A blood vessel imaging method, device, storage medium and electronic device
Through the pre-trained vascular signal intensity enhancement network, the curve formula of vascular image signal changes with concentration is used to solve the problem of insufficient edge detection and noise suppression ability of traditional methods, and improve the quality of vascular imaging, which is especially suitable for patients with renal insufficiency. Imaging.
Patent Information
- Application Number
- CN202410161913.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-05
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-02-05
AI Technical Summary
The existing Hessian matrix vascular enhancement method based on traditional machine learning has weak edge detection capabilities and insufficient noise suppression capabilities in vascular imaging, which affects the imaging quality of vascular images, and is especially challenging when using high concentration contrast agents for patients with renal insufficiency.
The pre-trained vascular signal intensity enhancement network is generated by using the curve formula of the vascular image signal with concentration to generate a pre-trained vascular signal intensity enhancement network, which only enhances the vascular signal and suppresses noise, thereby improving the quality of vascular imaging.
The imaging quality of the vascular image is significantly enhanced, and the noise suppression ability is improved, so that high-quality vascular images can be generated at different contrast agent concentrations.
Smart Images

Figure CN118587304B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of smart medical treatment and computer technology, and in particular to a vascular imaging method, device, storage medium and electronic equipment. Background Art
[0002] MRA (Magnetic Resonance Angiography) is a non-invasive medical imaging technology used to observe and evaluate the structure and function of the human vascular system. Magnetic resonance imaging is performed by injecting special magnetic resonance contrast agents, which can clearly show the location, morphology and blood flow of blood vessels, thereby providing doctors with data reference. The concentration of contrast agents during imaging is directly related to the quality of vascular imaging. The higher the concentration, the higher the quality of vascular imaging. However, high-concentration contrast agents increase the metabolic burden on the kidneys of the target subject, especially for subjects with renal insufficiency, it is difficult to use high-concentration contrast agents for imaging.
[0003] In the related art, the implementation scheme is a Hessian matrix blood vessel enhancement method based on traditional machine learning. This method uses the Hessian matrix to extract the three-dimensional or two-dimensional tubular features in the image, and enhances its features to improve the signal strength of the blood vessels. Since the Hessian matrix uses multi-scale image features, the edge range detection ability is weak and it is difficult to accurately enhance the blood vessels. At the same time, its noise suppression ability is weak. When enhancing the blood vessels, the noise level will be enhanced, and the noise will affect the imaging quality of the blood vessel image. Summary of the invention
[0004] The embodiments of the present application provide a vascular imaging method, apparatus, storage medium, and electronic device. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.
[0005] In a first aspect, an embodiment of the present application provides a vascular imaging method, the method comprising:
[0006] Controlling the imaging device to scan the object to be detected injected with a preset contrast agent to obtain a blood vessel image of the object to be detected;
[0007] Inputting a preset contrast agent concentration value and a vascular image of a target object into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated based on a curve formula of vascular image signal variation with concentration, the curve formula is fitted based on multiple vascular registration images of each target object, and the multiple vascular registration images of each target object are generated based on vascular image registration of each target object when contrast agents of different concentrations are injected;
[0008] An enhanced blood vessel image corresponding to each first concentration contrast agent corresponding to the object to be detected is output, wherein the concentration value of each first concentration contrast agent increases in sequence and the difference between adjacent concentration values is equal to the preset gradient concentration.
[0009] Optionally, a pre-trained vascular signal intensity enhancement network is generated according to the following steps, including:
[0010] Scanning blood vessel images of each target object when injected with contrast agents of different concentrations by an imaging device to obtain an imaging sequence of each target object; the contrast agents of different concentrations are set according to a preset gradient concentration;
[0011] Acquire a blood vessel image corresponding to a minimum concentration of contrast agent from an imaging sequence of each target object to obtain a reference blood vessel image of each target object;
[0012] Acquire other vascular images except the reference vascular image of each target object from the imaging sequence of each target object to obtain a plurality of vascular images to be registered of each target object;
[0013] Registering each to-be-registered blood vessel image of each target object with the reference blood vessel image of each target object one by one to obtain a plurality of registered blood vessel images of each target object;
[0014] Fitting a curve formula of the variation of the vascular image signal with the concentration according to the multiple vascular registration images of each target object;
[0015] A pre-trained vascular signal intensity enhancement network is generated by training based on multiple vascular registration images of each target object and a curve formula.
[0016] Optionally, according to the multiple vascular registration images of each target object, a curve formula of fitting the vascular image signal with the concentration variation includes:
[0017] Extracting vascular tissue from each vascular registration image of each target object to obtain multiple vascular tissue templates of each target object;
[0018] Analyze the average signal intensity or enhanced intensity in the blood vessel area of each blood vessel tissue template of each target object to obtain the signal value corresponding to each blood vessel tissue template of each target object;
[0019] According to the signal value corresponding to each vascular tissue template of each target object and the actual contrast agent concentration corresponding to each vascular tissue template of each target object, the relationship between the contrast agent concentration and the signal value is constructed to obtain the curve formula of the vascular image signal changing with the concentration; wherein,
[0020] The actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered used when each vascular tissue template is generated.
[0021] Optionally, the curve formula of the vascular image signal changing with concentration is:
[0022] y=a×X 3 +b×X 2 +c×X+d;
[0023] Among them, y is the signal value, X is the contrast agent concentration, and a, b, c, and d are the formula coefficients of the curve formula respectively.
[0024] Optionally, model training is performed based on multiple vascular registration images of each target object and a curve formula to generate a pre-trained vascular signal intensity enhancement network, including:
[0025] The parameters of the preset neural network are replaced with the formula coefficients of the curve formula to obtain a vascular signal intensity enhancement network;
[0026] creating a sample sequence for each target object based on each blood vessel registered image of each target object;
[0027] Input the sample sequence of each target object into the vascular signal intensity enhancement network and output the model loss value;
[0028] When the loss value reaches the minimum, a pre-trained vascular signal intensity enhancement network is generated.
[0029] Optionally, a sample sequence for each target object is created based on each blood vessel registration image of each target object, including:
[0030] Acquire the actual contrast agent concentration corresponding to each vascular registration image of each target object, the actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered used when each vascular tissue template is generated;
[0031] Each blood vessel registered image of each target object is associated with an actual contrast agent concentration corresponding to each blood vessel registered image of each target object to obtain a sample sequence of each target object.
[0032] Optionally, the blood vessel signal intensity enhancement network includes an image processing module, a curve formula module, and a loss value calculation module;
[0033] The sample sequence of each target object is input into the vascular signal intensity enhancement network, and the model loss value is output, including:
[0034] Extracting features of each blood vessel registration image of each target object through an image processing module to obtain feature parameters corresponding to each blood vessel registration image of each target object;
[0035] Inputting the actual contrast agent concentration corresponding to each blood vessel registration image of each target object into the curve formula module for calculation, thereby obtaining the target signal value corresponding to each blood vessel registration image of each target object;
[0036] The characteristic parameters corresponding to each vascular registration image of each target object and the target signal value corresponding to each vascular registration image of each target object are input into the loss value calculation module for calculation to obtain the model loss value.
[0037] In a second aspect, an embodiment of the present application provides a vascular imaging device, the device comprising:
[0038] A blood vessel image acquisition module is used to control the imaging device to scan the object to be detected that has been injected with a preset contrast agent, so as to acquire a blood vessel image of the object to be detected;
[0039] A parameter input module is used to input a preset contrast agent concentration value and a vascular image of a target object to be detected into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated based on a curve formula of vascular image signal variation with concentration, the curve formula is fitted based on multiple vascular registration images of each target object, and the multiple vascular registration images of each target object are generated based on vascular image registration of each target object when contrast agents of different concentrations are injected;
[0040] The enhanced blood vessel image output module is used to output enhanced blood vessel images corresponding to each first concentration contrast agent corresponding to the object to be detected, the concentration value of each first concentration contrast agent increases in sequence and the difference between adjacent concentration values is equal to the preset gradient concentration.
[0041] In a third aspect, an embodiment of the present application provides a computer storage medium, which stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.
[0042] In a fourth aspect, an embodiment of the present application provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.
[0043] The technical solution provided by the embodiments of the present application may have the following beneficial effects:
[0044] In the embodiment of the present application, the concentration value of the preset contrast agent and the vascular image of the object to be detected are analyzed by a pre-trained vascular signal intensity enhancement network, and the corresponding enhanced vascular images when the object to be detected is injected with different concentrations of contrast agent can be predicted. Since the pre-trained vascular signal intensity enhancement network incorporates the curve formula of the vascular image signal changing with the concentration, the network only enhances the vascular signal, so that the ability to suppress noise is greatly enhanced, thereby improving the imaging quality of the vascular image.
[0045] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0047] Figure 1 is a flow chart of a vascular imaging method provided in an embodiment of the present application;
[0048] Figure 2 is a schematic diagram of vascular imaging provided in an embodiment of the present application;
[0049] Figure 3 It is a flowchart of a blood vessel signal strength enhancement network training method provided in an embodiment of the present application;
[0050] Figure 4 is a structural schematic diagram of a vascular imaging device provided by the present application;
[0051] Figure 5 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0052] The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them.
[0053] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are only examples of devices and methods consistent with some aspects of the present invention as detailed in the attached claims.
[0055] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In addition, in the description of the present invention, unless otherwise specified, "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previously associated objects are in an "or" relationship.
[0056] The present application provides a vascular imaging method, device, storage medium and electronic device to solve the problems existing in the above-mentioned related technical problems. In the technical solution provided by the present application, the present application analyzes the concentration value of the preset contrast agent and the vascular image of the object to be detected through a pre-trained vascular signal intensity enhancement network, and can predict the corresponding enhanced vascular images when the object to be detected is injected with different concentrations of contrast agent. Since the pre-trained vascular signal intensity enhancement network incorporates the curve formula of the vascular image signal changing with the concentration, the network only enhances the vascular signal, so that the ability to suppress noise is greatly enhanced, thereby improving the imaging quality of the vascular image. The following uses an exemplary embodiment for detailed description.
[0057] The following will be combined with the attached Figure 1 -Attached Figure 3 , the vascular imaging method provided by the embodiment of the present application is introduced in detail. The method can be implemented by relying on a computer program and can be run on a vascular imaging device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application.
[0058] See also Figure 1 , which is a flow chart of a vascular imaging method according to an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:
[0059] S101, controlling an imaging device to scan a subject to be detected that has been injected with a preset contrast agent to obtain a blood vessel image of the subject to be detected;
[0060] Among them, the imaging device can be an MRA (Magnetic Resonance Angiography) device, which is a medical imaging device that uses magnetic resonance imaging technology to display the vascular structure and hemodynamics of the human body; the signal emitted by the imaging device is a radio frequency signal emitted by an RF (Radio Frequency) coil, which can excite hydrogen nuclei in the human body to generate signals; the object to be detected is a subject who needs vascular imaging, such as a patient, and the preset contrast agent is a substance used for medical examination and diagnosis, and the preferred contrast agent is superparamagnetic nano iron oxide.
[0061] In an embodiment of the present application, before controlling the imaging device to scan the object to be detected that has been injected with a preset contrast agent to obtain a vascular image of the object to be detected, it is necessary to deploy a magnetic resonance coil at various parts of the object to be detected. The magnetic resonance coil is a key device for magnetic resonance imaging and is placed around the body of the object to be detected.
[0062] Specifically, the magnetic resonance coil in the embodiment of the present application is a cascaded wrapped magnetic resonance coil, including multiple sub-coils distributed on different parts of the target object, the multiple sub-coils are electrically connected in sequence, and the multiple sub-coils include at least a head wrapped head and neck coil, a shoulder and chest wrapped body coil, an abdomen and waist wrapped body coil, a thigh and knee joint wrapped body coil, and a calf and foot wrapped body coil.
[0063] In some embodiments, after the magnetic resonance coil is deployed on the subject to be detected, the computer system can control the imaging device to scan the subject to be detected injected with a preset contrast agent to obtain a vascular image of the subject to be detected.
[0064] It should be noted that in actual application scenarios, the preset contrast agent preferably is superparamagnetic nano-iron oxide with the smallest concentration value.
[0065] S102, inputting a preset contrast agent concentration value and a vascular image of the object to be detected into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated based on a curve formula of vascular image signal variation with concentration, the curve formula is fitted based on multiple vascular registration images of each target object, and the multiple vascular registration images of each target object are generated based on vascular image registration of each target object when injected with contrast agents of different concentrations;
[0066] Among them, the pre-trained blood vessel signal intensity enhancement network is a neural network that can use the blood vessel image scanned under the injection of low-concentration contrast agent to predict multiple blood vessel images corresponding to high-concentration contrast agents.
[0067] In an embodiment of the present application, a process of generating a pre-trained vascular signal intensity enhancement network includes: scanning, by an imaging device, a vascular image of each target object when injected with contrast agents of different concentrations to obtain an imaging sequence of each target object; the contrast agents of different concentrations are set according to a preset gradient concentration; from the imaging sequence of each target object, obtaining a vascular image corresponding to a contrast agent of a minimum concentration to obtain a reference vascular image of each target object; from the imaging sequence of each target object, obtaining other vascular images except the reference vascular image of each target object to obtain a plurality of vascular images to be registered for each target object; registering each vascular image to be registered for each target object with the reference vascular image of each target object one by one to obtain a plurality of vascular registered images of each target object; fitting a curve formula of vascular image signal changing with concentration according to the plurality of vascular registered images of each target object; and training according to the plurality of vascular registered images of each target object and the curve formula to generate a pre-trained vascular signal intensity enhancement network.
[0068] In some embodiments of the present application, after obtaining the vascular image of the object to be detected, the concentration value of the preset contrast agent can be obtained, and then the vascular image of the object to be detected and the concentration value of the preset contrast agent are associated into input data, and finally the input data is input into a pre-trained vascular signal intensity enhancement network for processing.
[0069] S103, outputting enhanced blood vessel images corresponding to the contrast agents of the first concentration corresponding to the object to be detected;
[0070] Among them, the concentration values of each first concentration contrast agent increase in sequence and the difference between adjacent concentration values is equal to the preset gradient concentration. The concentration values of each first concentration contrast agent are arranged in sequence according to the size of the concentration value. The concentration value of each first concentration contrast agent is based on the concentration value of the preset contrast agent and is generated in combination with the preset gradient concentration.
[0071] For example, the preset contrast agent concentration value is 30 mg / L, and the preset gradient concentration is 3 mg / L. At this time, the concentration values of the contrast agents of the first concentrations may be: 33 mg / L, 36 mg / L, and 39 mg / L.
[0072] In some embodiments of the present application, after being processed by a pre-trained vascular signal strength enhancement network, an enhanced vascular image corresponding to each first concentration of contrast agent corresponding to the object to be detected can be output; for example Figure 2 As shown, the left side is an enhanced blood vessel image, and the right side is a blood vessel image of the scanned object to be detected. For example, the input model is an MRA image when the concentration is C1, and the model outputs multiple MRA images when the high concentration contrast agent is C2, C3, C4, and C5.
[0073] In the embodiment of the present application, the concentration value of the preset contrast agent and the vascular image of the object to be detected are analyzed by a pre-trained vascular signal intensity enhancement network, and the corresponding enhanced vascular images when the object to be detected is injected with different concentrations of contrast agent can be predicted. Since the pre-trained vascular signal intensity enhancement network incorporates the curve formula of the vascular image signal changing with the concentration, the network only enhances the vascular signal, so that the ability to suppress noise is greatly enhanced, thereby improving the imaging quality of the vascular image.
[0074] See also Figure 3 , is a flow chart of a method for training a network to enhance blood vessel signal strength according to an embodiment of the present application. Figure 3 As shown, the method of the embodiment of the present application may include the following steps:
[0075] S201, scanning blood vessel images of each target object when injected with contrast agents of different concentrations by an imaging device to obtain an imaging sequence of each target object; the contrast agents of different concentrations are set according to a preset gradient concentration;
[0076] In some embodiments of the present application, Nc contrast agent concentration gradients can be set (the optimal value of Nc in this example is 5, but not limited to this), and each gradient concentration is C1, C2, C3, C4 and C5 respectively; then N healthy target objects are selected, and contrast agents with concentrations of C1, C2, C3, C4 and C5 are injected into each target object respectively, and finally the MRA imaging sequence is scanned in the imaging device D to obtain the imaging sequence of each target object; the images generated by all target objects when injected with contrast agents of different concentrations constitute an image data set Dg.
[0077] S202, acquiring a blood vessel image corresponding to a minimum concentration of contrast agent from an imaging sequence of each target object, to obtain a reference blood vessel image of each target object;
[0078] In some embodiments of the present application, an imaging sequence of each target object and images Ics generated by contrast agents of all concentrations of each target object are extracted from the image data set Dg; and then the image Ic1 generated at the minimum concentration C1 is used as the reference image.
[0079] S203, acquiring other vascular images except the reference vascular image of each target object from the imaging sequence of each target object, to obtain a plurality of vascular images to be registered of each target object;
[0080] In some embodiments of the present application, all images except the reference blood vessel image Ic1 of each target object in the imaging sequence Ics of each target object are registered to Ic1; the registration method may adopt the rigid registration method in the prior art.
[0081] S204, registering each to-be-registered blood vessel image of each target object with the reference blood vessel image of each target object one by one, to obtain a plurality of registered blood vessel images of each target object;
[0082] In the embodiment of the present application, after the registration is completed, multiple blood vessel registered images of each target object can be obtained.
[0083] S205, fitting a curve formula of the variation of the vascular image signal with the concentration according to the multiple vascular registration images of each target object;
[0084] In an embodiment of the present application, the process of fitting a curve formula of vascular image signal changing with concentration based on multiple vascular registration images of each target object specifically includes: extracting vascular tissue from each vascular registration image of each target object to obtain multiple vascular tissue templates of each target object; analyzing the average signal intensity or enhanced intensity in the vascular area of each vascular tissue template of each target object to obtain a signal value corresponding to each vascular tissue template of each target object; constructing a relationship between contrast agent concentration and signal value based on the signal value corresponding to each vascular tissue template of each target object and the actual contrast agent concentration corresponding to each vascular tissue template of each target object to obtain a curve formula of vascular image signal changing with concentration; wherein the actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered used when each vascular tissue template is generated.
[0085] Specifically, the curve formula of the vascular image signal changing with concentration is:
[0086] y=a×X 3 +b×X 2 +c×X+d; where y is the signal value, X is the contrast agent concentration, and a, b, c, and d are the formula coefficients of the curve formula respectively.
[0087] In some embodiments of the present application, vascular tissue is first extracted from multiple vascular registration images of each target object to obtain multiple vascular tissue templates of each target object, wherein the vascular tissue extraction method can use a manual outlining method or other automated or semi-automated extraction methods; then, the average signal intensity or enhanced intensity in the vascular area of each vascular tissue template can be extracted from the multiple vascular tissue templates of each target object to obtain a concentration image signal value Scs, and by fitting the curve relationship between the concentration value and the signal value, a curve formula for the change of vascular image signal with concentration can be established.
[0088] S206, training is performed according to the multiple vascular registration images and the curve formula of each target object to generate a pre-trained vascular signal intensity enhancement network.
[0089] In an embodiment of the present application, the process of generating a pre-trained vascular signal strength enhancement network by training based on multiple vascular registration images and a curve formula of each target object includes: replacing the parameters of the preset neural network with the formula coefficients of the curve formula to obtain the vascular signal strength enhancement network; creating a sample sequence of each target object according to each vascular registration image of each target object; inputting the sample sequence of each target object into the vascular signal strength enhancement network and outputting the model loss value; generating a pre-trained vascular signal strength enhancement network when the loss value reaches the minimum. The preset neural network is preferably created by an LSTM model.
[0090] In some embodiments of the present application, the coefficients a, b, c, and d in the curve formula of the change of vascular image signal with concentration can be used as the initial parameters of the vascular signal intensity enhancement network of the model. The model can be trained through multiple vascular images to be registered of each target object to find the optimal solution of the formula coefficients a, b, c, and d.
[0091] Specifically, the process of creating a sample sequence for each target object based on each vascular registration image of each target object includes: obtaining an actual contrast agent concentration corresponding to each vascular registration image of each target object, the actual contrast agent concentration corresponding to each vascular tissue template being the contrast agent concentration corresponding to the vascular image to be registered used when each vascular tissue template is generated; associating each vascular registration image of each target object with the actual contrast agent concentration corresponding to each vascular registration image of each target object, to obtain a sample sequence for each target object.
[0092] Specifically, the blood vessel signal intensity enhancement network includes an image processing module, a curve formula module and a loss value calculation module.
[0093] Specifically, the specific process of inputting the sample sequence of each target object into the vascular signal intensity enhancement network and outputting the model loss value includes: extracting the features of each vascular registration image of each target object through the image processing module to obtain the feature parameters corresponding to each vascular registration image of each target object; inputting the actual contrast agent concentration corresponding to each vascular registration image of each target object into the curve formula module for calculation to obtain the target signal value corresponding to each vascular registration image of each target object; inputting the feature parameters corresponding to each vascular registration image of each target object and the target signal value corresponding to each vascular registration image of each target object into the loss value calculation module for calculation to obtain the model loss value.
[0094] Specifically, the loss value calculation formula contained in the loss value calculation module is:
[0095]
[0096] Where W is the network weight parameter of the vascular signal intensity enhancement network, i, j, k are the step concentration levels, x, y are the pixel coordinates of image I, C is the concentration, and Loss is the loss value. The characteristic parameter corresponding to each blood vessel registration image of each target object, mu j The target signal value corresponding to each blood vessel of each target object is registered in the image.
[0097] In the embodiment of the present application, the concentration value of the preset contrast agent and the vascular image of the object to be detected are analyzed by a pre-trained vascular signal intensity enhancement network, and the corresponding enhanced vascular images when the object to be detected is injected with different concentrations of contrast agent can be predicted. Since the pre-trained vascular signal intensity enhancement network incorporates the curve formula of the vascular image signal changing with the concentration, the network only enhances the vascular signal, so that the ability to suppress noise is greatly enhanced, thereby improving the imaging quality of the vascular image.
[0098] The following are embodiments of the device of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments of the present invention, please refer to the method embodiments of the present invention.
[0099] See also Figure 4 , which shows a schematic diagram of the structure of a vascular imaging device provided by an exemplary embodiment of the present invention. The vascular imaging device can be implemented as all or part of an electronic device through software, hardware or a combination of both. The device 1 includes a vascular image acquisition module 10, a parameter input module 20, and an enhanced vascular image output module 30.
[0100] The blood vessel image acquisition module 10 is used to control the imaging device to scan the object to be detected that has been injected with a preset contrast agent to obtain a blood vessel image of the object to be detected;
[0101] The parameter input module 20 is used to input the preset contrast agent concentration value and the vascular image of the object to be detected into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated according to a curve formula of the vascular image signal changing with the concentration, and the curve formula is fitted according to multiple vascular registration images of each target object, and the multiple vascular registration images of each target object are generated according to the vascular image registration of each target object when the contrast agent of different concentrations is injected;
[0102] The enhanced blood vessel image output module 30 is used to output enhanced blood vessel images corresponding to each first concentration contrast agent corresponding to the object to be detected, wherein the concentration value of each first concentration contrast agent increases in sequence and the difference between adjacent concentration values is equal to a preset gradient concentration.
[0103] It should be noted that the vascular imaging device provided in the above embodiment only uses the division of the above functional modules as an example when executing the vascular imaging method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the vascular imaging device provided in the above embodiment and the vascular imaging method embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.
[0104] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0105] In the embodiment of the present application, the concentration value of the preset contrast agent and the vascular image of the object to be detected are analyzed by a pre-trained vascular signal intensity enhancement network, and the corresponding enhanced vascular images when the object to be detected is injected with different concentrations of contrast agent can be predicted. Since the pre-trained vascular signal intensity enhancement network incorporates the curve formula of the vascular image signal changing with the concentration, the network only enhances the vascular signal, so that the ability to suppress noise is greatly enhanced, thereby improving the imaging quality of the vascular image.
[0106] The present invention also provides a computer-readable medium on which program instructions are stored. When the program instructions are executed by a processor, the vascular imaging methods provided by the above-mentioned various method embodiments are implemented.
[0107] The present invention also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the vascular imaging method of each of the above method embodiments.
[0108] See also Figure 5 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 5 As shown, the electronic device 1000 may include: at least one processor 1001 , at least one network interface 1004 , a user interface 1003 , a memory 1005 , and at least one communication bus 1002 .
[0109] The communication bus 1002 is used to realize the connection and communication between these components.
[0110] The user interface 1003 may include a display screen (Display) and a camera (Camera), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
[0111] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).
[0112] Among them, the processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect various parts within the entire electronic device 1000, and executes various functions and processes data of the electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, and it can be implemented separately through a chip.
[0113] Among them, the memory 1005 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1005 may optionally be at least one storage device located away from the aforementioned processor 1001. As Figure 5 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a vascular imaging application.
[0114] exist Figure 5In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 1001 can be used to call the vascular imaging application stored in the memory 1005 and specifically perform the following operations:
[0115] Controlling the imaging device to scan the object to be detected injected with a preset contrast agent to obtain a blood vessel image of the object to be detected;
[0116] Inputting a preset contrast agent concentration value and a vascular image of a target object into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated based on a curve formula of vascular image signal variation with concentration, the curve formula is fitted based on multiple vascular registration images of each target object, and the multiple vascular registration images of each target object are generated based on vascular image registration of each target object when contrast agents of different concentrations are injected;
[0117] An enhanced blood vessel image corresponding to each first concentration contrast agent corresponding to the object to be detected is output, wherein the concentration value of each first concentration contrast agent increases in sequence and the difference between adjacent concentration values is equal to a preset gradient concentration.
[0118] In one embodiment, when the processor 1001 generates a pre-trained blood vessel signal strength enhancement network, the processor 1001 specifically performs the following operations:
[0119] Scanning blood vessel images of each target object when injected with contrast agents of different concentrations by an imaging device to obtain an imaging sequence of each target object; the contrast agents of different concentrations are set according to a preset gradient concentration;
[0120] Acquire a blood vessel image corresponding to a minimum concentration of contrast agent from an imaging sequence of each target object to obtain a reference blood vessel image of each target object;
[0121] Acquire other vascular images except the reference vascular image of each target object from the imaging sequence of each target object to obtain a plurality of vascular images to be registered of each target object;
[0122] Registering each to-be-registered blood vessel image of each target object with the reference blood vessel image of each target object one by one to obtain a plurality of registered blood vessel images of each target object;
[0123] Fitting a curve formula of the change of the vascular image signal with the concentration according to the multiple vascular registration images of each target object;
[0124] A pre-trained vascular signal intensity enhancement network is generated by training based on multiple vascular registration images of each target object and a curve formula.
[0125] In one embodiment, when the processor 1001 performs fitting of a curve formula of changes in vascular image signals with concentrations according to multiple vascular registration images of each target object, the following operations are specifically performed:
[0126] Extracting vascular tissue from each vascular registration image of each target object to obtain multiple vascular tissue templates of each target object;
[0127] Analyze the average signal intensity or enhanced intensity in the blood vessel area of each blood vessel tissue template of each target object to obtain the signal value corresponding to each blood vessel tissue template of each target object;
[0128] According to the signal value corresponding to each vascular tissue template of each target object and the actual contrast agent concentration corresponding to each vascular tissue template of each target object, the relationship between the contrast agent concentration and the signal value is constructed to obtain the curve formula of the vascular image signal changing with the concentration; wherein,
[0129] The actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered used when each vascular tissue template is generated.
[0130] In one embodiment, when the processor 1001 performs model training according to multiple vascular registration images of each target object and the curve formula to generate a pre-trained vascular signal intensity enhancement network, the following operations are specifically performed:
[0131] The parameters of the preset neural network are replaced with the formula coefficients of the curve formula to obtain a vascular signal intensity enhancement network;
[0132] creating a sample sequence for each target object based on each blood vessel registered image of each target object;
[0133] Input the sample sequence of each target object into the vascular signal intensity enhancement network and output the model loss value;
[0134] When the loss value reaches the minimum, a pre-trained vascular signal intensity enhancement network is generated.
[0135] In one embodiment, when the processor 1001 creates a sample sequence for each target object according to each blood vessel registration image of each target object, the processor 1001 specifically performs the following operations:
[0136] Acquire the actual contrast agent concentration corresponding to each vascular registration image of each target object, the actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered used when each vascular tissue template is generated;
[0137] Each blood vessel registered image of each target object is associated with an actual contrast agent concentration corresponding to each blood vessel registered image of each target object to obtain a sample sequence of each target object.
[0138] In one embodiment, when the processor 1001 inputs the sample sequence of each target object into the blood vessel signal strength enhancement network and outputs the model loss value, the processor 1001 specifically performs the following operations:
[0139] Extracting features of each blood vessel registration image of each target object through an image processing module to obtain feature parameters corresponding to each blood vessel registration image of each target object;
[0140] Inputting the actual contrast agent concentration corresponding to each blood vessel registration image of each target object into the curve formula module for calculation, thereby obtaining the target signal value corresponding to each blood vessel registration image of each target object;
[0141] The characteristic parameters corresponding to each vascular registration image of each target object and the target signal value corresponding to each vascular registration image of each target object are input into the loss value calculation module for calculation to obtain the model loss value.
[0142] In the embodiment of the present application, the concentration value of the preset contrast agent and the vascular image of the object to be detected are analyzed by a pre-trained vascular signal intensity enhancement network, and the corresponding enhanced vascular images when the object to be detected is injected with different concentrations of contrast agent can be predicted. Since the pre-trained vascular signal intensity enhancement network incorporates the curve formula of the vascular image signal changing with the concentration, the network only enhances the vascular signal, so that the ability to suppress noise is greatly enhanced, thereby improving the imaging quality of the vascular image.
[0143] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the vascular imaging program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes in the above-mentioned embodiments of the methods. The storage medium can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.
[0144] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.
Claims
1. A method for vascular imaging, It is characterized in that The method comprises: Controlling an imaging device to scan the object to be detected injected with a preset contrast agent to obtain a blood vessel image of the object to be detected; Inputting the preset contrast agent concentration value and the vascular image of the object to be detected into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated according to a curve formula of vascular image signal changing with concentration, the curve formula is fitted according to multiple vascular registration images of each target object, and the multiple vascular registration images of each target object are generated according to the vascular image registration of each target object when the contrast agent of different concentrations is injected; Outputting enhanced blood vessel images corresponding to the first concentration contrast agents corresponding to the object to be detected, wherein the concentration values of the first concentration contrast agents increase in sequence and the difference between adjacent concentration values is equal to a preset gradient concentration; The pre-trained vascular signal strength enhancement network is generated according to the following steps, including: Scanning blood vessel images of each target object when injected with contrast agents of different concentrations by an imaging device to obtain an imaging sequence of each target object; the contrast agents of different concentrations are set according to a preset gradient concentration; Acquire a blood vessel image corresponding to a contrast agent with a minimum concentration from an imaging sequence of each target object to obtain a reference blood vessel image of each target object; Acquire other vascular images except the reference vascular image of each target object from the imaging sequence of each target object to obtain a plurality of vascular images to be registered of each target object; Registering each to-be-registered blood vessel image of each target object with the reference blood vessel image of each target object one by one to obtain a plurality of registered blood vessel images of each target object; Fitting a curve formula of the variation of the vascular image signal with the concentration according to the multiple vascular registration images of each target object; Training is performed based on the multiple vascular registration images of each target object and the curve formula to generate a pre-trained vascular signal intensity enhancement network; wherein, The method of fitting the curve formula of the variation of the vascular image signal with the concentration according to the multiple vascular registration images of each target object includes: Extracting vascular tissue from each vascular registered image of each target object to obtain multiple vascular tissue templates of each target object; Analyzing the average signal intensity or enhanced intensity in the blood vessel area of each blood vessel tissue template of each target object to obtain a signal value corresponding to each blood vessel tissue template of each target object; According to the signal value corresponding to each vascular tissue template of each target object and the actual contrast agent concentration corresponding to each vascular tissue template of each target object, a relationship between the actual contrast agent concentration and the signal value is constructed to obtain a curve formula of the change of vascular image signal with concentration; wherein, The actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered and used when each vascular tissue template is generated.
2. The method according to claim 1, It is characterized in that The curve formula of the blood vessel image signal changing with concentration is: y=a×X 3 +b×X 2 +c×X+d; Among them, y is the signal value, X is the contrast agent concentration, and a, b, c, and d are the formula coefficients of the curve formula respectively.
3. The method according to claim 1, It is characterized in that The performing model training according to the multiple vascular registration images of each target object and the curve formula to generate a pre-trained vascular signal intensity enhancement network includes: Replacing the parameters of the preset neural network with the formula coefficients of the curve formula to obtain a blood vessel signal intensity enhancement network; Creating a sample sequence for each target object according to each blood vessel registered image of each target object; Inputting the sample sequence of each target object into the blood vessel signal strength enhancement network, and outputting a model loss value; When the loss value reaches the minimum, a pre-trained vascular signal intensity enhancement network is generated.
4. The method according to claim 3, It is characterized in that The step of creating a sample sequence for each target object according to each blood vessel registration image of each target object comprises: Acquire the actual contrast agent concentration corresponding to each vascular registration image of each target object, wherein the actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered used when generating each vascular tissue template; Each blood vessel registered image of each target object is associated with an actual contrast agent concentration corresponding to each blood vessel registered image of each target object to obtain a sample sequence of each target object.
5. The method according to claim 4, It is characterized in that The blood vessel signal strength enhancement network includes an image processing module, a curve formula module and a loss value calculation module; The step of inputting the sample sequence of each target object into the blood vessel signal strength enhancement network and outputting a model loss value comprises: Extracting features of each blood vessel registration image of each target object by the image processing module to obtain feature parameters corresponding to each blood vessel registration image of each target object; Inputting the actual contrast agent concentration corresponding to each blood vessel registered image of each target object into the curve formula module for calculation to obtain the target signal value corresponding to each blood vessel registered image of each target object; The characteristic parameters corresponding to each blood vessel registration image of each target object and the target signal value corresponding to each blood vessel registration image of each target object are input into the loss value calculation module for calculation to obtain the model loss value.
6. A vascular imaging device, It is characterized in that The device comprises: A blood vessel image acquisition module, used for controlling an imaging device to scan a subject to be detected injected with a preset contrast agent, so as to acquire a blood vessel image of the subject to be detected; A parameter input module, used for inputting the preset contrast agent concentration value and the vascular image of the object to be detected into a pre-trained vascular signal intensity enhancement network; the pre-trained vascular signal intensity enhancement network is generated according to a curve formula of vascular image signal changing with concentration, the curve formula is fitted according to a plurality of vascular registration images of each target object, and the plurality of vascular registration images of each target object are generated according to the vascular image registration of each target object when the contrast agent of different concentrations is injected; An enhanced blood vessel image output module, used to output enhanced blood vessel images corresponding to each first concentration contrast agent corresponding to the object to be detected, wherein the concentration value of each first concentration contrast agent increases in sequence and the difference between adjacent concentration values is equal to a preset gradient concentration; The pre-trained vascular signal strength enhancement network is generated according to the following steps, including: Scanning blood vessel images of each target object when injected with contrast agents of different concentrations by an imaging device to obtain an imaging sequence of each target object; the contrast agents of different concentrations are set according to a preset gradient concentration; Acquire a blood vessel image corresponding to a contrast agent with a minimum concentration from an imaging sequence of each target object to obtain a reference blood vessel image of each target object; Acquire other vascular images except the reference vascular image of each target object from the imaging sequence of each target object to obtain a plurality of vascular images to be registered of each target object; Registering each to-be-registered blood vessel image of each target object with the reference blood vessel image of each target object one by one to obtain a plurality of registered blood vessel images of each target object; Fitting a curve formula of the variation of the vascular image signal with the concentration according to the multiple vascular registration images of each target object; Training is performed based on the multiple vascular registration images of each target object and the curve formula to generate a pre-trained vascular signal intensity enhancement network; wherein, The method of fitting the curve formula of the variation of the vascular image signal with the concentration according to the multiple vascular registration images of each target object includes: Extracting vascular tissue from each vascular registered image of each target object to obtain multiple vascular tissue templates of each target object; Analyzing the average signal intensity or enhanced intensity in the blood vessel area of each blood vessel tissue template of each target object to obtain a signal value corresponding to each blood vessel tissue template of each target object; According to the signal value corresponding to each vascular tissue template of each target object and the actual contrast agent concentration corresponding to each vascular tissue template of each target object, a relationship between the actual contrast agent concentration and the signal value is constructed to obtain a curve formula of the change of vascular image signal with concentration; wherein, The actual contrast agent concentration corresponding to each vascular tissue template is the contrast agent concentration corresponding to the vascular image to be registered and used when each vascular tissue template is generated.
7. A computer storage medium, It is characterized in that The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 5.
8. An electronic device, It is characterized in that include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 5.
Citation Information
Patent Citations
Method and system for calculating quantitative parameter of magnetic resonance imaging
CN101912262A
Apparatus and system for calculating blood vessel blood flow parameters based on angiography
CN108550388A
Blood vessel physiological parameter acquisition method and device, electronic equipment and storage medium
CN114420286A
Contrast agent residual quantity analysis method and device, storage medium and terminal
CN115629348A
Method for co-registering and displaying multiple imaging modalities
US20190029624A1
