Flexible puncture needle motion trail control prediction system and method

Abstract

The invention discloses a flexible puncture needle motion trail control prediction system and method. The system comprises a mechanical arm, a needle feeding mechanism, a flexible puncture needle, an ultrasonic probe, a signal demodulation instrument and an upper computer. Wherein the mechanical arm is used for adjusting the posture of a needle feeding point; the needle inserting mechanism is used for adjusting the puncture movement radius and needle inserting and rotating operations; the head rigidity of the flexible puncture needle is adjustable, the bending radius of the flexible puncture needle is in direct proportion to the extension length, and an optical fiber sensor is arranged to collect the body posture and cutting force of the needle body; the ultrasonic probe is used for collecting tissue information; the signal demodulator is responsible for acquiring and processing optical fiber and ultrasonic signals; the upper computer comprehensively processes operation area information such as the position and posture of the puncture needle, the bending radius, the needle tip cutting force and the contact tissue type in the puncture process, a neural network algorithm is adopted for predicting and visually displaying the puncture track, and obstacle avoidance and precise puncture are achieved by adjusting the puncture track of the puncture needle. The puncture track can be effectively adjusted and visually predicted, the puncture time is shortened, and the puncture quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of medical equipment and robot navigation technology, and in particular to a flexible puncture needle motion trajectory control prediction system and method. Background Art

[0002] Percutaneous puncture is a typical minimally invasive interventional surgery. It has extremely important clinical applications in the diagnosis and treatment of diseased tissue biopsy, ablation, radiotherapy, etc. It is the main method for early diagnosis and treatment of cancer. Flexible needle puncture can use the bending of the flexible needle to avoid sensitive areas such as blood vessels and nerves, and flexibly reach lesions that are difficult to reach with traditional rigid needles. It greatly expands the needle insertion options for safe puncture and reduces surgical risks. It is of great clinical significance. However, due to the diversity, heterogeneity, viscoelasticity and other mechanical properties of biological tissues, the force on the flexible needle during the puncture process is extremely complex, which increases the uncertainty of the flexible needle's motion trajectory and seriously affects the puncture accuracy.

[0003] The inventors found in actual research that the traditional conventional oblique flexible needle has a single structure, and the curvature of the needle body movement guided by the reaction force of the tissue is relatively fixed, which greatly limits its path deviation correction ability and the puncture path is not flexible enough. In addition, the lack of real-time interactive information between the flexible needle and the soft tissue during the puncture process, such as the contact tissue type, cutting force, relative position, etc., makes it difficult to predict and control the puncture motion trajectory, resulting in an unsatisfactory needle insertion trajectory, and repeated punctures will increase local damage. Summary of the invention

[0004] In order to address the deficiencies of the prior art, the present invention provides a flexible puncture needle motion trajectory control prediction system and method, which changes the turning radius of the puncture path by adjusting the structural stiffness of the flexible needle, and predicts and visualizes the puncture path by combining visual and tactile information with intelligent calculation, which is of great significance to improving the puncture surgical robot's ability to perceive information in the surgical area and the flexibility and controllability of the puncture movement.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] In a first aspect, the present invention provides a flexible puncture needle motion trajectory control prediction system, including a robotic arm, a needle insertion mechanism, a flexible puncture needle, an ultrasonic probe, a signal demodulation instrument and a host computer;

[0007] The end of the mechanical arm is fixed with the needle insertion mechanism and the ultrasonic probe, and is used to adjust the insertion position and angle of the puncture needle;

[0008] The flexible puncture needle comprises a flexible needle core with an oblique opening having a variable diameter and a sleeve for providing support;

[0009] The needle feeding mechanism has three degrees of freedom, which are used for cannula feeding, needle core feeding, and needle core rotation. The flexible needle core is grooved and equipped with an optical fiber sensor.

[0010] The optical fiber sensor collects information on needle shape changes and needle tip puncture force in real time;

[0011] The ultrasonic probe is used for real-time 3D imaging during the puncture process to obtain the shape information of the flexible puncture needle, the lesion target, and the obstacle tissue information;

[0012] The signal demodulator demodulates the information collected by the optical fiber sensor and the signal collected by the ultrasonic probe, and transmits them to the host computer for further processing;

[0013] The host computer aligns and fuses the flexible needle body shape information reconstructed by the optical fiber sensor with the three-dimensional medical image of the flexible puncture needle shape reconstructed by the ultrasonic probe to obtain clear needle body posture information; then the needle body posture information, lesion target, obstacle tissue information, needle tip puncture force information, needle core length, puncture speed, and needle insertion depth are input into the trajectory prediction model. The trajectory prediction model predicts the motion trajectory of the needle insertion mechanism and visualizes it to provide doctors with a more intuitive puncture motion trajectory and potential puncture risks; the doctor or operator makes a control decision based on the visualized predicted trajectory to control the needle insertion mechanism, and the needle insertion mechanism adjusts the length of the flexible needle core extending out of the sleeve to change the puncture motion trajectory and puncture until the target point.

[0014] As a further technical solution, the flexible needle core is a nickel-titanium needle core.

[0015] As a further technical solution, the sleeve is a nickel-titanium sleeve, and the flexible needle puncture trajectory can be adjusted by controlling the length of the nickel-titanium needle core extending out of the nickel-titanium sleeve.

[0016] As a further technical solution, the host computer includes a puncture motion control module, a biological tissue recognition module, an image fusion module and a trajectory prediction module; the biological tissue recognition module and the image fusion module communicate with the trajectory prediction module, and the trajectory prediction module communicates with the puncture motion control module.

[0017] As a further technical solution, the puncture motion control module drives the mechanical arm and the needle insertion mechanism to complete the adjustment of the needle insertion position, needle insertion angle, puncture speed, needle core length, and deflection direction.

[0018] As a further technical solution, the biological tissue recognition module is responsible for processing the received optical fiber force sensing signal to obtain the surgical area information during the puncture process.

[0019] As a further technical solution, the image fusion module registers and fuses the shape of the flexible needle body reconstructed by the optical fiber shape sensor with the three-dimensional medical image of the shape of the flexible puncture needle to obtain a clear needle body posture.

[0020] As a further technical solution, the trajectory prediction module inputs the measurable parameters related to the puncture motion trajectory into the trained neural network model, predicts and visualizes the motion trajectory, and drives the needle insertion mechanism to adjust the bending radius of the puncture needle according to the optimized trajectory.

[0021] In a second aspect, the present invention further provides a trajectory control prediction method based on the flexible puncture needle motion trajectory control prediction system, as follows:

[0022] In advance, the needle tip puncture force information and needle body bending deflection under different puncture speeds, feeding depths, needle core lengths, and bionic tissue types are collected and analyzed through puncture experiments to establish a sample data set; based on the observable information of puncture speed, needle core length, and needle tip puncture force, a flexible needle motion trajectory prediction module is generated through artificial intelligence technology;

[0023] During the puncture implementation phase, the optical fiber sensor collects information on needle shape changes, needle tip puncture force, puncture speed, needle core length, and other information in real time, and inputs them into the motion trajectory prediction module to generate a predicted trajectory.

[0024] The ultrasound probe performs real-time 3D imaging during the puncture process to obtain the shape information of the flexible puncture needle, the target lesion, and the obstacle tissue information;

[0025] The signal demodulator receives the information collected by the optical fiber sensor and the ultrasonic probe, demodulates the signal, and transmits it to the host computer for further processing;

[0026] The host computer aligns and fuses the shape of the flexible needle reconstructed by the optical fiber sensor with the shape information of the flexible puncture needle reconstructed by the ultrasonic probe to obtain clear needle body posture information; further, the predicted trajectory generated by the trajectory prediction module is visualized in the aligned and fused 3D image, and a control decision is made based on the visualized predicted trajectory. The needle insertion mechanism is then controlled to adjust the length of the flexible needle core extending out of the sheath to change the bending trajectory of the puncture movement and puncture until the target point.

[0027] As a further technical solution, the method for registering and fusing the shape of the flexible needle body reconstructed by the optical fiber sensor with the shape information of the flexible puncture needle reconstructed by the ultrasound probe is as follows:

[0028] Data preprocessing;

[0029] Obtain the bending degree and angle of the needle at each position during the puncture process, and reconstruct the three-dimensional shape curve of the needle by integrating the arc length; obtain the ultrasound image of the needle in the body, and use the edge detection algorithm to identify and extract the contour information of the needle;

[0030] Non-rigid registration is used to correctly align the three-dimensional needle shape data of the optical fiber sensor and the needle data of the ultrasound image in the same space;

[0031] The tip and inflection point of the needle in the fiber-optic reconstructed image and the three-dimensional ultrasound image are selected as registration features; the deformation between coordinate systems is simulated by B-spline deformation, and the non-rigid transformation is calculated by the gradient descent method to obtain a non-rigid transformation model of translation, rotation, distortion and stretching features; the fiber-optic sensor data is applied to the ultrasound image coordinate system after non-rigid transformation to complete the alignment of the origin position and direction of the two coordinate systems of the fiber-optic image and the ultrasound image; the boundary of the needle body is extracted by the ultrasound image segmentation method, and then the boundary is aligned with the needle body shape reconstructed by the fiber optic; the needle body shape reconstructed by the fiber optic is used as a constraint condition to assist the ultrasound image segmentation, so that the shape of the needle body can be accurately superimposed on the ultrasound image;

[0032] In three-dimensional space, the fiber optic sensor data and the needle position in the ultrasound image are weighted and fused, and the weights are assigned according to the clarity of the image and the reliability of the needle shape data. Using rendering technology in computer graphics, the needle shape reconstructed by the fiber optic sensor is superimposed on the three-dimensional ultrasound image in the form of a virtual object to obtain a comprehensive visual image.

[0033] The working principle of the present invention is as follows:

[0034] According to the preoperative plan, the robot arm is controlled to adjust the puncture needle to the appropriate insertion point and angle for puncture. During the puncture, the shape of the fiber-optic-flexible needle body is fused with the ultrasonic medical image to obtain information such as the needle tip posture, lesion target, and obstacle tissue. At the same time, the cutting force, tissue feature type, puncture speed, needle core length and other factors affecting the puncture bending radius are collected in real time during the puncture process. The trajectory is predicted through the trained neural network model and visualized in the reconstructed medical image; the system controls the needle insertion mechanism according to the predicted trajectory to adjust the length of the flexible needle core extending out of the sleeve to change the puncture movement bending radius, and punctures according to the predicted trajectory until the target point.

[0035] Beneficial effects of the present invention:

[0036] The variable stiffness flexible needle structure and the corresponding flexible needle control method proposed by the present invention are simple and convenient, and can significantly enhance the flexibility and controllability of the oblique flexible needle. Multi-sensor information fusion greatly improves the information perception ability of the surgical area during puncture. The trajectory prediction method based on visual-tactile information and artificial intelligence technology improves the predictability and visualization of the puncture movement, and provides a basis for puncture decision-making, greatly improving the accuracy and safety of puncture. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0038] Figure 1 Schematic diagram of a flexible puncture needle motion trajectory control and prediction system in an embodiment of the present application;

[0039] Figure 2 The overall structure and cross-sectional diagram of the flexible needle in the embodiment of the present application;

[0040] Figure 3 A schematic diagram of the bending prediction and obstacle avoidance of the flexible needle in the embodiment of the present application;

[0041] Figure 4 is a schematic diagram of a needle insertion mechanism in an embodiment of the present application;

[0042] Figure 5 It is a schematic diagram of the needle core rotating device of the needle insertion mechanism in the embodiment of the present application;

[0043] Figure 6 It is a schematic diagram of a needle core feeding device of a needle insertion mechanism in an embodiment of the present application;

[0044] Figure 7 This is a flowchart of the tissue characteristics and type identification process in the embodiment of the present application;

[0045] Figure 8 A flowchart of the fiber-ultrasound image fusion program for an example implementation of the present application;

[0046] Fig. 9 This is a flowchart of the flexible needle puncture trajectory prediction program for an example implementation of the present application.

[0047] In the figure: 1. multi-degree-of-freedom robotic arm, 2. needle insertion mechanism, 3. flexible puncture needle, 4. ultrasonic probe, 5. signal line, 6. signal demodulator, 7. host computer, 8. patient, 9. nickel-titanium needle core needle body, 10. nickel-titanium needle core head, 11. nickel-titanium sleeve, 12. optical fiber, 13. optical fiber sensor, 14. first puncture trajectory, 15. second puncture trajectory, 16. obstacle tissue, 7. needle core rotating device, 18. needle core feeding device, 19. puncture needle feeding device, 20 motor, 21 screw, 22 connecting block, 23 slide; 24 needle core clamping device, 25 gear set, 26 motor, 27 motor, 28 slide, 29 screw. DETAILED DESCRIPTION

[0048] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0049] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0050] In order to address the deficiencies of the prior art, the present invention provides a flexible puncture needle motion trajectory control prediction system and method, which changes the turning radius of the puncture path by adjusting the structural stiffness of the flexible needle, and predicts and visualizes the puncture path by combining visual and tactile information with intelligent calculation, which is of great significance to improving the puncture surgical robot's ability to perceive information in the surgical area and the flexibility and controllability of the puncture movement.

[0051] The flexible puncture needle motion trajectory control prediction system disclosed in this embodiment includes a multi-degree-of-freedom robotic arm 1, a needle insertion mechanism 2, a flexible puncture needle 3, an ultrasonic probe 4, a signal demodulator 6 and a host computer 7.

[0052] The end of the mechanical arm is fixed with a needle insertion mechanism 2 and an ultrasonic probe 4 for adjusting the insertion position and angle of the puncture needle.

[0053] The needle insertion mechanism 2 has three degrees of freedom, which are used for needle core rotation, needle core feeding, and puncture needle feeding, and should be able to complete the adjustment of the length of the needle core extending out of the sleeve and the measurement of the extension length. The needle insertion mechanism 2 mainly includes a needle core rotating device 17, a needle core feeding device 18, and a puncture needle feeding device 19.

[0054] Specifically, the needle core rotating device 17 includes a needle core clamping device 24, a gear set 25, and a motor 26. The nickel-titanium needle core needle body 9 is fixed on the needle core clamping device 24. The motor 26 rotates, driving the needle core clamping device 24 to rotate through the gear set 25, thereby realizing the rotation of the nickel-titanium needle core needle body 9 and achieving puncture steering adjustment.

[0055] The needle core feeding device 18 includes a motor 27, a slide 28, and a screw 29, which is an ultra-micro screw slide mechanism; the needle core rotating device 17 is fixed on the slide 28; the slide 28 and the screw 29 are threadedly matched, and the screw 29 is connected to the motor 27. A nickel-titanium sleeve 11 is fixed on the connecting plate 16, and the connecting plate 16 and the needle core feeding device 18 are installed at fixed positions. The motor 27 is controlled to rotate and drive the slide 28 to move, thereby controlling the needle core rotating device 17 to make a linear motion relative to the nickel-titanium sleeve 11, thereby realizing the adjustment of the length of the needle core extending out of the sleeve.

[0056] The puncture needle feeding device 19 includes a motor 20, a lead screw 21, a connecting block 22, and a slide 23. The needle core feeding device 18 and the connecting plate 16 are fixed on the slide 23, the slide 23 is connected to the connecting block 22, and the connecting block 22 is threadedly matched with the lead screw 21. By controlling the rotation of the motor 20, the linear motion of the connecting block 22 is realized, and then the puncture speed and position of the puncture needle (including the cannula and the needle core) can be adjusted.

[0057] The flexible puncture needle comprises a nickel-titanium bevel needle core 9 with a variable diameter and a nickel-titanium sleeve 11 for providing support; the nickel-titanium needle core head 10 is a conventional bevel tip, and its diameter is consistent with that of the sleeve, so as to provide sufficient deflection force for fully squeezing the tissue. Compared with the needle tip, the needle core needle body diameter is slightly thinner, which is equivalent to the inner diameter of the sleeve. The nickel-titanium bevel needle core 9 is inserted into the nickel-titanium sleeve 11, and the length L of the nickel-titanium bevel needle core 9 extending out of the sleeve 11 and the bevel direction are adjusted by the needle insertion mechanism 2: the length L of the needle core extending out of the sleeve is controlled, and the rigidity of the flexible puncture needle head (including the needle core head and part of the sleeve) is changed to achieve the purpose of changing the bending radius of the flexible needle: for example, when puncturing in a plane, there is an obstacle tissue 16. When the length of the needle core extending out of the sleeve is L1, the bending radius of the first puncture trajectory 14 is R1, and continued puncture will puncture the obstacle tissue. At this time, the extension length can be adjusted to L2, the rigidity of the flexible needle head is reduced, and it is easier to bend under the same tissue force, and the flexible needle will avoid the obstacle tissue along the second puncture trajectory 15. The puncture movement direction in three-dimensional space can be adjusted by rotating the needle core.

[0058] In addition, the flexible needle core is grooved and equipped with fiber optic sensors. The fiber optic sensor implanted in the nickel-titanium needle core head 10 is used to measure the cutting force during the puncture process, and the fiber optic sensor distributed in the nickel-titanium needle core needle body 9 is used to measure the shape of the needle body during the puncture process. The fiber optic sensor collects signals and transmits them to the host computer for processing to display the needle body shape and needle tip cutting force information.

[0059] Specifically, the optical fiber sensor may adopt a fiber Bragg grating (FBG) or a Fabry-Perot interferometer (FPI), wherein the FBG senses strain and temperature by measuring the change of the central wavelength of the grating, while the FPI senses strain and temperature by measuring the shift of the interference spectrum.

[0060] FBG sensor: When the optical fiber is stretched or compressed, the periodic structure of the grating changes, causing the wavelength of the reflected light to change. Multiple FBGs can be evenly arranged on the nickel-titanium needle core needle body 9. When the puncture needle is bent, the optical fiber will bend, which will change the central wavelength of the grating. The bending amount information can be obtained by analyzing the central wavelength, and the bending amount information is the change information of the needle body shape;

[0061] FPI sensor: When the optical fiber is stretched or compressed, the cavity length of the FPI changes, resulting in a shift in the interference spectrum. An FPI sensor can be arranged at the head 10 of the nickel-titanium needle core. During the puncture process, the Fabry-Perot interferometer sensor can sense the forces exerted on the puncture needle, such as cutting force. These forces will cause strain in the optical fiber, thereby changing the cavity length of the FPI and causing a shift in the spectrum. The information on the needle tip puncture force can be obtained by analyzing the spectrum;

[0062] The ultrasonic probe is fixed at the end of the mechanical arm, and the ultrasonic probe collects data and sends it to the host computer for processing, which is used for real-time 3D imaging during the puncture process; it can provide real-time position information of the puncture needle;

[0063] The signal demodulator demodulates the optical fiber spectrum and ultrasonic signal and transmits them to the host computer for further processing.

[0064] The host computer runs a puncture motion control program, a biological tissue recognition program, an image fusion program, and a trajectory prediction program.

[0065] The puncture motion control program drives the robotic arm and needle insertion mechanism to complete control and adjustment of the needle insertion position, needle insertion angle, puncture speed, needle core length, deflection direction, etc.

[0066] The biological tissue recognition module is responsible for processing the received optical fiber force sensing signal to obtain surgical area information such as cutting force and tissue type during the puncture process.

[0067] The image fusion module aligns and fuses the shape of the flexible needle reconstructed by the fiber optic shape sensor with the three-dimensional medical image reconstructed by the ultrasonic sensor to obtain clear information on the needle body posture, lesion target, and obstacle tissue.

[0068] Specifically, first, data preprocessing is performed under the premise of ensuring data synchronization, including: filtering: filtering the information collected by the optical fiber sensor and the ultrasonic probe to remove noise and interference; calibration: calibrating the optical fiber sensor and the ultrasonic probe to eliminate system errors;

[0069] Secondly, the fiber Bragg grating (FBG) technology and the Frenet-Serret formula are used to obtain the bending degree and angle of the needle at each position during the puncture process, and the arc length integral is used to reconstruct the three-dimensional shape curve of the needle. The target area is scanned using a three-dimensional ultrasonic imaging device to obtain an ultrasonic image of the needle in the body, and the edge detection algorithm is used to identify and extract the contour information of the needle.

[0070] Data registration and shape fusion method: The coordinate system of the fiber optic reconstructed image is the flexible needle base, and the ultrasound image coordinate system is based on the ultrasound probe as a reference. Non-rigid registration is used to correctly align the three-dimensional needle shape data of the fiber optic sensor and the ultrasound image needle data in the same space.

[0071] The tip and inflection point of the needle in the fiber-optic reconstructed image and three-dimensional ultrasound image are selected as registration features; B-spline deformation is used to simulate the deformation between coordinate systems, and the non-rigid transformation is calculated by the gradient descent method to obtain the non-rigid transformation model of translation, rotation, distortion and stretching features; the fiber-optic sensor data is applied to the ultrasound image coordinate system after non-rigid transformation to complete the alignment of the origin position and direction of the two coordinate systems of the fiber-optic image and ultrasound image. The boundary of the needle body is extracted using ultrasound image segmentation methods, such as threshold segmentation and region growth segmentation, and then these boundaries are aligned with the needle body shape reconstructed by the fiber optic. The needle body shape reconstructed by the fiber optic is used as a constraint condition to assist ultrasound image segmentation and improve matching accuracy, so that the shape of the needle body can be accurately superimposed on the ultrasound image.

[0072] Finally, the data fusion method:

[0073] In three-dimensional space, the fiber optic sensor data and the needle position in the ultrasound image are weighted and fused, and the weights are assigned according to the clarity of the image and the reliability of the needle shape data. Using rendering technology in computer graphics, the needle shape reconstructed by the fiber optic sensor is superimposed on the three-dimensional ultrasound image in the form of a virtual object. In this way, the information from the two data sources (the shape of the needle and the ultrasound image) can be combined into a comprehensive visual image.

[0074] The trajectory prediction model inputs measurable parameters related to the puncture motion trajectory, such as needle core length, needle body posture, puncture speed, needle insertion depth, tissue characteristics and type, into the trained neural network model, predicts and visualizes the motion trajectory, and drives the needle insertion mechanism to adjust the bending radius of the puncture needle according to the predicted motion trajectory.

[0075] The collected data are cleaned, and continuous features such as force signal, needle core extension length, puncture speed, and needle insertion depth are normalized or standardized to ensure that their scales are consistent.

[0076] It is planned to construct a 5-layer neural network, select stress signal, needle core length, puncture speed, needle insertion depth, and needle body posture as input, and bend curvature as output. The number of hidden layer neurons is planned to be set to 5, and each layer contains 20-100 neurons. It is planned to select the average variance as the loss function, the Sigmoid function as the activation function of the hidden layer and output layer, and use the Levenberg-Marquardt method combined with the gradient descent method and the Newton iteration method to train the model, and cross-validate and optimize the model with five-fold cross validation.

[0077] The working principle of the present invention is as follows:

[0078] According to the preoperative plan, the robot arm is controlled to adjust the puncture needle to the appropriate insertion point and angle for puncture. During the puncture process, the shape of the fiber-optic-flexible needle body is fused with the ultrasonic medical image to obtain information such as the needle tip posture, lesion target, and obstacle tissue. At the same time, the cutting force, tissue feature type, puncture speed, needle core length and other factors affecting the puncture bending radius are collected in real time during the puncture process. The trajectory is predicted through the trained neural network model and visualized in the reconstructed medical image. The system adjusts the needle insertion strategy based on the predicted trajectory and optimized path, controls the needle insertion mechanism to adjust the length of the flexible needle core extending out of the cannula to change the puncture movement bending radius, and punctures according to the optimized path until the target point.

[0079] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. Flexible puncture needle motion trajectory control prediction system, characterized in that: It includes a mechanical arm, a needle insertion mechanism, a flexible puncture needle, an ultrasonic probe, a signal demodulation instrument and a host computer; The end of the mechanical arm is fixed with the needle insertion mechanism and the ultrasonic probe, and is used to adjust the insertion position and angle of the puncture needle; The flexible puncture needle comprises a flexible needle core with an oblique opening having a variable diameter and a sleeve for providing support; The needle feeding mechanism has three degrees of freedom, which are used for cannula feeding, needle core feeding, and needle core rotation. The flexible needle core is grooved and equipped with an optical fiber sensor. The optical fiber sensor collects needle shape change information and needle tip puncture force information in real time; The ultrasonic probe is used for real-time 3D imaging during the puncture process to obtain the shape information of the flexible puncture needle, the lesion target, and the obstacle tissue information; The signal demodulator demodulates the information collected by the optical fiber sensor and the signal collected by the ultrasonic probe, and transmits them to the host computer for further processing; The host computer aligns and fuses the flexible needle body shape information reconstructed by the optical fiber sensor with the three-dimensional medical image of the flexible puncture needle shape reconstructed by the ultrasonic probe to obtain clear needle body posture information; then the needle body posture information, lesion target, obstacle tissue information, needle tip puncture force information, needle core length, puncture speed, and needle insertion depth are input into the trajectory prediction model, and the trajectory prediction model predicts the motion trajectory of the needle insertion mechanism and visualizes it; control decisions are made based on the visualized predicted trajectory to control the needle insertion mechanism, and the needle insertion mechanism adjusts the length of the flexible needle core extending out of the sleeve to change the puncture motion trajectory, and punctures until the target point.

2. The flexible puncture needle motion trajectory control prediction system according to claim 1, characterized in that: The flexible needle core is a nickel-titanium needle core.

3. The flexible puncture needle motion trajectory control prediction system according to claim 1, characterized in that: The sleeve is a nickel-titanium sleeve, and the flexible needle puncture trajectory can be adjusted by controlling the length of the nickel-titanium needle core extending out of the nickel-titanium sleeve.

4. The flexible puncture needle motion trajectory control prediction system according to claim 1, characterized in that: The host computer includes a puncture motion control module, a biological tissue recognition module, an image fusion module and a trajectory prediction module; the biological tissue recognition module and the image fusion module communicate with the trajectory prediction module, and the trajectory prediction module communicates with the puncture motion control module.

5. The flexible puncture needle motion trajectory control prediction system according to claim 4, characterized in that: The puncture motion control module drives the mechanical arm and the needle insertion mechanism to adjust the needle insertion position, needle insertion angle, puncture speed, needle core length, and deflection direction.

6. The flexible puncture needle motion trajectory control prediction system according to claim 4, characterized in that: The biological tissue recognition module is responsible for processing the needle tip puncture force information collected by the optical fiber sensor and obtaining the surgical area information during the puncture process.

7. The flexible puncture needle motion trajectory control prediction system according to claim 4, characterized in that: The image fusion module registers and fuses the three-dimensional medical images of the flexible needle body shape reconstructed by the optical fiber sensor and the flexible puncture needle shape reconstructed by the ultrasonic probe to obtain a clear needle body posture.

8. The flexible puncture needle motion trajectory control prediction system according to claim 4, characterized in that: The trajectory prediction module receives measurable parameters related to the puncture motion trajectory, predicts and visualizes the motion trajectory, and drives the needle insertion mechanism to adjust the bending radius of the puncture needle according to the optimized trajectory.

9. The trajectory control prediction method of the flexible puncture needle motion trajectory control prediction system according to any one of claims 1 to 8, characterized in that: as follows: In advance, the needle tip puncture force information and needle body bending deflection under different puncture speeds, feeding depths, needle core lengths, and bionic tissue types are collected and analyzed through puncture experiments to establish a sample data set; based on the observable information of puncture speed, needle core length, and needle tip puncture force, a flexible needle motion trajectory prediction module is generated through artificial intelligence technology; During the puncture implementation phase, the optical fiber sensor collects information on needle shape changes, needle tip puncture force, puncture speed, needle core length, and other information in real time, and inputs them into the motion trajectory prediction module to generate a predicted trajectory. The ultrasound probe performs real-time 3D imaging during the puncture process to obtain the shape information of the flexible puncture needle, the target lesion, and the obstacle tissue information; The signal demodulator receives the information collected by the optical fiber sensor and the ultrasonic probe, demodulates the signal, and transmits it to the host computer for further processing; The host computer aligns and fuses the shape of the flexible needle reconstructed by the optical fiber sensor with the shape information of the flexible puncture needle reconstructed by the ultrasonic probe to obtain clear needle body posture information; the predicted trajectory generated by the trajectory prediction module is visualized in the aligned and fused 3D image, and a control decision is made based on the visualized predicted trajectory. The needle insertion mechanism is then controlled to adjust the length of the flexible needle core extending out of the sheath to change the bending trajectory of the puncture movement and puncture until the target point is reached.

10. The trajectory control prediction method of the flexible puncture needle motion trajectory control prediction system according to claim 9, characterized in that: The method for registering and fusing the shape of the flexible needle body reconstructed by the optical fiber sensor and the shape information of the flexible puncture needle reconstructed by the ultrasound probe is as follows: Data preprocessing; Obtain the bending degree and angle of the needle at each position during the puncture process, and reconstruct the three-dimensional shape curve of the needle by integrating the arc length; obtain the ultrasound image of the needle in the body, and use the edge detection algorithm to identify and extract the contour information of the needle; Non-rigid registration is used to correctly align the three-dimensional needle shape data of the optical fiber sensor and the needle data of the ultrasound image in the same space; The tip and inflection point of the needle in the fiber-optic reconstructed image and the three-dimensional ultrasound image are selected as registration features; the deformation between coordinate systems is simulated by B-spline deformation, and the non-rigid transformation is calculated by the gradient descent method to obtain a non-rigid transformation model of translation, rotation, distortion and stretching features; the fiber-optic sensor data is applied to the ultrasound image coordinate system after non-rigid transformation to complete the alignment of the origin position and direction of the two coordinate systems of the fiber-optic image and the ultrasound image; the boundary of the needle body is extracted by the ultrasound image segmentation method, and then the boundary is aligned with the needle body shape reconstructed by the fiber optic; the needle body shape reconstructed by the fiber optic is used as a constraint condition to assist the ultrasound image segmentation, so that the shape of the needle body can be accurately superimposed on the ultrasound image; In three-dimensional space, the fiber optic sensor data and the needle position in the ultrasound image are weighted and fused, and the weights are assigned according to the clarity of the image and the reliability of the needle shape data. Using rendering technology in computer graphics, the needle shape reconstructed by the fiber optic sensor is superimposed on the three-dimensional ultrasound image in the form of a virtual object to obtain a comprehensive visual image.

Images