Joint cavity puncture positioning method and system based on ultrasonic image guidance

Abstract

The invention discloses an articular cavity puncture positioning method and system based on ultrasonic image guidance, and belongs to the technical field of articular cavity puncture positioning, and the method comprises the following steps: ultrasonic image acquisition, sound shadow area identification, articular cavity model construction, puncture path planning and path visual guidance. The method comprises the following steps: constructing three-dimensional ultrasonic body data through multi-view ultrasonic scanning, and establishing a space model of bones and sound shadow on this basis; then, an anatomical prior model is introduced for registration, so that intelligent form completion is carried out on the articular cavity under sound shadow shielding; then, double constraints of obstacle avoidance skeleton and sound shadow blind area avoidance are comprehensively considered, and optimization calculation is conducted on a puncture path; and finally, the optimized puncture path is mapped to a two-dimensional ultrasonic image to be displayed in an overlapped mode. Accurate positioning and safe puncture guiding of the articular cavity can be achieved under the sound shadow interference condition, and the problem that due to the fact that a blind area exists in an ultrasonic image, articular cavity puncture positioning is inaccurate is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of joint cavity puncture and localization technology, and in particular to a joint cavity puncture and localization method and system based on ultrasound image guidance. Background Technology

[0002] Joint aspiration is a commonly used diagnostic and therapeutic procedure widely applied in clinical fields such as orthopedics, rheumatology and immunology, and rehabilitation medicine. Due to the complex anatomy of joints and the significant individual differences among patients in terms of body type, anatomical structure, and pathological condition, the puncture procedure requires a high degree of precision in positioning.

[0003] In existing technologies, joint cavity puncture localization is mostly performed using ultrasound image guidance. Specifically, before performing joint cavity puncture, medical personnel first position the patient according to the type of the target joint to fully expose the joint area, and then perform routine disinfection and aseptic treatment of the puncture area. Subsequently, an ultrasound probe matching the joint depth is selected to perform multi-sectional scanning of the target joint to obtain ultrasound images of the joint capsule, joint cavity, and surrounding soft tissues, and to identify the location of joint effusion and nearby important anatomical structures such as blood vessels, nerves, and tendons. After obtaining clear ultrasound images, the location of the joint cavity and a suitable needle insertion plane are determined based on the image information, and the puncture point and direction are selected on the skin surface. During the puncture, the needle angle and depth are adjusted by observing the changes in the position of the needle echo in the ultrasound image, allowing the needle to gradually approach and enter the joint cavity. When the ultrasound image shows that the needle tip is inside the joint cavity, the puncture localization is confirmed. After the puncture, the needle is removed, and the condition inside the joint cavity can be confirmed again by ultrasound scanning.

[0004] For example, Chinese invention patent CN120318329B discloses a joint cavity localization method and device based on a deep learning model, including: S1, acquiring arthroscopic images and preprocessing the arthroscopic images to obtain enhanced images; S2, acquiring joint cavity location information based on a pre-trained joint cavity localization model and the enhanced images; the pre-training of the joint cavity localization model includes: A1, training a pre-constructed deep learning model using a pre-collected general medical image dataset to obtain a preliminary localization model; A2, expanding a pre-collected rare bone and joint disease arthroscopic image dataset to obtain an expanded image dataset, and then dynamically adjusting the regularization strength of the preliminary localization model based on the joint cavity morphology diversity index of the expanded image dataset according to an adaptive regularization coefficient function to obtain a joint cavity localization model.

[0005] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0006] In knee joints with severe degenerative changes, large osteophytes are often present at the patellar margin or in the intercondylar fossa of the femur. When ultrasound encounters dense bone, it produces total reflection, creating a significant acoustic shadow behind it. During puncture guidance, the acoustic shadow created by these osteophytes severely obscures the underlying tissue structures, making it impossible for the surgeon to observe the joint cavity or needle tip located within the acoustic shadow on the screen. If the needle insertion path happens to pass through the acoustic shadow area, the operator faces the risk of "blind puncture," which can easily lead to puncture failure or damage to the soft tissue surrounding the osteophytes.

[0007] Meanwhile, in cases of severe knee joint disease, especially with significant bone hyperplasia, bone destruction, or severe inflammation, the normal anatomical structure of the knee joint undergoes significant changes, with narrowing or morphological disturbances in the joint space, resulting in unclear joint cavity boundaries and complex sonographic features in ultrasound images. Existing ultrasound-guided joint cavity puncture localization methods mainly rely on the operator's empirical judgment of real-time images, which is easily affected by factors such as strong bony echoes and acoustic shadowing, synovial thickening, and soft tissue edema, making it difficult to accurately identify the joint cavity location and determine a stable puncture point and needle insertion path. Summary of the Invention

[0008] To address the technical problem in existing technologies where the presence of acoustic shadowing blind spots in ultrasound images makes accurate joint cavity puncture localization difficult, this invention provides a joint cavity puncture localization method and system based on ultrasound image guidance. The technical solution is as follows:

[0009] On the one hand, a method for joint cavity puncture localization based on ultrasound image guidance is provided. This method includes: performing multi-view ultrasound scanning of the target joint region to acquire ultrasound image data containing spatial location information; performing three-dimensional reconstruction of the ultrasound image data based on the spatial location information to construct three-dimensional ultrasound volume data containing multi-view tissue information; identifying the bony structures of the joint based on the three-dimensional ultrasound volume data to obtain a bone tissue spatial model; and determining the ultrasound-obstructed area of ​​the bone tissue in space according to the geometric features and ultrasound propagation characteristics of the bone tissue spatial model, constructing a sound-shadow spatial model spatially registered with the three-dimensional ultrasound volume data; and then performing... The procedure involves registration and, using the anatomical topological constraints in the prior anatomical model, performing 3D morphological inference and completion of the joint cavity region occluded by the acoustic shadowing model to generate a 3D joint cavity model. The puncture target point is then determined within this 3D model. Starting from the determined needle insertion point and ending at the puncture target point, a cost constraint for puncture path planning is established by combining the bone tissue spatial model and the acoustic shadowing spatial model. The puncture path is then optimized to output a path that avoids bone tissue and reduces the risk of the puncture path falling within the acoustic shadowing risk area. The puncture path is then spatially correlated with real-time acquired 2D ultrasound images, and the puncture path and acoustic shadowing risk area are marked on the 2D ultrasound images.

[0010] On the other hand, an ultrasound-guided joint cavity puncture localization system is provided, including: an ultrasound image acquisition module, a sound shadow region recognition module, a joint cavity model construction module, a puncture path planning module, and a path visualization guidance module; the ultrasound image acquisition module is used to perform multi-view ultrasound scanning of the target joint region, acquire ultrasound image data containing spatial location information, and perform three-dimensional reconstruction of the ultrasound image data based on the spatial location information to construct three-dimensional ultrasound volume data containing multi-view tissue information; the sound shadow region recognition module is used to identify the bony structures of the joint based on the three-dimensional ultrasound volume data to obtain a bone tissue spatial model, and determine the ultrasound-obstructed area of ​​the bone tissue in space according to the geometric features and ultrasound propagation characteristics of the bone tissue spatial model, and construct a sound shadow spatial model spatially registered with the three-dimensional ultrasound volume data; the joint cavity The model building module is used to register the bone tissue spatial model with the pre-set anatomical prior model, and use the anatomical topological constraints in the anatomical prior model to perform 3D morphological inference and completion of the joint cavity region occluded by the acoustic shadowing spatial model, generating a 3D joint cavity model, and determining the puncture target point in the 3D joint cavity model; the puncture path planning module is used to establish cost constraints for puncture path planning with the determined needle insertion point as the starting point and the puncture target point as the ending point, combining the bone tissue spatial model and the acoustic shadowing spatial model, and optimize the puncture path to output a puncture path that avoids bone tissue and reduces the puncture path from falling within the acoustic shadowing risk area; the path visualization guidance module is used to spatially associate and display the puncture path with the real-time acquired 2D ultrasound image, and mark the puncture path and acoustic shadowing risk area in the 2D ultrasound image.

[0011] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0012] This invention addresses the common problems of osteophyte formation, bony structural abnormalities, and the resulting ultrasound shadowing inaccurate joint cavity puncture positioning in severely degenerative knee joints. It proposes an ultrasound image-guided joint cavity puncture positioning method and system, which can effectively overcome the shortcomings of traditional ultrasound-guided puncture in complex lesion conditions, such as poor visibility, unstable positioning, and high operational risks.

[0013] Specifically, this invention, firstly, constructs three-dimensional ultrasound volume data containing rich spatial tissue information by performing multi-view ultrasound scanning combined with spatial location information for three-dimensional reconstruction. Compared to traditional methods that rely on a single two-dimensional section, this approach can comprehensively express the three-dimensional distribution relationship of osteophytes, femoral condyles, patella, and surrounding soft tissues within the knee joint from a spatial perspective. This significantly reduces the problem of the joint cavity being "invisible" due to local strong echoes or acoustic shadowing from a single viewpoint, providing a more comprehensive and stable structural perception basis for complex degenerative joints.

[0014] Secondly, by identifying bony structures and constructing a spatial model of bone tissue in 3D ultrasound volume data, and further combining the geometric features of bone tissue and ultrasound propagation characteristics, the range of ultrasound obstruction by dense bone such as osteophytes in space is clarified, forming a sound shadow spatial model precisely registered with the 3D ultrasound volume data. This sound shadow spatial model can quantitatively describe the sound shadow area generated by osteophytes, which helps to actively identify potential "blind spots" in ultrasound imaging, avoiding blind needle insertion when the operator cannot observe the needle tip and joint cavity structure, thus reducing the risk of "blind puncture" from the source.

[0015] Furthermore, this invention registers the bone tissue spatial model with a pre-set anatomical prior model and utilizes the anatomical topological constraints in the prior model to perform three-dimensional morphological inference and completion of joint cavity regions that are obscured by acoustic shadowing or have disordered image features. In cases where there is significant bone hyperplasia, joint space narrowing, or severe anatomical deformation in the knee joint, this method no longer relies solely on local sonographic features of real-time ultrasound images, but instead combines medical anatomical principles to rationally reconstruct the joint cavity space, thereby significantly improving the robustness and accuracy of joint cavity localization.

[0016] Based on this, the present invention automatically determines the puncture target point in the reconstructed three-dimensional model of the joint cavity, and combines the needle insertion point on the skin surface, the bone tissue spatial model, and the acoustic shadow spatial model to establish a puncture path planning cost function that includes bone avoidance constraints and acoustic shadow risk constraints, and optimizes the needle insertion path. The path planning result can avoid osteophytes and important bony structures while minimizing the length of the puncture path passing through high-risk acoustic shadow areas, thereby effectively reducing the probability of the needle tip being invisible, puncture failure, or damage to surrounding soft tissues due to osteophyte obstruction.

[0017] Finally, by spatially linking the optimized puncture path with real-time two-dimensional ultrasound images and visually marking the planned puncture path and acoustic risk areas within the two-dimensional ultrasound images, doctors can clearly perceive the needle insertion direction, target location, and potentially invisible risk areas while maintaining their existing ultrasound operation habits. This method significantly reduces reliance on the operator's personal experience and improves the safety, success rate, and repeatability of intra-articular puncture procedures in complex clinical scenarios such as severely degenerated knee joints, demonstrating high clinical application value. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0019] Figure 1A schematic diagram of the structure of the ultrasound image-guided joint cavity puncture positioning system provided in the embodiments of this application;

[0020] Figure 2 This is a flowchart illustrating the preprocessing logic of the ultrasound image-guided joint cavity puncture localization method provided in the embodiments of this application. Detailed Implementation

[0021] Embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present disclosure are shown in the drawings, it should be understood that embodiments of the present disclosure may be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure.

[0022] It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure. In the description of the embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "this embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects.

[0023] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0024] Joint aspiration, a common minimally invasive procedure for diagnosing and treating joint diseases, is widely used in clinical settings such as knee degenerative diseases, inflammatory joint diseases, and the management of joint effusions. Ultrasound imaging, due to its advantages of being radiation-free, real-time, and bedside-accessible, has become an important guiding tool for joint aspiration. However, in current clinical practice, ultrasound-guided joint aspiration still heavily relies on the operator's experience, requiring a high level of image interpretation and spatial reasoning skills. In cases of complex anatomical structures or severe lesions, the difficulty and risk of the procedure increase significantly.

[0025] In severe degenerative knee joint disease, significant osteophyte formation, bone deformities, and joint space narrowing are common complications. Dense bone tissue tends to produce strong echoes and significant acoustic shadowing during ultrasound imaging, obscuring deeper tissues and making it difficult to clearly visualize the joint cavity and needle tip in two-dimensional ultrasound images. Simultaneously, factors such as synovial thickening, soft tissue edema, and inflammatory responses further complicate ultrasound imaging, blurring joint cavity boundaries, destabilizing sonographic features, increasing the risk of puncture positioning errors and "blind puncture," and easily leading to puncture failure or damage to surrounding tissues.

[0026] Therefore, it is necessary to systematically analyze and improve the joint cavity puncture process in order to achieve more reliable puncture positioning and path planning under complex anatomy and severe lesion conditions, thereby improving the safety, accuracy and clinical repeatability of joint cavity puncture operations.

[0027] As a first aspect of the embodiment, this application provides a method for joint cavity puncture localization based on ultrasound image guidance. The method includes: performing multi-view ultrasound scanning on the target joint region to acquire ultrasound image data containing spatial location information; performing three-dimensional reconstruction of the ultrasound image data based on the spatial location information to construct three-dimensional ultrasound volume data containing multi-view tissue information; identifying the bony structures of the joint based on the three-dimensional ultrasound volume data to obtain a bone tissue spatial model; and determining the ultrasound-obstructed area of ​​the bone tissue in space according to the geometric features and ultrasound propagation characteristics of the bone tissue spatial model, constructing an acoustic-shadow spatial model spatially registered with the three-dimensional ultrasound volume data; and matching the bone tissue spatial model with a pre-set anatomical... The prior model is registered, and the anatomical topological constraints in the anatomical prior model are used to perform three-dimensional morphological inference and completion of the joint cavity region occluded by the acoustic shadow space model, generating a three-dimensional joint cavity model, and determining the puncture target point in the three-dimensional joint cavity model; taking the needle insertion point determined on the skin surface as the starting point and the puncture target point as the ending point, the cost constraints of puncture path planning are established by combining the bone tissue space model and the acoustic shadow space model, and the puncture path is optimized to output a puncture path that avoids bone tissue and reduces the puncture path from being located in the acoustic shadow risk area; the puncture path is spatially correlated with the real-time acquired two-dimensional ultrasound image and the puncture path and acoustic shadow risk area are marked in the two-dimensional ultrasound image.

[0028] As an embodiment of the second aspect, such as Figure 1 The diagram shown is a structural schematic of the ultrasound image-guided joint cavity puncture localization system provided in an embodiment of this application. (Refer to...) Figure 1 The system includes: an ultrasound image acquisition module, an acoustic shadow region recognition module, a joint cavity model construction module, a puncture path planning module, and a path visualization guidance module.

[0029] Specifically, the ultrasound image acquisition module is used to perform multi-view ultrasound scanning of the target joint area, acquire ultrasound image data containing spatial location information, perform three-dimensional reconstruction of the ultrasound image data based on the spatial location information, and construct three-dimensional ultrasound body data containing multi-view tissue information.

[0030] The acoustic shadow region recognition module is used to identify the bony structures of joints based on three-dimensional ultrasound volume data to obtain a spatial model of bone tissue. Based on the geometric features and ultrasound propagation characteristics of the bone tissue spatial model, it determines the ultrasound-blocked area of ​​bone tissue in space and constructs an acoustic shadow spatial model that is spatially registered with the three-dimensional ultrasound volume data.

[0031] The joint cavity model construction module is used to register the bone tissue spatial model with the pre-set anatomical prior model, and use the anatomical topological constraints in the anatomical prior model to perform three-dimensional morphological inference and completion of the joint cavity area occluded by the acoustic shadow spatial model, generate a three-dimensional joint cavity model, and determine the puncture target point in the three-dimensional joint cavity model.

[0032] The puncture path planning module is used to establish cost constraints for puncture path planning by taking the needle insertion point determined on the skin surface as the starting point and the puncture target point as the ending point, and combining the bone tissue spatial model and the acoustic shadow spatial model. It also optimizes the puncture path to output a puncture path that avoids bone tissue and reduces the puncture path from being located in the acoustic shadow risk area.

[0033] The path visualization guidance module is used to spatially correlate and display the puncture path with the real-time acquired two-dimensional ultrasound images, and to mark the puncture path and acoustic shadow risk areas in the two-dimensional ultrasound images.

[0034] It is important to understand that the steps of the method of this invention are not implemented in isolation, but rather form a closely coordinated and progressively enhanced technical chain around the core objective of "improving the visibility, localization, and safety of joint cavity puncture in complex degenerative joints." Through the organic connection and synergistic cooperation of the above steps, this invention achieves integrated support for the entire process from ultrasound data acquisition to puncture execution guidance, effectively reducing information fragmentation and error accumulation, and improving the accuracy, consistency, and safety of joint cavity puncture localization and path planning.

[0035] As a further approach, ultrasound image acquisition is the first step in the ultrasound-guided joint cavity puncture localization method. However, due to the influence of speckle noise, differences in tissue acoustic properties, and the strong reflection effect of bone tissue, the grayscale distribution of different tissue types in the original ultrasound image often overlaps. The boundary between bony hyperechoic areas and hypoechoic shadow areas is prone to blurred transitions, and soft tissue areas generally suffer from strong noise interference. Directly performing bone tissue identification, shadow modeling, and subsequent 3D reconstruction based on the original ultrasound image can easily lead to misclassification of tissue types, inaccurate estimation of shadow range, and accumulated errors in subsequent spatial analysis, affecting the reliability of joint cavity localization and puncture path planning.

[0036] Therefore, after acquiring ultrasound image data containing spatial location information using an ultrasound probe, the process also includes preprocessing the acquired ultrasound image data, such as... Figure 2 The diagram shown is a preprocessing logic flowchart of the ultrasound image-guided joint cavity puncture localization method provided in this application embodiment; see reference. Figure 2 Preprocessing includes the following steps:

[0037] First, pixel regions in the ultrasound image data with grayscale values ​​greater than a first preset threshold are designated as bony hyperechoic regions. In a preferred embodiment, for an 8-bit depth ultrasound image (grayscale range 0-255), the first preset threshold is set to a value between 200 and 240. To improve robustness, an adaptive threshold calculated using Otsu's method can be used as a reference, and the first preset threshold can be set to the adaptive threshold plus a fixed offset (e.g., +30) to ensure that only the brightest bony hyperechoic regions are captured.

[0038] Secondly, based on the direction of ultrasound propagation, in the deep direction of the bony hyperechoic region, pixel regions with gray values ​​less than a second preset threshold are identified as hypoechoic regions, and the remaining pixel regions are recorded as soft tissue regions. The first preset threshold is greater than the second preset threshold. For example, the second preset threshold can be set to 10 to 50. At the same time, in order to prevent the acoustic shadow region from extending indefinitely, a search depth limit is set. For example, the search is only conducted to the pixel distance corresponding to 20mm to 50mm below the current bone surface pixel, or until a pixel with a gray value that rises back to exceed the second preset threshold is encountered.

[0039] Next, edge enhancement processing is performed on the boundary between the bony hyperechoic region and the hypoechoic region, and smoothing and noise reduction processing is performed on the soft tissue region other than the boundary edge. The specific processing stages of edge enhancement processing include: first, edge detection is performed using the Canny operator, followed by sharpening enhancement using a high-pass filtering enhancement algorithm; smoothing and noise reduction processing is implemented using bilateral filtering or nonlocal mean filtering algorithms.

[0040] Through the above preprocessing and analysis steps, the strong echoes of bone, the low echoes of acoustic shadows, and the soft tissue regions are effectively distinguished and enhanced in terms of grayscale distribution and boundary features. This not only improves the stability and accuracy of subsequent three-dimensional ultrasound reconstruction, bone tissue identification, and acoustic shadow space modeling, but also provides clearer and more reliable basic data for joint cavity space inference and puncture path planning.

[0041] It should also be noted that the three-dimensional reconstruction of ultrasound image data based on spatial location information specifically includes:

[0042] First, a set number of two-dimensional ultrasound images are acquired by an electromagnetic positioning sensor connected to the ultrasound probe, and the corresponding spatial position information is collected simultaneously. Based on the spatial position information, the projection distribution range of the pixels in each frame of the two-dimensional ultrasound image in the three-dimensional coordinate space is calculated. The projection distribution range can be determined by the maximum and minimum coordinate values ​​of the pixels in the three-dimensional space.

[0043] Secondly, based on the projection distribution range, the echo grayscale values ​​in the two-dimensional ultrasound image are mapped to the corresponding three-dimensional voxel positions using an interpolation algorithm. Next, according to the preset fusion weight allocation rule, corresponding fusion weights are assigned to the echo grayscale values ​​mapped to the three-dimensional voxel positions. The fusion weight allocation rule is configured such that the fusion weight is negatively correlated with the spatial distance from the pixel to the voxel center and positively correlated with the perpendicularity of the ultrasound probe to the incident angle of the voxel. Then, the echo grayscale values ​​mapped to the same voxel position are weighted and summed based on the fusion weights to generate three-dimensional ultrasound volume data.

[0044] Specifically, according to the preset fusion weight allocation rules, a fusion weight is assigned to each echo grayscale value mapped to a voxel location. The fusion weight is determined by both a distance weight factor and an angle weight factor, i.e., their product. The distance weight factor is calculated using a Gaussian function. In the implementation of the angle weight factor, firstly, the beam direction vector of the ultrasound probe is calculated. For convex or linear array probes, this is typically perpendicular to the plane of the probe array and points deep into the tissue being measured. Secondly, the tissue surface normal vector at the voxel location is calculated. Since the reconstruction process is iterative, it can be approximated based on the grayscale gradient in the current voxel's neighborhood. Finally, the perpendicularity weight of the incident angle is calculated. The incident angle is defined as the angle between the beam direction vector and the surface normal vector. When the incident angle is 0, it indicates perpendicular incidence.

[0045] This three-dimensional reconstruction step achieves a highly consistent mapping from two-dimensional section information to three-dimensional volume data by tightly coupling two-dimensional ultrasound images with their corresponding spatial location information. This provides a more accurate three-dimensional data foundation for subsequent bone structure modeling, acoustic shadowing analysis, and puncture path planning, effectively improving the safety, accuracy, and repeatability of joint cavity puncture operations.

[0046] Furthermore, acoustic shadowing region identification is the second step in the ultrasound image-guided joint puncture localization method. It's important to note that constructing an acoustic shadowing space model registered with the three-dimensional ultrasound volume data space is a crucial step in the joint puncture guidance system. This aims to assess and eliminate acoustic shadowing regions in ultrasound images through accurate model reconstruction, thereby improving the accuracy and safety of subsequent analysis and path planning. Specifically, this includes:

[0047] First, the surface mesh structure of the bone tissue spatial model is extracted, and the normal vector information of each surface unit in the surface mesh is calculated. Second, the spatial pose of the ultrasound probe and the beam fan angle during multi-view scanning are obtained, and the spatial coverage of the incident sound beam is reconstructed in the three-dimensional coordinate system. Among them, with the probe position as the vertex and the probe orientation (usually defined as the normal vector direction of the probe) as the central axis, a virtual pyramid or wedge is constructed in the three-dimensional coordinate system according to the fan angle and imaging depth. The internal space region of the wedge represents the spatial coverage of the incident sound beam under the current view.

[0048] Next, taking the position of the ultrasound probe as the origin of the ray, a virtual detection ray is emitted along the propagation direction of the incident sound beam to the surface grid of the bone tissue spatial model. The target intersection point of the virtual detection ray and the surface grid is calculated using a ray-triangle intersection algorithm (such as the Möller-Trumbore algorithm). Then, the spatial voxels located behind the target intersection point along the propagation direction of the sound beam and within the preset imaging depth range are marked as acoustic shadow voxels. All acoustic shadow voxels constitute the acoustic shadow spatial model.

[0049] By accurately constructing a acoustic shadow space model, a complete three-dimensional structural view is provided for ultrasound images, enabling effective identification and processing of acoustic shadow regions within the images. By combining spatial location information, the ultrasound probe's incident angle, and the direction of sound beam propagation, the generated acoustic shadow model significantly improves the accuracy of subsequent image reconstruction, risk assessment, and puncture path planning, ensuring the safety and accuracy of the joint cavity puncture process.

[0050] Furthermore, the joint cavity model construction is the third step in the ultrasound image-guided joint cavity puncture localization method. This involves using the anatomical topological constraints in the prior anatomical model to perform three-dimensional morphological inference and completion of the joint cavity region occluded by the acoustic shadowing model. Specifically, this includes:

[0051] First, the set of bony feature points in the bone tissue spatial model is extracted. The iterative nearest point algorithm is used to rigidly register the bone tissue spatial model with the standard bone part in the anatomical prior model to obtain the spatial transformation matrix. The anatomical prior model not only contains the standard bone part, but also pre-stores the standard curved surface of the joint cavity that matches the bone (such as the surface of the meniscus bursa between the femur and tibia).

[0052] Secondly, based on the spatial transformation matrix, the pre-stored standard surface of the joint cavity in the anatomical prior model is mapped to the coordinate system where the three-dimensional ultrasound body data is located to generate a reference joint cavity model; using the obtained spatial transformation matrix, coordinate transformation (i.e., multiplication operation) is performed on each vertex of the pre-stored standard surface of the joint cavity in the anatomical prior model to generate a reference joint cavity model.

[0053] Then, the spatial relationship between the sound-shadow space model and the reference joint cavity model is calculated to identify the areas to be completed that are covered by the sound-shadow space model, as well as the visible areas that are not covered by the sound-shadow space model.

[0054] Meanwhile, in order to seamlessly connect the inferred region to be completed with the real visible region, this step uses anatomical topological constraints (i.e., the joint cavity surface is a continuous, smooth, tear-free manifold surface) to perform local surface reconstruction; that is, using the anatomical topological constraint relationship, with the visible region surface shape at the edge of the region to be completed as the boundary condition, the surface in the reference joint cavity model located within the region to be completed is subjected to local geometric deformation and smooth fitting, so that it can be geometrically continuous with the visible region surface, generating the completed three-dimensional model of the joint cavity.

[0055] By combining the anatomical prior model and the acoustic shadow space model, this step achieves reasonable inference and continuous completion of the three-dimensional morphology of the joint cavity even under severe acoustic shadow occlusion and incomplete imaging information. This not only significantly improves the integrity and anatomical rationality of the joint cavity space model, but also provides a stable and reliable spatial basis for subsequent puncture target point determination and path planning, thereby improving the overall safety and success rate of ultrasound-guided joint cavity puncture under complex degenerative lesion conditions.

[0056] It should also be added that the puncture target point is determined in the 3D model of the joint cavity, specifically including:

[0057] First, the three-dimensional model of the joint cavity is voxelized. All voxel points inside the three-dimensional model of the joint cavity are traversed, and the shortest Euclidean distance from each voxel point to the surface mesh of the bone tissue spatial model is calculated.

[0058] Secondly, within the 3D model of the joint cavity, a set of candidate voxel points with the shortest Euclidean distance greater than a preset safety distance is selected to avoid areas close to the bone surface. The preset safety distance is set according to clinical needs; for example, for the knee joint cavity, it can be set to 2mm to 5mm.

[0059] Next, in the candidate voxel point set, the voxel point corresponding to the maximum value of the shortest Euclidean distance is selected as the center position of the joint cavity, and this center position is marked as the puncture target point; if there are multiple voxel points with the maximum value of the shortest Euclidean distance, the distance from each voxel point to the surface of the joint capsule is calculated, and the voxel point closest to the geometric center of the joint capsule is selected as the puncture target point.

[0060] By introducing voxel-level spatial analysis, safety distance constraints, and multiple geometric criteria, this step achieves automated, quantitative, and anatomically rational determination of the puncture target point. This not only significantly improves the accuracy and consistency of puncture target localization but also provides a safe and stable core reference point for subsequent puncture path planning, thereby improving the overall reliability and clinical operability of ultrasound-guided intra-articular puncture under complex lesion conditions.

[0061] Furthermore, puncture path planning is the fourth step in the ultrasound image-guided joint cavity puncture localization method. This involves establishing cost constraints for puncture path planning by combining a bone tissue spatial model and a acoustic shadow spatial model. Specifically, this includes:

[0062] First, a collision cost function for the puncture path is constructed. This function is configured such that if the distance between any point on the puncture path and the surface mesh of the bone tissue spatial model is less than a preset safety threshold, the collision cost function is assigned an infinite value; otherwise, it is assigned a value of zero. Second, a sound and shadow risk cost function for the puncture path is constructed. This function is positively correlated with the path length within the sound and shadow spatial model and also positively correlated with the depth of voxel points on the path from the sound and shadow boundary. The specific calculation formula is as follows: ,in, This represents the cost function of sound and shadow risk. This represents the length penalty coefficient. Indicates the path length. Indicates the depth penalty coefficient. This represents the i-th voxel point on the puncture path. This represents the set of voxel points along the puncture path. Indicates the first path The depth of an individual pixel from the sound-shadow boundary. The length penalty coefficient and depth penalty coefficient are set by professionals according to industry standards, and all data in the formula have been de-normalized.

[0063] Next, the collision cost function and the acoustic shadow risk cost function are weighted and summed to generate the total cost function. The weight coefficient of the acoustic shadow risk cost function in the total cost function is higher than that of the collision cost function. In a specific embodiment, the weight coefficient of the acoustic shadow risk cost function can be set to 5, and the weight coefficient of the collision cost function can be set to 1. Since the collision cost function itself has a "switch" characteristic (0 for safe, infinity for unsafe), its main function is to provide a safety threshold. The purpose of this setting is to greatly amplify the impact of acoustic shadow risk on the total score among all candidate paths that meet the safety conditions, forcing the path planning algorithm to prioritize the path with the least acoustic shadow occlusion and the best visibility, thereby maximizing the visibility of ultrasound-guided operations while ensuring safety.

[0064] By combining bone tissue spatial models and acoustic shadow spatial models to impose cost constraints on puncture path planning, and by accurately quantifying the risk of bone structure collision and acoustic shadowing effects, the puncture path can effectively avoid entering dangerous areas, reducing operational risks. By setting reasonable cost functions and weighting mechanisms, not only is the safety of path planning improved, but the visibility and accuracy of ultrasound-guided puncture procedures are also optimized, thereby significantly improving the success rate of the puncture process and the safety of clinical operations.

[0065] It should also be noted that the puncture path is optimized to output a puncture path that avoids bone tissue and reduces the puncture path's location within the acoustic shadowing risk area. Specifically, this includes:

[0066] First, the three-dimensional space containing the bone tissue spatial model and the acoustic shadow spatial model is discretized into a voxel mesh. Then, a heuristic search algorithm (such as the A-Star Algorithm) is used to perform optimization in the voxel mesh, where the expansion cost of a node is calculated based on the total cost function. Finally, the path node sequence generated by the search is smoothed to generate the puncture path.

[0067] By using voxelized discrete modeling, heuristic search based on the total cost function, and path curve smoothing, this step achieves global optimization and physical executability constraints for the puncture path. This not only ensures that the puncture path avoids bone tissue in space and minimizes acoustic shadowing coverage, but also improves the continuity and clinical operability of the path, thereby enhancing the overall safety, real-time performance, and intelligence of ultrasound-guided intra-articular puncture path planning.

[0068] Furthermore, path visualization guidance is the fifth step in the ultrasound image-guided joint cavity puncture localization method. It also requires the addition of spatially correlated display of the puncture path with real-time acquired two-dimensional ultrasound images, and marking the puncture path and acoustic shadowing risk areas within the two-dimensional ultrasound images. Specifically, this includes:

[0069] First, the puncture path and acoustic shadow space model are projected and mapped onto the current two-dimensional ultrasound image plane. Then, the intersection of the acoustic shadow space model and the two-dimensional ultrasound image plane is calculated, and the corresponding acoustic shadow risk area is extracted. Second, the acoustic shadow risk area is superimposed on the two-dimensional ultrasound image, and the puncture path located within the acoustic shadow risk area is marked with a first color (such as red or yellow), and the puncture path located outside the acoustic shadow risk area is marked with a second color (such as green or blue).

[0070] This step uses a three-dimensional-to-two-dimensional spatial mapping and risk visualization overlay to present the complex puncture path planning results and acoustic-visual risk information in a way that conforms to clinical operating habits in a two-dimensional ultrasound image. This not only improves the operator's real-time perception of the safety and visibility of the puncture path, but also enhances the interpretability and controllability of the puncture decision-making process, thereby improving the overall safety, reliability and clinical practical value of ultrasound-guided intra-articular puncture.

[0071] In summary, this invention achieves a holistic perception of the complex anatomical structure of the joint region through multi-view ultrasound data fusion and 3D reconstruction. Based on this, it establishes a bone tissue spatial model and a sound shadow spatial model to effectively characterize ultrasound-obscured and invisible areas. By combining anatomical priors and topological constraints, it infers and completes the morphology of the joint cavity obscured by sound shadows, obtaining a continuous and reliable 3D structure of the joint cavity even under incomplete imaging conditions. By introducing path cost constraints for bone tissue avoidance and sound shadow risk suppression, it achieves global optimization planning of the puncture path, ensuring that the path simultaneously meets the requirements of accessibility, safety, and imaging visibility. Furthermore, it links the optimized path with real-time 2D ultrasound images for display, providing intuitive and continuous navigation guidance for intraoperative procedures.

[0072] Therefore, this invention can significantly improve the accuracy and stability of joint cavity puncture positioning and path planning, reduce the uncertainty caused by bony obstruction and acoustic interference, reduce the risk of accidental bone puncture and multiple trial needle insertions, shorten operation time, reduce reliance on operator experience and subjective judgment, and enhance consistency and repeatability under different equipment and operator conditions, thereby improving the overall safety, success rate and clinical application value of joint cavity puncture.

[0073] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the above functions can be divided into different functional modules to complete all or part of the functions described above.

[0074] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0075] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units, located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0076] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0077] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the solution, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for joint cavity puncture localization based on ultrasound image guidance, characterized in that, Includes the following steps: Multi-view ultrasound scanning of the target joint region is performed to acquire ultrasound image data containing spatial location information. Based on the spatial location information, the ultrasound image data is reconstructed in three dimensions to build three-dimensional ultrasound body data containing multi-view tissue information. Based on three-dimensional ultrasound volume data, the bony structures of the joints are identified to obtain a spatial model of the bone tissue. Based on the geometric features and ultrasound propagation characteristics of the bone tissue spatial model, the ultrasound-blocking area of ​​the bone tissue in space is determined, and a sound-shadow space model that is spatially registered with the three-dimensional ultrasound volume data is constructed. The bone tissue spatial model is registered with the pre-set anatomical prior model, and the anatomical topological constraints in the anatomical prior model are used to perform three-dimensional morphological inference and completion of the joint cavity region occluded by the acoustic shadow spatial model to generate a three-dimensional model of the joint cavity, and the puncture target point is determined in the three-dimensional model of the joint cavity. Starting from a defined needle insertion point and ending at the puncture target point, a cost constraint for puncture path planning is established by combining a bone tissue spatial model and a sound shadow spatial model. The puncture path is then optimized to output a puncture path that avoids bone tissue and reduces the puncture path from being located in the sound shadow risk area. The puncture path is spatially correlated with the real-time acquired two-dimensional ultrasound images, and the puncture path and acoustic shadow risk areas are marked in the two-dimensional ultrasound images.

2. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 1, characterized in that, The acquisition of ultrasound image data containing spatial location information is followed by preprocessing of the acquired ultrasound image data, which includes the following steps: Pixel regions in ultrasound image data with gray values ​​greater than a first preset threshold are defined as bony hyperechoic regions. Based on the direction of ultrasound propagation, in the deep direction of the bony strong echo region, the pixel region with a gray value less than the second preset threshold is identified as the acoustic shadow low echo region, and the remaining pixel region is recorded as the soft tissue region. The first preset threshold is greater than the second preset threshold. Edge enhancement processing is performed on the boundary between the bony hyperechoic region and the hypoechoic region, and smoothing and noise reduction processing is performed on the soft tissue region other than the boundary edge.

3. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 1, characterized in that, The three-dimensional reconstruction of ultrasound image data based on spatial location information specifically includes: Acquire a set number of two-dimensional ultrasound images and simultaneously collect the corresponding spatial location information. Calculate the projection distribution range of pixels in the three-dimensional coordinate space in each frame of the two-dimensional ultrasound image based on the spatial location information. Based on the projection distribution range, the echo gray values ​​in the two-dimensional ultrasound image are mapped to the corresponding three-dimensional voxel positions using an interpolation algorithm; According to the preset fusion weight allocation rule, the echo gray value mapped to the three-dimensional voxel position is assigned a corresponding fusion weight. The fusion weight allocation rule is configured as follows: the fusion weight is negatively correlated with the spatial distance from the pixel to the voxel center and positively correlated with the perpendicularity of the ultrasonic probe to the incident angle of the voxel. Based on the fusion weight, the echo gray values ​​mapped to the same voxel location are weighted and summed to generate three-dimensional ultrasound body data.

4. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 1, characterized in that, The construction of the acoustic-shadow space model registered with the three-dimensional ultrasound body data space specifically includes: Extract the surface mesh structure of the bone tissue spatial model and calculate the normal vector information of each surface element in the surface mesh; Acquire the spatial pose and beam fan angle of the ultrasonic probe during multi-view scanning, and reconstruct the spatial coverage of the incident beam in a three-dimensional coordinate system; Using the position of the ultrasound probe as the origin of the ray, a virtual detection ray is emitted along the propagation direction of the incident sound beam to the surface grid of the bone tissue spatial model, and the target intersection point between the virtual detection ray and the surface grid is calculated. The spatial voxels located behind the target intersection point along the direction of sound beam propagation and within the preset imaging depth range are marked as acoustic-shadow voxels, and the acoustic-shadow space model is composed of all acoustic-shadow voxels.

5. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 1, characterized in that, The method of using anatomical topological constraints in the prior anatomical model to perform three-dimensional morphological inference and completion of the joint cavity region occluded by the sound and shadow space model specifically includes: Extract the set of bony feature points from the spatial model of bone tissue, and use the iterative nearest point algorithm to rigidly register the spatial model of bone tissue with the standard bone part in the anatomical prior model to obtain the spatial transformation matrix; Based on the spatial transformation matrix, the standard surface of the joint cavity pre-stored in the anatomical prior model is mapped to the coordinate system where the three-dimensional ultrasound body data is located, and a reference joint cavity model is generated. Calculate the spatial relationship between the sound-shadow space model and the reference joint cavity model to identify the area to be completed that is covered by the sound-shadow space model, as well as the visible area that is not covered by the sound-shadow space model. By utilizing anatomical topological constraints and taking the visible surface morphology of the edge of the region to be completed as the boundary condition, the surface in the reference joint cavity model located within the region to be completed is subjected to local geometric deformation and smooth fitting, so that it can be geometrically continuous with the visible surface, thus generating the completed three-dimensional model of the joint cavity.

6. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 1, characterized in that, The determination of the puncture target point in the three-dimensional model of the joint cavity specifically includes: The joint cavity 3D model is voxelized, and all voxel points inside the joint cavity 3D model are traversed to calculate the shortest Euclidean distance from each voxel point to the surface mesh of the bone tissue spatial model. Within the 3D model of the joint cavity, a set of candidate voxel points with the shortest Euclidean distance greater than a preset safety distance is selected to avoid areas that are close to the bone surface. In the candidate voxel point set, the voxel point corresponding to the shortest Euclidean distance maximum value is selected as the center position of the joint cavity, and this center position is marked as the puncture target point; If the shortest Euclidean distance between multiple voxel points is the maximum value, then calculate the distance from each voxel point to the surface of the joint capsule, and select the voxel point closest to the geometric center of the joint capsule as the puncture target point.

7. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 1, characterized in that, The cost constraints for establishing the puncture path planning by combining the bone tissue spatial model and the acoustic shadow spatial model specifically include: A collision cost function for the puncture path is constructed. The collision cost function is configured such that when the distance between any point on the puncture path and the surface mesh of the bone tissue spatial model is less than a preset safety threshold, the collision cost function is assigned an infinite value; otherwise, it is assigned a value of zero. Construct a sound and shadow risk cost function for the puncture path. The sound and shadow risk cost function is positively correlated with the path length of the puncture path within the sound and shadow space model and positively correlated with the depth of the voxel points on the path from the sound and shadow boundary. The collision cost function and the sound and shadow risk cost function are weighted and summed to generate the total cost function, wherein the weight coefficient of the sound and shadow risk cost function in the total cost function is higher than the weight coefficient of the collision cost function.

8. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 7, characterized in that, The optimization calculation of the puncture path to output a puncture path that avoids bone tissue and reduces the puncture path's location within the acoustic shadowing risk area specifically includes: The three-dimensional space containing the bone tissue spatial model and the sound and shadow spatial model is discretized into a voxel mesh; A heuristic search algorithm is used to find the optimal solution in a voxel grid, where the expansion cost of a node is calculated based on the total cost function. The path node sequence generated by the search is smoothed to generate the puncture path.

9. The method for joint cavity puncture localization based on ultrasound image guidance as described in claim 8, characterized in that, The step of spatially associating the puncture path with real-time acquired two-dimensional ultrasound images and marking the puncture path and acoustic shadowing risk areas in the two-dimensional ultrasound images specifically includes: The puncture path and acoustic-shadow space model are projected and mapped onto the current two-dimensional ultrasound image plane; Calculate the intersection of the acoustic shadow space model and the two-dimensional ultrasound image plane, and extract the corresponding acoustic shadow risk area; The acoustic shadowing risk area is superimposed on the two-dimensional ultrasound image, and the puncture path located within the acoustic shadowing risk area is marked with a first color, while the puncture path located outside the acoustic shadowing risk area is marked with a second color.

10. A joint cavity puncture and positioning system based on ultrasound image guidance, using the joint cavity puncture and positioning method based on ultrasound image guidance as described in any one of claims 1-9, characterized in that, include: The system includes an ultrasound image acquisition module, an acoustic shadow region recognition module, a joint cavity model construction module, a puncture path planning module, and a path visualization guidance module. The ultrasound image acquisition module is used to perform multi-view ultrasound scanning on the target joint area, acquire ultrasound image data containing spatial location information, and perform three-dimensional reconstruction based on the spatial location information to construct three-dimensional ultrasound body data containing multi-view tissue information. The acoustic shadow region recognition module is used to identify the joint bony structure based on three-dimensional ultrasound volume data to obtain a bone tissue spatial model, and to determine the ultrasound-blocked area of ​​bone tissue in space according to the geometric features and ultrasound propagation characteristics of the bone tissue spatial model, and to construct an acoustic shadow space model that is spatially registered with the three-dimensional ultrasound volume data. The joint cavity model construction module is used to register the bone tissue spatial model with the preset anatomical prior model, and use the anatomical topological constraints in the anatomical prior model to perform three-dimensional morphological inference and completion of the joint cavity area occluded by the acoustic shadow spatial model, generate a three-dimensional joint cavity model, and determine the puncture target point in the three-dimensional joint cavity model. The puncture path planning module is used to establish cost constraints for puncture path planning by taking the determined needle insertion point as the starting point and the puncture target point as the ending point, combining the bone tissue spatial model and the acoustic shadow spatial model, and to optimize the puncture path to output a puncture path that avoids bone tissue and reduces the puncture path from being located in the acoustic shadow risk area. The path visualization guidance module is used to spatially correlate and display the puncture path with the real-time acquired two-dimensional ultrasound image, and to mark the puncture path and acoustic shadow risk area in the two-dimensional ultrasound image.

Images