An MR arteriovenous dual mode vascular puncture navigation system and method

Abstract

The application provides an MR arteriovenous double-mode blood vessel puncture navigation system and method, which combines multispectral near-infrared imaging, real-time biomechanical simulation and virtual-real fusion display and the like to create a novel blood vessel puncture navigation architecture, can directly present the position and direction of subcutaneous veins and arteries, and can update the display in real time along with the movement or tissue deformation of a patient, and can display the dynamically adjusted three-dimensional position of blood vessels in real time, overcomes the defects of incomplete blood vessel positioning and poor dynamic adaptability in the prior art, and can provide accurate and intuitive navigation, can promote efficient and accurate blood vessel puncture, and effectively reduces the risk.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical treatment, in particular to a MR arteriovenous dual-mode blood vessel puncture navigation system and method. BACKGROUND

[0002] Invasive operations such as intravenous injection, blood sampling and arterial sampling are common clinical techniques, but due to individual differences, the blood vessels of some patients are difficult to accurately locate by naked eyes and palpation, such as obese people, infants, shock low perfusion patients, etc., and repeated puncture failure not only increases the pain of patients, but also may delay the opportunity for emergency treatment. Therefore, a device for assisting blood vessel positioning has emerged as the times require.

[0003] At present, there are mainly near-infrared vein imaging instruments, which utilize the strong tissue penetration of 700nm-900nm near-infrared light and the blood absorption characteristics to make the subcutaneous vein clearly present in infrared imaging, and then project back to the skin surface for the reference of doctors. This type of device has significant help for vein puncture, but has limitations: it can only display veins and cannot identify deeper arteries, and once the projected image is calibrated, if the patient moves or the operator changes the viewing angle, the vein image may deviate from the true position, affecting accuracy. On the other hand, for deep blood vessels, especially arteries (such as radial artery blood sampling, central vein catheterization), clinical mainly relies on ultrasound guidance, but ultrasound devices are large in size and require professional skills to operate.

[0004] Therefore, there is an urgent need for a puncture navigation system that can visualize veins and arteries simultaneously and adaptively change to improve the effectiveness of blood vessel positioning assistance, ensure the success rate of various blood vessel puncture operations, and reduce risks. SUMMARY

[0005] The present application provides a MR arteriovenous dual-mode blood vessel puncture navigation system and method, which combines multispectral near-infrared imaging, real-time biomechanical simulation and virtual-real fusion display technologies to create a novel blood vessel puncture navigation architecture that can directly present the position and direction of subcutaneous veins and arteries and update the display in real time as the patient moves or the tissue deforms. The real-time display of the dynamically adjusted blood vessel three-dimensional position overcomes the shortcomings of incomplete blood vessel positioning and poor dynamic adaptability in the prior art, and provides precise and intuitive navigation, which can promote efficient and accurate blood vessel puncture and effectively reduce risks.

[0006] In a first aspect, the present application provides a MR arteriovenous dual-mode blood vessel puncture navigation system, which comprises a multispectral imaging unit, a force sensing unit, a computing processing unit and a mixed reality display unit.

[0007] The multispectral imaging unit comprises a camera assembly and a light source assembly, the camera assembly acquires multispectral images of the target part of the patient under the uniform illumination provided by the light source assembly, wherein the multispectral images comprise color images and near-infrared images;

[0008] The force sensing unit comprises a MEMS pressure sensor and an inertial measurement unit, the MEMS pressure sensor acquires the contact pressure between the probe and the contact part of the patient, and the inertial measurement unit is used to acquire three-dimensional attitude data;

[0009] The computing processing unit comprises an image processing module, a blood vessel identification module, a biomechanics module and a navigation control module, the image processing module is used to pre-process the multispectral images and obtain first blood vessel images, the blood vessel identification module is used to distinguish arteries and veins from the first blood vessel images by using multi-wavelength difference analysis and pulsatile signals, and then generate second blood vessel images with blood vessel markers, the biomechanics module is used to calculate compensation values of blood vessel stress deformation and motion displacement based on a preset biomechanics model of the target part of the patient, in combination with the contact pressure and the three-dimensional attitude data, and taking the heartbeat influence into account, and the navigation control module is used to update the blood vessel positions in combination with the second blood vessel images and the compensation values, and generate corresponding virtual three-dimensional blood vessel models;

[0010] The mixed reality display unit comprises a glasses device, the glasses device displays corresponding display pictures of the virtual three-dimensional blood vessel models under a preset mixed reality display configuration, so as to be superimposed and displayed in the field of view of the target part of the patient, thereby assisting the operator to perform the blood vessel puncture operation.

[0011] In a second aspect, the application provides an MR arterial and venous dual-mode blood vessel puncture navigation method, which is applied to an MR arterial and venous dual-mode blood vessel puncture navigation system, the MR arterial and venous dual-mode blood vessel puncture navigation system comprises a multispectral imaging unit, a force sensing unit, a computing processing unit and a mixed reality display unit, the multispectral imaging unit comprises a camera assembly and a light source assembly, the force sensing unit comprises a MEMS pressure sensor and an inertial measurement unit, the computing processing unit comprises an image processing module, a blood vessel identification module, a biomechanics module and a navigation control module, and the mixed reality display unit comprises a glasses device, and the method comprises the following steps:

[0012] The camera assembly acquires multispectral images of the target part of the patient under the uniform illumination provided by the light source assembly, wherein the multispectral images comprise color images and near-infrared images;

[0013] The MEMS pressure sensor acquires the contact pressure between the probe and the contact part of the patient, and the inertial measurement unit is used to acquire three-dimensional attitude data;

[0014] The image processing module is configured to pre-process the multispectral image and obtain a first blood vessel image, the blood vessel recognition module is configured to distinguish arteriovenous of the first blood vessel image by using multi-wavelength difference analysis and pulsatile signal, and then generate a second blood vessel image with blood vessel markers, the biomechanics module is configured to calculate compensation values of blood vessel stress deformation and motion displacement based on a preset biomechanics model of the target site of the patient, combined with contact pressure and three-dimensional posture data, and considering the influence of heartbeat, and the navigation control module is configured to update the blood vessel position combined with the second blood vessel image and the compensation values, and generate a corresponding virtual three-dimensional blood vessel model.

[0015] The glasses device displays a corresponding display screen of the virtual three-dimensional blood vessel model in the preset mixed reality display configuration to be superimposed and displayed in the field of view of the operator at the target site of the patient, assisting the operator to perform the blood vessel puncture operation.

[0016] In a fourth aspect, a computer readable storage medium is provided, which stores a plurality of instructions. The instructions are adapted to be loaded by a processor to execute the method provided in the first aspect or any possible implementation manner of the first aspect of the present application.

[0017] From the above, the present application has the following beneficial effects:

[0018] For the blood vessel puncture navigation target, the present application combines multispectral near-infrared imaging, real-time biomechanics simulation and virtual-real fusion display technologies to create a novel blood vessel puncture navigation architecture, which can directly present the position and direction of subcutaneous veins and arteries, and update the display in real time with the movement or tissue deformation of the patient, and display the dynamically adjusted three-dimensional position of the blood vessel in real time, overcoming the shortcomings of incomplete blood vessel positioning and poor dynamic adaptability in the prior art, and providing precise and intuitive navigation, promoting efficient and accurate blood vessel puncture, and effectively reducing the risk. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 It is a system architecture schematic diagram of the MR arteriovenous blood vessel puncture navigation system of the present application.

[0021] Figure 2 It is a scene schematic diagram of the multispectral imaging unit and the force sensing unit of the present application.

[0022] Figure 3A scene schematic diagram of the MR display effect of the forearm blood vessels of the present application;

[0023] Figure 4 A scene schematic diagram of the system work logic of the overall scheme of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0025] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or device that includes a series of steps or modules does not necessarily have to be limited to only those steps or modules clearly listed, but can include other steps or modules that are not clearly listed or inherent to the process, method, product, or device. The naming or numbering of the steps appearing in the present application does not mean that the steps in the method flow must be executed in the order / time sequence indicated by the naming or numbering, and the execution order of the named or numbered flow steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0026] The division of the modules appearing in the present application is a logical division, and in actual application, there can be another division manner, for example, multiple modules can be combined or integrated in another system, or some features can be ignored or not executed, in addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be through some interface, the indirect coupling or communication connection between the modules can be electrical or other similar forms, which are not limited in the present application. Moreover, the modules or sub-modules described as separate components can or can not be physically separated, can or can not be physical modules, or can be distributed into multiple circuit modules, and some or all of the modules can be selected according to actual needs to achieve the purpose of the present application scheme.

[0027] First, refer to Figure 1 A system architecture schematic diagram of the MR arteriovenous puncture navigation system of the present application is shown, from Figure 1It can be clearly seen that the MR arteriovenous puncture navigation system provided by the application includes four units, i.e., a multispectral imaging unit, a force sensing unit, a computing processing unit, and a mixed reality display unit. Among these units, corresponding to the data flow involved in the subsequent scheme description, appropriate communication connections are configured. Generally, wired connections are used, but wireless connections are not excluded in some cases. In addition, the specific communication connection method is not fixed, i.e., it can be switched / updated between wired and wireless connections in some cases, which can meet the data transmission function under the corresponding application condition.

[0028] Next, the four units are specifically expanded.

[0029] (1) Multispectral imaging unit

[0030] The multispectral imaging unit includes a camera assembly and a light source assembly. The camera assembly collects multispectral images of the target part of the patient under the uniform illumination provided by the light source assembly, wherein the multispectral images include color images and near-infrared images.

[0031] It can be seen that the purpose of the multispectral imaging unit is to provide the images required for the operation of the system of the application, and specifically involves two aspects / types of images, i.e., color images and near-infrared images. The multispectral images formed can be transmitted to the computing processing unit for subsequent data processing.

[0032] Corresponding to the collection capability of color images and near-infrared images, the multispectral imaging unit specifically includes a camera assembly and a light source assembly. The camera assembly collects multispectral images of the target part of the patient under the uniform illumination provided by the light source assembly.

[0033] Among them, the digital camera involved in the camera assembly can be one or multiple, which can be configured according to the specific imaging scheme in actual situation, and the multispectral light source involved in the light source assembly is the same.

[0034] As a specific implementation scheme here, the corresponding sensitive wavelength range of the camera assembly for collecting color images can be specifically 400nm-700nm, and the corresponding sensitive wavelength range of the camera assembly for collecting near-infrared images can be specifically 700nm-900nm. Here, it is not meant that the images are only collected in these wavelength ranges, but that the corresponding high-resolution CMOS sensor (such as Sony IMX586 series) has high sensitivity to these wavelength ranges. The collected images can be transmitted to the computing processing unit through wired connection (such as high-speed data bus (MIPI CSI high-speed interface));

[0035] The light source assembly can specifically include a visible light source, an 850nm wavelength narrowband near-infrared light-emitting diode (LED) array, and a 760nm wavelength narrowband near-infrared LED array, and the illumination areas of the two narrowband near-infrared LED arrays are obviously overlapped;

[0036] The light source assembly is coaxial with the lens of the camera assembly, or the light source assembly is arranged around the lens of the camera assembly to directly illuminate the target part of the patient at a vertical or inclined angle lower than a threshold (here, the inclined angle is a slight inclined angle), thereby providing uniform illumination conditions.

[0037] The image captured by the camera can be configured to have a resolution of no less than 1280x1024 and a frame rate of 30fps, so as to ensure clear imaging and real-time performance.

[0038] The selection of the two wavelengths of 760nm and 850nm for the light source is based on the light absorption spectrum of hemoglobin. At 760nm, the absorption of oxygenated hemoglobin is strong, and at 850nm, the absorption of deoxygenated hemoglobin is strong. The difference between the two helps to distinguish between arteries and veins.

[0039] The target part of the patient can be a specific part such as a forearm in a clinic, which can be adjusted and changed according to actual conditions.

[0040] As an example, the camera assembly can further be equipped with a switchable filter mechanism to switch different spectral imaging modes in a very short time. One mode transmits near-infrared light (cutting off visible light) for shooting infrared images of blood vessels, and the other mode transmits full-spectrum light for shooting images with skin surface details.

[0041] In one imaging cycle, the system controls the light source to alternately emit different wavelength near-infrared pulses, and synchronously controls the camera to capture grayscale images of corresponding wavelengths, and turns on full-spectrum illumination to capture color images.

[0042] After rapid switching, three images can be obtained in one imaging cycle (about 33ms): a color visible light image I vis , a wavelength λ1 near-infrared image I IR1 (such as 850nm), and a wavelength λ2 near-infrared image I IR2 (such as 760nm).

[0043] If three-dimensional information needs to be acquired, the multispectral imaging unit can also adopt a binocular stereo camera configuration, two cameras are installed with a certain baseline, and the 3D point cloud of the scene (including the depth information of the skin surface and blood vessels) is recovered through stereo matching. In the scheme of the present application, in order to simplify the system, monocular multispectral imaging is mainly adopted, and the depth of the blood vessels is approximately estimated in combination with the biomechanical model described later. Of course, the use of a depth camera can also provide more direct depth information. In this case, the depth information collected by the depth camera can be used as a new model input to assist in the approximate estimation of the depth of the blood vessels.

[0044] As another example, the point multispectral imaging unit can start to collect multispectral images at a preset frame rate under control. Each frame of multispectral image specifically includes: a visible light full-color image I vis , two infrared gray-scale images, i.e., a wavelength λ1 near-infrared image I IR1 (such as 850 nm) and a wavelength λ2 near-infrared image I IR2 (such as 760 nm). For example, at t=0, the 850 nm near-infrared LED array is turned on, and other light sources are turned off. The camera captures I IR1 , then at , the 760 nm near-infrared LED array is switched on, and I IR2 is captured, then the white light LED is turned on at the same time, the camera is switched to the IR filter mode, and I vis is captured.

[0045] At the same time, compared with the corresponding data processing method that can be used, in order to conveniently and efficiently eliminate the time difference influence of different images, as a specific implementation scheme, the camera assembly adopts a double-camera structure scheme, which involves a first camera for collecting color images and a second camera for collecting near-infrared images. The color image and the near-infrared image are collected at the same time, i.e., the visible light and the near-infrared light are acquired at the same time.

[0046] Alternatively, the camera assembly alternately collects different images under the condition that the frame rate is higher than a threshold value.

[0047] Of course, the scheme here can also continue to combine the subsequent corresponding data processing method, i.e., the time difference image elimination processing, to continue to eliminate the time difference influence of different images in the case that the time difference influence may still exist and is higher than a threshold value in actual situations.

[0048] (2) Force sensing unit

[0049] The force sensing unit includes a MEMS pressure sensor and an inertial measurement unit. The MEMS pressure sensor collects the contact pressure (or force) between the probe and the contact part of the patient, and the inertial measurement unit is used to collect three-dimensional attitude data.

[0050] Among them, the Micro-Electro-Mechanical System (MEMS) pressure sensor can be specifically installed in the part of the system that can be in direct contact with the patient's skin or in the device shell, for real-time monitoring of the pressure applied to the patient's skin surface by the device during operation.

[0051] As an example, the MEMS pressure sensor can be specifically configured with: selecting a high-sensitivity strain gauge or piezoelectric sensor, the sensor range is 0~10N, and the accuracy is not less than 0.05N, so as to well capture the small pressure changes, such as capturing the small pressure changes of the blood vessel position after the skin is pressed, and the small pressure wave of the arterial pulse transmitted to the body surface.

[0052] The Inertial Measurement Unit (IMU) can be specifically installed on the main circuit board inside the module, containing a three-axis accelerometer and a three-axis gyroscope, to measure the device attitude and spatial displacement changes in real time, and output corresponding three-dimensional attitude data.

[0053] The force sensing unit can be connected to the computing processing unit through a digital interface (such as SPI, I2C interface) for real-time transmission of pressure data and three-dimensional attitude data, plus the spatial positioning of the MR glasses, which can estimate whether the target part of the patient has moved as a whole, and can be used for subsequent dynamic correction calculation of the blood vessel position based on the biomechanical model, in addition, it can also provide navigation prompts, such as when the probe is pressed too hard, the system can remind the operator to reduce the pressure to prevent the vein from collapsing.

[0054] For the data acquisition link of the above two aspects, it can also be understood more intuitively through Figure 2 A scene schematic diagram of the multispectral imaging unit and the force sensing unit of the present application is shown.

[0055] In addition, for the force sensing unit, it can be specifically implemented through corresponding supports, semi-automatic mechanical arms (i.e. the blood vessel puncture operation based on semi-automatic mechanical arms still involves manual operation of medical personnel, but it also involves a part of automatic operation), etc. At the same time, the force sensing unit can be further integrated with the multispectral imaging unit in structure design, which can be understood that the specific structure of the force sensing unit and the multispectral imaging unit is relatively flexible, and in actual application, it can be adaptively designed according to actual needs.

[0056] (3) Computing processing unit

[0057] The computing processing unit comprises an image processing module, a blood vessel identification module, a biomechanics module and a navigation control module, the image processing module is used for pre-processing the multispectral image and obtaining a first blood vessel image, the blood vessel identification module is used for distinguishing arteriovenous of the first blood vessel image by using multi-wavelength difference analysis and pulsatile signal, and then generating a second blood vessel image with blood vessel markers, the biomechanics module is used for calculating compensation values of blood vessel stress deformation and motion displacement on the basis of a preset biomechanics model of a target site of a patient, in combination with contact pressure and three-dimensional posture data and in consideration of heartbeat influence, and the navigation control module is used for updating blood vessel positions in combination with the second blood vessel image and the compensation values and generating a corresponding virtual three-dimensional blood vessel model.

[0058] It is easy to see that the computing processing unit is a unit responsible for core data processing of the scheme, which uses the data collected by the above two units, i.e., the multispectral image, the contact pressure and the three-dimensional posture data, to complete image processing, data fusion, model calculation and result output and the like. The computing processing unit can be specifically composed of a high-performance embedded processor, such as an ARM embedded platform containing GPU acceleration, an FPGA+ARM heterogeneous SoC, an NVIDIA Jetson embedded AI computing board, wherein the NVIDIA Jetson embedded AI computing board contains an 8-core CPU and a CUDA parallel GPU, which is sufficient to support real-time image analysis and physical simulation, and well meets the data processing capability required by the scheme.

[0059] 3.1) Image processing module

[0060] The pre-processing carried out by the image processing module is to highlight the blood vessels in the image to form a clear blood vessel image, i.e., a pre-processed image, which can specifically involve denoising, enhancement, segmentation, white balance, distortion correction and the like. For enhancement and segmentation, specifically, there can be:

[0061] a) Filter denoising, histogram equalization and the like enhancement operations are respectively performed on the near-infrared image I IR1 and I IR2 to improve the contrast between blood vessels and background;

[0062] b) Multi-scale filtering + threshold segmentation method is used for feature extraction to extract the blood vessel pixel region.

[0063] 3.2) Blood vessel identification module

[0064] The blood vessel recognition module, on the basis of image recognition algorithm, to the first blood vessel image obtained by the foregoing preprocessing, specifically utilizes multi-wavelength difference analysis and pulsation signal (through the gray level change period of the image sequence) to distinguish the involved arteries and veins, and respectively gives different marks to mark whether the specific blood vessel is an artery or a vein, and is fused into the visible light image coordinate system to generate a second blood vessel image with blood vessel marks. For the data processing of this part, as an example, the following can be specifically used:

[0065] For the information of the two wavelengths of 760 nm and 850 nm: for example, calculate the corresponding pixel I diff =I IR1 -k×I IR2 (k is a calibration coefficient), and utilize the hemoglobin absorption difference to enhance the visualization of the veins;

[0066] Then, the morphological operation is used to connect the small paragraphs to obtain a continuous blood vessel binary image, thin the blood vessel center line, and record the shape, length and other geometric information of the blood vessel connected segments.

[0067] Then, the extracted blood vessels are classified: on the one hand, by analyzing the change of the gray level of the blood vessel pixels in the two near-infrared image I IR sequences with time, if the periodic fluctuation is detected at the heart rate (1 Hz), it is determined that the blood vessel is an artery (exists pulsation); otherwise, the gray level is stable and continuous, which is a vein; another criterion is to compare the average intensity ratio of I IR1 and I IR2 in the blood vessel region, and utilize the characteristics such as the relatively high brightness of the artery at 760 nm to assist in the judgment;

[0068] After the arteries and veins are recognized, different marks (such as mark ID or color) are given.

[0069] Then, the blood vessel information is aligned and fused with the color image I vis . It is considered that the color image I vis and the near-infrared image I IR (I IR1 and I IR2 ) have been aligned in the same coordinate system through hardware coaxiality or calibration in advance, so that the blood vessel contour can be directly drawn on the corresponding position of the near-infrared image I vis , to generate an augmented reality image frame, in which the skin surface texture is clear and visible, and the subcutaneous blood vessel path can also be displayed in lines or highlights (veins are blue and arteries are red).

[0070] If a binocular camera is used, the depth (z coordinate) of each blood vessel point can also be estimated using the parallax data in the above processing process; and in the case of using a monocular camera, the depth of each blood vessel point can also be estimated in combination with an empirical model, for example, the depth from the surface can be estimated according to the clarity and size of the vein imaging (e.g., the estimated depth is about 0.5 cm), the depth from the surface can be estimated according to the pulsation amplitude of the artery (e.g., the estimated depth is about 1.2 cm), etc.

[0071] The above image processing or identification steps can be completed in real time for each frame. Modern FPGA or GPU parallel computing can accelerate the segmentation operation to be completed within tens of milliseconds.

[0072] The processing results described above can be stored in the form of a predetermined corresponding data structure (such as a blood vessel centerline point set and attributes).

[0073] 3.3) Biomechanics module

[0074] The biomechanics module involves the application of a biomechanics model. It can be understood that for different patient target sites involved in clinical work, the present application can be configured with a corresponding biomechanics model for the compensation operation of the blood vessel position to be described.

[0075] As a specific implementation scheme, the biomechanics model involved in the present application can be a finite element model or an equivalent elastic model, and can be divided into skin, subcutaneous fat muscle tissue and blood vessels for different patient target sites as different layers or different material regions, and is configured with corresponding physical property parameters. The model boundary condition assumes that the distal end is fixed and the proximal end has a slight periodic displacement with the heartbeat.

[0076] As an example, for the forearm soft tissue patient target site, the physical property parameters are: skin Young's modulus (E) ≈ 0.5 MPa, fat Young's modulus (E) ≈ 0.05 MPa, vein vessel wall Young's modulus (E) ≈ 0.1 MPa, artery vessel wall Young's modulus (E) ≈ 0.5 MPa, intravenous pressure 5 mmHg-15 mmHg, arterial pressure 80 mmHg-120 mmHg (dynamic), etc.

[0077] For specific model processing, as a specific implementation scheme, the biomechanics module mentioned above is used to calculate the compensation value of the blood vessel stress deformation and motion displacement based on the pre-set biomechanics model of the patient target site in combination with the contact pressure and three-dimensional posture data and considering the influence of heartbeat. The specific processing content can include the following:

[0078] 1) The contact pressure collected by the MEMS pressure sensor is used to estimate the normal stress on the skin surface, and the model outputs the corresponding tissue compression deformation field after applying the surface stress;

[0079] Specifically, the reading F of the contact pressure (pressure size) collected by the MEMS pressure sensor press is used to estimate the normal stress p=F on the skin surface press / A (A is the contact area), and the model outputs the tissue compression deformation field Δ(x, y, z) after applying the surface stress.

[0080] 2) According to the three-dimensional attitude data collected by the inertial measurement unit and the pose of the glasses device, it is determined whether the target part of the patient has overall translation and overall rotation, and if so, the same rigid body motion is applied to the model coordinate system;

[0081] It can be understood that corresponding to the MR technology, the glasses device is usually configured with a corresponding inertial measurement unit itself, and the pose of the glasses device can be collected.

[0082] In this way, the three-dimensional attitude data and the pose of the glasses device are combined to assist in determining whether to perform the same rigid body motion processing, and when the same rigid body motion is needed, the processing is performed to fit the overall translation and rotation that occurs at the moment, to assist the model to more accurately locate the position of the blood vessel.

[0083] 3) For the consideration of the influence of heartbeat, a periodic pulse pressure is applied in the artery, and the periodic pulse pressure is obtained by a harmonic internal pressure source preset by the model, and the amplitude corresponds to the difference between systolic pressure and diastolic pressure, and the frequency is about 1Hz;

[0084] Of course, the periodic pulse pressure applied in the artery by the image processing layer here can also be obtained by an external harmonic internal pressure source configured by the model.

[0085] 4) The model synthesizes different inputs, calculates the change of the blood vessel center position and the change of the blood vessel radius in the three-dimensional space through finite element solving or analytical approximation, and in the calculation process, on the one hand, the analytical formula is used to quickly estimate, and on the other hand, more detailed calculation is performed in the background asynchronously, so that the calculation result is continuously corrected.

[0086] After the configuration work of the model input in the above several aspects is completed, the biomechanical model of the current target patient part is configured, the above inputs are synthesized, the change of the blood vessel center position Δr=(Δx, Δy, Δz) and the change of the blood vessel radius ΔR are calculated through finite element solving or analytical approximation.

[0087] For example, when the pressing force increases, the model can result in a 30% reduction in vein radius and a 2mm downward displacement of the center; when the patient's arm moves slightly by 1cm, the model outputs a 1cm translation of the entire vascular network.

[0088] In terms of details, since the model calculation takes a certain amount of time, in order to ensure real-time performance, a two-stage strategy can be adopted: on the one hand, a quick estimate is made using an analytical formula, such as the degree of vein flattening ≈ f(F press ) linear relationship under the assumption of small deformation, and on the other hand, more detailed finite element calculations or analytical approximations can be performed asynchronously in the background, and the quick estimate results for the same model input of the former aspect can be overlaid or fused, and the calculation results can be continuously corrected in this way.

[0089] Finally, the model can output a set of blood vessel position correction parameters, i.e., the compensation values for the blood vessel stress deformation and motion displacement mentioned earlier, which can specifically include the displacement vector Δr = (Δx, Δy, Δz) (three-dimensional) and the radius change ΔR of each blood vessel.

[0090] At the same time, the biomechanical model can further introduce new model inputs or other influencing factors that can contribute to the correction of blood vessel position, such as blood pressure, body mass index (BMI), etc., to promote better model processing accuracy.

[0091] 3.4) Navigation control module

[0092] The navigation control module is responsible for fusing the results of the previous processing to generate a corresponding virtual three-dimensional blood vessel model for the mixed reality display unit to perform screen display work and perform superimposed display. Based on the second blood vessel image obtained by the blood vessel recognition module and the compensation values for the blood vessel stress deformation and motion displacement calculated by the biomechanical module, the blood vessel position can be updated (which can involve blood vessel positioning and morphological changes), and a real-time three-dimensional virtual blood vessel model that adapts to the current field of view of the glasses can be generated, which can be accurately projected into the real field of view.

[0093] For the processing of the three-dimensional virtual blood vessel model, specifically:

[0094] If a vein is a centerline in image coordinates (2D pixel coordinates), the depth z0 is preliminarily estimated, and the three-dimensional coordinates (x, y, z) = (x0 + Δx, y0 + Δy, z0 + Δz) are preliminarily determined. The three-dimensional coordinates of the blood vessel can be added to the displacement vector Δr = (Δx, Δy, Δz) obtained by analysis to obtain the latest position of the vein in the actual space.

[0095] If MR glasses are used for display, the system can also transform the new three-dimensional coordinates into the display coordinate system of the glasses. Since the registration between the glasses and the camera has been completed during initialization, the same reference coordinates can be used. Therefore, the navigation control module can project the three-dimensional point coordinates of the center line of each blood vessel onto the viewing plane of the glasses according to the calibration matrix, and generate the corresponding 2D rendering instructions.

[0096] At the same time, the navigation control module can also select the corresponding rendering style according to the blood vessel attributes: veins are rendered with semi-transparent blue lines and arteries with red lines, and the brightness / thickness is encoded (for example, adjusting the brightness attenuation according to the depth to simulate the effect of observing through the skin).

[0097] Correspondingly, as a specific implementation method, in the corresponding display screen of the virtual three-dimensional blood vessel model, veins can be displayed as blue semi-transparent virtual lines, and arteries can be displayed as red dashed semi-transparent lines. The display effects are used to indicate the depth information of the blood vessels. The display effects include at least one of the following: line thickness, transparency, and brightness variation.

[0098] Furthermore, if needed, other navigation information can be overlaid, such as the optimal puncture vessel (e.g., a vein with a large diameter and a long straight segment, suitable for cannulation), the optimal puncture point (the optimal location that can be recommended based on the vessel diameter and depth, marked with a small green dot on the vessel), and needle trajectory guidance (displaying the needle insertion path from the skin surface to the target vessel as a virtual straight line). All the graphic elements involved can be synthesized into a single frame of augmented reality, which is sent to the mixed reality display unit at a frequency of approximately 30fps for loading and display.

[0099] Correspondingly, as a specific implementation method, the corresponding display screen of the virtual three-dimensional blood vessel model can also include navigation information such as the optimal puncture vessel, the optimal puncture point, needle trajectory guidance, needle position, needle insertion angle, needle insertion depth, needle insertion path, pressure reading, whether the puncture was successful, and simulated arterial pulsation. Among these, the needle is equipped with image-recognizable markers or corresponding sensors to assist in capturing the needle position, needle insertion angle, needle insertion depth, and needle insertion path.

[0100] Among them, image-recognizable markers refer to physical markers that can reflect specific features at the image level, such as reflective metal markers. Sensors can directly reflect the location of their own pillow through corresponding sensing processing. The specific sensing principles or sensing types involved are quite flexible.

[0101] Whether a puncture was successful can be determined by changes in pressure readings or by whether the resistance suddenly decreased during the puncture, or by observing changes in blood vessel deformation (e.g., a punctured vein will collapse). Specifically, changes in the color of the blood vessel can indicate whether the puncture was successful.

[0102] Thus, taking the relatively frequent arm puncture as an example, medical staff wearing glasses will see a vascular distribution map on the patient's arm as if "seeing through" it. The veins' course closely matches their anatomical location (including possible slight flattening due to pressure), and the arteries are also displayed at a slightly deeper level with a slight pulsating effect (simulated by brightening and darkening every second). When the medical staff or patient moves slightly, the glasses' head tracking and model correction work together to keep the virtual vascular image firmly attached to the correct position without drifting or misaligning, achieving a very good navigation and visual assistance effect. Furthermore, combined with further navigation information prompts, the navigation and visual assistance effects can be further enhanced.

[0103] (4) Mixed Reality Display Unit

[0104] The mixed reality display unit includes glasses, which display a virtual three-dimensional blood vessel model under a preset mixed reality display configuration. The display is superimposed on the patient's target site in the operator's field of vision to assist the operator in performing vascular puncture.

[0105] Specifically, the glasses device is a head-mounted augmented reality (AR) glasses or mixed reality (MR) glasses, featuring transparent diffractive waveguide lenses, high-resolution virtual image projection function, and a built-in spatial tracking module. It can stably overlay the virtual vascular image, i.e., the three-dimensional virtual vascular model, obtained through real-time processing onto the real patient's target site in real time within the field of vision, providing real-time and stable vascular position navigation services.

[0106] Thus, for reference Figure 3 The illustration shows a scenario diagram of the MR display effect of the forearm blood vessels according to this application. In specific applications, medical staff can simultaneously see the real patient's limb and the virtual blood vessel direction superimposed on it through glasses, intuitively guiding the needle direction. If glasses are not provided, it is understood that the system of this application can also allow the use of a miniature projector installed on the probe to directly project the virtual blood vessel image, i.e., the three-dimensional virtual blood vessel model, onto the skin surface, or use a corresponding screen to display augmented reality images (real-time playback after superimposing blood vessel video from a camera). These settings can serve as supplements to the solution to meet application needs in some situations.

[0107] At the same time, it is easy to understand that the different aspects of screen display content or different aspects of navigation information mentioned above can be switched / adjusted according to different actual needs in specific applications. For example, for arteries and veins, only the arteries can be displayed, and after viewing the artery condition, the vein condition can be viewed separately, and then the overall condition of arteries and veins can be viewed.

[0108] Furthermore, it is important to understand that the above solutions are typically implemented after system initialization and calibration have been completed.

[0109] In this regard, focusing on the calibration process involved, as a specific implementation method, the glasses device and camera components can be integrated into a single design to solidify their relative positions, thereby avoiding the need to re-perform calibration before subsequent image acquisition and processing after the initial calibration process.

[0110] Alternatively, calibration can be performed by setting visual markers on the target areas of the patient before formally acquiring multispectral images.

[0111] The visual markers can be understood as another type of image-identifiable marker specifically designed for calibration processing in this area.

[0112] Specifically, as an example, calibration settings can be performed before use:

[0113] a) Fix the multispectral imaging unit about 15-30cm above the patient's target site. This distance is determined according to the lens field of view and the size of the imaging area, so that the entire puncture area (e.g., the forearm) falls into the camera's field of view.

[0114] b) Power on the system and the camera captures a frame containing a calibration pattern (such as a checkerboard pattern). The processing unit then corrects the camera's intrinsic parameter distortion and establishes the transformation relationship between the camera image coordinate system and the real-world coordinate system.

[0115] c) Alignment of the coordinates between the glasses and the camera: An integrated design can be used to fix the relative positions of the camera and glasses, or a visual marker can be attached to the patient's skin. The camera recognizes the unknown of the marker, and the glasses simultaneously recognize the position of the marker. In this way, the coordinate systems of the two markers are unified. After calibration, the system enters the real-time navigation mode.

[0116] In this case, the entire solution outlined above can also be combined with... Figure 4 The following is a schematic diagram illustrating a scenario of the overall system working logic of this application, for a more vivid understanding.

[0117] Based on the above solutions, it can be seen that this application combines multispectral near-infrared imaging, real-time biomechanical simulation, and virtual-real fusion display technologies to create a novel vascular puncture navigation architecture. This architecture can directly present the location and direction of subcutaneous veins and arteries, and update the display in real time with the patient's movement or tissue deformation. It also displays the dynamically adjusted three-dimensional position of the blood vessels in real time, overcoming the shortcomings of incomplete vascular positioning and poor dynamic adaptability in existing technologies. By providing accurate and intuitive navigation, it can promote efficient and precise vascular puncture and effectively reduce risks.

[0118] More specifically, the above-mentioned solutions in this application can achieve the following specific beneficial effects:

[0119] 1. Dual-mode arteriovenous imaging: Employing multispectral optical imaging technology, the system performs label-free real-time vascular modeling and simultaneously acquires information on veins and arteries beneath the patient's body surface. Through the fusion processing of optical images at different wavelengths and analysis of pulsation characteristics, the system can distinguish between veins and arteries and display them with different colors or markers, helping medical staff avoid accidentally puncturing arteries or accurately locate arteries when needed.

[0120] 2. Dynamic biomechanical correction: A biomechanical model is introduced to dynamically simulate soft tissues and blood vessels. Combined with data such as patient movement, respiration, and probe contact pressure, the vascular model dynamically adjusts with body position / pressure, calculates the deformation and displacement of blood vessels in real time, and corrects the displayed position. This dynamic navigation ensures that the vascular projection always matches the actual anatomy. Even if the patient moves slightly or the tissue is pressed, the system can adaptively adjust, improving the puncture success rate and safety.

[0121] 3. Mixed Reality Intuitive Display: Using MR display devices (such as see-through AR glasses), the reconstructed 3D model of blood vessels is directly superimposed on the patient's body surface, achieving a "what you see is what you get" navigation effect. Compared with traditional ultrasound screens, operators do not need to frequently look down at the monitor and coordinate hand-eye positioning. Instead, they can directly see the location of blood vessels on the patient's body, making the operation process more natural and efficient.

[0122] 4. Integrated Puncture Navigation: This system integrates optical imaging, image processing, spatial positioning, and navigation display. It features multimodal feedback, requires no large equipment, and is suitable for bedside and field use. Information acquisition and display are completed in the same device coordinate system (with built-in coordinate registration), avoiding cumbersome multi-device registration. It is easy to use and can be widely applied to specific scenarios such as intravenous infusion, arterial blood collection, and deep vein catheterization. It provides intelligent decision support for medical staff, such as recommending the best puncture vessel and indicating the needle insertion angle and depth, effectively reducing trial and error.

[0123] This achieves the integration of visualization of veins and arteries with puncture navigation, which can significantly improve the visibility and success rate of clinical puncture operations, and has significant clinical practical value and innovative significance.

[0124] Furthermore, based on the MR arteriovenous dual-mode vascular puncture navigation system provided in this application, this application also provides an MR arteriovenous dual-mode vascular puncture navigation method from the perspective of control method process.

[0125] Specifically, the MR arteriovenous dual-mode vascular puncture navigation method provided in this application is applied to an MR arteriovenous dual-mode vascular puncture navigation system. The MR arteriovenous dual-mode vascular puncture navigation system includes a multispectral imaging unit, a force sensing unit, a computational processing unit, and a mixed reality display unit. The multispectral imaging unit includes a camera assembly and a light source assembly; the force sensing unit includes a MEMS pressure sensor and an inertial measurement unit; the computational processing unit includes an image processing module, a blood vessel recognition module, a biomechanical module, and a navigation control module; and the mixed reality display unit includes a glasses device. Based on this, the MR arteriovenous dual-mode vascular puncture navigation method provided in this application specifically includes the following steps:

[0126] (1) The camera assembly acquires multispectral images of the patient's target area under uniform illumination provided by the light source assembly, wherein the multispectral images include color images and near-infrared images;

[0127] (2) The MEMS pressure sensor collects the contact pressure between the probe and the patient contact area, and the inertial measurement unit is used to collect three-dimensional attitude data;

[0128] (3) The image processing module is used to preprocess the multispectral image and obtain the first blood vessel image. The blood vessel recognition module is used to distinguish the arteries and veins in the first blood vessel image by using multi-wavelength difference analysis and pulsation signal, and then generate the second blood vessel image with blood vessel marking. The biomechanics module is used to calculate the compensation value of blood vessel stress deformation and motion displacement based on the preset biomechanical model of the patient's target site, combined with contact pressure and three-dimensional posture data, and considering the influence of heartbeat. The navigation control module is used to update the blood vessel position by combining the second blood vessel image and the compensation value, and generate the corresponding virtual three-dimensional blood vessel model.

[0129] (4) The glasses device displays the corresponding display screen of the virtual three-dimensional blood vessel model under the preset mixed reality display configuration, so as to superimpose the display on the patient's target site in the operator's field of vision, and assist the operator in performing blood vessel puncture operation.

[0130] In one exemplary implementation, the sensitive wavelength range for the camera component to acquire color images is specifically 400nm-700nm, and the sensitive wavelength range for acquiring near-infrared images is specifically 700nm-900nm.

[0131] The light source components specifically include a visible light source, an 850nm wavelength narrowband near-infrared LED array, and a 760nm wavelength narrowband near-infrared LED array;

[0132] The light source component is coaxial with the lens of the camera component, or the light source component is arranged around the lens of the camera component to directly illuminate the target area of ​​the patient in a vertical or tilted manner below a threshold, providing uniform illumination conditions.

[0133] In one exemplary implementation, the camera component adopts a dual-camera structure, involving a first camera for acquiring color images and a second camera for acquiring near-infrared images, with color images and near-infrared images being acquired simultaneously.

[0134] Alternatively, the camera components can alternately capture different images at extremely high frame rates, exceeding a threshold.

[0135] In one exemplary implementation, the biomechanical model is specifically a finite element model or an equivalent elastic model. For different patient target sites, it is divided into skin, subcutaneous fat, muscle tissue, and blood vessels as different layers or different material regions, and is configured with corresponding physical property parameters. The model boundary conditions assume that the distal end is fixed and the proximal end has a slight periodic displacement with the heartbeat.

[0136] In one exemplary implementation, the biomechanics module is used to calculate compensation values ​​for vascular stress deformation and motion displacement based on a pre-defined biomechanical model of the patient's target site, combined with contact pressure and three-dimensional posture data, and taking into account the influence of heartbeat. This includes the following processing:

[0137] 1) The contact pressure collected by the MEMS pressure sensor is used to estimate the normal stress on the skin surface. After the model applies surface stress, it outputs the corresponding tissue compression deformation field.

[0138] 2) Based on the three-dimensional posture data collected by the inertial measurement unit and the pose of the glasses device, determine whether the patient's target part has been translated or rotated as a whole. If so, apply the same rigid body motion to the model coordinate system.

[0139] 3) To account for the effects of heartbeat, periodic pulse pressure is applied in the artery. The periodic pulse pressure is obtained from a pre-set simple harmonic internal pressure source in the model. The amplitude corresponds to the systolic-diastolic pressure difference, and the frequency is 1Hz.

[0140] 4) The model integrates different inputs and calculates the changes in the position of the blood vessel center and the changes in the blood vessel radius in three-dimensional space through finite element solution or analytical approximation. During the calculation process, on the one hand, analytical formulas are used for rapid estimation, and on the other hand, more refined calculations are performed asynchronously in the background, so as to continuously correct the calculation results.

[0141] In one exemplary implementation, the glasses device and camera assembly are integrated into a single design to solidify their relative positions, thereby avoiding the need to recalibrate before subsequent image acquisition and processing after the initial calibration process.

[0142] Alternatively, calibration can be performed by setting visual markers on the target areas of the patient before formally acquiring multispectral images.

[0143] In one exemplary implementation, in the corresponding display screen of the virtual three-dimensional blood vessel model, veins are displayed as blue semi-transparent virtual lines, and arteries are displayed as red dashed semi-transparent lines. The display effects are used to indicate the depth information of the blood vessels, wherein the display effects include at least one of line thickness, transparency, and brightness variations.

[0144] In one exemplary implementation, the corresponding display screen of the virtual three-dimensional blood vessel model also includes navigation information such as the optimal puncture vessel, the optimal puncture point, needle trajectory guidance, needle position, needle insertion angle, needle insertion depth, needle insertion path, pressure reading, whether the puncture was successful, and simulated arterial pulsation. The needle is equipped with image-recognizable markers or corresponding sensors to assist in capturing the needle position, needle insertion angle, needle insertion depth, and needle insertion path.

[0145] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the MR arteriovenous dual-mode vascular puncture navigation method described above can be found in, for example... Figure 1 The description of the MR arteriovenous dual-mode vascular puncture navigation system in the corresponding embodiment will not be repeated here.

[0146] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0147] Therefore, this application provides a computer-readable storage medium storing a plurality of instructions that can be loaded by a processor to execute the steps of the MR arteriovenous dual-mode vascular puncture navigation method described above. For specific operation, please refer to the description of the MR arteriovenous dual-mode vascular puncture navigation method described above, which will not be repeated here.

[0148] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0149] Since the instructions stored in the computer-readable storage medium can execute the steps of the MR arteriovenous dual-mode vascular puncture navigation method described above, the beneficial effects of the MR arteriovenous dual-mode vascular puncture navigation method described above can be achieved, as detailed in the preceding description, and will not be repeated here.

[0150] The above provides a detailed description of the MR arteriovenous dual-mode vascular puncture navigation system, method, and computer-readable storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An MR dual-mode arteriovenous vascular puncture navigation system, characterized in that, The MR arteriovenous dual-mode vascular puncture navigation system includes a multispectral imaging unit, a force sensing unit, a computing processing unit, and a mixed reality display unit. The multispectral imaging unit includes a camera assembly and a light source assembly. The camera assembly acquires multispectral images of the target area of ​​the patient under uniform illumination provided by the light source assembly. The multispectral images include color images and near-infrared images. The force sensing unit includes a microelectromechanical system (MEMS) pressure sensor and an inertial measurement unit. The MEMS pressure sensor collects the contact pressure between the probe and the patient's contact area, and the inertial measurement unit is used to collect three-dimensional attitude data. The computational processing unit includes an image processing module, a blood vessel recognition module, a biomechanical module, and a navigation control module. The image processing module is used to preprocess the multispectral image to obtain a first blood vessel image. The blood vessel recognition module is used to distinguish between arteries and veins in the first blood vessel image using multi-wavelength difference analysis and pulsation signals, thereby generating a second blood vessel image with blood vessel markings. The biomechanical module is used to calculate compensation values ​​for blood vessel stress deformation and motion displacement based on a preset biomechanical model of the patient's target site, combined with the contact pressure, the three-dimensional posture data, and the heartbeat. The navigation control module is used to update the blood vessel position by combining the second blood vessel image and the compensation values, and generate a corresponding virtual three-dimensional blood vessel model. The mixed reality display unit includes glasses, which display the corresponding display screen of the virtual three-dimensional blood vessel model under a preset mixed reality display configuration, so as to overlay the target site of the patient in the operator's field of vision, assisting the operator in performing blood vessel puncture. The biomechanical module is used to calculate compensation values ​​for vascular deformation and displacement based on a pre-set biomechanical model of the target site on the patient, combined with the contact pressure, the three-dimensional posture data, and the heartbeat. This includes the following processing steps: 1) The contact pressure collected by the microelectromechanical system pressure sensor is used to estimate the normal stress on the skin surface. After the model applies the normal stress, it outputs the corresponding tissue compression deformation field. 2) Based on the three-dimensional posture data of the inertial measurement unit and the pose of the glasses device, determine whether the patient's target part has been translated or rotated as a whole. If so, apply the same rigid body motion to the model coordinate system. 3) For the heartbeat, periodic pulse pressure is applied in the artery. The periodic pulse pressure is obtained by a simple harmonic internal pressure source preset by the model. The amplitude corresponds to the systolic pressure-diastolic pressure difference and the frequency is 1Hz. 4) The model integrates the above contact pressure, three-dimensional posture data, glasses device pose and heartbeat, and calculates the changes in the position of the blood vessel center and the changes in the blood vessel radius in three-dimensional space through finite element solution or analytical approximation. During the calculation process, on the one hand, analytical formulas are used for rapid estimation, and on the other hand, fine calculations are performed asynchronously in the background, so as to continuously correct the calculation results.

2. The MR arteriovenous dual-mode vascular puncture navigation system according to claim 1, characterized in that, The specific sensitive wavelength range for the camera component to acquire the color image is 400nm-700nm, and the specific sensitive wavelength range for the near-infrared image is 700nm-900nm. The light source assembly specifically includes a visible light source, an 850nm wavelength narrowband near-infrared LED array, and a 760nm wavelength narrowband near-infrared LED array; The light source assembly is coaxial with the lens of the camera assembly, or the light source assembly is arranged around the lens of the camera assembly to directly illuminate the patient's target area in a vertical or tilted manner below a threshold, providing the uniform illumination conditions.

3. The MR arteriovenous dual-mode vascular puncture navigation system according to claim 2, characterized in that, The camera assembly adopts a dual-camera structure, involving a first camera for capturing the color image and a second camera for capturing the near-infrared image, wherein the color image and the near-infrared image are captured simultaneously; Alternatively, the camera assembly may alternately acquire the color image and the near-infrared image while the frame rate is above a threshold.

4. The MR arteriovenous dual-mode vascular puncture navigation system according to claim 1, characterized in that, The biomechanical model is specifically a finite element model. For different target areas of patients, it is divided into skin, subcutaneous fat, muscle tissue and blood vessels as different layers or different material regions, and is configured with corresponding physical property parameters. The model boundary conditions assume that the distal end is fixed and the proximal end has a slight periodic displacement with the heartbeat.

5. The MR arteriovenous dual-mode vascular puncture navigation system according to claim 1, characterized in that, The glasses device and the camera assembly are integrated into a single design, thus fixing their relative positions. Alternatively, calibration can be performed by setting visual markers on the target area of ​​the patient before the actual acquisition of the multispectral images.

6. The MR arteriovenous dual-mode vascular puncture navigation system according to claim 1, characterized in that, In the corresponding display screen of the virtual three-dimensional blood vessel model, veins are displayed as blue semi-transparent virtual lines, and arteries are displayed as red dashed semi-transparent lines. The display effects are used to indicate the depth information of the blood vessels, including at least one of line thickness, transparency, and brightness variations.

7. The MR arteriovenous dual-mode vascular puncture navigation system according to claim 6, characterized in that, The corresponding display screen of the virtual three-dimensional blood vessel model also includes navigation information such as the optimal puncture vessel, the optimal puncture point, needle trajectory guidance, needle position, needle insertion angle, needle insertion depth, needle insertion path, pressure reading, whether the puncture was successful, and simulated arterial pulsation. The needle is equipped with image-recognizable markers to assist in capturing the needle position, the needle insertion angle, the needle insertion depth, and the needle insertion path.

Images