Automatic calibration methods, systems, equipment and media for ultrasound-guided puncture mechanisms

Abstract

This application discloses an automatic calibration method, system, device, and medium for an ultrasound-guided puncture mechanism, relating to the field of medical puncture technology. The method includes the following steps: based on at least one preset target depth, and using an ideal inverse kinematics model, obtaining at least one theoretical control quantity corresponding to the preset target depth; based on the theoretical control quantity, controlling the puncture mechanism to move to the preset target depth and collecting an error array; constructing a linear interpolation compensation model based on the mapping relationship between the target depth and the corresponding error array; and obtaining a compensation quantity based on the target depth and the linear interpolation compensation model, and dynamically correcting the output of the ideal inverse kinematics model. This constructs a data-driven real-time interpolation compensation model, dynamically correcting the output of the inverse kinematics solution, thereby upgrading a low-precision hardware system to a high-precision level. It overcomes processing and assembly errors, improves puncture accuracy, and increases the success rate of surgery.

Description

Technical Field

[0001] This application relates to the field of medical puncture technology, and in particular to an automatic calibration method, system, device and medium for an ultrasound-guided puncture mechanism. Background Technology

[0002] Ultrasound-guided puncture technology is widely used in the medical field. For surgeries involving puncture procedures, accurate puncture path and angle are crucial for surgical success and patient prognosis. The puncture mechanism, as the execution unit, typically includes an ultrasound probe, a drive transmission mechanism, a puncture needle clamping mechanism, and a puncture needle. The precision of the puncture mechanism directly affects the success or failure of the surgery.

[0003] However, existing puncture mechanisms generally suffer from a combination of positioning errors due to factors such as machining errors, assembly errors, and simplification of kinematic models. These errors cause deviations between the actual position of the puncture needle tip and the theoretical position calculated based on an ideal kinematic model. Especially in deep punctures, even small angular errors can lead to significant target deviations.

[0004] In related technologies, high-precision machining and assembly are typically used to reduce errors, but this significantly increases costs. Some solutions employ manual calibration, but this process is cumbersome, relies heavily on operator experience, and struggles to achieve full-workspace accuracy coverage. Therefore, how to effectively improve the final puncture accuracy of ordinary precision puncture mechanisms through an automated, low-cost calibration and compensation method has become a pressing issue. Summary of the Invention

[0005] To address the aforementioned issues, this application provides an automatic calibration method and system for ultrasound-guided puncture mechanisms.

[0006] On the one hand, the automatic calibration method for an ultrasound-guided puncture mechanism provided in this application adopts the following technical solution:

[0007] An automatic calibration method for an ultrasound-guided puncture mechanism includes the following steps:

[0008] Based on at least one preset target depth, at least one theoretical control quantity corresponding to the preset target depth is obtained based on the inverse kinematics model.

[0009] Based on the theoretical control quantity, the puncture mechanism is controlled to move to the preset target depth and an error array is collected;

[0010] Based on the mapping relationship between the target depth and the corresponding error array, a linear interpolation compensation model is constructed;

[0011] Based on the target depth and the linear interpolation compensation model, the compensation amount is obtained, and the output of the ideal kinematic inverse solution model is dynamically corrected.

[0012] Optionally, obtaining at least one theoretical control quantity corresponding to the preset target depth based on an ideal kinematics inverse model, according to at least one preset target depth, includes:

[0013] Multiple preset target depth points are set along the depth direction at fixed step intervals, ranging from the minimum preset target depth to the maximum preset target depth.

[0014] Optionally, controlling the puncture mechanism to move to a preset target depth and collecting an error array based on the theoretical control quantity includes:

[0015] Determine if the current target depth is less than the preset maximum depth. If yes, continue; otherwise, end the calibration.

[0016] Based on the inverse kinematics model, the theoretical control quantity is calculated to make the needle tip reach the target depth.

[0017] Control the puncture mechanism to execute the theoretical control quantity;

[0018] Automatically identify the actual position of the puncture needle tip in the image coordinate system based on image processing;

[0019] Calculate the error between the actual position of the needle tip and the target position, wherein the error includes at least the depth direction error, the needle insertion direction compensation amount, and the lateral error;

[0020] Store the error corresponding to the current target depth into an error array;

[0021] Increase the target depth by a fixed step size, and repeat the above steps until all depth points are collected.

[0022] Optionally, the calculation of the theoretical control quantity that causes the needle tip to reach the target depth based on the ideal kinematic inverse model includes:

[0023] The theoretical control quantity includes the theoretical control quantity of three motors. The three motors are a first motor for controlling the needle insertion of the puncture mechanism, a second motor for controlling the angle change of the puncture needle relative to the ultrasonic probe, and a third motor for controlling the vertical movement of the puncture needle.

[0024] Optionally, the calculation of the error between the actual position of the needle tip and the target position includes at least the depth direction error, the needle insertion direction compensation amount, and the lateral error, including:

[0025] The needle insertion direction compensation amount Δd1 is determined by an iterative search algorithm, based on the theoretical solution. Based on this, the feed amount of the first motor is finely adjusted until the area or gray intensity of the needle tip feature in the ultrasound image reaches the preset range.

[0026] Optionally, in the step of constructing a linear interpolation compensation model based on the mapping relationship between the target depth and the corresponding error array,

[0027] The algorithm for the linear interpolation compensation model is as follows:

[0028]

[0029] Where n is the index of the nth discrete interval where the vessel depth h is located. These are linear interpolation coefficients. h0 is the preset target depth, h0 is the preset minimum target depth, and s is the fixed step size; Let n be the depth error at the nth discrete point. This is the compensation amount added to the first motor at the nth discrete point.

[0030] Optionally, the fixed step size is 0.2 mm, ranging from the minimum depth h0 to the maximum depth h. max Dense calibration point collection is performed within the range using a fixed step size.

[0031] Secondly, the automatic calibration system for an ultrasound-guided puncture mechanism provided in this application adopts the following technical solution:

[0032] An automated calibration system for an ultrasound-guided puncture mechanism includes:

[0033] The control quantity calculation module is used to obtain at least one theoretical control quantity corresponding to the preset target depth based on the ideal kinematics inverse solution model, according to at least one preset target depth.

[0034] The motion control and data acquisition module is used to control the puncture mechanism to move to the preset target depth based on the theoretical control quantity, and to acquire the error array at the preset target depth;

[0035] The compensation model construction module is used to construct a linear interpolation compensation model based on the mapping relationship between the target depth and the corresponding error array;

[0036] The dynamic correction module is used to obtain the compensation amount based on the target depth to be punctured and the linear interpolation compensation model, and to dynamically correct the output of the ideal kinematic inverse solution model.

[0037] Thirdly, the computer device provided in this application adopts the following technical solution:

[0038] A computer device, comprising one or more processors and memory;

[0039] One or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the method.

[0040] Fourthly, the computer-readable storage medium provided in this application adopts the following technical solution:

[0041] A computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method.

[0042] In summary, this application includes at least one of the following beneficial technical effects:

[0043] By utilizing ultrasound images as a feedback sensor and employing an automated calibration process, multi-dimensional error distribution data of the puncture needle across the entire workspace is acquired. A data-driven real-time interpolation compensation model is then constructed to dynamically correct the output of the inverse kinematics solution, thereby elevating the low-precision hardware system to a high-precision level. Through data-driven compensation, the system can proactively overcome systematic errors caused by needle flexibility, manufacturing and assembly errors, and tissue deformation. This improved puncture accuracy directly increases the surgical success rate and reduces the risk of tissue damage due to repeated punctures or off-target procedures. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the puncture mechanism according to an embodiment of this application;

[0045] Figure 2 This is a schematic diagram of ultrasound images from an embodiment of this application;

[0046] Figure 3 This is a schematic diagram showing the actual and ideal positions of the needle tip under ultrasound in this application;

[0047] Figure 4 This is a schematic diagram of the calibration tooling in an embodiment of this application;

[0048] Figure 5 This is a flowchart of an automatic calibration method for an ultrasound-guided puncture mechanism according to an embodiment of this application;

[0049] Figure 6 This application Figure 5 A flowchart of one embodiment of step S20;

[0050] Figure 7 This is a structural block diagram of an automatic calibration system for an ultrasound-guided puncture mechanism according to an embodiment of this application;

[0051] Figure 8 This is a schematic diagram showing that the puncture needle of this application has not reached the ultrasonic cross-section;

[0052] Figure 9 This is a schematic diagram of the puncture needle to the ultrasonic cross section of this application. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0054] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0055] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0056] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the meaning of "and / or" throughout the text includes three parallel solutions; for example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0057] Ultrasound-guided puncture technology is widely used in the medical field. For surgeries involving puncture procedures, accurate puncture paths and angles are crucial for surgical success and patient prognosis. As the execution unit, the precision of the puncture mechanism directly impacts the success or failure of the surgery.

[0058] like Figure 1 As shown, the puncture mechanism consists of an ultrasound probe, a drive transmission mechanism, a puncture needle clamping mechanism, and a puncture needle. Its working principle is that when the ultrasound probe is laterally positioned above the blood vessel, it can collect images such as… Figure 2 The ultrasound image shown is used to calculate the joint position of each puncture motor based on the depth information of the blood vessel under the ultrasound image, and drive each joint motor to reach the target position, so that the puncture needle passes through the cross-section of the blood vessel. Figure 2 The white dot within the green circle represents the appearance of the puncture needle tip under ultrasonic cross-section. Due to manufacturing and assembly errors, the final position of the white dot at the needle tip deviates somewhat from the target position, thus affecting puncture accuracy and blood collection success rate.

[0059] Theoretically, without manufacturing and assembly errors, the plane of motion of the puncture needle should coincide with the longitudinal section of the ultrasonic probe. However, in reality, due to manufacturing and assembly errors, the puncture needle is not within the longitudinal section of the ultrasonic probe. Therefore, this may cause... Figure 3 The results shown indicate that we controlled the puncture needle passage. Figure 3 The ideal position circled in red, while the actual position of the needle tip under ultrasound is... Figure 3 Within the blue circle. If... Figure 3 The red circle indicates the location of blood vessels in the human body under ultrasound. Without error compensation, the puncture needle cannot accurately insert into the blood vessel.

[0060] Therefore, in order to measure the deviation between the ideal and actual positions of puncture needles at different depths under ultrasound, we designed a method as follows: Figure 4 The calibration fixture is shown. After calibration, the deviation of the puncture needle from the ideal position is measured and saved. Then, the corresponding deviation is interpolated and compensated into the corresponding inverse kinematic solution results. This eliminates the influence of machining and assembly errors on puncture accuracy. Before running the automatic calibration algorithm program, we need to fill the calibration fixture with ultrasonic coupling agent and correctly install the calibration fixture and calibration needle.

[0061] Example 1

[0062] This application discloses an automatic calibration method for an ultrasound-guided puncture mechanism.

[0063] Reference Figure 5 An automatic calibration method for an ultrasound-guided puncture mechanism is disclosed. In this embodiment, the puncture mechanism includes three motors: a first motor for controlling the needle insertion of the puncture mechanism, a second motor for controlling the angle change of the puncture needle relative to the ultrasound probe, and a third motor for controlling the vertical movement of the puncture needle.

[0064] To measure the deviation between the ideal and actual positions of puncture needles at different depths under ultrasound, a design was created as follows: Figure 1 The calibration fixture shown can present a series of reference features of known depth under ultrasound images; before performing the calibration steps, it is necessary to fill the calibration fixture with ultrasound coupling agent and correctly install the calibration fixture and calibration needle.

[0065] The automatic calibration method includes the following steps:

[0066] S10. Based on at least one preset target depth, obtain at least one theoretical control quantity corresponding to the preset target depth using an ideal kinematics inverse solution model.

[0067] Optionally, in this embodiment, the minimum preset target depth h0 to the maximum preset target depth h max Within the specified range, multiple preset target depth points are set along the depth direction at fixed step intervals. The fixed step is 0.2 mm.

[0068] The ideal kinematic inverse kinematics model is established based on the ideal geometric parameters of the puncture mechanism. Ideally, the parts have no machining errors, the assembly follows the designed 3D model without assembly errors, and the motion plane of the puncture needle coincides with the plane of the ultrasound probe. We only need to use a robotic arm to move the ultrasound probe laterally, adjusting the position of the blood vessel in the ultrasound image to the center of the image, i.e., the blood vessel is within the physical longitudinal plane of the ultrasound probe. Then, based on the blood vessel depth h in the ultrasound image, the values ​​of the lifting motor, the rotation joint, and the puncture motor can be inversely solved.

[0069] For a given target position, such as the depth coordinate h, the theoretical control quantities [d1, θ2, d3]^T required to drive the three motors can be calculated. The first motor controls the feed of the puncture needle, a linear motion d1; the second motor controls the swing angle θ2 of the puncture needle; and the third motor controls the vertical movement of the puncture needle, a linear motion d3.

[0070] In one specific implementation, a calibrated target depth range can be set, for example, from h0 = 10 mm to h max =100mm. Within this range, a series of densely spaced, pre-defined target depth points are set with a fixed step size s = 0.2mm. ,Right now =[10.0, 10.2, 10.4, ... 99.8, 100.0] mm. This dense sampling ensures that the subsequently constructed compensation model has high accuracy.

[0071] S20. Based on the theoretical control quantity, control the puncture mechanism to move to the preset target depth and collect the error array.

[0072] like Figure 6 As shown, this step includes:

[0073] S201. Determine whether the target depth is less than the preset maximum depth;

[0074] Specifically, initialize the target depth Given the minimum value h0, determine whether the current target depth is less than the preset maximum depth h. max If yes, continue; otherwise, end the calibration.

[0075] It should be noted that the maximum depth h max It should not be simply set as a mechanical limit, but should be determined based on clinical goals, such as the location of the blood vessel to be punctured.

[0076] S202. Based on the ideal kinematics inverse model, calculate the theoretical control quantity [d1, θ2, d3]^T when the needle tip reaches the target depth;

[0077] The ideal kinematic inverse model is as follows:

[0078] = .

[0079] S203. Control the puncture mechanism to execute the theoretical control quantity, and drive the puncture needle to move toward the target position;

[0080] S204. Automatically identify the actual position of the puncture needle tip in the image coordinate system based on image processing;

[0081] Specifically, after the motor motion stabilizes, images are acquired using an ultrasonic probe. Ultrasonic image processing algorithms automatically identify the characteristics of the puncture needle tip and calculate its actual position (x, y) in the ultrasonic coordinate system. n h n The image processing algorithm includes one or more combinations of threshold segmentation, morphological operations, Hough transform, or convolutional neural networks.

[0082] S205. Calculate the error between the actual position of the needle tip and the target position, wherein the error includes at least the depth direction error, the needle insertion direction compensation amount, and the lateral error.

[0083] The depth direction error Δh is the difference between the actual coordinates of the needle tip in the depth direction of the ultrasound image and the target depth h. The needle insertion direction compensation Δd1 is used to ensure clear imaging of the needle tip within the ultrasound imaging plane. It is determined through an iterative search algorithm, that is, based on the theoretical control value d1, the feed amount of the first motor is finely adjusted in small steps, such as 0.05 mm, while the area or grayscale intensity of the needle tip feature in the ultrasound image is analyzed in real time. When the area or grayscale intensity reaches its maximum value or a certain preset threshold range, the needle tip is considered to be in the optimal imaging position, and the adjustment amount of the first motor relative to the theoretical value d1 at this time is recorded as Δd1.

[0084] Lateral error Δx refers to the motion compensation of the robotic arm in the X direction of the tool coordinate system. In the compensation model of this embodiment, the main compensation is achieved through depth error Δh and needle insertion compensation Δd1. Lateral error is compensated by the motion compensation of the robotic arm driving the puncture mechanism along the X direction of the ultrasound probe image coordinate system. After compensating for depth error Δh, needle insertion compensation Δd1, and lateral error, the puncture needle can accurately reach the target blood vessel without adjusting the control of the rotation joint; therefore, the corresponding θ2 compensation is 0.

[0085] S206. Store the error corresponding to the current target depth into an error array, wherein the corresponding error includes at least one of Δh, Δd1 and Δx;

[0086] Error array structure:

[0087]

[0088] The step size 's' needs to strike a balance between calibration accuracy and efficiency. A smaller step size results in a denser mapping table, potentially leading to higher compensation accuracy, but also a longer calibration time. In this implementation, a step size of 0.2 mm is used.

[0089] S207. Increase the target depth by a fixed step size, and repeat the above steps until all depth points are collected; that is, increase the target depth h by a fixed step size s, then return to step S201 and repeat the above process until the depth from h0 to h0 is collected. max Data collection at all preset depth points within the range.

[0090] Through Δh and Δd1, the system learns the actual control values ​​needed to reach the target depth h. The actual needle tip position under ultrasound accurately falls on the location of the blood vessel. The search process of Δd1 actively ensures that the needle tip remains within the ultrasound imaging plane, solving the "blind puncture" problem of out-of-plane puncture. This enables real-time monitoring and subsequent automatic compensation throughout the process, significantly improving the absolute accuracy of the needle tip reaching the target position. Furthermore, the system automatically identifies the needle tip and calculates errors through image processing algorithms, reducing human intervention, decreasing reliance on operator experience, and improving the repeatability and standardization of the surgery.

[0091] S30. Based on the mapping relationship between the target depth and the corresponding error array, construct a linear interpolation compensation model.

[0092] In one embodiment, the target depth of the nth target point is preset during calibration. When the three puncture motors execute the motion control quantities obtained from the inverse kinematics solution, a white dot will appear on the ultrasound image as the puncture needle tip passes through the ultrasound interface; this white dot is called the needle white dot. The size of the needle white dot is positively correlated with the stroke of the needle insertion motor. For example... Figure 8As shown, when following the inverse solution of the theory, the needle tip may not reach the ultrasonic cross-section, in which case the area of ​​the needle white spot is 0. To ensure the needle's cutting edge is on the ultrasonic cross-section, the theoretical solution of the needle insertion motor (i.e., the first motor) is used... Increase compensation amount This ensures that the area of ​​the white dot is within a preset range. At this point, the position of the needle tip is as follows: Figure 9 As shown, depth discretization is then used to obtain the depth of different target points. The corresponding depth error . ,in, To calibrate the target depth The actual depth of the white dot on the needle. Through the calibration process described above, three error arrays can be obtained. , , .

[0093] In one embodiment, a linear interpolation compensation model is constructed for use at any target depth. (h0≤ ≤ h max When ), it provides compensation for the theoretical control quantity.

[0094] The algorithm for the linear interpolation compensation model is as follows:

[0095]

[0096]

[0097] in , The algorithm for the linear interpolation compensation model after compensation, used for compensating the first and second motors of the puncture mechanism for the theoretical inverse solution, is as follows:

[0098] ;

[0099] Where n is the index of the nth discrete interval where the vessel depth h is located. These are linear interpolation coefficients. h0 is the preset target depth, h0 is the preset minimum target depth, and s is the fixed step size; Let n be the depth error at the nth discrete point. This is the compensation amount added to the first motor at the nth discrete point.

[0100] In one specific embodiment, an array is set. The nth member of D array The nth member of H The linear interpolation compensation algorithm after calibration correction can be derived as follows:

[0101] ;

[0102] = + .

[0103] After calibration, the value of the second motor of the puncture mechanism was not changed. Therefore, the error in the depth of the puncture needle can be directly compensated by the third motor that controls the vertical movement of the puncture mechanism.

[0104] Optionally, in one embodiment, the lateral compensation amount is calculated based on the error array ΔX:

[0105] [d4]=[ΔX[n]+ fra×(ΔX[n+1] -ΔX[n])].

[0106] In this application, the linear interpolation model connects adjacent calibration points, at any two known depth points (h... n ) and (h n+1 An "error prediction line" was established between the two points. This allows the system to provide a smooth and continuous compensation for any target depth within the workspace, not just limited to a few calibrated points. This greatly expands the system's practicality and flexibility, enabling it to handle punctures of various depths in clinical practice. Secondly, linear interpolation, through the fra coefficient, ensures that the compensation amount... d1 and h transitions linearly and smoothly with changes in depth h. The rate of change of the compensation amount is constant, without abrupt changes. This ensures the smoothness and continuity of the puncture mechanism's movement, avoids mechanical vibrations caused by abrupt changes in control commands, and thus improves the overall control quality of the system and the compliance of the puncture process.

[0107] The linear interpolation compensation model algorithm is simple, involving only basic addition, subtraction, and multiplication operations, resulting in minimal computational complexity and high speed. This model is suitable for operation in embedded systems or real-time controllers. It can complete instruction compensation for each control cycle without consuming significant computing resources or introducing substantial time delays, thus meeting the stringent real-time requirements of robot systems.

[0108] Furthermore, the differential compensation in this application can cancel out errors before they occur, significantly reducing the steady-state error of the system. Compared to simple feedback control, which adjusts only after an error is detected, the feedforward-feedback combined control strategy responds faster and is more accurate. Ultimately, this results in a significant improvement in the absolute accuracy of the needle tip reaching the target position.

[0109] S40. Based on the target depth and the linear interpolation compensation model, obtain the compensation amount and dynamically correct the output of the ideal kinematics inverse solution model.

[0110] In actual puncture surgery, when it is necessary to puncture to a certain target depth h, the theoretical control quantity is first calculated using an ideal kinematic inverse model. Then, based on the current target depth h, the corresponding compensation amount [Δd1, 0, Δh]^T is queried or calculated using the above linear interpolation compensation model. Finally, the compensation amount is added to the theoretical control amount to obtain the corrected control amount. This allows for dynamic and precise compensation of system errors by controlling the movement of the puncture mechanism.

[0111] The bending of the needle in tissue is a complex nonlinear process. Through a linear interpolation compensation model, the system approximates this complex nonlinear error curve with a series of continuous linear segments. After compensation, the system's performance no longer strongly depends on precision machining or ideal tissue assumptions. For the same hardware and similar soft tissue, the system can stably reproduce high-precision puncture results, reducing reliance on the operator's personal experience. This allows for the standardized and reproducible application of precise puncture techniques in clinical practice, facilitating the promotion and standardization of the technique.

[0112] Example 2

[0113] Reference Figure 7 This application also discloses an automatic calibration system for an ultrasound-guided puncture mechanism, comprising:

[0114] The control quantity calculation module is used to obtain at least one theoretical control quantity corresponding to the preset target depth based on the ideal kinematic inverse solution model.

[0115] In one specific embodiment, the module incorporates an ideal inverse kinematic model of the puncture mechanism. It receives one or more preset target depths, such as a sequence from 10mm to 100mm in 0.2mm increments, as input. For each target depth, the module assumes an ideal needle tip target position and calculates the driving parameters for the three joints. The theoretical control quantity required to reach this position. The output of this module is a series of values: target depth h, theoretical control quantity [ The pairing data of ]^T provides a command reference for subsequent physical motion.

[0116] The motion control and data acquisition module is used to control the puncture mechanism to move to the preset target depth based on the theoretical control quantity, and to acquire the error array at the preset target depth.

[0117] In one specific embodiment, after the motor motion stabilizes, the module is activated to perform the following precise measurement process:

[0118] Image Triggering and Acquisition: Controls the ultrasound equipment to acquire the current pattern image.

[0119] Needle tip recognition: The image processing algorithm automatically identifies the puncture needle tip in the image and calculates its precise centroid position (x, y) in the ultrasound image coordinate system. n h n ).

[0120] Error calculation: based on the identified actual position (x) n h n ) and the preset target position, calculate multi-dimensional error;

[0121] Depth orientation error Δh: Δh = h n - h directly reflects the accuracy of the needle insertion depth.

[0122] The needle insertion direction compensation amount Δd1 is obtained by iteratively searching through the micro-motion of the first motor and the analysis of image clarity, such as the maximum grayscale gradient of the needle tip area, to ensure that the needle tip is always in the optimal imaging plane.

[0123] Lateral error Δx: Δx = x n Let the target be at x=0, which is used to evaluate the lateral offset.

[0124] This module ultimately outputs an error array covering the entire target depth range, establishing a mapping database between "target depth" and "multi-dimensional error".

[0125] The compensation model construction module is used to construct a linear interpolation compensation model based on the mapping relationship between the target depth and the corresponding error array.

[0126] In one specific embodiment, the module receives a discrete error array from the data acquisition module. H, D1, X).

[0127] Using these discrete data points, a continuous compensation function applicable to the entire workspace is constructed. Through linear connections between adjacent calibration points, the system can calculate smoothly varying compensation amounts for arbitrary target depths, not just calibrated depths. The module's output is a compensation model that can be invoked in real time, encapsulating systematic error knowledge specific to the particular mechanism and needle.

[0128] The dynamic correction module is used to obtain the compensation amount based on the target depth to be punctured and the linear interpolation compensation model, and to dynamically correct the output of the ideal kinematic inverse solution model.

[0129] In this application, through data-driven compensation, the system can proactively overcome systematic errors caused by needle flexibility, processing and assembly errors, tissue deformation, etc. Precise control of puncture accuracy directly improves the success rate of surgery and reduces the risk of tissue damage caused by repeated punctures or off-target procedures.

[0130] Example 3

[0131] This application also discloses a computer device, including one or more processors and a memory;

[0132] One or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the methods described above.

[0133] Example 4

[0134] This application also discloses a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method.

[0135] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a device including one or any combination of the above-mentioned memories. The computer may be a variety of computing devices, including smart terminals and servers.

[0136] In the above embodiments of this disclosure, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0137] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0138] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0139] Furthermore, the functional units in the various embodiments of this disclosure 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.

[0140] 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 computer-readable non-volatile storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a non-volatile storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this disclosure. The aforementioned non-volatile storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0141] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. An automatic calibration system for an ultrasound-guided puncture mechanism, characterized by, Comprise: a control amount calculation module configured to obtain at least one theoretical control amount corresponding to a preset target depth based on an ideal kinematic inverse solution model according to at least one preset target depth; a plurality of preset target depth points are set at fixed step intervals along the depth direction within the range from the minimum preset target depth to the maximum preset target depth; a motion control and data acquisition module configured to control the puncture mechanism to move to the preset target depth based on the theoretical control amount and acquire an error array at the preset target depth; a compensation model construction module configured to construct a linear interpolation compensation model according to the mapping relationship between the target depth and the corresponding error array; a dynamic correction module configured to obtain a compensation amount based on the target depth and the linear interpolation compensation model, and dynamically correct the output of the ideal kinematic inverse solution model; Wherein, the control of the puncture mechanism to move to the preset target depth based on the theoretical control amount and the acquisition of the error array, comprising: determine whether the current target depth is less than the preset maximum depth, if yes, continue; if not, the calibration is ended; calculate the theoretical control amount that makes the needle tip reach the target depth based on the ideal kinematic inverse solution model; control the puncture mechanism to execute the theoretical control amount; automatically identify the actual position of the puncture needle tip in the image coordinate system based on image processing; calculate the error between the actual position of the needle tip and the target position, the error at least including depth direction error, needle insertion direction compensation amount and lateral error; store the error corresponding to the current target depth to the error array; increase the target depth by a fixed step, repeat the above steps until all depth point acquisition is completed; In the step of constructing a linear interpolation compensation model according to the mapping relationship between the target depth and the corresponding error array, The algorithm of the linear interpolation compensation model is: wherein n is the index of the n-th discrete interval in which the vessel depth h is located, is a linear interpolation coefficient, is a preset target depth, h0 is a preset minimum target depth, and s is a fixed step length; is the depth error of the n-th discrete point, is the compensation amount added by the first motor at the n-th discrete point.

2. The automatic calibration system of claim 1, wherein The calculation of the theoretical control amount that makes the needle tip reach the target depth based on the ideal kinematic inverse solution model, comprising: The theoretical control amount includes the theoretical control amount of three motors, the first motor for controlling the puncture mechanism to insert the needle, the second motor for controlling the angle change of the puncture needle relative to the ultrasonic probe, and the third motor for controlling the vertical movement of the puncture needle.

3. The automatic calibration system of claim 2, wherein The calculation of the error between the actual position of the needle tip and the target position, the error at least including depth direction error, needle insertion direction compensation amount and lateral error, comprising: The needle insertion direction compensation amount Δd1 is determined by an iterative search algorithm, and the first motor feed amount is fine-tuned on the basis of the theoretical solution of the needle tip feature in the ultrasound image reaches a preset range.

4. The automatic calibration system of claim 3, wherein The fixed step size is 0.2 mm, and the dense calibration point collection is performed at a fixed step size within the range from the minimum depth ho to the maximum depth h max .

Images