Spine endoscopic surgery path planning method based on ultrasonic guidance

Abstract

The invention relates to the technical field of construction of a spinal internal three-dimensional model, in particular to a spinal endoscopic surgery path planning method based on ultrasonic guidance. The method comprises the following steps: synchronously acquiring ultrasonic images containing an endoscope probe at various angles in real time; according to the positions and shapes of the human tissue edge lines in the adjacent ultrasonic images obtained synchronously, the shape consistency degree of the human tissue edge lines and matched edge lines are obtained; acquiring matched pixel points of the edge pixel points according to the position relation between the human tissue edge line in each ultrasonic image and the edge pixel points on the matched edge line in the position relative ultrasonic image; according to the positions of the edge pixel points and the matched pixel points, the real distance between the edge pixel points and the endoscope probe is obtained, and then a three-dimensional model in the spine endoscopic surgery process is constructed in real time. According to the method, the real distance between the edge pixel point on the edge line of the human tissue and the endoscope probe is accurately obtained in real time, so that a three-dimensional model in the spine endoscopic surgery process can be accurately constructed in real time.

Description

Technical Field

[0001] This invention relates to the field of constructing a three-dimensional model of the interior of the spine, and specifically to a method for planning the surgical path of spinal endoscopic surgery based on ultrasound guidance. Background Technology

[0002] With the prevalence of sedentary lifestyles, the incidence of lumbar disc herniation is rising year by year. Due to its advantages of minimal invasiveness and rapid recovery, spinal endoscopic surgery has gradually replaced traditional open surgery as the mainstream treatment option. Spinal endoscopic surgery requires inserting an endoscope probe percutaneously into the spine to reach the lesion area (such as the herniated disc) while avoiding nerves and blood vessels.

[0003] Current methods utilize ultrasound images of the spine to guide and control the endoscopic probe, allowing it to avoid important blood vessels and tissues. This enables the probe to approach the lesion, observe the affected tissue, and directly remove or excise it for treatment. However, in practice, this approach fails to consider that ultrasound images are two-dimensional cross-sectional images. Given the three-dimensional structure of the human vertebrae, which have numerous grooves, ultrasound images at certain angles cannot accurately represent the tortuous information of these grooves. This can easily lead to misjudgment of the edges of important tissues, making it difficult to accurately plan the path of the endoscopic probe and potentially causing damage to human tissue. Summary of the Invention

[0004] To address the technical problem of inaccurate endoscopic probe path planning during spinal endoscopic surgery, which can easily cause damage to human tissues, the present invention aims to provide an ultrasound-guided spinal endoscopic surgical path planning method. The specific technical solution adopted is as follows: This invention provides a method for planning the surgical path in spinal endoscopy based on ultrasound guidance, the method comprising the following steps: Real-time synchronous acquisition of ultrasound images from various angles of the endoscope probe during spinal endoscopic surgery; Based on the positional distribution and shape similarity of the human tissue edge lines in each synchronously acquired ultrasound image and its adjacent ultrasound images, the shape consistency and matching edge lines of each human tissue edge line are obtained. Based on the positional relationship between each edge pixel and the edge pixel on the matching edge in each ultrasound image, the matching pixel of each edge pixel is obtained. Based on the shape consistency of the edge of the human tissue where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, the positional relationship between each edge pixel and its matching pixel in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, and the positional relationship between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, the true distance between each edge pixel and the endoscope probe is obtained. Based on the actual distance and the angle corresponding to the ultrasound image of each edge pixel, a three-dimensional model of the spinal endoscopic surgery process is constructed in real time, thereby planning the spinal endoscopic surgery path.

[0005] Furthermore, the method for obtaining the degree of shape consistency is as follows: For any given moment, the ultrasound images acquired synchronously at that moment are all taken as target images, and two target images that are adjacent in position are taken as a set of matching images; For any set of matching images, the two target images in the set of matching images are overlapped, and a preset number of human tissue edge lines in another target image that are close to the position of each human tissue edge line are used as reference edge lines for each human tissue edge line. The two target images in the matching image set are overlaid and mapped onto a two-dimensional coordinate system to obtain the pixel coordinate sequence corresponding to each human tissue edge line; For any human tissue edge line and any of its reference edge lines, the DTW value of the pixel coordinate sequence of the human tissue edge line and the reference edge line is obtained by the dynamic time rule algorithm; the result of negatively correlated and normalized DTW value is used as the matching degree between the human tissue edge line and the reference edge line. The degree of matching between the edge line of the human tissue and each of its reference edge lines is obtained, and the highest degree of matching is taken as the degree of shape consistency of the edge line of the human tissue.

[0006] Furthermore, the method for obtaining the matching edge line is as follows: For any human tissue edge line, the reference edge line corresponding to the degree of shape consistency of the human tissue edge line is used as the matching edge line of the human tissue edge line. Among them, the matching edge line of the matching edge line of the human tissue edge line is also used as the matching edge line of the human tissue edge line; the human tissue edge line corresponds to a matching edge line and a degree of shape consistency in other ultrasound images.

[0007] Furthermore, the method for obtaining the matching pixels is as follows: For any edge pixel on the edge line of any human tissue in any ultrasound image, the position of that ultrasound image relative to the ultrasound image is used as the analysis image; The ultrasound image and the analysis image are overlaid and mapped onto a two-dimensional coordinate system. The Euclidean distance between the edge pixel and the matching edge line of the human tissue edge in the analysis image is obtained, and each edge pixel is used as the matching distance of the edge pixel. The edge pixel on the matching edge line of the human tissue edge line in the analysis image corresponding to the smallest matching distance is taken as the matching pixel of the edge pixel.

[0008] Furthermore, the method for obtaining the true distance is as follows: Based on the consistency of the shape of the human tissue edge line where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, and the positional relationship between each edge pixel and its matching pixel in the ultrasound image of the endoscope probe, the positional deviation between each edge pixel and its matching pixel is obtained. Based on the distance between each edge pixel and the endoscopic probe in its ultrasound image and the corresponding ultrasound probe in the ultrasound image, and the distance between each edge pixel and the matching pixel and the endoscopic probe in its ultrasound image and the corresponding ultrasound probe in the ultrasound image, the reference weights of each edge pixel and its matching pixel are obtained respectively. Based on the reference weights of each edge pixel and its matching pixel, and the positional relationship between each edge pixel and its matching pixel and the endoscope probe in the ultrasound image, the initial corrected distance between each edge pixel and the endoscope probe is obtained. The result of negatively correlating and normalizing the positional deviation between each edge pixel and its matching pixel is used as the second correction weight for each edge pixel. The product of the second correction weight and the initial correction distance is used as the true distance between each edge pixel and the endoscope probe.

[0009] Furthermore, the method for obtaining the degree of positional deviation is as follows: For any edge pixel, the matching distance between the edge pixel and its matching pixel is taken as the first distance; The Euclidean distance between the edge pixel and the corresponding coordinate point of the endoscope probe in the ultrasound image where the matching pixel is located is obtained, and this distance is used as the second distance. The sum of the first distance and the second distance is used as the distance deviation between the edge pixel and its matching pixel. The average value of the shape consistency of the human tissue edge line where the edge pixel is located is used as the reference consistency of the human tissue edge line where the edge pixel is located. The ratio of the distance deviation value to the reference consistency is used as the degree of positional deviation between the edge pixel and its matching pixel.

[0010] Furthermore, the method for obtaining the reference weights of each edge pixel and its matching pixel is as follows: For any edge pixel, the ultrasound image containing that edge pixel is used as the first designated image; The distance between the edge pixel and the ultrasound probe corresponding to the first specified image is taken as the first specified distance; The distance between the endoscope probe in the first specified image and the corresponding ultrasound probe in the first specified image is used as the second specified distance; The ultrasound image containing the matching pixel of the edge pixel is used as the second specified image; The distance between the matching pixel of the edge pixel and the ultrasound probe corresponding to the second specified image is used as the third specified distance; The distance between the endoscope probe in the second specified image and the corresponding ultrasound probe in the second specified image is taken as the fourth specified distance; The sum of the first specified distance and the second specified distance is used as the distance analysis value of the edge pixel. The sum of the third and fourth specified distances is used as the distance analysis value of the matching pixel for that edge pixel. The sum of the distance analysis values ​​of the edge pixel and its matching pixel is used as the overall distance analysis value; The result of negatively correlating the ratio of the distance analysis value of the edge pixel to the overall distance analysis value is used as the reference weight of the edge pixel. The result of negatively correlating the ratio of the distance analysis value of the matching pixel of the edge pixel to the overall distance analysis value is used as the reference weight of the matching pixel of the edge pixel.

[0011] Furthermore, the method for obtaining the initial correction distance is as follows: For any edge pixel, the Euclidean distance between the edge pixel and the coordinate point corresponding to the endoscope probe in the ultrasound image is used as the first reference distance between the edge pixel and the endoscope probe. The Euclidean distance between the matching pixel of the edge pixel and the coordinate point corresponding to the endoscope probe in the ultrasound image is used as the second reference distance between the edge pixel and the endoscope probe. The product of the reference weight of the edge pixel and the first reference distance is used as the first corrected distance; The product of the reference weight of the matching pixel of the edge pixel and the second reference distance is used as the second correction distance; The sum of the first correction distance and the second correction distance is used as the initial correction distance between the edge pixel and the endoscope probe.

[0012] Furthermore, the method for obtaining the endoscopic probe in the ultrasound image is as follows: The endoscopic probe in each ultrasound image is obtained using a threshold segmentation algorithm.

[0013] Furthermore, the method for obtaining the edge line of the human tissue is as follows: The edge lines of human tissue in each ultrasound image are obtained using the Canny edge detection algorithm.

[0014] The present invention has the following beneficial effects: This invention first obtains the shape consistency and matching edge lines of each human tissue edge line based on the positional distribution and shape similarity of the human tissue edge lines in each synchronously acquired ultrasound image and its adjacent ultrasound images. This accurately reflects the reliability of each human tissue edge line and its distribution in each synchronously acquired ultrasound image, which is beneficial for the subsequent real-time and accurate construction of a three-dimensional model during spinal endoscopic surgery. Then, based on the positional relationship between the edge pixels of each human tissue edge line in each ultrasound image and the matching edge lines in the corresponding ultrasound images, the matching pixels of each edge pixel are obtained, accurately determining the location of the same human tissue in two synchronously acquired ultrasound images, preparing for the subsequent accurate determination of the three-dimensional model of the spine's interior. Furthermore, based on the shape consistency of the human tissue edge line where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, and the distribution of each edge pixel in the corresponding ultrasound images, this invention provides a basis for the subsequent accurate determination of the three-dimensional model of the spine's interior. The system obtains the true distance between each edge pixel and the endoscope probe in the ultrasound image, including the positional relationship between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, and the positional relationship between each edge pixel and its matching pixel and the endoscope probe in the ultrasound image. This accurately reflects the distance between the edge pixels on the edge line of the human tissue and the endoscope probe at each moment, which is beneficial for more accurately constructing a three-dimensional model during spinal endoscopic surgery. Furthermore, based on the true distance and the angle corresponding to each edge pixel in the ultrasound image, a three-dimensional model of the spinal endoscopic surgery process can be constructed in real time and accurately, thereby accurately and rationally planning the spinal endoscopic surgery path and effectively avoiding damage to human tissue caused by the endoscope probe. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages 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.

[0016] Figure 1 This is a schematic flowchart illustrating an ultrasound-guided spinal endoscopic surgical path planning method according to an embodiment of the present invention. Figure 2 A flowchart illustrating a method for obtaining real distance according to an embodiment of the present invention; Figure 3This is a structural diagram of an ultrasound-guided spinal endoscopic surgical path planning system provided in one embodiment of the present invention. Figure 4 This is a schematic diagram of a computer device provided according to an embodiment of the present invention. Detailed Implementation

[0017] 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 the ultrasound-guided spinal endoscopic surgical path planning method 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.

[0018] 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.

[0019] The specific scheme of the ultrasound-guided spinal endoscopic surgical path planning method provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Example 1: This invention proposes an ultrasound-guided method for planning the surgical path in spinal endoscopic surgery. Please refer to [link to relevant documentation]. Figure 1 The diagram illustrates a schematic flowchart of an ultrasound-guided spinal endoscopic surgical path planning method according to an embodiment of the present invention. The method includes the following steps: Step S1: Real-time synchronous acquisition of ultrasound images from various angles of the endoscopic probe during spinal endoscopic surgery.

[0021] Specifically, during spinal endoscopic surgery, doctors can view the internal condition of the patient's spine in real time using a high-definition camera mounted on the endoscope probe. However, when there are human tissue walls obstructing the camera's view, the camera's field of view is limited, and it is impossible to determine whether other human tissue walls obstruct the area around the endoscope probe based solely on the camera's view. Therefore, an ultrasound probe is needed to scan the area around the endoscope probe to plan its path. It is known that the path of the endoscope probe can be planned using ultrasound images during spinal endoscopic surgery, i.e., a three-dimensional model of the spine's interior can be constructed using ultrasound images. However, the three-dimensional structure of the spine's interior is prone to information loss in two-dimensional ultrasound images. To more accurately analyze the corresponding three-dimensional model of the spine's interior during spinal endoscopic surgery, and to more accurately plan the path of the endoscope probe, avoiding damage to the human tissues inside the spine, this embodiment acquires ultrasound images of the endoscope probe from various angles during spinal endoscopic surgery in real time.

[0022] The specific method for synchronously acquiring ultrasound images from various angles is as follows: Centered on the patient, an ultrasound probe is installed at 30° intervals clockwise around the patient, ensuring that the probes uniformly surround the patient 360°. The operator can set the interval angle and position of the ultrasound probes according to the actual situation; no limitation is imposed here. Simultaneously, the ultrasound probes are all installed at the same height. The ultrasound images acquired by all ultrasound probes at the same moment, including the endoscopic probe, are the synchronously acquired ultrasound images. In this embodiment, an ultrasound image is synchronously acquired every 1 second during spinal endoscopic surgery. The operator can set the time interval between two adjacent synchronously acquired ultrasound images according to the actual situation; no limitation is imposed here.

[0023] Step S2: Based on the positional distribution and shape similarity of the human tissue edge lines in each synchronously acquired ultrasound image and its adjacent ultrasound images, obtain the shape consistency and matching edge lines of each human tissue edge line.

[0024] Specifically, the more similar the corresponding spinal internal regions in the ultrasound images acquired by adjacent ultrasound probes, the more accurate the information about the spine's internal structure contained in the ultrasound images becomes. This is crucial for accurately constructing a three-dimensional model of the spine's interior and accurately identifying the human tissues within it. Therefore, this embodiment first uses the Canny edge detection algorithm to obtain the edge lines of human tissues in each ultrasound image. The more similar the position and shape of the edge lines of human tissues in two synchronously acquired adjacent ultrasound images, the more accurate the information in the corresponding ultrasound image, which is beneficial for accurately constructing a three-dimensional model of the spine's interior. The Canny edge detection algorithm is a well-known technique and will not be described in detail here.

[0025] To accurately construct a local 3D model of the spine where the endoscopic probe is located in real time, this embodiment obtains the shape consistency and matching edge lines of each human tissue edge line based on the positional distribution and shape similarity of the human tissue edge lines in each synchronously acquired ultrasound image and its adjacent ultrasound images. A greater degree of shape consistency indicates more accurate overall information carried by the corresponding human tissue edge line. Simultaneously obtaining the matching edge lines of each human tissue edge line facilitates more detailed analysis of the human tissue's location in subsequent steps.

[0026] Preferably, in one feasible embodiment of this invention, the method for obtaining the degree of shape consistency is as follows: For any given moment, all ultrasound images acquired synchronously at that moment are taken as target images, and two adjacent target images are taken as a set of matching images. For any set of matching images, two target images in the set of matching images are overlapped, and a predetermined number of human tissue edge lines in another target image that are close to each human tissue edge line are taken as reference edge lines for each human tissue edge line; in this embodiment, the predetermined number is set to 5, and the implementer can set the size of the predetermined number according to the actual situation, which is not limited here. It should be noted that in this embodiment, when obtaining the reference edge line of each human tissue edge line, the contour of each human tissue edge line is first obtained, and then the contour points corresponding to each human tissue edge line are determined. The reference edge line of each human tissue edge line is determined by the 5 shortest Euclidean distances between the corresponding contour points of each human tissue edge line and each human tissue edge line in another target image after overlap. The methods for obtaining the contour of the edge line and the contour points, as well as the methods for obtaining the Euclidean distances, are all known technologies and will not be described in detail here. By obtaining the reference edge line of each human tissue edge line, it is beneficial to accurately and efficiently determine the corresponding human tissue edge line in adjacent ultrasound images. Further, the two target images in the matching image set are overlaid and mapped onto a two-dimensional coordinate system to obtain the pixel coordinate sequence corresponding to each human tissue edge line. It should be noted that the projection position of the target image in the two-dimensional coordinate system can be set according to the actual situation and is not limited here. For any human tissue edge line and any of its reference edge lines, the DTW (Dynamic Time Warping) value of the pixel coordinate sequence of the human tissue edge line and the reference edge line is obtained through a dynamic time rule algorithm. The smaller the DTW value, the more similar the position and shape of the human tissue edge line and the reference edge line are. The result of negatively correlated and normalized DTW values ​​is then used as the matching degree between the human tissue edge line and the reference edge line. In this embodiment, the negative number of the above DTW value is used as the power of an exponential function with the natural constant as the base. The output of the exponential function is the result of negatively correlated and normalized DTW values. In addition, the dynamic time rule algorithm is a well-known technology and will not be described in detail here. The greater the degree of matching, the more accurate the information represented by the human tissue edge line in the ultrasound image. In order to accurately determine the accuracy of the human tissue edge line, this embodiment obtains the degree of matching between the human tissue edge line and each of its reference edge lines, and takes the maximum degree of matching as the shape consistency of the human tissue edge line.

[0027] Preferably, in one feasible embodiment of this invention, the method for obtaining the matching edge line is as follows: for any human tissue edge line, the reference edge line corresponding to the shape consistency of the human tissue edge line is used as the matching edge line of the human tissue edge line. Since there must be a common ultrasound image in two adjacent sets of matching images, the matching edge line of the matching edge line of the human tissue edge line is essentially also the matching edge line of the human tissue edge line; therefore, the human tissue edge line corresponds to a matching edge line and a shape consistency in other ultrasound images.

[0028] At this point, the shape consistency and matching edge lines of each human tissue edge line are obtained.

[0029] Step S3: Based on the positional relationship between each human tissue edge line in each ultrasound image and the edge pixels on the matching edge line in the ultrasound image, obtain the matching pixel of each edge pixel; based on the shape consistency of the human tissue edge line where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, the positional relationship between each edge pixel and its matching pixel in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, and the positional relationship between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, obtain the true distance between each edge pixel and the endoscope probe.

[0030] Specifically, in practice, ultrasound images acquired by ultrasound probes at relatively opposite positions essentially correspond to the same area inside the spine; for example, ultrasound images acquired by probes at 0° and 180° angles represent the same region inside the spine. However, when using ultrasound probes to probe the internal spinal region around an endoscope probe, it cannot be guaranteed that the ultrasound images acquired by each probe represent the same local area of ​​the endoscope probe. Therefore, the distance between the human tissue wall and the endoscope probe in the obtained ultrasound images is not accurate. To accurately determine the distance between the human tissue wall and the endoscope probe, and to accurately construct a three-dimensional model of the internal spine around the endoscope probe, thereby more accurately planning the path of the endoscope probe, this embodiment obtains the matching pixel of each edge pixel based on the positional relationship between the edge pixels of each human tissue edge line in each ultrasound image and the matching edge lines in the relatively opposite ultrasound image; based on the positional relationship between each edge pixel and its matching pixel, the distance between each edge pixel and the endoscope probe is further corrected.

[0031] Preferably, in one feasible embodiment of this invention, the method for obtaining the matching pixel is as follows: For any edge pixel on any edge line of any human tissue in any ultrasound image, the position of the ultrasound image relative to the ultrasound image is used as the analysis image; the ultrasound image and the analysis image are overlaid and mapped onto a two-dimensional coordinate system, and the Euclidean distance between the edge pixel and the corresponding coordinate point of each edge pixel on the matching edge line of the human tissue edge in the analysis image is obtained, all of which are used as the matching distance of the edge pixel; the edge pixel on the matching edge line of the human tissue edge in the analysis image corresponding to the smallest matching distance is taken as the matching pixel of the edge pixel. It should be noted that each edge pixel and its matching pixel are essentially the same edge pixel detected by the ultrasound probe.

[0032] The greater the consistency between the shape of the edge pixel and the edge line of the human tissue, the more accurate the position of the edge pixel. The smaller the Euclidean distance between the edge pixel and its matching pixel in the two-dimensional coordinate system, the more accurate the position of the edge pixel. The more consistent the positions of the edge pixel and its matching pixel in the ultrasound image with the endoscopic probe in the two-dimensional coordinate system, the more accurate the position of the edge pixel. Meanwhile, considering that in practice, the smaller the distance between the edge pixel and the ultrasound probe, the more accurate the position of the edge pixel, because the ultrasound probe is less affected by noise when emitting and receiving sound waves, and the more accurate the data obtained; therefore, this embodiment obtains the true distance between each edge pixel and the endoscope probe based on the consistency of the shape of the human tissue edge line where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, the positional relationship between each edge pixel and its matching pixel in the ultrasound image of the endoscope probe, the distance between each edge pixel and its matching pixel in the ultrasound image of the corresponding ultrasound probe, the distance between each edge pixel and its matching pixel in the ultrasound image of the corresponding ultrasound probe, and the positional relationship between each edge pixel and its matching pixel in the ultrasound image of the corresponding ultrasound probe, and the positional relationship between each edge pixel and its matching pixel in the ultrasound image of the corresponding ultrasound probe, which is beneficial for accurately constructing the three-dimensional model of the spinal endoscopic surgery process.

[0033] Preferably, in one feasible implementation of this embodiment, the method for obtaining the actual distance is described in [reference needed]. Figure 2 The document presents a flowchart of a method for obtaining real distance provided in this embodiment, which includes the following steps: Step S201: Based on the consistency of the shape of the human tissue edge line where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, and the positional relationship between each edge pixel and its matching pixel in the ultrasound image of the endoscope probe, obtain the positional deviation between each edge pixel and its matching pixel.

[0034] The greater the positional deviation, the less accurate the relative position of the corresponding edge pixels and the endoscope probe.

[0035] In one possible implementation of this embodiment, the method for obtaining the degree of positional deviation is as follows: For any edge pixel, the matching distance between the edge pixel and its matching pixel is taken as the first distance; the larger the first distance, the more inconsistent the positions of the edge pixel and its matching pixel are, indirectly indicating that the position of the edge pixel is less accurate; the Euclidean distance between the edge pixel and the corresponding coordinate point of the endoscope probe in the ultrasound image where the matching pixel is located is obtained, as the second distance; the larger the second distance, the less accurate the position of the edge pixel and its matching pixel relative to the endoscope probe is. It should be noted that, because the endoscope probe is significantly different from human tissue, this embodiment uses a threshold segmentation algorithm to obtain the endoscope probe in each ultrasound image, and uses the centroid of the endoscope probe as the position point of the endoscope probe. The threshold segmentation algorithm and the method for obtaining the centroid are well-known technologies and will not be described in detail here. The larger the first distance and the larger the second distance, and the smaller the shape consistency of the human tissue edge line corresponding to the edge pixel, the more inconsistent and inaccurate the position of the edge pixel and its matching pixel, indirectly indicating that the distance between the edge pixel and the endoscope probe is less accurate. Therefore, the sum of the first distance and the second distance is taken as the distance deviation value between the edge pixel and its matching pixel. The average value of the shape consistency of the human tissue edge line corresponding to the edge pixel is taken as the reference consistency of the human tissue edge line corresponding to the edge pixel. Then, the ratio of the distance deviation value to the reference consistency is taken as the positional deviation degree between the edge pixel and its matching pixel.

[0036] At this point, the positional deviation between each edge pixel and its matching pixel is obtained.

[0037] Step S202: Based on the distance between each edge pixel and the endoscopic probe in the ultrasound image and the corresponding ultrasound probe in the ultrasound image, and the distance between each edge pixel and the matching pixel and the endoscopic probe in the ultrasound image and the corresponding ultrasound probe in the ultrasound image, obtain the reference weights of each edge pixel and its matching pixel.

[0038] The higher the reference weight, the more accurate the position of the corresponding edge pixel.

[0039] In one possible implementation of this embodiment, the method for obtaining the reference weight of each edge pixel and its matching pixel is as follows: For any edge pixel, the ultrasound image in which the edge pixel is located is taken as the first specified image; the distance between the edge pixel and the ultrasound probe corresponding to the first specified image is taken as the first specified distance; the smaller the first specified distance, the more accurate the position of the edge pixel; the distance between the endoscope probe in the first specified image and the ultrasound probe corresponding to the first specified image is taken as the second specified distance; the smaller the second specified distance, the more accurate the position of the endoscope probe in the first specified image; the ultrasound image in which the matching pixel of the edge pixel is located is taken as the second specified image; the distance between the matching pixel of the edge pixel and the ultrasound probe corresponding to the second specified image is taken as the third specified distance; the smaller the third specified distance, the more accurate the position of the matching pixel of the edge pixel; the distance between the endoscope probe in the second specified image and the ultrasound probe corresponding to the second specified image is taken as the fourth specified distance; the smaller the fourth specified distance, the more accurate the position of the endoscope probe in the second specified image. To accurately determine the distance between the edge pixel and the endoscope probe, the sum of the first and second specified distances is used as the distance analysis value of the edge pixel; the sum of the third and fourth specified distances is used as the distance analysis value of the matching pixel of the edge pixel; the sum of the distance analysis values ​​of the edge pixel and its matching pixel is used as the overall distance analysis value; then, the ratio of the distance analysis value of the edge pixel to the overall distance analysis value is negatively correlated, and the result is used as the reference weight of the edge pixel; the ratio of the distance analysis value of the matching pixel of the edge pixel to the overall distance analysis value is also negatively correlated, and the result is used as the reference weight of the matching pixel of the edge pixel. In this embodiment, 1 minus the above ratios is used as the result of the negative correlation of the above ratios.

[0040] At this point, the reference weights of each edge pixel and its matching pixel are obtained.

[0041] Step S203: Based on the reference weights of each edge pixel and its matching pixel, and the positional relationship between each edge pixel and its matching pixel and the endoscope probe in the ultrasound image, obtain the initial correction distance between each edge pixel and the endoscope probe.

[0042] It is known that the larger the reference weight, the more accurate the position of the corresponding edge pixel. Therefore, in this embodiment, the initial correction distance between each edge pixel and the endoscope probe is obtained based on the reference weight of each edge pixel and its matching pixel, as well as the positional relationship between each edge pixel and its matching pixel and the endoscope probe in the ultrasound image.

[0043] In one possible implementation of this embodiment, the method for obtaining the initial correction distance is as follows: For any edge pixel, the Euclidean distance between the edge pixel and the coordinate point corresponding to the endoscope probe in the ultrasound image is used as the first reference distance between the edge pixel and the endoscope probe; the Euclidean distance between the matching pixel of the edge pixel and the coordinate point corresponding to the endoscope probe in the ultrasound image is used as the second reference distance between the edge pixel and the endoscope probe; the product of the reference weight of the edge pixel and the first reference distance is used as the first correction distance; the product of the reference weight of the matching pixel of the edge pixel and the second reference distance is used as the second correction distance; finally, the sum of the first correction distance and the second correction distance is used as the initial correction distance between the edge pixel and the endoscope probe.

[0044] At this point, the initial corrected distance between each edge pixel and the endoscope probe is obtained.

[0045] Step S204: The result of negatively correlating and normalizing the positional deviation between each edge pixel and its matching pixel is used as the second correction weight for each edge pixel; the product of the second correction weight and the initial correction distance is used as the true distance between each edge pixel and the endoscope probe.

[0046] As can be seen from step S201 above, the greater the positional deviation, the less accurate the relative position of the corresponding edge pixel and the endoscope probe, that is, the less accurate the distance between the corresponding edge pixel and the endoscope probe. To accurately determine the true distance between each edge pixel and the endoscope probe, the positional deviation of each edge pixel and its matching pixel is negatively correlated and normalized, and used as the second correction weight for each edge pixel. In this embodiment, the negative of the aforementioned positional deviation is used as the power of an exponential function with a natural constant as its base. The output of this specified function is the result of negatively correlated and normalized positional deviation. The larger the second correction weight, the more accurate the initial corrected distance between the corresponding edge pixel and the endoscope probe; therefore, the product of the second correction weight corresponding to each edge pixel and the initial corrected distance is used as the true distance between each edge pixel and the endoscope probe.

[0047] Step S4: Based on the actual distance and the angle corresponding to the ultrasound image of each edge pixel, a three-dimensional model of the spinal endoscopic surgery process is constructed in real time, and then the spinal endoscopic surgery path is planned.

[0048] To accurately construct a three-dimensional model of the spinal endoscopic surgery process in real time, this embodiment uses the centroid of the endoscope probe as the origin and constructs a three-dimensional coordinate system along the basic axes of the human body (including the vertical axis, sagittal axis, and coronal axis). Then, based on the actual distance between each edge pixel in the synchronously acquired ultrasound image at each moment and the endoscope probe, as well as the angle between the ultrasound image of each edge pixel and the sagittal axis, the coordinates of each edge pixel in the three-dimensional coordinate system at each moment are determined. This allows for a more accurate planning of the endoscope probe's path and effectively avoids damage caused by collisions between the endoscope probe and the patient's tissue walls. The determination of the basic axes of the human body and the coordinates of each edge pixel in the three-dimensional coordinate system are well-known and will not be elaborated further.

[0049] In summary, this embodiment acquires ultrasound images from various angles, including those of the endoscopic probe, in real time. Based on the position and shape of the human tissue edge lines in adjacent ultrasound images, it obtains the consistency of the human tissue edge lines' shapes and matching edge lines. Based on the positional relationship between the human tissue edge lines in each ultrasound image and the edge pixels on the matching edge lines in the corresponding ultrasound images, it obtains the matching pixels for each edge pixel. Based on the positions of the edge pixels and their matching pixels, it obtains the true distance between the edge pixels and the endoscopic probe, thereby constructing a three-dimensional model of the spinal endoscopic surgery process in real time. This invention, by accurately acquiring the true distance between the edge pixels on the human tissue edge lines and the endoscopic probe in real time, facilitates the accurate construction of a three-dimensional model of the spinal endoscopic surgery process.

[0050] Example 2: This invention also proposes an ultrasound-guided spinal endoscopic surgical path planning system; please refer to [link to relevant documentation]. Figure 3 The diagram illustrates a structural diagram of an ultrasound-guided spinal endoscopic surgery path planning system according to an embodiment of the present invention. The system includes: an image acquisition module 10, a human tissue edge line analysis module 20, a real distance acquisition module 30, and a spinal endoscopic surgery path planning module 40.

[0051] The image acquisition module 10 is used to acquire ultrasound images from various angles, including the endoscope probe, during spinal endoscopic surgery in real time and synchronously.

[0052] The human tissue edge line analysis module 20 is used to obtain the shape consistency and matching edge line of each human tissue edge line based on the position distribution and shape similarity of the human tissue edge lines in each synchronously acquired ultrasound image and its adjacent ultrasound images.

[0053] The true distance acquisition module 30 is used to acquire the matching pixel of each edge pixel based on the positional relationship between each human tissue edge line in each ultrasound image and the edge pixel on the matching edge line in the ultrasound image. Based on the shape consistency of the human tissue edge line where each edge pixel is located, the positional relationship between each edge pixel and its matching pixel, the positional relationship between each edge pixel and its matching pixel in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, the distance between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, and the positional relationship between each edge pixel and its matching pixel and the corresponding ultrasound probe in the ultrasound image, the true distance between each edge pixel and the endoscope probe in the ultrasound image, and the positional relationship between each edge pixel and its matching pixel and the endoscope probe in the ultrasound image, the true distance between each edge pixel and the endoscope probe is acquired.

[0054] The spinal endoscopic surgery path planning module 40 is used to construct a three-dimensional model of the spinal endoscopic surgery process in real time based on the actual distance and the angle corresponding to the ultrasound image of each edge pixel, and then plan the spinal endoscopic surgery path.

[0055] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the ultrasound-guided spinal endoscopic surgery path planning system and the ultrasound-guided spinal endoscopic surgery path planning method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0056] Example 3: This invention also proposes an ultrasound-guided spinal endoscopic surgery path planning device. The device includes a memory and a processor. The memory stores executable program code, and the processor calls and executes the executable program code to perform the ultrasound-guided spinal endoscopic surgery path planning method provided in the embodiments of this application. Specifically, the device may be a chip, component, or module. The chip may include a connected processor and memory; the memory stores instructions, and when the processor calls and executes the instructions, the chip can perform the ultrasound-guided spinal endoscopic surgery path planning method provided in the above embodiments.

[0057] Furthermore, this application also protects a computer device; please refer to [link to relevant documentation]. Figure 4The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402. When the processor 402 executes the computer program 403, the computer device can perform any of the ultrasound-guided spinal endoscopic surgical path planning methods described above.

[0058] Example 4: This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-described related method steps to implement the ultrasound-guided spinal endoscopic surgery path planning method provided in the above embodiment.

[0059] Example 5: This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the ultrasound-guided spinal endoscopic surgery path planning method provided in the above embodiment.

[0060] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0061] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0062] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An ultrasound-guided endoscopic spinal surgery path planning method, characterized by, The method comprises the following steps: Real-time synchronous acquisition of ultrasonic images of each angle containing an endoscope probe in a spinal endoscopic surgery process; According to the position distribution and shape similarity of the human tissue edge lines in each synchronous acquired ultrasonic image and its adjacent ultrasonic image, the shape consistency degree and matching edge line of each human tissue edge line are obtained; According to the position relationship between each edge pixel point on each human tissue edge line and the matching edge line in the position relative ultrasonic image, the matching pixel point of each edge pixel point is obtained; according to the shape consistency degree of the human tissue edge line where each edge pixel point is located, the position relationship between each edge pixel point and its matching pixel point, the position relationship between the endoscope probe in the ultrasonic image where each edge pixel point and its matching pixel point are located and the corresponding ultrasonic probe in the ultrasonic image, the distance between each edge pixel point and its matching pixel point and the corresponding ultrasonic probe in the ultrasonic image, and the position relationship between each edge pixel point and its matching pixel point and the corresponding ultrasonic probe in the ultrasonic image, the real distance between each edge pixel point and the endoscope probe is obtained; According to the real distance and the angle corresponding to each edge pixel point in the ultrasonic image, a three-dimensional model in the spinal endoscopic surgery process is constructed in real time, and then a spinal endoscopic surgery path is planned.

2. The ultrasound-guided endoscopic spinal surgery path planning method of claim 1, wherein, The shape consistency degree is obtained by: For any time, the ultrasonic images acquired synchronously at this time are all taken as target images, and the two target images adjacent in position are all taken as a group of matching images; For any group of matching images, the two target images in the group of matching images are overlapped, and the preset number of human tissue edge lines in another target image close in position to each human tissue edge line are all taken as the reference edge line of each human tissue edge line; The two target images in the group of matching images are overlapped and mapped into a two-dimensional coordinate system to obtain the pixel point coordinate sequence corresponding to each human tissue edge line; For any human tissue edge line and any reference edge line thereof, the DTW value of the pixel point coordinate sequence of the human tissue edge line and the reference edge line is obtained by a dynamic time warping algorithm; The result of negative correlation and normalization of the DTW value is taken as the matching degree of the human tissue edge line and the reference edge line; The matching degree of the human tissue edge line and each reference edge line thereof is obtained, and the maximum matching degree is taken as the shape consistency degree of the human tissue edge line.

3. The ultrasound-guided endoscopic spinal surgery path planning method of claim 2, wherein, The matching edge line is obtained by: For any human tissue edge line, the reference edge line corresponding to the shape consistency degree of the human tissue edge line is taken as the matching edge line of the human tissue edge line; Wherein, the matching edge line of the matching edge line of the human tissue edge line is also taken as the matching edge line of the human tissue edge line; the human tissue edge line corresponds to a matching edge line and a shape consistency degree in other ultrasonic images.

4. The ultrasound-guided endoscopic spinal surgery path planning method of claim 1, wherein, The matching pixel point is obtained by: For any edge pixel point on any human tissue edge line in any ultrasound image, the position of the ultrasound image is taken as an analysis image relative to the ultrasound image; The ultrasound image and the analysis image are overlaid and mapped into a two-dimensional coordinate system, the Euclidean distance between the edge pixel point and each edge pixel point on the matching edge line of the human tissue edge line in the analysis image is obtained, and the Euclidean distance is taken as the matching distance of the edge pixel point; The edge pixel point on the matching edge line of the human tissue edge line in the analysis image corresponding to the smallest matching distance is taken as the matching pixel point of the edge pixel point.

5. The ultrasound-guided endoscopic spinal surgery path planning method of claim 4, wherein, The method for obtaining the real distance is: According to the shape consistency degree of the human tissue edge line where each edge pixel point is located, the positional relationship between each edge pixel point and its matching pixel point, and the positional relationship between the endoscope probe in the ultrasound image where each edge pixel point and its matching pixel point are located, the positional deviation degree of each edge pixel point and its matching pixel point is obtained; According to the distances between each edge pixel point and the endoscope probe in the ultrasound image where each edge pixel point and its matching pixel point are located, and the distances between the corresponding ultrasound probe and the endoscope probe in the ultrasound image, the reference weights of each edge pixel point and its matching pixel point are obtained respectively; According to the reference weights of each edge pixel point and its matching pixel point, and the positional relationship between each edge pixel point and the endoscope probe in the ultrasound image where each edge pixel point and its matching pixel point are located, the initial correction distance between each edge pixel point and the endoscope probe is obtained; The result of negative correlation and normalization of the positional deviation degree of each edge pixel point and its matching pixel point is taken as the second correction weight of each edge pixel point; The product of the second correction weight and the initial correction distance is taken as the real distance between each edge pixel point and the endoscope probe.

6. The ultrasound-guided endoscopic spinal surgery path planning method of claim 5, wherein, The method for obtaining the positional deviation degree is: For any edge pixel point, the matching distance between the edge pixel point and its matching pixel point is taken as a first distance; The Euclidean distance between the endoscope probe in the ultrasound image where the edge pixel point and its matching pixel point are located and the corresponding coordinate point is obtained as a second distance; The addition result of the first distance and the second distance is taken as the distance deviation value of the edge pixel point and its matching pixel point; The mean value of the shape consistency degree of the human tissue edge line where the edge pixel point is located is taken as the reference consistency degree of the human tissue edge line where the edge pixel point is located; The ratio of the distance deviation value and the reference consistency degree is taken as the positional deviation degree of the edge pixel point and its matching pixel point.

7. The ultrasound-guided endoscopic spinal surgery path planning method of claim 5, wherein, The method for obtaining the reference weight of each edge pixel point and its matching pixel point is: For any edge pixel point, the ultrasound image where the edge pixel point is located is taken as a first designated image; The distance between the edge pixel point and the corresponding ultrasound probe of the first designated image is taken as a first designated distance; The distance between the endoscope probe in the first designated image and the corresponding ultrasound probe of the first designated image is taken as a second designated distance; take the ultrasound image in which the matching pixel point of the edge pixel point is located as a second designated image; take the distance between the matching pixel point of the edge pixel point and the ultrasound probe corresponding to the second designated image as a third designated distance; take the distance between the endoscope probe in the second designated image and the ultrasound probe corresponding to the second designated image as a fourth designated distance; take the sum of the first designated distance and the second designated distance as the distance analysis value of the edge pixel point; take the sum of the third designated distance and the fourth designated distance as the distance analysis value of the matching pixel point of the edge pixel point; take the sum of the distance analysis value of the edge pixel point and the distance analysis value of the matching pixel point of the edge pixel point as the overall distance analysis value; take the result of negatively correlating the ratio of the distance analysis value of the edge pixel point to the overall distance analysis value as the reference weight of the edge pixel point; take the result of negatively correlating the ratio of the distance analysis value of the matching pixel point of the edge pixel point to the overall distance analysis value as the reference weight of the matching pixel point of the edge pixel point.

8. The ultrasound-guided endoscopic spinal surgery path planning method of claim 5, wherein, The method for obtaining the initial correction distance is: for any edge pixel point, take the Euclidean distance between the edge pixel point and the coordinate point corresponding to the endoscope probe in the ultrasound image in which the edge pixel point is located as the first reference distance between the edge pixel point and the endoscope probe; take the Euclidean distance between the matching pixel point of the edge pixel point and the coordinate point corresponding to the endoscope probe in the ultrasound image in which the matching pixel point is located as the second reference distance between the edge pixel point and the endoscope probe; take the product of the reference weight of the edge pixel point and the first reference distance as a first correction distance; take the product of the reference weight of the matching pixel point of the edge pixel point and the second reference distance as a second correction distance; take the sum of the first correction distance and the second correction distance as the initial correction distance between the edge pixel point and the endoscope probe.

9. The ultrasound-guided endoscopic spinal surgery path planning method of claim 1, wherein, The method for obtaining the endoscope probe in the ultrasound image is: obtain the endoscope probe in each ultrasound image through a threshold segmentation algorithm.

10. The ultrasound-guided endoscopic spinal surgery path planning method of claim 1, wherein, The method for obtaining the human tissue edge line is: obtain the human tissue edge line in each ultrasound image through a Canny edge detection algorithm.

Images