Auxiliary femoral artery puncture device

Abstract

The invention discloses a femoral artery auxiliary puncture device, and relates to the technical field of medical instruments, and the femoral artery auxiliary puncture device comprises an ultrasonic image acquisition module which is responsible for acquiring real-time ultrasonic images of a femoral artery and surrounding tissues; the A I identification and analysis module is responsible for identifying and analyzing an anatomical structure in the ultrasonic image; the positioning and navigation module is responsible for guiding puncture operation according to an ultrasonic image and an A I recognition result; the puncture equipment control module is responsible for accurately controlling the position and depth of puncture equipment; the safety and monitoring module is responsible for recording and analyzing ultrasonic images, AI recognition results and puncture process data of each operation, when the femoral artery of a patient is punctured, a motor is started, the motor enables a universal shaft to rotate, a plurality of rotating wheels can be made to get close to one another finally, the puncture head is limited, the puncture head is prevented from shaking, and the puncture accuracy is improved. Extra tissue damage and pain stimulation are prevented, and the comfort level of a patient is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and particularly to a femoral artery assisted puncture device. Background Art

[0002] The femoral artery is the main trunk of the lower limb arteries, continuing from the external iliac artery and entering the femoral triangle deep to the midpoint of the inguinal ligament. It gives off the superficial epigastric artery, superficial circumflex iliac artery and external pudendal artery anteriorly, and the deep femoral artery posteriorly. The deep femoral artery gives off the medial and lateral circumflex femoral arteries and the perforating arteries. The femoral artery and the deep femoral artery are almost on the same vertical line, with the characteristics of thick diameter, shallow position and being in front of the femoral head. Transfemoral artery puncture is one of the common approaches in interventional diagnosis and treatment, especially playing an important role in cardiovascular interventional therapy. As the main trunk of the lower limb arteries, the femoral artery has the advantages of thick diameter, shallow position, being easy to touch and puncture, making it the main puncture site. The transfemoral artery puncture technique is applied in almost most arterial cavity treatments.

[0003] After retrieval, the patent with Chinese patent number CN216535407U discloses a femoral artery assisted puncture device for cerebrovascular angiography in neurology, belonging to the technical field of medical devices. The key points of its technical solution include a flat plate. The left and right sides of the flat plate are respectively fixedly connected with a left stay cord and a right stay cord. A screw cylinder is fixedly connected through the center of the flat plate. A screw rod is threadedly connected inside the screw cylinder. Three movable rods move inside three movable cylinders respectively. Pushing the outer plate inward can drive the inner plate to move inward, so as to reduce the size of the semi-circular structure formed by the flat plate, the left stay cord and the right stay cord from the inside, so as to adapt to patients with different body types and weights, make the device more firmly fixed, have good stability, be convenient for auxiliary hemostasis work, and have strong applicability. It solves the problem that most doctors in current hospitals choose sandbags for compression hemostasis. Due to the easy displacement of sandbags, poor stability and poor hemostasis effect, various complications may occur at the transfemoral artery puncture site of patients, and even local skin damage or necrosis may occur.

[0004] However, in the actual use process of the above device, due to individual differences in the anatomical structure of patients, such as the position and depth of arteries, etc., it may be difficult for the device to accurately locate the puncture point on some patients. At the same time, it is difficult to limit the needle during puncture and difficult to avoid the needle from shaking, which may cause additional tissue damage and pain stimulation. Therefore, a femoral artery assisted puncture device needs to be proposed. Summary of the Invention

[0005] The object of the present invention is to solve the disadvantages in the prior art that due to individual differences in the anatomical structure of patients, such as the position and depth of arteries, etc., it may be difficult to accurately locate the puncture point on some patients, and at the same time, it is difficult to limit the position of the needle during puncture, and it is difficult to avoid the needle from shaking, which may cause additional tissue damage and pain stimulation, and to propose a femoral artery assisted puncture device.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A femoral artery assisted puncture device, comprising:

[0008] An ultrasonic image acquisition module: responsible for acquiring real-time ultrasonic images of the femoral artery and surrounding tissues;

[0009] An AI recognition and analysis module: responsible for recognizing and analyzing the anatomical structure in the ultrasonic image;

[0010] A positioning and navigation module: responsible for guiding the puncture operation based on the ultrasonic image and the AI recognition result;

[0011] A puncture device control module: responsible for precisely controlling the position and depth of the puncture device;

[0012] A safety and monitoring module: responsible for recording and analyzing the ultrasonic images, AI recognition results and puncture process data of each operation;

[0013] The puncture device control module is arranged on one side of the flat plate. The flat plate is provided with a puncture opening for puncture. One side of the flat plate is fixedly connected with a fixed ring. One side of the fixed ring is fixedly installed with a motor. The output end of the motor is provided with a universal shaft. The side of the universal shaft far from the motor is fixedly connected with a threaded rod. The threaded rod is threadedly connected with a second connecting plate. One side of the second connecting plate close to the puncture opening is fixedly connected with a second connecting ring. The upper part of the second connecting ring is fixedly connected with a plurality of fixing blocks. One side of the fixing block far from the second connecting ring is fixedly connected with a first connecting ring. The first connecting ring is rotatably connected with a plurality of first rotating shafts. The lower part of the first rotating shaft is fixedly connected with a sliding sleeve. The sliding sleeve is slidably connected with a first connecting plate. The first connecting plate is rotatably connected with the fixed ring. One side of the first connecting plate far from the fixed ring is rotatably connected with a rotating wheel for preventing the puncture needle from shaking.

[0014] The above technical solution further includes:

[0015] The universal shaft is rotatably connected with the fixed ring. The threaded rod is rotatably connected with the fixed ring. The second connecting ring is rotatably connected with the fixed ring.

[0016] One side of the flat plate is fixedly connected with a first extension belt, and a first magic tape is arranged on the side of the first extension belt away from the flat plate. The side of the flat plate away from the first extension belt is fixedly connected with a second extension belt, and a second magic tape is arranged on the side of the second extension belt away from the flat plate.

[0017] The upper part of the flat plate is fixedly connected with a disinfectant storage box for storing disinfectant, and a liquid adding port for adding disinfectant is opened on the upper part of the disinfectant storage box.

[0018] One side of the disinfectant storage box is fixedly connected with a second infusion tube, and a pump body is arranged at the connection between the disinfectant storage box and the second infusion tube.

[0019] The side of the second infusion tube away from the disinfectant storage box is fixedly connected with a first infusion tube. The first infusion tube is fixedly connected to the flat plate. A plurality of disinfection nozzles for spraying disinfectant are fixedly installed on the side of the first infusion tube away from the second infusion tube.

[0020] The AI recognition and analysis module uses a convolutional neural network algorithm to extract features from ultrasonic images, identify blood vessel structures and lesion areas. The specific steps of the convolutional neural network algorithm are as follows:

[0021] Input data representation: Assume that the input data is a multi-dimensional array, denoted as X. The dimensions of X are batch size, number of channels, height, and width, representing the number of input samples, the number of channels, the image height, and the image width respectively;

[0022] Convolution operation: Perform a sliding window convolution operation on the input data through a convolution kernel to extract features from the input data. Assume that the dimensions of the convolution kernel are number of channels, convolution kernel height, and convolution kernel width, denoted as W. The convolution operation can be represented by the following formula: Z[i,j] = sum(sum(sum(X[:,:,i:i+h,j:j+w]*W))) + b, where Z is the output of the convolution operation, representing the feature map, i and j represent the positions of the feature map, h and w represent the height and width of the convolution kernel respectively, and b is the bias term;

[0023] Activation function: After the convolution operation, apply ReLU to perform a non-linear transformation on the feature map to increase the expression ability of the network;

[0024] Pooling operation: Select the average value as the pooling result within the local perception area to reduce the spatial dimension of the feature map and retain important feature information;

[0025] Fully connected operation: After multiple convolution and pooling operations, the feature map is unfolded into a one-dimensional vector, and classification or regression tasks are performed through a fully connected layer. Each neuron in the fully connected layer is connected to all neurons in the previous layer, and the mapping relationship between the input and output is established by learning weights and biases;

[0026] Loss function: In CNN, the cross-entropy loss function is used to measure the gap between the prediction result of the model and the true label;

[0027] Backpropagation algorithm: The backpropagation algorithm is used to calculate the gradient of the loss function with respect to the model parameters, and gradient descent is used to update the model parameters to continuously optimize the performance of the model.

[0028] The specific steps of the positioning and navigation module are as follows:

[0029] Positioning and planning: Determine the optimal puncture point position on the ultrasound image, and combine the AI recognition result and the puncture point position to plan the optimal puncture path

[0030] Navigation and adjustment: During the puncture process, use ultrasound imaging and AI recognition technology to real-time track the position and attitude of the puncture needle, compare the real-time information with the preset puncture path, adjust the puncture direction and depth in a timely manner, and perform fine-tuning on the puncture device according to the real-time navigation result and AI recognition feedback. At the same time, monitor various safety parameters during the puncture process;

[0031] Complete the puncture: When the puncture needle reaches the predetermined position, stop the puncture operation and perform necessary treatment on the puncture site.

[0032] The present invention has the following beneficial effects:

[0033] 1. In the present invention, when puncturing the femoral artery of a patient, the motor is started, and the motor will cause the universal shaft to rotate. Eventually, multiple rotating wheels can approach each other to achieve the limiting effect on the puncture head, prevent the puncture head from shaking, prevent additional tissue damage and pain stimulation, and improve the comfort of the patient.

[0034] 2. In the present invention, the real-time ultrasound image of the patient's femoral artery and surrounding tissues is obtained through the ultrasound image acquisition module, and then the accurate position of the femoral artery is automatically identified and located through the AI recognition and analysis module, and parameters such as blood vessel structure and blood flow velocity are analyzed, providing suggestions for doctors on puncture path planning, puncture depth control, etc., and assisting doctors to make more accurate and safe decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a schematic structural diagram of a femoral artery assisted puncture device proposed by the present invention;

[0036] Figure 2This is the first structural schematic diagram in the present invention;

[0037] Figure 3 This is the second structural schematic diagram in the present invention;

[0038] Figure 4 is Figure 2 the enlarged schematic diagram of the structure at position A in

[0039] Figure 5 is Figure 3 the enlarged schematic diagram of the structure at position B in

[0040] Figure 6 This is the system block diagram of the present invention.

[0041] In the figure: 1, flat plate; 2, first extension belt; 3, second extension belt; 4, first magic tape; 5, second magic tape; 6, disinfectant storage box; 7, fixing ring; 8, puncture port; 9, liquid filling port; 10, first infusion tube; 11, disinfection spray head; 12, first connecting plate; 13, first rotating shaft; 14, rotating wheel; 15, second infusion tube; 16, pump body; 17, first connecting ring; 18, second connecting ring; 19, universal shaft; 20, threaded rod; 21, second connecting plate; 22, sliding sleeve; 23, fixing block; 24, motor. Detailed implementation manners

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] As Figures 1-6 shown, a femoral artery assisted puncture device includes:

[0044] Ultrasound image acquisition module: responsible for acquiring real-time ultrasound images of the femoral artery and surrounding tissues;

[0045] AI recognition and analysis module: responsible for recognizing and analyzing anatomical structures in ultrasound images;

[0046] Positioning and navigation module: responsible for guiding the puncture operation based on ultrasound images and AI recognition results;

[0047] Puncture device control module: responsible for precisely controlling the position and depth of the puncture device;

[0048] Safety and monitoring module: responsible for recording and analyzing ultrasound images, AI recognition results, and puncture process data of each operation;

[0049] The puncture device control module is arranged on one side of the flat plate 1. The flat plate 1 is provided with a puncture port 8 for puncture. One side of the flat plate 1 is fixedly connected with a fixing ring 7. One side of the fixing ring 7 is fixedly installed with a motor 24. The output end of the motor 24 is provided with a universal shaft 19. The side of the universal shaft 19 away from the motor 24 is fixedly connected with a threaded rod 20. The threaded rod 20 is threadedly connected with a second connecting plate 21. One side of the second connecting plate 21 close to the puncture port 8 is fixedly connected with a second connecting ring 18. The upper part of the second connecting ring 18 is fixedly connected with a plurality of fixing blocks 23. One side of the fixing block 23 away from the second connecting ring 18 is fixedly connected with a first connecting ring 17. The first connecting ring 17 is rotatably connected with a plurality of first rotating shafts 13. The lower part of the first rotating shaft 13 is fixedly connected with a sliding sleeve 22. The sliding sleeve 22 is slidably connected with a first connecting plate 12. The first connecting plate 12 is rotatably connected with the fixing ring 7. One side of the first connecting plate 12 away from the fixing ring 7 is rotatably connected with a rotating wheel 14 for preventing the puncture needle from shaking.

[0050] The working principle of a femoral artery assisted puncture device proposed by the present invention is that the ultrasonic image acquisition module obtains real-time ultrasonic images of the femoral artery and surrounding tissues through an ultrasonic probe, and then the ultrasonic imaging device receives, processes and displays ultrasonic signals to generate images for doctors to refer to. The AI recognition and analysis module automatically recognizes and locates the accurate position of the femoral artery, and analyzes parameters such as blood vessel structure and blood flow velocity to provide puncture suggestions for doctors. The puncture device control module controls the advancing, retreating, rotating and other actions of the puncture device, and precisely controls the position and depth of the puncture device according to the information guided by AI and ultrasound to ensure a safe and effective puncture process.

[0051] During puncture, the motor 24 fixedly installed outside the fixing ring 7 is started. The motor 24 will cause the universal shaft 19 arranged at the output end to rotate. The rotation of the universal shaft 19 will cause the fixedly connected threaded rod 20 to rotate. The rotation of the threaded rod 20 will cause the second connecting plate 21 connected by thread to move on the threaded rod 20. The movement of the second connecting plate 21 will drive the fixedly connected second connecting ring 18 to rotate on the fixing ring 7. The second connecting ring 18 will drive the first connecting ring 17 to rotate together through a plurality of fixedly connected fixing blocks 23.

[0052] When the first connecting ring 17 rotates, it will cause the rotatably connected first rotating shaft 13 to rotate. The first rotating shaft 13 will drive the fixedly connected sliding sleeve 22 to slide on the first connecting plate 12. The first connecting plate 12 is rotatably connected with the fixing ring 7, that is, the inclination angle of the first connecting plate 12 can be changed, that is, the rotating wheels 14 rotatably connected to the outside of the plurality of first connecting plates 12 can be made to approach the center of the puncture port 8, and the rotating wheels 14 can be in rolling connection with the puncture needle, which is convenient for puncturing the patient's femoral artery.

[0053] In one embodiment, for the above-mentioned universal shaft 19, the universal shaft 19 is rotatably connected to the fixed ring 7, the threaded rod 20 is rotatably connected to the fixed ring 7, and the second connecting ring 18 is rotatably connected to the fixed ring 7.

[0054] In one embodiment, for the above-mentioned flat plate 1, one side of the flat plate 1 is fixedly connected with a first extension belt 2, a first magic tape 4 is arranged on the side of the first extension belt 2 far away from the flat plate 1, the other side of the flat plate 1 far away from the first extension belt 2 is fixedly connected with a second extension belt 3, and a second magic tape 5 is arranged on the side of the second extension belt 3 far away from the flat plate 1.

[0055] The working principle of a femoral artery assisted puncture device proposed by the present invention is that by pulling the first extension belt 2 and the second extension belt 3, the device can be fixed at the puncture site of the patient through the first magic tape 4 arranged outside the first extension belt 2 and the second magic tape 5 arranged outside the second extension belt 3.

[0056] In one embodiment, for the above-mentioned flat plate 1, a disinfectant storage box 6 for storing disinfectant is fixedly connected to the upper part of the flat plate 1, and a liquid adding port 9 for adding disinfectant is arranged on the upper part of the disinfectant storage box 6.

[0057] In one embodiment, for the above-mentioned disinfectant storage box 6, a second infusion tube 15 is fixedly connected to one side of the disinfectant storage box 6, and a pump body 16 is arranged at the connection between the disinfectant storage box 6 and the second infusion tube 15.

[0058] In one embodiment, for the above-mentioned second infusion tube 15, a first infusion tube 10 is fixedly connected to the side of the second infusion tube 15 far away from the disinfectant storage box 6, the first infusion tube 10 is fixedly connected to the flat plate 1, and a plurality of disinfection spray heads 11 for spraying disinfectant are fixedly installed on the side of the first infusion tube 10 far away from the second infusion tube 15.

[0059] The working principle of a femoral artery assisted puncture device proposed by the present invention is that the disinfectant storage box 6 is convenient for storing disinfectant, and disinfectant can be added into the disinfectant storage box 6 when the liquid adding port 9 is opened. When it is necessary to disinfect the puncture site of the patient, the pump body 16 is started, and the pump body 16 will make the disinfectant inside the disinfectant storage box 6 enter the second infusion tube 15. The disinfectant enters the first infusion tube 10 through the second infusion tube 15, and finally the disinfectant can be sprayed at the puncture site of the patient through a plurality of disinfection spray heads 11 arranged on one side of the first infusion tube 10.

[0060] In one embodiment, for the above-mentioned AI recognition and analysis module, the AI recognition and analysis module uses a convolutional neural network algorithm to extract features from ultrasonic images, identify blood vessel structures and lesion areas. The specific steps of the convolutional neural network algorithm are as follows:

[0061] Input data representation: Assume the input data is a multi-dimensional array, denoted as X. The dimensions of X are batch size, number of channels, height, and width, representing the number of input samples, the number of channels, the image height, and the image width respectively;

[0062] Convolution operation: Perform a sliding window convolution operation on the input data through a convolution kernel to extract features from the input data. Assume the dimensions of the convolution kernel are number of channels, convolution kernel height, and convolution kernel width, denoted as W. The convolution operation can be expressed by the following formula: Z[i,j] = sum(sum(sum(X[:,:,i:i+h,j:j+w]*W))) + b, where Z is the output of the convolution operation, representing the feature map, i and j represent the positions in the feature map, h and w represent the height and width of the convolution kernel respectively, and b is the bias term;

[0063] Activation function: After the convolution operation, apply ReLU to perform a non-linear transformation on the feature map to increase the expressive power of the network;

[0064] Pooling operation: Select the average value within the local perception area as the pooling result to reduce the spatial dimension of the feature map and retain important feature information;

[0065] Fully connected operation: After multiple convolution and pooling operations, expand the feature map into a one-dimensional vector and perform classification or regression tasks through a fully connected layer. Each neuron in the fully connected layer is connected to all neurons in the previous layer, and a mapping relationship between the input and output is established by learning weights and biases;

[0066] Loss function: In a CNN, use the cross-entropy loss function to measure the gap between the prediction result of the model and the true label;

[0067] Backpropagation algorithm: Use the backpropagation algorithm to calculate the gradient of the loss function with respect to the model parameters, and use gradient descent to update the model parameters to continuously optimize the performance of the model.

[0068] In one embodiment, for the above positioning and navigation module, the specific steps of the positioning and navigation module are as follows:

[0069] Positioning and planning: Determine the optimal puncture point position on the ultrasound image, and combine the AI recognition result and the puncture point position to plan the optimal puncture path

[0070] Navigation and adjustment: During the puncture process, use ultrasound imaging and AI recognition technology to real-time track the position and attitude of the puncture needle, compare the real-time information with the preset puncture path, timely adjust the puncture direction and depth, and perform fine-tuning on the puncture device according to the real-time navigation result and AI recognition feedback, while monitoring various safety parameters during the puncture process;

[0071] Complete puncture: When the puncture needle reaches the predetermined position, stop the puncture operation and perform necessary treatment on the puncture site.

[0072] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A femoral artery auxiliary puncture device, characterized in that: include: Ultrasonic image acquisition module: responsible for obtaining real-time ultrasonic images of the femoral artery and surrounding tissues; AI recognition and analysis module: responsible for recognizing and analyzing anatomical structures in ultrasound images; Positioning and navigation module: responsible for guiding puncture operations based on ultrasound images and AI recognition results; Puncture device control module: responsible for accurately controlling the position and depth of the puncture device; Safety and monitoring module: responsible for recording and analyzing ultrasound images, AI recognition results and puncture process data of each operation; The puncture device control module is arranged on one side of the plate (1); the plate (1) is provided with a puncture opening (8) for puncture; a fixing ring (7) is fixedly connected to one side of the plate (1); a motor (24) is fixedly installed on one side of the fixing ring (7); a universal shaft (19) is provided at the output end of the motor (24); a threaded rod (20) is fixedly connected to the side of the universal shaft (19) away from the motor (24); the threaded rod (20) is threadedly connected to a second connecting plate (21); a second connecting ring (18) is fixedly connected to the side of the second connecting plate (21) close to the puncture opening (8); and the second connecting A plurality of fixed blocks (23) are fixedly connected to the upper part of the ring (18); a first connecting ring (17) is fixedly connected to the side of the fixed block (23) away from the second connecting ring (18); the first connecting ring (17) is rotatably connected to a plurality of first rotating shafts (13); a sliding sleeve (22) is fixedly connected to the lower part of the first rotating shaft (13); the sliding sleeve (22) is slidably connected to the first connecting plate (12); the first connecting plate (12) is rotatably connected to the fixed ring (7); a rotating wheel (14) for preventing the puncture needle from shaking is rotatably connected to the side of the first connecting plate (12) away from the fixed ring (7).

2. A femoral artery auxiliary puncture device according to claim 1, characterized in that: The universal shaft (19) and the fixing ring (7) are rotatably connected, the threaded rod (20) and the fixing ring (7) are rotatably connected, and the second connecting ring (18) and the fixing ring (7) are rotatably connected.

3. A femoral artery auxiliary puncture device according to claim 1, characterized in that: A first extension belt (2) is fixedly connected to one side of the flat plate (1), a first Velcro (4) is provided on the side of the first extension belt (2) away from the flat plate (1), a second extension belt (3) is fixedly connected to the side of the flat plate (1) away from the first extension belt (2), and a second Velcro (5) is provided on the side of the second extension belt (3) away from the flat plate (1).

4. A femoral artery auxiliary puncture device according to claim 1, characterized in that: A disinfectant storage box (6) for storing disinfectant is fixedly connected to the upper part of the flat plate (1), and a liquid adding port (9) for adding disinfectant is provided on the upper part of the disinfectant storage box (6).

5. A femoral artery auxiliary puncture device according to claim 4, characterized in that: A second infusion tube (15) is fixedly connected to one side of the disinfectant storage box (6), and a pump body (16) is provided at the connection between the disinfectant storage box (6) and the second infusion tube (15).

6. A femoral artery auxiliary puncture device according to claim 5, characterized in that: The side of the second infusion tube (15) away from the disinfectant storage box (6) is fixedly connected to the first infusion tube (10), the first infusion tube (10) is fixedly connected to the flat plate (1), and a plurality of disinfection nozzles (11) for spraying disinfectant are fixedly installed on the side of the first infusion tube (10) away from the second infusion tube (15).

7. A femoral artery auxiliary puncture device according to claim 1, characterized in that: The AI ​​recognition and analysis module uses a convolutional neural network algorithm to extract features from ultrasound images and identify vascular structures and lesion areas. The specific steps of the convolutional neural network algorithm are as follows: Input data representation: Assume that the input data is a multidimensional array, denoted as X, where the dimensions of X are batch size, number of channels, height, and width, which represent the number of input samples, number of channels, image height, and image width, respectively; Convolution operation: Perform sliding window convolution operation on the input data through the convolution kernel to extract the features in the input data. Assuming that the dimensions of the convolution kernel are the number of channels, the height of the convolution kernel, and the width of the convolution kernel, recorded as W, the convolution operation can be expressed by the following formula: Z[i,j]=sum(sum(sum(X[:,:,i:i+h,j:j+w]*W)))+b, where, Z is the output of the convolution operation, representing the feature map, i and j represent the location of the feature map, h and w represent the height and width of the convolution kernel respectively, and b is the bias term; Activation function: After the convolution operation, ReLU is applied to perform nonlinear transformation on the feature map to increase the network’s expressive power; Pooling operation: By selecting the average value in the local perception area as the pooling result, the spatial dimension of the feature map is reduced and important feature information is retained; Fully connected operation: After multiple convolution and pooling operations, the feature map is expanded into a one-dimensional vector and passed through a fully connected layer for classification or regression tasks. Each neuron in the fully connected layer is connected to all neurons in the previous layer, and the mapping relationship between input and output is established by learning weights and biases. Loss function: In CNN, the cross entropy loss function is used to measure the gap between the model's prediction results and the true label; Back propagation algorithm: Use the back propagation algorithm to calculate the gradient of the loss function with respect to the model parameters, and use gradient descent to update the model parameters to continuously optimize the performance of the model.

8. The femoral artery auxiliary puncture device according to claim 1, characterized in that: The specific steps of the positioning and navigation module are: Positioning and planning: Determine the best puncture point location on the ultrasound image, combine AI recognition results and puncture point location to plan the optimal puncture path Navigation and adjustment: During the puncture process, the position and posture of the puncture needle are tracked in real time using ultrasound imaging and AI recognition technology, and the real-time information is compared with the preset puncture path to adjust the puncture direction and depth in time. The puncture device is fine-tuned based on the real-time navigation results and AI recognition feedback, while monitoring various safety parameters during the puncture process; Complete puncture: When the puncture needle reaches the predetermined position, stop the puncture operation and perform necessary treatment on the puncture site.

Images