Needle body tracking type intelligent acupuncture system

Abstract

The invention belongs to the technical field of acupuncture and moxibustion, and discloses a needle body tracking type intelligent acupuncture and moxibustion system which comprises a data collecting and processing module, a needle body positioning module, a 3D model registration and dynamic updating module, a real-time rendering display module and an interaction module. Three-dimensional coordinates, puncture depth and angle information of a needle body are obtained in real time with high precision through a multi-modal composite positioning technology, and real-time visual association of the needle body with key anatomical structures such as bones, blood vessels and nerves in the in-vivo penetrating process is achieved by combining a dynamically-registered 3D anatomical model and the actual body position of a patient. The needle inserting precision and the operation safety are obviously improved; meanwhile, through multi-plane fusion view presentation, a dynamic breathing compensation mechanism and a layered rendering updating strategy, it is ensured that a physician can visually master the real-time space relation between a needle body and surrounding tissue, and important structures are effectively prevented from being accidentally injured; and diversified touch interaction and instant operation feedback are supplemented, so that the operation complexity is greatly reduced.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of acupuncture and moxibustion, and particularly relates to an intelligent acupuncture and moxibustion system with needle body tracking. BACKGROUND

[0002] Acupuncture and moxibustion is a treasure in traditional Chinese medicine therapy, which stimulates specific acupoints of the human body with a filiform needle or combines the warm action of moxibustion, dredges the meridians, regulates the blood and qi, balances yin and yang, stimulates the self-healing ability of the body, and achieves the purpose of preventing and treating diseases.

[0003] As an important part of traditional Chinese medicine, acupuncture and moxibustion has a wide range of clinical applications in pain management, chronic disease conditioning and many other fields. Its treatment effect is closely related to the accuracy of needle insertion. Precise understanding of the relationship between the needle body position and the surrounding anatomical structure is the key to improving the treatment effect and reducing the risk. With the development of modern medical technology, the demand for the accuracy, safety and visualization of acupuncture and moxibustion operation is increasing. At present, the operation of acupuncture and moxibustion mainly depends on the anatomical knowledge, clinical experience and surface markers of doctors for positioning. Although some auxiliary technologies such as ultrasound guidance and X-ray fluoroscopy are applied in clinical practice, the existing technology often relies on a single positioning method and cannot comprehensively and accurately obtain the three-dimensional spatial position, needle insertion depth and angle information of the needle body. In terms of anatomical structure presentation, it is mostly static preoperative images or isolated real-time images, which cannot realize the effective correlation between the preoperative anatomical model and the real-time body position of the patient and the dynamic position of the needle body. As a result, doctors cannot intuitively understand the relative relationship between the needle body and the surrounding key structures such as blood vessels, nerves and bones during the process of the needle body in the body. These deficiencies to some extent affect the accuracy and safety of acupuncture and moxibustion operation and limit the further application of acupuncture and moxibustion technology in complex clinical scenarios. Therefore, it is necessary to improve it. SUMMARY

[0004] The purpose of the present application is to provide an intelligent acupuncture and moxibustion system with needle body tracking to solve the problems raised in the background art.

[0005] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme: an intelligent acupuncture and moxibustion system with needle body tracking, comprising the following modules: a data acquisition and processing module, a needle body positioning module, a 3D model registration and dynamic updating module, a real-time rendering and display module, and an interaction module.

[0006] The data acquisition and processing module is responsible for acquiring medical image data of CT and MRI of the patient, and converting two-dimensional tomographic images into a three-dimensional anatomical model containing key structures such as bones, blood vessels and nerves through a pre-processing process.

[0007] The needle body positioning module: through multi-modal composite positioning technology, real-time capture of three-dimensional coordinates, puncture depth and angle information of the needle body, through data fusion and precision calibration technology, ensure the accuracy and stability of the needle body position tracking, provide data support for the association of needle body and anatomical structure;

[0008] The 3D model registration and dynamic updating module: through body surface scanning and anatomical landmark matching, the constructed 3D anatomical model is accurately aligned with the actual body position of the patient, and according to the movement of the needle body and the change of the patient's body position, the position relationship of each structure in the model is dynamically adjusted, so that the model always keeps consistent with the real state of the patient, and accurately reflects the relative position of the needle body in the body;

[0009] The real-time rendering display module: for fusing the real-time position information of the needle body with the 3D anatomical model, generating coronal, sagittal and transverse views, and presenting through the device to realize the visualization of the anatomical structure around the needle body;

[0010] The interaction module: equipped with diversified interaction modes, touch operation supports single-click for selecting structure, double-finger zoom for adjusting model size, and sliding for rotating view; the system interface adopts iconized design, the touch screen button has vibration feedback, and the interaction operation also has feedback, ensuring that the physician can accurately judge whether the operation is effective.

[0011] Preferably, the data acquisition and processing module includes a positioning unit composed of an infrared positioning component, an ultrasonic detection component and an electromagnetic induction component. The infrared positioning component includes four infrared cameras distributed around the treatment bed with a sampling frequency of 30Hz. By identifying the micro active infrared light marking point at the tail of the needle body, the coordinate of the needle tip is calculated. The ultrasonic detection component adopts a 7.5MHz linear array probe fixed on a rotatable mechanical arm to generate real-time cross-sectional ultrasonic images of the needle body, extract the needle axis through image edge detection algorithm, and determine the puncture angle. The electromagnetic induction component includes a transmitting coil generating uniform alternating magnetic field and a micro receiving coil implanted in the needle body, which calculates the depth of the needle body through the change of magnetic field strength. Three kinds of technical data are collected synchronously, and a complete set of positioning data is generated every 10 milliseconds.

[0012] Preferably, the data acquisition processing module comprises a positioning data fusion unit, which firstly aligns the time stamps of infrared, ultrasonic and electromagnetic data to eliminate the synchronization error between devices by establishing a multi-source data fusion algorithm framework; adopts an extended Kalman filter algorithm to perform weighted fusion on the three types of data, wherein the weight of infrared data accounts for 40% (responsible for plane coordinates), the weight of ultrasonic data accounts for 30% (responsible for angle calibration), and the weight of electromagnetic data accounts for 30% (responsible for depth calculation); the stability of each data source is monitored in real time during the fusion process, and when the data fluctuation of a certain device exceeds a threshold value, the weight thereof is automatically reduced to ensure the reliability of the fusion result; the output data after fusion includes the three-dimensional coordinates of the needle tip, the angle between the needle and the horizontal plane, and the puncture depth, and the display interface is updated every 50 milliseconds.

[0013] Preferably, the 3D model registration and dynamic updating module comprises a patient position registration unit, which adopts a registration method combining optical surface scanning and anatomical landmark points, marks six anatomical landmark points on the surface of the patient's skin, including the suprasternal fossa, the navel, the bilateral anterior superior iliac spines and the bilateral shoulder peaks, uses a laser scanner to obtain the point cloud data of the surface of the upper body of the patient, matches the point cloud data with the surface grid of the 3D model, calculates the optimal transformation matrix by using an improved iterative closest point algorithm, and makes the model coincide with the actual position of the patient; the registration process allows manual adjustment by the doctor, aligns the landmark points on the model with the surface markers on the patient by dragging, and automatically records the transformation parameters after adjustment as the reference coordinate system for subsequent positioning.

[0014] Preferably, the 3D model registration and dynamic updating module comprises a needle position mapping unit, which converts the real-time coordinates output by the needle positioning module to the 3D anatomical model coordinate system through a rigid transformation matrix by establishing a double-coordinate system conversion mechanism; the transformation process considers the body micro-motion caused by the patient's breathing, adjusts the organ position in real time according to the chest and abdomen fluctuation data collected by the breathing sensor by setting an elastic deformation area in the model, so that the relative position of the needle and the surrounding structure is always accurate; the mapping result is superimposed on the 3D model in the form of dynamic lines, the line color changes with the depth, and the path of the needle in the tissue is clearly displayed.

[0015] Preferably, the 3D model registration and dynamic updating module comprises a model dynamic updating unit, which adopts a hierarchical rendering updating strategy to divide the 3D model into a static layer (skeleton), a semi-dynamic layer (muscle, blood vessel), and a full-dynamic layer (needle body, skin), and only the full-dynamic layer is updated when the patient's body position changes less than 0.5 mm, the semi-dynamic layer and the full-dynamic layer are synchronously updated when the change is between 0.5-2 mm, and the complete model is updated when the change exceeds 2 mm, and the updating process adopts an incremental rendering technology to only re-render the changed area, so that the picture refresh rate is kept above 25 frames / second, the lagging effect is avoided, the operation is affected, the needle body movement trajectory is recorded, and the history record of the puncture path is generated, so that the needle body position change in the past 30 seconds can be viewed at any time.

[0016] Preferably, the real-time rendering display module comprises a multi-modal image fusion unit for the fusion display of the 3D anatomical model and the real-time ultrasound image, and a pixel-level-based fusion algorithm is adopted to superimpose the anatomical structure contour in the model on the ultrasound image in a semi-transparent line, wherein the skeleton is displayed as a white solid line, the blood vessel is displayed as a red dashed line, and the nerve is displayed as a yellow dotted line; the fusion transparency can be adjusted through a sliding bar, and the doctor can switch the display mode according to the need: pure model mode (only display the 3D model and the needle body), pure ultrasound mode (only display the real-time ultrasound image), and fusion mode (both are superimposed).

[0017] Preferably, the real-time rendering display module comprises a multi-plane view generation unit capable of automatically generating three orthogonal plane views related to the needle body: a transverse plane (horizontal layer), a sagittal plane (front-back layer), and a coronal plane (left-right layer) with the needle body tip as the center; the display range of each view can be zoomed in or out through the mouse wheel, the needle body position is labeled in real time in the view, the puncture direction is indicated by an arrow, and the anatomical structure name within a range of 2 cm around is displayed; when the needle body moves, the three planes are automatically updated according to the tip position, so that the latest anatomical relationship is always displayed.

[0018] Preferably, the real-time rendering display module comprises a display device adaptation unit adopting a dual-mode wireless transmission architecture, short-distance data is transmitted through Bluetooth technology, and large-capacity data is transmitted through Wi-Fi 6; the type of the connected device can be automatically detected, the resolution and color parameters can be adjusted, and the display effect on different screens can be ensured to be consistent; a four-split screen mode is adopted on the display screen, the left side is the 3D model, and the right side is the three plane views; the tablet computer and the mobile phone adopt a vertical screen layout, the real-time image is above, and the operation button is below.

[0019] The beneficial effects of the application are as follows:

[0020] By acquiring real-time, high-precision three-dimensional coordinates, puncture depth, and angle information of the needle through multimodal composite positioning technology, and combining it with dynamically registered 3D anatomical models and the patient's actual position, the system achieves real-time visual association between the needle and key anatomical structures such as bones, blood vessels, and nerves during the needle's passage through the body, significantly improving needle insertion accuracy and operational safety. Simultaneously, through multi-planar fusion view presentation, dynamic respiratory compensation mechanisms, and layered rendering update strategies, the system ensures that physicians can intuitively grasp the real-time spatial relationship between the needle and surrounding tissues, effectively avoiding accidental damage to important structures. Furthermore, the system incorporates diverse touch-screen interactions and instant operational feedback, greatly reducing operational complexity and enhancing the reliability and efficiency of clinical procedures, thereby promoting the in-depth application of acupuncture technology in modern precision medicine. Attached Figure Description

[0021] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0022] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, this embodiment of the invention provides a needle tracking intelligent acupuncture system, including the following modules: a data acquisition and processing module, a needle positioning module, a 3D model registration and dynamic update module, a real-time rendering and display module, and an interaction module;

[0024] Data acquisition and processing module: responsible for acquiring the patient's CT and MRI medical image data, and converting the two-dimensional tomographic images into three-dimensional anatomical models containing key structures such as bones, blood vessels, and nerves through a preprocessing process;

[0025] Needle positioning module: Through multimodal composite positioning technology, it captures the three-dimensional coordinates, puncture depth and angle information of the needle in real time. Through data fusion and precision calibration technology, it ensures the accuracy and stability of needle position tracking and provides data support for the association between the needle and anatomical structures.

[0026] 3D model registration and dynamic update module: Through body surface scanning and anatomical landmark matching, the constructed 3D anatomical model is precisely aligned with the patient's actual position. At the same time, the positional relationship of each structure in the model is dynamically adjusted according to the movement of the needle and changes in the patient's position to ensure that the model always remains consistent with the patient's real state and accurately reflects the relative position of the needle in the body.

[0027] Real-time rendering display module: used to fuse the real-time needle position information with the 3D anatomical model, generate multi-planar views such as coronal, sagittal, transverse, etc., and present through the device to realize the visualization of the anatomical structure around the needle;

[0028] Interaction module: equipped with diversified interaction methods, touch operation supports single-click for selecting structure, double-finger zoom for adjusting model size, and sliding for rotating view; the system interface adopts iconized design, the touch screen button has vibration feedback, and the interactive operation also has feedback to ensure that the physician can accurately judge whether the operation is effective.

[0029] Through the data acquisition and processing module, a 3D model containing key anatomical structures is constructed to provide a basis for accurate positioning; with the multi-modal composite positioning and data fusion technology of the needle positioning module, the three-dimensional coordinates, depth and angle of the needle are accurately and real-time tracked; through the 3D model registration and dynamic updating module, the model is accurately aligned with the actual body position of the patient and dynamically adjusted with the movement of the needle and the change of the body position, ensuring the accurate relative position of the needle and the surrounding structure; the real-time rendering display module is used to generate multi-planar views to realize the visualization of the anatomical structure around the needle; combined with the diversified operation methods and feedback mechanism of the interaction module, the convenience and accuracy of the physician's operation are improved, and the overall accuracy, safety and intuitiveness of the acupuncture operation are effectively improved, better meeting the clinical demand for precision and visualization of acupuncture operation.

[0030] Among them, the data acquisition and processing module includes a positioning unit, which is composed of an infrared positioning component, an ultrasonic detection component and an electromagnetic induction component. The infrared positioning component includes four infrared cameras distributed around the treatment bed, with a sampling frequency of 30 Hz. By recognizing the micro active infrared light marking point at the tail of the needle, the coordinates of the needle tip are calculated. The ultrasonic detection component uses a 7.5 MHz linear array probe fixed on a rotatable mechanical arm to generate real-time cross-sectional ultrasonic images of the needle, extract the needle axis through image edge detection algorithm, and determine the puncture angle. The electromagnetic induction component includes a transmitting coil that generates a uniform alternating magnetic field and a micro receiving coil implanted inside the needle. The depth of the needle is calculated by the change of magnetic field strength. Three kinds of technical data are collected synchronously, and a complete set of positioning data is generated every 10 milliseconds.

[0031] Through the cooperation of infrared, ultrasonic and electromagnetic positioning technologies, the positioning advantages of each technology in different dimensions can be exerted, and the overall positioning comprehensiveness and accuracy can be improved through multi-source data complementation.

[0032] The data acquisition and processing module comprises a positioning data fusion unit. The positioning data fusion unit establishes a multi-source data fusion algorithm framework, first aligns the time stamps of infrared, ultrasonic and electromagnetic data, and eliminates the synchronization error between devices. The three types of data are weighted and fused by using an extended Kalman filter algorithm, wherein the weight of the infrared data accounts for 40% (responsible for plane coordinates), the weight of the ultrasonic data accounts for 30% (responsible for angle calibration), and the weight of the electromagnetic data accounts for 30% (responsible for depth calculation). During the fusion process, the stability of each data source is monitored in real time. When the data fluctuation of a certain device exceeds the threshold, the weight of the device is automatically reduced to ensure the reliability of the fusion result. The output data after fusion includes the three-dimensional coordinates of the needle tip, the angle between the needle and the horizontal plane, and the puncture depth. The display interface is updated every 50 milliseconds.

[0033] The weighted fusion algorithm of the extended Kalman filter realizes accurate integration of data, and can dynamically adjust the weight according to the real-time stability of each data source, effectively cope with the situation of single device data fluctuation, and ensure that the output needle coordinates, angle and depth information are always stable and reliable.

[0034] The 3D model registration and dynamic updating module comprises a patient position registration unit. The patient position registration unit adopts a registration method combining optical surface scanning and anatomical landmark points. Six anatomical landmark points are marked on the surface of the patient's skin, including the suprasternal fossa, the navel, the bilateral anterior superior iliac spines and the bilateral shoulder peaks. The point cloud data of the patient's upper body surface is obtained by using a laser scanner, and the point cloud data is matched with the surface grid of the 3D model. An improved iterative closest point algorithm is used to calculate the optimal transformation matrix, so that the model coincides with the actual position of the patient. During the registration process, the doctor is allowed to manually adjust the landmark points by aligning the landmark points on the model with the patient's surface markers. After the adjustment is completed, the system automatically records the transformation parameters as the reference coordinate system for subsequent positioning.

[0035] Allowing the doctor to manually adjust the landmark points further optimizes the efficiency of automatic registration and the flexibility of manual adjustment, ensuring that the 3D model highly coincides with the actual position of the patient.

[0036] The 3D model registration and dynamic updating module comprises a needle position mapping unit. The needle position mapping unit converts the real-time coordinates output by the needle positioning module to the 3D anatomical model coordinate system through a rigid transformation matrix by establishing a double-coordinate system conversion mechanism. During the transformation process, the body micro-motion caused by the patient's breathing is considered. By setting an elastic deformation area in the model, the organ position is adjusted in real time according to the chest and abdomen fluctuation data collected by the breathing sensor, so that the relative position of the needle and the surrounding structure is always accurate. The mapping result is superimposed on the 3D model in the form of dynamic lines. The line color changes with the depth, clearly showing the path of the needle in the tissue.

[0037] By setting the elastic deformation area and combining with the respiratory sensor data to adjust the organ position in real time, the influence of patient's body movement caused by breathing on positioning is effectively offset.

[0038] The 3D model registration and dynamic updating module includes a model dynamic updating unit, which divides the 3D model into static layers (skeleton), semi-dynamic layers (muscles, blood vessels), and full-dynamic layers (needle body, skin) using a hierarchical rendering update strategy. When the patient's body position changes by less than 0.5 mm, only the full-dynamic layer is updated. When the change is between 0.5-2 mm, the semi-dynamic layer and the full-dynamic layer are updated simultaneously. When the change exceeds 2 mm, the complete model is updated, and the update process uses incremental rendering technology to only re-render the changed area, keeping the frame refresh rate above 25 frames / second to avoid lag and affect operation. At the same time, the needle movement trajectory is recorded to generate a historical record of the puncture path, which can be reviewed at any time to check the needle position changes in the past 30 seconds.

[0039] The hierarchical rendering update strategy updates according to the degree of patient's body position change, significantly reducing the system's computational load, and the incremental rendering technology ensures stable frame refresh rate, avoiding lag problems.

[0040] The real-time rendering display module includes a multi-modal image fusion unit for the fusion display of the 3D anatomical model and real-time ultrasound images. A pixel-level fusion algorithm is used to superimpose the anatomical structure outlines in the model as semi-transparent lines on the ultrasound image, with the skeleton displayed as a white solid line, the blood vessels as a red dashed line, and the nerves as a yellow dotted line. The fusion transparency can be adjusted through a sliding bar, and the physician can switch the display mode as needed: pure model mode (only display 3D model and needle body), pure ultrasound mode (only display real-time ultrasound image), and fusion mode (both superimposed).

[0041] Different colors and line types are used to clearly distinguish different types of anatomical structures, and the fusion transparency can be adjusted and support three display mode switching, which can meet the observation needs of physicians for different anatomical information and improve the information acquisition efficiency during operation through flexible display control.

[0042] The real-time rendering display module includes a multi-plane view generation unit that can automatically generate three orthogonal plane views related to the needle body: a transverse plane (horizontal section), a sagittal plane (front-back section), and a coronal plane (left-right section) centered on the needle tip. The display range of each view can be scaled by the mouse wheel, and the needle position is labeled in real time in the view, with an arrow indicating the puncture direction and displaying the names of the anatomical structures within a 2 cm range. When the needle moves, the three planes automatically update the tip position to ensure that the latest anatomical relationship is always displayed.

[0043] The automatically generated cross-section, sagittal plane and coronal plane are centered on the needle tip, fully covering the key anatomical areas around the needle, allowing the practitioner to accurately grasp the spatial relationship between the needle and adjacent tissues, and improving the accuracy of acupuncture.

[0044] Among them, the real-time rendering display module includes a display device adaptation unit, which adopts a dual-mode wireless transmission architecture, short-distance data is transmitted through Bluetooth technology, and large-capacity data is transmitted through Wi-Fi 6; It can automatically detect the type of connected device, adjust the resolution and color parameters, and ensure consistent display effects on different screens; Four-screen mode is adopted on the display screen, 3D model on the left and three plane views on the right; Tablet computer and mobile phone adopt portrait screen layout, real-time image on top and operation button on bottom.

[0045] The positioning parameters and control instructions are transmitted through Bluetooth technology, and the 3D model and real-time image are transmitted through Wi-Fi 6, taking into account transmission efficiency and stability.

[0046] It should be noted that in this text, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual such relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device.

[0047] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A needle tracking intelligent acupuncture system, characterized in that, It comprises the following modules: data acquisition and processing module, needle positioning module, 3D model registration and dynamic updating module, real-time rendering display module, interactive module; The data acquisition and processing module is responsible for obtaining CT and MRI medical image data of a patient, and converts two-dimensional tomographic images into a three-dimensional anatomical model containing key structures such as bones, blood vessels and nerves through a preprocessing process; The needle positioning module captures the three-dimensional coordinates, puncture depth and angle information of the needle body in real time through multi-modal composite positioning technology, and ensures the accuracy and stability of needle position tracking through data fusion and precision calibration technology, providing data support for the association of the needle body and the anatomical structure; The 3D model registration and dynamic updating module precisely aligns the constructed 3D anatomical model with the actual body position of the patient through body surface scanning and anatomical landmark matching, and dynamically adjusts the positional relationship of each structure in the model according to the movement of the needle body and the change of the patient's body position, ensuring that the model always remains consistent with the actual state of the patient and accurately reflects the relative position of the needle body in the body; The real-time rendering display module is used to fuse the real-time position information of the needle body with the 3D anatomical model to generate multiple plane views such as coronal, sagittal and transverse views, and presents them through the device to realize the visualization of the anatomical structure around the needle body; The interactive module is equipped with diversified interaction methods, and touch operation supports single-click for selecting structures, double-finger zoom for adjusting model size, and sliding for rotating views; The system interface adopts iconized design, and the touch screen buttons have vibration feedback, and the interactive operations also have feedback, ensuring that the physician can accurately determine whether the operation is effective.

2. The needle tracking intelligent acupuncture system according to claim 1, wherein: The data acquisition and processing module comprises a positioning unit composed of an infrared positioning component, an ultrasonic detection component and an electromagnetic induction component. The infrared positioning component includes four infrared cameras distributed around the treatment bed with a sampling frequency of 30 Hz. It calculates the needle tip coordinates by recognizing the micro active infrared light marker point at the tail of the needle body. The ultrasonic detection component uses a 7.5 MHz linear array probe fixed on a rotatable mechanical arm to generate real-time cross-sectional ultrasonic images of the needle body, extract the needle axis through image edge detection algorithm, and determine the puncture angle. The electromagnetic induction component includes a transmitting coil generating a uniform alternating magnetic field and a micro receiving coil implanted in the needle body, which calculates the needle depth through the change of magnetic field strength. Three kinds of technical data are collected synchronously, and a complete set of positioning data is generated every 10 milliseconds.

3. The needle tracking intelligent acupuncture system according to claim 1, wherein: The data acquisition processing module comprises a positioning data fusion unit, which firstly aligns the time stamps of infrared, ultrasonic and electromagnetic data by establishing a multi-source data fusion algorithm framework, and eliminates the synchronization error between devices; the three types of data are weighted and fused by using an extended Kalman filtering algorithm, wherein the weight of infrared data accounts for 40% (responsible for plane coordinates), the weight of ultrasonic data accounts for 30% (responsible for angle calibration), and the weight of electromagnetic data accounts for 30% (responsible for depth calculation); the stability of each data source is monitored in real time during the fusion process, and when the data fluctuation of a certain device exceeds the threshold, the weight thereof is automatically reduced to ensure the reliability of the fusion result; the output data after fusion includes the three-dimensional coordinates of the needle tip, the angle between the needle and the horizontal plane, and the puncture depth, and the display interface is updated every 50 milliseconds.

4. The needle tracking intelligent acupuncture system according to claim 1, wherein: The 3D model registration and dynamic updating module comprises a patient position registration unit, which adopts a registration mode combining optical surface scanning and anatomical landmark points, marks six anatomical landmark points on the patient's skin surface, including the suprasternal fossa, the navel, the bilateral anterior superior iliac spines and the bilateral shoulder peaks, acquires the surface point cloud data of the patient's upper body by using a laser scanner, matches the point cloud data with the surface grid of the 3D model, calculates the optimal transformation matrix by using an improved iterative closest point algorithm, and makes the model coincide with the actual position of the patient; the registration process allows manual adjustment by the doctor, and the landmark points on the model are aligned with the patient's surface markers by dragging, and the system automatically records the transformation parameters after adjustment as the reference coordinate system for subsequent positioning.

5. The needle tracking intelligent acupuncture system according to claim 1, wherein: The 3D model registration and dynamic updating module comprises a needle position mapping unit, which converts the real-time coordinates output by the needle positioning module to the 3D anatomical model coordinate system through a rigid transformation matrix by establishing a double-coordinate system conversion mechanism; the body micro-motion caused by patient's breathing is considered during the conversion process, the organ position is adjusted in real time according to the chest and abdomen fluctuation data collected by the breathing sensor by setting an elastic deformation area in the model, so that the relative position of the needle and the surrounding structure always remains accurate; the mapping result is superimposed on the 3D model in the form of dynamic lines, and the line color changes with the depth, clearly showing the running path of the needle in the tissue.

6. The needle tracking intelligent acupuncture system according to claim 1, wherein: The 3D model registration and dynamic updating module comprises a model dynamic updating unit, which adopts a hierarchical rendering updating strategy, divides the 3D model into static layer (skeleton), semi-dynamic layer (muscle, blood vessel) and full-dynamic layer (needle, skin), and only updates the full-dynamic layer when the patient's position changes less than 0.5 mm; when the change is between 0.5-2 mm, the semi-dynamic layer and the full-dynamic layer are updated simultaneously; when the change exceeds 2 mm, the complete model is updated, the update process adopts incremental rendering technology, only re-renders the changed area, so that the frame refresh rate is maintained above 25 frames / second, avoiding the influence of lag on operation, and recording the needle movement trajectory to generate a history record of the puncture path, which can be reviewed at any time to check the needle position change in the past 30 seconds.

7. The needle tracking intelligent acupuncture system according to claim 1, wherein: The real-time rendering display module includes a multi-modal image fusion unit, which is used for the fusion display of a 3D anatomical model and a real-time ultrasound image. A pixel-level fusion algorithm is adopted to superimpose the anatomical structure profile in the model on the ultrasound image in semi-transparent lines, wherein the skeleton is displayed as a white solid line, the blood vessel is displayed as a red dotted line, and the nerve is displayed as a yellow dotted line. The fusion transparency can be adjusted by a sliding bar, and the doctor can switch the display mode according to the need: pure model mode (only 3D model and needle body are displayed), pure ultrasound mode (only real-time ultrasound image is displayed), and fusion mode (both are superimposed).

8. The needle tracking intelligent acupuncture system according to claim 1, wherein: The real-time rendering display module includes a multi-plane view generation unit, which can automatically generate three orthogonal plane views related to the needle body: a transverse plane (horizontal layer), a sagittal plane (front and back layer), and a coronal plane (left and right layer) with the needle body tip as the center. The display range of each view can be zoomed in or out by a mouse wheel, the needle body position is labeled in real time in the view, the puncture direction is indicated by an arrow, and the anatomical structure name within a range of 2 cm around the needle body is displayed. When the needle body moves, the three planes automatically update the tip position to ensure that the latest anatomical relationship is always displayed.

9. The needle tracking intelligent acupuncture system according to claim 1, wherein: The real-time rendering display module includes a display device adaptation unit, which adopts a dual-mode wireless transmission architecture. Short-distance data is transmitted through Bluetooth technology, and large-capacity data is transmitted through Wi-Fi 6. The device type connected can be automatically detected, and the resolution and color parameters can be adjusted to ensure consistent display effects on different screens. A four-split screen mode is adopted on the display screen, with a 3D model on the left and three plane views on the right. A tablet computer and a mobile phone adopt a vertical screen layout, with a real-time image on the top and operation buttons on the bottom.

Images