Method, device, electronic device and storage medium for obtaining vascular physiological parameters

By obtaining contrast agent concentration characteristic information to calculate blood flow, and generating a model to obtain vascular physiological parameters, solving the problem of inaccurate vascular physiological parameters caused by inaccurate estimation of vascular blood flow in the prior art, achieving higher accuracy.

CN114420286BActive Publication Date: 2024-11-22SHENZHEN RAYSIGHT INTELLIGENT MEDICAL TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210045153.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-14
Publication Date
2024-11-22
Estimated Expiration
2042-01-14

AI Technical Summary

Technical Problem

The estimate of vascular blood flow in the prior art depends on experience, resulting in inaccuracy of vascular physiological parameters.

Method used

By obtaining the concentration characteristic information of the contrast agent of the target blood vessel in the time dimension, the blood flow information is calculated, and a model is generated based on the blood flow information to obtain the physiological parameters of the blood vessel.

Benefits of technology

It improves the accuracy of vascular physiological parameters and can better conform to personalized characteristics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114420286B_ABST
    Figure CN114420286B_ABST
Patent Text Reader

Abstract

The present application provides a method, device, electronic device and storage medium for obtaining vascular physiological parameters, the method comprising: obtaining concentration characteristic information of a contrast agent at a preset position of a target blood vessel in a time dimension; obtaining blood flow information of the target blood vessel based on the concentration characteristic information; obtaining a model generated for the target blood vessel; and obtaining physiological parameters of the target blood vessel based on the model and the blood flow information of the target blood vessel. The embodiments of the present application can improve the accuracy of the obtained vascular physiological parameters.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the medical field, and specifically to a method, device, electronic device and storage medium for acquiring vascular physiological parameters. Background Art

[0002] In the medical field, it is often necessary to obtain the patient's vascular physiological parameters in order to make a diagnosis or other medical activities based on this. In the prior art, the blood flow of the blood vessels is usually first estimated based on experience, and then the patient's vascular physiological parameters are estimated based on the blood flow of the blood vessels. Since the blood flow in the prior art can only be estimated based on experience, the blood flow in the prior art is usually not accurate enough, which leads to the fact that the vascular physiological parameters in the prior art are usually not accurate enough. Summary of the invention

[0003] One purpose of the present application is to provide a method, device, electronic device and storage medium for obtaining vascular physiological parameters, which can improve the accuracy of the obtained vascular physiological parameters.

[0004] According to one aspect of an embodiment of the present application, a method for acquiring vascular physiological parameters is disclosed, the method comprising:

[0005] Acquiring concentration characteristic information of the contrast agent at a preset position of the target blood vessel in a time dimension;

[0006] Based on the concentration characteristic information, acquiring blood flow information of the target blood vessel;

[0007] Acquire a model generated for the target blood vessel;

[0008] Based on the model and the blood flow information of the target blood vessel, the physiological parameters of the target blood vessel are obtained.

[0009] According to one aspect of an embodiment of the present application, a device for acquiring vascular physiological parameters is disclosed, the device comprising:

[0010] A concentration feature acquisition module, configured to acquire concentration feature information of the contrast agent at a preset position of the target blood vessel in a time dimension;

[0011] a blood flow acquisition module, configured to acquire blood flow information of the target blood vessel based on the concentration characteristic information;

[0012] A model acquisition module, configured to acquire a model generated for the target blood vessel;

[0013] The physiological parameter acquisition module is configured to acquire the physiological parameters of the target blood vessel based on the model and the blood flow information of the target blood vessel.

[0014] In an exemplary embodiment of the present application, the concentration feature acquisition module is configured as follows:

[0015] Obtaining a curve of the concentration of the contrast agent changing with time;

[0016] Based on the change curve, the concentration characteristic information is acquired.

[0017] In an exemplary embodiment of the present application, the blood flow acquisition module is configured as follows:

[0018] Acquiring the volume of the functional tissue where the target blood vessel is located;

[0019] Based on the concentration characteristic information, obtaining the time difference between contrast agent concentration nodes;

[0020] Based on the volume and the time difference, blood flow information of the target blood vessel is calculated.

[0021] In an exemplary embodiment of the present application, the blood flow acquisition module is configured as follows:

[0022] Based on the original medical image of the target blood vessel, acquiring intraluminal attenuation gradient information of the target blood vessel;

[0023] Based on the concentration characteristic information, obtaining the time difference between contrast agent concentration nodes;

[0024] The blood flow information of the target blood vessel is calculated based on the intracavitary attenuation gradient information and the time difference.

[0025] In an exemplary embodiment of the present application, the contrast agent concentration nodes include: a node where the contrast agent concentration is zero, a node where the contrast agent concentration peak is, a node where the contrast agent concentration mean is, and a node where the contrast agent concentration plateau is.

[0026] In an exemplary embodiment of the present application, the model acquisition module is configured to: establish a physical model of the target blood vessel based on an original medical image of the target blood vessel;

[0027] The physiological parameter acquisition module is configured to solve the physical model of the target blood vessel based on the blood flow information of the target blood vessel to obtain the physiological parameters of the target blood vessel.

[0028] In an exemplary embodiment of the present application, the physical model of the target blood vessel includes: a three-dimensional model of the target blood vessel, a one-dimensional model of the target blood vessel, a reduced-order model of the target blood vessel, and a circuit-like model of the target blood vessel.

[0029] In an exemplary embodiment of the present application, the model acquisition module is configured to: use the original medical image including the sample blood vessel and the first information of the blood flow information of the sample blood vessel as the input of the preset machine learning model, and train the machine learning model with the physiological parameters of the sample blood vessel as the target output to obtain a trained machine learning model;

[0030] The physiological parameter acquisition module is configured to: input the second information including the original medical image of the target blood vessel and the blood flow information of the target blood vessel into the trained machine learning model to obtain the physiological parameters of the target blood vessel.

[0031] In an exemplary embodiment of the present application, the first information further includes geometric feature information of the sample blood vessel and lesion feature information of the sample blood vessel, and the second information further includes geometric feature information of the target blood vessel and lesion feature information of the target blood vessel.

[0032] According to one aspect of an embodiment of the present application, an electronic device is disclosed, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device implements any one of the above embodiments.

[0033] According to one aspect of an embodiment of the present application, a computer program medium is disclosed, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a computer, the computer is caused to execute any one of the above embodiments.

[0034] According to one aspect of the embodiments of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the above various optional implementations.

[0035] In the embodiment of the present application, since the blood flow information is obtained based on the concentration characteristic information of the contrast agent in the time dimension, the obtained blood flow information can more accurately meet the personalized characteristics of the target blood vessel, so that the physiological parameters obtained on this basis can more accurately meet the personalized characteristics of the target blood vessel, thereby improving the accuracy of the obtained vascular physiological parameters.

[0036] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0037] It should be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above and other objects, features and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0039] Figure 1 A flow chart of a method for acquiring vascular physiological parameters according to an embodiment of the present application is shown.

[0040] Figure 2 A curve showing the change of contrast agent concentration with blood vessel length according to one embodiment of the present application is shown.

[0041] Figure 3 A schematic diagram of training a deep neural network for image segmentation according to an embodiment of the present application is shown.

[0042] Figure 4 A schematic diagram of a deep neural network trained to obtain blood flow information according to an embodiment of the present application is shown.

[0043] Figure 5 A schematic diagram of physiological parameter visualization results according to an embodiment of the present application is shown.

[0044] Figure 6 A framework diagram of a process for acquiring vascular physiological parameters according to an embodiment of the present application is shown.

[0045] Figure 7 A block diagram of a device for acquiring vascular physiological parameters according to an embodiment of the present application is shown.

[0046] Figure 8 A hardware diagram of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0047] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present application will be more comprehensive and complete and the concepts of the example embodiments will be fully conveyed to those skilled in the art. The accompanying drawings are only schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0048] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present application. However, those skilled in the art will appreciate that the technical solution of the present application may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring the present application and making the various aspects of the present application obscure.

[0049] Some of the blocks shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0050] This application proposes a method for obtaining vascular physiological parameters, which is mainly used in the medical field to obtain vascular physiological parameters. The vascular physiological parameters include: vascular blood pressure, vascular blood flow velocity, vascular blood flow reserve fraction, vascular wall shear force, vascular axial plaque stress, and vascular oscillation shear index.

[0051] Figure 1 The flowchart of the method for obtaining vascular physiological parameters according to an embodiment of the present application is shown. The execution subject of the method can be any device with sufficient computing power (for example, a personal computer, a server, etc.). The method includes:

[0052] Step S110, obtaining concentration characteristic information of the contrast agent at a preset position of the target blood vessel in the time dimension;

[0053] Step S120, obtaining blood flow information of the target blood vessel based on the concentration characteristic information;

[0054] Step S130, obtaining a model generated for the target blood vessel;

[0055] Step S140: Acquire physiological parameters of the target blood vessel based on the model and the blood flow information of the target blood vessel.

[0056] Specifically, after mixing into the blood flow, the distribution and movement of the contrast agent can accurately reflect the distribution and movement of the blood flow in real time. Considering that after a certain amount of contrast agent is mixed into the blood flow, the concentration of the contrast agent at a certain position of the blood vessel will change over time in accordance with the laws of physics under the influence of the blood flow, therefore, according to the change of the contrast agent concentration in the time dimension, the corresponding blood flow can be accurately obtained.

[0057] Therefore, in the process of obtaining the physiological parameters of the target blood vessel, in order to accurately evaluate the blood flow of the target blood vessel, in the embodiment of the present application, the blood flow information of the target blood vessel is obtained based on the concentration characteristic information of the contrast agent at the preset position of the target blood vessel in the time dimension. For example: the entrance of the target blood vessel is pre-selected as the reference position of the concentration characteristic information. Then, the concentration characteristic information of the contrast agent at the entrance of the target blood vessel in the time dimension is obtained. Then, based on the concentration characteristic information, the blood flow of the target blood vessel is accurately obtained.

[0058] In the embodiment of the present application, a model generated for the target blood vessel is also obtained. The model is mainly used to simulate the performance of the target blood vessel in a specific space, for example: the performance in the physical space, and the performance in the neural network space. The model generated for the target blood vessel in the embodiment of the present application can be a physical model for simulating the performance of the target blood vessel in the physical space, or a machine learning model for simulating the performance of the target blood vessel in the neural network space.

[0059] Furthermore, in the embodiment of the present application, physiological parameters of the target blood vessel are obtained based on the model of target blood vessel generation and the blood flow information of the target blood vessel.

[0060] It can be seen that in the embodiment of the present application, since the blood flow information is obtained based on the concentration characteristic information of the contrast agent in the time dimension, the obtained blood flow information can more accurately meet the personalized characteristics of the target blood vessel, so that the physiological parameters obtained on this basis can more accurately meet the personalized characteristics of the target blood vessel, thereby improving the accuracy of the obtained vascular physiological parameters.

[0061] The optional embodiments of the present application are described in detail below. It should be noted that the embodiments shown below are only exemplary descriptions of the optional implementation schemes of the present application, and do not mean that the present application can only implement the corresponding functions according to the embodiments shown below, and should not limit the functions and scope of use of the present application.

[0062] In one embodiment, the preset position of the target blood vessel includes: an inlet of the target blood vessel, an outlet of the target blood vessel, and a bifurcation of the target blood vessel.

[0063] Specifically, the entrance of the target blood vessel can be pre-selected as the reference position of the concentration characteristic information. Then, the original medical image of the target blood vessel is subjected to image recognition to locate the entrance of the target blood vessel in the image. Then, by monitoring the entrance of the target blood vessel in the image, the concentration characteristic information of the contrast agent at the entrance of the target blood vessel in the time dimension is obtained. Then, according to the concentration characteristic information of the contrast agent at the entrance of the target blood vessel in the time dimension, the blood flow information of the target blood vessel is obtained.

[0064] The original medical images refer to the images of the internal tissues of organisms obtained in the medical field. Common original medical images include but are not limited to: CT (Computed Tomography) images and NMRI (Nuclear Magnetic Resonance Imaging) images.

[0065] Similarly, the outlet of the target blood vessel may be preselected as the reference position of the concentration characteristic information, and then the blood flow information of the target blood vessel is obtained according to the concentration characteristic information of the contrast agent at the outlet of the target blood vessel in the time dimension.

[0066] It should be noted that in other embodiments, positions other than the inlet, outlet and bifurcation may also be selected, for example: the middle position between the inlet of the blood vessel and the bifurcation of the blood vessel.

[0067] Similarly, the bifurcation of the target blood vessel may be preselected as the reference position of the concentration characteristic information, and then the blood flow information of the target blood vessel is obtained according to the concentration characteristic information of the contrast agent at the bifurcation of the target blood vessel in the time dimension.

[0068] In one embodiment, the preset position of the target blood vessel may be one or more. When the preset position of the target blood vessel is more than one, the concentration characteristic information at each preset position is obtained respectively, and then the corresponding blood flow information is obtained based on each concentration characteristic information respectively, and then all the blood flow information is integrated to obtain the final blood flow information of the target blood vessel.

[0069] In one embodiment, the concentration value of the contrast agent at the preset position is periodically collected to obtain a concentration value sequence, and then based on the collected concentration value sequence, the concentration characteristic information is obtained.

[0070] Specifically, the preset position of the target blood vessel in the monitoring image is detected. During the monitoring process, the concentration value of the contrast agent at the preset position is collected every preset time. After the monitoring is completed, the collected concentration values ​​are arranged in the order of collection to obtain a corresponding concentration value sequence. The concentration value sequence is used to discretely describe the value of the concentration of the contrast agent at the preset position in the time dimension.

[0071] Then, based on the concentration value sequence, concentration characteristic information is obtained, so as to obtain blood flow information of the target blood vessel on this basis.

[0072] In one embodiment, a change curve of the concentration of the contrast agent over time is obtained, and concentration characteristic information is obtained based on the change curve.

[0073] Specifically, the preset position of the target blood vessel in the monitoring image is used to generate a change curve of the concentration of the contrast agent at the preset position over time based on the information obtained from the monitoring. The change curve is used to continuously describe the value of the concentration of the contrast agent at the preset position in the time dimension. One optional change curve is the arterial input function (AIF).

[0074] Then, based on the change curve, the concentration characteristic information is obtained, so as to obtain the blood flow information of the target blood vessel on this basis.

[0075] The advantage of this embodiment is that, based on the change curve of the contrast agent concentration over time, the concentration of the contrast agent in the time dimension can be completely and accurately described, thereby ensuring the accuracy of the blood flow information obtained on this basis.

[0076] In one embodiment, the volume of the functional tissue where the target blood vessel is located is obtained. Based on the concentration characteristic information, the time difference between the contrast agent concentration nodes is obtained. Based on the volume and the time difference, the blood flow information of the target blood vessel is calculated.

[0077] Specifically, the functional tissue where the target blood vessel is located refers to the vascular tissue to which the target blood vessel belongs, or the functional tissue that wraps the target blood vessel (e.g., muscle tissue). When the target blood vessel is part of a coronary artery, the functional tissue where the target blood vessel is located may be the coronary artery, the ventricle, or the myocardium.

[0078] The contrast agent concentration node refers to a node that has a trend identification significance for the contrast agent concentration. The contrast agent concentration nodes include: the node where the contrast agent concentration is zero, the node where the contrast agent concentration is peak, the node where the contrast agent concentration is mean, and the node where the contrast agent concentration is stable. Among them, the node where the contrast agent concentration is zero refers to the node where the contrast agent concentration is 0; the node where the contrast agent concentration is peak refers to the node where the contrast agent concentration is the highest value; the node where the contrast agent concentration is mean refers to the node where the contrast agent concentration is the average value; the node where the contrast agent concentration is stable refers to the node where the contrast agent concentration is stable within a certain period of time.

[0079] At least two contrast agent concentration nodes are identified through the concentration characteristic information, and then the time difference between the contrast agent concentration nodes is obtained.

[0080] Since the size of blood flow directly affects the speed of change of contrast agent concentration, and the volume of the functional tissue where the target blood vessel is located describes the volume of the space where the blood flow is located to a certain extent, the blood flow information of the target blood vessel can be calculated based on the volume of the functional tissue where the target blood vessel is located and the time difference between the contrast agent concentration nodes.

[0081] In one embodiment, the blood flow information of the target blood vessel is calculated using the following formula:

[0082] Q=V / t

[0083] Among them, Q is the blood flow information of the target blood vessel; V can be the coronary volume of the heart, the ventricular volume, the myocardial volume, or a combination of the above volumes; t is the time difference between two contrast agent concentration nodes.

[0084] In one embodiment, based on the original medical image of the target blood vessel, the intraluminal attenuation gradient information of the target blood vessel is obtained. Based on the concentration characteristic information, the time difference between the contrast agent concentration nodes is obtained. Based on the intraluminal attenuation gradient information and the time difference, the blood flow information of the target blood vessel is calculated.

[0085] Specifically, the transluminal attenuation gradient information (TAG) of the target vessel refers to the gradient of the contrast agent concentration attenuating along the center line of the target vessel along the length of the vessel. The center line of the target vessel refers to the line segment used to mark the center of the target vessel. The center line is also called the midline, and is usually represented by a line segment connected by spaced points and short line segments.

[0086] refer to Figure 2 The curve of the contrast agent concentration changing with the length of the blood vessel is shown. After obtaining the original medical image of the target blood vessel, the center line of the target blood vessel is captured. The contrast agent concentration corresponding to the point on the center line is recorded, and then a curve of the contrast agent concentration changing with the length of the blood vessel is drawn. Then a linear fit is performed on the change curve to obtain the corresponding fitting oblique line. The slope of the fitting oblique line is the TAG value.

[0087] Since the blood flow directly affects the speed of change of contrast agent concentration, it will also affect the TAG value of the target blood vessel. Therefore, the blood flow information of the target blood vessel can be calculated based on the intraluminal attenuation gradient information of the target blood vessel and the time difference between contrast agent concentration nodes.

[0088] The contrast agent concentration nodes include: a node where the contrast agent concentration is zero, a node where the contrast agent concentration peak is, a node where the contrast agent concentration mean is, and a node where the contrast agent concentration plateau is.

[0089] In one embodiment, the blood flow information of the target blood vessel is calculated using the following formula:

[0090] Q=a*t b *(-TAG) c

[0091] Among them, Q is the blood flow information of the target blood vessel; a is a coefficient related to the geometric size (area, length, etc.) of the target blood vessel; b and c are constants; TAG is usually a negative number, so the minus sign is taken here to make it a positive value; t is the time difference between two contrast agent concentration nodes.

[0092] It should be noted that in other embodiments, the time difference between contrast agent concentration nodes can be directly read from the contrast agent concentration monitoring device, and the read time difference between contrast agent concentration nodes is directly used as concentration characteristic information. Then, combined with the volume of the functional tissue where the target blood vessel is located, or combined with the intraluminal attenuation gradient information of the target blood vessel, the blood flow information of the target blood vessel is calculated.

[0093] In one embodiment, a physical model of the target blood vessel is established based on the original medical image of the target blood vessel, and the physical model of the target blood vessel is solved based on the blood flow information of the target blood vessel to obtain the physiological parameters of the target blood vessel.

[0094] Specifically, the original medical image of the target blood vessel is obtained. Then, based on the original medical image of the target blood vessel, a physical model of the target blood vessel is established. The established physical model includes multiple physiological parameters. Since the blood flow information of the target blood vessel has been obtained based on the concentration characteristic information, the blood flow information of the target blood vessel can be used as a known physiological parameter to solve the physical model, thereby obtaining other unknown physiological parameters.

[0095] In one embodiment, the physical model of the target blood vessel includes: a three-dimensional model of the target blood vessel, a one-dimensional model of the target blood vessel, a reduced-order model of the target blood vessel, and a circuit-like model of the target blood vessel.

[0096] In one embodiment, a three-dimensional model of the target blood vessel is established based on the original medical image of the target blood vessel, and the physiological parameters of the target blood vessel are obtained based on the solution of the three-dimensional model.

[0097] Specifically, the three-dimensional model can be established by using methods such as threshold method, filtering method, region growing method, minimum cost path method, and skelton method.

[0098] When the threshold method is used to establish a three-dimensional model, a suitable image intensity threshold interval is selected according to the image intensity of the original medical image, and then the original medical image is segmented into foreground and background according to the selected image intensity threshold interval. The segmented foreground is the three-dimensional model of the corresponding blood vessel.

[0099] When using the filtering method to establish a three-dimensional model, a filter is used to process the original medical image. Depending on the filter used, different features in the image are enhanced to highlight some features in the image, thereby facilitating image segmentation. Filters include but are not limited to: median filter, mean filter, bilateral filter, Frangi filter, Gaussian filter, Laplace filter, Sobel filter, Gabor filter, Canny filter, etc. The gradient method is a type of filtering method. The original medical image is first converted into a gradient distribution image using a gradient filter, and then the high gradient is segmented using the threshold method to obtain the contour, and then the foreground and background are segmented according to the contour. The segmented foreground is the three-dimensional model of the corresponding blood vessel.

[0100] When using the region growing method to build a 3D model, a "seed" point is selected from the original medical image, that is, the point with the most certain characteristics, such as the point with the highest image intensity, the point with the largest gradient change, etc. Then the segmented graphics are grown around the "seed" point. The growth process is based on information such as image intensity value and texture. The final result of the growth is the 3D model of the corresponding blood vessel.

[0101] When the minimum cost path method is used to build a 3D model, the "cost value" corresponding to each pixel is obtained based on the image information. Then, the path with the minimum cost is found based on the "cost value". In medical imaging, the minimum cost path method is generally used to determine the center line of blood vessels or slender tissues. Taking the 3D modeling of cardiovascular system as an example, the minimum cost path method obtains the center line of cardiovascular system. Then, the outline of cardiovascular system is segmented along the center line, and finally the 3D model of blood vessels is obtained.

[0102] When using the skeleton method to build a 3D model, morphological operations are performed on the image, the image is thinned, the topological structure of the image is extracted, and the skeleton of the image is obtained. For blood vessels, skeletonization of the blood vessels can obtain the center line of the blood vessels. Taking the 3D modeling of cardiovascular as an example, the skeletonization method obtains the center line of the cardiovascular. Then the outline of the cardiovascular is segmented along the center line, and finally the 3D model of the blood vessel is obtained.

[0103] After obtaining the three-dimensional model of the target blood vessel, the calculation area is discretized using fluid dynamics simulation, and then the fluid control equation is solved using numerical methods. Finally, through continuous iterations, the physical values ​​in the calculation area, such as velocity and pressure, are obtained. The physiological parameters of the target blood vessel are then calculated based on these physical values.

[0104] In one embodiment, a machine learning model is used to segment the original medical image of the target blood vessel to obtain a three-dimensional model of the target blood vessel according to the segmentation result.

[0105] Figure 3A schematic diagram of a deep neural network trained for image segmentation according to an embodiment of the present application is shown.

[0106] In this embodiment, the original medical image of the sample blood vessel is input into the deep neural network to obtain the prediction result of the deep neural network for image segmentation, and then the prediction result is compared with the manual annotation and fed back to the deep neural network to update the deep neural network so that it evolves in the direction of reducing errors. This process is repeated until a deep neural network capable of accurately performing image segmentation is obtained.

[0107] In one embodiment, a one-dimensional model of the target blood vessel is established based on the original medical image of the target blood vessel, and the physiological parameters of the target blood vessel are obtained based on the solution of the one-dimensional model.

[0108] Specifically, a one-dimensional module can be regarded as a simplification of a three-dimensional model. For a blood vessel, the one-dimensional model is generally the centerline of the blood vessel. A one-dimensional model can be obtained by first obtaining a three-dimensional model and then simplifying the three-dimensional model, for example, extracting the centerline of the blood vessel from the three-dimensional model of the blood vessel; a one-dimensional model can also be directly extracted from an image, for example, directly extracting the centerline of the blood vessel from the original medical image through the minimum path cost method or the skeletonization method.

[0109] Similarly, fluid mechanics simulation can be used to solve a three-dimensional model, so can fluid mechanics simulation.

[0110] In one embodiment, a reduced-order model of the target blood vessel is established based on the original medical image of the target blood vessel, and the physiological parameters of the target blood vessel are obtained based on solving the reduced-order model.

[0111] Specifically, the reduced-order model is similar to a one-dimensional model and is also a simplification of a three-dimensional model. However, compared to a one-dimensional model, a reduced-order model retains more information. Common reduced-order models in vascular simulation include pressure drop models, 1D-0D models, etc. The pressure drop model uses a formula to express the pressure drop of a vascular segment or vascular bifurcation. The parameters in the pressure drop formula generally come from the geometric information of the blood vessel itself, such as the diameter and length of the blood vessel. The 1D-0D model combines the one-dimensional model and the zero-dimensional model (lumped model) together and obtains the result through coupling solution. The lumped model simplifies the vascular information into a combination of one or several lumped elements. The parameters of the lumped elements generally come from the geometric information of various parts of the patient's heart, such as coronary artery area, myocardial volume, etc.

[0112] Similar to the one-dimensional model, there are two ways to obtain a reduced-order model. One method is to first obtain a three-dimensional model, then simplify the three-dimensional model, such as obtaining the diameter and length of the blood vessel, and then obtain the parameters of the reduced-order model to establish a reduced-order model. The other method is to directly obtain the parameters required for the reduced-order model from the image and then establish the reduced-order model.

[0113] After obtaining the reduced-order model of the target blood vessel, the physical values ​​of the calculation area, such as velocity and pressure, are obtained by solving the reduced-order formula, and then the physiological parameters of the target blood vessel are calculated based on these obtained physical values.

[0114] In one embodiment, a circuit-like model of the target blood vessel is established based on the original medical image of the target blood vessel, and the physiological parameters of the target blood vessel are obtained based on the solution of the circuit-like model.

[0115] Specifically, there is a high degree of analogy between hydraulic problems and electrical problems, for example, water flow is similar to electric current, water pressure is similar to voltage, water flow resistance is similar to resistance, etc. Therefore, fluid mechanics problems can be converted into circuit problems, and such models are called circuit-like models. There are also two ways to establish circuit-like models. One is to first establish a three-dimensional model and then simplify it into a circuit-like model; the other is to directly obtain the parameters required for the circuit-like model from the image and then establish the circuit-like model.

[0116] After obtaining the circuit system corresponding to the target blood vessel, the blood flow value, blood pressure value or other parameters of the corresponding point in the blood vessel can be obtained by solving the circuit formula, and then the physiological parameters of the target blood vessel can be obtained on this basis.

[0117] In one embodiment, the first information including the original medical image of the sample blood vessel and the blood flow information of the sample blood vessel is used as the input of the preset machine learning model, and the machine learning model is trained with the physiological parameters of the sample blood vessel as the target output to obtain the trained machine learning model. The second information including the original medical image of the target blood vessel and the blood flow information of the target blood vessel is input into the trained machine learning model to obtain the physiological parameters of the target blood vessel.

[0118] Specifically, the first information at least includes the original medical image of the sample blood vessel and the blood flow information of the sample blood vessel. The first information is used as the input of the machine learning model, and the physiological parameters of the sample blood vessel are used as the target output of the machine learning model. The machine learning model is trained to map the first information to the physiological parameters of the sample blood vessel with an accuracy rate that meets the conditions, thereby obtaining a trained machine learning model.

[0119] After obtaining the second information (the second information at least includes the original medical image of the target blood vessel and the blood flow of the target blood vessel), the second information is input into the trained machine learning model to obtain the physiological parameters of the target blood vessel output by the machine learning model.

[0120] Figure 4 A schematic diagram of a deep neural network trained to obtain blood flow information according to an embodiment of the present application is shown.

[0121] In this embodiment, the original medical image of the sample blood vessel and the blood flow information Q of the sample blood vessel are input into the deep neural network to obtain the prediction result made by the deep neural network for the physiological parameters of the sample blood vessel, and then the prediction result is compared with the manual annotation, and fed back to the deep neural network to update the deep neural network so that it evolves in the direction of reducing errors. This process is repeated until a deep neural network that can accurately obtain blood flow information is obtained.

[0122] In one embodiment, the first information further includes geometric feature information of the sample blood vessel and lesion feature information of the sample blood vessel, and the second information further includes geometric feature information of the target blood vessel and lesion feature information of the target blood vessel.

[0123] Specifically, the geometric feature information of the blood vessel includes: blood vessel volume information, blood vessel diameter information, blood vessel length information, blood vessel area information, blood vessel bifurcation angle information, and blood vessel bifurcation position information, etc. The vascular lesion feature information includes: blood vessel stenosis rate information, blood vessel lesion type information, blood vessel lesion area information, blood vessel lesion size information, blood vessel lesion texture information, and blood vessel calcification score, etc.

[0124] The information composition structure of the second information is consistent with the information composition structure of the first information. If the first information is composed of the original medical image of the sample blood vessel, the blood flow information of the sample blood vessel, and the volume information of the sample blood vessel, then the second information is composed of the original medical image of the target blood vessel, the blood flow information of the target blood vessel, and the volume information of the target blood vessel. Similarly, the composition of other information will not be described in detail.

[0125] The advantage of this embodiment is that by adding geometric feature information and lesion feature information in addition to the original medical image and blood flow information, and then training and using the model, the accuracy of the vascular physiological parameters output by the model is improved.

[0126] In one embodiment, reference Figure 5 Schematic diagram of physiological parameter visualization result is shown. After obtaining the physiological parameters of the target blood vessel, the physiological parameters of the target blood vessel are projected onto the three-dimensional model of the target blood vessel for visualization, so as to intuitively display the physiological parameters of the target blood vessel. Specifically, different values ​​of physiological parameters can be displayed by different colors.

[0127] Figure 6 A framework diagram of a vascular physiological parameter acquisition process according to an embodiment of the present application is shown.

[0128] refer to Figure 6As shown, in this embodiment, starting from the original medical image, a model for the target blood vessel generation is obtained. There are two ways to do this process: first, based on the original medical image of the target blood vessel, a physical model of the target blood vessel is established; second, based on the original medical image of the sample blood vessel and the blood flow information of the sample blood vessel, a machine learning model is trained to accurately map the original medical image and blood flow information to the physiological parameters of the corresponding blood vessel.

[0129] In addition, starting from the curve of the contrast agent concentration changing with time, the blood flow information of the target blood vessel is obtained. Then, combined with the model generated for the target blood vessel, the physiological parameters of the target blood vessel are obtained. Depending on the way the model is obtained, there are also two ways of this process: First, if the physical model of the target blood vessel is obtained, the physical model of the target blood vessel is solved based on the blood flow information of the target blood vessel to obtain the physiological parameters of the target blood vessel; second, if the machine learning model is obtained, the original medical image of the target blood vessel and the blood flow information of the target blood vessel are input into the machine learning model to obtain the physiological parameters of the target blood vessel output by the machine learning model.

[0130] After the physiological parameters of the target blood vessel are obtained, the physiological parameters of the target blood vessel may optionally be projected onto a three-dimensional model of the target blood vessel, thereby visually displaying the physiological parameters of the target blood vessel.

[0131] Figure 7 A device for acquiring vascular physiological parameters according to an embodiment of the present application is shown, and the device includes:

[0132] A concentration feature acquisition module 210 is configured to acquire concentration feature information of the contrast agent at a preset position of the target blood vessel in a time dimension;

[0133] A blood flow acquisition module 220 is configured to acquire blood flow information of a target blood vessel based on concentration characteristic information;

[0134] A model acquisition module 230, configured to acquire a model generated for a target blood vessel;

[0135] The physiological parameter acquisition module 240 is configured to acquire the physiological parameters of the target blood vessel based on the model and the blood flow information of the target blood vessel.

[0136] In an exemplary embodiment of the present application, the concentration feature acquisition module is configured as follows:

[0137] Obtaining a curve of the concentration of the contrast agent changing with time;

[0138] Based on the change curve, the concentration characteristic information is obtained.

[0139] In an exemplary embodiment of the present application, the blood flow acquisition module is configured as follows:

[0140] Obtaining the volume of the functional tissue where the target blood vessel is located;

[0141] Based on the concentration characteristic information, the time difference between the contrast agent concentration nodes is obtained;

[0142] Based on the volume and time difference, the blood flow information of the target blood vessel is calculated.

[0143] In an exemplary embodiment of the present application, the blood flow acquisition module is configured as follows:

[0144] Based on the original medical image of the target blood vessel, the intraluminal attenuation gradient information of the target blood vessel is obtained;

[0145] Based on the concentration characteristic information, the time difference between the contrast agent concentration nodes is obtained;

[0146] Based on the intracavitary attenuation gradient information and the time difference, the blood flow information of the target blood vessel is calculated.

[0147] In an exemplary embodiment of the present application, the contrast agent concentration nodes include: a node where the contrast agent concentration is zero, a node where the contrast agent concentration peak is, a node where the contrast agent concentration mean is, and a node where the contrast agent concentration plateau is.

[0148] In an exemplary embodiment of the present application, the model acquisition module is configured to: establish a physical model of the target blood vessel based on the original medical image of the target blood vessel;

[0149] The physiological parameter acquisition module is configured to solve the physical model of the target blood vessel based on the blood flow information of the target blood vessel to obtain the physiological parameters of the target blood vessel.

[0150] In an exemplary embodiment of the present application, the physical model of the target blood vessel includes: a three-dimensional model of the target blood vessel, a one-dimensional model of the target blood vessel, a reduced-order model of the target blood vessel, and a circuit-like model of the target blood vessel.

[0151] In an exemplary embodiment of the present application, the model acquisition module is configured to: use the original medical image of the sample blood vessel and the first information of the blood flow information of the sample blood vessel as the input of the preset machine learning model, and train the machine learning model with the physiological parameters of the sample blood vessel as the target output to obtain a trained machine learning model;

[0152] The physiological parameter acquisition module is configured to input second information including the original medical image of the target blood vessel and the blood flow information of the target blood vessel into the trained machine learning model to obtain the physiological parameters of the target blood vessel.

[0153] In an exemplary embodiment of the present application, the first information further includes geometric feature information of the sample blood vessel and lesion feature information of the sample blood vessel, and the second information further includes geometric feature information of the target blood vessel and lesion feature information of the target blood vessel.

[0154] Reference below Figure 8 The electronic device 30 according to the embodiment of the present application is described. Figure 8 The electronic device 30 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0155] like Figure 8 As shown, the electronic device 30 is in the form of a general computing device. The components of the electronic device 30 may include but are not limited to: at least one processing unit 310, at least one storage unit 320, and a bus 330 connecting different system components (including the storage unit 320 and the processing unit 310).

[0156] The storage unit stores program codes, which can be executed by the processing unit 310, so that the processing unit 310 performs the steps according to various exemplary embodiments of the present invention described in the description of the exemplary method described above in this specification. For example, the processing unit 310 can perform the following steps: Figure 1 The steps shown in .

[0157] The storage unit 320 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 3201 and / or a cache memory unit 3202 , and may further include a read-only memory unit (ROM) 3203 .

[0158] The storage unit 320 may also include a program / utility 3204 having a set (at least one) of program modules 3205, such program modules 3205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0159] Bus 330 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0160] The electronic device 30 may also communicate with one or more external devices 400 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 30, and / or may communicate with any device that enables the electronic device 30 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 350. The input / output (I / O) interface 350 is connected to a display unit 340. Furthermore, the electronic device 30 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 360. As shown, the network adapter 360 communicates with other modules of the electronic device 30 via a bus 330. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 30, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0161] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0162] In an exemplary embodiment of the present application, a computer-readable storage medium is further provided, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a computer, the computer executes the method described in the above method embodiment.

[0163] According to one embodiment of the present application, a program product for implementing the method in the above method embodiment is also provided, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited to this. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0164] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0165] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0166] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0167] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as JAVA, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0168] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0169] In addition, although the steps of the method in the present application are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0170] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0171] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the appended claims.

Claims

1. A method for obtaining vascular physiological parameters, characterized in that: The method comprises: Acquire concentration characteristic information of contrast agent at at least two of the inlet, outlet and bifurcation of the target blood vessel in the time dimension; Based on the original medical image of the target blood vessel, acquiring intraluminal attenuation gradient information of the target blood vessel; Based on the concentration characteristic information, obtaining the time difference between contrast agent concentration nodes; Calculating the blood flow information of the target blood vessel based on the intracavitary attenuation gradient information and the time difference; Based on the original medical image of the target blood vessel, a physical model of the target blood vessel is established; the physical model of the target blood vessel includes a three-dimensional model of the target blood vessel, a one-dimensional model of the target blood vessel, a reduced-order model of the target blood vessel, and a circuit-like model of the target blood vessel; Based on the blood flow information of the target blood vessel, the physical model of the target blood vessel is solved to obtain the physiological parameters of the target blood vessel.

2. The method according to claim 1, characterized in that The step of obtaining the concentration characteristic information of the contrast agent at at least two of the inlet, outlet and bifurcation of the target blood vessel in the time dimension includes: Obtaining a curve of the concentration of the contrast agent changing with time; Based on the change curve, the concentration characteristic information is acquired.

3. The method according to claim 1, characterized in that The contrast agent concentration nodes include: a node where the contrast agent concentration is zero, a node where the contrast agent concentration peak is, a node where the contrast agent concentration mean is, and a node where the contrast agent concentration plateau is.

4. The method according to claim 1, characterized in that After calculating the blood flow information of the target blood vessel based on the intracavitary attenuation gradient information and the time difference, the method further includes: Taking the original medical image of the sample blood vessel and the first information of the blood flow information of the sample blood vessel as the input of the preset machine learning model, and taking the physiological parameters of the sample blood vessel as the target output to train the machine learning model, so as to obtain a trained machine learning model; The second information including the original medical image of the target blood vessel and the blood flow information of the target blood vessel is input into the trained machine learning model to obtain the physiological parameters of the target blood vessel.

5. The method according to claim 4, characterized in that The first information further includes geometric feature information of the sample blood vessel and lesion feature information of the sample blood vessel, and the second information further includes geometric feature information of the target blood vessel and lesion feature information of the target blood vessel.

6. A device for acquiring vascular physiological parameters, characterized in that: The device comprises: A concentration feature acquisition module, configured to acquire concentration feature information of the contrast agent at at least two of the inlet, outlet and bifurcation of the target blood vessel in a time dimension; a blood flow acquisition module configured to acquire intraluminal attenuation gradient information of the target blood vessel based on the original medical image of the target blood vessel; acquire the time difference between contrast agent concentration nodes based on the concentration characteristic information; and calculate the blood flow information of the target blood vessel based on the intraluminal attenuation gradient information and the time difference; a model acquisition module, configured to establish a physical model of the target blood vessel based on the original medical image of the target blood vessel; the physical model of the target blood vessel includes a three-dimensional model of the target blood vessel, a one-dimensional model of the target blood vessel, a reduced-order model of the target blood vessel, and a circuit-like model of the target blood vessel; The physiological parameter acquisition module is configured to solve the physical model of the target blood vessel based on the blood flow information of the target blood vessel to obtain the physiological parameters of the target blood vessel.

7. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enables the electronic device to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: Computer-readable instructions are stored thereon, and when the computer-readable instructions are executed by a processor of a computer, the computer is caused to execute the method according to any one of claims 1 to 5.