VR meridian and acupoint three-dimensional digital human model construction method and application

Abstract

The invention relates to the field of meridian and acupoint digital human model construction, in particular to a VR meridian and acupoint three-dimensional digital human model construction method, which comprises the following steps: step 1, multi-dimensional high-precision data acquisition, and step 2, next-generation digital modeling. Step 3, hierarchical construction of a meridian system; step 4, VR scene body fusion; step 5, interaction function integration; through multi-dimensional high-precision data acquisition, next-generation digital modeling, channel and collateral system layered construction, VR scene body fusion and interaction function integration, a highly accurate, vivid and highly interactive three-dimensional digital human model is constructed. The model can be widely applied to medical education, clinical simulation and scientific research analysis, meanwhile, through dynamic data capture and fusion and cross-platform development and efficient rendering, the practicability, immersion and application range of the model are remarkably improved, and an efficient, accurate and safe solution is provided for medical education, clinical practice and scientific research analysis.

Description

Technical Field

[0001] The present invention belongs to the field of meridian and acupoint digital human model construction, and specifically relates to a VR meridian and acupoint three-dimensional digital human model construction method and application. Background Art

[0002] With the development of virtual reality (VR) technology and computer graphics, digital human models have shown tremendous potential in a variety of fields, including medical education, clinical simulation, and scientific research analysis. In particular, in the research and teaching of Traditional Chinese Medicine meridians and acupoints, traditional two-dimensional images and text descriptions cannot fully convey the complex anatomical structures and precise locations of acupuncture points, limiting learning and understanding.

[0003] Traditional meridian and acupoint models often rely on materials like plastic and glass to simulate surface structures, or use two-dimensional computer models to display anatomical relationships. While some studies have attempted to incorporate three-dimensional reconstruction technology (such as the VOXEL-MAN platform) to locate acupoints, these models still suffer from issues such as insufficient model accuracy, a single layer structure, and a lack of dynamic interaction. In recent years, digital human technology and virtual reality (VR) have been gradually applied to medical education. However, existing solutions often rely on direct modeling based on static cross-sectional data, making it difficult to faithfully reproduce the anatomical details and dynamic characteristics of the human meridian system and lacking an immersive user experience. Therefore, addressing these technical issues and shortcomings is a pressing need. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects described in the background technology, thereby realizing a method for constructing a VR meridian and acupoint three-dimensional digital human model, so as to solve the problems in the prior art of direct modeling of static tomographic data that are difficult to truly restore the anatomical details and dynamic characteristics of the human meridian system and lack of immersive operation experience.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution of the present invention is: A method for constructing a VR meridian and acupoint three-dimensional digital human model comprises the following steps: Step 1: Multi-dimensional high-precision data acquisition: Based on real human plastination specimens, a CT scanner is used to collect tomographic data of human meridians and acupoints, and simultaneously obtain anatomical structure data in the horizontal, sagittal, and coronal directions.

[0006] Step 2: Next-generation digital modeling: Based on the data from step 1, a three-dimensional model is constructed using next-generation modeling technology.

[0007] Step 3: The meridian system is constructed in layers. Based on the meridian theory of traditional Chinese medicine, the meridians, collaterals, tendons and skin parts are modeled in layers, and the meridian paths are defined using parameterized curve equations.

[0008] Step 4: VR view volume fusion, establishing a parallel binocular asymmetric view volume model.

[0009] Step 5: Integrate interactive functions, implement real-time multi-gesture recognition through the Leap Motion sensor, and define response functions for gesture control model rotation, scaling, and translation.

[0010] In the above-mentioned VR meridian and acupoint three-dimensional digital human model construction method, in step 1, the CT scanner collects the tomographic data of the human meridian and acupoints with an accuracy of layer thickness ≤ 0.5 mm, and the resolution of the anatomical structure data in the horizontal, sagittal and coronal directions is ≥ 2048×2048 pixels.

[0011] In the above-mentioned VR meridian and acupoint three-dimensional digital human model construction method, the number of faces of a single model of the three-dimensional model in step 2 is ≥ 6000, the texture map uses a PBR material map of 4096×4096 pixels, and the surface details are enhanced by normal mapping and displacement mapping, and the model error rate is ≤ 0.1%.

[0012] In the above-mentioned VR meridian and acupoint 3D digital human model construction method, the parameterized curve equation is used in step 3 to define the meridian path, and the path equation is: Where, is the Bessel basis function, To control the point coordinates, ensure that the path is smooth and conforms to anatomical standards.

[0013] In the above-mentioned VR meridian acupoint 3D digital human model construction method, the visual body parameters in step 4 can be dynamically adjusted according to the user's pupil distance. d Distance from viewpoint s The relationship is: .

[0014] In the above-mentioned VR meridian and acupoint three-dimensional digital human model construction method, in step one, data acquisition also includes dynamic capture of muscle and joint morphology in motion, using optical marker tracking technology with a sampling frequency of ≥120Hz, and using the fusion of dynamic data and static model through a skeleton skinning algorithm to capture subtle changes in the body during motion.

[0015] In the above-mentioned VR meridian and acupoint three-dimensional digital human model construction method, in step 2, the model topology optimization adopts the Quadrilateral Edge Collapse algorithm to ensure that the edge collapse error threshold is ≤0.01mm, and the three-dimensional reconstruction of the tomographic data is achieved based on the MarchingCubes algorithm.

[0016] In the above-mentioned VR meridian acupoint 3D digital human model construction method, in step 3, the meridian path control point The coordinates of are matched with the anatomical standard database, and the error correction function is: ; Where, is the correction amount of the control point coordinates; is the actual control point coordinate obtained by matching the anatomical standard database, which is a three-dimensional vector; is the coordinate of the theoretical control point defined in the TCM meridian theory, which is a three-dimensional vector; To correct the weight coefficient, its value was determined through matching experiments with the anatomical standard database.

[0017] In the above-mentioned VR meridian and acupoint 3D digital human model construction method, the distance between the left and right eye viewpoints of the parallel binocular asymmetric viewing volume model in step 4 is set to 65 mm, and the projection matrix formula is: ; Where, n is the near clipping surface, f is the far clipping plane, l , r , b , t is the viewing volume boundary parameter.

[0018] In the above-mentioned VR meridian and acupoint three-dimensional digital human model construction method, in step 5, gesture recognition adopts a convolutional neural network classification model, the input is a 128×128 pixel depth map, the output is 6 types of gesture actions, and the recognition accuracy rate is ≥98%.

[0019] In the above-mentioned VR meridian and acupoint 3D digital human model construction method, when gestures are recognized and the user approaches dangerous acupoints, a 200Hz vibration feedback is applied through the Haptic Feedback tactile device, and the intensity is linearly related to the distance: ; Where, is the tactile feedback intensity; The current position coordinates of the acupuncture needle tip; The coordinates of dangerous acupuncture points; is a proportionality constant whose value is calibrated based on the physical characteristics of the force feedback device.

[0020] In the above-mentioned VR meridian and acupoint three-dimensional digital human model construction method, after the VR meridian and acupoint three-dimensional digital human model is constructed, a cross-platform APP is developed based on the Unity engine, and GPU instantiation technology is used to reduce the rendering load. The frame rate is stable at ≥90fps, multi-user collaborative operation is supported, and a distributed rendering architecture is adopted. Each client synchronizes the model status through the WebRTC protocol, and the delay is ≤50ms.

[0021] The VR meridian and acupoint 3D digital human model constructed by the above construction method is specifically used for: In the field of medical education, VR equipment is used to achieve immersive acupoint anatomy teaching, supporting real-time assessment and error correction.

[0022] Clinical simulation: Combined with force feedback equipment to simulate the acupuncture needle insertion process and provide real-time tissue resistance feedback.

[0023] Scientific research analysis: Based on the model data, the spatial relationship between acupoints, nerves and blood vessels is statistically analyzed, and a three-dimensional probability distribution map is output.

[0024] Compared with the existing technology, the VR meridian and acupoint three-dimensional digital human model construction method of the present invention has at least the following beneficial effects: 1. The present invention's method for constructing a VR meridian and acupoint 3D digital human model utilizes multi-dimensional, high-precision data acquisition, next-generation digital modeling, layered construction of the meridian system, VR scene fusion, and integrated interactive functions to create a highly accurate, realistic, and interactive 3D digital human model. This model can be widely used in medical education, clinical simulation, and scientific research analysis. Furthermore, through dynamic data capture and fusion, cross-platform development, and efficient rendering, the model's practicality, immersiveness, and application scope are significantly enhanced. This provides an efficient, accurate, and safe solution for medical education, clinical practice, and scientific research analysis, faithfully recreating the anatomical details and dynamic characteristics of the human meridian system while addressing the lack of an immersive user experience.

[0025] 2. This invention uses a CT scanner to collect cross-sectional data of human meridians and acupoints, and combines this with optical marker tracking technology to capture muscle and joint morphology in motion, achieving the fusion of static and dynamic data. This multidimensional, high-precision data acquisition method overcomes the limitations of traditional methods, providing an accurate and comprehensive data foundation for 3D modeling and enhancing the model's practicality and application range.

[0026] 3. Utilizing next-generation modeling technology, combined with PBR material mapping, normal mapping, and displacement mapping, we constructed a highly accurate and detailed 3D model. By strictly controlling the model error rate to ≤ 0.1% and utilizing advanced topology optimization algorithms, we ensured model accuracy and rendering efficiency, significantly improving the model's visual quality and interactive performance, and overcoming the limitations of traditional modeling techniques.

[0027] 4. Based on Traditional Chinese Medicine meridian theory, the system modeled the meridians, collaterals, tendons, and skin in layers. The meridian paths were defined using parameterized curve equations. Combined with an anatomical standard database and error correction functions, this approach enabled precise matching and dynamic adjustment of the meridian system. This layered modeling and parameterized path definition approach not only improved the model's accuracy but also enhanced its flexibility and visualization, providing reliable data support for medical education and scientific research analysis.

[0028] 5. By establishing a parallel binocular asymmetric viewing volume model and dynamically adjusting the viewpoint spacing based on the user's interpupillary distance, a more realistic stereoscopic visual effect is provided. Combined with the Leap Motion sensor, real-time multi-gesture recognition and high-precision tactile feedback significantly enhance the user experience and safety in VR environments. This immersive interaction not only enhances the user experience but also provides a more intuitive and efficient platform for medical education, clinical simulation, and scientific research analysis. DETAILED DESCRIPTION

[0029] The following is a more detailed description of the VR meridian and acupoint three-dimensional digital human model construction method of the present invention through specific implementation methods. Example 1

[0030] This embodiment discloses a method for constructing a VR meridian and acupoint three-dimensional digital human model, which is used to solve the problems in the existing technology that direct modeling of static tomographic data is difficult to truly restore the anatomical details and dynamic characteristics of the human meridian system and lacks an immersive operating experience.

[0031] The present invention adopts the following technical solutions to solve the technical problems: A method for constructing a VR meridian and acupoint three-dimensional digital human model comprises the following steps: Step 1: Multi-dimensional high-precision data acquisition: Based on real human plastination specimens, a CT scanner is used to collect tomographic data of human meridians and acupoints, and simultaneously obtain anatomical structure data in horizontal, sagittal, and coronal directions.

[0032] In this embodiment, the CT scanner collects cross-sectional data of human meridians and acupoints with a slice thickness of ≤0.5 mm, and the resolution of the anatomical structure data in the horizontal, sagittal, and coronal directions is ≥2048×2048 pixels.

[0033] By using a CT scanner to collect cross-sectional data of human meridians and acupoints with a slice thickness of ≤0.5mm, the accuracy and richness of the data are ensured. This high-precision data acquisition method provides a solid foundation for subsequent 3D modeling, enabling the model to highly restore the real human anatomical structure.

[0034] The simultaneous acquisition of horizontal, sagittal, and coronal anatomical structure data can provide multi-angle references for three-dimensional modeling, ensuring the accuracy and consistency of the model from all perspectives.

[0035] Step 2: Next-generation digital modeling: Based on the data from step 1, a three-dimensional model is constructed using next-generation modeling technology.

[0036] In this embodiment, the number of faces of a single model of the three-dimensional model is ≥6000, the texture map uses a PBR material map of 4096×4096 pixels, and the surface details are enhanced by normal mapping and displacement mapping, and the model error rate is ≤0.1%.

[0037] It should be noted that when implementing the above steps, the model topology optimization uses the Quadrilateral EdgeCollapse algorithm to ensure that the edge collapse error threshold is ≤ 0.01 mm, and the three-dimensional reconstruction of the fault data is achieved based on the Marching Cubes algorithm.

[0038] Utilizing next-generation modeling technology to construct 3D models, we generate highly accurate and detailed models, with a single model boasting ≥6,000 faces. Texture mapping utilizes 4,096×4,096 pixel PBR textures, with surface detail enhanced through normal and displacement mapping. This high-precision modeling technology allows the models to closely replicate the anatomy of the real human body, resulting in realistic visuals.

[0039] Traditional 3D modeling techniques usually find it difficult to achieve such high accuracy and detail. However, this method uses next-generation modeling technology, combined with PBR material mapping, normal mapping and displacement mapping, to generate highly realistic 3D models, breaking through the limitations of traditional modeling technology.

[0040] Furthermore, by strictly controlling the model error rate to ≤0.1%, the accuracy and reliability of the model are ensured. This high-precision model can provide accurate data support for medical education, clinical simulation, and scientific research analysis, reducing misleading or errors caused by model errors.

[0041] The Quadrilateral Edge Collapse algorithm was used to optimize the model's topology, ensuring an edge collapse error threshold of ≤0.01mm. Three-dimensional reconstruction of the slice data was achieved using the Marching Cubes algorithm. This optimization and reconstruction technology reduces the model's face count and computational load without sacrificing detail, improving rendering efficiency and real-time interactive performance.

[0042] Traditional 3D modeling often finds it difficult to strike a balance between accuracy and efficiency in topology optimization and 3D reconstruction. However, this method achieves a balance between high accuracy and high efficiency through advanced algorithms.

[0043] Using 4096×4096 pixel PBR texture maps, combined with normal maps and displacement maps, can enhance the surface detail of the model, making the model more visually realistic. This high-detail texture and material performance can enhance the user experience, especially in medical education and clinical simulation, and can provide more realistic visual feedback.

[0044] By applying next-generation modeling technology, combined with high-precision error control, topology optimization and 3D reconstruction technology, as well as highly detailed texture and material performance, a highly accurate and realistic 3D model is constructed. Step 3: Layered construction of the meridian system: Based on the meridian theory of traditional Chinese medicine, the meridians, collaterals, tendons, and skin are modeled layer by layer, and the meridian paths are defined using parameterized curve equations. The path equation is: Where, is the Bessel basis function, To control the point coordinates, ensure that the path is smooth and conforms to anatomical standards.

[0045] In this embodiment, the meridian path control points The coordinates of are matched with the anatomical standard database, and the error correction function is: Where, is the correction amount of the control point coordinates; is the actual control point coordinate obtained by matching the anatomical standard database, which is a three-dimensional vector; is the coordinate of the theoretical control point defined in the TCM meridian theory, which is a three-dimensional vector; To correct the weight coefficient, its value was determined through matching experiments with the anatomical standard database.

[0046] According to the meridian theory of traditional Chinese medicine, the meridians, collaterals, meridian tendons and skin parts are modeled in layers. Through layered modeling and parameterized path definition, multi-level visualization of the meridian system can be achieved. Users can clearly observe the structure and distribution of meridians, collaterals, meridian tendons and skin parts at different levels.

[0047] The meridian paths are defined using parameterized curve equations, ensuring smooth paths that conform to anatomical standards. This mathematical path definition not only improves model accuracy but also allows for more flexible adjustment and optimization of meridian paths, facilitating modification based on different anatomical standards. Through parameterized curve equations and error correction functions, meridian paths can be dynamically adjusted and optimized to ensure smoothness and anatomical accuracy.

[0048] The control point coordinates of the meridian path are matched by the anatomical standard database, and error correction is performed in combination with the error correction function to ensure the accuracy of the meridian path. The database-based matching and correction method can reduce human errors and improve the reliability of the model.

[0049] Traditional meridian path modeling usually lacks a systematic error correction mechanism. However, this method achieves accurate matching and correction of meridian paths by combining an anatomical standard database and an error correction function.

[0050] An accurate and flexible meridian system model was constructed through technical means such as hierarchical modeling based on traditional Chinese medicine meridian theory, defining meridian paths with parameterized curve equations, and matching and error correction with anatomical standard databases.

[0051] Step 4: VR visual volume fusion, establish a parallel binocular asymmetric visual volume model. In this embodiment, the visual volume model parameters can be dynamically adjusted according to the user's pupil distance. d Distance from viewpoint s The relationship is; .

[0052] By establishing a parallel binocular asymmetric viewing volume model, VR devices can be provided with a more realistic stereoscopic visual effect. This viewing volume model can dynamically adjust the viewpoint distance based on the user's pupil distance, ensuring that the images seen by the user in the VR environment are consistent with the actual human visual experience, reducing visual fatigue and discomfort.

[0053] At the same time, the viewing volume parameters can be dynamically adjusted based on the user's interpupillary distance to ensure that the viewpoint distance matches the user's interpupillary distance. This dynamic adjustment mechanism can improve the comfort of the VR experience and reduce visual discomfort and fatigue caused by interpupillary distance mismatch.

[0054] The frustum boundary parameters (left, right, top, and bottom) can be flexibly configured based on actual needs, ensuring that images within the VR environment are fully presented within the user's field of view. This flexible configuration approach can adapt to different application scenarios and device configurations, improving the versatility and adaptability of the VR system.

[0055] Step 5: Integrate interactive functions, implement real-time multi-gesture recognition through the Leap Motion sensor, and define response functions for gesture control model rotation, scaling, and translation.

[0056] In this embodiment, gesture recognition uses a convolutional neural network classification model with a 128×128 pixel depth map as input and 6 types of gesture actions as output, with a recognition accuracy of ≥98%.

[0057] The Leap Motion sensor enables real-time multi-gesture recognition, allowing users to control the model's rotation, scaling, translation, and other operations through natural gestures. This intuitive and easy-to-use interaction significantly enhances the user experience in VR environments.

[0058] By defining gesture-controlled model rotation, scaling, and translation response functions, we ensure the accuracy and real-time nature of user gestures. These response functions adjust the model's state in real time based on user gestures, providing a smooth interactive experience.

[0059] A convolutional neural network (CNN) classification model is used for gesture recognition. The input is a 128×128 pixel depth map, and the output is six types of gestures, with a recognition accuracy of ≥98%. This high-precision recognition model ensures the accuracy and reliability of gesture operations, reducing the possibility of misoperation.

[0060] This embodiment also discloses the application scope of the VR meridian and acupoint three-dimensional digital human model constructed by the above steps and methods. The VR meridian and acupoint three-dimensional digital human model constructed by the above steps and methods can be applied in the field of medical education, and immersive acupoint anatomy teaching can be realized through VR equipment, supporting real-time assessment and error correction.

[0061] Through VR devices, students can intuitively observe and learn the distribution, structure, and function of meridian acupoints in a virtual environment, significantly improving learning outcomes. The system also supports real-time assessments, providing timely feedback on student operational errors and corrective suggestions, helping students quickly master correct acupoint positioning and manipulation techniques. Traditional anatomy teaching relies on physical specimens or models, which are costly and difficult to reuse, while VR models can be reused endlessly, reducing teaching costs.

[0062] The VR meridian and acupoint three-dimensional digital human model constructed through the above steps can also be used in clinical simulations, combined with force feedback equipment to simulate the acupuncture needle insertion process and provide real-time tissue resistance feedback.

[0063] By integrating force feedback devices, the system simulates tissue resistance during acupuncture needle insertion, providing realistic feedback and helping doctors or students master the force and depth of needle insertion. Furthermore, through repeated simulation practice, doctors can master acupuncture techniques in a virtual environment, reducing the risk of errors in actual practice. Furthermore, simulating procedures in a virtual environment avoids the risks associated with operating on real patients, making it particularly suitable for beginners.

[0064] At the same time, it can also be used for scientific research analysis, statistically analyzing the spatial relationship between acupoints and nerves and blood vessels based on model data, and outputting a three-dimensional probability distribution map.

[0065] Based on model data, the system can calculate the spatial relationship between acupoints and surrounding nerves and blood vessels, and output a three-dimensional probability distribution map, providing researchers with accurate data support. It can also assist in the study of Traditional Chinese Medicine (TCM) theory. Through the three-dimensional probability distribution map, researchers can more intuitively analyze the spatial relationship between acupoints and surrounding tissues, providing a scientific basis for the modern research of TCM meridian theory. The three-dimensional probability distribution map can intuitively display the relationship between acupoints and surrounding tissues, helping researchers better understand and analyze data.

[0066] The VR meridian and acupoint three-dimensional digital human model constructed by the above steps and methods can be applied not only to the above-mentioned fields, but also to other fields, including but not limited to health management, intelligent diagnosis and treatment systems, telemedicine, etc. The above is only an example and does not exhaust all possible application scenarios. Without deviating from the core of the present invention, all reasonable application extensions fall within the scope of protection of the present invention. Example 2

[0067] The same points as the above embodiment will not be repeated here, but the differences are as follows: In this embodiment, the multi-dimensional high-precision data acquisition in step one also includes dynamic capture of muscle and joint morphology in motion; optical marker tracking technology is used with a sampling frequency of ≥120Hz, and dynamic data is fused with a static model through a skeleton skinning algorithm to capture subtle changes in the body during motion.

[0068] Optical marker tracking technology captures dynamic data on muscle and joint morphology during motion, capturing subtle changes in the body during movement. This dynamic data capture allows the model to not only display static anatomical structures but also simulate changes in the human body during movement, enhancing its practicality and application range.

[0069] By fusing dynamic data with static models through a skeletal skinning algorithm, we can accurately simulate subtle changes in the human body during movement. This fusion approach allows the model to be better applied in clinical simulations and scientific research analysis, providing a more realistic experience and data support.

[0070] Through high-precision, multi-dimensional data collection, combined with the capture and fusion of dynamic data, the combination of static and dynamic data can be achieved, providing an accurate and comprehensive data foundation for subsequent three-dimensional modeling. The above data collection method breaks through the limitations of traditional methods and provides a more comprehensive data collection solution for the construction of VR meridian and acupoint three-dimensional digital human models. Example 3

[0071] The similarities with the above embodiments and their combinations are not repeated here, and the differences are as follows: In this embodiment, the distance between the left and right eye viewpoints of the parallel binocular asymmetric viewing volume model in step 4 is set to 65 mm, and the projection matrix formula is: Where, n is the near clipping surface, f is the far clipping plane, l , r , b , t is the viewing volume boundary parameter.

[0072] Precise projection matrix calculations ensure the visual accuracy and consistency of 3D models in VR environments. The projection matrix formula includes parameters for the near and far clipping planes, as well as the frustum boundaries, and can generate images that conform to perspective projection principles, enhancing the realism of the VR experience.

[0073] The parallel binocular asymmetric viewing volume model and precise projection matrix calculation can significantly enhance the immersiveness of VR environments. The images users see in VR environments are more realistic and three-dimensional, allowing them to better integrate into the virtual scene and improve the user experience. Example 4

[0074] The similarities with the above embodiments and their combinations are not repeated here, and the differences are as follows: When gestures are recognized and the user approaches dangerous acupuncture points, the Haptic Feedback device applies vibration feedback at a frequency of 200Hz, with the intensity being linearly related to the distance: Where, is the tactile feedback intensity; The current position coordinates of the acupuncture needle tip; The coordinates of dangerous acupuncture points; is a proportionality constant whose value is calibrated based on the physical characteristics of the force feedback device.

[0075] In this embodiment, real-time tactile feedback allows users to promptly sense when approaching dangerous acupuncture points during operation, avoiding misoperation and improving operational safety. Tactile feedback provides an intuitive warning method, enhancing the user's immersion and operational experience in the VR environment. Example 5

[0076] The similarities with the above embodiments and their combinations are not repeated here, and the differences are as follows: In this embodiment, after the VR meridian and acupoint three-dimensional digital human model is constructed, a cross-platform APP is developed based on the Unity engine, and GPU instantiation technology is used to reduce the rendering load. The frame rate is stable at ≥90fps, multi-user collaborative operation is supported, and a distributed rendering architecture is adopted. Each client synchronizes the model status through the WebRTC protocol, and the delay is ≤50ms.

[0077] Cross-platform apps developed with the Unity engine can run on different operating systems and devices, such as Windows, macOS, iOS, and Android, greatly expanding the scope of app use. Users can use the app on multiple devices, improving its versatility and convenience.

[0078] At the same time, GPU instancing significantly reduces rendering overhead, improves rendering efficiency, and ensures a stable frame rate of ≥90fps. This efficient rendering technology provides a smooth visual experience, reduces lag and latency, and is particularly suitable for the high frame rate requirements of VR environments.

[0079] In addition, in this embodiment, the cross-platform app supports multi-user collaboration, allowing multiple users to operate and interact simultaneously in the VR environment. This function is particularly suitable for scenarios such as medical education, team collaboration, and clinical simulation, and can improve collaboration efficiency and learning effects.

[0080] In addition, unless otherwise defined, technical or scientific terms used herein shall have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0081] The exemplary embodiments of the present invention are described in detail above with reference to preferred embodiments. The scope of protection of the present invention is not limited thereto. Those skilled in the art will understand that, without departing from the concept of the present invention, various modifications and variations can be made to the above-mentioned specific embodiments, and various combinations of the various technical features proposed in the present invention can be made without exceeding the scope of protection of the present invention.

Claims

1. A method for constructing a VR meridian and acupoint three-dimensional digital human model, characterized in that: The method comprises the following steps: Step 1: Multi-dimensional, high-precision data acquisition. Based on real human plastination specimens, a CT scanner is used to collect cross-sectional data of human meridians and acupoints, and simultaneously obtain anatomical structure data in the horizontal, sagittal, and coronal directions. Step 2: Next-generation digital modeling: Based on the data from step 1, a 3D model is constructed using next-generation modeling technology. Step 3: The meridian system is constructed in layers. Based on the meridian theory of traditional Chinese medicine, the meridians, collaterals, tendons, and skin are modeled in layers, and the meridian paths are defined using parameterized curve equations. Step 4: VR visual volume fusion, establishing a parallel binocular asymmetric visual volume model; Step 5: Integrate interactive functions. Use the Leap Motion sensor to achieve real-time multi-gesture recognition and define the response functions for gesture control model rotation, scaling, and translation.

2. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 1, characterized in that: In the step 1, the CT scanner collects cross-sectional data of the human meridians and acupoints with a slice thickness of ≤0.5 mm, and the resolution of the anatomical structure data in the horizontal, sagittal, and coronal directions is ≥2048×2048 pixels; The number of faces of a single 3D model in step 2 is ≥ 6000, the texture map uses a 4096×4096 pixel PBR material map, and the surface details are enhanced by normal map and displacement map, and the model error rate is ≤ 0.1%; In step 3, a parameterized curve equation is used to define the meridian path, and the path equation is: Where, is the Bessel basis function, To control the point coordinates, ensure the path is smooth and conforms to anatomical standards; In the step 4, the viewing volume parameters can be dynamically adjusted according to the user's pupil distance. d Distance from viewpoint s The relationship is; 。 3. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 2, characterized in that: In step 1, data collection also includes dynamic capture of muscle and joint morphology in motion, using optical marker tracking technology with a sampling frequency of ≥120 Hz, and fusion of dynamic data with static models through a skeleton skinning algorithm to capture subtle changes in the body during motion.

4. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 3, characterized in that: In step 2, the model topology optimization uses the Quadrilateral Edge Collapse algorithm to ensure that the edge collapse error threshold is ≤ 0.01 mm, and the three-dimensional reconstruction of the fault data is achieved based on the Marching Cubes algorithm.

5. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 4, characterized in that: In step 3, the meridian path control point The coordinates of are matched through the anatomical standard database, and the error correction function is; Where, is the correction amount of the control point coordinates; is the actual control point coordinate obtained by matching the anatomical standard database, which is a three-dimensional vector; is the coordinate of the theoretical control point defined in the TCM meridian theory, which is a three-dimensional vector; To correct the weight coefficient, its value was determined through matching experiments with the anatomical standard database.

6. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 5, characterized in that: In the step 4, the distance between the left and right eye viewpoints of the parallel binocular asymmetric viewing volume model is set to 65 mm, and the projection matrix formula is: ； Where, n is the near clipping surface, f is the far clipping plane, l , r , b , t is the viewing volume boundary parameter.

7. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 6, characterized in that: In step 5, gesture recognition uses a convolutional neural network classification model with a 128×128 pixel depth map as input and 6 types of gesture actions as output, with a recognition accuracy of ≥98%.

8. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 7, characterized in that: When gestures are recognized and the user approaches dangerous acupuncture points, the Haptic Feedback device applies vibration feedback at a frequency of 200Hz, with the intensity linearly related to the distance. Where, is the tactile feedback intensity; The current position coordinates of the acupuncture needle tip; The coordinates of dangerous acupuncture points; is a proportionality constant whose value is calibrated based on the physical characteristics of the force feedback device.

9. The method for constructing a VR meridian and acupoint 3D digital human model according to claim 6, characterized in that: After the VR meridian and acupoint 3D digital human model is built, a cross-platform APP is developed based on the Unity engine, using GPU instantiation technology to reduce the rendering load, with a stable frame rate of ≥90fps, supporting multi-user collaborative operation, and using a distributed rendering architecture. Each client synchronizes the model status through the WebRTC protocol, with a delay of ≤50ms.

10. An application of a VR meridian and acupoint 3D digital human model, comprising a VR meridian and acupoint 3D digital human model constructed by the construction method according to any one of claims 1 to 9, characterized in that: Applied in the field of medical education; using VR equipment to achieve immersive acupoint anatomy teaching, supporting real-time assessment and error correction; Applied to clinical simulation; combined with force feedback equipment to simulate the acupuncture needle insertion process and provide real-time tissue resistance feedback; Applied to scientific research analysis; based on the model data, the spatial relationship between acupoints, nerves and blood vessels is statistically analyzed, and a three-dimensional probability distribution map is output.

Images