Real-scene interactive anesthesia simulation training management method and system based on VR technology

Abstract

The invention discloses a VR technology-based live-action interactive anesthesia simulation training management method and system, and relates to the technical field of intelligent medical simulation, and the method comprises the steps: based on a patient MRI / CT image and a medical record, generating a virtual three-dimensional solution cutaway view through a generative adversarial network, employing a linear regression method, configuring a linear mapping function, and generating virtual patient physiological parameters, the virtual patient physiological parameters are embedded into the virtual three-dimensional solution cutaway view to obtain a virtual patient model; the interaction force of the puncture needle in the virtual patient model is calculated in real time through macroscopic finite element analysis and microcosmic molecular dynamics; based on the interaction force and the drug effect curve, an interaction force deviation value and an anesthetic deviation value are calculated, and a trainee is prompted in real time through a VR head-mounted display and a touch glove. By generating the virtual patient model, accurate modeling of individualized anatomical structures and physiological parameters is realized, and the authenticity and personalization of patient simulation are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical simulation technology, and in particular to a real-life interactive anesthesia simulation training management method and system based on VR technology. Background Art

[0002] In recent years, with the rapid development of virtual reality (VR) technology and its in-depth application in medical education, VR-based medical simulation training systems have gradually become an important supplement to medical practice, especially in the field of anesthesiology. Traditional clinical training is limited by bottlenecks such as the scarcity of real patient resources, high operational risks, and ethical issues, prompting researchers to explore new models for anesthesia skills training using high-fidelity simulation environments.

[0003] Currently, relevant technologies mainly include 3D reconstruction of medical images, immersive interaction, tactile feedback devices, and dynamic simulation of physiological parameters. For example, medical image processing technology can already convert patients' MRI or CT scans into 3D visualization models to support surgical planning. At the same time, the development of tactile feedback devices also allows users to experience a near-realistic sense of touch in virtual environments. However, in terms of virtual patient modeling, although existing technologies can generate three-dimensional anatomical models based on medical images, these models often lack personalized features and cannot fully reflect the differences between individuals. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a real-life interactive anesthesia simulation training management method based on VR technology to solve the problem that virtual patient models lack personalization and are difficult to reflect individual differences.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a real-life interactive anesthesia simulation training management method based on VR technology, which comprises: Based on the patient's MRI / CT images and medical records, a generative adversarial network is used to generate a virtual 3D anatomical view. A linear regression method is used to configure a linear mapping function to generate the virtual patient's physiological parameters. The virtual patient's physiological parameters are then embedded in the virtual 3D anatomical view to obtain a virtual patient model. The interaction force and tissue node displacement of the puncture needle in the virtual patient model are calculated in real time through macroscopic finite element analysis and microscopic molecular dynamics. The interaction force includes macroscopic resistance and microscopic resistance. Trainees use VR headsets and tactile gloves to perform anesthesia puncture operations, using quantum algorithms to solve nonlinear differential equations in pharmacodynamics and generate pharmacodynamic curves; Based on the drug efficacy curve, the linear mapping method is used to calculate the changes in the patient's physiological parameters and update the virtual patient model; Based on the interaction force and drug efficacy curves, the interaction force deviation value and anesthetic drug deviation value are calculated, and real-time prompts are given to the trainee through the VR headset and tactile gloves.

[0007] As a preferred solution of the VR-based real-life interactive anesthesia simulation training management method of the present invention, wherein: the virtual three-dimensional anatomical view is generated by using a generative adversarial network, and the specific steps are as follows: Use DICOM standard parsing tools to preprocess patient MRI / CT images to obtain three-dimensional image tensors; Based on 3D CycleGAN, a generative adversarial network architecture is configured, wherein the generative adversarial network architecture includes a generator, an inverse generator, and a discriminator; The 3D image tensor is used as the source domain input and a virtual 3D anatomical view is generated through the generator; The virtual 3D anatomical view generated by the generator is converted back into a 3D image tensor through an inverse generator; Using the discriminator, the virtual 3D anatomical view is compared with the real 3D anatomical view, and the authenticity probability of the virtual 3D anatomical view being the real 3D anatomical view is output. Calculate the Cycle consistency loss value of the generator and inverse generator, as well as the adversarial loss value of the discriminator; Based on the Cycle consistency loss and adversarial loss, the weighted summation method is used to calculate the total loss of the generative adversarial network and update the parameters of the generative adversarial network. Based on the convergence criterion of the generative adversarial network, a total loss value threshold is set. When the total loss value of the generative adversarial network is less than the total loss value threshold, the training of the generative adversarial network is completed and a virtual three-dimensional anatomical view is output.

[0008] As a preferred solution of the VR-based real-life interactive anesthesia simulation training management method of the present invention, the specific steps of generating the virtual patient physiological parameters are as follows: Clean and normalize the medical records to generate a three-dimensional conditional vector; Based on the three-dimensional conditional vector, the linear regression method is adopted to configure the linear mapping function to generate the physiological parameters of the virtual patient.

[0009] As a preferred solution of the VR-based real-life interactive anesthesia simulation training management method of the present invention, wherein: the virtual patient physiological parameters are embedded in the virtual three-dimensional anatomical view to obtain a virtual patient model, the specific steps are as follows: Using the Marching Cubes algorithm, the virtual 3D anatomical view is converted into a 3D mesh model; Use JSON serialization tools to convert the virtual patient's physiological parameters into key-value pairs and create a physiological parameter attribute table; Through the vertex attribute editing function of Unreal Engine 5, the physiological parameter attribute table is embedded into the 3D mesh model to obtain a virtual patient model.

[0010] As a preferred solution of the VR-based real-life interactive anesthesia simulation training management method of the present invention, the specific steps of calculating the interaction force of the puncture needle on the virtual patient model are as follows: The elastic modulus of the tissue in the anesthesia puncture area is set according to the Young's modulus of soft tissue in biomechanics; Based on the elastic modulus of the tissue in the anesthesia puncture area, the macroscopic resistance was calculated by finite element analysis using Hooke's law; Based on the molecular structure of collagen fibers in the protein database, the collagen fiber molecular model was constructed using GROMACS software; Traverse the atomic pairs in the collagen fiber molecular model, use the Lennard-Jones potential function to calculate the interatomic force, accumulate the interatomic forces of all atomic pairs, and obtain the microscopic resistance; Based on macro resistance and micro resistance, the interaction force is calculated using the weighted summation method.

[0011] As a preferred solution of the VR-based real-life interactive anesthesia simulation training management method of the present invention, the specific steps of generating the drug efficacy curve are as follows: Real-time recording of the trainer's anesthetic injection dose and injection rate during the training process; Configure the variational quantum algorithm in the Qiskit framework to transform the nonlinear differential equation of pharmacodynamics into a Hamiltonian; Based on the VQE algorithm, the discretization method is used to calculate the central chamber anesthetic concentration and the peripheral chamber anesthetic concentration of each step, and the concentration time series of the central chamber and the peripheral chamber are obtained; The JSON serialization tool was used to integrate the time series of anesthetic agent concentrations in the central and peripheral chambers to generate the drug effect curve.

[0012] As a preferred solution of the VR-based real-life interactive anesthesia simulation training management method of the present invention, the specific steps of updating the virtual patient model are as follows: Extract the central chamber anesthetic concentration from the pharmacodynamic curve, construct the pharmacodynamic-physiological response mapping function based on the linear mapping method, and update the virtual patient's physiological parameters; The updated physiological parameters of the virtual patient are embedded into the three-dimensional mesh model to update the virtual patient model.

[0013] In a second aspect, the present invention provides a real-life interactive anesthesia simulation training management system based on VR technology, comprising a patient virtual module, a calibration module, an interactive force module, a drug efficacy curve module, a model update module, and a prompt module. The patient virtualization module is used to generate a virtual three-dimensional anatomical view based on the patient's MRI / CT images and medical records using a generative adversarial network, adopt a linear regression method, configure a linear mapping function, generate virtual patient physiological parameters, and embed the virtual patient physiological parameters into the virtual three-dimensional anatomical view to obtain a virtual patient model; The interaction force module is used to calculate the interaction force and tissue node displacement of the puncture needle in the virtual patient model in real time through macroscopic finite element analysis and microscopic molecular dynamics, wherein the interaction force includes macroscopic resistance and microscopic resistance; The pharmacodynamic curve module is used to train people to perform anesthesia puncture operations using VR headsets and tactile gloves, using quantum algorithms to solve nonlinear pharmacodynamic differential equations and generate pharmacodynamic curves; The model updating module is used to calculate the changes in the patient's physiological parameters based on the drug efficacy curve and adopt a linear mapping method to update the virtual patient model; The prompt module is used to calculate the interaction force deviation value and the anesthetic drug deviation value based on the interaction force and the drug efficacy curve, and provide real-time prompts to the trainee through the VR head display and tactile gloves.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the real-life interactive anesthesia simulation training management method based on VR technology as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the real-life interactive anesthesia simulation training management method based on VR technology as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention generates a virtual patient model through a generative adversarial network, realizes the precise modeling of individualized anatomical structures and physiological parameters, significantly improves the authenticity and personalization of patient simulation, and adopts macroscopic finite element analysis combined with microscopic molecular dynamics to calculate the interaction force and tissue deformation during the puncture process, which not only captures the macroscopic mechanical behavior of soft tissue, but also takes into account the nonlinear response of microstructures such as collagen fibers, providing a more realistic tactile feedback experience, and uses quantum algorithms to efficiently solve pharmacodynamic differential equations, generate pharmacodynamic curves in real time and update the changes in the physiological state of virtual patients, ensuring that the dynamic simulation of drug effects is accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a flowchart of the management method of real-life interactive anesthesia simulation training based on VR technology.

[0019] Figure 2 This is a module diagram of the real-life interactive anesthesia simulation training management system based on VR technology.

[0020] Figure 3 Flowchart for generating virtual 3D anatomical views using generative adversarial networks.

[0021] Figure 4 A flow chart for embedding virtual patient physiological parameters into a virtual three-dimensional anatomical view to obtain a virtual patient model. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides a real-life interactive anesthesia simulation training management method based on VR technology, comprising the following steps: S1. Based on the patient's MRI / CT images and medical records, a generative adversarial network is used to generate a virtual three-dimensional anatomical view. A linear regression method is used to configure a linear mapping function to generate virtual patient physiological parameters. The virtual patient physiological parameters are embedded in the virtual three-dimensional anatomical view to obtain a virtual patient model.

[0026] Extract patients' MRI / CT images and medical records from the hospital's local server; Preprocess patient MRI / CT images using DICOM standard parsing tools, such as the ITK library; Specifically, the DICOM header information is verified, including the DICOM header information containing the patient ID, scanning date, and modality type. The patient identification information is removed by replacing the patient ID with a random UUID string to obtain the MRI / CT image matrix. The pixel values of the MRI / CT image matrix are linearly mapped to the range of 0-1 using the maximum-minimization method to generate a normalized MRI / CT image matrix. A three-dimensional Gaussian filtering algorithm is applied to perform a convolution operation on the normalized MRI / CT image matrix to generate a smoothed filtered MRI / CT image matrix. The image-to-tensor function of the ITK library is used to convert the preprocessed patient MRI / CT image into a three-dimensional tensor format to obtain a three-dimensional image tensor. Clean and normalize the medical records to generate a three-dimensional conditional vector; Specifically, the data in the medical records are normalized using the maximum-minimum method. For example, weight is scaled to 0-1, gender is encoded as a binary value, such as 1 for male and 0 for female, and allergy history is encoded as a binary value, such as 1 for presence and 0 for absence. A three-dimensional conditional vector is generated to process missing values in the medical records. For example, for missing fields in the medical records, such as weight or allergy history, which are empty, a default value filling method is used. Based on the average adult weight, the default value of weight is set to 70 kg, the default value of gender is set to 0, and the default value of allergy history is set to 0. The data in the medical records are standardized, and the unit of weight is uniformly converted to kilograms. For records in non-kilogram units, the weight unit conversion formula is used for conversion. The integrity of the JSON format of the medical records is verified to ensure the field name. Based on 3D CycleGAN, configure the generative adversarial network architecture; Specifically, the generative adversarial network includes a generator, an inverse generator, and a discriminator. The generator network is a U-Net architecture with 32 layers, including 16 downsampling layers and 16 upsampling layers. The convolution kernel size of each layer is 3×3×3, the stride is 1, the padding is 1, and the activation function is ReLU. The structure of the inverse generator is the same as the generator. The discriminator network is a PatchGAN architecture with 5 convolution layers. The convolution kernel size of each layer is 4×4×4, the stride is 2, the receptive field is 70×70×70, the activation function is LeakyReLU, and Xavier is used to initialize the weights of the generative adversarial network. The three-dimensional image tensor is used as the source domain input, and the forward propagation calculation is performed. The 16 downsampling layers of the generator are used, for example, each downsampling reduces the resolution by 2 times, and the features of the MRI / CT scan data are extracted. The 16 upsampling layers are used, for example, each upsampling increases the resolution by 2 times, and a virtual three-dimensional anatomical view is generated. For example, the size is 512×512×256, and the pixel values are normalized to 0-1, including the distribution of skin, fat, muscle and blood vessels. The virtual three-dimensional anatomical view generated by the generator is converted back to a three-dimensional image tensor through the inverse generator. The cycle consistency error between the generator and the inverse generator is calculated, that is, the generated data, such as the generated results of the generator and the inverse generator, such as the virtual three-dimensional anatomical view and the three-dimensional image tensor and the original data, such as the input of the generator and the inverse generator, such as the sum of the absolute values of the pixel differences between the three-dimensional image tensor and the virtual three-dimensional anatomical view, is used as the generation error. The calculation formula is as follows: ; in, Indicates the Cycle consistency loss value, represents a 3D image tensor, represents the inverse generator, Represented by the generator, represents a virtual 3D anatomical view, and It's not a corresponding relationship. is generated by the generator based on different 3D image tensors (non ) generated for unsupervised learning, and denote the expected values of the 3D image tensor and the virtual 3D anatomical view, respectively. and Respectively express and The batch pixel average of Further, Indicates that it passes the generator first The three-dimensional image tensor Convert to virtual 3D anatomical view (non ), and then through the inverse generator, the virtual 3D anatomical view is converted back to a 3D image tensor , Similarly, since traditional generative adversarial networks require paired data for supervised learning, the virtual 3D anatomical views generated in this scheme lack paired training data, that is, a direct mapping between the 3D image tensor and the corresponding virtual 3D anatomical view. Therefore, through the inverse generator, a bidirectional mapping is performed to achieve unsupervised learning, thereby improving the quality and diversity of the generated virtual 3D anatomical views. In the discriminator, the virtual 3D anatomical view is compared with the real 3D anatomical view, and the authenticity probability of the virtual 3D anatomical view being the real 3D anatomical view is output. The adversarial loss value is calculated as follows: ; in, represents the adversarial loss value, express The discriminator determines the virtual 3D anatomical view (generated by the generator based on different 3D image tensors (non ) generates) the probability of authenticity, Representation discriminator decision The probability of truth; The total loss value of the generative adversarial network is calculated by the weighted sum method, and the calculation formula is as follows: ; in, is the total loss value, is the weight of the Cycle consistency loss value, the value range is , To counter the loss weight, the value range is ,and ; Based on the total loss value, the Adam optimizer is used to calculate the weight increment of the generative adversarial network and update the weight of the generative adversarial network. The calculation formula is as follows: ; in, represents the weight increment of the generative adversarial network, is the learning rate, Initialize the weights of the generated adversarial network; Based on the convergence criteria of the generative adversarial network, such as Nash equilibrium and early stopping, a total loss threshold is set. When the total loss of the generative adversarial network is less than the total loss threshold, the training of the generative adversarial network is completed, and a virtual 3D anatomical view is output. Based on the three-dimensional conditional vector, the linear regression method is used to configure the linear mapping function to generate the virtual patient's physiological parameters, where the virtual patient's physiological parameters include heart rate, blood pressure (systolic pressure) and respiratory rate. The calculation formula is as follows: ; in, Indicates the Virtual patient physiological parameters, The index variable representing the physiological parameters of the virtual patient, specifically, when hour, Indicates heart rate, when hour, Indicates blood pressure, when hour, Represents the respiratory rate, Indicates the The weight coefficient of the physiological parameters of a virtual patient, is the normalized weight value, Indicates the The gender coding coefficients of the physiological parameters of virtual patients, Code for gender, Indicates the The allergy history coding coefficient of the physiological parameters of a virtual patient, Code for allergy history, For the constant term of the physiological parameters of a virtual patient, 、 and The value range is , The value range is ; Using the Marching Cubes algorithm, the virtual 3D anatomical view is converted into a 3D mesh model; Specifically, the tensor reading function of the VTK library is used to convert the virtual 3D anatomical view into a voxel grid format that can be processed by the Marching Cubes algorithm, and the equal face threshold is set, for example, fixed to 0.5, and the range is , representing the boundary of the anatomical structure, such as the skin-fat interface, setting the vertex accuracy of the triangular mesh, initializing the lookup table of the Marching Cubes algorithm, where the lookup table contains 256 voxel configurations, each configuration defines 0-4 triangles, using the Marching Cubes algorithm to traverse each voxel of the virtual three-dimensional anatomical view in the voxel grid format, comparing the 8 vertex pixel values of each voxel with the isosurface threshold of 0.5, determining the configuration index of the voxel, generating triangles according to the lookup table, where each triangle contains 3 vertices, and the vertex coordinates are calculated by linear interpolation, merging all triangles, and outputting a three-dimensional mesh model; The three-dimensional mesh model is smoothed using the Laplace smoothing method; Using Unreal Engine 5's texture generation tool, initialize the texture map, set the background color of the texture map to a neutral gray to avoid subsequent color deviation, and assign different areas to the skin, fat, muscle, and blood vessels of the 3D mesh model; According to the pixels of each voxel in the 3D mesh model, the voxel labeling method is used to assign a label value to the pixels of each voxel. For example, the skin label value is , the fat label value is , the muscle label value is , the vessel label value is , Through the UV mapping method, the pixel label value is projected to the texture map pixel corresponding to the UV coordinate to generate the texture area division. The RGB color is set according to the pixel label value. The skin area is set to semi-transparent, for example, the RGB value is 200, 180, 150, and the transparency is 0.5. The fat area is set to yellow, for example, the RGB value is 240, 220, 100, and the transparency is 1.0. The blood vessel area is set to red, for example, the RGB value is 200, 50, 50, and the transparency is 1.0. Linear interpolation is used to smooth the color transition of adjacent texture areas, and the texture map after color assignment is output; Using the material binding function of Unreal Engine 5, create material instances and assign materials to the skin, fat, muscle, and blood vessel areas of the 3D mesh model. Set material parameters. For example, enable subsurface scattering for the skin material, with a scattering coefficient of 0.2 and a range of 0-1; enable diffuse reflection for the fat material, with a reflectivity of 0.4; enable metallic luster for the muscle material, with a metallicity of 0.1; and enable self-illumination for the blood vessel material, with a luminous intensity of 0.3. Use UV mapping to map the color-assigned texture maps to the corresponding areas of the 3D mesh model. Using JSON serialization tools, the virtual patient's physiological parameters were converted into key-value pairs to create a physiological parameter attribute table. Using the vertex attribute editing function of Unreal Engine 5, the physiological parameter attribute table was embedded into the 3D mesh model to obtain a virtual patient model. It should be noted that the above information is obtained with the user's consent and is used for legal purposes.

[0027] S2. The interaction force and tissue node displacement of the puncture needle in the virtual patient model are calculated in real time through macroscopic finite element analysis and microscopic molecular dynamics. The interaction force includes macroscopic resistance and microscopic resistance.

[0028] Divide the anesthesia puncture area in the virtual patient model and determine the anesthesia puncture position; The elastic modulus of the skin, fat, and muscle in the anesthesia puncture area is set based on the Young's modulus of soft tissue in biomechanics, for example, the skin elastic modulus MPa, fat elastic modulus MPa, muscle elastic modulus MPa, based on Hooke's law, the macro resistance and tissue node displacement were calculated by finite element analysis; Based on protein databases, such as Protein Data Bank (PDB), human skin collagen is screened and entries containing complete triple helical structures are selected to obtain the molecular structure of collagen fibers; Using visual molecular dynamics software, such as VMD (Visual Molecular Dynamics), to pre-process the collagen fiber molecular structure and convert it into a topology file recognizable by GROMACS software, such as the .top format. The topology file contains atom types, bond connections, and potential energy parameters. Specifically, preprocessing the collagen fiber molecular structure includes repairing missing atoms in the initial collagen fiber molecular structure data, for example, by using the VMD Auto PSF plug-in to add hydrogen atoms, with a target number of about 7,000 hydrogen atoms, adjusting the molecular conformation to ensure the stability of the triple helical structure and a bond length error of <0.01 nanometers, and setting the molecular center coordinates, for example, moving to the origin x=0, y=0, z=0, with an accuracy of 0.001 nanometers; The genbox tool of GROMACS software was used to place the collagen fiber molecular structure in the topology file into a simulation water box, and sodium and chloride ions were added to neutralize the charge to generate a solvated topology file. The steepest descent method was used to minimize the energy of collagen fiber molecules and generate the lowest energy collagen fiber molecular configuration. The gmx check tool of GROMACS software was used to check the integrity of the atomic coordinates, chemical bond stability, and rationality of the molecular configuration of the collagen fiber molecular structure to generate a collagen fiber molecular model. The tissue node displacement is used as a displacement constraint, and the displacement constraint is applied to the boundary atoms of the collagen fiber molecular model; Traverse the atomic pairs in the collagen fiber molecular model, use the Lennard-Jones potential function to calculate the interatomic force, accumulate the interatomic forces of all atomic pairs, and obtain the microscopic resistance; Based on macro resistance and micro resistance, the interaction force is calculated using the weighted summation method; Using the vertex buffer editing function of Unreal Engine 5, the vertices of the triangular mesh of the anesthesia puncture area are traversed, and the coordinates of the triangular mesh vertices of the anesthesia puncture area are updated according to the displacement of the tissue nodes in the anesthesia puncture area in the virtual patient model; Reset the color of the texture map of the anesthesia puncture area, for example, light red, RGB value 220,150,150, transparency 0.8, range , the closer to the anesthesia puncture site, the lower the transparency; Using real-time ray tracing, the virtual patient model is dynamically rendered based on the displacement changes of tissue nodes in the anesthesia puncture area to obtain a tissue deformation view. By generating a tissue deformation view, the tissue changes caused by the puncture operation are reflected in real time, the visual effect of the deformation area is highlighted, the realism of the skin indentation is enhanced, the trainers can be more easily identified by the puncture impact, the interaction is more intuitive, and ultra-high-precision tactile and visual feedback is provided to the trainers, simulating real tissue interactions such as skin elasticity, muscle viscosity, and fiber rupture.

[0029] S3. Trainees perform anesthesia puncture operations through VR headsets and tactile gloves, and use quantum algorithms to solve nonlinear pharmacodynamic differential equations to generate pharmacodynamic curves.

[0030] The virtual patient model is imported into the VR device. The trainee uses the VR headset to confirm the anesthesia puncture point, wears tactile gloves to perform the anesthesia puncture operation, and records the anesthetic injection dose and injection rate in real time. The calculation formula for the pharmacodynamic nonlinear differential equation is as follows: ; ; in, Indicates the concentration of anesthetic agent in the central chamber, indicates the concentration of anesthetic agent in the blood, is the time step, is the metabolic rate, with a value range of , represents the clearance rate of anesthetic agents from the central chamber, is the distribution rate, with a value range of , represents the transfer rate of anesthetic agents from the central chamber to the peripheral chamber, is the return rate, the value range is , represents the rate at which the drug returns from the peripheral chamber to the central chamber, is the concentration of anesthetic agent in the peripheral chamber, which indicates the concentration of anesthetic agent in the tissue. For the injection dose of anesthetic, is the injection rate factor, with a value range of , represents the injection rate of anesthetic agents, is the volume of the central chamber; Configure a variational quantum algorithm, such as the VQE algorithm, in the Qiskit framework, initialize the quantum circuit, set the time step to 0.01 minutes, convert the pharmacodynamic nonlinear differential equation into a Hamiltonian, and use a discretization method, such as the forward Euler method, based on the VQE algorithm to calculate the central and peripheral anesthetic concentrations for each step to obtain the central and peripheral concentration time series. The Runge-Kutta method is used to solve the nonlinear differential equation of pharmacodynamics, and the residual value between the VQE solution and the Runge-Kutta solution is calculated to verify the accuracy of the VQE solution. The calculation formula is as follows: ; in, is the time step The residual value when and The time steps The concentration of anesthetic agent in the central and peripheral ventricles of the VQE solution is and The time steps The concentration of anesthetic agent in the central and peripheral chambers of the Runge-Kutta solution at 37°C; According to the clinical requirement of concentration accuracy based on the pharmacokinetics of anesthetic agents, the error is required to be <1% of the peak concentration, approximately 0.01 mg / L, and the residual threshold is set; If the residual value is greater than the residual threshold, adjust the VQE optimizer learning rate, for example, increase it by 10% each time, from , and recalculate the central chamber anesthetic agent concentration and the peripheral chamber anesthetic agent concentration until the residual value is no greater than the residual threshold; The JSON serialization tool was used to integrate the time series of anesthetic agent concentrations in the central and peripheral chambers to generate a pharmacodynamic curve with time as the horizontal axis and anesthetic agent concentration as the vertical axis.

[0031] S4. Based on the drug efficacy curve, the linear mapping method is used to calculate the changes in the physiological parameters of the virtual patient and update the virtual patient model.

[0032] The concentration of the central anesthetic agent is extracted from the pharmacodynamic curve. Based on the linear mapping method, a pharmacodynamic-physiological response mapping function is constructed to update the physiological parameters of the virtual patient and further update the virtual patient model. The calculation formula is as follows: ; in, After the update Virtual patient physiological parameters, For the The effect coefficient of the physiological parameters of a virtual patient is within the range of ; The updated physiological parameters of the virtual patient are embedded into the three-dimensional mesh model to update the virtual patient model.

[0033] S5. Based on the interaction force and drug efficacy curve, the interaction force deviation value and the anesthetic drug deviation value are calculated, and real-time prompts are given to the trainee through the VR headset and tactile gloves.

[0034] Using the segmented deviation analysis method, the anesthesia puncture process was divided into three stages: skin layer penetration, fat layer penetration, and muscle layer penetration. Based on the elastic modulus and tissue thickness of the corresponding tissue in each anesthesia puncture process, Hooke's law was used to calculate the target interaction force value for each anesthesia puncture process. The absolute value of the difference between the actual interaction force value and the target interaction force value during each anesthesia puncture process is calculated as the interaction force deviation value during each anesthesia puncture process; The real-time central ventricular anesthetic concentration is extracted from the pharmacodynamic curve. The real-time target central ventricular anesthetic concentration is calculated based on the physiological parameters in the virtual patient model and the patient's medical records. The calculation formula is as follows: ; ; in, is the time step The target ventricular anesthetic concentration at The clinically recommended dose is The dose per unit body weight is is the patient's weight; Based on the real-time central ventricular anesthetic concentration and the real-time target central ventricular anesthetic concentration, the anesthetic deviation value is calculated using the root mean square error method; Using Unreal Engine 5, a puncture path is generated based on the trajectory of the needle tip's coordinate changes. The width of the puncture path is the same as the cross-sectional size of the puncture needle. The puncture path is mapped and rendered using the same rendering and mapping methods as the tissue in the virtual patient model, and is highlighted. Highlighting the puncture path does not mean increasing the display brightness of the puncture path, but rather selecting a color with high contrast with the rendering color of the surrounding tissue, such as red with an RGB value of 255,0,0. The transparency setting should be lower than that of the surrounding tissue, and the path should be eye-catching, such as 0.8. The vibration of the HaptX haptic glove is triggered. The larger the interactive force deviation value, the higher the vibration frequency of the HaptX haptic glove. For the deviation value of the anesthetic agent, the VR headset displays the drug efficacy curve and physiological abnormality warnings in real time, such as the text prompt "Drug efficacy exceeds the standard."

[0035] This embodiment also provides a real-life interactive anesthesia simulation training and management system based on VR technology, comprising: a patient virtualization module, an interaction force module, a drug efficacy curve module, a model update module, and a prompt module. The patient virtualization module is used to generate a virtual three-dimensional anatomical view based on the patient's MRI / CT images and medical records using a generative adversarial network, and adopts a linear regression method to configure a linear mapping function to generate virtual patient physiological parameters. The virtual patient physiological parameters are embedded in the virtual three-dimensional anatomical view to obtain a virtual patient model. The interaction force module is used to calculate the interaction force and tissue node displacement of the puncture needle in the virtual patient model in real time through macroscopic finite element analysis and microscopic molecular dynamics. The interaction force includes macroscopic resistance and microscopic resistance. The drug efficacy curve module is used to train people to perform anesthesia puncture operations through a VR headset and tactile gloves, and uses quantum algorithms to solve nonlinear differential equations in pharmacodynamics to generate drug efficacy curves. The model update module is used to calculate changes in patient physiological parameters based on the drug efficacy curve using a linear mapping method and update the virtual patient model. The prompt module is used to calculate interaction force deviation values and anesthetic agent deviation values based on the interaction force and drug efficacy curve, and provide real-time prompts to the trainees through the VR headset and tactile gloves.

[0036] This embodiment also provides a computer device, which is suitable for the real-life interactive anesthesia simulation training management method based on VR technology, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the real-life interactive anesthesia simulation training management method based on VR technology proposed in the above embodiment.

[0037] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0038] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the real-life interactive anesthesia simulation training management method based on VR technology as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0039] In summary, the present invention generates a virtual patient model through a generative adversarial network, achieves accurate modeling of individualized anatomical structures and physiological parameters, significantly improves the authenticity and personalization of patient simulation, and uses macroscopic finite element analysis combined with microscopic molecular dynamics to calculate the interaction force and tissue deformation during the puncture process. It not only captures the macroscopic mechanical behavior of soft tissue, but also considers the nonlinear response of microstructures such as collagen fibers, providing a more realistic tactile feedback experience. With the help of quantum algorithms, the pharmacodynamic differential equations are efficiently solved, the pharmacodynamic curve is generated in real time, and the changes in the physiological state of the virtual patient are updated to ensure that the dynamic simulation of drug effects is accurate.

[0040] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A real-life interactive anesthesia simulation training management method based on VR technology, characterized by: include, Based on the patient's MRI / CT images and medical records, a generative adversarial network is used to generate a virtual 3D anatomical view. A linear regression method is used to configure a linear mapping function to generate the virtual patient's physiological parameters. The virtual patient's physiological parameters are then embedded in the virtual 3D anatomical view to obtain a virtual patient model. The interaction force and tissue node displacement of the puncture needle in the virtual patient model are calculated in real time through macroscopic finite element analysis and microscopic molecular dynamics. The interaction force includes macroscopic resistance and microscopic resistance. Trainees use VR headsets and tactile gloves to perform anesthesia puncture operations, using quantum algorithms to solve nonlinear differential equations in pharmacodynamics and generate pharmacodynamic curves; Based on the drug efficacy curve, the linear mapping method is used to calculate the changes in the patient's physiological parameters and update the virtual patient model; Based on the interaction force and drug efficacy curves, the interaction force deviation value and anesthetic drug deviation value are calculated, and real-time prompts are given to the trainee through the VR headset and tactile gloves.

2. The VR-based real-life interactive anesthesia simulation training management method according to claim 1, characterized in that: The specific steps of using the generative adversarial network to generate a virtual three-dimensional anatomical view are as follows: Use DICOM standard parsing tools to preprocess patient MRI / CT images to obtain three-dimensional image tensors; Based on 3D CycleGAN, a generative adversarial network architecture is configured, wherein the generative adversarial network architecture includes a generator, an inverse generator, and a discriminator; The 3D image tensor is used as the source domain input and a virtual 3D anatomical view is generated through the generator; The virtual 3D anatomical view generated by the generator is converted back into a 3D image tensor through an inverse generator; Using a discriminator, the virtual 3D anatomical view is compared with the real 3D anatomical view, and the authenticity probability of the virtual 3D anatomical view being the real 3D anatomical view is output; Calculate the Cycle consistency loss value of the generator and inverse generator, as well as the adversarial loss value of the discriminator; Based on the Cycle consistency loss and adversarial loss, the weighted summation method is used to calculate the total loss of the generative adversarial network and update the parameters of the generative adversarial network. Based on the convergence criterion of the generative adversarial network, a total loss value threshold is set. When the total loss value of the generative adversarial network is less than the total loss value threshold, the training of the generative adversarial network is completed and a virtual three-dimensional anatomical view is output.

3. The VR-based real-life interactive anesthesia simulation training management method according to claim 2, characterized in that: The specific steps of generating the virtual patient physiological parameters are as follows: Clean and normalize the medical records to generate a three-dimensional conditional vector; Based on the three-dimensional conditional vector, the linear regression method is adopted to configure the linear mapping function to generate the physiological parameters of the virtual patient.

4. The VR-based real-life interactive anesthesia simulation training management method according to claim 3, characterized in that: The virtual patient model is obtained by embedding the virtual patient physiological parameters into the virtual three-dimensional anatomical view. The specific steps are as follows: Using the Marching Cubes algorithm, the virtual 3D anatomical view is converted into a 3D mesh model; Use JSON serialization tools to convert the virtual patient's physiological parameters into key-value pairs and create a physiological parameter attribute table; Through the vertex attribute editing function of Unreal Engine 5, the physiological parameter attribute table is embedded into the 3D mesh model to obtain a virtual patient model.

5. The VR-based real-life interactive anesthesia simulation training management method according to claim 4, characterized in that: The specific steps for calculating the interaction force of the puncture needle on the virtual patient model are as follows: The elastic modulus of the tissue in the anesthesia puncture area is set according to the Young's modulus of soft tissue in biomechanics; Based on the elastic modulus of the tissue in the anesthesia puncture area, the macroscopic resistance was calculated by finite element analysis using Hooke's law; Based on the molecular structure of collagen fibers in the protein database, the collagen fiber molecular model was constructed using GROMACS software; Traverse the atomic pairs in the collagen fiber molecular model, use the Lennard-Jones potential function to calculate the interatomic force, accumulate the interatomic forces of all atomic pairs, and obtain the microscopic resistance; Based on macro resistance and micro resistance, the interaction force is calculated using the weighted summation method.

6. The VR-based real-life interactive anesthesia simulation training management method according to claim 5, characterized in that: The specific steps of generating the drug efficacy curve are as follows: Real-time recording of the trainer's anesthetic injection dose and injection rate during the training process; Configure the variational quantum algorithm in the Qiskit framework to transform the nonlinear differential equation of pharmacodynamics into a Hamiltonian; Based on the VQE algorithm, the discretization method is used to calculate the central chamber anesthetic concentration and the peripheral chamber anesthetic concentration of each step, and the concentration time series of the central chamber and the peripheral chamber are obtained; The JSON serialization tool was used to integrate the time series of anesthetic agent concentrations in the central and peripheral chambers to generate the drug effect curve.

7. The VR-based real-life interactive anesthesia simulation training management method according to claim 6, characterized in that: The specific steps of updating the virtual patient model are as follows: Extract the central chamber anesthetic concentration from the pharmacodynamic curve, construct the pharmacodynamic-physiological response mapping function based on the linear mapping method, and update the virtual patient's physiological parameters; The updated physiological parameters of the virtual patient are embedded into the three-dimensional mesh model to update the virtual patient model.

8. A VR-based real-life interactive anesthesia simulation training management system, based on the VR-based real-life interactive anesthesia simulation training management method according to any one of claims 1 to 7, characterized in that: Including patient virtual module, interaction module, drug efficacy curve module, model update module and prompt module, The patient virtualization module is used to generate a virtual three-dimensional anatomical view based on the patient's MRI / CT images and medical records using a generative adversarial network, adopt a linear regression method, configure a linear mapping function, generate virtual patient physiological parameters, and embed the virtual patient physiological parameters into the virtual three-dimensional anatomical view to obtain a virtual patient model; The interaction force module is used to calculate the interaction force and tissue node displacement of the puncture needle in the virtual patient model in real time through macroscopic finite element analysis and microscopic molecular dynamics, wherein the interaction force includes macroscopic resistance and microscopic resistance; The pharmacodynamic curve module is used to train people to perform anesthesia puncture operations using VR headsets and tactile gloves, using quantum algorithms to solve nonlinear pharmacodynamic differential equations and generate pharmacodynamic curves; The model updating module is used to calculate the changes in the patient's physiological parameters based on the drug efficacy curve and adopt a linear mapping method to update the virtual patient model; The prompt module is used to calculate the interaction force deviation value and the anesthetic drug deviation value based on the interaction force and the drug efficacy curve, and provide real-time prompts to the trainee through the VR head display and tactile gloves.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the real-life interactive anesthesia simulation training management method based on VR technology are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the real-life interactive anesthesia simulation training management method based on VR technology are implemented.

Images