Blood vessel puncture system and method based on handheld ultrasonic equipment

Abstract

The invention provides a blood vessel puncture system and method based on handheld ultrasonic equipment, and relates to the technical field of puncture guidance, and the method comprises the steps: obtaining a real-time ultrasonic image of a to-be-punctured part of a patient by the handheld ultrasonic equipment, and recognizing a blood vessel boundary and a center line in the real-time ultrasonic image; calculating to obtain an optimal puncture route according to the blood vessel boundary and the center line, and then adjusting the needle point position of a puncture needle of a puncture positioning device of the handheld ultrasonic equipment according to the optimal puncture route to perform blood vessel puncture. The handheld ultrasonic equipment has the beneficial effects that the puncture positioning device is additionally arranged in the handheld ultrasonic equipment, and the image acquisition module and the puncture module are configured, so that the problem that a traditional system depends on experience and manual operation skills of medical personnel and is easily influenced by human factors is solved; compared with a robot auxiliary system, the system has very close puncture accuracy and success rate, and the system is low in complexity, low in cost and convenient to carry.

Description

Technical Field

[0001] The present invention relates to the technical field of puncture guidance, and in particular to a blood vessel puncture system and method based on a handheld ultrasonic device. Background Art

[0002] In the medical field, vascular puncture is a common and important operation, widely used in diagnosis, treatment and emergency scenarios. Traditionally, vascular puncture mainly relies on the experience and technical level of medical staff, and is completed by manually operating the ultrasound probe and puncture needle. However, with the continuous advancement of medical technology and the increasing needs of patients, higher requirements are placed on the accuracy, safety and efficiency of vascular puncture. Therefore, ultrasound-guided puncture systems came into being, aiming to assist medical staff in more accurately locating blood vessels and improving the success rate of punctures through real-time ultrasound imaging technology. The prior art includes traditional ultrasound-guided puncture systems and robot-assisted ultrasound-guided puncture systems.

[0003] The traditional ultrasound-guided puncture system is mainly composed of an ultrasound probe, a display, and a puncture needle. The system uses the ultrasound probe to obtain ultrasound images of blood vessels in real time and displays them on the display for medical staff to observe and operate. However, this system has obvious limitations. First, it is highly dependent on the experience and manual operation skills of medical staff, and is prone to operating errors due to human factors. Secondly, medical staff need to observe the display and manually operate the puncture needle at the same time during the operation, which places high demands on their coordination and attention. In addition, in complex or difficult puncture situations (such as deep blood vessels, small-caliber blood vessels, etc.), traditional systems often find it difficult to ensure the accuracy and success rate of puncture.

[0004] In order to overcome the limitations of traditional systems, robot-assisted ultrasound-guided puncture systems have emerged. This system realizes the automation or semi-automation of puncture operations through a robotic arm and a control algorithm. The robotic arm can accurately control the movement and angle of the puncture needle, and the ultrasound probe is also installed on the robotic arm to obtain vascular images in real time. The control system is responsible for processing the image and executing the control algorithm to achieve automated puncture. However, this system also has some problems. First, the system is highly complex and expensive, which limits its widespread clinical application. Second, the system is large and inconvenient to carry, and is not suitable for use at the bedside or in emergency situations. In addition, although the robot-assisted system improves the accuracy and success rate of puncture, it still requires specially trained professional operators to operate and maintain the system.

[0005] In summary, the existing ultrasound-guided puncture systems all have limitations to varying degrees. Traditional systems rely on the experience and manual operation skills of medical staff and are easily affected by human factors; while the robot-assisted system improves the accuracy and success rate of puncture, the system is highly complex, expensive, and inconvenient to carry. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention provides a vascular puncture system based on a handheld ultrasound device, comprising:

[0007] An image acquisition module, used to acquire a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device, and identify the blood vessel boundary and center line in the real-time ultrasound image;

[0008] The puncture module is connected to the image acquisition module and is used to calculate the best puncture route according to the blood vessel boundary and the center line, and then adjust the needle tip position of the puncture needle of the puncture positioning device according to the best puncture route to perform blood vessel puncture.

[0009] Preferably, the image acquisition module includes:

[0010] An image processing unit, used for acquiring a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device and performing image preprocessing on the real-time ultrasound image to obtain a preprocessed image;

[0011] The image recognition unit is used to perform image recognition on the preprocessed image using a blood vessel recognition model to identify the blood vessel boundaries of all blood vessels in the preprocessed image, and then for each identified blood vessel, mark the blood vessel to be punctured according to the selection instruction input by the medical staff, and calculate the center line of the blood vessel to be punctured.

[0012] Preferably, the image processing unit comprises:

[0013] A contrast enhancement subunit, configured to perform contrast adjustment on the real-time ultrasound image using an adaptive contrast network model, and then update the real-time ultrasound image to the real-time ultrasound image after contrast adjustment;

[0014] a resolution reconstruction subunit, connected to the contrast enhancement subunit, for performing super-resolution reconstruction on the real-time ultrasound image by using an adversarial network model, and then updating the real-time ultrasound image to the real-time ultrasound image after resolution reconstruction;

[0015] an image denoising subunit, connected to the resolution reconstruction subunit, and configured to perform denoising processing on the real-time ultrasonic image using a convolutional neural network, and then update the real-time ultrasonic image to the real-time ultrasonic image after the denoising processing;

[0016] The edge enhancement subunit is connected to the image denoising subunit and is used to perform edge enhancement on the real-time ultrasound image using an edge enhancement network model, and then use the real-time ultrasound image after edge enhancement as the preprocessed image.

[0017] Preferably, the puncture module includes:

[0018] a path calculation unit, for receiving the puncture starting area drawn by the medical staff on the skin surface in the real-time ultrasound image and the puncture target point in the blood vessel to be punctured, then selecting a number of starting points in the puncture starting area and connecting them with the puncture target point to obtain a plurality of puncture paths, then screening out the puncture paths that do not intersect with other blood vessels from all the puncture paths, and then screening out the puncture paths that have the smallest angle with the skin surface and are less than a preset angle range from each of the puncture paths that do not intersect with other blood vessels as the optimal puncture path;

[0019] The needle adjustment unit is connected to the path calculation unit and is used to move the needle tip of the puncture needle of the puncture positioning device to the starting point of the optimal puncture path, and adjust the angle between the puncture needle and the skin surface to the angle between the optimal puncture path and the skin surface, and then perform blood vessel puncture.

[0020] The present invention also provides a blood vessel puncture method based on a handheld ultrasonic device, which is applied to the above-mentioned blood vessel puncture system, comprising:

[0021] Step S1, the vascular puncture system obtains a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device, and identifies the blood vessel boundary and center line in the real-time ultrasound image;

[0022] Step S2, the blood vessel puncture system calculates the best puncture route according to the blood vessel boundary and the center line, and then adjusts the needle tip position of the puncture needle of the puncture positioning device according to the best puncture route to perform blood vessel puncture.

[0023] Preferably, the step S1 comprises:

[0024] Step S11, the vascular puncture system obtains a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device and performs image preprocessing on the real-time ultrasound image to obtain a preprocessed image;

[0025] In step S12, the vascular puncture system uses a vascular recognition model to perform image recognition on the preprocessed image, identifies the vascular boundaries of all blood vessels in the preprocessed image, and then for each identified blood vessel, identifies the blood vessel to be punctured according to the selection instruction input by the medical staff, and calculates the center line of the blood vessel to be punctured.

[0026] Preferably, the step S11 includes:

[0027] Step S111, the vascular puncture system uses an adaptive contrast network model to perform contrast adjustment on the real-time ultrasound image, and then updates the real-time ultrasound image to the real-time ultrasound image after contrast adjustment;

[0028] Step S112, the vascular puncture system uses an adversarial network model to perform super-resolution reconstruction on the real-time ultrasound image, and then updates the real-time ultrasound image to the real-time ultrasound image after resolution reconstruction;

[0029] Step S113, the vascular puncture system uses a convolutional neural network to perform denoising on the real-time ultrasonic image, and then updates the real-time ultrasonic image to the real-time ultrasonic image after the denoising process;

[0030] In step S114, the vascular puncture system uses an edge enhancement network model to perform edge enhancement on the real-time ultrasound image, and then uses the edge-enhanced real-time ultrasound image as the preprocessed image.

[0031] Preferably, the step S2 includes:

[0032] Step S21, the blood vessel puncture system receives the puncture starting area marked by the medical staff on the skin surface in the real-time ultrasound image and the puncture target point in the blood vessel to be punctured, then selects several starting points in the puncture starting area and connects them with the puncture target point to obtain multiple puncture paths, then selects the puncture path that does not intersect with other blood vessels from all the puncture paths, and then selects the puncture path with the smallest angle with the skin surface and less than a preset angle range from each of the puncture paths that does not intersect with other blood vessels as the optimal puncture path;

[0033] In step S22, the blood vessel puncture system moves the needle tip of the puncture needle of the puncture positioning device to the starting point of the optimal puncture path, and adjusts the angle between the puncture needle and the skin surface to the angle between the optimal puncture path and the skin surface, and then performs blood vessel puncture.

[0034] The above technical solution has the following advantages or beneficial effects: a puncture positioning device is added to the handheld ultrasound device, and an image acquisition module and a puncture module are configured. When the ultrasound image is subjected to image recognition processing to obtain the best puncture route, the puncture needle in the puncture positioning device is adjusted for puncture. This not only solves the problem that the traditional system relies on the experience and manual operation skills of medical staff and is easily affected by human factors; compared with the robot-assisted system, it can not only have very close puncture accuracy and success rate, but also the system of the present invention has low complexity, low cost and is easy to carry. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a structural schematic diagram of a vascular puncture system based on a handheld ultrasound device in a preferred embodiment of the present invention;

[0036] Figure 2 This is a schematic structural diagram of a puncture needle positioning device in a preferred embodiment of the present invention;

[0037] Figure 3 A schematic diagram of the puncture process in a preferred embodiment of the present invention;

[0038] Figure 4 The figure is a flow chart of a blood vessel puncture method based on a handheld ultrasound device in a preferred embodiment of the present invention;

[0039] Figure 5 A schematic diagram of a sub-process of step S1 in a preferred embodiment of the present invention;

[0040] Figure 6 This is a schematic diagram of a sub-process of step S11 in a preferred embodiment of the present invention;

[0041] Figure 7 This is a schematic diagram of a sub-process of step S2 in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0042] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. The present invention is not limited to this embodiment, and other embodiments may also fall within the scope of the present invention as long as they conform to the gist of the present invention.

[0043] In a preferred embodiment of the present invention, based on the above problems existing in the prior art, a vascular puncture system based on a handheld ultrasound device is provided, comprising:

[0044] The image acquisition module 1 is used to acquire a real-time ultrasound image of the patient's puncture site by a handheld ultrasound device, and identify the blood vessel boundary and center line in the real-time ultrasound image;

[0045] The puncture module 2 is connected to the image acquisition module 1 and is used to calculate the best puncture route according to the blood vessel boundary and center line, and then adjust the needle tip position of the puncture needle of the puncture positioning device according to the best puncture route to perform blood vessel puncture.

[0046] Specifically, in the present embodiment, a puncture positioning device 100 is added to the handheld ultrasound device, an image acquisition module 1 and a puncture module 2 are configured in the handheld ultrasound device, and when the ultrasound image is subjected to image recognition processing to obtain the best puncture route, the puncture needle in the puncture positioning device is adjusted to perform puncture. This not only solves the problem that the traditional system relies on the experience and manual operation skills of medical staff and is easily affected by human factors; compared with the robot-assisted system, it can not only have very close puncture accuracy and success rate, but also the system of the present invention has low complexity, low cost and is easy to carry.

[0047] The puncture needle positioning device 100 in this embodiment is installed on the ultrasound probe of the handheld ultrasound device, and specifically includes a position adjustment structure 110 and a propulsion mechanism. The position adjustment mechanism is mainly used to adjust the position of the puncture point of the puncture needle on the patient's skin surface and the angle of the puncture needle, such as Figure 2 As shown, it includes a two-axis driving mechanism 1110 and an angle adjustment mechanism 1120 fixed on the two-axis driving structure. The two driving mechanisms are used to adjust the position of the puncture point of the puncture needle 120 on the patient's skin surface, and the angle adjustment mechanism is used to change the inclination angle of the puncture needle to change the angle between the puncture needle and the skin surface; the propulsion mechanism is mainly used to push the puncture needle into the patient's skin for blood vessel puncture after the puncture point and the optimal puncture path are determined.

[0048] In a preferred embodiment of the present invention, the image acquisition module 1 comprises:

[0049] An image processing unit 11 is used to obtain a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device and perform image preprocessing on the real-time ultrasound image to obtain a preprocessed image;

[0050] The image recognition unit 12 is used to perform image recognition on the preprocessed image using a blood vessel recognition model, identify the blood vessel boundaries of all blood vessels in the preprocessed image, and then for each identified blood vessel, mark the blood vessel to be punctured according to the selection instruction input by the medical staff, and calculate the center line of the blood vessel to be punctured.

[0051] In this embodiment, the image processing unit 11 includes:

[0052] The contrast enhancement subunit 111 is used to perform contrast adjustment on the real-time ultrasound image using an adaptive contrast network model, and then update the real-time ultrasound image to a contrast-adjusted real-time ultrasound image;

[0053] The resolution reconstruction subunit 112 is connected to the contrast enhancement subunit 111 and is used to perform super-resolution reconstruction on the real-time ultrasound image using an adversarial network model, and then update the real-time ultrasound image to the real-time ultrasound image after resolution reconstruction;

[0054] The image denoising subunit 113 is connected to the resolution reconstruction subunit 112 and is used to perform denoising on the real-time ultrasound image using a convolutional neural network, and then update the real-time ultrasound image to the real-time ultrasound image after the denoising process;

[0055] The edge enhancement subunit 114 is connected to the image denoising subunit 113 and is used to perform edge enhancement on the real-time ultrasound image using an edge enhancement network model, and then use the edge-enhanced real-time ultrasound image as a preprocessed image.

[0056] Specifically, in the medical practice of vascular puncture, the application of the image processing unit can significantly improve the accuracy and safety of vascular puncture under ultrasound guidance. The following is a specific embodiment to illustrate how the above-mentioned image processing unit plays a role in the process of vascular puncture:

[0057] Background of the Embodiments

[0058] During vascular puncture (such as central venous catheterization, arterial blood sampling, etc.), doctors rely on ultrasound images to locate blood vessels and ensure that the puncture needle accurately enters the target blood vessel. However, traditional ultrasound images may be affected by problems such as low contrast, insufficient resolution, and noise interference, resulting in blurred blood vessel boundaries, increasing the difficulty and risk of puncture.

[0059] Image Processing Unit Application Steps

[0060] During the vascular puncture preparation phase, the doctor first obtains a real-time ultrasound image of the patient.

[0061] The contrast enhancement subunit processes the image using an adaptive contrast network model, automatically adjusting the contrast between the blood vessels and surrounding tissues in the image to make the blood vessel boundaries clearer.

[0062] The updated image is displayed on the ultrasound machine screen, allowing doctors to more easily identify the location and depth of blood vessels.

[0063] Next, the resolution reconstruction subunit uses an adversarial network model to perform super-resolution reconstruction on the contrast-enhanced image.

[0064] By learning features from a large number of high-resolution images, the model is able to generate ultrasound images with higher resolution and more details, further refining vascular structures and reducing blur and artifacts.

[0065] The images with improved resolution provide doctors with more detailed information about the vascular structure, helping to plan the puncture path more accurately.

[0066] During the image transmission or acquisition process, noise may be introduced, affecting the image quality.

[0067] The image denoising subunit uses a convolutional neural network to denoise the image, effectively removing random noise and speckle noise in the image and maintaining the integrity and clarity of the blood vessel contour.

[0068] The denoised images are cleaner and smoother, providing doctors with clearer visual guidance.

[0069] Finally, the edge enhancement subunit uses the edge enhancement network model to further process the image.

[0070] This model can enhance the contrast of blood vessel edges, make blood vessel contours more prominent, and improve the visual effect and readability of images.

[0071] In edge-enhanced images, the boundaries between blood vessels and surrounding tissues are clearer, allowing doctors to more accurately determine the puncture point and direction.

[0072] Through the comprehensive application of the above image processing units, doctors can obtain clearer, more accurate, and higher-resolution ultrasound images during vascular puncture. This not only improves the success rate of puncture, but also reduces the risk of complications caused by mispuncture. At the same time, the automation and intelligence of image processing technology also reduces the workload of doctors and improves the efficiency and quality of medical services.

[0073] After image preprocessing, it is necessary to identify the blood vessel boundaries and the center lines of the blood vessels in the image;

[0074] The pre-processed images are updated and displayed on the screen of the ultrasound device for further analysis and operation by medical staff.

[0075] The image recognition unit receives the preprocessed image and uses a blood vessel recognition model to perform image recognition on it. After extensive training, the model can automatically recognize the blood vessel structure in the image and mark the boundaries of the blood vessels.

[0076] For each identified blood vessel, the image recognition unit will further calculate its center line. The center line calculation is based on the boundary information of the blood vessel and the geometric center path of the blood vessel is obtained through algorithm processing.

[0077] For this step, in the process of calculating the centerline of the blood vessel, the image recognition unit will use the boundary information of the blood vessel to extract the geometric center path of the blood vessel through a series of image processing and algorithm analysis steps. The following is a specific embodiment to illustrate the process of calculating the centerline:

[0078] Background of the embodiment: In ultrasound images, blood vessels are usually presented as dark areas with a certain width and length, and their boundaries are formed by the contrast difference between the blood vessel wall and the surrounding tissue. In order to perform accurate blood vessel puncture or other medical operations, it is necessary to determine the centerline of the blood vessel, that is, the geometric center path of the blood vessel.

[0079] Centerline calculation process:

[0080] 1. Boundary extraction:

[0081] The image recognition unit first processes the preprocessed ultrasound image using an edge detection algorithm (such as the Canny edge detector) to extract the boundaries of the blood vessels. This step generates a binary image in which the blood vessel boundaries are marked as white (or highlighted) and the rest are black (or low-brightness).

[0082] 2. Boundary refinement:

[0083] Next, the extracted blood vessel boundaries are thinned. A thinning algorithm (such as the Zhang-Suen thinning algorithm) is used to reduce the boundary pixels to a single pixel width to form the skeleton of the blood vessel. The result of this step is a thinner line that roughly represents the center location of the blood vessel, but may contain some spurs and branches.

[0084] 3. Centerline extraction:

[0085] After the skeleton is refined, further processing is required to extract smooth centerlines. This usually involves steps such as removing burrs, smoothing curves, and processing branch points.

[0086] A common method is to use morphological operations (such as corrosion and dilation) to smooth the skeleton lines and use connected component analysis to identify and process branch points. At the branch points, the main branches can be selected or some branches can be ignored according to the actual morphology of the blood vessels and the puncture requirements.

[0087] Another method is to use an algorithm based on graph theory or path search (such as Dijkstra algorithm or A* algorithm) to search for the shortest path or optimal path from one end of the blood vessel to the other end on the skeleton. This path is the desired center line.

[0088] 4. Post-processing:

[0089] Finally, the extracted centerlines are post-processed to ensure their smoothness and accuracy. This may include removing isolated short line segments, smoothing minor fluctuations on the curve, and adjusting the position of the centerlines to better reflect the geometric center of the vessel.

[0090] After the above steps, the image recognition unit will output one or more smooth curves, which represent the geometric center path of the blood vessel in the ultrasound image. Medical staff can plan the puncture path based on this centerline information to ensure that the puncture needle can enter along the center of the blood vessel, thereby improving the accuracy and safety of the puncture.

[0091] It should be noted that since the morphology of blood vessels and the quality of ultrasound images may vary from patient to patient, in practical applications, it may be necessary to adjust and optimize the algorithm parameters according to the specific situation to obtain the best centerline extraction effect.

[0092] Further, the medical staff inputs a selection instruction to identify the blood vessel to be punctured based on the blood vessel image and centerline information displayed on the ultrasound device screen. This is usually done through a touch screen or an external controller, and the medical staff can select the clearest, straightest and easiest blood vessel to puncture as the target.

[0093] The blood vessel to be punctured is clearly marked and its center line is clearly visible, which provides precise visual guidance for subsequent puncture operations.

[0094] By applying the image acquisition module including the image processing unit and the image recognition unit, the vascular puncture process becomes more accurate, efficient and safe. Medical staff can intuitively see the boundary and centerline information of the blood vessel, making it easier to select the best puncture point and puncture path. This not only improves the success rate of puncture, but also reduces the risk of complications caused by mispuncture. At the same time, the application of this module also reduces the workload of medical staff and improves the efficiency and quality of medical services.

[0095] In a preferred embodiment of the present invention, the puncture module 2 comprises:

[0096] The path calculation unit 21 is used to receive the puncture starting area and the puncture target point in the blood vessel to be punctured drawn by the medical staff on the skin surface in the real-time ultrasound image, and then select a number of starting points in the puncture starting area to connect with the puncture target point to obtain multiple puncture paths, and then select the puncture path that does not intersect with other blood vessels from all the puncture paths, and then select the puncture path with the smallest angle with the skin surface and less than a preset angle range from each puncture path that does not intersect with other blood vessels as the best puncture path;

[0097] The needle adjustment unit 22 is connected to the path calculation unit 21 and is used to move the needle tip of the puncture needle of the puncture positioning device to the starting point of the optimal puncture path, and adjust the angle between the puncture needle and the skin surface to the optimal angle between the puncture path and the skin surface, and then perform blood vessel puncture.

[0098] Specifically, during the implementation of blood vessel puncture, the application of the puncture module including the path calculation unit and the needle adjustment unit can greatly improve the accuracy and safety of the puncture. The following is a specific embodiment, which details the application process of these modules during blood vessel puncture:

[0099] Patients need to undergo precise vascular puncture operations, such as central venous catheterization. Medical staff use a puncture system combined with real-time ultrasound images, which includes an image processing unit, an image recognition unit, and a puncture module.

[0100] 1. Application of Path Computation Unit

[0101] Step 1: Select the puncture starting area and target point

[0102] Medical staff use the touch screen of the ultrasound device or an external controller to mark a puncture starting area on the patient's skin surface in the real-time ultrasound image.

[0103] At the same time, select a clear puncture target point inside the blood vessel to be punctured, which is usually the center of the blood vessel or a site that is easy to puncture.

[0104] Step 2: Generate and filter puncture paths

[0105] After receiving the information of the starting area and the target point, the path calculation unit automatically selects several starting points in the starting area and connects them with the target point to generate multiple possible puncture paths.

[0106] Subsequently, the path calculation unit uses the blood vessel boundary information provided by the image processing unit to screen out those puncture paths that do not intersect with other blood vessels.

[0107] Among these non-intersecting paths, the paths with the smallest angle with the skin surface and less than a preset angle range (such as 30 degrees) are further screened out as candidate paths.

[0108] Finally, the path with the smallest angle and meeting all conditions is selected as the optimal puncture path.

[0109] 2. Application of needle adjustment unit

[0110] Step 3: Adjust the position of the puncture needle

[0111] The needle adjustment unit receives the optimal puncture path information calculated by the path calculation unit, including the starting point position and the puncture angle.

[0112] The needle adjustment unit drives the puncture positioning device (the two-axis driving mechanism in the aforementioned embodiment and the angle adjustment mechanism fixed on the two-axis driving structure) to accurately move the needle tip of the puncture needle to the starting point of the optimal puncture path.

[0113] Step 4: Adjust the puncture angle

[0114] The needle adjustment unit is also responsible for adjusting the angle between the puncture needle and the skin surface to ensure that the angle is consistent with the angle between the optimal puncture path and the skin surface.

[0115] After the adjustment is completed, Figure 3 As shown, the puncture needle is at the puncture point B of the best puncture path in the starting area A and at an angle α with the skin, and a puncture operation is performed toward the puncture target point C.

[0116] Step 5: Perform the puncture operation

[0117] After the medical staff confirms that the position and angle of the puncture needle are correct, they start the puncture operation.

[0118] The puncture needle enters the blood vessel along the optimal puncture path. Due to the precise path planning and appropriate angle, the puncture process is smoother and safer.

[0119] Implementation effect: By applying the puncture module including the path calculation unit and the needle adjustment unit, the vascular puncture process becomes more intelligent and precise. Medical staff do not need to rely on experience and feel to perform puncture operations, but rely on the optimal puncture path and precise angle adjustment provided by the system to achieve puncture. This not only improves the success rate of puncture, but also reduces the risk of complications caused by mispuncture. At the same time, the application of this module also reduces the workload of medical staff and improves the efficiency and quality of medical services.

[0120] The present invention also provides a blood vessel puncture method based on a handheld ultrasound device, which is applied to the above-mentioned blood vessel puncture system, such as Figure 4 As shown, including:

[0121] Step S1, the vascular puncture system obtains a real-time ultrasound image of the patient's puncture site by a handheld ultrasound device, and identifies the blood vessel boundary and center line in the real-time ultrasound image;

[0122] Step S2, the blood vessel puncture system calculates the best puncture route according to the blood vessel boundary and center line, and then adjusts the needle tip position of the puncture needle of the puncture positioning device according to the best puncture route to perform blood vessel puncture.

[0123] In a preferred embodiment of the present invention, Figure 5 As shown, step S1 includes:

[0124] Step S11, the vascular puncture system obtains a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device and performs image preprocessing on the real-time ultrasound image to obtain a preprocessed image;

[0125] In step S12, the vascular puncture system uses a vascular recognition model to perform image recognition on the preprocessed image, identifies the vascular boundaries of all blood vessels in the preprocessed image, and then for each identified blood vessel, identifies the blood vessel to be punctured according to the selection instruction input by the medical staff, and calculates the center line of the blood vessel to be punctured.

[0126] In a preferred embodiment of the present invention, Figure 6 As shown, step S11 includes:

[0127] Step S111, the vascular puncture system uses an adaptive contrast network model to perform contrast adjustment on the real-time ultrasound image, and then updates the real-time ultrasound image to the real-time ultrasound image after contrast adjustment;

[0128] Step S112, the vascular puncture system uses the adversarial network model to perform super-resolution reconstruction on the real-time ultrasound image, and then updates the real-time ultrasound image to the real-time ultrasound image after resolution reconstruction;

[0129] Step S113, the vascular puncture system uses a convolutional neural network to perform denoising on the real-time ultrasonic image, and then updates the real-time ultrasonic image to the real-time ultrasonic image after denoising;

[0130] In step S114, the vascular puncture system uses an edge enhancement network model to perform edge enhancement on the real-time ultrasound image, and then uses the edge-enhanced real-time ultrasound image as a preprocessed image.

[0131] In a preferred embodiment of the present invention, Figure 7 As shown, step S2 includes:

[0132] Step S21, the blood vessel puncture system receives the puncture starting area and the puncture target point in the blood vessel to be punctured drawn by the medical staff on the skin surface in the real-time ultrasound image, then selects several starting points in the puncture starting area and connects them with the puncture target point to obtain multiple puncture paths, then selects the puncture path that does not intersect with other blood vessels from all the puncture paths, and then selects the puncture path with the smallest angle with the skin surface and less than a preset angle range from each puncture path that does not intersect with other blood vessels as the best puncture path;

[0133] In step S22, the blood vessel puncture system moves the needle tip of the puncture needle of the puncture positioning device to the starting point of the optimal puncture path, and adjusts the angle between the puncture needle and the skin surface to the angle between the optimal puncture path and the skin surface, and then performs blood vessel puncture.

[0134] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the contents of this specification and illustrations should be included in the protection scope of the present invention.

Claims

1. A vascular puncture system based on a handheld ultrasound device, characterized in that: include: An image acquisition module, used to acquire a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device, and identify the blood vessel boundary and center line in the real-time ultrasound image; The puncture module is connected to the image acquisition module and is used to calculate the best puncture route according to the blood vessel boundary and the center line, and then adjust the needle tip position of the puncture needle of the puncture positioning device of the handheld ultrasound device according to the best puncture route to perform blood vessel puncture.

2. The vascular puncture system according to claim 1, characterized in that: The image acquisition module comprises: An image processing unit, used for acquiring a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device and performing image preprocessing on the real-time ultrasound image to obtain a preprocessed image; The image recognition unit is used to perform image recognition on the preprocessed image using a blood vessel recognition model to identify the blood vessel boundaries of all blood vessels in the preprocessed image, and then for each identified blood vessel, mark the blood vessel to be punctured according to the selection instruction input by the medical staff, and calculate the center line of the blood vessel to be punctured.

3. The vascular puncture system according to claim 2, characterized in that: The image processing unit comprises: A contrast enhancement subunit, configured to perform contrast adjustment on the real-time ultrasound image using an adaptive contrast network model, and then update the real-time ultrasound image to the real-time ultrasound image after contrast adjustment; a resolution reconstruction subunit, connected to the contrast enhancement subunit, for performing super-resolution reconstruction on the real-time ultrasound image by using an adversarial network model, and then updating the real-time ultrasound image to the real-time ultrasound image after resolution reconstruction; an image denoising subunit, connected to the resolution reconstruction subunit, and configured to perform denoising processing on the real-time ultrasonic image using a convolutional neural network, and then update the real-time ultrasonic image to the real-time ultrasonic image after the denoising processing; The edge enhancement subunit is connected to the image denoising subunit and is used to perform edge enhancement on the real-time ultrasound image using an edge enhancement network model, and then use the real-time ultrasound image after edge enhancement as the preprocessed image.

4. The vascular puncture system according to claim 2, characterized in that: The puncture module includes: a path calculation unit, for receiving the puncture starting area drawn by the medical staff on the skin surface in the real-time ultrasound image and the puncture target point in the blood vessel to be punctured, then selecting a number of starting points in the puncture starting area and connecting them with the puncture target point to obtain a plurality of puncture paths, then screening out the puncture paths that do not intersect with other blood vessels from all the puncture paths, and then screening out the puncture paths that have the smallest angle with the skin surface and are less than a preset angle range from each of the puncture paths that do not intersect with other blood vessels as the optimal puncture path; The needle adjustment unit is connected to the path calculation unit and is used to move the needle tip of the puncture needle of the puncture positioning device to the starting point of the optimal puncture path, and adjust the angle between the puncture needle and the skin surface to the angle between the optimal puncture path and the skin surface, and then perform blood vessel puncture.

5. A blood vessel puncture method based on a handheld ultrasound device, characterized in that: A vascular puncture system as claimed in any one of claims 1 to 4, comprising: Step S1, the vascular puncture system obtains a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device, and identifies the blood vessel boundary and center line in the real-time ultrasound image; Step S2, the blood vessel puncture system calculates the best puncture route according to the blood vessel boundary and the center line, and then adjusts the needle tip position of the puncture needle of the puncture positioning device according to the best puncture route to perform blood vessel puncture.

6. The blood vessel puncture method according to claim 5, characterized in that: The step S1 comprises: Step S11, the vascular puncture system obtains a real-time ultrasound image of the patient's puncture site by the handheld ultrasound device and performs image preprocessing on the real-time ultrasound image to obtain a preprocessed image; In step S12, the vascular puncture system uses a vascular recognition model to perform image recognition on the preprocessed image, identifies the vascular boundaries of all blood vessels in the preprocessed image, and then for each identified blood vessel, identifies the blood vessel to be punctured according to the selection instruction input by the medical staff, and calculates the center line of the blood vessel to be punctured.

7. The blood vessel puncture method according to claim 6, characterized in that: The step S11 comprises: Step S111, the vascular puncture system uses an adaptive contrast network model to perform contrast adjustment on the real-time ultrasound image, and then updates the real-time ultrasound image to the real-time ultrasound image after contrast adjustment; Step S112, the vascular puncture system uses an adversarial network model to perform super-resolution reconstruction on the real-time ultrasound image, and then updates the real-time ultrasound image to the real-time ultrasound image after resolution reconstruction; Step S113, the vascular puncture system uses a convolutional neural network to perform denoising on the real-time ultrasonic image, and then updates the real-time ultrasonic image to the real-time ultrasonic image after the denoising process; In step S114, the vascular puncture system uses an edge enhancement network model to perform edge enhancement on the real-time ultrasound image, and then uses the edge-enhanced real-time ultrasound image as the preprocessed image.

8. The blood vessel puncture method according to claim 6, characterized in that: The step S2 includes: Step S21, the blood vessel puncture system receives the puncture starting area marked by the medical staff on the skin surface in the real-time ultrasound image and the puncture target point in the blood vessel to be punctured, then selects several starting points in the puncture starting area and connects them with the puncture target point to obtain multiple puncture paths, then selects the puncture path that does not intersect with other blood vessels from all the puncture paths, and then selects the puncture path with the smallest angle with the skin surface and less than a preset angle range from each of the puncture paths that does not intersect with other blood vessels as the optimal puncture path; In step S22, the blood vessel puncture system moves the needle tip of the puncture needle of the puncture positioning device to the starting point of the optimal puncture path, and adjusts the angle between the puncture needle and the skin surface to the angle between the optimal puncture path and the skin surface, and then performs blood vessel puncture.

Images