Monitoring system for lymph node biopsy puncture and biopsy puncture sampling method

Abstract

The invention relates to the technical field of medical image navigation puncture, in particular to a lymph node biopsy puncture monitoring system and a biopsy puncture sampling method.The lymph node biopsy puncture monitoring method comprises the steps that real-time ultrasound, elastography and color Doppler image data streams of a lymph node area are obtained and fused, and the real-time ultrasound, elastography and color Doppler image data streams of the lymph node area are obtained; and generating a comprehensive feature map integrating the lymph node boundary contour, the internal hardness distribution and the blood vessel walking path. And calculating a safe puncture path and a target spot which avoid the blood vessel and pass through the target hardness region based on the feature map. In the puncture process, the needle inserting track is tracked in real time and dynamically compared with a preset path, and deviation warning and correction guidance are automatically generated. And when the needle point reaches the target spot, triggering sampling, and continuously monitoring the sampling area. According to the method, through multi-source information fusion and real-time navigation deviation correction, the accuracy of puncture sampling and the operation safety are improved.

Description

Technical Field

[0001] This invention relates to the field of medical image-guided puncture technology, and in particular to a monitoring system and a method for lymph node biopsy puncture sampling. Background Technology

[0002] In clinical lymph node biopsy, ultrasound imaging has become a core guiding tool due to its real-time capability. To comprehensively assess the lesion, the procedure often requires the combination of multiple imaging modalities, such as using color Doppler to identify blood flow distribution and avoid blood vessels, and using elastography to assess tissue stiffness and locate suspicious areas. Current standard techniques rely on physicians simultaneously observing or switching between ultrasound images, elastography, and color Doppler flow imaging displayed on separate screens during the procedure. These images from different modalities are not precisely registered spatially and are asynchronous in time. Physicians must mentally synthesize this discrete information to form a holistic judgment of the target structure, stiffness properties, and blood vessel location, and plan the approximate puncture route accordingly.

[0003] This approach, relying on subjective experience to integrate multi-source information, has inherent flaws. The separate display and processing of multimodal information increases the operator's cognitive burden, easily leading to the omission or misjudgment of crucial subtle information in real-time dynamic environments, affecting the objectivity and accuracy of target area selection. Simultaneously, path planning under non-fusion imaging lacks precise quantitative spatial relationship support; whether the path effectively avoids all important blood vessels and accurately traverses the target hardness zone depends primarily on the physician's personal experience, introducing uncertainty and operator variability. Furthermore, during the puncture execution phase, current technology lacks continuous, automated monitoring and feedback of deviations between the instrument trajectory and the predetermined path. Physicians can only rely on visual observation of real-time ultrasound images to estimate needle position. When tissue displacement or needle deflection occurs, it is difficult to immediately and quantitatively identify the deviation and determine the specific correction direction, thus affecting the success rate and safety of the puncture. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and to propose a monitoring system and a biopsy sampling method for lymph node biopsy puncture.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a sampling method for lymph node biopsy, comprising: Acquire real-time multimodal imaging data stream of lymph node region, wherein the real-time multimodal imaging data stream includes ultrasound image sequence, elastography data and color Doppler blood flow signal; The real-time multimodal image data stream is fused to generate a lymph node structural feature map, which includes the contour of the lymph node boundary, the hardness distribution area of ​​the internal tissue, and the spatial course of the main blood vessels. Based on the lymph node structural feature map, the safe puncture path and target location are calculated and determined. The safe puncture path avoids the spatial course of the main blood vessels and passes through the target sampling area in the hardness distribution region of the internal tissue. The actual needle insertion trajectory of the puncture instrument is tracked in real time, and the actual needle insertion trajectory is dynamically compared with the safe puncture path to generate trajectory deviation alarm information and correction suggestions; When the tip of the actual needle trajectory reaches the target location, the biopsy sampling device is triggered to perform tissue sampling, and real-time image changes in the sampling area are continuously monitored during the sampling process.

[0006] As a further aspect of the present invention, the step of fusing the real-time multimodal image data stream to generate a lymph node structural feature map includes: Morphological boundary information of lymph nodes is extracted from the ultrasound image sequence, and an initial geometric model of the lymph nodes is constructed using edge detection and contour tracking algorithms; The elastic imaging data is analyzed, and the hardness values ​​of different regions are mapped to pseudo-color or grayscale levels, which are then superimposed onto the initial geometric model to form a lymph node enhancement model that includes hardness distribution. The color Doppler blood flow signal is analyzed to identify and locate the vascular structures inside and around the lymph node, and the spatial course and diameter information of the main blood vessels are marked in the lymph node enhancement model. The morphological boundary information, the hardness distribution, and the spatial course of the main blood vessels are registered and fused in three dimensions in a unified spatial coordinate system to output a lymph node structural feature map with multi-dimensional structural attributes.

[0007] As a further aspect of the present invention, the step of calculating and determining the safe puncture path and target location based on the lymph node structural feature map includes: Centered on the target sampling area in the lymph node structural feature map, the target sampling area is determined according to the preset hardness threshold range in the hardness distribution area or by manual annotation by the user. Starting from the target sampling area, a spatial straight line or curve is planned in reverse to reach the preset puncture point on the skin; Virtual detection is performed along the described straight line or curve to check whether it intersects with the spatial course of the main blood vessels and whether it passes through specific areas inside the lymph nodes that cause excessive damage. If there is an intersection or passage through a specific area, the angle and depth of the straight line or curve in the space are dynamically adjusted until a path that meets the preset safety distance requirement and has the shortest path length is found, and this path is determined as the safe puncture path. The geometric center point of the target sampling area or the feature point specified by the user is determined as the target point location.

[0008] As a further aspect of the present invention, the real-time tracking of the actual needle insertion trajectory of the puncture instrument and the dynamic comparison of the actual needle insertion trajectory with the safe puncture path include: The spatial coordinates and attitude angles of the positioning probe installed on the puncture instrument are obtained in real time through electromagnetic positioning sensors or optical navigation systems. Based on the spatial coordinates and attitude angle of the positioning probe, combined with the known physical dimensions of the puncture instrument, the real-time spatial position of the tip of the puncture instrument and the direction of the long axis of the instrument are calculated to form the actual needle insertion trajectory. In the same spatial coordinate system where the lymph node structure feature map is located, the actual needle insertion trajectory is drawn and updated in real time. Calculate the spatial distance between the tip of the actual needle insertion trajectory and the corresponding depth point on the safe puncture path, as the lateral deviation value; calculate the angle between the major axis direction of the actual needle insertion trajectory and the direction of the safe puncture path, as the angular deviation value; When the lateral deviation value or the angular deviation value exceeds the respective set allowable threshold, a trajectory deviation alarm message and correction suggestion containing the deviation value and deviation direction are immediately generated.

[0009] As a further aspect of the present invention, the generation of trajectory deviation alarm information and correction suggestions includes: The lateral deviation value and the angular deviation value are compared with preset multi-level alarm thresholds to determine the alarm level; Select the corresponding alarm method according to the alarm level. The alarm method includes color change of visual cues, graphic flashing, and frequency and pitch change of sound cues. Based on the vector synthesis result of the lateral deviation value and the angular deviation value, a theoretical correction motion vector is calculated; The correction motion vector is converted into guidance instructions for the operator. These instructions are displayed as graphic arrows overlaid on the real-time image or expressed through voice prompts, forming the correction suggestions. The correction suggestions include the direction in which the needle insertion angle should be adjusted and the approximate adjustment range.

[0010] As a further aspect of the present invention, the triggered biopsy sampling device performs tissue sampling operations and continuously monitors real-time image changes in the sampling area during the sampling process, including: When the system detects that the spatial distance between the tip coordinates of the actual needle trajectory and the target position coordinates is less than the trigger threshold, it automatically or after waiting for user confirmation sends a firing command to the biopsy sampling device. The biopsy sampling device, upon firing command, drives the internal cutting cannula and sampling needle to perform rapid forward, cutting, and retraction actions to obtain cylindrical tissue samples. The ultrasound image sequence and the color Doppler blood flow signal were continuously acquired and analyzed before, during and shortly after the firing action; Monitor the echo changes, morphological displacement, and presence of new blood flow signals in the target sampling area and surrounding tissues to assess whether the sampling procedure has caused unexpected bleeding or tissue damage.

[0011] As a further aspect of the present invention, the step of extracting the morphological boundary information of lymph nodes from the ultrasound image sequence and constructing an initial geometric model of the lymph nodes through edge detection and contour tracking algorithms includes: Each frame of the ultrasound image sequence is preprocessed, including noise reduction filtering and contrast enhancement. On the preprocessed image, the active contour model algorithm is applied to initialize a closed curve that encloses the lymph node region; Drive the closed curve to evolve toward the true boundary of the lymph node under the combined action of the image gradient field and internal constraint force; When the evolution of the closed curve tends to stabilize, that is, when the energy function of the curve reaches its minimum value, the shape of the curve at this time is locked. The locked closed curve is tracked and reconstructed in three dimensions in consecutive frame images to obtain a continuous three-dimensional point cloud or mesh model representing the outer surface of the lymph node, which serves as the initial geometric model.

[0012] As a further aspect of the present invention, the monitoring of echo changes, morphological displacement, and the presence of new blood flow signals in the target sampling area and its surrounding tissues to assess whether the sampling operation has caused unexpected bleeding or tissue damage includes: Before firing, the baseline echo characteristics and baseline blood flow distribution map of the target sampling area are recorded; After the firing action is executed, images are continuously acquired within a preset time period, and the target sampling area and the area along the needle path are analyzed to see if new anechoic or hypoechoic areas appear. The new anechoic or hypoechoic areas represent fresh hematomas. Simultaneously, the color Doppler blood flow signal is monitored to check whether abnormal punctate or patchy blood flow signals appear outside the original vascular structure. The abnormal punctate or patchy blood flow signals indicate active bleeding. Based on the occurrence of the new anechoic or hypoechoic areas and the occurrence of the abnormal punctate or patchy blood flow signals, a sampling safety assessment report is generated.

[0013] As a further aspect of the present invention, the method further includes a sample quality pre-assessment step: Immediately after the biopsy sampling device completes the tissue sampling operation and is withdrawn from the body, in-situ optical imaging is performed on the obtained tissue sample. The in-situ optical imaging is used to preliminarily analyze the macroscopic morphology, color, length and diameter of the tissue sample, and to quickly compare it with a pre-stored database of normal lymph node tissue samples. Based on the comparison results, a representativeness score for the sample is calculated, which is based on the sample's completeness, size, and whether it contains characteristic structures of lymph nodes. If the representative score is lower than the qualified threshold, a prompt message suggesting resampling is generated, and the prompt message is recorded in the case log along with the current operating parameters.

[0014] As a further aspect of the present invention, the present invention also includes a monitoring system for lymph node biopsy puncture, the system including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements the steps of the sampling method for lymph node biopsy puncture as described above.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: By fusing ultrasound image sequences, elastography data, and color Doppler blood flow signals in real time, a lymph node structural feature map integrating morphology, stiffness, and blood flow information is generated. This technology directly visualizes anatomical structures, tissue mechanical properties, and vascular distribution in the same coordinate system. Operators can obtain a unified and accurate spatial understanding of lymph node boundaries, internal stiffness distribution, and the three-dimensional course of blood vessels without switching between multiple independent image windows or performing mental synthesis. This reduces the information integration burden on physicians, making target identification and the determination of their spatial relative positions to key blood vessels direct and accurate. It provides a reliable and intuitive data foundation for subsequently planning a puncture path that effectively avoids blood vessels and precisely reaches the target area.

[0016] The system tracks the actual needle insertion trajectory of the puncture instrument in real time and dynamically compares this trajectory data with the preset safe puncture path, automatically generating deviation alarms and correction suggestions. This process transforms the traditional "open-loop" operation, which mainly relies on the doctor's experience and hand-eye coordination, into a "closed-loop" process of "intelligent navigation" with real-time monitoring and feedback from the system. The system can continuously quantify the degree of deviation between the actual trajectory and the ideal path, and promptly issue warnings when clinically significant deviations occur, while providing specific spatial adjustment suggestions. This allows the operator to proactively and promptly correct deviations during the puncture process, reducing the risk of path deviation caused by factors such as tissue displacement and unstable operation, ensuring that the puncture needle accurately reaches the target point along the predetermined safe path, and improving the controllability, accuracy, and safety of the entire operation. Attached Figure Description

[0017] Figure 1 This is a flowchart of the lymph node biopsy puncture sampling method described in this invention; Figure 2 A flowchart for determining a safe puncture path and target location; Figure 3 A monitoring graph for deviations during lymph node biopsy puncture; Figure 4 A bar chart showing the echo intensity analysis after lymph node biopsy. Figure 5 A heatmap showing the similarity score of the characteristic structural texture of lymph node biopsy samples. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1The overall implementation scheme of the monitoring system and biopsy sampling method for lymph node biopsy puncture of the present invention includes the following steps: scanning the target lymph node area of ​​the patient with an ultrasound probe, and simultaneously acquiring a real-time multimodal image data stream including ultrasound image sequences, elastography data, and color Doppler blood flow signals. This data stream is transmitted to an image processing unit. The image processing unit performs fusion processing on the received real-time multimodal image data stream to generate a lymph node structural feature map integrating multiple information, including the contour of the lymph node boundary, the hardness distribution area of ​​the internal tissue, and the spatial course of the main blood vessels. The planning module calculates based on the generated lymph node structural feature map to determine a safe puncture path that avoids the spatial course of the main blood vessels and traverses the target sampling area within the hardness distribution area of ​​the internal tissue, while simultaneously identifying the target point location. During the puncture, the spatial positioning and navigation unit tracks the spatial coordinates and attitude angle of the positioning probe attached to the puncture instrument in real time, thereby calculating the actual needle insertion trajectory of the puncture instrument. The actual needle insertion trajectory is overlaid on the lymph node structural feature map in real time and dynamically compared with the pre-planned safe puncture path. The system generates trajectory deviation alarm information and correction suggestions based on the deviations identified during the comparison. When the system determines that the tip of the actual needle insertion trajectory has reached the target location, it sends a trigger command to the biopsy sampling device, driving it to complete the tissue sampling operation. Throughout the entire sampling process, the system continuously acquires and analyzes real-time image changes in the target area to monitor the operation.

[0021] In one embodiment of the present invention, the generation process of a lymph node structural feature map is described, specifically including morphological boundary extraction and multimodal data fusion. Morphological boundary information of the lymph nodes is extracted from an ultrasound image sequence, and an initial geometric model of the lymph nodes is constructed using edge detection and contour tracking algorithms. Specifically, each frame of the ultrasound image sequence is preprocessed, including noise reduction filtering and contrast enhancement. On the preprocessed image, an active contour model algorithm is applied to initialize a closed curve enclosing the lymph node region. This closed curve is driven to evolve towards the true boundary of the lymph node under the combined action of the image gradient field and internal constraints. When the evolution of the closed curve tends to stabilize, i.e., when the energy function of the curve reaches its minimum value, the curve morphology is locked. The locked closed curve is tracked and 3D reconstructed in consecutive frame images to obtain a continuous 3D point cloud or mesh model representing the outer surface of the lymph node, which serves as the initial geometric model. In the tracking of the locked closed curve across consecutive frames, a deep learning-based optical flow estimation algorithm predicts the motion vectors of lymph node boundary points between adjacent frames. This optical flow estimation network employs an encoder-decoder structure. The encoder extracts image features using convolutional layers, while the decoder progressively upsamples through deconvolutional layers and fuses shallow features with skip connections to output a dense optical flow field. The network's training data consists of numerous ultrasound image sequence pairs labeled with the correspondence of lymph node boundary points. The loss function is the sum of the Euclidean distance between the predicted optical flow and the actual displacement vector, plus a smoothing constraint term. Each boundary point on the closed curve in the current frame, whose displacement vector is calculated based on the optical flow field, is mapped to its corresponding position in the next frame, achieving frame-by-frame transmission and position update of the closed curve within the sequence. After tracking the closed curve across consecutive frames, 3D reconstruction is performed to generate an initial geometric model. The system transforms the boundary points contained in the locked closed curve in each frame from 2D image coordinates to a unified 3D spatial coordinate system. This transformation is based on the spatial positioning parameters of the ultrasound probe and the geometric calibration parameters of the image acquisition. The 3D boundary points from all frames are fused to form a dense 3D point cloud representing the outer surface of the lymph node. When converting the 3D point cloud into a continuous mesh model, a Poisson surface reconstruction algorithm is applied. This algorithm fits an implicit surface by calculating the gradient field of the indicator function of the point cloud and solving the Poisson equation. An octree depth parameter is set to control the level of detail in the reconstruction. Then, the isosurfaces of the implicit surface are extracted to generate a triangular mesh model. The generated triangular mesh model may contain noise or non-manifold structures, which are optimized through subsequent processing. A Laplacian smoothing algorithm is applied to iteratively adjust the vertex positions of the mesh to reduce surface noise. Simultaneously, a side-length-based mesh simplification algorithm is used to reduce the number of triangles while maintaining the overall shape, improving the efficiency of subsequent processing. Finally, an optimized, continuous 3D mesh model of the lymph node's outer surface is obtained as the initial geometric model.

[0022] Elastography data is analyzed, and the stiffness values ​​of different regions are mapped to pseudo-color or grayscale levels, which are then superimposed onto the initial geometric model to form a lymph node enhancement model containing stiffness distribution. Simultaneously, color Doppler blood flow signals are analyzed to identify and locate vascular structures within and around the lymph node, and the spatial paths and diameters of major blood vessels are marked in the lymph node enhancement model. Finally, morphological boundary information, stiffness distribution, and the spatial paths of major blood vessels are three-dimensionally registered and fused in a unified spatial coordinate system to output a lymph node structural feature map with multi-dimensional structural attributes. In specific implementation, after acquiring real-time multimodal image data streams of the lymph node region, the real-time multimodal image data streams are fused to generate the lymph node structural feature map; this process involves multiple image processing and data integration steps. Morphological boundary information of lymph nodes is extracted from ultrasound image sequences. An initial geometric model of the lymph nodes is constructed using edge detection and contour tracking algorithms. This process begins with preprocessing each frame of the ultrasound image sequence. Preprocessing operations include noise reduction filtering and contrast enhancement to improve image quality for subsequent analysis. On the preprocessed image, an active contour model algorithm is applied to initialize a closed curve enclosing the lymph node region. This closed curve is driven to evolve towards the true boundary of the lymph node under the combined influence of the image gradient field and internal constraints. The evolution of the closed curve is driven by minimizing an energy function. When the evolution of the closed curve tends to stabilize, i.e., when the energy function of the curve reaches its minimum value, the curve morphology at this point is locked. The expression is as follows: in: Represents a parameterized closed curve, parameter It is the arc length. and These are the weighting coefficients that control the elasticity and stiffness of the curve. and These represent the first and second derivatives of the curve, respectively. It is an external energy field originating from the image gradient field. The locked closed curve is tracked and 3D reconstructed in consecutive frame images. Through inter-frame registration and point cloud fusion algorithms, a continuous 3D point cloud or mesh model representing the outer surface of the lymph node is obtained. This model serves as the initial geometric model.

[0023] In some embodiments, elastography data is parsed and the hardness values ​​of different regions are mapped to pseudo-color or grayscale levels. This mapping process is completed according to a preset hardness-color reference table. The mapped pseudo-color or grayscale data is then superimposed onto the initial geometric model as a texture map, thereby forming a lymph node enhancement model containing hardness distribution. Simultaneously, color Doppler blood flow signals are analyzed to identify and locate the vascular structures inside and around the lymph node. The identification process is based on blood flow signal intensity and continuity analysis. In the lymph node enhancement model, the spatial paths of major blood vessels are marked using three-dimensional spatial curves, and the blood vessel diameter information estimated from the signals is associated and stored. It can be understood that the final step is to perform three-dimensional registration and fusion of morphological boundary information, hardness distribution, and the spatial paths of major blood vessels in a unified spatial coordinate system. Registration is achieved by extracting feature points from each modality of data and calculating the spatial transformation matrix. After fusion, a lymph node structural feature map with multi-dimensional structural attributes is output, which is presented in a three-dimensional visualization form. Optionally, the initialization of the active contour model algorithm can be done manually by setting seed points so that the system can automatically generate closed curves, or by having the operator sketch a rough outline on a single frame image and then refine it using the algorithm. In practice, optical flow or feature point matching methods are used to track the locking curve in consecutive frame images to ensure the continuity and accuracy of the 3D reconstruction. When analyzing elastography data, the quantification of hardness values ​​needs to be calibrated using the system's built-in standard module before each examination. During the analysis of color Doppler blood flow signals, adaptive filters are used for filtering and noise suppression of the blood flow signals. The unified spatial coordinate system used for 3D registration and fusion is established based on the spatial positioning of the ultrasound probe. It can be understood that the final output lymph node structure feature map supports multi-planar reconstruction display and rotational viewing; the hardness distribution area is displayed as a semi-transparent overlay layer inside the lymph node model, and the spatial course of the main blood vessels is embedded within it as a bright colored pipeline model. Optionally, all processing steps are accelerated by dedicated parallel computing hardware in the image processing unit to ensure real-time performance. In practice, the generated lymph node structure feature map serves as the basic data layer, which is then used and displayed by the subsequent path planning module.

[0024] See Figure 2In one embodiment of the present invention, a target sampling area is taken as the center in the lymph node structural feature map. The target sampling area is determined based on a preset hardness threshold range in the hardness distribution area or by manual annotation by the user through an interactive interface. Starting from this target sampling area, a spatial straight line or curve is planned backward to reach the preset puncture point on the skin. Then, virtual detection is performed along this spatial straight line or curve to check whether it intersects with the spatial path of major blood vessels and whether it passes through a specific area inside the lymph node that would cause excessive damage. If there is an intersection or passage through a specific area, the angle and depth of the spatial straight line or curve are dynamically adjusted until a path that meets the preset safety distance requirement and has the shortest path length is found, which is then determined as the safe puncture path. Finally, the geometric center point of the target sampling area or the feature point specified by the user is determined as the target point location. In specific implementation, the process of calculating and determining the safe puncture path and target point location based on the lymph node structural feature map is executed by the planning module to generate geometric data to guide the puncture. The planning module calculates based on the target sampling area in the lymph node structural feature map. The target sampling area is determined by a preset hardness threshold range in the hardness distribution area or by manual annotation by the user through the interactive interface. When determined by the hardness threshold range, the system automatically identifies and delineates a continuous spatial region with hardness values ​​within the preset range as the target sampling area. Starting from the target sampling area, a spatial straight line or curve is planned backward to reach the preset skin puncture point. Backward planning uses a ray detection method from the geometric center of the target sampling area to the preset skin puncture point to initially generate a candidate path. Virtual detection is performed along the spatial straight line or curve. Virtual detection is achieved by calculating the minimum distance between the path segment and the 3D model of the spatial course of the main blood vessels. It checks whether the candidate path intersects with the spatial course of the main blood vessels and whether it passes through specific areas inside the lymph node that would cause excessive damage. Specific areas include the central area of ​​the lymph node hilum and non-target tissues pre-marked by the user.

[0025] In practice, if virtual detection detects intersections or crossings of specific areas, the angle and depth of the spatial straight line or curve are dynamically adjusted. This adjustment process follows an iterative optimization algorithm, generating new candidate paths and re-performing virtual detection by changing the tangential angle or curvature of the path at the skin entry point. This continues until a path that meets the preset safety distance requirement and has the shortest length is found and designated as the safe puncture path. The preset safety distance requirement refers to the minimum distance between the path and the surface of any major vascular spatial path model being greater than a set value. The path length is calculated from the three-dimensional coordinates of the path control points. The geometric center point of the target sampling area or a user-specified feature point is determined as the target location. The target location is stored in three-dimensional coordinates and associated with the endpoint of the safe puncture path. In some embodiments, the path optimization process is completed by solving a constrained optimization problem, with the objective function... Defined as path length With penalty items Weighted sum: in: Represents a series of discrete points Defined path, It is the total length of the path. It is a penalty function for violations of the distance between the path and all obstacles (including blood vessels and specific regions). and This is a weighting coefficient that determines the path length and safety priority level. The process of determining a safe puncture path is interactive; users can manually fine-tune the automatically planned safe puncture path, after which the system immediately re-executes virtual detection and safety verification. Optionally, preset skin puncture points can be automatically recommended by the system within a specific skin area based on the shortest path principle, or the operator can directly select them on the 3D body surface model. In some embodiments, the path that meets the preset safety distance requirements and has the shortest path length may not be unique; the planning module will provide multiple suitable paths for the operator to choose from based on clinical preferences. The safe puncture path and target location calculated by the planning module will be displayed superimposed on the lymph node structural feature map as a highlighted 3D dashed line trajectory and a spherical marker. In specific implementations, when virtual detection checks whether it has passed through a specific area inside the lymph node that has caused excessive damage, the definition of the specific area can be extended to include the low-hardness annular band surrounding the high-hardness core area. It's understandable that algorithms that dynamically adjust the angle and depth of straight lines or curves in space will prompt the user to redefine the target sampling area or the preset skin puncture point if the number of adjustments exceeds a set limit and a safe path is still not found. Optionally, the priority of the shortest path length can be configured. In complex vascular environments, the system may prioritize finding an absolutely safe path rather than the absolutely shortest path.

[0026] In one embodiment of the present invention, the process of real-time needle insertion trajectory tracking and dynamic comparison and correction is described. The spatial coordinates and attitude angles of a positioning probe mounted on the puncture instrument are acquired in real time using an electromagnetic positioning sensor or an optical navigation system. Based on the spatial coordinates and attitude angles of the positioning probe, combined with the known physical dimensions of the puncture instrument, the real-time spatial position of the instrument tip and the direction of the instrument's long axis are calculated to form the actual needle insertion trajectory. This actual needle insertion trajectory is drawn and updated in real time within the same spatial coordinate system as the lymph node structural feature map. The spatial distance between the tip of the actual needle insertion trajectory and the corresponding depth point on the safe puncture path is calculated as the lateral deviation value; the angle between the direction of the long axis of the actual needle insertion trajectory and the direction of the safe puncture path is calculated as the angular deviation value. When the lateral deviation value or the angular deviation value exceeds its respective set allowable threshold, a trajectory deviation alarm message and correction suggestion containing the deviation value and deviation direction are immediately generated. When generating the trajectory deviation alarm message and correction suggestion, the lateral deviation value and the angular deviation value are compared with preset multi-level alarm thresholds to determine the alarm level. The appropriate alarm method is selected based on the alarm level. Alarm methods include color changes and graphic flashing for visual cues, and frequency and tone changes for audio cues. A theoretical correction motion vector is calculated based on the vector synthesis results of lateral and angular deviation values. This correction motion vector is then converted into guidance instructions for the operator. These instructions are displayed as graphic arrows overlaid on the real-time image or expressed via voice prompts, forming correction suggestions. These suggestions include the direction in which the needle insertion angle should be adjusted and the approximate adjustment range. In practice, the actual needle insertion trajectory of the puncture instrument is tracked in real time, and the actual trajectory is dynamically compared with the safe puncture path. This function is achieved collaboratively by the spatial positioning and navigation unit and the display control unit. The spatial coordinates and attitude angles of the positioning probe mounted on the puncture instrument are acquired in real time using an electromagnetic positioning sensor or optical navigation system. The positioning probe is rigidly connected to the handle of the puncture instrument, and its spatial coordinates and attitude angle data are transmitted to the processing core at a high frequency (e.g., tens of times per second) through a data interface. Based on the spatial coordinates and attitude angle of the positioning probe, combined with the known physical dimensions of the puncture instrument (including the length of the puncture needle and the fixed offset of the probe relative to the needle tip), the real-time spatial position of the tip of the puncture instrument and the direction of the long axis of the instrument are calculated through spatial coordinate system transformation, forming the actual needle insertion trajectory. In the same spatial coordinate system where the lymph node structure feature map is located, the actual needle insertion trajectory is drawn and updated in real time. The actual needle insertion trajectory is usually displayed on the navigation interface as a dynamically extending colored line superimposed on it.

[0027] In practice, the spatial distance between the tip of the actual needle insertion trajectory and the corresponding depth point on the safe puncture path is calculated. The corresponding depth point is defined as a point on the safe puncture path whose distance from the skin entry point is the same as the distance from the current tip of the actual needle insertion trajectory to the skin entry point. This spatial distance is used as the lateral deviation value. The angle between the major axis of the actual needle insertion trajectory and the direction of the safe puncture path is calculated. This angle is obtained by calculating the angle between two spatial vectors and is used as the angular deviation value. When the lateral deviation value or the angular deviation value exceeds its respective set allowable threshold, a trajectory deviation alarm message and correction suggestion containing the deviation value and deviation direction are immediately generated. It can be understood that generating trajectory deviation alarm message and correction suggestion includes subsequent processing steps, comparing the lateral deviation value and the angular deviation value with preset multi-level alarm thresholds to determine the alarm level, such as setting a yellow warning level and a red alarm level. The corresponding alarm method is selected according to the alarm level. Alarm methods include color changes for visual cues (such as the trajectory line changing from green to yellow and then to red), graphic flashing (such as the target icon flashing), and frequency and tone changes for audio cues. A theoretical correction motion vector is calculated based on the vector synthesis result of the lateral deviation and angular deviation values. The formula for its calculation is: in: This represents the lateral deviation vector from the tip of the actual needle insertion trajectory to the corresponding depth point on the safe puncture path. Represents the angular deviation value (in radians). It is a reference unit vector perpendicular to the current needle insertion direction and pointing towards the safe puncture path. and These are weighted coefficients that control the intensity of translational and angular corrections. The correction motion vector is converted into guiding instructions for the operator, which are displayed as graphic arrows overlaid on the real-time image or expressed through voice prompts, forming correction suggestions. The correction suggestions include the direction in which the needle insertion angle should be adjusted and the approximate adjustment range. For example, a graphic arrow indicates a slight adjustment to the left of the patient, with a note "2 mm to the left".

[0028] In specific implementations, the spatial positioning and navigation unit continuously displays the calculated lateral and angular deviation values ​​in numerical form on a fixed area of ​​the screen. It is understood that thresholds can be preset according to clinical operational precision requirements and can be differentiated at different surgical stages. Optionally, the dynamic comparison process not only compares the tip position but also simulates and calculates the expected deviation between the path and the safe puncture path over a predetermined distance if the needle continues to be inserted at the current angle. In some embodiments, there is a brief delay in the generation of trajectory deviation alarm information and correction suggestions, the delay time of which is determined by the system data processing and refresh cycle. Optionally, the content of the audio prompts can be selected by the user from pre-recorded voice phrases or specific prompt sound sequences. In specific implementations, the calculation of the correction motion vector Φ considers the interaction model between the instrument and tissue to avoid excessively large suggested corrections that could lead to tissue tearing.

[0029] See Figure 3 This is a deviation monitoring chart for lymph node biopsy puncture. Its core function is to track lateral and angular deviations during needle insertion in real time, aiding in clinical safety control. As needle depth increases, changes in tissue resistance and slight instrument movement can easily lead to accumulated deviations, consistent with clinical practice. Within the 40–60 mm depth range, angular deviations approach the threshold, requiring a visual / audible alarm to prompt the operator to adjust the needle angle. When the lateral deviation approaches 3.0 mm, the system should output a correction instruction to "adjust 2–3 mm to the left" to prevent the needle tip from deviating from the safe path. This chart is the core output module of the puncture navigation system. By quantifying the dynamics of deviation, it helps the operator monitor needle insertion accuracy in real time, reducing the risk of complications such as vascular injury and sampling failure. Its dual-dimensional deviation monitoring design covers the two most critical accuracy indicators in clinical operation, providing data support for surgical safety.

[0030] In one embodiment of the present invention, the specific implementation of trigger sampling and sampling process monitoring is described. When the system detects that the spatial distance between the tip coordinates of the actual needle trajectory and the target position coordinates is less than the trigger threshold, it automatically or after waiting for user confirmation sends a firing command to the biopsy sampling device. According to the firing command, the biopsy sampling device drives the internal cutting cannula and sampling needle to complete rapid forward, cutting, and retraction movements to obtain a cylindrical tissue sample. Before, during, and briefly after the firing action, ultrasound image sequences and color Doppler blood flow signals are continuously acquired and analyzed. The echo changes, morphological displacement, and the appearance of new blood flow signals in the target sampling area and its surrounding tissues are monitored to assess whether the sampling operation has caused unexpected bleeding or tissue damage. Specifically, before the firing action, the baseline echo characteristics and baseline blood flow distribution map of the target sampling area are recorded. After the firing action, images are continuously acquired within a preset time period to analyze whether new anechoic or hypoechoic areas appear in the target sampling area and along the needle path; new anechoic or hypoechoic areas represent fresh hematoma. Simultaneously, color Doppler blood flow signals are monitored to check for any abnormal punctate or patchy blood flow signals appearing outside the original vascular structure. Abnormal punctate or patchy blood flow signals indicate active bleeding. A sampling safety assessment report is generated based on the combined findings of the appearance of new anechoic or hypoechoic areas and the presence of abnormal punctate or patchy blood flow signals.

[0031] In practice, the biopsy sampling device is triggered to perform tissue sampling and continuously monitors real-time image changes in the sampling area during the sampling process. This series of actions is coordinated and executed by the system control unit. When the system detects that the spatial distance between the tip coordinates of the actual needle trajectory and the target location coordinates is less than the trigger threshold, it automatically or after waiting for user confirmation sends a firing command to the biopsy sampling device. The trigger threshold is a preset small distance value, such as 1.5 mm. According to the firing command, the biopsy sampling device drives the internal cutting cannula and sampling needle to complete rapid forward, cutting, and retraction movements to obtain a cylindrical tissue sample. This mechanical action is completed within hundreds of milliseconds. Before, during, and briefly after the firing action, the system control unit instructs the image acquisition module to continuously acquire and analyze ultrasound image sequences and color Doppler blood flow signals. The acquisition process is performed at a high frame rate to ensure that no instantaneous changes are missed. The system monitors echogenic changes, morphological displacement, and the presence of new blood flow signals in the target sampling area and surrounding tissues to assess whether the sampling procedure has caused unexpected bleeding or tissue damage. The monitoring area includes a spherical space centered on the target location with a predetermined radius. The safety assessment of the sampling procedure is achieved through the following specific steps: Before firing, the baseline echogenicity and baseline blood flow distribution map of the target sampling area are recorded. The baseline echogenicity is derived by analyzing the average grayscale value of several ultrasound images immediately preceding firing. The baseline blood flow distribution map records the intensity and distribution coordinates of the color Doppler blood flow signal within the area. After firing, images are continuously acquired over a preset time period. The system analyzes whether new anechoic or hypoechoic areas appear in the target sampling area and along the needle path. New anechoic or hypoechoic areas represent fresh hematoma, and the analysis is achieved by comparing the pixel grayscale differences between the post-firing images and the baseline images. Simultaneously, color Doppler blood flow signals are monitored to check for abnormal punctate or patchy blood flow signals outside the original vascular structure. Abnormal punctate or patchy blood flow signals indicate active bleeding, which is determined by detecting the presence of newly added color pixel clusters discontinuous with the baseline vascular distribution in the post-firing blood flow signal image. Based on the combined findings of new anechoic or hypoechoic areas and abnormal punctate or patchy blood flow signals, a sampling safety assessment report is generated. The report includes the location, size, blood flow signal intensity, and trend of the abnormal area. The detection of new anechoic or hypoechoic areas uses an image difference algorithm, expressed by the formula: in: Image pixel coordinates, Time after firing The pixel grayscale value at time 10:00. This represents the pixel grayscale value at the corresponding location in the baseline image. This represents the grayscale difference value. When a certain region differs across multiple consecutive frames... If the threshold is exceeded, the area is determined to be a new area of ​​echo change. It is understood that the preset time period can be set according to clinical needs, typically covering several seconds to tens of seconds after firing. Abnormal information identified during monitoring is marked on the image screen in real time and alerted to the operator with different alarm levels. The alarm level is related to the size of the abnormal area and the intensity of the blood flow signal, as shown in Table 1 below.

[0032] Table 1: Example of the correspondence between anomaly detection and alarm level after sampling Optionally, the recording of baseline echo characteristics and baseline blood flow distribution maps can be manually triggered or automatically performed by the system when it detects that the puncture needle is stable at the target site. In some embodiments, the sampling safety assessment report is not only displayed in real time but also automatically archived into the patient's surgical record. It is understood that the continuous monitoring process will only stop after the user manually disables it or the system determines that the risk has been eliminated. In specific implementations, the three-dimensional spatial coordinates of the detected abnormal areas are recorded and can be projected back onto the initial lymph node structural feature map to update the anatomical risk labeling of that area.

[0033] See Figure 4 This is a bar chart analyzing the echogenicity of lymph node biopsy puncture sites. It assesses changes in tissue echogenicity in the sampling area and helps determine whether the sampling procedure caused bleeding or tissue damage. The highest total echogenicity is observed at the target site center and lowest in areas further away, indicating that the sampling procedure has the most significant impact on the echogenicity of the target site and surrounding tissues. The difference in echogenicity decreases rapidly with increasing distance from the target site, suggesting that the impact of the sampling procedure is mainly concentrated within a 2mm radius of the target site, consistent with the local effects of clinical biopsies. The significant difference in echogenicity at the target site center suggests that local tissue may experience slight edema or minor bleeding due to sampling, but no large anechoic area was observed, which is a normal postoperative reaction. This chart is a core basis for assessing the quality and safety of biopsy sampling. By quantifying the echogenicity changes in different areas, the accuracy of the sampling procedure and the risk of potential complications can be objectively assessed, providing data support for clinical decision-making.

[0034] In one embodiment of the present invention, immediately after the biopsy sampling device completes the tissue sampling operation and withdraws from the body, in-situ optical imaging is performed on the acquired tissue sample. This in-situ optical imaging allows for preliminary analysis of the macroscopic morphology, color, length, and diameter of the tissue sample, and rapid comparison with a pre-stored database of normal lymph node tissue samples. Based on the comparison results, a representativeness score is calculated for the sample, based on its integrity, size, and whether it contains characteristic lymph node structures. If the representativeness score is below a passing threshold, a suggestion to resample is generated, and this suggestion, along with the current operating parameters, is recorded in the case log.

[0035] In practice, the sample quality pre-assessment step is initiated immediately after the biopsy sampling device completes the tissue sampling operation and is withdrawn from the body. This step is completed collaboratively by a miniature optical imaging module integrated near the sampling device and a quality analysis module. Immediately after the biopsy sampling device completes the tissue sampling operation and is withdrawn from the body, in-situ optical imaging is performed on the acquired tissue sample. In-situ optical imaging is performed while the tissue sample is still in a sterile sampling tank or a specific sample-carrying platform. The imaging module automatically focuses and acquires digital images of the tissue sample from multiple angles. Through in-situ optical imaging, the macroscopic morphology, color, length, and diameter of the tissue sample are preliminarily analyzed. Macroscopic morphology analysis includes assessing whether the sample is a continuous cylinder and whether there is obvious breakage or crushing. Color analysis compares the RGB color space values ​​of the image with the preset normal tissue color range. Length and diameter are calculated using a scale reference in the image. A rapid comparison is then performed with a pre-stored database of normal lymph node tissue samples. This database contains a large number of pathologically confirmed normal lymph node tissues with morphological, color, and size parameters under the same imaging conditions.

[0036] In practice, a representativeness score is calculated based on the comparison results. This score is based on the sample's integrity, size, and whether it contains characteristic lymph node structures. The representativeness score is calculated using a quantitative formula. The integrity score is calculated based on image analysis to determine the continuity and surface smoothness of the tissue sample. The size fit score is calculated based on whether the sample length and diameter conform to a preset acceptable sample size range. The characteristic structure score is calculated based on image texture analysis to detect the presence of macroscopic or microscopic texture signs of characteristic structures such as lymph node cortex, medulla, or lymphoid follicles. These three scores are then weighted and summed to calculate the representativeness score Γ, using the following formula: in, Represents a normalized score based on integrity. Represents a size-fitted normalized score. Represents the normalized score based on feature structure recognition. , , These are preset weighting coefficients that assign importance to integrity, size fit, and characteristic structure, and satisfy... If the representativeness score is below the acceptable threshold, a suggestion to resample is generated and recorded in the case log along with the current operating parameters. The acceptable threshold is a pre-defined value that has been clinically validated. A pre-stored database of normal lymph node tissue samples is categorized by lymph node location (e.g., neck, axilla, groin) and patient age group. During comparison, a subset of the database from the same location and age group is prioritized. It is understood that the illumination conditions and focal length of in-situ optical imaging are standardized to eliminate the influence of imaging variations on the analysis results. Optionally, the representativeness score calculation result, along with a macroscopic image of the sample, is displayed in real-time in a specific area of ​​the operating interface, with acceptable and unacceptable samples distinguished by different colored borders. In some embodiments, the determination of whether the sample contains characteristic lymph node structures can be aided by image texture analysis, combined with cross-sectional information obtained from a very brief low-power optical coherence tomography scan. It is understood that the acceptable threshold can be slightly adjusted by the user according to different clinical diagnostic needs (e.g., suspected lymphoma versus suspected metastatic cancer).

[0037] In practice, if the representativeness score falls below the acceptable threshold, the generated suggestion for resampling includes both textual explanation and visual warnings. For example, it may flash "Insufficient sample representativeness, resampling is recommended" on the screen and highlight the relevant score item. It is understood that the current operating parameters recorded in the case log include, but are not limited to, the target location coordinates, puncture path, sampling device model, and detailed component values ​​of the calculated representativeness score Γ. Optionally, the miniature optical imaging module has an automatic cleaning function, cleaning the lens after each imaging session to prevent sample residue contamination from affecting subsequent imaging analysis.

[0038] See Figure 5 This is a heatmap showing the similarity of the characteristic structure texture of a lymph node biopsy sample. It's used to assess whether a sample contains the core characteristic structures of a lymph node through texture analysis using in-situ optical imaging, a crucial step in sample quality pre-assessment. The consistent trend in the three types of characteristic structure scores across all samples indicates the good stability of this texture analysis method, avoiding misjudgments caused by single-structure identification bias. The heatmap visually distinguishes between qualified and unqualified samples, preventing invalid samples from being transferred before pathological testing and improving diagnostic efficiency. The completeness of the characteristic structure in high-scoring samples provides a more reliable histological basis for pathological diagnosis, reducing the risk of missed diagnoses. This heatmap is a product of the combination of in-situ optical imaging and AI texture analysis technology. By quantifying visual features, it transforms the subjective "sample representativeness" into an objective similarity score, providing a quantifiable basis for clinical decision-making and serving as a core tool for intelligent quality control in lymph node biopsies.

[0039] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for sampling lymph node biopsy, characterized in that, Includes the following steps: Acquire real-time multimodal imaging data stream of lymph node region, wherein the real-time multimodal imaging data stream includes ultrasound image sequence, elastography data and color Doppler blood flow signal; The real-time multimodal image data stream is fused to generate a lymph node structural feature map, which includes the contour of the lymph node boundary, the hardness distribution area of ​​the internal tissue, and the spatial course of the main blood vessels. Based on the lymph node structural feature map, the safe puncture path and target location are calculated and determined. The safe puncture path avoids the spatial course of the main blood vessels and passes through the target sampling area in the hardness distribution region of the internal tissue. The actual needle insertion trajectory of the puncture instrument is tracked in real time, and the actual needle insertion trajectory is dynamically compared with the safe puncture path to generate trajectory deviation alarm information and correction suggestions; When the tip of the actual needle trajectory reaches the target location, the biopsy sampling device is triggered to perform tissue sampling, and real-time image changes in the sampling area are continuously monitored during the sampling process.

2. The sampling method for lymph node biopsy puncture according to claim 1, characterized in that, The process of fusing the real-time multimodal image data stream to generate a lymph node structural feature map includes: Morphological boundary information of lymph nodes is extracted from the ultrasound image sequence, and an initial geometric model of the lymph nodes is constructed using edge detection and contour tracking algorithms; The elastic imaging data is analyzed, and the hardness values ​​of different regions are mapped to pseudo-color or grayscale levels, which are then superimposed onto the initial geometric model to form a lymph node enhancement model that includes hardness distribution. The color Doppler blood flow signal is analyzed to identify and locate the vascular structures inside and around the lymph node, and the spatial course and diameter information of the main blood vessels are marked in the lymph node enhancement model. The morphological boundary information, the hardness distribution, and the spatial course of the main blood vessels are registered and fused in three dimensions in a unified spatial coordinate system to output a lymph node structural feature map with multi-dimensional structural attributes.

3. The sampling method for lymph node biopsy puncture according to claim 2, characterized in that, The step of calculating and determining the safe puncture path and target location based on the lymph node structural feature map includes: Centered on the target sampling area in the lymph node structural feature map, the target sampling area is determined according to the preset hardness threshold range in the hardness distribution area or by manual annotation by the user. Starting from the target sampling area, a spatial straight line or curve is planned in reverse to reach the preset puncture point on the skin; Virtual detection is performed along the described straight line or curve to check whether it intersects with the spatial course of the main blood vessels and whether it passes through specific areas inside the lymph nodes that cause excessive damage. If there is an intersection or passage through a specific area, the angle and depth of the straight line or curve in the space are dynamically adjusted until a path that meets the preset safety distance requirement and has the shortest path length is found, and this path is determined as the safe puncture path. The geometric center point of the target sampling area or the feature point specified by the user is determined as the target point location.

4. The sampling method for lymph node biopsy puncture according to claim 1, characterized in that, The real-time tracking of the actual needle insertion trajectory of the puncture instrument, and the dynamic comparison of the actual needle insertion trajectory with the safe puncture path, includes: The spatial coordinates and attitude angles of the positioning probe installed on the puncture instrument are obtained in real time through electromagnetic positioning sensors or optical navigation systems. Based on the spatial coordinates and attitude angle of the positioning probe, combined with the known physical dimensions of the puncture instrument, the real-time spatial position of the tip of the puncture instrument and the direction of the long axis of the instrument are calculated to form the actual needle insertion trajectory. In the same spatial coordinate system where the lymph node structure feature map is located, the actual needle insertion trajectory is drawn and updated in real time. Calculate the spatial distance between the tip of the actual needle insertion trajectory and the corresponding depth point on the safe puncture path, as the lateral deviation value; calculate the angle between the major axis direction of the actual needle insertion trajectory and the direction of the safe puncture path, as the angular deviation value; When the lateral deviation value or the angular deviation value exceeds the respective set allowable threshold, a trajectory deviation alarm message and correction suggestion containing the deviation value and deviation direction are immediately generated.

5. The sampling method for lymph node biopsy puncture according to claim 4, characterized in that, The generated trajectory deviation alarm information and correction suggestions include: The lateral deviation value and the angular deviation value are compared with preset multi-level alarm thresholds to determine the alarm level; Select the corresponding alarm method according to the alarm level. The alarm method includes color change of visual cues, graphic flashing, and frequency and pitch change of sound cues. Based on the vector synthesis result of the lateral deviation value and the angular deviation value, a theoretical correction motion vector is calculated; The correction motion vector is converted into guidance instructions for the operator. These instructions are displayed as graphic arrows overlaid on the real-time image or expressed through voice prompts, forming the correction suggestions. The correction suggestions include the direction in which the needle insertion angle should be adjusted and the approximate adjustment range.

6. The sampling method for lymph node biopsy puncture according to claim 1, characterized in that, The trigger biopsy sampling device performs tissue sampling and continuously monitors real-time image changes in the sampling area during the sampling process, including: When the system detects that the spatial distance between the tip coordinates of the actual needle trajectory and the target position coordinates is less than the trigger threshold, it automatically or after waiting for user confirmation sends a firing command to the biopsy sampling device. The biopsy sampling device, upon firing command, drives the internal cutting cannula and sampling needle to perform rapid forward, cutting, and retraction actions to obtain cylindrical tissue samples. The ultrasound image sequence and the color Doppler blood flow signal were continuously acquired and analyzed before, during and shortly after the firing action; Monitor the echo changes, morphological displacement, and presence of new blood flow signals in the target sampling area and surrounding tissues to assess whether the sampling procedure has caused unexpected bleeding or tissue damage.

7. The sampling method for lymph node biopsy puncture according to claim 2, characterized in that, The step of extracting morphological boundary information of lymph nodes from the ultrasound image sequence and constructing an initial geometric model of the lymph nodes using edge detection and contour tracking algorithms includes: Each frame of the ultrasound image sequence is preprocessed, including noise reduction filtering and contrast enhancement. On the preprocessed image, the active contour model algorithm is applied to initialize a closed curve that encloses the lymph node region; Drive the closed curve to evolve toward the true boundary of the lymph node under the combined action of the image gradient field and internal constraint force; When the evolution of the closed curve tends to stabilize, that is, when the energy function of the curve reaches its minimum value, the shape of the curve at this time is locked. The locked closed curve is tracked and reconstructed in three dimensions in consecutive frame images to obtain a continuous three-dimensional point cloud or mesh model representing the outer surface of the lymph node, which serves as the initial geometric model.

8. A method for lymph node biopsy sampling according to claim 6, characterized in that, The monitoring of echogenicity changes, morphological displacement, and the presence of new blood flow signals in the target sampling area and surrounding tissues is used to assess whether the sampling procedure has caused unexpected bleeding or tissue damage, including: Before firing, the baseline echo characteristics and baseline blood flow distribution map of the target sampling area are recorded; After the firing action is executed, images are continuously acquired within a preset time period, and the target sampling area and the area along the needle path are analyzed to see if new anechoic or hypoechoic areas appear. The new anechoic or hypoechoic areas represent fresh hematomas. Simultaneously, the color Doppler blood flow signal is monitored to check whether abnormal punctate or patchy blood flow signals appear outside the original vascular structure. The abnormal punctate or patchy blood flow signals indicate active bleeding. Based on the occurrence of the new anechoic or hypoechoic areas and the occurrence of the abnormal punctate or patchy blood flow signals, a sampling safety assessment report is generated.

9. The sampling method for lymph node biopsy puncture according to claim 1, characterized in that, The method also includes a sample quality pre-assessment step: Immediately after the biopsy sampling device completes the tissue sampling operation and is withdrawn from the body, in-situ optical imaging is performed on the obtained tissue sample. The in-situ optical imaging is used to preliminarily analyze the macroscopic morphology, color, length and diameter of the tissue sample, and to quickly compare it with a pre-stored database of normal lymph node tissue samples. Based on the comparison results, a representativeness score for the sample is calculated, which is based on the sample's completeness, size, and whether it contains characteristic structures of lymph nodes. If the representative score is lower than the qualified threshold, a prompt message suggesting resampling is generated, and the prompt message is recorded in the case log along with the current operating parameters.

10. A monitoring system for lymph node biopsy puncture, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the sampling method for lymph node biopsy puncture as described in any one of claims 1 to 9.

Images