalent diameter and reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, and then determining the narrow region of the first blood vessel segmentation model based on the equivalent diameter and reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, the automatic identification of the narrow region is realized.

[0079] Based on the above embodiments, optionally, obtaining the equivalent blood vessel diameter and reference blood vessel diameter corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model includes: obtaining the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model; determining the equivalent blood vessel diameter based on the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model; and determining the reference blood vessel diameter based on pre-configured constraints and the equivalent blood vessel diameter.

[0080] The cross-sectional area of ​​a blood vessel refers to the area of ​​its cross-section; in other words, the cross-section of a blood vessel is perpendicular to its centerline.

[0081] For example, the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch can be obtained from the first blood vessel segmentation model. Substituting this cross-sectional area into the formula for calculating the area of ​​a circle, the equivalent diameter of the blood vessel can be obtained. The determination of the reference diameter of the blood vessel must satisfy the following two constraints:

[0082] 1) At the centerline point of any two vascular branches, the reference diameter of the proximal vessel is greater than or equal to the reference diameter of the distal vessel.

[0083] 2) All points on the center line satisfy Minimum;

[0084] Among them, R ref R_represents the initial reference diameter of the blood vessel, which can be customized based on experience; R_real represents the equivalent diameter of the blood vessel.

[0085] Based on the above embodiments, optionally, determining the narrow region of the first blood vessel segmentation model based on the equivalent blood vessel diameter and reference blood vessel diameter corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model includes: comparing the equivalent blood vessel diameter corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model with the reference blood vessel diameter corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, determining the blood vessel stenosis rate based on the comparison result; and determining the narrow region of the first blood vessel segmentation model based on the blood vessel stenosis rate.

[0086] Among them, the vascular stenosis rate is used to characterize the degree of stenosis of blood vessels. It is understandable that blood vessels are irregular, and when the vascular stenosis rate meets the criteria for a stenotic region, that region can be determined to be a stenotic region to avoid misjudgment.

[0087] For example, R_ref represents the reference diameter of the blood vessel, R_real represents the equivalent diameter of the blood vessel, and R_diff = R_ref - R_real represents the difference between the reference diameter of the blood vessel and the equivalent diameter of the blood vessel. If R_diff > 0, it indicates that the centerline point is a potential stenosis location, and the potential stenosis location can be represented by S1, S2, ... Sn. Specifically, if R_diff > 0 at the m-th centerline point and R_diff = 0 at the (m-1)-th centerline point, then this centerline point is the starting position of the narrow region. If R_diff > 0 at all centerline points from m to m+k, and R_diff = 0 at the (m+k+1)-th centerline point, then the (m+k)-th centerline point is the ending position of the narrow region. The distance between the m-th and m+k-th centerline points can be determined based on the number of centerline points and the distance between them. Taking Max((R_ref - R_real) / R_ref) over the region between these centerline points yields the maximum stenosis rate. Furthermore, when the maximum stenosis rate exceeds a preset stenosis rate threshold, the region between these nodes is determined as the narrow region of the first vessel segmentation model. The preset stenosis rate threshold can be determined empirically.

[0088] The technical solution of this invention obtains the equivalent diameter and reference diameter of the blood vessels corresponding to the centerline points of each blood vessel branch in the first blood vessel segmentation model, and then determines the narrow region of the first blood vessel segmentation model based on the equivalent diameter and reference diameter of the blood vessels corresponding to the centerline points of each blood vessel branch in the first blood vessel segmentation model, thereby realizing the automatic identification of the narrow region.

[0089] Example 3

[0090] Figure 4This is a flowchart of a method for determining the fractional flow reserve (FVRRP) according to Embodiment 3 of the present invention. The method in this embodiment can be combined with various optional schemes in the methods for determining FVRRP provided in the above embodiments. The method for determining FVRRP provided in this embodiment has been further optimized. Optionally, the step of destenosis processing on the narrow region of the first vessel segmentation model to obtain the second vessel segmentation model includes: replacing the equivalent vessel diameter corresponding to the vessel centerline point of the narrow region of the first vessel segmentation model with the vessel reference diameter to obtain the second vessel segmentation model.

[0091] like Figure 4 As shown, the method includes:

[0092] S310. Obtain an initial blood vessel image and input the initial blood vessel image into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model.

[0093] S320. Determine the narrow region of the first blood vessel segmentation model, and replace the equivalent blood vessel diameter corresponding to the blood vessel centerline point of the narrow region of the first blood vessel segmentation model with the blood vessel reference diameter to obtain the second blood vessel segmentation model.

[0094] S330. Obtain the flow data of the stenotic region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model.

[0095] S340. Determine the fractional flow reserve based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model.

[0096] In this embodiment, the equivalent diameter of the blood vessel corresponding to the center line point of the narrow region in the first blood vessel segmentation model is replaced with the reference diameter of the blood vessel, so that the diameter of the narrow region is restored to the normal state, thereby obtaining a second blood vessel segmentation model without stenosis.

[0097] Specifically, by inputting an initial vascular image into a pre-trained vascular segmentation model, a first vascular segmentation model is obtained. Then, the stenotic region of the first vascular segmentation model is determined. The equivalent vascular diameter corresponding to the vascular centerline point of the stenotic region in the first vascular segmentation model is replaced with the vascular reference diameter, restoring the stenotic region diameter to a normal state, thus obtaining a second vascular segmentation model without stenosis. Flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model are then obtained. Based on the flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model, the fractional flow reserve (FRR) is determined. Compared with existing technologies, this invention determines the FRR without introducing pressure information with large errors, making the obtained FRR more accurate and thus solving the problem of large FRR errors.

[0098] Based on the above embodiments, optionally, obtaining the flow data of the narrow region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model includes: performing three-dimensional simulation solutions on the first blood vessel segmentation model and the second blood vessel segmentation model respectively to obtain the flow data of the narrow region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model; or, performing reduced-order model solutions on the first blood vessel segmentation model and the second blood vessel segmentation model respectively to obtain the flow data of the narrow region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model.

[0099] The 3D simulation solution includes two processes: mesh discretization and fluid dynamics equation solving. Mesh discretization refers to discretizing the internal space of the segmented blood vessel model to output the model's mesh information. The specific meshing process includes: setting the global mesh size, setting local mesh sizes for different blood vessels, and finally meshing the entire model's surface and volumetric space. Fluid dynamics equation solving mainly refers to solving the Navier-Stokes (NS) equations. Before solving the fluid dynamics equations, simulation parameters need to be set, including: fluid and pipe wall material properties; initial conditions of the flow field; calculating arterial blood flow based on myocardial mass and deriving the inlet flow field distribution; calculating the flow resistance of each outlet based on the end face area and making appropriate adjustments for congestion conditions; determining the boundary condition type and allocating the flow resistance and flow capacity of each outlet according to a certain proportion; in addition, other parameters required by the fluid simulation solver need to be set. The setting of simulation parameters can be done manually or automatically by the system based on preset parameters.

[0100] Solving the reduced-order model can include solving the one-dimensional model and solving the zero-dimensional model. For the one-dimensional model, the blood vessel segmentation model is simplified to a single-scale calculation along the blood vessel centerline, and the mass conservation equation and momentum conservation equation are solved on the single scale. The equations are as follows:

[0101]

[0102]

[0103] Where A represents the cross-sectional area of ​​the blood vessel, U represents the average axial velocity over the cross-sectional area, P represents the average pressure over the cross-sectional area, and f represents the average frictional resistance per unit length. Similar to the three-dimensional simulation solution process, solving the one-dimensional equations also requires setting boundary conditions and other parameters. These other parameters can include the inlet flow rate and multiple outlet flow resistances. The inlet flow rate can be calculated based on the myocardial volume, and then the flow rate and pressure information in the entire one-dimensional field can be solved based on the flow resistance of each outlet.

[0104] For the zero-dimensional model solution method: the blood vessel segmentation model is equivalent to a circuit model. In the circuit model, the resistance R is equivalent to the flow resistance in the blood vessel, the current is equivalent to the blood flow in the blood vessel, and the voltage is equivalent to the blood pressure. Then, according to Poiseuille's law, the resistance in the circular tube... Where μ represents the fluid viscosity coefficient, r is the radius, and dl represents the length; thus, the relationship between flow resistance, blood flow, and blood pressure is obtained according to Kirchhoff's laws, and the flow rate and pressure field of the entire vascular segmentation model are then solved.

[0105] The technical solution of this invention is to replace the equivalent diameter of the blood vessel corresponding to the center line point of the narrow region of the first blood vessel segmentation model with the reference diameter of the blood vessel, so that the diameter of the narrow region is restored to a normal state, thereby obtaining a second blood vessel segmentation model without stenosis.

[0106] Example 4

[0107] Figure 5 This is a schematic diagram of a device for determining the fractional blood flow reserve provided in Embodiment 4 of the present invention. Figure 5 As shown, the device includes:

[0108] The first blood vessel segmentation model determination module 410 is used to acquire an initial blood vessel image and input the initial blood vessel image into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model.

[0109] The second blood vessel segmentation model determination module 420 is used to determine the narrow region of the first blood vessel segmentation model, perform destenosis processing on the narrow region of the first blood vessel segmentation model, and obtain the second blood vessel segmentation model.

[0110] The model flow data acquisition module 430 is used to acquire the flow data of the stenotic region of the first blood vessel segmentation model and the flow data of the destenotic region of the second blood vessel segmentation model.

[0111] The fractional flow reserve determination module 440 is used to determine the fractional flow reserve based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model.

[0112] The technical solution of this invention involves acquiring an initial vascular image, inputting it into a pre-trained vascular segmentation model to obtain a first vascular segmentation model, identifying the stenotic region of the first vascular segmentation model, performing destenosis processing on the stenotic region of the first vascular segmentation model to obtain a second vascular segmentation model, acquiring flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model, and determining the fractional flow reserve (FLR) based on the flow data from the stenotic region of the first vascular segmentation model and the destenotic region of the second vascular segmentation model. Compared with the prior art, this invention determines the FLR without introducing pressure information with large errors, making the obtained FLR more accurate, thereby solving the problem of large errors in the FLR.

[0113] In some optional implementations, the training steps of the blood vessel segmentation model include:

[0114] Obtain an initial blood vessel sample image and a corresponding blood vessel segmentation and annotation image;

[0115] The initial neural network model is trained based on the initial blood vessel sample image and the corresponding blood vessel segmentation annotation image to obtain the blood vessel segmentation model.

[0116] In some alternative implementations, the second vessel segmentation model determination module 420 includes:

[0117] The pipe diameter information acquisition unit is used to acquire the equivalent pipe diameter and reference pipe diameter of the blood vessel corresponding to the center line point of each blood vessel branch in the first blood vessel segmentation model.

[0118] The narrow region determination unit is used to determine the narrow region of the first blood vessel segmentation model based on the equivalent diameter of the blood vessel and the reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model.

[0119] In some optional implementations, the pipe diameter information acquisition unit is further configured to:

[0120] Obtain the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model;

[0121] The equivalent diameter of the blood vessel is determined based on the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model.

[0122] The reference diameter of the blood vessel is determined based on pre-configured constraints and the equivalent diameter of the blood vessel.

[0123] In some alternative implementations, the narrow region determination unit is further configured to:

[0124] The equivalent diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model is compared with the reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, and the blood vessel stenosis rate is determined based on the comparison results.

[0125] The narrow region of the first blood vessel segmentation model is based on the vascular stenosis rate.

[0126] In some optional implementations, the second vessel segmentation model determination module 420 is further configured to:

[0127] The equivalent diameter of the blood vessel corresponding to the centerline point of the narrow region in the first blood vessel segmentation model is replaced with the reference diameter of the blood vessel to obtain the second blood vessel segmentation model.

[0128] In some alternative implementations, the model traffic data acquisition module 430 is further configured to:

[0129] Three-dimensional simulations were performed on the first blood vessel segmentation model and the second blood vessel segmentation model respectively to obtain the flow data of the narrow region of the first blood vessel segmentation model and the flow data of the denarrow region of the second blood vessel segmentation model.

[0130] Alternatively, the first blood vessel segmentation model and the second blood vessel segmentation model can be solved by reducing their order to obtain the flow data of the narrow region of the first blood vessel segmentation model and the flow data of the denarrow region of the second blood vessel segmentation model.

[0131] In some alternative implementations, the fractional flow reserve determination module 440 is further configured to:

[0132] The blood flow reserve fraction is obtained by dividing the flow data of the stenotic region in the first blood vessel segmentation model by the flow data of the destenotic region in the second blood vessel segmentation model.

[0133] The fractional flow reserve determination device provided in the embodiments of the present invention can execute the fractional flow reserve determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0134] Example 5

[0135] Figure 6 This is a schematic diagram of the structure of the fractional blood flow reserve determination system provided in an embodiment of the present invention, as shown below. Figure 6 As shown, the system includes a processor 501 and a memory 502, which stores computer programs. The system can have one or more processors 501. Figure 6 Take the 501 processor as an example;

[0136] The memory 502, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the method for determining the fractional blood flow reserve in this embodiment of the invention. The processor 501 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 502, namely, executing a method for determining the fractional blood flow reserve, which includes:

[0137] An initial blood vessel image is obtained, and the initial blood vessel image is input into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model;

[0138] The narrow region of the first blood vessel segmentation model is determined, and the narrow region of the first blood vessel segmentation model is de-stenotic to obtain the second blood vessel segmentation model.

[0139] Obtain flow data of the stenotic region from the first vessel segmentation model and flow data of the destenotic region from the second vessel segmentation model;

[0140] The fractional flow reserve is determined based on the flow data of the stenotic region from the first vessel segmentation model and the flow data of the destenotic region from the second vessel segmentation model.

[0141] Of course, when the processor provided in the embodiments of the present invention executes computer program instructions, it is not limited to the method operation described above, but can also execute related operations in the method for determining the fractional blood flow reserve provided in any embodiment of the present invention.

[0142] Memory 502 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on terminal usage. Furthermore, memory 502 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, memory 502 may further include memory remotely located relative to processor 501, which can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0143] The system also includes an input device 503 and an output device 504; the processor 501, memory 502, input device 503, and output device 504 in the device can be connected via a bus or other means. Figure 6 Taking the example of a connection between China and Israel via a bus.

[0144] The input device 503 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the device.

[0145] The output device 504 may include a display device such as a display screen, for example, the display screen of a user terminal.

[0146] Example 6

[0147] This invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a method for determining a fractional blood flow reserve, the method comprising:

[0148] An initial blood vessel image is obtained, and the initial blood vessel image is input into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model;

[0149] The narrow region of the first blood vessel segmentation model is determined, and the narrow region of the first blood vessel segmentation model is de-stenotic to obtain the second blood vessel segmentation model.

[0150] Obtain flow data of the stenotic region from the first vessel segmentation model and flow data of the destenotic region from the second vessel segmentation model;

[0151] The fractional flow reserve is determined based on the flow data of the stenotic region from the first vessel segmentation model and the flow data of the destenotic region from the second vessel segmentation model.

[0152] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the method operations described above, but can also perform related operations in the method for determining the fractional blood flow reserve provided in any embodiment of the present invention.

[0153] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the method for determining the fractional blood flow reserve described in the various embodiments of the present invention.

[0154] It is worth noting that in the embodiments of the above-mentioned fractional blood flow reserve determination device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0155] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A system for determining fractional blood flow reserve, characterized in that, The system includes a processor and a memory, the memory storing a computer program, and the processor performing the following steps when running the computer program: An initial blood vessel image is obtained, and the initial blood vessel image is input into a pre-trained blood vessel segmentation model to obtain a first blood vessel segmentation model; The narrow region of the first blood vessel segmentation model is determined, and the narrow region of the first blood vessel segmentation model is de-stenotic to obtain the second blood vessel segmentation model. Obtain flow data of the stenotic region from the first vessel segmentation model and flow data of the destenotic region from the second vessel segmentation model; The fractional flow reserve is determined based on the flow data of the stenotic region in the first vessel segmentation model and the flow data of the destenotic region in the second vessel segmentation model. The step of obtaining the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model includes: Three-dimensional simulations were performed on the first blood vessel segmentation model and the second blood vessel segmentation model respectively to obtain the flow data of the narrow region of the first blood vessel segmentation model and the flow data of the denarrow region of the second blood vessel segmentation model. Alternatively, a reduced-order model can be solved for both the first and second vessel segmentation models to obtain flow data for the stenotic region of the first vessel segmentation model and flow data for the destenotic region of the second vessel segmentation model; wherein, the flow data for the stenotic region of the first vessel segmentation model refers to the flow rate of the branch vessel cross-section where the stenotic region is located in the first vessel segmentation model; and the flow data for the destenotic region of the second vessel segmentation model refers to the flow rate of the branch vessel cross-section where the destenotic region is located in the second vessel segmentation model. The step of determining the narrow region of the first blood vessel segmentation model includes: Obtain the equivalent diameter and reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model. The equivalent diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model is compared with the reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, and the blood vessel stenosis rate is determined based on the comparison results. The narrow region of the first blood vessel segmentation model is determined based on the blood vessel stenosis rate; wherein the blood vessel stenosis rate is used to characterize the degree of narrowing of the blood vessel.

2. The system according to claim 1, characterized in that, The training steps for the blood vessel segmentation model include: Obtain an initial blood vessel sample image and a corresponding blood vessel segmentation and annotation image; The initial neural network model is trained based on the initial blood vessel sample image and the corresponding blood vessel segmentation annotation image to obtain the blood vessel segmentation model.

3. The system according to claim 1, characterized in that, The step of obtaining the equivalent diameter and reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model includes: Obtain the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model; The equivalent diameter of the blood vessel is determined based on the cross-sectional area of ​​the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model. The reference diameter of the blood vessel is determined based on pre-configured constraints and the equivalent diameter of the blood vessel.

4. The system according to claim 1, characterized in that, The step of destenosis processing on the narrow region of the first blood vessel segmentation model to obtain the second blood vessel segmentation model includes: The equivalent diameter of the blood vessel corresponding to the centerline point of the narrow region in the first blood vessel segmentation model is replaced with the reference diameter of the blood vessel to obtain the second blood vessel segmentation model.

5. The system according to claim 1, characterized in that, The determination of the fractional flow reserve based on the flow data of the stenotic region from the first vessel segmentation model and the flow data of the destenotic region from the second vessel segmentation model includes: The blood flow reserve fraction is obtained by dividing the flow data of the stenotic region in the first blood vessel segmentation model by the flow data of the destenotic region in the second blood vessel segmentation model.

6. A device for determining fractional blood flow reserve, characterized in that, include: The first blood vessel segmentation model determination module is used to acquire an initial blood vessel image and input the initial blood vessel image into a pre-trained blood vessel segmentation model to obtain the first blood vessel segmentation model. The second blood vessel segmentation model determination module is used to determine the narrow region of the first blood vessel segmentation model, and to perform destenosis processing on the narrow region of the first blood vessel segmentation model to obtain the second blood vessel segmentation model. The model flow data acquisition module is used to acquire flow data of the stenotic region of the first blood vessel segmentation model and flow data of the destenotic region of the second blood vessel segmentation model; The fractional flow reserve (FRR) determination module is used to determine the FRR based on the flow data of the stenotic region of the first vessel segmentation model and the flow data of the destenotic region of the second vessel segmentation model. The model flow data acquisition module is further configured to: perform three-dimensional simulation solutions on the first blood vessel segmentation model and the second blood vessel segmentation model respectively to obtain flow data of the stenotic region of the first blood vessel segmentation model and flow data of the destenotic region of the second blood vessel segmentation model; or, perform reduced-order model solutions on the first blood vessel segmentation model and the second blood vessel segmentation model respectively to obtain flow data of the stenotic region of the first blood vessel segmentation model and flow data of the destenotic region of the second blood vessel segmentation model; wherein, the flow data of the stenotic region of the first blood vessel segmentation model refers to the flow of the branch blood vessel section where the stenotic region is located in the first blood vessel segmentation model; the flow data of the destenotic region of the second blood vessel segmentation model refers to the flow of the branch blood vessel section where the destenotic region is located in the second blood vessel segmentation model; The second vessel segmentation model determination module includes: The pipe diameter information acquisition unit is used to acquire the equivalent pipe diameter and the reference pipe diameter of the blood vessel corresponding to the center line point of each blood vessel branch in the first blood vessel segmentation model. The narrowing region determination unit is used to compare the equivalent diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model with the reference diameter of the blood vessel corresponding to the centerline point of each blood vessel branch in the first blood vessel segmentation model, and determine the vascular stenosis rate based on the comparison result; and determine the narrowing region of the first blood vessel segmentation model based on the vascular stenosis rate; wherein, the vascular stenosis rate is used to characterize the degree of vascular stenosis.

Images