Method and system for generating disaster case 3D virtual visualization patient based on data driving
Through data-driven methods and artificial intelligence algorithms, 3D virtual disaster patients are dynamically constructed and displayed, solving the problem of insufficient time and flexibility in creating traditional 3D visual patients, achieving efficient and detailed disaster injury display and interaction, and improving medical care The effectiveness of diagnosis and education and training.
Patent Information
- Application Number
- CN202410965054.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-08
AI Technical Summary
The creation of traditional 3D visualized patients in the field of medical first aid requires a lot of time and manpower, and virtual disaster patients lack random variability, which cannot effectively meet the display and interaction needs of complex trauma effects.
Using a data-driven method, abstractly extract key data, build basic databases, segment and modular 3D models, combine artificial intelligence algorithms and 3D real-time rendering technology to dynamically build and display disaster patients, realizing the splicing and fusion of Mesh grid data and real-time Interaction.
It improves the efficiency and flexibility of 3D visualization, provides a variety of variable virtual disaster patient models, enhances the efficiency and accuracy of medical diagnosis, promotes medical education and training, and improves the first aid thinking and decision-making capabilities of medical staff.
Smart Images

Figure CN118919083A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of 3D virtual visualization, and in particular to a method and system for generating 3D virtual visualization patients of disaster cases based on data drive. Background Art
[0002] 3D visualization is a tool used to display, describe and understand the characteristics of many underground and surface geological phenomena. It is widely used in all fields of geology and geophysics. 3D visualization is a means of depicting and understanding models and a representation form of data bodies, not a simulation technology. It can use large amounts of data, check the continuity of data, identify the authenticity of data, discover and propose useful anomalies, provide useful tools for analyzing, understanding and repeating data, and serve as a bridge for multidisciplinary communication and collaboration.
[0003] Traditional 3D visualized patients are manually created using traditional 3D artificial modeling technology. In particular, the creation of virtual disaster patients in the field of medical emergency requires different trauma effects on different parts. Each part and each trauma effect requires modelers to spend a lot of time drawing and modeling to complete. At the same time, the completed virtual disaster patients do not have random variability. This part of the work requires a lot of manpower, time and funds. Therefore, a method and system for generating 3D virtual visualization patients for disaster cases based on data-driven is proposed to solve the above problems. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present invention provides a method and system for generating 3D virtual visualization patients of disaster cases based on data-driven, which has the advantages of improving the efficiency and flexibility of 3D visualization, and solves the problem that traditional 3D visualization patients are manually created by using traditional 3D artificial modeling technology. In particular, the creation of virtual disaster patients in the field of medical emergency requires different trauma effects on different parts, and the modelers spend a lot of time drawing and modeling for each part and each trauma effect. At the same time, the completed virtual disaster patients do not have random variability, and this part of the work requires a lot of manpower, time and funds.
[0005] To achieve the above object, the present invention provides the following technical solution: a method for generating 3D virtual visualization patients of disaster cases based on data drive, comprising the following steps:
[0006] Step 1: For the creation of 3D virtual disaster patients, abstract and extract key data, such as patient gender, age group, trauma site, and trauma type, and build a basic database.
[0007] Step 2: For the creation of 3D virtual disaster patients, this part of the 3D visualized 3D model is planned and segmented, and basic module resources such as patient torso, patient limbs, patient residual limbs, patient surface trauma, clothing and props are modularized. 3D modeling technology is used to create the basic 3D resources for creating 3D virtual disaster patients to form a 3D resource library.
[0008] Step 3: Develop a case editing module to select and associate case generation based on background key data to form a complete case injury data set.
[0009] Step 4: The front end requests case data from the back end through network communication. After the back end pushes the key data of the corresponding case data to the front end, the front end retrieves the corresponding 3D resources based on the data.
[0010] Step 5: The front end uses artificial intelligence algorithms to retrieve the corresponding 3D resources and integrate Mesh modules such as position, location, and trauma effect to form a complete 3D resource of disaster patients.
[0011] Step 6: Based on 3D real-time rendering technology, the complete disaster patient will be presented in 3D visualization and real-time interaction, and the result of the interaction will also be presented in 3D visualization real-time rendering.
[0012] Furthermore, in step one, the planning and source of data come from data from multiple different sources, such as medical records, physiological data, pathological data, site data, sensor data, and environmental data, and are dynamically constructed based on real-time data streams in a data-driven manner.
[0013] Furthermore, in the step five, the mesh data splicing and fusion method merges multiple independent mesh data into a continuous and seamless three-dimensional model, combining data preprocessing, mesh alignment and merging steps. There are n models, and their prediction results are y1, y2, ..., yn, and the corresponding weights are w1, w2, ..., wn (satisfying w1+w2+...+wn=1). The fusion result of the weighted average method is: [\text{Fus i onResult}=\sum_{i=1}^{n}w_i\cdoty_i]. The key weights can be determined by cross-validation and model performance evaluation to reflect the contribution of each model to the final prediction result.
[0014] Furthermore, the data preprocessing includes denoising, simplification and standardization. In the denoising and simplification, for each input Mesh grid, denoising is first performed to remove unnecessary details and noise, and the model is simplified to reduce the amount of calculation. In the standardization, all Mesh grids are converted to a unified coordinate system and measurement unit for subsequent processing.
[0015] Furthermore, the mesh registration includes feature extraction and alignment. In the feature extraction, key feature points or feature lines, such as corner points and edges, are extracted from each Mesh mesh. In the alignment, rigid transformations (such as translation, rotation, and scaling) are used to align the extracted feature points or feature lines to the same coordinate system to ensure that the positions and directions of different meshes are consistent.
[0016] Furthermore, the fusion parameters include disposeSource: a Boolean value indicating whether to destroy the source mesh after merging, all32BitsIndices: a Boolean value indicating whether to allow the use of 32-bit indexes, meshSubclass (optional): specifying the subclass of the merged mesh, subdividingWithSubMeshes (optional): a Boolean value indicating whether to subdivide the merged mesh to generate sub-meshes, and multiMultiMaterials (optional): a Boolean value indicating whether to use multi-material merging. The splicing fusion function is:
[0017]
[0018]
[0019] Furthermore, the subsequent processing includes smoothing and optimization and compression. In the smoothing, the merged Mesh mesh is smoothed to eliminate seams and uneven edges. In the optimization and compression, the mesh structure is optimized and the number of vertices and faces is reduced to improve rendering performance and reduce memory usage.
[0020] Furthermore, the visualization includes spatial coordinate calculation, color mapping calculation, rendering calculation and interactive calculation, and real-time interaction with the virtual patient is carried out through an interactive interface. When constructing a 3D model, it is necessary to determine the position of each data point in the three-dimensional space, which usually involves calculations such as projection and rotation in the three-dimensional coordinate system. Taking 3D surface visualization as an example, a formula similar to the following can be used to describe the shape of the surface in 3D space:
[0021] Z value calculation: For given X and Y values, use a function to calculate the corresponding Z value. This function can be a simple mathematical expression (such as f(x,y)=x2+y2) or a complex mathematical model (such as f(x,y)=(1-y5+x5)*exp(-x2-y2));
[0022] Mesh generation: Use the X and Y grid values (such as X, Y = np.meshgrid(X_values, Y_values)) and the above Z value calculation function to generate the grid point coordinates of the surface;
[0023] Drawing surfaces: Write shader programs (vertex shaders and fragment shaders) to control the appearance and lighting effects of 3D models, and use 3D rendering engines (such as OpenGL, DirectX, etc.) to render 3D models to the screen;
[0024] Interactive computing: mainly involves operation response and data update. For example, when the user interacts with the visualization interface through a mouse or touch screen, the system needs to calculate the response of the user operation (such as rotation, zooming, and translation) and update the data in real time to reflect the impact of these operations on the visualization results. For example, the user can perform treatment operations on the virtual patient and observe its reaction and effect. The real-time data obtained is used as the basis for driving the 3D visualization virtual patient. These data are input into pre-designed algorithms and models to dynamically update the status and behavior of the virtual patient. Using advanced 3D modeling and rendering technology, the real-time status of the virtual patient is displayed in a realistic 3D form, including the patient's vital signs, wound condition, appearance, movements, expressions and other details. Real-time visualization and update, and adjust the treatment strategy based on feedback.
[0025] Furthermore, the use of algorithms and models helps users collect, analyze and interpret various data, thereby providing a basis and suggestions for decision-making:
[0026] z_j^{(2)}=\sum_{i=1}^{n}w_{ij}^{(2)}\cdot a_i^{(1)}+b_j^{(2)}
[0027] a_j^{(2)}=\si gma(z_j^{(2)})
[0028] zj(2) is the weighted sum of the jth neuron in the hidden layer, wi j(2) is the weight from the i-th neuron in the input layer to the j-th neuron in the hidden layer, ai(1) is the output (i.e., input data) of the i-th neuron in the input layer, and bj(2) is the bias term of the j-th neuron in the hidden layer.
[0029] A system for generating 3D virtual visualization patients for disaster cases based on data-driven method includes a management-end browser module, the management-end browser module is bidirectionally electrically connected to a system module, the system module is bidirectionally electrically connected to a user-end program module, the system module includes a 3D resource module, a Mesh splicing and fusion module, a real-time rendering module, a simulation interaction module, a network communication module, a data service storage module and an auxiliary AI module, a data interface communication is established between the management-end browser module and the client, and the generated patient data group information is saved, the system module obtains database data, and then edits and communicates the data, edits and defines digital patients to generate patient data groups, the system module generates or updates digital patients and renders in real time, and interacts with visualized patients, and the user-end program module communicates with the server and feeds back interactive information.
[0030] Compared with the prior art, the technical solution of this application has the following beneficial effects:
[0031] This method and system for generating 3D virtual visualization patients for disaster cases based on data-driven means can effectively improve the efficiency and accuracy of medical diagnosis by using a large number of diverse and variable 3D disaster virtual patients for visualization and virtual interactive training methods. Especially when dealing with complex disaster rescue cases, it can provide medical staff with more efficient, intuitive and detailed disaster injury presentations, promote medical education and training, and provide a more realistic, vivid and authentic learning experience through virtual reality technology, thereby reducing costs while improving the medical staff's emergency thinking and decision-making ability in clinical situations. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 It is a schematic diagram of the system structure of the present invention;
[0033] Figure 2 It is a structural schematic diagram of the system block diagram in the present invention;
[0034] Figure 3 for Figure 2 An enlarged schematic diagram of the system module. DETAILED DESCRIPTION
[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0036] See also Figure 1-3 In this embodiment, a method for generating 3D virtual visualization patients of disaster cases based on data drive includes the following steps:
[0037] Step 1: For the creation of 3D virtual disaster patients, key data are abstracted and extracted, such as patient gender, age group, trauma site and trauma type, and a basic database is built. In step 1, data planning and sources come from multiple different sources, such as medical records, physiological data, pathological data, site data, sensor data and environmental data, etc., and are dynamically constructed according to real-time data streams in a data-driven manner.
[0038] Step 2: For the creation of 3D virtual disaster patients, this part of the 3D visualized 3D model is planned and segmented, and basic module resources such as patient torso, patient limbs, patient residual limbs, patient surface trauma, clothing and props are modularized. 3D modeling technology is used to create the basic 3D resources for creating 3D virtual disaster patients to form a 3D resource library.
[0039] Step 3: Develop a case editing module to select and associate case generation based on background key data to form a complete case injury data set.
[0040] Step 4: The front end requests case data from the back end through network communication. After the back end pushes the key data of the corresponding case data to the front end, the front end retrieves the corresponding 3D resources based on the data.
[0041] Step 5: The front end uses artificial intelligence algorithms to stitch and fuse the corresponding 3D resources retrieved, such as position, location, and trauma effects, to form a complete 3D resource of disaster patients. In step 5, the mesh data splicing and fusion method is to merge multiple independent mesh data into a continuous and seamless 3D model, combining data preprocessing, mesh registration and merging steps. There are n models, and their prediction results are y1, y2, ..., yn, respectively, and the corresponding weights are w1, w2, ..., wn (satisfying w1+w2+...+wn=1). The fusion result of the weighted average method is: [\text{FusionResult}=\sum_{i=1}^{n}w_i\cdoty_i], which means "fusion result (FusionResu lt)" is calculated based on the weight (w) and the corresponding value (y). The key weight can be determined by cross-validation and model performance evaluation to reflect the contribution of each model to the final prediction result. Data preprocessing includes denoising, simplification and standardization. In denoising and simplification, for each input Mesh grid, denoising is first performed to remove unnecessary details and noise, and the model is simplified to reduce the amount of calculation. In standardization, all Mesh grids are converted to a unified coordinate system and measurement unit for subsequent processing. Mesh registration includes feature extraction and alignment. In feature extraction, key feature points or feature lines, such as corners and edges, are extracted from each Mesh grid. In alignment, rigid transformations (such as translation, rotation and scaling) are used to align the extracted feature points or feature lines to the same coordinate system to ensure that the positions and directions of different meshes are consistent. Subsequent processing includes smoothing and optimization and compression. In smoothing, the merged Mesh grid is smoothed to eliminate seams and uneven edges. In optimization and compression, the mesh structure is optimized to reduce the number of vertices and faces to improve rendering performance and reduce memory usage.
[0042] Step 6: Based on 3D real-time rendering technology, the complete disaster patient will be presented in 3D visualization and real-time interaction, and the result of the interaction will also be presented in 3D visualization real-time rendering.
[0043] Among them, the fusion parameters include disposeSource: Boolean value, indicating whether to destroy the source mesh after merging, all32BitsIndexes: Boolean value, indicating whether to allow the use of 32-bit indexes, meshSubclass (optional): specify the subclass of the merged mesh, subdivideWithSubMeshes (optional): Boolean value, indicating whether to subdivide the merged mesh to generate sub-meshes, and multiMultiMaterials (optional): Boolean value, indicating whether to use multi-material merging. The splicing fusion function is:
[0044]
[0045]
[0046] Among them, this section of function code for merging multiple meshes first calls a function called Mesh.MergeMeshes, which may be a function in Babylon.js. It is used to merge multiple meshes. If multi-material processing is required (multi MultiMaterials is true), the material will be further set. In this case, assuming that arrayOfMaterials is the material array corresponding to arrayOfMeshes, these materials are merged and applied to the merged mesh mergedMesh. Finally, the merged mesh mergedMesh is returned.
[0047] Secondly, visualization includes spatial coordinate calculation, color mapping calculation, rendering calculation and interactive calculation. Real-time interaction with virtual patients is carried out through an interactive interface. When building a 3D model, it is necessary to determine the position of each data point in the three-dimensional space, which usually involves calculations such as projection and rotation in the three-dimensional coordinate system. Taking 3D surface visualization as an example, a formula similar to the following can be used to describe the shape of the surface in 3D space;
[0048] Z value calculation: For given X and Y values, use a function to calculate the corresponding Z value. This function can be a simple mathematical expression (such as f(x,y)=x2+y2) or a complex mathematical model (such as f(x,y)=(1-y5+x5)*exp(-x2-y2));
[0049] Mesh generation: Use the X and Y grid values (such as X, Y = np.meshgrid(X_values, Y_values)) and the above Z value calculation function to generate the grid point coordinates of the surface;
[0050] Drawing surfaces: Write shader programs (vertex shaders and fragment shaders) to control the appearance and lighting effects of 3D models, and use 3D rendering engines (such as OpenGL, DirectX, etc.) to render 3D models to the screen;
[0051] Interactive computing: mainly involves operation response and data update. For example, when the user interacts with the visualization interface through a mouse or touch screen, the system needs to calculate the response of the user operation (such as rotation, zooming, and translation) and update the data in real time to reflect the impact of these operations on the visualization results. For example, the user can perform treatment operations on the virtual patient and observe its reaction and effect. The real-time data obtained is used as the basis for driving the 3D visualization virtual patient. These data are input into the pre-designed algorithms and models to dynamically update the status and behavior of the virtual patient. The real-time status of the virtual patient is displayed in a realistic 3D form using advanced 3D modeling and rendering technology. This includes the real-time visualization and update of the patient's vital signs, wound condition, appearance, movements, and expressions, and adjusts the treatment strategy based on feedback. The algorithm and model are used to help users collect, analyze and interpret various data, thereby providing a basis and suggestions for decision-making:
[0052] z_j^{(2)}=\sum_{i=1}^{n}w_{ij}^{(2)}\cdot a_i^{(1)}+b_j^{(2)}
[0053] a_j^{(2)}=\s igma(z_j^{(2)})
[0054] zj(2) is the weighted sum of the jth neuron in the hidden layer, wi j(2) is the weight from the ith neuron in the input layer to the jth neuron in the hidden layer, ai(1) is the output of the ith neuron in the input layer (i.e., input data), bj(2) is the bias term of the jth neuron in the hidden layer, the first formula calculates the weighted input of the second layer neurons, and the second formula calculates the output of the second layer neurons.
[0055] Specifically, by using a large number of diverse and variable 3D disaster virtual patients' visualization presentation and virtual interactive training methods, the efficiency and accuracy of medical diagnosis can be effectively improved. Especially when dealing with complex disaster rescue cases, it can provide medical staff with more efficient, intuitive and detailed disaster injury presentations, promote medical education and training, and provide a more realistic, vivid and real learning experience through virtual reality technology, thereby reducing costs and improving the medical staff's emergency thinking and decision-making ability in clinical situations.
[0056] A system for generating 3D virtual visualization patients for disaster cases based on data-driven method includes a management-end browser module, the management-end browser module is bidirectionally electrically connected to a system module, the system module is bidirectionally electrically connected to a user-end program module, the system module includes a 3D resource module, a Mesh splicing and fusion module, a real-time rendering module, a simulation interaction module, a network communication module, a data service storage module and an auxiliary AI module, a data interface communication is established between the management-end browser module and the client, the generated patient data group information is saved, the system module obtains database data, and then edits and communicates the data, edits and defines digital patients to generate patient data groups, the system module generates or updates digital patients and renders in real time, and interacts with visualized patients, and the user-end program module communicates with the server and feeds back interactive information.
[0057] The working principle of the above embodiment is:
[0058] By using a large number of diverse and variable 3D disaster virtual patients' visualization and virtual interactive training methods, the efficiency and accuracy of medical diagnosis can be effectively improved. Especially when dealing with complex disaster rescue cases, it can provide medical staff with more efficient, intuitive and detailed disaster injury presentations, promote medical education and training, and provide a more realistic, vivid and real learning experience through virtual reality technology, reducing costs while improving medical staff's emergency thinking and decision-making ability in clinical situations.
[0059] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0060] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A method for generating 3D virtual visualization patients of disaster cases based on data-driven, characterized in that: The following steps are involved: Step 1: For the creation of 3D virtual disaster patients, abstract and extract key data, such as patient gender, age group, trauma site, and trauma type, and build a basic database; Step 2: For the creation of 3D virtual disaster patients, the 3D model of this part of 3D visualization is planned and segmented, and basic module resources such as patient torso, patient limbs, patient residual limbs, patient surface trauma, clothing and props are modularized. The basic 3D resources for creating 3D virtual disaster patients are produced using 3D modeling technology to form a 3D resource library; Step 3: Develop a case editing module to select and associate case generation based on background key data to form a complete case injury data set; Step 4: The front end requests case data from the back end through network communication. After the back end pushes the key data of the corresponding case data to the front end, the front end retrieves the corresponding 3D resources based on the data. Step 5: The front end uses artificial intelligence algorithms to stitch and fuse the corresponding 3D resources retrieved, such as position, location, and trauma effect, to form a complete 3D resource of disaster patients; Step 6: Based on 3D real-time rendering technology, the complete disaster patient will be presented in 3D visualization and real-time interaction, and the result of the interaction will also be presented in 3D visualization real-time rendering.
2. The method for generating 3D virtual visualization patients of disaster cases based on data drive according to claim 1, characterized in that: In step 1, the planning and source of data come from multiple different sources, such as medical records, physiological data, pathological data, site data, sensor data, and environmental data, and are dynamically constructed based on real-time data streams in a data-driven manner.
3. The method for generating 3D virtual visualization of patients in disaster cases based on data drive according to claim 1, characterized in that: In the step five, the mesh data splicing and fusion method merges multiple independent mesh data into a continuous and seamless three-dimensional model, combining data preprocessing, mesh alignment and merging steps. There are n models, and their prediction results are y1, y2, ..., yn, and the corresponding weights are w1, w2, ..., wn (satisfying w1+w2+...+wn=1). The fusion result of the weighted average method is: [\text{FusionResult}=\sum_{i=1}^{n}w_i\cdoty_i]. The key weights can be determined by cross-validation and model performance evaluation to reflect the contribution of each model to the final prediction result.
4. The method for generating 3D virtual visualization patients of disaster cases based on data drive according to claim 3, characterized in that: The data preprocessing includes denoising, simplification and standardization. In the denoising and simplification, for each input Mesh grid, denoising is first performed to remove unnecessary details and noise, and the model is simplified to reduce the amount of calculation. In the standardization, all Mesh grids are converted to a unified coordinate system and measurement unit for subsequent processing.
5. The method for generating 3D virtual visualization of patients in disaster cases based on data drive according to claim 4, characterized in that: The mesh registration includes feature extraction and alignment. In the feature extraction, key feature points or feature lines, such as corner points and edges, are extracted from each Mesh mesh. In the alignment, rigid transformations (such as translation, rotation, and scaling) are used to align the extracted feature points or feature lines to the same coordinate system to ensure that the positions and directions of different meshes are consistent.
6. The method for generating 3D virtual visualization patients of disaster cases based on data drive according to claim 5, characterized in that: The fusion parameters include disposeSource: Boolean value, indicating whether to destroy the source mesh after merging, allow32BitsIndices: Boolean value, indicating whether to allow the use of 32-bit indexes, meshSubclass (optional): specify the subclass of the merged mesh, subdivideWithSubMeshes (optional): Boolean value, indicating whether to subdivide the merged mesh to generate sub-meshes, and multiMultiMaterials (optional): Boolean value, indicating whether to use multi-material merging. The splicing fusion function is:
7. The method for generating 3D virtual visualization patients of disaster cases based on data drive according to claim 4, characterized in that: The subsequent processing includes smoothing and optimization and compression. In the smoothing process, the merged Mesh grid is smoothed to eliminate seams and uneven edges. In the optimization and compression, the grid structure is optimized and the number of vertices and facets is reduced to improve rendering performance and reduce memory usage.
8. The method for generating 3D virtual visualization of patients in disaster cases based on data drive according to claim 1, characterized in that: The visualization includes spatial coordinate calculation, color mapping calculation, rendering calculation and interactive calculation. Real-time interaction with the virtual patient is carried out through an interactive interface. When constructing a 3D model, it is necessary to determine the position of each data point in the three-dimensional space, which usually involves calculations such as projection and rotation in the three-dimensional coordinate system. Taking 3D surface visualization as an example, a formula similar to the following can be used to describe the shape of the surface in 3D space: Z value calculation: For given X and Y values, use a function to calculate the corresponding Z value. This function can be a simple mathematical expression (such as f(x,y)=x2+y2) or a complex mathematical model (such as f(x,y)=(1-y5+x5)*exp(-x2-y2)); Mesh generation: Use the X and Y grid values (such as X, Y = np.meshgrid (X_values, Y_values)) and the above Z value calculation function to generate the grid point coordinates of the surface; Drawing surfaces: Write shader programs (vertex shaders and fragment shaders) to control the appearance and lighting effects of 3D models, and use 3D rendering engines (such as OpenGL, DirectX, etc.) to render 3D models to the screen; Interactive computing: mainly involves operation response and data update. For example, when the user interacts with the visualization interface through a mouse or touch screen, the system needs to calculate the response of the user operation (such as rotation, zooming, and translation) and update the data in real time to reflect the impact of these operations on the visualization results. For example, the user can perform treatment operations on the virtual patient and observe its reaction and effect. The real-time data obtained is used as the basis for driving the 3D visualization virtual patient. These data are input into pre-designed algorithms and models to dynamically update the status and behavior of the virtual patient. Using advanced 3D modeling and rendering technology, the real-time status of the virtual patient is displayed in a realistic 3D form, including the patient's vital signs, wound condition, appearance, movements, expressions and other details. Real-time visualization and update, and adjust the treatment strategy based on feedback.
9. The method for generating 3D virtual visualization patients of disaster cases based on data drive according to claim 8, characterized in that: The use of algorithms and models helps users collect, analyze and interpret various data, thereby providing a basis and suggestions for decision-making: z_j^{(2)}=\sum_{i=1}^{n}w_{ij}^{(2)}\cdot a_i^{(1)}+b_j^{(2)} a_j^{(2)}=\sigma(z_j^{(2)}) zj(2) is the weighted sum of the jth neuron in the hidden layer, wij(2) is the weight from the i-th neuron in the input layer to the j-th neuron in the hidden layer, ai(1) is the output (i.e., input data) of the i-th neuron in the input layer, and bj(2) is the bias term of the j-th neuron in the hidden layer.
10. The system for generating a 3D virtual visualization patient method for disaster cases based on data drive according to any one of claims 1 to 9, characterized in that: It includes a management-end browser module, which is bidirectionally electrically connected to the system module, and the system module is bidirectionally electrically connected to the user-end program module. The system module includes a 3D resource module, a Mesh splicing and fusion module, a real-time rendering module, a simulation interaction module, a network communication module, a data service storage module and an auxiliary AI module. The management-end browser module establishes a data interface communication with the client to save the generated patient data group information. The system module obtains database data, and then edits and communicates the data, edits and defines digital patients to generate patient data groups, and the system module generates or updates digital patients and renders them in real time, and interacts with visualized patients. The user-end program module communicates with the server and feeds back interactive information.
