Puncture path planning system and equipment adopting puncture path planning system

Abstract

The invention relates to a puncture path planning system and equipment adopting the puncture path planning system, which directly solve the difficulty of peripheral artery puncture and laterally solve the defects of some deep vein catheterization and vasostomy, and the puncture path planning system comprises at least one memory in which a computer program is stored, the at least one processor is configured to implement the puncture path planning method when executing a computer program, and the method comprises the following specific steps: generating a path leading to a target position from different starting points and angles of a blood vessel by adopting a preset geometric algorithm based on a preset safety margin constraint, blood vessel feature constraint information and surrounding tissue constraint information; the generated puncture needle inserting path is integrated into a blood vessel 3D reconstruction model to be displayed in different colors or line styles. The method has the advantages that the blood vessel model is accurately constructed, the puncture path is safely generated, and catheterization of the peripheral artery indwelling needle is more efficient.

Description

Technical Field

[0001] The present invention relates to the technical field of medical auxiliary equipment, and in particular to a puncture path planning system and equipment using the puncture path planning system. Background Art

[0002] In the medical field, for specific patient groups, especially those who need treatments such as plasma exchange, hemodialysis, and blood purification but have poor peripheral venous conditions, arterial catheterization has become a key vascular access option. Arterial catheterization has many advantages, such as relatively small damage, no need for long-term retention, and can be inserted when needed and removed after use, which is very convenient. At the same time, because arterial blood pressure is higher than venous blood pressure, it can provide sufficient pressure for blood purification, which is better than venous catheterization in this respect.

[0003] Currently, the main methods include deep vein catheterization and vascular fistula. Deep vein catheterization provides patients with long-term vascular access by inserting a catheter into a deep vein, such as the internal jugular vein and subclavian vein. Deep vein catheterization is mainly an alternative to deep veins (without peripherals) for patients with poor peripheral venous conditions. However, deep vein catheterization is expensive, requires high operation requirements, and is time-consuming and troublesome. If it is only used 3 to 5 times, it is very uneconomical. At the same time, deep vein catheterization requires special care to prevent thrombosis, infection, and over-heparinization, which requires too much care in general wards.

[0004] Vascular fistula is a surgical procedure that connects arteries and veins to form a long-term vascular access. It is mainly suitable for patients with renal failure and long-term hemodialysis. Specifically, the distal end of the brachial artery and the cephalic vein are anastomosed to form an arteriovenous short circuit to facilitate dialysis for patients. However, this method is constructed surgically, and the preparation time is at least 2 months, and there are many complications. It cannot be quickly deployed in an emergency, and the cost is also high.

[0005] The disadvantages of deep venous catheterization and vascular fistula are mainly reflected in the following aspects: high cost, which is a heavy economic burden for some patients who only need short-term use of vascular access; complex operation, high technical requirements for medical staff, and increased medical risks; special care is required, which increases the workload and difficulty of nursing; and limited applicability to emergency situations. Summary of the invention

[0006] In order to accurately construct a vascular model, safely generate a puncture path, and help make peripheral arterial catheter placement more efficient, the present application provides a puncture path planning system and a device using the puncture path planning system.

[0007] In a first aspect, the present application provides a puncture path planning system, which adopts the following technical solution: A puncture path planning system, comprising: B-ultrasound equipment; at least one processor; At least one memory storing a computer program, and at least one processor configured to implement a puncture path planning method when executing the computer program, the specific steps are as follows: Acquire a multi-angle B-ultrasound image of a target blood vessel region acquired by a probe of a B-ultrasound device, wherein the target blood vessel region involves at least one of a radial artery, a dorsalis pedis artery, and a brachial artery; For each acquired B-ultrasound image, the feature-based image matching algorithm finds the two-dimensional coordinates of the matching feature points, and then converts these two-dimensional coordinates into three-dimensional coordinates in a unified coordinate system according to the spatial position and angle information of the probe when acquiring the image. The edge detection algorithm is used to extract the vascular boundary information from the acquired B-ultrasound image, and the vascular contour is preliminarily constructed in combination with the converted three-dimensional coordinates of the feature points. Finally, the acquired B-ultrasound image is subjected to grayscale analysis to obtain the internal structure information of the blood vessels, and the preliminarily constructed vascular contour is refined to form a 3D reconstruction model of the blood vessels. Performing a vascular feature analysis based on the 3D vascular reconstruction model to obtain key vascular feature data, and simultaneously performing a surrounding tissue analysis based on the 3D vascular reconstruction model to obtain key surrounding tissue feature data; Generate vascular feature constraint information based on the acquired vascular key feature data and a preset vascular feature special evaluation method, and simultaneously generate surrounding tissue constraint information based on the acquired surrounding tissue key feature data and a preset surrounding tissue feature special evaluation method; Based on the preset safety margin constraints, vascular feature constraint information, and surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to the target location from different starting points and angles of the blood vessel; The generated puncture needle path is integrated into the 3D reconstruction model of the blood vessel for display and presented in different colors or line styles.

[0008] By adopting the above technical solution, firstly, through multi-angle B-ultrasound image acquisition and advanced algorithm processing, a 3D reconstruction model of blood vessels can be accurately constructed to provide detailed blood vessel and surrounding tissue information for arterial catheterization. Secondly, the generated vascular feature constraint information and surrounding tissue constraint information, combined with the safety margin constraint, can ensure the safety of the puncture path and reduce the risk of damage to surrounding tissues. Furthermore, a preset geometric algorithm is used to generate paths from different starting points and angles, and displayed in a 3D reconstruction model, so that doctors can intuitively select the best puncture path and improve the success rate of arterial catheterization. Finally, this method is relatively simple to operate, low in cost, and relatively low in nursing requirements for ordinary wards. It provides a better choice for arterial catheterization for patients who need plasma exchange, hemodialysis, blood purification and other treatments but have poor peripheral venous conditions.

[0009] Optionally, obtaining a multi-angle B-ultrasound image of a target blood vessel region acquired by a probe of a B-ultrasound device includes: Start the probe of the B-ultrasound device to perform longitudinal and transverse section scanning on the target area on the skin surface to obtain a scanned B-ultrasound image; Scanning the probe of the B-ultrasound device at different angles in the longitudinal and transverse directions to obtain a scanned B-ultrasound image; The B-ultrasound images acquired at different scanning angles are summarized as multi-angle B-ultrasound images of the target blood vessel region.

[0010] By adopting the above technical solutions, longitudinal and transverse section scanning combined with scanning at different angles can capture the target blood vessel area information in all directions. It can clearly present the direction, structure and relationship of blood vessels with surrounding tissues. Summarizing images at different angles provides rich data for subsequent processing, helps to build a more accurate 3D model of blood vessels and improve the accuracy of puncture path planning. It enables doctors to understand blood vessel conditions more intuitively, reduce puncture risks, reduce patient pain, improve the safety and success rate of medical operations, and provide strong support for clinical diagnosis and treatment.

[0011] Optionally, the method further includes the following steps after the collected B-ultrasound image is subjected to grayscale analysis to obtain the internal structure information of the blood vessel and before the initially constructed blood vessel contour is refined to form a 3D reconstruction model of the blood vessel: A spatial positioning module preset in the probe is used to record the specific position and angle information of the probe when collecting each image, and the specific position and angle information when collecting each image is defined as spatial positioning information; The relative position relationship of the blood vessel boundaries in different images is determined based on the spatial positioning information, and the edge detection algorithm is used to refine the blood vessel boundaries. At the same time, the spatial positioning information is used to enhance the feature points found by the feature-based image matching algorithm. The three-dimensional model of the blood vessel is refined according to the refined blood vessel boundaries, enhanced feature points and spatial positioning information.

[0012] By adopting the above technical solution, the spatial positioning module records the position and angle information, and can accurately determine the relative position of the blood vessel boundaries in different images. The edge detection algorithm refines the blood vessel boundaries and enhances the feature points, and further refines the vascular three-dimensional model in combination with the spatial positioning information. This makes the constructed vascular 3D reconstruction model more accurate and helps to clearly present the internal structure and spatial relationship of the blood vessel. It provides a more reliable model foundation for subsequent puncture path planning, improves puncture accuracy and safety, reduces medical risks caused by inaccurate models, and improves the quality and efficiency of medical operations.

[0013] Optionally, refining the initially constructed vascular contour to form a 3D vascular reconstruction model includes: acquiring infrared imaging data and magnetic resonance imaging data; Extracting blood vessel related features from infrared imaging and extracting blood vessel morphological features from magnetic resonance imaging, wherein the blood vessel related features include temperature features and the morphological features include length, diameter, curvature and branching angle of the blood vessel; Establish a feature fusion matrix to correspond and combine the features extracted from infrared imaging, the features extracted from magnetic resonance imaging, and the internal structure information of blood vessels obtained from grayscale analysis of B-ultrasound images to form fusion features; The blood vessel adjustment parameters are obtained based on the fusion feature analysis, and the blood vessel three-dimensional model is adjusted according to the blood vessel adjustment parameters and the blood vessel internal structure information obtained by grayscale analysis. At the same time, the geometric shape of the blood vessel three-dimensional model is adjusted using the morphological features in the fusion features.

[0014] By adopting the above technical solutions, combining infrared imaging and magnetic resonance imaging data, extracting and fusing multi-dimensional features, the state of blood vessels can be more comprehensively reflected. By adjusting parameters and using morphological information to adjust the model geometry, the model can be made more accurate. It helps doctors to accurately understand the condition of blood vessels and provide a more reliable basis for puncture path planning. It improves puncture accuracy and safety, reduces operational risks caused by inaccurate models, improves medical quality, provides strong support for clinical diagnosis and treatment, and brings better treatment effects and experience to patients.

[0015] Optionally, based on preset safety margin constraints, vascular feature constraint information, and surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to a target location from different starting points and angles of the blood vessel, including: Collect the patient's current body position information and information related to physiological indicators, and collect multimodal imaging data of the patient in different body positions and physiological states, where the multimodal imaging data includes infrared imaging data, magnetic resonance imaging data, and B-ultrasound image data; body position information includes supine, lateral, prone, and semi-sitting positions; and physiological indicator information includes heart rate, blood pressure, respiratory rate, and blood oxygen saturation; The collected patient's current body position information, physiological index-related information and corresponding multimodal imaging data are input into the trained vascular feature analysis model to extract key features of blood vessels and surrounding tissues; Generate the current blood vessel feature constraint information according to the extracted blood vessel key feature data, and simultaneously generate the current surrounding tissue constraint information according to the acquired surrounding tissue key feature data; Based on the preset safety margin constraint, the current vascular feature constraint information, and the current surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to the target position from different starting points and angles of the blood vessel.

[0016] By adopting the above technical solutions, multimodal imaging data and patient posture and physiological index information are collected to fully understand the patient's condition. The data is input into the trained model to extract features and generate constraint information, and the puncture path is planned in combination with the geometric algorithm, which improves the accuracy and safety of the path. Different postures and physiological states are considered to make the path planning more in line with the actual situation and reduce the risk of puncture. Providing doctors with scientific puncture plans will help improve the success rate of puncture and subsequent treatment, reduce patient pain and complications, improve medical quality and efficiency, and provide strong technical support for clinical puncture operations.

[0017] Optionally, generating a path to a target location includes: Obtain different consideration angles, including safety priority, efficiency priority and patient comfort; According to the mapping relationship between the consideration angle and the genetic algorithm adjustment scheme, the genetic algorithm adjustment scheme under different consideration angles is analyzed and determined, wherein the genetic algorithm adjustment scheme includes adjusting the chromosome encoding method, the fitness function and the genetic algorithm parameter adjustment; For different consideration angles, the genetic algorithm is run separately to obtain the optimal puncture path solutions for different angles.

[0018] By adopting the above technical solutions, different consideration angles can be obtained to meet diverse needs. By adjusting the solution through genetic algorithms, accurate optimization can be performed for safety, efficiency and patient comfort. The genetic algorithms are run separately to obtain the optimal path solutions at different angles, providing doctors with a variety of options. The safety, efficiency and patient experience of puncture are improved, making path planning more scientific and reasonable.

[0019] Optionally, the method further includes steps after integrating the generated puncture needle path into the 3D reconstructed model of the blood vessel for display and presenting it in different colors or line styles, as follows: Obtain the patient's physiological indicators, the position and angle of the puncture needle, and the status of the blood vessels during the puncture process; The patient's physiological index curve is displayed on the operation interface according to the patient's physiological index obtained during the puncture process, and the position and path of the puncture needle are displayed in real time using a 3D model, and the result information of the puncture needle executing the preset puncture path plan is analyzed and displayed.

[0020] By adopting the above technical solution, the patient's physiological indicators, puncture needle position angle and blood vessel status information can be obtained in real time, and the puncture situation can be timely understood. The physiological indicator curve and the puncture needle position path and result information in the 3D model are displayed on the operation interface, which is convenient for doctors to monitor intuitively. It can improve the accuracy and safety of puncture and adjust the plan in time.

[0021] Optionally, the method further includes displaying the patient's physiological index curve on the operation interface according to the patient's physiological index obtained during the puncture process, and simultaneously using the 3D model to display the position and path of the puncture needle in real time, and analyzing the result information of the puncture needle executing the preset puncture path plan for display, which are as follows: Analyze whether to detect and obtain information related to the failure of equipment components involved in the puncture process; If not, maintain the original setting; If yes, an alarm is sounded, and the type and location of the faulty equipment are determined and displayed based on the acquired fault-related information.

[0022] By adopting the above technical solution, the safety of the puncture process can be effectively guaranteed. The equipment component failure information is analyzed in real time, and if no failure is detected, the operation is ensured to proceed smoothly. If there is a failure, an alarm is issued in time and the type and location of the faulty equipment are determined for quick processing. Medical risks caused by equipment failure can be avoided and the success rate of puncture can be improved.

[0023] In a second aspect, the present application provides a device using a puncture path planning system, which adopts the following technical solution: A device using a puncture path planning system, comprising: An acquisition module, used for acquiring multi-angle B-ultrasound images; A memory, used to store a program of a puncture path planning system as described in the first aspect; A processor, wherein the program in the memory can be loaded and executed by the processor to implement the puncture path planning method of the puncture path planning system as described in the first aspect; Medical puncture instruments; The driving device is controlled by the processor to receive the program of the puncture path planning method and drive the medical puncture instrument to perform puncture according to the puncture path. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of a puncture path planning method in a puncture path planning system in an embodiment of the present application.

[0025] Figure 2 It is a flowchart of another embodiment of the present application for obtaining multi-angle B-ultrasound images of a target blood vessel area collected by a probe of a B-ultrasound device.

[0026] Figure 3 It is a flowchart of another embodiment of the present application, which is located after the grayscale analysis of the collected B-ultrasound image is finally performed to obtain the internal structure information of the blood vessel, and before the preliminary constructed blood vessel contour is refined to form a 3D reconstruction model of the blood vessel.

[0027] Figure 4 It is a schematic diagram of a process of refining a preliminarily constructed blood vessel contour to form a 3D reconstructed model of the blood vessel according to another embodiment of the present application. DETAILED DESCRIPTION

[0028] The present application is further described in detail below in conjunction with the accompanying drawings.

[0029] Reference Figure 1 , is a puncture path planning system disclosed in the present application, comprising a B-ultrasound device, at least one processor and at least one memory storing a computer program, wherein the at least one processor is configured to implement a puncture path planning method when executing the computer program, and the specific steps are as follows: Step S100, obtaining a multi-angle B-ultrasound image of a target blood vessel region acquired by a probe of a B-ultrasound device.

[0030] The target blood vessel region involves at least one of the radial artery, the dorsalis pedis artery and the brachial artery.

[0031] Among them, the target blood vessel area refers to the area of ​​the body where the radial artery, dorsalis pedis artery and brachial artery are located in the part where the puncture path planning is required. B-ultrasound image: an image generated by the principle of ultrasonic reflection, which can show the structure of internal tissues of the human body.

[0032] The acquisition method is as follows: the probe of the B-ultrasound device is placed near the target blood vessel area, and the ultrasound is emitted and the reflected signal is received, and the B-ultrasound image is generated after being processed by the device. The probe can be moved from different angles to obtain B-ultrasound images from multiple angles.

[0033] For example: when planning the puncture path for a patient's arm, the probe is placed at a specific position on the patient's arm, and scans are performed from different angles such as the front and side to obtain multiple B-ultrasound images, which may show the position and morphology of the brachial artery.

[0034] Step S200: for each acquired B-ultrasound image, the two-dimensional coordinates of the matching feature points are found by a feature-based image matching algorithm. Then, according to the spatial position and angle information of the probe when acquiring the image, these two-dimensional coordinates are converted into three-dimensional coordinates in a unified coordinate system. The edge detection algorithm is used to extract the vascular boundary information from the acquired B-ultrasound image, and the vascular contour is preliminarily constructed in combination with the converted three-dimensional coordinates of the feature points. Finally, the acquired B-ultrasound image is subjected to grayscale analysis to obtain the internal structure information of the blood vessel, and the preliminarily constructed vascular contour is refined to form a 3D reconstruction model of the blood vessel.

[0035] Among them, the feature image matching algorithm: an algorithm that determines the correspondence between images by finding specific feature points in the image and comparing the similarity of these feature points in different images. Two-dimensional coordinates: two values ​​that determine the position of a point on a plane. Spatial position and angle information: information such as the position of the probe in three-dimensional space (such as X, Y, Z coordinates) and the tilt angle of the probe.

[0036] Unified coordinate system: A common coordinate system established to integrate information from different images. Three-dimensional coordinates: Three values ​​that determine the position of a point in three-dimensional space.

[0037] In the process of converting two-dimensional coordinates into three-dimensional coordinates in a unified coordinate system, the following steps can be followed to determine the initial coordinates and realize the three-dimensional coordinate conversion: 1. Determine the initial coordinates: For each acquired B-ultrasound image, establish an image coordinate system. Usually, the upper left corner of the image is taken as the origin, the horizontal right is the positive direction of the x-axis, and the vertical downward is the positive direction of the y-axis. The two-dimensional coordinates of the matching feature points found by the feature-based image matching algorithm are the coordinates in this image coordinate system. These coordinates can be used as the initial coordinates for the conversion. 2. Use the probe spatial position and angle information: Determine the probe's spatial position information: The probe's position coordinates (Xp, Yp, Zp) in three-dimensional space can be obtained through the probe's built-in positioning sensor or external positioning device. This position coordinate is usually determined in a unified world coordinate system. The world coordinate system can be defined according to actual needs, for example, a rectangular coordinate system is established with a fixed point as the origin. Determine the probe's angle information: The probe's angle information includes pitch, yaw, and roll. These angles can be measured by sensors or entered manually. The pitch angle indicates the angle of rotation of the probe around the x-axis, the yaw angle indicates the angle of rotation of the probe around the y-axis, and the roll angle indicates the angle of rotation of the probe around the z-axis. 3. Coordinate conversion process: First, convert the angle information of the probe into a rotation matrix. The rotation matrix R can be calculated based on the pitch angle, yaw angle, and roll angle using the formula of the rotation matrix. The rotation matrix R is a 3x3 matrix used to describe the rotation of the probe in three-dimensional space. Then, the initial two-dimensional coordinates (x, y) are expanded to three-dimensional coordinates (x, y, 0). This is because in the two-dimensional image coordinate system, there are only two coordinates, x and y, while three coordinates are required in three-dimensional space. Set the z coordinate to 0, indicating that the initial point is on the probe plane. Next, multiply the expanded three-dimensional coordinates by the rotation matrix R to obtain the rotated coordinates. Let the rotated coordinates be (x', y', z'), then (x', y', z') = (x, y, 0) × R. Finally, add the rotated coordinates to the position coordinates of the probe to obtain the three-dimensional coordinates in the unified coordinate system. That is (X, Y, Z) = (x', y', z') + (Xp, Yp, Zp).

[0038] Edge detection algorithm: an algorithm used to detect the edge of an object in an image, which determines the boundary of the object by analyzing the change in pixel values ​​in the image. Grayscale analysis: a method of analyzing based on the grayscale value of pixels in an image, which reflects the brightness of the pixels. Vascular 3D reconstruction model: a 3D model of blood vessels is constructed by processing multiple B-ultrasound images, which can more intuitively display the morphology and structure of blood vessels.

[0039] First, for each acquired B-ultrasound image, the feature image matching algorithm is used to find the two-dimensional coordinates of the matching feature points, and then these two-dimensional coordinates are converted into three-dimensional coordinates in a unified coordinate system according to the spatial position and angle information of the probe when the image is acquired. Next, the edge detection algorithm is used to extract the vascular boundary information from the B-ultrasound image, and the vascular contour is reconstructed in combination with the converted three-dimensional coordinates of the feature points. Finally, the grayscale analysis of the B-ultrasound image is used to obtain the internal structure information of the blood vessel, and the three-dimensional model of the blood vessel is refined to form a 3D reconstruction model of the blood vessel. Among them, the internal features of the blood vessel include information such as the inner diameter and arterial plaque.

[0040] For example, in a B-ultrasound image, some obvious feature points, such as specific spots on the blood vessel wall, are found through the feature image matching algorithm. Then, the two-dimensional coordinates of these feature points are converted into three-dimensional coordinates based on the position and angle information of the probe. The edge detection algorithm can be used to determine the edge contour of the blood vessel, and then combined with the three-dimensional coordinates of the feature points to construct a preliminary three-dimensional model of the blood vessel. Through grayscale analysis, the grayscale distribution inside the blood vessel can be understood, and the three-dimensional model can be further refined, such as determining whether there are plaques in the blood vessel.

[0041] Step S300 , performing a vascular feature analysis based on the 3D reconstructed model of the blood vessel to obtain key vascular feature data, and simultaneously performing a surrounding tissue analysis based on the 3D reconstructed model of the blood vessel to obtain key surrounding tissue feature data.

[0042] Among them, blood vessel key feature data: numerical values ​​describing important features of blood vessels, such as blood vessel diameter, length, curvature, etc. Surrounding tissue key feature data: numerical values ​​describing important features of tissue surrounding blood vessels, such as the hardness of surrounding tissue, distance from blood vessels, etc. The surrounding tissue is mainly skin tissue.

[0043] The key feature data of blood vessels can be obtained in the following ways: First, use the 3D reconstruction model of blood vessels to measure the blood vessels in the model through measurement tools or algorithms. For example, the diameter of the blood vessels can be measured at different locations in the model to calculate the total length of the blood vessels and the length of specific sections. For the degree of curvature of the blood vessels, the bending angle or curvature can be calculated by analyzing the direction of the blood vessels in the model. The thickness of the blood vessel wall can also be determined by analyzing the grayscale information. For example, the area with large grayscale changes at the boundary of the blood vessel may correspond to the blood vessel wall, and the thickness of the blood vessel wall can be estimated by measuring the width of these areas.

[0044] The key feature data of the surrounding tissue can be obtained in the following way: Based on the 3D reconstruction model of the blood vessels, the scope of the tissue around the blood vessels is determined. Different tissues can be distinguished by identifying different grayscale values ​​or texture features. For the hardness of the surrounding tissue, other medical examination data, such as the results of elastic imaging technology, can be combined, or the hardness of specific tissues can be evaluated based on clinical experience. To measure the distance between the surrounding tissue and the blood vessels, the boundaries of the tissue and the boundaries of the blood vessels can be determined in the model, and then the shortest distance between the two can be calculated. If there are important organs or nerves and other structures around, the positions of these structures in the model can be marked and their distance from the blood vessels can be measured.

[0045] Suppose we are planning a puncture path for a patient's arm, and the target blood vessel is the radial artery. The relevant information for obtaining the key feature data of the blood vessel is as follows: In the 3D reconstruction model of the blood vessel, the diameter of the radial artery is measured to be 2.5 mm, 2.8 mm, etc. at different locations. By analyzing the direction of the blood vessels in the model, it is found that there is a bend angle of about 30 degrees. According to the grayscale analysis, the thickness of the blood vessel wall is estimated to be about 0.5 mm. The relevant information for obtaining the key feature data of the surrounding tissue is as follows: It is determined that the surrounding tissues are mainly muscles and tendons. Through elastic imaging technology, it is known that the hardness of muscle tissue is moderate, while the tendon is relatively hard. The shortest distance between the muscle tissue and the radial artery is measured to be 1.5 mm, and the shortest distance between the tendon and the radial artery is 0.8 mm. At the same time, the nearby nerve structure is marked in the model, and the distance between the nerve and the radial artery is measured to be 3 mm.

[0046] Step S400, generating vascular feature constraint information based on the acquired vascular key feature data and a preset vascular feature special evaluation method, and generating surrounding tissue constraint information based on the acquired surrounding tissue key feature data and a preset surrounding tissue feature special evaluation method.

[0047] Among them, the vascular feature constraint information is the restriction conditions on the puncture path determined based on the key feature data of the vascular. For example, if the blood vessel is too thin, the size of the puncture needle may be limited; if the blood vessel is very curved, the puncture angle may need to be adjusted to avoid damaging the blood vessel wall.

[0048] Vascular Characteristic Specific Assessment Methodology: A systematic process for quantitative analysis and assessment of key vascular characteristic data to determine the constraints imposed on the vessels during specific medical procedures.

[0049] The specific steps and quantitative analysis process of the special evaluation method for vascular characteristics are as follows: 1. Data collection and collation: Through high-precision medical imaging technology, such as high-resolution ultrasound, CT angiography, magnetic resonance angiography, etc., the key characteristic data of the patient's blood vessels are obtained. These data are accurately measured and recorded to ensure the accuracy and reliability of the data. 2. Establish a quantitative evaluation index system: Determine the specific quantitative indicators and safety ranges for evaluating vascular characteristics. For vascular diameter, set puncture needle size limits corresponding to different diameter ranges. For example, when the vascular diameter is 2-3 mm, the puncture needle diameter must not exceed 1 mm; when the diameter is 3-4 mm, the puncture needle diameter can be between 1-1.2 mm. For vascular curvature, determine the safe range of the puncture angle based on the curvature angle. For example, when the vascular curvature is 30-45 degrees, the puncture angle should be between 10-20 degrees with the direction of the blood vessel; when the curvature is 45-60 degrees, the angle is between 5-15 degrees. For wall thickness, combine mechanical analysis to determine the upper limit of the operating force. For example, when the wall thickness is 0.5-0.8 mm, the operating force must not exceed X Newtons; when the wall thickness is 0.8-1 mm, the operating force can be between XY Newtons. 3. Data analysis and processing: Use mathematical models and statistical methods to analyze the collected quantitative key feature data of blood vessels. For the curvature of the blood vessel, the optimal path angle of the puncture needle in blood vessels with different curvatures can be calculated by trigonometric functions. For example, if the curvature angle of the blood vessel is known to be α, the angle β between the puncture needle and the direction of the blood vessel can be calculated by the formula β=arcsin (sinα / ratio of the puncture needle length to the blood vessel radius). For wall thickness and operating force, a mechanical model can be established to calculate the safe operating force range under different wall thicknesses based on parameters such as the elastic modulus and Poisson's ratio of the wall. 4. Generate vascular feature constraint information: Based on the results of quantitative data analysis, determine specific constraint information, including precise limits on puncture needle size, puncture angle, operating force, etc. For example, if the blood vessel diameter is measured to be 2.8 mm, the curvature is 40 degrees, and the wall thickness is 0.7 mm. According to the evaluation index system, the generated vascular characteristic constraint information is: the diameter of the puncture needle shall not exceed 1 mm, the angle between the puncture angle and the direction of the blood vessel should be between 12-18 degrees, and the operating force shall not exceed Z Newtons.

[0050] Key characteristic data of surrounding tissues: including the type of surrounding tissues, hardness value (measured in specific units, such as Pascals), distance from blood vessels (in millimeters), level of importance of physiological functions, etc.

[0051] Specialized assessment method for surrounding tissue characteristics: A method for determining the specific constraints on surrounding tissues during medical operations by quantitatively analyzing and comprehensively evaluating the key characteristic data of surrounding tissues.

[0052] Surrounding tissue restraint information: Clearly stipulates the specific restrictions and protection measures that should be taken for surrounding tissues during medical operations to ensure safe operations.

[0053] The specific steps and quantitative analysis process of the special assessment method for peripheral tissue characteristics are as follows: 1. Data collection and quantification: Use high-resolution medical imaging technology (such as magnetic resonance elastic imaging to measure tissue hardness, three-dimensional CT to accurately measure distance, etc.) to obtain key characteristic data of peripheral tissues. For example, peripheral tissues are divided into muscle tissue, nerve tissue, bone tissue and other types, and different codes are assigned to each type. Measure tissue hardness. Muscle tissue hardness may be between 1000-5000 Pascals. Nerve tissue is relatively soft and may be between 500-1000 Pascals. Bone tissue is very hard and can be as high as tens of thousands of Pascals. Measure the distance between tissue and blood vessels, accurate to millimeter level. Quantitatively rate the importance of the physiological function of tissues. For example, important nerve tissue can be rated as level 5 (the highest level), tissues surrounding important organs can be rated as level 4, general muscle tissue can be rated as level 2-3, and tissues such as fat can be rated as level 1. 2. Risk assessment and determination of quantitative indicators: Determine quantitative risk assessment indicators based on factors such as tissue type, hardness, distance and importance level. For example, the risk value calculation formula is set as follows: risk value = tissue hardness value × inverse of distance × importance level coefficient. Among them, the inverse of distance can make the closer the distance, the higher the risk, and the importance level coefficient can be set to different values ​​according to different levels, such as 0.8 for level 5 importance level coefficient and 0.6 for level 4. According to the calculated risk value, different risk level ranges are divided, such as high risk (risk value greater than or equal to 1000), medium risk (500-999), and low risk (less than 500). 3. Generate surrounding tissue constraint information: Determine specific constraint information according to the risk level. For high-risk surrounding tissue: If the distance to the blood vessel is less than or equal to 5 mm and the risk value is greater than or equal to 1000, the constraint information may be that the puncture path must maintain a safe distance of at least 10 mm from the tissue, the operation force must not exceed X Newtons, and the deviation of the puncture angle must not exceed ±5 degrees. For medium-risk surrounding tissue: If the distance to the blood vessel is between 5-10 mm and the risk value is between 500-999, the constraint information may be to maintain a safe distance of at least 8 mm, the operation force must not exceed Y Newtons, and the deviation of the puncture angle must not exceed ±8 degrees. For low-risk surrounding tissues: the distance to the blood vessel is greater than 10 mm, the risk value is less than 500, and the constraint information is relatively loose. For example, if a safe distance of at least 5 mm is maintained, the operation force can be appropriately relaxed, but care must still be taken to avoid excessive squeezing of the tissue.

[0054] Step S500, based on the preset safety margin constraint, blood vessel feature constraint information, and surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to the target position from different starting points and angles of the blood vessel.

[0055] Among them, safety margin constraints: some pre-set distance, angle and other restrictions to ensure the safety of the puncture process. For example, a certain distance needs to be maintained between the puncture path and the blood vessel wall to avoid damaging the blood vessel; the puncture angle also needs to be within a certain range to ensure the stability and safety of the operation. Geometric algorithm: an algorithm that uses geometric principles and mathematical methods to calculate and is used to determine the optimal puncture path.

[0056] The process of determining the safety margin constraint is as follows: Based on medical experience and safety standards, determine the minimum safe distance between the puncture path and the vessel wall. For example, the distance between the puncture path and the vessel wall can be set to not be less than 1 mm. Determine the safe range of the puncture angle. Generally speaking, the puncture angle should not be too large or too small to avoid increasing the difficulty of puncture or damaging surrounding tissues. A suitable puncture angle range can be determined based on the specific puncture site and the characteristics of the target vessel.

[0057] Select a geometric algorithm: You can select an optimization algorithm, such as a genetic algorithm or a simulated annealing algorithm. Determine the algorithm parameters and objective function. Determine the algorithm parameters and objective function based on the safety margin constraints, vascular feature constraints, and surrounding tissue constraints. For example, the length of the puncture path, the distance from the vascular wall, the distance from the surrounding tissue, etc. can be used as part of the objective function, and the optimization algorithm can be used to find the best puncture path that meets all constraints.

[0058] The puncture path is generated as follows: Starting from different starting points and angles of the blood vessel, the selected geometric algorithm is used for calculation. Based on the output of the algorithm, the puncture path leading to the target location is determined. The generated puncture path is evaluated and adjusted. Check whether the puncture path meets all the constraints. If not, it is necessary to adjust the algorithm parameters or reselect the starting point and angle and calculate again until a puncture path that meets the requirements is found.

[0059] Taking the simulated annealing algorithm as an example, the specific process is as follows: 1. Define the problem and solution space: transform the puncture path planning problem into the problem of finding the optimal path from the starting point of the blood vessel to the target position in a certain space. The solution space can be defined as a set of all possible paths, each path consisting of a series of points, representing the coordinates of the puncture needle at different positions. 2. Initialization parameters: set parameters such as initial temperature, cooling rate, and termination temperature. The initial temperature is high so that the algorithm can accept large changes in the early stage of the search; the cooling rate controls the speed of temperature drop; the termination temperature is the condition for the algorithm to stop searching. Randomly generate an initial path as the current solution. 3. Calculate the energy function: define an energy function to measure the quality of the path. The energy function should comprehensively consider the safety margin constraint, blood vessel feature constraint information, and surrounding tissue constraint information. For example, if the path violates the safety margin constraint, such as the distance from the blood vessel wall is less than the specified value, or the distance from the important surrounding tissue is less than the safety distance, a larger penalty value is added to the energy function. If the blood vessel characteristics are not conducive to puncture, such as the blood vessel is too thin, the curvature is too large, etc., the energy function value is also increased accordingly. On the contrary, if the path satisfies various constraints and has advantages such as shorter length and better operating angle, the energy function value is reduced. 4. Iterative search: In each iteration, the current path is randomly perturbed to generate a new path. Calculate the difference between the energy function value of the new path and the energy function value of the current path. If the energy function value of the new path is lower, that is, better, the new path is accepted as the current path. If the energy function value of the new path is higher, that is, worse, the new path is accepted with a certain probability. This probability gradually decreases as the temperature decreases to simulate the process of solid materials gradually stabilizing during annealing. 5. Cooling: Reduce the temperature according to the set cooling rate. When the temperature drops to the termination temperature, the algorithm stops searching, and the current path at this time is the generated path to the target location.

[0060] Suppose that the puncture path of the patient's brachial artery is to be planned. The specific process is as follows: 1. Define the problem and solution space: divide the area where the brachial artery is located into three-dimensional space, and the path consists of a series of three-dimensional coordinate points. The starting point can be a specific position on the brachial artery, and the target position can be a specific area where the puncture operation needs to be performed. 2. Initialize parameters: set the initial temperature to 100, the cooling rate to 0.95, and the end temperature to 1. Randomly generate an initial path from the starting point to the target position, which may be a relatively tortuous path. 3. Calculate the energy function: Assume that the safety margin constraint stipulates that the distance between the puncture path and the blood vessel wall must not be less than 1.5 mm, and the distance from the surrounding important nerve tissue must not be less than 2 mm. If the path violates these constraints, a penalty value of 10 units is added to the energy function for each violation. Considering the vascular feature constraint information, assuming that the diameter of the brachial artery is 3 mm, if the diameter of the blood vessel in the area where the path passes is less than 2.5 mm, a penalty value of 5 units is added to the energy function. If the area where the blood vessel is too curved is also increased accordingly. At the same time, the energy function increases by 1 unit for every increase of 1 mm in the path length. If the operating angle of the path is more appropriate, such as between 30 and 60 degrees, the energy function value can be reduced by 2 units. 4. Iterative search: In each iteration, the current path is randomly perturbed, such as randomly changing the coordinates of a point on the path. Calculate the difference between the energy function value of the new path and the energy function value of the current path. If the new path is better, accept the new path directly. If the new path is worse, accept the new path with a certain probability. For example, when the temperature is high, the probability of accepting a worse path is greater, and this probability gradually decreases as the temperature decreases. 5. Cooling: Lower the temperature according to the cooling rate. After multiple iterations and cooling, when the temperature drops to the termination temperature, the algorithm stops searching. At this time, the path obtained is as close to the optimal path as possible while comprehensively considering various constraint information. This path not only meets the safety margin constraint, but also takes into account the limitations of vascular characteristics and surrounding tissues. At the same time, it is also more optimized in terms of length and operating angle.

[0061] Step S600: Integrate the generated puncture needle path into the 3D reconstructed model of the blood vessel for display in different colors or line styles.

[0062] Puncture needle insertion path: The optimal path for puncture operation generated by planning in the previous steps, including starting point, direction, angle and other information.

[0063] Vascular 3D reconstruction model: A three-dimensional model of blood vessels and their surrounding tissues constructed using data such as B-ultrasound images, which can intuitively display the morphology, position, and relationship of blood vessels with surrounding tissues.

[0064] The process of integrating the puncture needle path is as follows: match and integrate the generated coordinate data of the puncture needle path with the coordinate system of the 3D reconstruction model of the blood vessel. Ensure that the position of the puncture needle path in three-dimensional space is accurately displayed in the 3D reconstruction model of the blood vessel. This can be done by adding data points of the puncture needle path in the database of the model and associating it with the data of the blood vessel and surrounding tissue.

[0065] The process of presenting the puncture needle path is as follows: using visualization software or tools, a 3D reconstructed model of the blood vessel incorporating the puncture needle path is displayed. The model can be displayed in the form of a three-dimensional graphic on a computer screen, or the model can be projected into the actual operation space by projection or the like.

[0066] Reference Figure 2 , obtaining multi-angle B-ultrasound images of the target blood vessel area collected by the probe of the B-ultrasound device includes: Step S110, starting the probe of the B-ultrasound device to perform longitudinal and transverse section scanning on the target area on the skin surface to obtain a scanned B-ultrasound image.

[0067] Among them, the probe of the B-ultrasound equipment: a device for transmitting and receiving ultrasonic waves, which transmits ultrasonic waves into the human body by contacting the human skin and receives the reflected signals, thereby generating B-ultrasound images. Longitudinal section scanning: B-ultrasound scanning along the long axis of the human body, usually used to observe the length and direction of blood vessels and their relationship with surrounding tissues. Transverse section scanning: B-ultrasound scanning perpendicular to the long axis of the human body, which can be used to observe the diameter of blood vessels, cross-sectional morphology and the structure of surrounding tissues.

[0068] The process of performing a longitudinal section scan is as follows: Gently place the probe of the B-ultrasound device on the target area on the patient's skin surface, ensuring that the probe is in good contact with the skin. Slowly move the probe along the long axis of the human body while observing the B-ultrasound image on the device display. The position and angle of the probe can be adjusted as needed to obtain the best longitudinal section image. During the scan, record important image information, such as the location, morphology, and blood flow of the blood vessels.

[0069] The process of performing a transverse section scan is as follows: rotate the probe 90 degrees so that it is perpendicular to the long axis of the human body and perform a transverse section scan. Similarly, move the probe slowly, observe the image on the display, and record important information. Transverse section scans can provide information such as the diameter of the blood vessel, the cross-sectional shape, and the structure of the surrounding tissue.

[0070] Step S120 , scanning the probe of the B-ultrasound device at different angles in the longitudinal and transverse directions to obtain a scanned B-ultrasound image.

[0071] Scanning at different angles between longitudinal and transverse sections: Based on the longitudinal section scanning and transverse section scanning, the target area is scanned at various tilted angles to obtain more B-ultrasound images from different perspectives, thereby gaining a more comprehensive understanding of the structure and characteristics of the target vascular area.

[0072] The acquisition process of the scanned B-ultrasound image is as follows: First, after completing the longitudinal section scan and the transverse section scan, keep the probe in contact with the skin surface and start adjusting the angle of the probe. The scanning angle can be changed by slightly rotating and tilting the probe. You can start with a smaller angle change, such as gradually increasing the tilt at an angle of 10 degrees, 15 degrees, etc., to scan the target area. At the same time, observe the image changes on the display screen of the B-ultrasound device to ensure that the target vascular area can be clearly seen. Then, at different angles, slowly move the probe to cover different parts of the target vascular area. Pay attention to keeping the probe stable to avoid excessive jitter or moving too fast to cause image blur. For each angle scan, record important image information, including the morphology, position, and relationship with surrounding tissues of the blood vessels. You can use the marking function of the device or manually record key information for subsequent analysis. Finally, the process of adjusting the probe angle and scanning can be repeated as many times as needed. Try different angle combinations to obtain as many B-ultrasound images from different perspectives as possible.

[0073] Step S130 , summarizing the B-ultrasound images acquired at different angles through scanning as a multi-angle B-ultrasound image of the target blood vessel region.

[0074] The specific process can be divided into the following steps: 1. Image collection: First, ensure that the B-ultrasound device has completed various scanning operations in steps S110 and S120, and all scanned B-ultrasound images have been correctly stored in the storage system of the device or can be obtained through data transmission. The image storage function of the device can be used, such as storing the image on a hard disk, a memory card, or transmitting it to a specific storage location through a network. Ensure the integrity and accuracy of the image to avoid loss or damage. 2. Image sorting: Sort the collected B-ultrasound images from different angles. They can be classified and sorted according to the scanning angle, time sequence, or specific identification. Check the quality of the image and exclude images that are blurred, unclear, or have obvious interference. If necessary, some simple preprocessing can be performed on the image, such as adjusting the contrast, brightness, etc., to improve the readability of the image. 3. Summarization operation: Use specific software tools or data processing methods to aggregate the sorted B-ultrasound images from different angles. An image data set can be created to store all images in a specific folder or database.

[0075] Reference Figure 3A puncture path planning system further includes steps after the collected B-ultrasound image is finally subjected to grayscale analysis to obtain the internal structure information of the blood vessel and before the initially constructed blood vessel contour is refined to form a 3D reconstruction model of the blood vessel, which are specifically as follows: Step S210, using a spatial positioning module preset in the probe to record the specific position and angle information of the probe when acquiring each image, and defining the specific position and angle information when acquiring each image as spatial positioning information.

[0076] Spatial Positioning Module: A device installed in the B-ultrasound probe that can record the specific position and angle information of the probe when collecting each image. It usually uses sensor technology, such as accelerometers, gyroscopes, etc., to determine the position and direction of the probe in three-dimensional space. Spatial Positioning Information: The specific position and angle data of the probe when collecting each image recorded by the spatial positioning module.

[0077] When using the probe to collect B-ultrasound images, the spatial positioning module will automatically record the specific position and angle information of the probe when collecting each image. This information can be transmitted to a computer or other data storage device in real time and saved together with the corresponding B-ultrasound image. After the acquisition is completed, the spatial positioning information can also be associated with the image through a specific data transmission method.

[0078] Step S220, determining the relative position relationship of the blood vessel boundaries in different images based on the spatial positioning information, and using an edge detection algorithm to refine the blood vessel boundaries. At the same time, using the spatial positioning information, the feature points found by the feature-based image matching algorithm are enhanced.

[0079] Blood vessel boundary: In B-ultrasound images, the boundary between blood vessels and surrounding tissues. The location of blood vessel boundaries can be determined by edge detection algorithms. Feature points: Points with specific features in B-ultrasound images. They are found by feature-based image matching algorithms and can be used for image alignment and 3D model construction.

[0080] The relative position relationship of the blood vessel boundaries is determined as follows: First, read the spatial positioning information corresponding to each B-ultrasound image. This information includes the position and angle of the probe. According to the spatial positioning information, the different images are converted to the same coordinate system. This can be achieved through mathematical transformations, such as translation and rotation. In the unified coordinate system, the position and morphology of the blood vessels in different images are compared. By analyzing the appearance of blood vessels in different images, the relative position relationship between them can be determined.

[0081] The process of thinning the blood vessel boundary is as follows: For each B-ultrasound image, an edge detection algorithm is applied to determine the blood vessel boundary. Commonly used edge detection algorithms include Sobel operator, Canny operator, etc. The edge detection algorithm determines the edge of the object by analyzing the grayscale change of the image. In the process of thinning the blood vessel boundary, it is necessary to adjust the parameters of the algorithm according to the characteristics of the blood vessel to obtain the best effect. The detected blood vessel boundary is further processed, such as removing noise, smoothing edges, etc., to improve the accuracy and clarity of the boundary.

[0082] The process of feature point enhancement is as follows: After the feature points are found in different images by the feature-based image matching algorithm, these feature points are enhanced using the spatial positioning information. Based on the spatial positioning information, the position and direction of the feature points in the three-dimensional space can be determined. By analyzing the spatial distribution of feature points, those feature points that are important for the construction of the three-dimensional model can be further highlighted and strengthened.

[0083] Step S230, the three-dimensional model of the blood vessel is refined according to the refined blood vessel boundary, the enhanced feature points and the spatial positioning information.

[0084] The specific process is as follows: Integrate information: Integrate the refined vascular boundaries, enhanced feature points, and spatial positioning information. A data structure can be established to store this information together for subsequent processing. Ensure the accuracy and consistency of vascular boundaries, feature points, and spatial positioning information. If necessary, data calibration and verification can be performed.

[0085] Optimize the 3D model: Use the refined vascular boundaries to correct the surface of the initially constructed vascular 3D model. The model's surface can be made to fit the actual vascular boundaries better by adjusting the model's vertex positions, normal directions and other parameters. The enhanced feature points can be used as constraints to further optimize the shape and structure of the 3D model. For example, the model can be made to better fit the vascular morphology reflected by the feature points by adjusting the model's curvature, bending degree and other parameters. Spatial positioning information can be used to determine the positional relationship between different images in 3D space, thereby better fusing the data of multiple images and further refining the 3D model.

[0086] Reference Figure 4 , the initially constructed vascular contour is refined to form a 3D vascular reconstruction model including: Step S2a0, acquiring infrared imaging data and magnetic resonance imaging data.

[0087] Infrared imaging data: Thermal radiation image data of human tissue collected by infrared imaging equipment. Different tissues will show different image characteristics in infrared imaging due to differences in temperature and thermal radiation characteristics. In vascular-related applications, infrared imaging can provide temperature information about vascular areas, etc. The specific method of obtaining infrared imaging data is as follows: Use an infrared imaging device and aim it at the target vascular area. The device generates an image by detecting infrared radiation emitted by the human body. Adjust the parameters of the device, such as resolution, contrast, temperature range, etc., to obtain the best image quality. Appropriate adjustments may be required depending on the patient and examination site.

[0088] During the acquisition process, ensure that the patient remains still to avoid blurring of the image. The patient may be asked to be examined in a specific posture or a fixation device may be used to restrict the patient's movement. After the acquisition is completed, the infrared imaging data is stored in a computer or other storage device for subsequent processing.

[0089] Magnetic resonance imaging data: Image data of the internal structure of the human body obtained using magnetic resonance imaging technology. Magnetic resonance imaging can provide high-resolution soft tissue images and is very effective in showing the morphological characteristics of blood vessels. The acquisition of magnetic resonance imaging data is as follows: The patient enters the scanning room of the magnetic resonance imaging device and lies on the scanning bed. The technician sets the parameters of the magnetic resonance imaging device, such as magnetic field strength, pulse sequence, scanning time, etc., according to the purpose and location of the examination. The device is started for scanning. During the scanning process, the device will emit radio frequency pulses into the patient's body and receive magnetic resonance signals emitted by human tissue. The device processes the received signals to generate magnetic resonance image data.

[0090] Step S2b0, extracting blood vessel related features from infrared imaging, and extracting blood vessel morphological features from magnetic resonance imaging.

[0091] Among them, blood vessel-related features include temperature features, which refer to the temperature difference between the blood vessel area and the surrounding tissue due to blood flow and other reasons. This temperature difference can be used as an important feature to analyze the blood vessel condition. Morphological features include the length, diameter, curvature, and branching angle of the blood vessel.

[0092] The process of extracting blood vessel-related features from infrared imaging is as follows: First, the acquired infrared imaging data is preprocessed, including noise removal, contrast enhancement and other operations to improve image quality. Then, the blood vessel region is identified through specific algorithms and techniques. The temperature difference can be used to distinguish blood vessels from surrounding tissues. For example, the blood vessel region is usually slightly hotter than the surrounding tissue. The temperature features of the blood vessel region are extracted, and parameters such as the average temperature of the blood vessel region and the temperature variation range can be calculated as temperature features.

[0093] The process of extracting morphological features of blood vessels from MRI is as follows: MRI data is processed, including image segmentation, denoising and other operations, to highlight the vascular structure. Image analysis algorithms are used to measure the length of the blood vessel. The measurement can be performed along the centerline of the blood vessel to calculate the distance from the start to the end of the blood vessel. To determine the diameter of the blood vessel, the width of the blood vessel can be measured at multiple locations and the average value is taken as the diameter feature. To analyze the curvature of the blood vessel, the degree of curvature can be determined by calculating the curvature of the centerline of the blood vessel. For branching angles, the angle between branching blood vessels can be measured at the branch of the blood vessel.

[0094] Step S2c0, establishing a feature fusion matrix, matching and combining the features extracted from infrared imaging, the features extracted from magnetic resonance imaging, and the internal structure information of blood vessels obtained from grayscale analysis of B-ultrasound images, to form fusion features.

[0095] The feature fusion matrix is ​​a mathematical structure used to integrate features from different sources. It corresponds and combines features extracted from multiple imaging methods through specific rules to form fused features that can more comprehensively and accurately describe vascular characteristics.

[0096] Blood vessel internal structure information obtained by grayscale analysis of B-ultrasound images: characteristic information about the internal structure of blood vessels obtained by grayscale analysis of B-ultrasound images, such as the thickness of the blood vessel wall, the shape of the blood vessel cavity, etc. B-ultrasound imaging can observe the internal structure of blood vessels in real time.

[0097] The process of establishing the feature fusion matrix is ​​as follows: Infrared imaging features: Assume that the extracted temperature features are quantified in degrees Celsius (°C), for example, the average temperature value is T1°C. If there is also a temperature distribution feature, it can be represented by a vector, such as [T11°C, T12°C, ..., T1n°C], where n is the number of sampling points of the temperature distribution.

[0098] Magnetic resonance imaging features: Vessel length is quantified in millimeters (mm), recorded as Lmm. Vessel diameter is quantified in millimeters (mm), recorded as Dmm. Curvature can be quantified in angles, recorded as θ degrees. Branching angles can be quantified in degrees, recorded as α degrees.

[0099] Grayscale analysis features of B-ultrasound images: The thickness of the blood vessel wall is quantified in millimeters (mm), recorded as Wmm. The area of ​​the blood vessel cavity is quantified in square millimeters (mm²), recorded as Amm².

[0100] The size of the feature fusion matrix is ​​determined to be 3×m, where 3 represents the three imaging methods and m is the sum of all feature dimensions.

[0101] Initialize the matrix: Create an initial feature fusion matrix. The initial values ​​of all elements can be set to zero or set according to a specific initialization strategy. For example, create a 3×m matrix M, and the initial value can be set to zero.

[0102] Data standardization: For each feature dimension, Z-score standardization is performed.

[0103] Assume that for a certain feature dimension X (which can be any of the above quantitative features), its original data is [X1,X2,...,Xn].

[0104] Calculate the mean μ=(X1+X2+...+Xn) / n.

[0105] Calculate the standard deviation σ=sqrt(((X1-μ)²+(X2-μ)²+...+(Xn-μ)²) / n). The standardized value is (Xi-μ) / σ, where i=1,2,...,n. Fill the standardized eigenvalues ​​into the corresponding positions of the feature fusion matrix.

[0106] The process of feature matching and combination is as follows: Feature correspondence: Determine the correspondence between features extracted by different imaging methods. For example, the average temperature feature of infrared imaging is matched with the blood vessel diameter feature of magnetic resonance imaging, assuming that the blood vessels may expand and increase in diameter when the temperature is higher. This correspondence can be determined through clinical experience or preliminary research.

[0107] 2. The following are examples of feature combination method selection and quantification: Linear combination: Assume that the fusion feature F=w1*F1+w2*F2+w3*F3, where F1, F2, and F3 are features from three imaging methods, and w1, w2, and w3 are corresponding weight coefficients. For example, F1 is the average temperature feature T1 (standardized value) of infrared imaging, F2 is the blood vessel diameter feature D (standardized value) of magnetic resonance imaging, and F3 is the blood vessel wall thickness feature W (standardized value) of B-ultrasound images. Assume that the weight coefficients w1=0.3, w2=0.5, and w3=0.2 are determined experimentally. Then the fusion feature F=0.3T1+0.5D+0.2*W.

[0108] Nonlinear combination (taking neural network as an example): Suppose a simple three-layer neural network is constructed, with m neurons in the input layer, corresponding to the m feature dimensions of the feature fusion matrix; k neurons in the hidden layer; and 1 neuron in the output layer, representing the fusion feature. The network is trained with the input as the standardized feature vector [F1, F2, ..., Fm] and the target output as the known fusion feature value (the fusion feature value of a part of the samples can be pre-determined by other methods for training). The network weights are adjusted by the back propagation algorithm, and after multiple iterations of training, a neural network model that can accurately predict the fusion feature is obtained. Combination based on feature selection: First, calculate the correlation or importance index of each feature. For example, the Pearson correlation coefficient between the feature and the target variable (such as the diagnosis result of vascular disease) can be calculated. Suppose that after calculation, it is found that the temperature feature of infrared imaging has a low correlation with the target variable, while the vascular diameter and curvature features of magnetic resonance imaging and the vascular wall thickness features of B-ultrasound images have a high correlation. Select the features with higher correlation for combination, such as F=w1D+w2θ+w3*W, where w1, w2, and w3 are weight coefficients, which can be determined by experiments or optimization algorithms.

[0109] Step S2d0, obtaining blood vessel adjustment parameters based on fusion feature analysis, and adjusting the blood vessel three-dimensional model according to the blood vessel adjustment parameters and the blood vessel internal structure information obtained by grayscale analysis, and adjusting the geometric shape of the blood vessel three-dimensional model using the morphological features in the fusion features.

[0110] Vascular adjustment parameters: Numerical parameters used to adjust the vascular 3D model based on fusion feature analysis. These parameters can change the location, shape, size and other properties of the blood vessels to more accurately reflect the actual state of the blood vessels.

[0111] The process of obtaining the vascular adjustment parameters based on the fusion feature analysis is as follows: the fusion feature matrix is ​​analyzed to extract information related to the adjustment of the vascular three-dimensional model. For example, the eigenvalues ​​related to the size and shape of the blood vessel can be found from the fusion features through statistical analysis, machine learning algorithms, etc. According to the extracted eigenvalues, the corresponding vascular adjustment parameters are calculated. These parameters can be length adjustment coefficients, diameter adjustment coefficients, curvature adjustment angles, etc. For example, if the fusion feature shows that the blood vessel is thicker in a certain area than the blood vessel in the model, then a diameter adjustment coefficient can be calculated to increase the diameter of the blood vessel in the area. A variety of methods can be used to determine the vascular adjustment parameters, such as regression analysis, neural networks, etc. These methods can automatically learn and generate adjustment parameters based on the relationship between the fusion features and the actual blood vessel features.

[0112] The process of adjusting the vascular three-dimensional model according to the vascular adjustment parameters and the vascular internal structure information obtained by grayscale analysis is as follows: 1. Adjustment using vascular adjustment parameters: According to the obtained vascular adjustment parameters, the vascular three-dimensional model is adjusted accordingly. If the adjustment parameter is a size adjustment coefficient, the length, diameter and other dimensions of the vascular three-dimensional model can be scaled according to the coefficient. If it is an angle adjustment value, the bending angle of the blood vessel can be adjusted. For example, if the size adjustment coefficient is 1.2, all dimensions of the vascular three-dimensional model are multiplied by 1.2 to make the model larger. If the angle adjustment value is 5 degrees, the bending angle of the blood vessel can be increased by 5 degrees. 2. Adjustment of the vascular internal structure information obtained by grayscale analysis: The vascular internal structure information obtained by grayscale analysis can provide information about the thickness of the vascular wall, the vascular cavity morphology and other aspects. Based on this information, the vascular three-dimensional model can be further adjusted. For example, if the grayscale analysis finds that the vascular wall thickness is uneven, the wall thickness of the vascular three-dimensional model can be adjusted accordingly to make it more consistent with the actual situation. If the vascular cavity morphology is irregular, the shape of the model can be adjusted to better match the actual vascular cavity morphology.

[0113] The process of adjusting the geometric shape of the 3D model of the blood vessel using the morphological features in the fused features: 1. Extract morphological features: Extract the morphological features of the magnetic resonance imaging from the fused features, i.e., the length, diameter, curvature, branch angle, etc. of the blood vessel. 2. Adjust the geometric shape: According to these morphological features, adjust the geometric shape of the 3D model of the blood vessel. For example, if the length changes, the overall length of the model can be adjusted. If the diameter changes, the thickness of the model can be adjusted. Changes in curvature and branch angles can be achieved by adjusting the positions of the curved parts and branch points of the model.

[0114] For example, if magnetic resonance imaging shows that the curvature of a blood vessel increases by 10 degrees, the curvature angle of the corresponding part of the three-dimensional model of the blood vessel can also be increased by 10 degrees to make the model more consistent with the actual vascular morphology.

[0115] Based on the preset safety margin constraints, vascular feature constraint information, and surrounding tissue constraint information, a preset geometric algorithm is used to generate paths to the target location from different starting points and angles of the blood vessels, including: Step S510 , collecting the patient's current body position information and physiological index related information, and collecting multimodal imaging data of the patient in different body positions and physiological states.

[0116] Among them, multimodal imaging data includes infrared imaging data, magnetic resonance imaging data and B-ultrasound image data; body position information includes supine, lateral, prone and semi-sitting positions; and information related to physiological indicators includes heart rate, blood pressure, respiratory rate and blood oxygen saturation.

[0117] Information related to physiological indicators can be measured using appropriate medical equipment to measure the patient's heart rate, blood pressure, respiratory rate, blood oxygen saturation and other physiological indicators.

[0118] Step S520 , input the collected patient's current body position information, physiological index related information and corresponding multimodal imaging data into the trained vascular feature analysis model to extract key features of the blood vessels and surrounding tissues.

[0119] Vascular feature analysis model: A trained model that can extract vascular features and surrounding tissue features from input information.

[0120] The training process of the vascular feature analysis model is as follows: 1. Data collection and labeling: collect a large amount of patient position information, physiological index related information, infrared imaging data, magnetic resonance imaging data, and B-ultrasound image data. Professional doctors or medical imaging experts label these data and mark the corresponding vascular features (such as vascular diameter, curvature, blood flow, etc.) and surrounding tissue features (such as surrounding tissue hardness, distance from the blood vessel, location of important organs, etc.). 2. Data preprocessing: Preprocess the collected raw data, including encoding and converting body position information, standardizing information related to physiological indicators, and performing image enhancement and denoising on multimodal imaging data, so that it is suitable for input into the model for training. 3. Model architecture selection: Select the appropriate model architecture according to the task requirements. For example, you can choose a deep neural network, such as a combination of a convolutional neural network (CNN) and a recurrent neural network (RNN), or other complex deep learning model architectures.

[0121] Determine the model's parameters such as the number of layers, number of neurons, activation function, etc. 4. Model initialization: Randomly initialize the model's parameters to prepare for model training. 5. Training process: Divide the labeled training data into several batches, and input one batch of data into the model each time. The model processes the input data and outputs the predicted vascular features and surrounding tissue features. Calculate the loss function, which is the difference between the predicted features and the true labeled features. For example, mean square error (MSE), cross entropy, etc. can be used as loss functions. According to the loss function, use an optimization algorithm (such as Adam) to update the model's parameters so that the loss function gradually decreases, that is, the model's prediction results are getting closer and closer to the true labeling. Repeat the above process until the model's loss function on the training set reaches a stable minimum value or after a certain number of training rounds (epochs).

[0122] The extraction process of key features of blood vessels and surrounding tissues is as follows: After the model receives input data, it extracts key features of blood vessels and surrounding tissues through a series of calculations and processing. For image data, the model may use structures such as convolutional layers to extract features in the image, such as the edges and textures of blood vessels. For numerical data, such as physiological indicators, the model may process through fully connected layers to extract information related to blood vessel features. The output of the model can be a feature vector, which contains the numerical representation of blood vessel features and surrounding tissue features.

[0123] Step S530, generating current blood vessel feature constraint information according to the extracted blood vessel key feature data, and generating current surrounding tissue constraint information according to the acquired surrounding tissue key feature data.

[0124] The method of generating the blood vessel feature constraint information and the surrounding tissue constraint information in step S530 is the same as that described in step S400 and will not be described in detail here.

[0125] Step S540, based on the preset safety margin constraint, the current blood vessel feature constraint information, and the current surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to the target position from different starting points and angles of the blood vessel.

[0126] The process of step S540 is basically similar to that of step S500 and will not be described in detail here.

[0127] Generating a path to a target location involves: Step S541, obtaining different consideration angles, including a safety priority angle, an efficiency priority angle, and a patient comfort angle.

[0128] Safety-first perspective: The primary consideration in path planning is the safety of the puncture process to ensure that the risk to the patient is minimized. Efficiency-first perspective: Focus on the efficiency of puncture path generation, find a feasible path as quickly as possible, and reduce surgery time. Patient comfort perspective: Consider the patient's feelings during the puncture process and try to choose a path that causes less discomfort to the patient's body.

[0129] Step S542, analyzing and determining genetic algorithm adjustment schemes under different consideration angles according to the mapping relationship between the consideration angles and the genetic algorithm adjustment schemes, wherein the genetic algorithm adjustment schemes include adjusting the chromosome encoding method, the fitness function, and the genetic algorithm parameter adjustment.

[0130] Taking the patient comfort as an example, chromosome coding: information related to the patient comfort is added to the coding method. For example, for neck blood vessel puncture, if the path passes through the sensitive area of ​​the neck, it is indicated by a specific mark "1" in the chromosome coding, otherwise it is indicated by "0". At the same time, the body position information that the patient needs to maintain under a specific path can also be encoded, such as using digital codes to represent different body position states.

[0131] Fitness function: Add the evaluation of patient comfort to the fitness function. Consider factors such as the tissue type and patient position requirements of the path. Assume that the score of passing through soft tissue is s1, the score of not requiring the patient to twist the body excessively is s2, and the scores of other non-comfort factors are other_factors.

[0132] The fitness function can be designed as F=0.6*s1+0.3*s2+0.1*other_factors. For example, when passing through soft tissue, s1=1, when passing through harder tissue, s1=0.5; when the patient does not need to twist the body excessively, s2=1, when the patient needs to twist the body to a greater extent, s2=0.3.

[0133] Parameter adjustment: Appropriately reduce the crossover probability and mutation probability, reduce the crossover probability from 0.7 to 0.6, and the mutation probability from 0.1 to 0.08, to maintain the stability of the path and avoid discomfort to the patient. At the same time, add constraints, such as the path cannot pass through the extremely sensitive area of ​​the patient, or the patient's position requirement on a specific path cannot exceed a certain comfort threshold. Assume that this threshold is expressed in angles, such as the patient's position angle cannot exceed 30 degrees.

[0134] Step S543, running the genetic algorithm for different consideration angles respectively to obtain the optimal puncture path solutions for different angles.

[0135] Taking the patient comfort angle as an example, the parameters of the genetic algorithm are set according to the genetic algorithm adjustment scheme determined under the patient comfort angle. When running the genetic algorithm, the fitness function considers comfort-related factors such as the tissue type passed by the path and the patient's body position requirements. Paths that pass through the comfort area and have low patient position requirements have higher fitness values. After multiple generations of evolution, the genetic algorithm will generate puncture path solutions with higher comfort. The best one is selected as the optimal puncture path solution under the patient comfort angle.

[0136] For example, the fitness function can be F=0.6*comfort_factor+0.4*other_factors, where comfort_factor is the evaluation value of the comfort factor and other_factors is the evaluation value of other non-comfort factors. If a path plan passes through soft tissue and the patient position requirement is low, comfort_factor=0.8 and other_factors=0.7.

[0137] Then the fitness value F = 0.6*0.8+0.4*0.7=0.48+0.28=0.76. Through continuous evolution, the path plan with the highest fitness value is found as the optimal solution from the perspective of patient comfort.

[0138] A puncture path planning system also includes steps after integrating the generated puncture needle path into the 3D reconstructed model of the blood vessel for display and presenting it in different colors or line styles, specifically as follows: Step S700, obtaining the patient's physiological indicators, the position and angle of the puncture needle, and the state information of the blood vessel during the puncture process.

[0139] The process of obtaining the position and angle of the puncture needle is as follows: The position and angle of the puncture needle can be determined using an optical positioning system, an electromagnetic positioning system, or other positioning technologies. These systems usually consist of a sensor and a receiver, where the sensor is mounted on the puncture needle and the receiver is placed in the surrounding environment.

[0140] The sensor sends a signal to the receiver, which calculates the position and angle of the puncture needle based on the received signal. The data can be transmitted to a computer or other display device via wired or wireless means.

[0141] The state information of blood vessels is obtained as follows: Before puncture, the initial state information of blood vessels can be obtained through multimodal imaging technology (such as infrared imaging, magnetic resonance imaging, B-ultrasound, etc.), including the diameter of blood vessels, blood flow velocity, elasticity of blood vessel walls, etc. During the puncture process, ultrasound imaging technology can be used to monitor the state changes of blood vessels in real time. The ultrasound probe can be placed near the puncture site to generate an image of the blood vessel by emitting ultrasound and receiving reflected signals. The ultrasound image is analyzed to extract the state information of the blood vessel. Image processing algorithms can be used to measure the diameter of blood vessels, calculate blood flow velocity, etc.

[0142] Step S800, displaying the patient's physiological index curve on the operation interface according to the patient's physiological index acquired during the puncture process, and at the same time using a 3D model to display the position and path of the puncture needle in real time, and analyzing the result information of the puncture needle executing the preset puncture path plan for display.

[0143] The process of displaying the patient's physiological index curve on the operation interface according to the patient's physiological index obtained during the puncture process is as follows: Continuously receive the patient's physiological index data from physiological monitoring equipment (such as electrocardiograph, sphygmomanometer, etc.). Plot these data in real time on the operation interface to form a dynamic physiological index curve. This can be achieved using common drawing software libraries or specialized medical data visualization tools. For example, for heart rate data, time can be used as the horizontal coordinate, heart rate value as the vertical coordinate, and data points can be updated once a second to draw a continuous heart rate curve. When the heart rate changes, the curve will fluctuate accordingly, and the doctor can intuitively see the dynamic changes of the patient's heart rate.

[0144] The process of using a 3D model to display the position and path of the puncture needle in real time is as follows: The position and angle information of the puncture needle is obtained in real time through positioning technology (such as electromagnetic positioning, optical positioning, etc.). This information is transmitted to the computer, and the position and path of the puncture needle are displayed in real time in a pre-built 3D reconstruction model of the patient's blood vessels. A three-dimensional graphics rendering engine can be used to achieve this function. For example, when the puncture needle moves in the patient's body, the positioning system will continuously send new position information. After receiving this information, the computer updates the position of the puncture needle in the 3D model to keep it synchronized with the actual position. The doctor can observe the position and path of the puncture needle in the blood vessel from different angles so as to adjust the puncture direction in time.

[0145] The process of analyzing and displaying the result information of the puncture needle executing the preset puncture path plan is as follows: During the puncture process, the position of the puncture needle and the state of the surrounding tissue are continuously monitored. Imaging technology (such as ultrasound imaging) can be used to observe the relationship between the puncture needle and the blood vessels and surrounding tissues. Compare and analyze the actual puncture results with the preset puncture path plan. For example, determine whether the puncture needle has accurately reached the target position, whether it has deviated from the preset path, whether it has caused damage to the surrounding tissues, etc. The analysis results are displayed on the operation interface in the form of text, graphics or color markings. For example, if the puncture needle deviates from the preset path, the deviation position can be marked in red in the 3D model, and the distance and direction of the deviation can be displayed. If the puncture needle causes damage to the surrounding tissues, special icons or colors can be used to indicate the degree and scope of the damage.

[0146] A puncture path planning also includes the steps of displaying the patient's physiological index curve on the operation interface according to the patient's physiological index obtained during the puncture process, and using the 3D model to display the position and path of the puncture needle in real time, and analyzing the result information of the puncture needle executing the preset puncture path plan for display, which are as follows: Step SA00, analyzing whether to detect and obtain information related to the failure of equipment components involved in the puncture process.

[0147] Fault-related information: Information corresponding to various fault conditions that may occur in the equipment components involved during the puncture process. This includes hardware failure, software failure, sensor failure, etc., which may manifest as equipment failure to work properly, data anomalies, signal interruption, etc.

[0148] The specific process is as follows: Establish a monitoring system: Establish a comprehensive monitoring system for various equipment involved in the puncture process. This system can include hardware sensors, software monitoring modules, etc. For example, for the positioning device of the puncture needle, a position sensor and an angle sensor can be installed to monitor the position and angle information of the puncture needle in real time. At the same time, a monitoring module is set in the control software to monitor the stability and accuracy of data transmission.

[0149] Data collection and analysis: The monitoring system continuously collects various data during the operation of the equipment, including the equipment's working status parameters, sensor data, communication signals, etc. The collected data is analyzed in real time to determine whether there are any abnormalities. For example, analyze the stability of sensor data, whether the data change trend is in line with expectations, whether the communication signal is interrupted, etc. Data analysis algorithms and models, such as statistical analysis and machine learning algorithms, can be used to identify potential signs of failure.

[0150] Fault detection judgment: Based on the results of data analysis, determine whether fault-related information has been detected. If the data fluctuates abnormally, exceeds the normal range, or the device exhibits obvious abnormal behavior, it is considered that fault-related information may have been detected. For example, if the positioning data of the puncture needle suddenly fluctuates greatly, or the communication between the positioning device and the control computer is interrupted, it can be determined that a fault may have occurred.

[0151] Step SB00, if no, maintain the original settings.

[0152] Step SC00: If yes, an alarm is issued, and the type and location of the faulty device are determined and displayed based on the acquired fault-related information.

[0153] Specifically, once the relevant information of the fault is detected, the alarm system is triggered immediately. Multiple alarm modes can be set to be performed simultaneously, such as a sharp sound alarm, a flashing warning light in the operating room, and a prominent warning prompt box popping up on the operation interface.

[0154] The process of determining the type and location of the faulty device and displaying it is as follows: According to the specific content of the fault-related information, the type of the faulty device is analyzed and determined. It can be judged by the unique identification and fault characteristics of each device in the monitoring system. For example, if the electromagnetic positioning system fails, it may be manifested as a sudden loss or abnormal fluctuation of the puncture needle position data. By identifying these characteristics, it can be determined that the faulty device is a puncture needle positioning device. The location of the faulty device can be represented by the physical layout of the device in the operating room or the logical location in the system. For the physical location, the layout diagram of the operating room can be displayed on the operation interface, and the specific location of the faulty device can be marked.

[0155] A puncture path planning system also includes the steps of displaying the patient's physiological index curve on the operation interface according to the patient's physiological index obtained during the puncture process, and simultaneously using the 3D model to display the position and path of the puncture needle in real time, and analyzing the result information of the puncture needle executing the preset puncture path plan for display, which are as follows: Step Sa00, according to the patient's physiological indicators during the puncture process, the position and angle of the puncture needle, and the status information of the blood vessels, a preset risk quantification analysis scheme is used to analyze and determine the risk quantification values ​​of different dimensions. The preset risk quantification analysis scheme includes a risk quantification analysis scheme based on physiological indicators, a risk quantification scheme based on the position and angle of the puncture needle, and a risk quantification scheme based on the status of the blood vessels. The dimensions include the patient's physiological indicators during the puncture process, the position and angle of the puncture needle, and the status information of the blood vessels.

[0156] Preset risk quantification analysis plan: a pre-set method system for analyzing risks in different dimensions, including risk quantification analysis plans based on physiological indicators, puncture needle position and angle, and vascular status.

[0157] Risk quantification analysis scheme based on physiological indicators: First, determine the normal range and risk threshold of physiological indicators. For example, the normal range of heart rate is 60-100 beats / minute. When the heart rate exceeds 120 beats / minute or is less than 50 beats / minute, it is considered to be at a higher risk. Continuously monitor the changes in the patient's physiological indicators during the puncture process. Appropriate medical equipment, such as electrocardiographs, sphygmomanometers, pulse oximeters, etc., can be used to collect physiological indicator data in real time. Determine the risk quantification value based on the relationship between physiological indicators and risk thresholds. For example, linear or nonlinear quantification methods can be used. If the heart rate is 110 beats / minute, the corresponding risk quantification value can be calculated as 0.5 according to the preset formula (assuming that 0 represents no risk and 1 represents the highest risk).

[0158] Risk quantification scheme based on puncture needle position and angle: Determine the ideal position and angle range of the puncture needle. This can be determined based on medical experience, anatomical knowledge, and puncture targets. For example, for a specific blood vessel puncture, the angle between the puncture needle and the blood vessel should be between 30-60 degrees. Use positioning technology (such as electromagnetic positioning, optical positioning, etc.) to obtain the position and angle information of the puncture needle in real time. Calculate the degree of deviation between the actual position and angle of the puncture needle and the ideal range. The greater the deviation, the higher the risk. Indicators such as distance and angle difference can be used to quantify the deviation. For example, if the puncture needle deviates from the ideal angle by 10 degrees, the corresponding risk quantification value can be calculated as 0.3 according to the preset formula.

[0159] Risk quantification scheme based on vascular status: Analyze the status information of the blood vessels, including diameter, blood flow velocity, elasticity, etc. This information can be obtained through multimodal imaging technology (such as infrared imaging, magnetic resonance imaging, B-ultrasound, etc.). Determine the degree of risk under different vascular conditions. For example, a blood vessel diameter that is too small or a blood flow velocity that is too fast may increase the difficulty and risk of puncture. Determine the risk quantification value based on the relationship between vascular status and risk. For example, if the blood vessel diameter is 50% of the normal range, the corresponding risk quantification value can be calculated as 0.6 according to the preset formula.

[0160] Step Sb00, calculating a comprehensive risk quantification value according to the risk quantification values ​​of different dimensions and a preset weight allocation scheme for different dimensions.

[0161] Determine the weight distribution scheme: First, analyze the importance of each dimension in puncture risk assessment. For example, for punctures in certain key areas, the state of the blood vessels may be more important, so a higher weight is given; while for punctures in general areas, the stability of the patient's physiological indicators may be relatively less important. Based on the analysis results, determine the weight values ​​of each dimension. The sum of the weight values ​​should be 1. For example, the weight of the physiological indicator dimension can be set to 0.3, the weight of the puncture needle position and angle dimension can be set to 0.3, and the weight of the blood vessel status dimension can be set to 0.4. Record the determined weight distribution scheme for use in subsequent calculations.

[0162] Calculate the comprehensive risk quantification value: obtain the risk quantification values ​​of different dimensions obtained in step Sa00. Assume that the risk quantification value of the physiological index is r1, the risk quantification value of the puncture needle position and angle is r2, and the risk quantification value of the vascular status is r3. According to the preset weight distribution scheme, calculate the comprehensive risk quantification value R. The calculation formula is: R=w1*r1+w2*r2+w3*r3, where w1, w2, and w3 are the weights of the physiological index, puncture needle position and angle, and vascular status dimensions, respectively. For example, assume that r1=0.4, r2=0.5, r3=0.6, and the weight distribution scheme is w1=0.3, w2=0.3, w3=0.4. Then the comprehensive risk quantification value R=0.3*0.4+0.3*0.5+0.4*0.6=0.12+0.15+0.24=0.51.

[0163] Step Sc00, if the obtained comprehensive risk quantification value exceeds the preset comprehensive risk quantification value, the latest infrared imaging data, magnetic resonance imaging data, B-ultrasound image data, and the patient's current body position information and physiological index related information are collected, and the process jumps to the step of extracting blood vessel-related features from infrared imaging and extracting blood vessel morphological features from magnetic resonance imaging, re-analyzing the 3D model of the blood vessel, and calculating the optimal puncture path plan at different angles.

[0164] First, the calculated comprehensive risk quantification value is compared with the preset comprehensive risk quantification value. The preset comprehensive risk quantification value is a threshold determined based on clinical experience and safety standards. If it exceeds the threshold, it means that the risk is high and further measures need to be taken.

[0165] Calculate the optimal puncture path plan at different angles: Based on the reconstructed 3D vascular model, a preset geometric algorithm is used to generate a path to the target location from different starting points and angles of the vascular. Considering factors such as safety margin constraints, vascular feature constraint information, and surrounding tissue constraint information, the generated path is evaluated and optimized. Finally, the optimal puncture path plan at different angles is obtained.

[0166] Step Sd00, according to the preset benefit quantification evaluation scheme, the calculated optimal puncture path scheme at different angles is subjected to benefit analysis, and the quantitative values ​​of different benefit angles are obtained through analysis, and the benefit angles include puncture success rate, patient recovery time, and treatment effect.

[0167] Benefit analysis: Evaluate the optimal puncture path plans at different angles to determine the benefits they bring in terms of puncture success rate, patient recovery time, treatment effect, etc.

[0168] Puncture success rate: refers to the proportion of successful puncture operations and achieving the expected goal. A higher puncture success rate means fewer repeated operations and lower risks.

[0169] Patient recovery time: The time it takes for a patient to return to normal after a puncture procedure. A shorter recovery time can reduce patient pain and the risk of complications.

[0170] Therapeutic effect: The therapeutic effect of the puncture operation on the patient's disease, including the degree of symptom relief, control of disease progression, etc.

[0171] Quantification of benefit perspective: Quantify the collected data to determine the quantitative value of each benefit perspective. For the puncture success rate, it can be expressed as a percentage value, for example, 80% means that similar puncture operations have an 80% success rate in historical data. For patient recovery time, it can be quantified according to the predicted recovery time of different path schemes, for example, a short recovery time can be assigned a higher quantitative value, and a long recovery time can be assigned a lower quantitative value. For the treatment effect, it can be quantified according to the degree of disease relief, the examination results after treatment, etc. For example, complete relief can be assigned a higher quantitative value, partial relief can be assigned a medium quantitative value, and no relief can be assigned a lower quantitative value. Furthermore, the treatment effect can be quantified by blood flow velocity. For patients with specific diseases, blood flow velocity data before and after treatment are collected to analyze the impact of the disease on blood flow velocity. The evaluation criteria for treatment effect are formulated as follows: Complete relief: After treatment, the blood flow velocity returns to the normal reference range, and the symptoms completely disappear. Partial relief: After treatment, the blood flow velocity has improved significantly, but has not reached the normal reference range, and the symptoms have been relieved. No relief: After treatment, there is no significant change in blood flow velocity, and the symptoms persist.

[0172] Step Se00, according to the preset risk quantification analysis scheme and the preset weight allocation scheme for different dimensions, analyze and calculate the comprehensive risk quantification value of different paths one by one, and calculate the risk-benefit ratio of different paths according to the preset risk-benefit ratio formula, and display the risk-benefit of different paths in the form of a chart, wherein the preset risk-benefit ratio formula is as follows: Risk-benefit ratio = (quantified value of puncture success rate + quantitative value of patient recovery time + quantitative value of treatment effect) / comprehensive risk quantification value.

[0173] Risk-benefit ratio: An indicator used to comprehensively evaluate puncture path plans, which is obtained by dividing the sum of the quantitative values ​​of benefits such as puncture success rate, patient recovery time, and treatment effect by the comprehensive risk quantification value.

[0174] The calculation process for obtaining the risk-benefit ratio of different paths is as follows: 1. Analyze and calculate the comprehensive risk quantification value of different paths: According to the method of step Sb00, for different puncture path schemes, according to the patient's physiological indicators during the puncture process, the position and angle of the puncture needle, the state information of the blood vessels and other dimensions, the preset risk quantification analysis scheme and the weight allocation scheme of different dimensions are used to calculate the comprehensive risk quantification value of each path scheme. 2. Calculate the benefit quantification value of different paths: For the three benefit angles of puncture success rate, patient recovery time, and treatment effect, collect and quantify the relevant data according to the method of step Sd00 to obtain the quantitative values ​​of different path schemes in these three aspects. 3. Calculate the risk-benefit ratio of different paths: According to the risk-benefit ratio formula, add the quantification value of the puncture success rate, the quantification value of the patient recovery time, and the quantification value of the treatment effect of a specific path scheme, and then divide it by the comprehensive risk quantification value of the path scheme to obtain the risk-benefit ratio of the path.

[0175] For example, the quantified value of the puncture success rate of a certain path plan is 0.8, the quantified value of the patient recovery time is 0.7, the quantified value of the treatment effect is 0.9, and the quantified value of the comprehensive risk is 0.6. Then the risk-benefit ratio of this path is (0.8+0.7+0.9) / 0.6=4.

[0176] Display in chart form: Display the risk-benefit ratio of different pathways in the form of a chart to intuitively compare the advantages and disadvantages of different pathways. You can use a bar chart, line chart, etc., with the risk-benefit ratio of different pathways as the vertical axis and the pathway number or name as the horizontal axis. The chart can also mark the quantitative values ​​of each pathway in terms of puncture success rate, patient recovery time, treatment effect, etc., so as to have a more comprehensive understanding of the characteristics of each pathway.

[0177] Step Sf00, if the obtained comprehensive risk quantification value is less than the preset comprehensive risk quantification value, then according to the correspondence between the interval range in which the comprehensive risk quantification value falls and the early warning scheme, analyze and determine the early warning scheme corresponding to the comprehensive risk quantification value, and execute the corresponding early warning scheme.

[0178] The specific process is as follows: 1. Determine the comprehensive risk quantification value: First, determine the comprehensive risk quantification value currently calculated. This value is obtained by analyzing and calculating the patient's physiological indicators, the position and angle of the puncture needle, the state information of the blood vessels, etc. during the puncture process through the previous steps. 2. Determine the interval range: pre-set the interval range of the comprehensive risk quantification value. For example, the risk can be divided into three levels: low, medium, and high. The corresponding interval range can be set as 0-0.3 for the low risk interval, 0.3-0.6 for the medium risk interval, and 0.6 or above for the high risk interval. More detailed interval division can also be performed according to the specific puncture scenario and clinical experience. 3. Analyze and determine the early warning plan: Determine the corresponding early warning plan according to the interval range in which the comprehensive risk quantification value falls. For the low-risk interval, the early warning plan can be routine monitoring, continue to operate according to the established puncture path plan, but keep an eye on the risk factors. For example, check the patient's physiological indicators and the position of the puncture needle at regular intervals to ensure that everything is normal. For the medium-risk interval, the early warning plan can be to strengthen monitoring and adjust some operating parameters to reduce risks. For example, the patient's physiological indicators can be checked more frequently, and the advancement speed or angle of the puncture needle can be adjusted to ensure a safer puncture process. For high-risk intervals, the early warning plan may be to immediately stop the puncture operation, reassess the risk and take emergency measures. This may include collecting more patient information, reanalyzing the 3D model of the blood vessels, adjusting the treatment plan, etc.

[0179] The present application also provides a device using a puncture path planning system, comprising: an acquisition module for acquiring multi-angle B-ultrasound images; a memory for storing Figures 1 to 4 A program of a puncture path planning method shown; a processor, the program in the memory can be loaded and executed by the processor and implement the following Figures 1 to 4 A puncture path planning method is shown.

[0180] A device using a puncture path planning system also includes a medical puncture instrument and a drive device. The drive device is controlled by a processor to receive a program of a puncture path planning method and drive the medical puncture instrument to perform puncture according to the puncture path. The medical puncture instrument can be a disposable indwelling needle or other puncture needle.

[0181] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A puncture path planning system, characterized in that: include: B-ultrasound equipment; at least one processor; At least one memory storing a computer program, and at least one processor configured to implement a puncture path planning method when executing the computer program, the specific steps are as follows: Acquire a multi-angle B-ultrasound image of a target blood vessel region acquired by a probe of a B-ultrasound device, wherein the target blood vessel region involves at least one of a radial artery, a dorsalis pedis artery, and a brachial artery; For each acquired B-ultrasound image, the feature-based image matching algorithm finds the two-dimensional coordinates of the matching feature points, and then converts these two-dimensional coordinates into three-dimensional coordinates in a unified coordinate system according to the spatial position and angle information of the probe when acquiring the image. The edge detection algorithm is used to extract the vascular boundary information from the acquired B-ultrasound image, and the vascular contour is preliminarily constructed in combination with the converted three-dimensional coordinates of the feature points. Finally, the acquired B-ultrasound image is subjected to grayscale analysis to obtain the internal structure information of the blood vessels, and the preliminarily constructed vascular contour is refined to form a 3D reconstruction model of the blood vessels. Performing a vascular feature analysis based on the 3D vascular reconstruction model to obtain key vascular feature data, and simultaneously performing a surrounding tissue analysis based on the 3D vascular reconstruction model to obtain key surrounding tissue feature data; Generate vascular feature constraint information based on the acquired vascular key feature data and a preset vascular feature special evaluation method, and simultaneously generate surrounding tissue constraint information based on the acquired surrounding tissue key feature data and a preset surrounding tissue feature special evaluation method; Based on the preset safety margin constraints, vascular feature constraint information, and surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to the target location from different starting points and angles of the blood vessel; The generated puncture needle path is integrated into the 3D reconstruction model of the blood vessel for display and presented in different colors or line styles.

2. A puncture path planning system according to claim 1, characterized in that: The multi-angle B-ultrasound images of the target blood vessel area acquired by the probe of the B-ultrasound device include: Start the probe of the B-ultrasound device to perform longitudinal and transverse section scanning on the target area on the skin surface to obtain a scanned B-ultrasound image; Scanning the probe of the B-ultrasound device at different angles in the longitudinal and transverse directions to obtain a scanned B-ultrasound image; The B-ultrasound images acquired at different scanning angles are summarized as multi-angle B-ultrasound images of the target blood vessel region.

3. A puncture path planning system according to claim 2, characterized in that: The method also includes the following steps after the collected B-ultrasound image is subjected to grayscale analysis to obtain the internal structure information of the blood vessel and before the initially constructed blood vessel contour is refined to form a 3D reconstruction model of the blood vessel. A spatial positioning module preset in the probe is used to record the specific position and angle information of the probe when collecting each image, and the specific position and angle information when collecting each image is defined as spatial positioning information; The relative position relationship of the blood vessel boundaries in different images is determined based on the spatial positioning information, and the edge detection algorithm is used to refine the blood vessel boundaries. At the same time, the spatial positioning information is used to enhance the feature points found by the feature-based image matching algorithm. The three-dimensional model of the blood vessel is refined according to the refined blood vessel boundaries, enhanced feature points and spatial positioning information.

4. A puncture path planning system according to any one of claims 1 to 3, characterized in that: The preliminary constructed vascular contour is refined to form a 3D vascular reconstruction model including: acquiring infrared imaging data and magnetic resonance imaging data; Extracting blood vessel related features from infrared imaging and extracting blood vessel morphological features from magnetic resonance imaging, wherein the blood vessel related features include temperature features and the morphological features include length, diameter, curvature and branching angle of the blood vessel; Establish a feature fusion matrix to correspond and combine the features extracted from infrared imaging, the features extracted from magnetic resonance imaging, and the internal structure information of blood vessels obtained from grayscale analysis of B-ultrasound images to form fusion features; The blood vessel adjustment parameters are obtained based on the fusion feature analysis, and the blood vessel three-dimensional model is adjusted according to the blood vessel adjustment parameters and the blood vessel internal structure information obtained by grayscale analysis. At the same time, the geometric shape of the blood vessel three-dimensional model is adjusted using the morphological features in the fusion features.

5. A puncture path planning system according to claim 4, characterized in that: Based on the preset safety margin constraints, vascular feature constraint information, and surrounding tissue constraint information, a preset geometric algorithm is used to generate paths to the target location from different starting points and angles of the blood vessels, including: Collect the patient's current body position information and information related to physiological indicators, and collect multimodal imaging data of the patient in different body positions and physiological states, where the multimodal imaging data includes infrared imaging data, magnetic resonance imaging data, and B-ultrasound image data; body position information includes supine, lateral, prone, and semi-sitting positions; and physiological indicator information includes heart rate, blood pressure, respiratory rate, and blood oxygen saturation; The collected patient's current body position information, physiological index-related information and corresponding multimodal imaging data are input into the trained vascular feature analysis model to extract key features of blood vessels and surrounding tissues; Generate the current blood vessel feature constraint information according to the extracted blood vessel key feature data, and simultaneously generate the current surrounding tissue constraint information according to the acquired surrounding tissue key feature data; Based on the preset safety margin constraint, the current vascular feature constraint information, and the current surrounding tissue constraint information, a preset geometric algorithm is used to generate a path to the target position from different starting points and angles of the blood vessel.

6. A puncture path planning system according to claim 5, characterized in that: Generating a path to a target location involves: Obtain different consideration angles, including safety priority, efficiency priority and patient comfort; According to the mapping relationship between the consideration angle and the genetic algorithm adjustment scheme, the genetic algorithm adjustment scheme under different consideration angles is analyzed and determined, wherein the genetic algorithm adjustment scheme includes adjusting the chromosome encoding method, the fitness function and the genetic algorithm parameter adjustment; For different consideration angles, the genetic algorithm is run separately to obtain the optimal puncture path solutions for different angles.

7. A puncture path planning system according to claim 6, characterized in that: The steps after integrating the generated puncture needle path into the 3D reconstruction model of the blood vessel for display and presenting it in different colors or line styles are also included, as follows: Obtain the patient's physiological indicators, the position and angle of the puncture needle, and the status of the blood vessels during the puncture process; The patient's physiological index curve is displayed on the operation interface according to the patient's physiological index obtained during the puncture process, and the position and path of the puncture needle are displayed in real time using a 3D model, and the result information of the puncture needle executing the preset puncture path plan is analyzed and displayed.

8. A puncture path planning system according to claim 7, characterized in that: The method also includes steps of displaying the patient's physiological index curve on the operation interface according to the patient's physiological index obtained during the puncture process, and using the 3D model to display the position and path of the puncture needle in real time, and analyzing the result information of the puncture needle executing the preset puncture path plan for display, which are as follows: Analyze whether to detect and obtain information related to the failure of equipment components involved in the puncture process; If not, maintain the original setting; If yes, an alarm is sounded, and the type and location of the faulty equipment are determined and displayed based on the acquired fault-related information.

9. A puncture path planning system according to claim 7, characterized in that: The method also includes steps of displaying the patient's physiological index curve on the operation interface according to the patient's physiological index obtained during the puncture process, and using the 3D model to display the position and path of the puncture needle in real time, and analyzing the result information of the puncture needle executing the preset puncture path plan for display, which are as follows: According to the patient's physiological indicators, the position and angle of the puncture needle, and the state information of the blood vessels during the puncture process, a preset risk quantification analysis scheme is used to analyze and determine the risk quantification values ​​of different dimensions. The preset risk quantification analysis scheme includes a risk quantification analysis scheme based on physiological indicators, a risk quantification scheme based on the position and angle of the puncture needle, and a risk quantification scheme based on the state of the blood vessels. The dimensions include the patient's physiological indicators, the position and angle of the puncture needle, and the state information of the blood vessels during the puncture process; According to the risk quantification values ​​of different dimensions and the preset weight allocation schemes of different dimensions, the comprehensive risk quantification value is calculated; If the obtained comprehensive risk quantification value exceeds the preset comprehensive risk quantification value, the latest infrared imaging data, magnetic resonance imaging data, B-ultrasound image data, and the patient's current body position information and physiological index related information are collected, and the process jumps to the step of extracting blood vessel-related features from infrared imaging and extracting blood vessel morphological features from magnetic resonance imaging, re-analyzing the 3D model of the blood vessel, and calculating the optimal puncture path plan at different angles; According to the preset benefit quantification evaluation scheme, the calculated optimal puncture path schemes at different angles are analyzed for benefits, and the quantitative values ​​of different benefit angles are obtained. The benefit angles include puncture success rate, patient recovery time, and treatment effect; According to the preset risk quantification analysis scheme and the preset weight allocation scheme for different dimensions, the comprehensive risk quantification value of different paths is analyzed and calculated one by one, and the risk-benefit ratio of different paths is calculated according to the preset risk-benefit ratio formula, and the risk-benefit of different paths is displayed in the form of a chart, where the preset risk-benefit ratio formula is as follows: risk-benefit ratio = (quantified value of puncture success rate + quantitative value of patient recovery time + quantitative value of treatment effect) / comprehensive risk quantification value; If the obtained comprehensive risk quantification value is less than the preset comprehensive risk quantification value, the warning plan corresponding to the comprehensive risk quantification value is analyzed and determined according to the correspondence between the interval range in which the comprehensive risk quantification value falls and the warning plan, and the corresponding warning plan is executed.

10. A device using a puncture path planning system, characterized in that: include: An acquisition module, used for acquiring multi-angle B-ultrasound images; A memory, used to store a program of a puncture path planning method in a puncture path planning system according to any one of claims 1 to 9; A processor, wherein the program in the memory can be loaded and executed by the processor and implements a puncture path planning method in a puncture path planning system according to any one of claims 1 to 9; Medical puncture instruments; The driving device is controlled by the processor to receive the program of the puncture path planning method and drive the medical puncture instrument to perform puncture according to the puncture path.

Images