Multi-mode whole body blood vessel magnetic resonance imaging method and device

Abstract

The invention discloses a multi-mode whole-body blood vessel magnetic resonance imaging method and device, and the method comprises the steps: displaying prompt information for carrying out the cascade wrapping of a whole-body magnetic resonance coil and the input of a contrast agent for a target object, and enabling the contrast agent to be polysaccharide superparamagnetic iron oxide; setting a signal mark point on the constructed whole body positioning image and configuring a multi-modal imaging sequence; emitting a magnetic resonance radio frequency pulse, and receiving a coil electric signal transmitted by the whole body magnetic resonance coil; determining the signal intensity of each set signal mark point according to the coil electric signal; when the signal intensity is larger than a preset threshold value, whole-body blood vessel scanning is conducted on the target object through the multi-mode imaging sequence; and performing image reconstruction according to the scanned target image sequence to generate a whole body blood vessel image. By adopting the embodiment of the invention, one-stop imaging of large and small blood vessels of the whole body can be realized at one time by utilizing a scanning mode of cascading wrapping of the whole body magnetic resonance coil, so that the scanning time is greatly shortened, and the blood vessel imaging efficiency is improved.

Description

Technical Field

[0001] The present application relates to the fields of smart medical technology and computer technology, and in particular to a multimodal whole-body vascular magnetic resonance imaging method and device. Background Art

[0002] In the field of medical diagnosis, panvascular disease is a type of vascular system disease that affects multiple important organs throughout the body, including the heart, brain, kidneys, and limbs. These diseases cause changes in organ function due to factors such as ischemia or hemorrhage, seriously threatening the patient's life and health. To achieve accurate diagnosis of panvascular disease, complete and clear imaging of blood vessels throughout the body is required to comprehensively assess the scope and extent of vascular lesions. Therefore, the development of a technology that can achieve one-stop imaging of blood vessels throughout the body is of great clinical significance.

[0003] Among related technologies, magnetic resonance angiography (MRA) is an important method widely used in clinical practice for diagnosing vascular diseases. This technology increases the imaging device's sensitivity to vascular tissue by injecting a contrast agent into the subject, thereby achieving clear imaging of blood vessels. In existing technologies, superparamagnetic gadolinium-based contrast agents (GBCAs) are commonly used as contrast agents to image the blood vessels of a single or small number of organs.

[0004] However, GBCA is a contrast agent containing the rare earth element gadolinium, and its toxicity and side effects in the human body have always been a concern. High-dose use of GBCA may lead to nephrogenic systemic fibrosis (NSF) or heavy metal deposition in deep brain nuclei, posing potential health risks to patients. GBCA has a rapid clearance rate in the blood pool, resulting in a narrow imaging time window, which cannot meet the long scanning requirements required for systemic vascular imaging. Therefore, existing technologies make it difficult to complete systemic vascular imaging under single-dose administration conditions. Summary of the Invention

[0005] The present embodiments provide a method and apparatus for multimodal whole-body vascular magnetic resonance imaging. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is provided below. This summary is not intended to be a comprehensive review, identify key or important elements, or delineate the scope of protection for these embodiments. Its sole purpose is to present some concepts in a simplified form, serving as a prelude to the detailed description that follows.

[0006] In a first aspect, an embodiment of the present application provides a multimodal whole-body vascular magnetic resonance imaging method, the method comprising:

[0007] During the initialization of the magnetic resonance imaging device, prompt information is displayed, including a first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and a second prompt information for inputting a contrast agent, wherein the contrast agent is polysaccharide superparamagnetic iron oxide;

[0008] After initialization is completed, a whole-body scout image is constructed for determining the positions of target object markers, signal markers are set on the whole-body scout image, and a multimodal imaging sequence for magnetic resonance imaging is configured;

[0009] Controlling the radio frequency system to transmit magnetic resonance radio frequency pulses to the target subject and receiving in real time coil electrical signals transmitted from a cascaded whole-body magnetic resonance coil; wherein the cascaded whole-body magnetic resonance coil comprises multiple sub-coils covering various parts of the target subject, and the whole-body magnetic resonance coil is electrically connected to the magnetic resonance imaging device;

[0010] Determine the signal strength of each set signal mark point based on the coil electrical signal;

[0011] When the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, a whole-body vascular scan of the target object is performed through a multimodal imaging sequence to obtain a scanned target image sequence;

[0012] Image reconstruction is performed based on the target image sequence to generate a whole-body vascular image.

[0013] Optionally, the signal strength of each set signal marking point is determined based on the coil electrical signal, including:

[0014] detecting a resonance signal returned from the target object through a coil system and converting the resonance signal into a resonance electrical signal;

[0015] Perform signal feature fusion based on the resonance electrical signal and the coil electrical signal to generate signal strengths at multiple locations;

[0016] From the signal strengths of multiple locations, the signal strengths of the set signal marker points are filtered out.

[0017] Optionally, the coil electrical signal includes a feedback electrical signal of each sub-coil in the plurality of sub-coils; the plurality of sub-coils include at least a head and neck coil wrapped around the head, a body coil wrapped around the shoulders and chest, a body coil wrapped around the abdomen and waist, a body coil wrapped around the thighs and knee joints, and a body coil wrapped around the calves and feet;

[0018] The signal characteristics of the resonance electrical signal and the coil electrical signal are fused to generate the signal strength of multiple locations, including:

[0019] The feedback electrical signal of each sub-coil is amplified, filtered and A / D converted in sequence to obtain a position digital signal;

[0020] One-hot encoding is performed on the part digital signal to generate a feature vector of the part digital signal;

[0021] Using historical resonance electrical signals to pre-train the neural network to determine the target parameter values ​​of the neural network;

[0022] Extract the features of the resonance electric signal according to the target parameter value to generate a resonance feature vector;

[0023] Based on the eigenvector and resonance eigenvector of the digital signal of the part, the signal strength of multiple position points is constructed.

[0024] Optionally, based on the eigenvector and resonance eigenvector of the position digital signal, the signal strength of multiple locations is constructed, including:

[0025] A target matrix is ​​constructed according to the characteristic vector of the position digital signal and the connection order of multiple sub-coils;

[0026] Construct a connecting line between the first and last elements in the target matrix, and perform weighted summing and averaging of all elements on the connecting line to obtain the vector reference value;

[0027] Perform vector splicing processing based on the vector reference value, the eigenvector of the digital signal of each part, and the resonance eigenvector to obtain a splicing vector for each part;

[0028] The signal intensity of the splicing vector of each part is quantified to obtain the signal intensity of multiple position points.

[0029] Optionally, image reconstruction is performed based on the target image sequence to generate a whole-body vascular image, including:

[0030] Input the target image sequence into the pre-trained blood vessel extraction model;

[0031] Output the grayscale values ​​of blood vessels and human tissues corresponding to the target image sequence;

[0032] Blood vessel segmentation is performed based on the grayscale values ​​of blood vessels and human tissues to generate a whole-body blood vessel image.

[0033] Optionally, generate a pre-trained vessel extraction model by following these steps:

[0034] Acquire historical scan images pre-labeled with blood vessel grayscale values ​​and human tissue grayscale values;

[0035] Create a blood vessel extraction model;

[0036] Construct the objective loss function of the blood vessel extraction model;

[0037] The blood vessel extraction model is trained based on the historical scanned images and the target loss function to generate a pre-trained blood vessel extraction model. The target loss function is expressed as:

[0038] is the cross entropy loss function for separating blood vessels from human tissues, λh(i)||f i || 2 It represents the constraint loss on the length of the feature vector when the blood vessels are separated from the human tissue incorrectly, e is a natural number, w is the fully connected layer, T is the transposition process, and f i It represents the features extracted by the model for the input image, h(i) is an indicator function. When the separation between blood vessels and human tissue is incorrect, h(i) = 1, otherwise h(i) = 0; λ is the weight of the constraint;

[0039] It is the constraint loss on the length of the feature vector when the blood vessels are correctly separated from the human tissue, γ is the weight of the constraint, and ε is the number to prevent the denominator from being 0, which is set to a fixed value.

[0040] Optionally, the vessel extraction model is trained based on historical scanned images and a target loss function to generate a pre-trained vessel extraction model, including:

[0041] Integrate the objective loss function into the blood vessel extraction model;

[0042] Input the historical scanned image into the blood vessel extraction model that integrates the target loss function, and output the loss value of the model;

[0043] When the loss value reaches the minimum, a pre-trained blood vessel extraction model is generated; or when the loss value does not reach the minimum, the step of inputting the historical scan image into the blood vessel extraction model that integrates the target loss function is continued until the loss value reaches the minimum.

[0044] Optionally, constructing a whole-body locator image for determining the positions of target object marker points, and setting signal marker points on the whole-body locator image, including:

[0045] Performing an initial scan of the target subject using a magnetic resonance imaging device to obtain a positioning image containing the entire body's anatomical structures;

[0046] Performing image registration on the whole-body positioning image and a plurality of pre-generated magnetic resonance three-dimensional template images to obtain a plurality of registration results;

[0047] For each registration result, the similarity between the whole-body positioning image and each magnetic resonance 3D template image is calculated;

[0048] The 3D magnetic resonance template image with the greatest similarity is used as the target template image;

[0049] Setting signal markers in the whole-body positioning image based on pre-marked locations in the target template image; or

[0050] Display positioning images;

[0051] Receive a parameter configuration request from the user for the client, where the parameter configuration request carries the location of the mark point manually selected by the user;

[0052] Based on the manually selected marker positions by the user, signal markers are set in the whole body positioning image.

[0053] Optionally, the multimodal imaging sequence includes T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging;

[0054] Perform a full-body vascular scan of the target subject using multimodal imaging sequences, including:

[0055] The scanning parameters corresponding to T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging are switched in sequence to perform a whole-body vascular scan on the target object.

[0056] In a second aspect, an embodiment of the present application provides a multimodal whole-body vascular magnetic resonance imaging device, the device comprising:

[0057] An information display module is used to display prompt information during the initialization process of the magnetic resonance imaging device, the prompt information including a first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and a second prompt information for inputting a contrast agent, wherein the contrast agent is polysaccharide superparamagnetic iron oxide;

[0058] A whole-body locator image construction module is used to construct a whole-body locator image for determining the position of target object marker points after initialization, set signal marker points on the whole-body locator image, and configure a multimodal imaging sequence for magnetic resonance imaging;

[0059] an MRI radio frequency pulse transmission module, configured to control the radio frequency system to transmit MRI radio frequency pulses to the target subject and to receive in real time coil electrical signals transmitted from a cascaded, whole-body MRI coil comprising multiple sub-coils covering various parts of the target subject, the whole-body MRI coil being electrically connected to the MRI imaging device;

[0060] A signal strength determination module is used to determine the signal strength of each set signal marking point based on the coil electrical signal;

[0061] A whole-body vascular scanning module is used to perform a whole-body vascular scan on the target object through a multimodal imaging sequence when the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, thereby obtaining a scanned target image sequence;

[0062] The whole-body blood vessel generation module is used to perform image reconstruction based on the target image sequence to generate a whole-body blood vessel image.

[0063] The technical solutions provided by the embodiments of the present application may have the following beneficial effects:

[0064] In an embodiment of the present application, on the one hand, after initialization is completed, a whole-body positioning image is constructed and signal markers are set, and a multimodal imaging sequence is configured. The cascaded wrapped magnetic resonance coil can cover all parts of the body of the target object, and the multimodal imaging sequence can provide richer vascular information, including vascular wall structure, hemodynamics, and pathological characteristics. Therefore, the multi-position coil cascade method is combined with the multimodal imaging sequence to achieve one-stop imaging of large and small blood vessels (including arteries and veins) throughout the body. This method greatly reduces the scanning time and improves imaging efficiency. After using polysaccharide superparamagnetic iron oxide contrast agent at the same time, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the image are significantly improved, and the image quality score is also greatly improved. High-quality images can more clearly show the edges of blood vessels and anatomical structures. On the other hand, the contrast agent is polysaccharide superparamagnetic iron oxide, which is modified by polysaccharides to improve biocompatibility, reduce the risk of cytotoxicity, hemolytic reaction, and complement activation-related pseudo-allergic reaction (CARPA), and avoid potential health risks to patients.

[0065] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0067] Figure 1 This is a schematic diagram of a method flow of a multimodal whole-body vascular magnetic resonance imaging method provided in an embodiment of the present application;

[0068] Figure 2 This is a schematic diagram of a whole-body positioning image provided in an embodiment of the present application;

[0069] Figure 3 This is a schematic diagram of a whole-body vascular image provided by an embodiment of the present application;

[0070] Figure 4 This is a schematic flowchart of a multimodal whole-body vascular magnetic resonance imaging process provided by an embodiment of the present application;

[0071] Figure 5 1 is a flow chart of a model training method for a blood vessel extraction model provided in an embodiment of the present application;

[0072] Figure 6 1 is a schematic structural diagram of a multimodal whole-body vascular magnetic resonance imaging device provided in an embodiment of the present application;

[0073] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0074] The following description and the drawings sufficiently illustrate specific embodiments of the application to enable those skilled in the art to practice them.

[0075] It should be clear that the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0076] When the following description refers to the accompanying 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 application. Instead, they are merely examples of devices and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0077] In the description of this application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances. In addition, in the description of this application, unless otherwise specified, "multiple" 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 previous and subsequent associated objects are in an "or" relationship.

[0078] The present application provides a multimodal whole-body vascular magnetic resonance imaging method and apparatus to solve the problems existing in the above-mentioned related technical problems. In an embodiment of the present application, on the one hand, after the initialization is completed, a whole-body positioning image is constructed and signal markers are set, and a multimodal imaging sequence is configured. The cascaded wrapped magnetic resonance coil can cover all parts of the body of the target object, and the multimodal imaging sequence can provide richer vascular information, including vascular wall structure, hemodynamics and pathological characteristics. Therefore, the multi-position coil cascade method combined with the multimodal imaging sequence can achieve one-stop imaging of large and small blood vessels (including arteries and veins) throughout the body. This method greatly reduces the scanning time and improves imaging efficiency. After using the polysaccharide superparamagnetic iron oxide contrast agent at the same time, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the image are significantly improved, and the image quality score is also greatly improved. The high-quality image can more clearly display the vascular edges and anatomical structures. On the other hand, the contrast agent is polysaccharide superparamagnetic iron oxide. Polysaccharide superparamagnetic iron oxide is modified with polysaccharides to improve biocompatibility, reduce the risk of cytotoxicity, hemolytic reaction and complement activation-related pseudoallergic reaction (CARPA), and avoid potential health risks to patients. The following exemplary embodiments are used to explain in detail.

[0079] The following will be combined with the Figure 1 -Attached Figure 5 This article details the multimodal whole-body vascular magnetic resonance imaging method provided in the embodiments of this application. This method can be implemented using a computer program and run on a multimodal whole-body vascular magnetic resonance imaging device based on the von Neumann architecture. This computer program can be integrated into an application or run as a standalone tool application.

[0080] See Figure 1 , provides a flow chart of a multimodal whole-body vascular magnetic resonance imaging method for an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:

[0081] S101, during the initialization process of the magnetic resonance imaging device, displaying prompt information, including first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and second prompt information for inputting a contrast agent, where the contrast agent is polysaccharide superparamagnetic iron oxide;

[0082] Among them, magnetic resonance imaging equipment initialization refers to a series of system self-tests, parameter settings, and equipment calibration operations performed before the magnetic resonance imaging (MRI) equipment is started and ready for scanning, to ensure that the equipment is in the best working condition. The cascade wrapping of the whole-body magnetic resonance coil is to cover multiple magnetic resonance coils on various parts of the subject's body (such as the head, chest, abdomen, limbs, etc.), and connect these coils in series through connecting wires to achieve continuous acquisition of whole-body signals. Polysaccharide superparamagnetic iron oxide contrast agent is a new type of magnetic resonance imaging contrast agent, which is composed of polysaccharide-modified superparamagnetic iron oxide nanoparticles. It has the advantages of high biocompatibility, low toxicity, and long in vivo half-life, and is suitable for whole-body vascular imaging. Prompt information is the guidance information displayed by the system to the operator during the equipment initialization process, which is used to prompt the operator to perform necessary preparations, such as coil wrapping and contrast agent input.

[0083] In some embodiments of the present application, the device first performs initialization operations, including system self-test, magnetic resonance coil calibration, and imaging parameter setting. During the magnetic resonance imaging device initialization process, the device displays prompts to guide the operator through the cascade wrapping of the whole-body magnetic resonance coil and the administration of contrast agent.

[0084] Specifically, the operator follows the device's prompts and wraps the subject with a cascade of full-body MRI coils. This involves using multiple sub-coils to cover the subject's head, neck, chest, abdomen, thighs, calves, and other areas. These sub-coils are connected in series via cables to ensure continuous signal transmission. All coils are then electrically connected to the MRI equipment to enable real-time signal acquisition.

[0085] Specifically, the device prompts the operator to prepare and administer a polysaccharide superparamagnetic iron oxide contrast agent. The device prompts the operator to prepare a polysaccharide superparamagnetic iron oxide contrast agent (iron agent) and configure it according to preset parameters (e.g., concentration 5 mg / mL, injection volume 3 mg / kg, injection rate 0.1 mL / s). The contrast agent is injected into the subject's vein using a manual push or pressure syringe.

[0086] Specifically, polysaccharide superparamagnetic iron oxide is a nano-sized iron oxide particle coated with a polysaccharide component (i.e., polysaccharide superparamagnetic iron oxide nanoparticles). The iron oxide core is trivalent iron with five lone pairs of electrons outside the core. Polysaccharide superparamagnetic iron oxide nanoparticles are non-stoichiometric iron ore (superparamagnetic iron oxide nanoparticles) coated with a layer of polyglucose sorbitol carboxymethylether (polyglucose sorbitol carboxymethylether), and the entire colloid particle size is 17-31 nm. Optionally, the polysaccharide superparamagnetic iron oxide nanoparticles are prepared by the following methods: chemical coprecipitation, laser treatment, microemulsion, ultrasonic emulsification, gamma ray radiation, or high-temperature decomposition. The polysaccharide superparamagnetic iron oxide nanoparticles are injected at a concentration of 510 mg Fe / 17 mL / vial. The injection excipient is mannitol, there is no preservative, and the pH is 6-8. Each 17 mL vial contains 510 mg of elemental iron. Each mL of sterile colloidal solution contains 30 mg of elemental iron, 30 mg of polydextrose sorbitan carboxymethyl ether, and 44 mg of mannitol. The injection is isotonic, with an osmotic pressure of 270 to 330 mOsm / kg.

[0087] The recommended dose of polysaccharide superparamagnetic iron oxide nanoparticles is 0.56 mg / kg to 4 mg / kg (calculated as iron, e.g., 1 mg / kg, 2 mg / kg, 2.5 mg / kg, 3 mg / kg, 4 mg / kg, or a range consisting of two of the aforementioned values). Alternatively, the polysaccharide superparamagnetic iron oxide nanoparticles can be diluted in 100 mL of a 5% glucose solution and injected over 30 minutes at a rate of 2 to 4 mL / min. Enhanced imaging can be acquired immediately after injection of the polysaccharide superparamagnetic iron oxide nanoparticles and beginning 0.5 hours after injection.

[0088] During MRI, the application of an external magnetic field (B0) magnetizes the iron oxide core, generating a magnetization intensity M with a vector direction aligned with the external magnetic field (B0), resulting in an increase in the effective magnetic field. During human MRI, applying radiofrequency pulses in the opposite direction of the main magnetic field (B0) shortens the longitudinal relaxation time (T1) and transverse relaxation time (T2) of the region containing the polysaccharide superparamagnetic iron oxide due to proton-proton and proton-electron interactions, thereby enhancing the corresponding weighted image signal on T1 / T2-weighted imaging.

[0089] Polysaccharides are sugars that can produce more than 10 monosaccharide molecules when one polysaccharide molecule is hydrolyzed. Superparamagnetic iron oxide is a microparticle-based contrast agent used for angiography and intravascular administration. Polysaccharide superparamagnetic iron oxide is a compound formed from polysaccharides and superparamagnetic iron oxide.

[0090] S102, after initialization is completed, constructing a whole-body scout image for determining the positions of target object markers, setting signal markers on the whole-body scout image, and configuring a multimodal imaging sequence for magnetic resonance imaging;

[0091] The target object refers to the subject undergoing an MRI scan. A whole-body locator image is a low-resolution preliminary image obtained before a whole-body scan. It is used to determine the subject's anatomical structure and scanning range, and to help locate the region of interest for subsequent imaging. Marker points are specific locations set on the whole-body locator image to monitor contrast agent flow or signal changes to trigger subsequent imaging sequences. Signal markers are specific marker points used to monitor changes in signal intensity in real time. When the signal intensity reaches a preset threshold, the imaging device is triggered to start scanning.

[0092] In some embodiments of the present application, the specific process of constructing a whole-body locator image for determining the locations of target object marker points and setting signal marker points on the whole-body locator image includes: performing an initial scan of the target object using a magnetic resonance imaging device to obtain a locator image containing the entire body's anatomical structures; performing image registration on the whole-body locator image with multiple pre-generated magnetic resonance 3D template images to obtain multiple registration results; for each registration result, calculating the similarity between the whole-body locator image and each magnetic resonance 3D template image; using the magnetic resonance 3D template image with the greatest similarity as the target template image; and setting signal marker points in the whole-body locator image based on the pre-marked locations in the target template image. This automated configuration method can improve the efficiency of marker point setting.

[0093] In other embodiments of the present application, the specific process of constructing a full-body positioning image for determining the locations of target object markers and setting signal markers on the full-body positioning image includes: displaying the positioning image; receiving a parameter configuration request input by a user for a client, the parameter configuration request carrying the user's manually selected marker locations; and setting signal markers in the full-body positioning image based on the user's manually selected marker locations. This manual configuration method can improve configuration accuracy.

[0094] In some embodiments, the device automatically starts a whole-body scanning program, uses a low-resolution imaging sequence to quickly scan the target object, and generates a whole-body positioning image. This image is mainly used to determine the anatomical structure of the subject. On the generated whole-body positioning image, the operator or the device automatically selects one or more marker positions. These markers are usually selected at prominent locations on the blood vessels, such as the aortic arch, to monitor the flow of contrast agents. The device configures a multimodal imaging sequence based on preset imaging requirements. These sequences may include: T1-weighted imaging (T1WI): used to display the structure of the blood vessel wall. T2-weighted imaging (T2WI): used to detect edema around blood vessels. Among them, whole-body positioning images, for example Figure 2shown.

[0095] S103, controlling the radio frequency system to transmit magnetic resonance radio frequency pulses to the target subject and receiving in real time coil electrical signals transmitted from a cascaded whole-body magnetic resonance coil; wherein the cascaded whole-body magnetic resonance coil comprises multiple sub-coils covering various parts of the target subject, and the whole-body magnetic resonance coil is electrically connected to the magnetic resonance imaging device;

[0096] Among them, the radio frequency system is a component in the magnetic resonance imaging (MRI) device used to transmit and receive radio frequency signals, including the radio frequency transmission system and the radio frequency control system. The control radio frequency system is a device that transmits magnetic resonance radio frequency pulses during the imaging process. The coil electrical signal includes the electrical signal fed back by each sub-coil in multiple sub-coils. The magnetic resonance radio frequency pulse is an electromagnetic pulse emitted by the radio frequency system, which is used to excite hydrogen nuclei (protons) in human tissue into an excited state. The frequency of this pulse matches the Larmor frequency of protons, thereby achieving the magnetic resonance phenomenon.

[0097] In some embodiments of the present application, a radio frequency system is controlled to transmit magnetic resonance radio frequency pulses to a target object to excite the spin-non-zero iron nuclei of the target to be imaged, thereby generating resonance.

[0098] In the embodiment of the present application, before controlling the radio frequency system to transmit magnetic resonance radio frequency pulses to the imaging target injected with contrast agent, the whole body of the target object is first wrapped with a magnetic resonance coil, specifically the head is wrapped with a head and neck coil C. head-neck , shoulder and chest wrap body coil C shoulder-chest , abdomen and waist wrapped body coil C stomach-waist , thigh and knee joint wrapped body coil C thigh-knee and calf and foot wrap coil C shank-foot The above coil types and positions are examples; then use the connecting coil C link Connect C head-neck and C shoulder-chest , connect C shoulder-chest and C stomach-waist , connect C stomach-waist and C thigh-knee , connect C thigh-knee and C shank-foot The above is an example of the connection sequence between the various parts, but is not limited to the above connection sequence.

[0099] Furthermore, after the connection is completed, the imaging device D is used to scan the human body and generate an image, wherein the best imaging device for D is MR, but not limited to this device; first, a monitoring mark point L is set in the device D a , where L aThe optimal position is the aortic arch, but is not limited to the optimal position; then, a multimodal imaging sequence S is set in the device D.

[0100] Furthermore, after setting the monitoring mark point, the concentration of polysaccharide superparamagnetic iron oxide C is set according to the characteristics of polysaccharide superparamagnetic iron oxide and relevant usage regulations. a and injection volume Q a , and injection speed V a , where C a The optimal value is 5 mg / mL, Q a The optimal value is 3mg / kg, V a The optimal value of is 0.1 mL / s, but is not limited to the optimal value.

[0101] In a possible implementation, when performing vascular imaging, the injection speed V a The target to be imaged is injected with polysaccharide superparamagnetic iron oxide. The radiofrequency system then transmits magnetic resonance radiofrequency pulses of a specific frequency and timing to the target according to preset imaging sequence parameters. The frequency of these pulses matches the Larmor frequency of hydrogen nuclei in human tissue, exciting them into an excited state. After the radiofrequency pulses excite the hydrogen nuclei in the target, they return to equilibrium and release energy. This energy is detected as an electromagnetic signal by the magnetic resonance coil covering the target's entire body and converted into an electrical signal. The device's receiving system collects these coil electrical signals in real time.

[0102] S104, determining the signal strength of each set signal marking point based on the coil electrical signal;

[0103] In some embodiments of the present application, the specific process of determining the signal strength of each set signal marking point based on the coil electrical signal includes: detecting the resonance signal returned from the target object through the coil system, and converting the resonance signal into a resonance electrical signal; performing signal feature fusion based on the resonance electrical signal and the coil electrical signal to generate the signal strength of multiple position points; and screening out the signal strength of each set signal marking point from the signal strength of the multiple position points.

[0104] Among them, the coil electrical signal includes the feedback electrical signal of each sub-coil in the multiple sub-coils; the multiple sub-coils include at least a head and neck coil wrapped around the head, a body coil wrapped around the shoulders and chest, a body coil wrapped around the abdomen and waist, a body coil wrapped around the thigh and knee joint, and a body coil wrapped around the calf and foot.

[0105] In some embodiments of the present application, the specific process of performing signal feature fusion based on the resonance electric signal and the coil electric signal to generate the signal strength of multiple position points includes: amplifying, filtering and A / D converting the feedback electric signal of each sub-coil in turn to obtain a position digital signal; one-hot encoding the position digital signal to generate a feature vector of the position digital signal; using historical resonance electric signals to pre-train the neural network to determine the target parameter value of the neural network; extracting features of the resonance electric signal according to the target parameter value to generate a resonance feature vector; and constructing the signal strength of multiple position points based on the feature vector of the position digital signal and the resonance feature vector.

[0106] In some embodiments of the present application, the specific process of constructing the signal strength of multiple position points based on the eigenvector and resonance eigenvector of the part digital signal includes: constructing a target matrix according to the eigenvector of the part digital signal and the connection order of multiple sub-coils; constructing a connecting line between the first element and the last element in the target matrix, and taking a weighted sum and average of all elements on the connecting line to obtain a vector reference value; performing vector splicing processing according to the vector reference value, the eigenvector of the digital signal of each part, and the resonance eigenvector to obtain a splicing vector of each part; performing signal strength quantization processing on the splicing vector of each part to obtain the signal strength of multiple position points.

[0107] S105, when the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, performing a whole-body vascular scan on the target object through a multimodal imaging sequence to obtain a scanned target image sequence;

[0108] Among them, the multimodal imaging sequence includes T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging.

[0109] In some embodiments of the present application, the specific process of performing a whole-body vascular scan on a target object through a multimodal imaging sequence includes: switching the scanning parameters corresponding to T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging in sequence to perform a whole-body vascular scan on the target object.

[0110] Among them, when the signal intensity of each signal marker point is greater than or equal to the preset intensity threshold, it means that the best time for image scanning has arrived. At this time, it is necessary to start the image scanning service to perform a whole-body vascular scan on the target object to obtain a scanned target image sequence.

[0111] In some other embodiments provided in the present application, when the signal strength of each signal marker point is less than a preset intensity threshold, it indicates that this is not the best time for image scanning, and the process can continue to return to the step of receiving in real time the coil electrical signal transmitted from the cascaded whole-body magnetic resonance coil.

[0112] For example, monitoring point L a The signal strength changes when the mark point L a The signal strength I a Exceeding the threshold T a The imaging device starts scanning when T a The optimal value of is 10, but it is not limited to this value.

[0113] S106: Perform image reconstruction based on the target image sequence to generate a whole-body vascular image.

[0114] In some embodiments of the present application, the specific process of performing image reconstruction based on the target image sequence to generate a whole-body vascular image includes: inputting the target image sequence into a pre-trained vascular extraction model; outputting the vascular grayscale values ​​and human tissue grayscale values ​​corresponding to the target image sequence; performing vascular segmentation based on the vascular grayscale values ​​and human tissue grayscale values ​​to generate a whole-body vascular image. Whole-body vascular image V vessel ,like Figure 3 shown.

[0115] In an embodiment of the present application, the specific process of generating a pre-trained blood vessel extraction model includes: obtaining historical scan images that are pre-labeled with blood vessel grayscale values ​​and human tissue grayscale values; creating a blood vessel extraction model; constructing a target loss function for the blood vessel extraction model; and training the blood vessel extraction model based on the historical scan images and the target loss function to generate a pre-trained blood vessel extraction model.

[0116] Specifically, the objective loss function expression is:

[0117] is the cross entropy loss function for separating blood vessels from human tissues, λh(i)||f i || 2 It represents the constraint loss on the length of the feature vector when the blood vessels are separated from the human tissue incorrectly, e is a natural number, w is the fully connected layer, T is the transposition process, and f i It represents the features extracted by the model for the input image, h(i) is an indicator function. When the separation between blood vessels and human tissue is incorrect, h(i) = 1, otherwise h(i) = 0; λ is the weight of the constraint;

[0118] It is the constraint loss on the length of the feature vector when the blood vessels are correctly separated from the human tissue, γ is the weight of the constraint, and ε is the number to prevent the denominator from being 0, which is set to a fixed value.

[0119] Specifically, the blood vessel extraction model is trained according to the historical scanned images and the target loss function, and the specific process of generating a pre-trained blood vessel extraction model includes: integrating the target loss function into the blood vessel extraction model; inputting the historical scanned images into the blood vessel extraction model integrated with the target loss function, and outputting the loss value of the model; when the loss value reaches the minimum, generating the pre-trained blood vessel extraction model; or when the loss value does not reach the minimum, continuing to execute the step of inputting the historical scanned images into the blood vessel extraction model integrated with the target loss function until the loss value reaches the minimum.

[0120] In one possible implementation, for example Figure 4 As shown, Figure 4 This is a schematic flowchart of the process of multimodal whole-body vascular magnetic resonance imaging provided by this application. First, the subject is wrapped with a cascade magnetic resonance coil to cover all parts of the subject's body. Then, the multimodal imaging sequence and signal marker points are set on the imaging device. Secondly, a polysaccharide superparamagnetic iron oxide contrast agent is injected. Finally, a vascular imaging scan is performed to obtain a one-stop full vascular image of the subject.

[0121] In an embodiment of the present application, on the one hand, after initialization is completed, a whole-body positioning image is constructed and signal markers are set, and a multimodal imaging sequence is configured. The cascaded wrapped magnetic resonance coil can cover all parts of the body of the target object, and the multimodal imaging sequence can provide richer vascular information, including vascular wall structure, hemodynamics, and pathological characteristics. Therefore, the multi-position coil cascade method is combined with the multimodal imaging sequence to achieve one-stop imaging of large and small blood vessels (including arteries and veins) throughout the body. This method greatly reduces the scanning time and improves imaging efficiency. After using polysaccharide superparamagnetic iron oxide contrast agent at the same time, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the image are significantly improved, and the image quality score is also greatly improved. High-quality images can more clearly show the edges of blood vessels and anatomical structures. On the other hand, the contrast agent is polysaccharide superparamagnetic iron oxide, which is modified by polysaccharides to improve biocompatibility, reduce the risk of cytotoxicity, hemolytic reaction, and complement activation-related pseudo-allergic reaction (CARPA), and avoid potential health risks to patients.

[0122] See Figure 5 , is a flow chart of a model training method for a blood vessel extraction model provided in an embodiment of the present application. Figure 5 As shown, the method of the embodiment of the present application may include the following steps:

[0123] S201, obtaining a historical scan image pre-labeled with blood vessel grayscale values ​​and human tissue grayscale values;

[0124] S202, creating a blood vessel extraction model;

[0125] S203, constructing a target loss function for the blood vessel extraction model;

[0126] S204 , training the blood vessel extraction model according to the historical scanned images and the target loss function to generate a pre-trained blood vessel extraction model.

[0127] In an embodiment of the present application, on the one hand, after initialization is completed, a whole-body positioning image is constructed and signal markers are set, and a multimodal imaging sequence is configured. The cascaded wrapped magnetic resonance coil can cover all parts of the body of the target object, and the multimodal imaging sequence can provide richer vascular information, including vascular wall structure, hemodynamics, and pathological characteristics. Therefore, the multi-position coil cascade method is combined with the multimodal imaging sequence to achieve one-stop imaging of large and small blood vessels (including arteries and veins) throughout the body. This method greatly reduces the scanning time and improves imaging efficiency. After using polysaccharide superparamagnetic iron oxide contrast agent at the same time, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the image are significantly improved, and the image quality score is also greatly improved. High-quality images can more clearly show the edges of blood vessels and anatomical structures. On the other hand, the contrast agent is polysaccharide superparamagnetic iron oxide, which is modified by polysaccharides to improve biocompatibility, reduce the risk of cytotoxicity, hemolytic reaction, and complement activation-related pseudo-allergic reaction (CARPA), and avoid potential health risks to patients.

[0128] The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0129] See Figure 6 , which shows a schematic structural diagram of a multimodal whole-body vascular magnetic resonance imaging device provided by an exemplary embodiment of the present application. This multimodal whole-body vascular magnetic resonance 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 an information display module 10, a whole-body scouting image construction module 20, a magnetic resonance radiofrequency pulse transmission module 30, a signal strength determination module 40, a whole-body vascular scanning module 50, and a whole-body angiogenesis module 60.

[0130] An information display module 10 is used to display prompt information during the initialization process of the magnetic resonance imaging device, the prompt information including a first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and a second prompt information for inputting a contrast agent, wherein the contrast agent is polysaccharide superparamagnetic iron oxide;

[0131] A whole-body scout image construction module 20 is configured to construct a whole-body scout image for determining the positions of target object markers after initialization, set signal markers on the whole-body scout image, and configure a multimodal imaging sequence for magnetic resonance imaging.

[0132] The MRI radio frequency pulse transmitting module 30 is used to control the radio frequency system to transmit MRI radio frequency pulses to the target subject and to receive in real time coil electrical signals transmitted from a cascaded whole-body MRI coil. The cascaded whole-body MRI coil comprises multiple sub-coils covering various parts of the target subject, and the whole-body MRI coil is electrically connected to the MRI device.

[0133] A signal strength determination module 40 is used to determine the signal strength of each set signal marking point based on the coil electrical signal;

[0134] The whole-body vascular scanning module 50 is configured to perform a whole-body vascular scan of the target subject through a multimodal imaging sequence when the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, thereby obtaining a scanned target image sequence;

[0135] The whole body blood vessel generation module 60 is used to perform image reconstruction according to the target image sequence to generate a whole body blood vessel image.

[0136] It should be noted that the multimodal whole-body vascular MRI apparatus provided in the above embodiments, when performing a multimodal whole-body vascular MRI method, is merely illustrated by the division of the aforementioned functional modules. In actual applications, the aforementioned functions can be assigned to different functional modules as needed, i.e., the internal structure of the apparatus can be divided into different functional modules to perform all or part of the functions described above. Furthermore, the multimodal whole-body vascular MRI apparatus provided in the above embodiments and the multimodal whole-body vascular MRI method embodiments are based on the same concept. The implementation process is detailed in the method embodiments and will not be further described here.

[0137] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0138] In an embodiment of the present application, on the one hand, after initialization is completed, a whole-body positioning image is constructed and signal markers are set, and a multimodal imaging sequence is configured. The cascaded wrapped magnetic resonance coil can cover all parts of the body of the target object, and the multimodal imaging sequence can provide richer vascular information, including vascular wall structure, hemodynamics, and pathological characteristics. Therefore, the multi-position coil cascade method is combined with the multimodal imaging sequence to achieve one-stop imaging of large and small blood vessels (including arteries and veins) throughout the body. This method greatly reduces the scanning time and improves imaging efficiency. After using polysaccharide superparamagnetic iron oxide contrast agent at the same time, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the image are significantly improved, and the image quality score is also greatly improved. High-quality images can more clearly show the edges of blood vessels and anatomical structures. On the other hand, the contrast agent is polysaccharide superparamagnetic iron oxide, which is modified by polysaccharides to improve biocompatibility, reduce the risk of cytotoxicity, hemolytic reaction, and complement activation-related pseudo-allergic reaction (CARPA), and avoid potential health risks to patients.

[0139] The present application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implement the multimodal whole-body vascular magnetic resonance imaging method provided by the above-mentioned various method embodiments.

[0140] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the multimodal whole-body vascular magnetic resonance imaging method of each of the above method embodiments.

[0141] See Figure 7 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7 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 .

[0142] The communication bus 1002 is used to implement the connection and communication between these components.

[0143] The user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface.

[0144] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0145] The processor 1001 may include one or more processing cores. The processor 1001 utilizes various interfaces and circuits to connect the various components within the entire 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, the processor 1001 performs various functions of the electronic device 1000 and processes data. Optionally, the processor 1001 may be implemented in at least one hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 1001 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily 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; and the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 1001 and may be implemented separately on a single chip.

[0146] Among them, the memory 1005 may include a random access memory (RAM) or 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 also be optionally at least one storage system located away from the aforementioned processor 1001. As Figure 7 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 multi-modality whole-body vascular magnetic resonance imaging application.

[0147] exist Figure 7In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain user input data; and the processor 1001 can be used to call the multimodal whole-body vascular magnetic resonance imaging application stored in the memory 1005 and specifically perform the following operations:

[0148] During the initialization of the magnetic resonance imaging device, prompt information is displayed, including a first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and a second prompt information for inputting a contrast agent, wherein the contrast agent is polysaccharide superparamagnetic iron oxide;

[0149] After initialization is completed, a whole-body scout image is constructed for determining the positions of target object markers, signal markers are set on the whole-body scout image, and a multimodal imaging sequence for magnetic resonance imaging is configured;

[0150] Controlling the radio frequency system to transmit magnetic resonance radio frequency pulses to the target subject and receiving in real time coil electrical signals transmitted from a cascaded whole-body magnetic resonance coil; wherein the cascaded whole-body magnetic resonance coil comprises multiple sub-coils covering various parts of the target subject, and the whole-body magnetic resonance coil is electrically connected to the magnetic resonance imaging device;

[0151] Determine the signal strength of each set signal mark point based on the coil electrical signal;

[0152] When the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, a whole-body vascular scan of the target object is performed through a multimodal imaging sequence to obtain a scanned target image sequence;

[0153] Image reconstruction is performed based on the target image sequence to generate a whole-body vascular image.

[0154] In one embodiment, when the processor 1001 determines the signal strength of each set signal marker point based on the coil electrical signal, it specifically performs the following operations:

[0155] detecting a resonance signal returned from the target object through a coil system and converting the resonance signal into a resonance electrical signal;

[0156] Perform signal feature fusion based on the resonance electrical signal and the coil electrical signal to generate signal strengths at multiple locations;

[0157] From the signal strengths of multiple locations, the signal strengths of the set signal marker points are filtered out.

[0158] In one embodiment, when the processor 1001 performs signal feature fusion based on the resonance electrical signal and the coil electrical signal to generate signal strengths at multiple locations, the processor 1001 specifically performs the following operations:

[0159] The feedback electrical signal of each sub-coil is amplified, filtered and A / D converted in sequence to obtain a position digital signal;

[0160] One-hot encoding is performed on the part digital signal to generate a feature vector of the part digital signal;

[0161] Using historical resonance electrical signals to pre-train the neural network to determine the target parameter values ​​of the neural network;

[0162] Extract the features of the resonance electric signal according to the target parameter value to generate a resonance feature vector;

[0163] Based on the eigenvector and resonance eigenvector of the digital signal of the part, the signal strength of multiple position points is constructed.

[0164] In one embodiment, when the processor 1001 constructs the signal strengths of multiple locations based on the feature vectors and the resonance feature vectors of the location digital signal, the processor 1001 specifically performs the following operations:

[0165] A target matrix is ​​constructed according to the characteristic vector of the position digital signal and the connection order of multiple sub-coils;

[0166] Construct a connecting line between the first and last elements in the target matrix, and perform weighted summing and averaging of all elements on the connecting line to obtain the vector reference value;

[0167] Perform vector splicing processing based on the vector reference value, the eigenvector of the digital signal of each part, and the resonance eigenvector to obtain a splicing vector for each part;

[0168] The signal intensity of the splicing vector of each part is quantified to obtain the signal intensity of multiple position points.

[0169] In one embodiment, when the processor 1001 performs image reconstruction based on the target image sequence to generate a whole-body vascular image, the processor 1001 specifically performs the following operations:

[0170] Input the target image sequence into the pre-trained blood vessel extraction model;

[0171] Output the grayscale values ​​of blood vessels and human tissues corresponding to the target image sequence;

[0172] Blood vessel segmentation is performed based on the grayscale values ​​of blood vessels and human tissues to generate a whole-body blood vessel image.

[0173] In one embodiment, when generating a pre-trained blood vessel extraction model, the processor 1001 specifically performs the following operations:

[0174] Acquire historical scan images pre-labeled with blood vessel grayscale values ​​and human tissue grayscale values;

[0175] Create a blood vessel extraction model;

[0176] Construct the objective loss function of the blood vessel extraction model;

[0177] The blood vessel extraction model is trained based on the historical scanned images and the target loss function to generate a pre-trained blood vessel extraction model. The target loss function is expressed as:

[0178] is the cross entropy loss function for separating blood vessels from human tissues, λh(i)||f i || 2 It represents the constraint loss on the length of the feature vector when the blood vessels are separated from the human tissue incorrectly, e is a natural number, w is the fully connected layer, T is the transposition process, and f i It represents the features extracted by the model for the input image, h(i) is an indicator function. When the separation between blood vessels and human tissue is incorrect, h(i) = 1, otherwise h(i) = 0; λ is the weight of the constraint;

[0179] It is the constraint loss on the length of the feature vector when the blood vessels are correctly separated from the human tissue, γ is the weight of the constraint, and ε is the number to prevent the denominator from being 0, which is set to a fixed value.

[0180] In one embodiment, when the processor 1001 trains the blood vessel extraction model based on the historical scanned images and the target loss function to generate a pre-trained blood vessel extraction model, the processor 1001 specifically performs the following operations:

[0181] Integrate the objective loss function into the blood vessel extraction model;

[0182] Input the historical scanned image into the blood vessel extraction model that integrates the target loss function, and output the loss value of the model;

[0183] When the loss value reaches the minimum, a pre-trained blood vessel extraction model is generated; or when the loss value does not reach the minimum, the step of inputting the historical scan image into the blood vessel extraction model that integrates the target loss function is continued until the loss value reaches the minimum.

[0184] In one embodiment, when constructing a whole-body scout image for determining the positions of target object markers and setting signal markers on the whole-body scout image, the processor 1001 specifically performs the following operations:

[0185] Performing an initial scan of the target subject using a magnetic resonance imaging device to obtain a positioning image containing the entire body's anatomical structures;

[0186] Performing image registration on the whole-body positioning image and a plurality of pre-generated magnetic resonance three-dimensional template images to obtain a plurality of registration results;

[0187] For each registration result, the similarity between the whole-body positioning image and each magnetic resonance 3D template image is calculated;

[0188] The 3D magnetic resonance template image with the greatest similarity is used as the target template image;

[0189] Setting signal markers in the whole-body positioning image based on pre-marked locations in the target template image; or

[0190] Display positioning images;

[0191] Receive a parameter configuration request from the user for the client, where the parameter configuration request carries the location of the mark point manually selected by the user;

[0192] Based on the manually selected marker positions by the user, signal markers are set in the whole body positioning image.

[0193] In one embodiment, when performing a whole-body vascular scan of a target object using a multimodal imaging sequence, the processor 1001 specifically performs the following operations:

[0194] The scanning parameters corresponding to T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging are switched in sequence to perform a whole-body vascular scan on the target object.

[0195] In an embodiment of the present application, on the one hand, after initialization is completed, a whole-body positioning image is constructed and signal markers are set, and a multimodal imaging sequence is configured. The cascaded wrapped magnetic resonance coil can cover all parts of the body of the target object, and the multimodal imaging sequence can provide richer vascular information, including vascular wall structure, hemodynamics, and pathological characteristics. Therefore, the multi-position coil cascade method is combined with the multimodal imaging sequence to achieve one-stop imaging of large and small blood vessels (including arteries and veins) throughout the body. This method greatly reduces the scanning time and improves imaging efficiency. After using polysaccharide superparamagnetic iron oxide contrast agent at the same time, the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the image are significantly improved, and the image quality score is also greatly improved. High-quality images can more clearly show the edges of blood vessels and anatomical structures. On the other hand, the contrast agent is polysaccharide superparamagnetic iron oxide, which is modified by polysaccharides to improve biocompatibility, reduce the risk of cytotoxicity, hemolytic reaction, and complement activation-related pseudo-allergic reaction (CARPA), and avoid potential health risks to patients.

[0196] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program for multimodal whole-body vascular magnetic resonance imaging can be stored in a computer-readable storage medium. When executed, the program can include the processes in the above-described method embodiments. The storage medium for the program for multimodal whole-body vascular magnetic resonance imaging can be a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0197] The above disclosure is only a preferred embodiment of the present application, and 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 multimodal whole-body vascular magnetic resonance imaging method, characterized in that: The method comprises: During the initialization process of the magnetic resonance imaging device, prompt information is displayed, wherein the prompt information includes a first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and a second prompt information for inputting a contrast agent, wherein the contrast agent is polysaccharide superparamagnetic iron oxide; After the initialization is completed, a whole-body scout image is constructed for determining the positions of the target object's marker points, signal marker points are set on the whole-body scout image, and a multimodal imaging sequence for magnetic resonance imaging is configured; controlling a radio frequency system to transmit magnetic resonance radio frequency pulses to the target object and receiving in real time coil electrical signals transmitted from a cascaded whole-body magnetic resonance coil; wherein the cascaded whole-body magnetic resonance coil comprises a plurality of sub-coils covering various parts of the target object, and the whole-body magnetic resonance coil is electrically connected to the magnetic resonance imaging device; Determining the signal strength of each set signal marking point based on the coil electrical signal; When the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, performing a whole-body blood vessel scan on the target object through the multimodal imaging sequence to obtain a scanned target image sequence; Image reconstruction is performed according to the target image sequence to generate a whole-body vascular image.

2. The method according to claim 1, characterized in that The step of determining the signal strength of each set signal marking point according to the coil electrical signal includes: detecting a resonance signal returned from the target object through a coil system and converting the resonance signal into a resonance electrical signal; Performing signal feature fusion based on the resonance electrical signal and the coil electrical signal to generate signal intensities at multiple locations; The signal strength of each set signal marking point is screened out from the signal strengths of the multiple position points.

3. The method according to claim 2, characterized in that The coil electrical signal includes a feedback electrical signal of each sub-coil in a plurality of sub-coils; the plurality of sub-coils include at least a head and neck coil wrapped around the head, a body coil wrapped around the shoulders and chest, a body coil wrapped around the abdomen and waist, a body coil wrapped around the thighs and knee joints, and a body coil wrapped around the calves and feet; The performing signal feature fusion according to the resonance electrical signal and the coil electrical signal to generate signal strengths at multiple locations includes: Amplifying, filtering, and A / D converting the feedback electrical signal of each sub-coil in sequence to obtain a position digital signal; One-hot encoding is performed on the part digital signal to generate a feature vector of the part digital signal; Pre-training a neural network using historical resonant electrical signals to determine target parameter values ​​of the neural network; Extracting features of the resonance electric signal according to the target parameter value to generate a resonance feature vector; Based on the characteristic vector of the digital signal of the part and the resonance characteristic vector, the signal strengths of multiple position points are constructed.

4. The method according to claim 3, characterized in that The constructing of signal strengths of multiple locations based on the characteristic vector of the location digital signal and the resonance characteristic vector includes: constructing a target matrix according to the characteristic vector of the digital signal of the part and the connection order of the plurality of sub-coils; Constructing a connecting line between the first element and the last element in the target matrix, and performing weighted summing and averaging of all elements on the connecting line to obtain a vector reference value; Performing vector splicing processing according to the vector reference value, the characteristic vector of the digital signal of each part, and the resonance characteristic vector to obtain a splicing vector for each part; The signal intensity of the splicing vector of each part is quantified to obtain the signal intensity of multiple position points.

5. The method according to claim 1, characterized in that The step of reconstructing the target image sequence to generate a whole-body vascular image includes: Inputting the target image sequence into a pre-trained blood vessel extraction model; Outputting the blood vessel grayscale value and human tissue grayscale value corresponding to the target image sequence; Blood vessel segmentation is performed based on the blood vessel grayscale value and the human tissue grayscale value to generate a whole-body blood vessel image.

6. The method according to claim 5, characterized in that Follow these steps to generate a pre-trained vessel extraction model, including: Acquire historical scan images pre-labeled with blood vessel grayscale values ​​and human tissue grayscale values; Create a blood vessel extraction model; Constructing a target loss function of the blood vessel extraction model; The blood vessel extraction model is trained according to the historical scan image and the target loss function to generate a pre-trained blood vessel extraction model; wherein the target loss function is expressed as: is the cross entropy loss function for separating blood vessels from human tissues, λh(i)||f i || 2 It represents the constraint loss on the length of the feature vector when the blood vessels are separated from the human tissue incorrectly, e is a natural number, w is the fully connected layer, T is the transposition process, and f i It represents the features extracted by the model for the input image, h(i) is an indicator function. When the separation between blood vessels and human tissue is incorrect, h(i) = 1, otherwise h(i) = 0; λ is the weight of the constraint; It is the constraint loss on the length of the feature vector when the blood vessels are correctly separated from the human tissue, γ is the weight of the constraint, and ε is the number to prevent the denominator from being 0, which is set to a fixed value.

7. The method according to claim 6, characterized in that The training of the blood vessel extraction model according to the historical scanned image and the target loss function to generate a pre-trained blood vessel extraction model includes: Integrating the target loss function into the blood vessel extraction model; Inputting the historical scanned image into a blood vessel extraction model integrated with the target loss function, and outputting the loss value of the model; When the loss value reaches the minimum, a pre-trained blood vessel extraction model is generated; or when the loss value does not reach the minimum, the step of inputting the historical scan image into the blood vessel extraction model that integrates the target loss function is continued until the loss value reaches the minimum.

8. The method according to claim 1, characterized in that The step of constructing a whole-body positioning image for determining the positions of the target object's marker points and setting signal marker points on the whole-body positioning image includes: Performing an initial scan on the target object using the magnetic resonance imaging device to obtain a positioning image containing whole-body anatomical structures; Performing image registration on the whole-body positioning image and a plurality of pre-generated magnetic resonance three-dimensional template images to obtain a plurality of registration results; For each registration result, calculating the similarity between the whole-body positioning image and each magnetic resonance three-dimensional template image; The 3D magnetic resonance template image with the greatest similarity is used as the target template image; Setting signal marker points in the whole-body positioning image according to pre-marked position points in the target template image; or, displaying the positioning image; Receiving a parameter configuration request input by a user for a client, wherein the parameter configuration request carries a location of a mark point manually selected by the user; Based on the positions of the markers manually selected by the user, signal markers are set in the whole-body positioning image.

9. The method according to claim 1, characterized in that The multimodal imaging sequence includes T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging; The performing of a whole-body vascular scan on the target object by using the multimodal imaging sequence includes: The scanning parameters corresponding to the T1-weighted imaging, T2-weighted imaging, and T2*-weighted imaging are switched in sequence to perform a whole-body vascular scan on the target object.

10. A multimodal whole-body vascular magnetic resonance imaging device, characterized in that: The device comprises: An information display module is used to display prompt information during the initialization process of the magnetic resonance imaging device, wherein the prompt information includes a first prompt information for cascade wrapping of the whole-body magnetic resonance coil for the target object and a second prompt information for inputting a contrast agent, wherein the contrast agent is polysaccharide superparamagnetic iron oxide; A whole-body scout image construction module is used to construct a whole-body scout image for determining the positions of the target object's marker points after initialization, set signal marker points on the whole-body scout image, and configure a multimodal imaging sequence for magnetic resonance imaging; a magnetic resonance radio frequency pulse transmission module, configured to control the radio frequency system to transmit magnetic resonance radio frequency pulses to the target subject and to receive in real time coil electrical signals transmitted from a cascaded whole-body magnetic resonance coil; wherein the cascaded whole-body magnetic resonance coil comprises a plurality of sub-coils covering various parts of the target subject, and the whole-body magnetic resonance coil is electrically connected to the magnetic resonance imaging device; A signal strength determination module, configured to determine the signal strength of each set signal marking point based on the coil electrical signal; a whole-body vascular scanning module, configured to perform a whole-body vascular scan on the target object using the multimodal imaging sequence when the signal intensity of each signal marker point is greater than or equal to a preset intensity threshold, thereby obtaining a scanned target image sequence; The whole-body blood vessel generation module is used to perform image reconstruction according to the target image sequence to generate a whole-body blood vessel image.

Images