An interactive method and system for guiding anesthesia and puncture.

Abstract

This invention relates to the field of image data processing technology, specifically to an interactive anesthesia puncture assistance guidance method and system, comprising: acquiring an ultrasound image sequence of the tissue region to be anesthetized at a preset initial frequency of an ultrasound probe; determining the imaging frequency penetration of each frame of ultrasound image to obtain an optimized frequency for the ultrasound probe; acquiring an updated ultrasound image sequence of the tissue region to be anesthetized; fusing the images with the same sequence values ​​in the updated ultrasound image sequence to obtain a fused ultrasound image sequence; performing tissue identification on the fused ultrasound image sequence; calculating the puncture path based on the tissue identification results; and generating puncture assistance guidance suggestions. This invention, by updating the ultrasound imaging frequency and fusing the images to obtain new imaging results, eliminates the problems of imaging blurring and identification errors caused by complex conditions within the tissue, thereby improving the safety of the anesthesia puncture assistance guidance process.

Description

Technical Field

[0001] This invention relates to the field of image data processing technology, and specifically to an interactive method and system for assisting in anesthesia and puncture. Background Technology

[0002] In the field of anesthesiology, "interactive" anesthesia refers to a system that interacts with the physician in real time during anesthesia procedures, providing dynamic visualization, positioning, navigation, and feedback to help the physician complete puncture or catheterization procedures. Specifically, this includes: (1) real-time information feedback: the system displays the needle tip position and tissue structure layers in real time based on data obtained from imaging equipment (ultrasound); (2) proactive prompts: the system issues visual or audible prompts when the needle approaches the target nerve, blood vessel, or spinal canal; and (3) safety assistance: the system identifies potential risks (such as accidental puncture of a blood vessel) and provides reminders.

[0003] In existing technologies, real-time navigation based on medical images is generally adopted. Among them, ultrasound guidance is the most common. By acquiring ultrasound images in real time and using existing image analysis algorithms to identify the puncture area, the brightness of the system-suggested puncture area is improved, thus achieving the purpose of interactive anesthesia puncture assistance guidance.

[0004] Existing problems: During interactive anesthesia guidance, the tissue area to be anesthetized is highly complex. When there is excessive fat layer or excessive bleeding, the fixed frequency of ultrasound cannot present a high level of penetration in thick fat areas or clear ultrasound results in noisy areas with excessive bleeding. Therefore, when providing puncture guidance based on real-time ultrasound images, errors in guidance due to image clarity issues are very likely to occur, and even medical accidents may occur. Summary of the Invention

[0005] This invention provides an interactive anesthesia puncture guidance method and system to solve existing problems.

[0006] The interactive anesthesia puncture guidance method and system of the present invention adopts the following technical solution:

[0007] One embodiment of the present invention provides an interactive anesthesia puncture guidance method, the method comprising the following steps:

[0008] At the preset initial frequency of the ultrasound probe, acquire an ultrasound image sequence of the area to be anesthetized and punctured within a preset duration.

[0009] The tissue heterogeneity of each ultrasound image is determined based on the gradient amplitude of pixels within each frame of the ultrasound image sequence and edge pixels.

[0010] Based on the tissue heterogeneity of each ultrasound image and the grayscale value of the pixels in the ultrasound image, the imaging frequency penetration of each ultrasound image is determined; based on the imaging frequency penetration of all ultrasound images in the ultrasound image sequence, combined with the preset initial frequency of the ultrasound probe, the optimized frequency of the ultrasound probe is obtained; at the optimized frequency of the ultrasound probe, the updated ultrasound image sequence of the tissue area to be anesthetized and punctured within a preset time period is re-acquired.

[0011] The ultrasound image sequence and the updated ultrasound image sequence with the same numerical values ​​are fused to obtain the fused ultrasound image sequence; tissue identification is performed on the fused ultrasound image sequence, and the puncture path is calculated based on the tissue identification results to generate puncture-assisted guidance suggestions.

[0012] Furthermore, the specific steps for determining the tissue heterogeneity of each ultrasound image frame are as follows:

[0013] The tissue absorption degree of each ultrasound image is determined based on the gradient amplitude of all edge pixels within each frame of ultrasound image.

[0014] Connected component labeling is performed on all edge pixels in each frame of ultrasound image to obtain several connected components;

[0015] The tissue heterogeneity of each ultrasound image is determined based on the number of connected components in each frame, the gradient magnitude of all pixels, and the degree of tissue absorption in each frame.

[0016] Furthermore, the specific steps for determining the degree of tissue absorption in each frame of ultrasound image are as follows:

[0017] The inversely proportional normalized value of the mean gradient amplitude of all edge pixels in each frame of ultrasound image is recorded as the tissue absorption degree of each frame of ultrasound image.

[0018] Furthermore, the specific steps for determining the tissue heterogeneity of each ultrasound image frame based on the number of connected components within each frame, the gradient magnitude of all pixels, and the tissue absorption degree of each frame are as follows:

[0019] Within each frame of ultrasound image, the mean of the gradient magnitude of all pixels is obtained and denoted as the first mean. The ratio of the number of connected components to the first mean is obtained and denoted as the first ratio. The normalized value of the product of the first ratio and the tissue absorption degree of each frame of ultrasound image is denoted as the tissue heterogeneity of each frame of ultrasound image.

[0020] Furthermore, the specific steps for determining the imaging frequency penetration of each frame of ultrasound image are as follows:

[0021] Within each frame of ultrasound image, the number of pixels with gray values ​​less than a preset gray value threshold is counted and recorded as the first quantity value. The ratio of the first quantity value to the total number of pixels is obtained and recorded as the second ratio.

[0022] Based on the second ratio and the tissue heterogeneity of each ultrasound image, the imaging frequency penetration of each ultrasound image is obtained.

[0023] Furthermore, the specific steps for obtaining the imaging frequency penetration of each ultrasound image based on the second ratio and the tissue heterogeneity of each ultrasound image frame are as follows:

[0024] The inversely proportional normalized value of the product of the second ratio and the tissue heterogeneity of each ultrasound image frame is denoted as the imaging frequency penetration of each ultrasound image frame.

[0025] Furthermore, the specific steps for obtaining the optimized frequency of the ultrasound probe are as follows:

[0026] In the ultrasound image sequence, the average value of the imaging frequency penetration of all frames of ultrasound images is obtained and denoted as the second average value.

[0027] When the second mean value is less than the preset penetration threshold, the frequency adjustment coefficient is obtained based on the preset initial frequency of the ultrasound probe and the second mean value.

[0028] The optimal frequency of the ultrasonic probe is determined based on the frequency adjustment coefficient and the preset base frequency.

[0029] Furthermore, the specific steps for obtaining the frequency adjustment coefficient are as follows:

[0030] The product of the preset initial frequency of the ultrasound probe and the second mean value is obtained and denoted as the frequency adjustment coefficient.

[0031] Furthermore, the specific steps for determining the optimal frequency of the ultrasonic probe based on the frequency adjustment coefficient and the preset fundamental frequency are as follows:

[0032] The sum of the frequency adjustment coefficient and the preset base frequency is used as the optimized frequency of the ultrasonic probe.

[0033] The present invention also proposes an interactive anesthesia puncture auxiliary guidance system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned interactive anesthesia puncture auxiliary guidance method.

[0034] The beneficial effects of the technical solution of the present invention are:

[0035] In this embodiment of the invention, at a preset initial frequency of the ultrasound probe, an ultrasound image sequence of the tissue area to be anesthetized and punctured within a preset time period is acquired. The imaging frequency penetration of each frame of the ultrasound image is determined to obtain an optimized frequency for the ultrasound probe. At the optimized frequency of the ultrasound probe, an updated ultrasound image sequence of the tissue area to be anesthetized and punctured within a preset time period is acquired again. Images with the same sequence values ​​in the updated ultrasound image sequence are fused to obtain a fused ultrasound image sequence. Tissue identification is performed on the fused ultrasound image sequence, and the puncture path is calculated based on the tissue identification results to generate puncture assistance guidance suggestions. Thus, this invention, by analyzing the main imaging characteristics of real-time imaging results, obtains the heterogeneity of the tissue and the penetration corresponding to the ultrasound frequency, updates the ultrasound imaging frequency based on this penetration, and fuses the results to obtain new imaging results. This eliminates the problems of imaging blurring and identification errors caused by complex conditions within the tissue, improves the safety of the assistance guidance process, and, in the process of analyzing the penetration effect, by analyzing the basic characteristics of imaging and combining the general reflection imaging effect of the tissue area, determines the unsuitable frequency and its further suggested frequency update results. The updated frequency ultrasound imaging results are then subjected to image fusion processing, improving the image recognizability. Therefore, targeted fusion processing is performed on key areas (including fat and other areas), while other areas generate normal ultrasound images. This reduces the energy consumption of hardware and increases the efficiency and accuracy of patient tissue area identification. Finally, after obtaining the fused image results, the system's terminal neural network identifies the images and provides relevant puncture assistance guidance, improving puncture efficiency and safety during the assisted identification process, reducing patient waiting time, and providing a learnable educational effect. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart illustrating the steps of an interactive anesthesia puncture guidance method according to the present invention;

[0038] Figure 2 This is a schematic diagram of the fusion process between ultrasound images and updated ultrasound images. Detailed Implementation

[0039] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an interactive anesthesia puncture guidance method and system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0041] The following description, in conjunction with the accompanying drawings, details the specific scheme of the interactive anesthesia puncture auxiliary guidance method and system provided by the present invention.

[0042] Please see Figure 1 The diagram illustrates a flowchart of an interactive anesthesia puncture guidance method according to an embodiment of the present invention, the method comprising the following steps:

[0043] Step S001: At the preset initial frequency of the ultrasound probe, acquire an ultrasound image sequence of the area to be anesthetized and punctured within a preset time period.

[0044] In this embodiment, an anesthesia puncture-assisted image data acquisition device and its corresponding analysis system are first configured, and the acquired real-time image data is transmitted to the analysis system. Then, real-time ultrasound image frames are acquired, and the real-time ultrasound imaging frequency penetration is determined based on the heterogeneity of the image tissue. The ultrasound imaging frequency and its probe angle range are updated according to the penetration. The regenerated ultrasound images are fused to obtain high-definition ultrasound images. Finally, combined with the recognition network, the tissue area is identified, and puncture positioning guidance suggestions are provided to improve the safety and efficiency of puncture-assisted guidance in complex tissue areas.

[0045] It should be noted that: in configuring the anesthesia puncture auxiliary image data acquisition equipment and its corresponding analysis system, and in the process of transmitting the acquired real-time image data to the analysis system, an ultrasound probe is used to acquire images of the puncture area in real time. The frequency of the probe is variable, which is used to optimize the clarity of the generated ultrasound images. The high frequency range of the probe is 5 to 15 MHz, and the low frequency range is 1 to 5 MHz. In addition, a linear array probe is selected, which is suitable for imaging of superficial tissues, such as planar scanning of nerves, blood vessels, etc. The ultrasound machine is used to process the images and transmit the data to the analysis system through an interface (USB). The position of the puncture needle is tracked using a sensor and a puncture guide device to ensure accuracy. The real-time image data is transmitted to the analysis system using Wi-Fi, and the data transmission requires low latency (i.e., less than 50 milliseconds, which is described as an example) to ensure real-time performance. The analysis system includes: (1) Image processing: Real-time image analysis is performed through a deep learning model (CNN) to identify tissue structures. (2) Computing platform: Data processing is performed through an IPGPU workstation. (3) User interface: Displays real-time images and puncture path suggestions to assist doctors in performing precise punctures. Therefore, images of the puncture area are first acquired in real time, then transmitted to the analysis system for real-time analysis and suggestions on the puncture path. Finally, the doctor performs a precise puncture operation based on the suggestions.

[0046] Therefore, at the preset initial frequency of the ultrasound probe, an ultrasound image sequence of the area to be anesthetized and punctured within a preset time period can be obtained.

[0047] It should be noted that in this embodiment, the preset initial frequency of the ultrasound probe is 10MHz. Higher frequencies result in better spatial resolution but reduced penetration depth. The preset duration is 2 seconds, and the ultrasound image acquisition frequency is 24 frames per second, meaning there are 48 frames of ultrasound images in the sequence. The ultrasound images are grayscale images. The interactive puncture assistance guidance device, configured as described above, provides real-time generation of ultrasound images of the patient's anesthetized area and auxiliary guidance for the puncture area. In actual interactive operation, the complexity of the tissue area to be punctured, such as excessive fat or bleeding, can lead to unclear ultrasound imaging and numerous noise spots. Analyzing the principle of ultrasound imaging with the above device, it obtains the imaging result by comprehensively acquiring and analyzing the reflection effect of ultrasound waves on the tissue surface. Within the tissue, different fat thicknesses provide different levels of penetration for ultrasound waves. Areas with thicker fat layers or more bleeding are suitable for ultrasound waves with stronger penetration, i.e., low-frequency ultrasound waves. The resulting ultrasound images lack detailed characteristics and only show the basic structure of the tissue within the fat layer. Conversely, for areas with less fat or bleeding, high-frequency ultrasound, which has lower penetration but richer imaging details, is more suitable. The resulting images have higher resolution, allowing for detailed observation of superficial structures such as blood vessels and nerves. Therefore, complex tissue areas are often accompanied by blurry, noisy ultrasound images. To address this issue, the system must first be able to identify specific complex tissue areas, i.e., areas where the complexity or heterogeneity leads to blurry ultrasound images.

[0048] Step S002: Determine the tissue heterogeneity of each ultrasound image frame based on the gradient amplitude of pixels and edge pixels within each frame of the ultrasound image sequence.

[0049] It should be noted that: real-time ultrasound image frames are obtained, and the heterogeneity of the patient's tissue region reflected by the visual characteristics of the ultrasound image under the current frame is judged. The higher the heterogeneity, the thicker the fat layer or the higher the bleeding volume corresponding to the puncture area, which will lead to blurry ultrasound images or high noise problems. Specifically, when the penetration effect is poor, the image will present more regular features for the system to accurately identify, including: (1) blurry boundaries. Due to the weak penetration ability of low-frequency ultrasound, the image will lack clear boundaries, especially in areas with thick fat layers, and the contrast is insufficient. That is, when low-frequency imaging is used, superficial tissues (such as skin and fat) have low recognition, making it difficult to distinguish the boundaries between tissues. (2) The effect of increased noise spots is accompanied by the fact that when the penetration depth is large, the ultrasound echo signal is relatively weak, resulting in an increase in noise spots during imaging, especially in the transition area between the fat layer and the muscle layer, and when there is excessive bleeding, it also presents high reflected noise.

[0050] Preferably, in one embodiment of the present invention, the method for obtaining tissue heterogeneity of each frame of ultrasound image includes:

[0051] In ultrasound imaging sequences, with the first... Taking the first frame of ultrasound image as an example, the Canny edge detection algorithm is used to perform edge detection on the first frame. Edge detection is performed on the first frame of ultrasound image to obtain the second frame. The gradient magnitude of each pixel in a frame of ultrasound image, as well as several edge pixels.

[0052] Using the connected component labeling algorithm (Two-pass algorithm), according to the first... All edge pixels within a frame of ultrasound image are marked as connected components to obtain several connected components.

[0053] The Canny edge detection algorithm and the connected component labeling algorithm (Two-pass algorithm) are both well-known technologies, and their specific methods will not be described here.

[0054] It should be noted that when the penetration effect of ultrasound is poor, the edges of the ultrasound image are extremely blurred and the contrast is low due to the absorption of ultrasound by the fat layer. Secondly, there are many isolated connected regions within the image frame. This is because excessive bleeding and the inability to penetrate part of the fat layer lead to increased echo noise spots or speckle noise, which appear as isolated connected regions in the image. Therefore, when there are many connected regions, the corresponding tissue heterogeneity is also high.

[0055] In the Within a frame of ultrasound image, obtain the mean gradient magnitude of all edge pixels. The inverse proportional normalized value is denoted as the th The degree of tissue absorption in a frame of ultrasound images.

[0056] It should be noted that in this embodiment, the following is used: As The inverse proportional normalized value, where, This is a linear normalization function used to normalize data values ​​to a range between 0 and 1. The larger the gradient magnitude of edge pixels, the clearer and more distinct the inner edges of the image. A larger inverse normalized value indicates blurred edges within the image, meaning that the ultrasound waves were absorbed more by the thicker fat layer within the tissue during image generation, resulting in higher tissue absorption. Higher tissue absorption corresponds to higher tissue heterogeneity, necessitating the determination of real-time ultrasound penetration and subsequent ultrasound frequency update adjustments.

[0057] In the Within a frame of ultrasound image, the mean of the gradient magnitudes of all pixels is obtained and denoted as the first mean. The ratio of the number of connected components to the first mean is also obtained and denoted as the first ratio. The first ratio is then compared with the first frame of ultrasound image. The product of tissue absorption in each frame of ultrasound image The normalized value, denoted as the th Tissue heterogeneity in frame ultrasound images.

[0058] It should be noted that: with As The normalized value, representing the average gradient amplitude of all pixels, reflects the contrast of the ultrasound image. A smaller contrast value indicates a blurrier image and higher tissue heterogeneity. Correspondingly, a greater number of connected regions within the image, i.e., more noise patches or speckles, suggests that excessive bleeding and thick fat layers can lead to poor echo quality and noise speckle formation. The closer the normalized value is to 1, the better the... The lower the visibility of a frame of ultrasound image, the more necessary it is to determine the penetration of the current ultrasound frequency into the tissue and the frequency update results in order to optimize the identifiability of the ultrasound image.

[0059] Step S003: Determine the imaging frequency penetration of each ultrasound image based on the tissue heterogeneity of each frame and the grayscale value of the pixels in the ultrasound image; obtain the optimized frequency of the ultrasound probe based on the imaging frequency penetration of all frames of ultrasound images in the ultrasound image sequence and the preset initial frequency of the ultrasound probe; at the optimized frequency of the ultrasound probe, re-acquire the updated ultrasound image sequence of the tissue area to be anesthetized within a preset time period.

[0060] It should be noted that, further, the frequency penetration needs to be determined. Firstly, it can be determined that for areas with higher tissue heterogeneity, the frequency penetration of ultrasound imaging is lower. This is because ultrasound cannot penetrate thicker fat layers and noisy areas with significant bleeding, resulting in blurred and noisy imaging results. Additionally, analyzing the real-time ultrasound imaging process, when ultrasound waves have difficulty penetrating highly attenuating media (such as fat deposits or blood clots), a "shadow region" is formed in the image. This manifests as a low-echo or anechoic "dark shadow" region behind fat or blood clots, blocking the display of deeper structures. Correspondingly, in the imaging results, for ultrasound image frames with high tissue heterogeneity, when there are many dark pixel blocks within them, it is considered that there are more "dark shadow" regions and lower penetration.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining updated ultrasound image sequences includes:

[0062] The preset grayscale threshold is 30, and we will use this as an example for explanation.

[0063] In the Within a frame of ultrasound image, the number of pixels with grayscale values ​​less than a preset grayscale threshold is counted and recorded as the first quantity value. The ratio of the first quantity value to the total number of pixels is obtained and recorded as the second ratio. The second ratio is then compared with the first... The product of tissue heterogeneity in frame ultrasound images The inverse proportional normalized value is denoted as the th The imaging frequency penetration of ultrasound images.

[0064] It should be noted that in this embodiment, the following is used: As The inversely proportional normalized value, quantified by the second ratio, yields the "shadow" effect caused by lower penetration when facing fat deposits and blood clots. A higher second ratio indicates a more pronounced "shadow" effect in the image, indicating lower penetration. Greater tissue heterogeneity results in lower penetration. Therefore, the product... The larger the inverse proportional normalization value, the greater the imaging frequency penetration.

[0065] The preset penetration threshold is 0.2, and the preset base frequency is 2MHz. This will be used as an example for the description.

[0066] In the ultrasound image sequence, the average value of the imaging frequency penetration of all frames of ultrasound images is obtained and recorded as the second average value. When the second average value is less than the preset penetration threshold, the product of the preset initial frequency of the ultrasound probe and the second average value is obtained and recorded as the frequency adjustment coefficient. The sum of the frequency adjustment coefficient and the preset base frequency is used as the optimized frequency of the ultrasound probe.

[0067] At the optimized frequency of the ultrasound probe, the updated ultrasound image sequence of the tissue area to be anesthetized and punctured within a preset time period is reacquired. The acquisition frequency is 24 frames per second, and the size of the updated ultrasound image is consistent with that of the original ultrasound image.

[0068] It should be noted that when the imaging frequency penetration of all frames in an ultrasound imaging sequence is generally low, it can be determined that the tissue area is a complex fat deposit or blood clot. In this case, a penetration value update ultrasound operation is required, which optimizes the real-time ultrasound frequency to generate a more suitable and visually clear ultrasound image. Since the preset initial frequency of the ultrasound probe is 10MHz, the optimized frequency range of the ultrasound probe is 2 to 4MHz. The lower the penetration value, the lower the corresponding frequency optimization value, because low frequencies have strong tissue penetration capabilities and are suitable for imaging deep tissues (such as the abdomen, deep blood vessels, etc.). When regenerating the ultrasound image, it is also necessary to increase the imaging angle range by 20% of the original angle range. This increase in the probe angle range is mainly to improve the imaging coverage area, allowing observation of a larger area of ​​tissue structure, with the aim of making the imaging results clearer while maintaining the same image size. When the second mean is greater than or equal to the preset penetration threshold, it indicates that the quality of the ultrasound image in the ultrasound image sequence is appropriate, and subsequent tissue identification can be performed directly on the ultrasound image sequence. Based on the tissue identification results, the puncture path is calculated, and puncture-assisted guidance suggestions are generated.

[0069] Step S004: Fuse the ultrasound image sequence with the image under the same sequence value in the updated ultrasound image sequence to obtain the fused ultrasound image sequence; perform tissue identification on the fused ultrasound image sequence, calculate the puncture path based on the tissue identification results, and generate puncture-assisted guidance suggestions.

[0070] It should be noted that after obtaining new ultrasound imaging results for complex tissue areas, these results need to be fused with the original high-frequency imaging results. This is because lower frequencies increase the penetration depth of the image, but reduce spatial resolution. In other words, after frequency optimization, the generated image has higher penetration and is clearer, but lower resolution and fewer imaging details, offering advantages only in areas with thicker fat layers and deeper structures. On the other hand, the original high-frequency ultrasound images have higher resolution and can help to observe superficial structures such as blood vessels and nerves in detail. Therefore, fusion processing is necessary to obtain high-resolution local ultrasound images.

[0071] Preferably, in one embodiment of the present invention, the method for obtaining puncture-assisted guidance suggestions includes:

[0072] The alpha fusion algorithm is used to fuse ultrasound image sequences with ultrasound images of the same order in the updated ultrasound image sequence to obtain a fused ultrasound image sequence.

[0073] It should be noted that the Alpha fusion algorithm is a simple image fusion technique that achieves image overlay by adjusting the image transparency. The Alpha value is an important parameter of this algorithm, and this is a well-known technique. In this embodiment, the Alpha value is adjusted based on the gradient magnitude of the pixels. Pixels with higher gradient magnitudes play a more critical role in the tissue, such as revealing key information like blood vessels. This allows for the fusion and highlighting of high-frequency details, such as blood vessels.

[0074] Preset base Alpha value The value is 0.5, and we will use this as an example for explanation. In the ultrasound imaging sequence, the... The first frame of ultrasound image and the second frame of the updated ultrasound image sequence During the fusion process of frame-updated ultrasound images, the first frame is obtained. The first frame of ultrasound image Pixel Alpha value update ,in For the first The first frame of ultrasound image Gradient magnitude of a pixel The value range is from 0.5 to 1. Then the... In frame-fused ultrasound images grayscale value of a pixel ,in For the first The first frame of ultrasound image The grayscale value of a pixel. For the first Frame update ultrasound image The grayscale values ​​of each pixel are used to sequentially obtain each frame of the fused ultrasound image, forming a fused ultrasound image sequence. A schematic diagram of the fusion process between ultrasound images and updated ultrasound images is shown below. Figure 2 As stated above. Figure 2 In this process, multiple frames of low-penetration images (ultrasound image sequence) are first acquired for the patient using ultrasound frequency f (the preset initial frequency of the ultrasound probe), and then multiple frames of high-penetration images (updated ultrasound image sequence) are acquired using ultrasound frequency F (the optimized frequency of the ultrasound probe). The low-penetration images and high-penetration images are then fused to obtain the image fusion result. Based on the image fusion result, assisted guidance for localization is achieved.

[0075] Tissue identification is performed on the fused ultrasound image sequence, and the puncture path is calculated based on the tissue identification results to generate puncture-assisted guidance suggestions.

[0076] It should be noted that the final fusion result possesses real-time, efficient, and highly recognizable visual features, using a recognition network to identify tissue regions. In this embodiment, the U-Net semantic segmentation network is used to train the fused ultrasound images, outputting the classification results of each pixel in the fused ultrasound images, such as fat, muscle, blood vessels, and nerves, thereby obtaining the various tissue regions in the fused ultrasound images. The dataset used by the recognition network is a sequence of fused ultrasound images; the network's task is classification, so the cross-entropy loss function is used. Then, based on the recognition results, the Dijkstra algorithm is used to calculate the shortest and safest puncture path. Finally, the visualization output displays the tissue region, recommended path, puncture angle, and risk warnings, providing real-time feedback and dynamically adjusting the puncture strategy, thus generating puncture assistance guidance suggestions to improve puncture safety and operational efficiency, and has practical teaching significance. Through the above operations, it is possible to achieve auxiliary guidance during the anesthesia puncture process through real-time interactive methods based on the ultrasound image recognition results. The U-Net semantic segmentation network and the Dijkstra algorithm are both well-known technologies, and their specific methods will not be described here.

[0077] The present invention also provides an interactive anesthesia puncture assistance guidance system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned interactive anesthesia puncture assistance guidance method.

[0078] This invention is now complete.

[0079] In summary, in this embodiment of the invention, at a preset initial frequency of the ultrasound probe, an ultrasound image sequence of the tissue area to be anesthetized and punctured within a preset time period is acquired. The imaging frequency penetration of each frame of the ultrasound image is determined to obtain an optimized frequency for the ultrasound probe. At the optimized frequency of the ultrasound probe, an updated ultrasound image sequence of the tissue area to be anesthetized and punctured within the preset time period is acquired again. Images with the same sequence values ​​in the updated ultrasound image sequence are fused to obtain a fused ultrasound image sequence. Tissue identification is performed on the fused ultrasound image sequence, and the puncture path is calculated based on the tissue identification results to generate puncture-assisted guidance suggestions. This invention, by updating the ultrasound imaging frequency and fusing to obtain new imaging results, eliminates the problems of imaging blurring and identification errors caused by complex conditions within the tissue, thereby improving the safety of the anesthesia puncture-assisted guidance process.

[0080] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An interactive method for guiding anesthesia puncture, characterized in that, The method includes the following steps: At the preset initial frequency of the ultrasound probe, acquire an ultrasound image sequence of the area to be anesthetized and punctured within a preset duration. The tissue heterogeneity of each ultrasound image is determined based on the gradient amplitude of pixels within each frame of the ultrasound image sequence and edge pixels. The specific steps involved in determining the tissue heterogeneity of each ultrasound image frame are as follows: The tissue absorption degree of each ultrasound image is determined based on the gradient amplitude of all edge pixels in each frame of ultrasound image; wherein the specific steps for determining the tissue absorption degree of each ultrasound image are as follows: the inversely proportional normalized value of the mean gradient amplitude of all edge pixels in each frame of ultrasound image is recorded as the tissue absorption degree of each frame of ultrasound image. Connected component labeling is performed on all edge pixels in each frame of ultrasound image to obtain several connected components; The tissue heterogeneity of each ultrasound image is determined based on the number of connected components in each frame, the gradient magnitude of all pixels, and the tissue absorption degree of each frame. The specific steps for determining the tissue heterogeneity of each ultrasound image based on the number of connected components in each frame, the gradient magnitude of all pixels, and the tissue absorption degree of each frame are as follows: In each ultrasound image, the mean of the gradient magnitude of all pixels is obtained and recorded as the first mean; the ratio of the number of connected components to the first mean is obtained and recorded as the first ratio; the normalized value of the product of the first ratio and the tissue absorption degree of each frame is recorded as the tissue heterogeneity of each frame. Based on the tissue heterogeneity of each ultrasound image and the grayscale value of the pixels in the ultrasound image, the imaging frequency penetration of each ultrasound image is determined; based on the imaging frequency penetration of all ultrasound images in the ultrasound image sequence, combined with the preset initial frequency of the ultrasound probe, the optimized frequency of the ultrasound probe is obtained; at the optimized frequency of the ultrasound probe, the updated ultrasound image sequence of the tissue area to be anesthetized and punctured within a preset time period is re-acquired. The ultrasound image sequence and the updated ultrasound image sequence with the same numerical values ​​are fused to obtain the fused ultrasound image sequence; tissue identification is performed on the fused ultrasound image sequence, and the puncture path is calculated based on the tissue identification results to generate puncture-assisted guidance suggestions.

2. The interactive anesthesia puncture guidance method according to claim 1, characterized in that, The specific steps for determining the imaging frequency penetration of each ultrasound image frame are as follows: Within each frame of ultrasound image, the number of pixels with gray values ​​less than a preset gray value threshold is counted and recorded as the first quantity value. The ratio of the first quantity value to the total number of pixels is obtained and recorded as the second ratio. Based on the second ratio and the tissue heterogeneity of each ultrasound image, the imaging frequency penetration of each ultrasound image is obtained.

3. The interactive anesthesia puncture guidance method according to claim 2, characterized in that, The specific steps for obtaining the imaging frequency penetration of each ultrasound image based on the second ratio and the tissue heterogeneity of each ultrasound image frame are as follows: The inversely proportional normalized value of the product of the second ratio and the tissue heterogeneity of each ultrasound image frame is denoted as the imaging frequency penetration of each ultrasound image frame.

4. The interactive anesthesia puncture guidance method according to claim 1, characterized in that, The specific steps for obtaining the optimized frequency of the ultrasound probe are as follows: In the ultrasound image sequence, the mean value of the imaging frequency penetration of all frames of ultrasound images is obtained and denoted as the second mean value. When the second mean value is less than the preset penetration threshold, the frequency adjustment coefficient is obtained based on the preset initial frequency of the ultrasound probe and the second mean value. The optimal frequency of the ultrasonic probe is determined based on the frequency adjustment coefficient and the preset base frequency.

5. The interactive anesthesia puncture guidance method according to claim 4, characterized in that, The specific steps for obtaining the frequency adjustment coefficient are as follows: The product of the preset initial frequency of the ultrasound probe and the second mean value is obtained and denoted as the frequency adjustment coefficient.

6. The interactive anesthesia puncture guidance method according to claim 4, characterized in that, The specific steps for determining the optimal frequency of the ultrasonic probe based on the frequency adjustment coefficient and the preset fundamental frequency are as follows: The sum of the frequency adjustment coefficient and the preset base frequency is used as the optimized frequency of the ultrasonic probe.

7. An interactive anesthesia puncture guidance system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of an interactive anesthesia puncture-assisted guidance method as described in any one of claims 1-6.

Images