A basic medical virtual experiment teaching method and system

Abstract

This invention relates to the field of virtual reality technology, and more particularly to a method and system for basic medical virtual experimental teaching. The method includes constructing a physiological digital twin model; achieving real-time bidirectional coupling of macroscopic and microscopic physiology through a dynamic simulation engine; compiling trainee operations based on force feedback devices and providing interaction; utilizing a knowledge graph to trace errors and create competency profiles for the operations, and generating personalized feedback. This invention significantly improves the physiological fidelity and teaching relevance of virtual experiments.

Description

Technical Field

[0001] This invention relates to the field of virtual reality technology, and in particular to a method and system for basic medical virtual experiment teaching. Background Technology

[0002] Virtual experimental teaching, as a crucial bridge connecting theoretical knowledge and practical operation, demonstrates significant advantages in improving students' hands-on skills, reducing experimental costs, and ensuring biosafety. Currently, most mainstream virtual experimental platforms are built upon static 3D models or pre-set interactive processes, with their core logic relying on linear operation sequences driven by fixed scripts. However, basic medical experiments (such as biochemical reaction kinetics determination, molecular cloning, cell culture and observation) are inherently highly dynamic and uncertain. Experimental results are affected by various variables such as reagent concentration, temperature control precision, operation timing, and cell state, and students often encounter unexpected operational paths during the exploration process.

[0003] Virtual experiments in the fields of biochemistry, molecular biology, and cell biology particularly emphasize the visualization of microscopic processes and the interpretability of causal logic. These experiments typically involve invisible but deducible dynamic processes such as changes in enzyme-catalyzed reaction rates, DNA fragment amplification efficiency, and cell cycle regulation. Summary of the Invention

[0004] The purpose of this invention is to provide a basic medical virtual experiment teaching method and system to solve the technical problems of distorted physiological feedback mechanisms, lack of personalized teaching guidance, and mismatch between virtual operations and actual physiological changes in virtual experiment teaching.

[0005] One aspect of the present invention provides a virtual experimental teaching method for basic medical sciences, comprising:

[0006] Constructing a physiological digital twin model of the target organism;

[0007] Based on the physiological digital twin model, a dynamic simulation based on a dynamic simulation engine is initiated. The dynamic simulation engine includes at least a mechanical simulation unit, a microscopic cellular and molecular dynamics simulation unit, and a cross-scale data coupling interface.

[0008] The cross-scale data coupling interface enables information interaction between the physical field and the microscopic reaction field. The cross-scale data coupling interface uses the samples of the stress field, temperature field and material concentration field calculated by the mechanical simulation unit as the local boundary conditions of the microscopic cell and molecular dynamics simulation unit, and uses the samples of cell state transition and metabolite generation rate calculated by the microscopic cell and molecular dynamics simulation unit to update the material property parameters of the corresponding tissue voxels in the mechanical simulation unit.

[0009] It receives the operation instructions from the trainee and compiles them into force boundary conditions, heat sources, and material sources applied to specific nodes of the physiological digital twin model. At the same time, it transmits the contact reaction force calculated by the mechanical simulation unit back to the trainee.

[0010] After monitoring the trainees' operations, the physiological digital twin model showed multidimensional changes and conducted online competency assessment. The online competency assessment synchronously recorded the trainees' operation sequence and key physiological indicator time series data, and compared the operation sequence with standard surgical operation procedures based on basic medical knowledge graph. When the physiological indicators exceeded the safety threshold, the most direct abnormal operation was identified and the consequences were predicted.

[0011] Based on the results of the online competency assessment, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structure on a stereoscopic display device and explains the nature of the error, the microscopic damage mechanism, and the macroscopic physiological effects. The intervention instructions adjust the parameters of subsequent virtual experiments according to the learner's competency profile.

[0012] In some embodiments, the process of constructing a physiological digital twin model of the target organism includes at least the following:

[0013] Magnetic resonance imaging and computed tomography data were acquired, and three-dimensional surface reconstruction was performed using the surface element advancement algorithm. The constrained Delaunay triangulation algorithm was used to generate tetrahedral volume meshes, which served as the computational domain for mechanical simulation units.

[0014] The tissue slice images are segmented and feature extracted to obtain cell density, collagen fiber orientation and matrix component ratio. Based on the cell density, collagen fiber orientation and matrix component ratio, a constitutive model characterizing the mechanical and physiological properties of the tissue is established.

[0015] Based on metabolic pathways, signal transduction cascades, and pharmacokinetic numbers from public genomics and proteomics databases, a system of ordinary differential equations was constructed:

[0016]

[0017] Where C is the molecular concentration vector, P is the local physical environment parameter vector, E is the external intervention event encoding vector, and f is the nonlinear biochemical reaction kinetic function;

[0018] By integrating tetrahedral volume meshes, constitutive relation models, and ordinary differential equations, and through data interfaces and time synchronization mechanisms, a physiological digital twin model is constructed.

[0019] In some embodiments, the microscopic cellular and molecular dynamics simulation unit employs the Hodgkin-Huxley model to simulate the electrical activity of nerve cells and uses a multi-compartment pharmacokinetic model to calculate the distribution and metabolism of drugs in different tissues.

[0020] In some embodiments, the cross-scale data coupling interface uses a hierarchical octree data structure to organize physical field data, aggregates the physical field data into a microscopic simulation region as boundary conditions, and uses a homogenization algorithm to transform microscopic discrete events into macroscopic continuous field variables to update material property parameters.

[0021] In some embodiments, the process of receiving the trainee's operation instructions, compiling them into force boundary conditions, heat sources, and material sources applied to specific nodes of the physiological digital twin model, and simultaneously transmitting the contact reaction force calculated by the mechanical simulation unit back to the trainee includes:

[0022] The system acquires the position, posture, and torque information of the force feedback operation device of the trainee, and converts it into motion parameters of the virtual surgical instrument in the model based on the virtual instrument mapper.

[0023] Contact detection is performed based on the intersection test of bounding box hierarchy and triangular facets, and contact events are compiled into different physical boundary conditions according to the interaction type.

[0024] The contact reaction force is transmitted back to the force feedback operation device operated by the trainee through a force feedback rendering algorithm.

[0025] In some embodiments, the operation sequence includes instrument trajectory, force magnitude, duration of action, and interaction type encoding; the key physiological indicator time-series data includes heart rate, blood pressure, blood oxygen saturation, respiratory rate, electroencephalogram power spectral density, and regional tissue perfusion and metabolite concentration.

[0026] In some embodiments, the basic medical knowledge graph is constructed based on a resource description framework and includes at least standard anatomical structure entities, pathophysiological state entities, standard surgical procedure entities, causal relationships, spatial topological relationships, and functional dependencies.

[0027] Another aspect of the present invention provides a basic medical virtual experimental teaching system, comprising:

[0028] The physiological model building module is used to build physiological digital twin models;

[0029] The dynamic simulation engine module, based on the physiological digital twin model, initiates dynamic simulation based on the dynamic simulation engine. The dynamic simulation engine includes at least a mechanical simulation unit, a microscopic cellular and molecular dynamics simulation unit, and a cross-scale data coupling interface.

[0030] The student interaction and operation module is used to receive operation instructions from the trainees and compile them into force boundary conditions, heat sources and material sources applied to specific nodes of the physiological digital twin model. At the same time, it transmits the contact reaction force calculated by the mechanical simulation unit back to the trainees.

[0031] The trainee competency assessment and adaptive intervention module monitors multidimensional changes in the physiological digital twin model after a trainee's operation and conducts online competency assessment. This online assessment synchronously records the trainee's operation sequence and key physiological indicator time-series data, and compares the operation sequence with standard surgical procedures based on a basic medical knowledge graph. When physiological indicators exceed safety thresholds, the module identifies the most direct abnormal operation and predicts its consequences. Based on the online competency assessment results, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structure on a stereoscopic display device and explains the nature of the error, the microscopic damage mechanism, and the macroscopic physiological impact. The intervention instructions adjust subsequent virtual experiment parameters according to the trainee's competency profile.

[0032] In some embodiments, the physiological model construction module acquires image data through a medical digital imaging and communication standard interface, and generates a tetrahedral volume mesh using a surface element advancement algorithm and a constrained Delaunay triangulation algorithm.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. Construct a physiological digital twin model and adopt a dynamic simulation engine to overcome the shortcomings of existing technologies where virtual experimental models are seriously disconnected from real physiological processes, thereby greatly improving the fidelity and immersion of the simulation;

[0035] 2. By introducing a basic medical knowledge graph to compare the operation sequence with the standard surgical procedure, the trainees' operation behavior can be linked with deeper medical knowledge, realizing the shift from "whether the operation is right or wrong" to "why it is wrong", making the teaching feedback more targeted;

[0036] 3. By monitoring and evaluating trainees' operations, a trainee competency profile can be established, and the difficulty of experiments can be adjusted accordingly to achieve personalized teaching and improve the learning outcomes of trainees.

[0037] 4. Link the trainees' operations with microscopic physiological changes to help them understand more intuitively the application of basic medical knowledge in clinical practice. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of the basic medical virtual experiment teaching method of the present invention;

[0040] Figure 2 This is an architecture diagram of the basic medical virtual experimental teaching system of this invention. Detailed Implementation

[0041] The following will be based on embodiments of the present invention. Figures 1-2 The technical solutions in the embodiments of the present invention will be clearly and completely described together. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0042] Partial interpretation:

[0043] 1. The surface-advancement algorithm is a core algorithm for unstructured mesh generation, which mainly constructs the mesh step by step through the expansion of the leading edge surface;

[0044] 2. Constrained Delaunay triangulation is an extension algorithm of basic Delaunay triangulation. Its core is to force the satisfaction of preset geometric constraints (such as specified edges and faces) while preserving the "optimal mesh quality" characteristic of Delaunay triangulation.

[0045] 3. Multicompartment pharmacokinetic models are the core analytical models of pharmacokinetics. By abstracting the body into multiple functional "compartments", they describe the dynamic processes of drug absorption, distribution, metabolism and excretion in the body. They are key tools for drug development and clinical dosing regimen design.

[0046] 4. The bounding box hierarchy is the core spatial indexing structure for accelerating 3D geometric intersection testing. By constructing a hierarchical bounding box for the set of triangular faces in a "coarse-grained → fine-grained" manner, it significantly reduces the calculation of direct intersection of invalid triangular faces. Triangular face intersection testing is the core algorithm for determining whether two triangular faces (or rays, line segments and triangular faces) have spatial overlap.

[0047] 5. In the material point method, the fracture criterion is the core mechanism for simulating the transition of materials from continuous deformation to discontinuous failure.

[0048] Example 1

[0049] This embodiment provides a virtual experimental teaching method for basic medical science, which includes: constructing a physiological digital twin model of the target organism; based on the physiological digital twin model, initiating dynamic simulation based on a dynamic simulation engine, which includes at least a mechanical simulation unit, a microscopic cell and molecular dynamics simulation unit, and a cross-scale data coupling interface; realizing information interaction between the physical field and the microscopic reaction field through the cross-scale data coupling interface, which uses samples of the stress field, temperature field, and substance concentration field calculated by the mechanical simulation unit as local boundary conditions of the microscopic cell and molecular dynamics simulation unit, and uses samples of cell state transitions and metabolic product generation rates calculated by the microscopic cell and molecular dynamics simulation unit to update the material property parameters of the corresponding tissue voxels in the mechanical simulation unit; receiving operation commands from the trainee and compiling them into an application... The physiological digital twin model incorporates force boundary conditions, heat sources, and material sources at specific nodes, while simultaneously transmitting the contact reaction forces calculated by the mechanical simulation unit back to the trainee. After monitoring the trainee's actions, the physiological digital twin model exhibits multidimensional changes and performs online competency assessments. These assessments synchronously record the trainee's operational sequence and key physiological indicators, comparing the sequence with standard surgical procedures based on a basic medical knowledge graph. When physiological indicators exceed safety thresholds, the most direct abnormal operation is identified, and its consequences are predicted. Based on the online competency assessment results, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structures on a stereoscopic display device, explaining the nature of the error, the microscopic damage mechanism, and the macroscopic physiological impact. The intervention instructions adjust subsequent virtual experiment parameters based on the trainee's competency profile.

[0050] The method of this embodiment is further described in detail below:

[0051] First, a physiological digital twin model of the target organism is constructed. Specifically, multi-level biomedical data of the target organism is acquired, including at least three scales: macroscopic anatomical structure data, histological microstructure data, and molecular-level physiological and biochemical pathway data.

[0052] Among them, macroscopic anatomical structure data are obtained through medical digital imaging and communication standard interfaces to acquire magnetic resonance imaging and computed tomography data. After preprocessing, the acquired raw image data is used to perform three-dimensional surface reconstruction using the surface element advancement algorithm to generate surface geometric models of organs, blood vessels and neural networks. Furthermore, the constrained Delaunay triangulation algorithm is used to transform the surface model into a tetrahedral volume mesh. This volume mesh serves as the computational domain for subsequent macroscopic continuum mechanics simulation, and its nodes and elements have spatial coordinates and topological connection relationships.

[0053] Histological microstructure data were obtained through image segmentation and feature extraction of high-resolution tissue slices. The acquired parameters included cell density, collagen fiber orientation, and matrix component ratios for different tissue types. Based on these parameters, a constitutive model characterizing the mechanical and physiological properties of tissues was constructed. This model, based on a nonlinear hyperelastic material model, has a strain energy function composed of multiple invariants, accurately describing the nonlinear response of soft tissues under tensile, compressive, and shear loads.

[0054] Molecular-level physiological and biochemical pathway data were obtained by integrating public genomics and proteomics databases. The focus was on extracting metabolic pathways, signal transduction cascades, and pharmacokinetic parameters relevant to the target experiments. Based on this data, a set of ordinary differential equations was constructed to describe the dynamic changes in the concentrations of key molecules such as calcium ions, adenosine triphosphate (ATP), and reactive oxygen species inside and outside the cell. The set of ordinary differential equations is as follows:

[0055]

[0056] Wherein, C is the molecular concentration vector, P is the local physical environment parameter vector, E is the external intervention event encoding vector, and f is the nonlinear biochemical reaction kinetic function; after completing the modeling of the above three types of data, the tetrahedral volume grid, constitutive relation model and ordinary differential equation system are integrated, and a physiological digital twin model is constructed through data interface and time synchronization mechanism.

[0057] Next, based on the physiological digital twin model, a dynamic simulation based on a dynamic simulation engine is initiated. Specifically, this step is executed by the dynamic simulation engine. The mechanical simulation unit receives the generated tetrahedral volume mesh and performs real-time calculations using a continuum mechanics solver based on the material point method. The advantage of the material point method is that it can naturally handle the deformation and topological changes of tissues caused by cutting, tearing, and suturing operations, solving the problem of numerical instability in the traditional finite element method when mesh distortion occurs.

[0058] The microscopic cellular and molecular dynamics simulation unit uses local physicochemical environmental parameters as input to drive the solution of the aforementioned set of ordinary differential equations. These local physicochemical environmental parameters include at least local temperature, pH value, oxygen partial pressure, and mechanical stress. In some optional implementations, for the simulation of neural electrical activity, the microscopic cellular and molecular dynamics simulation unit adopts the Hodgkin-Huxley model, whose membrane potential dynamic equation is:

[0059]

[0060] Among them, C m V is the membrane capacitance; V is the membrane potential. , , These are the maximum conductivities of the sodium, potassium, and leakage channels, respectively; ENa E K E L I represents the equilibrium potential of the corresponding ion; m, ℎ, and n are gating variables, and I represents the equilibrium potential of the ion. ext It is stimulated by an external electric current.

[0061] For the simulation of drug distribution, the microscopic cell and molecular dynamics simulation unit adopts a multi-compartment pharmacokinetic model, which divides the body into a central compartment and a peripheral compartment, and describes the transport and metabolism of drugs between the compartments through first-order differential equations.

[0062] Next, information exchange between the physical field and the microscopic reaction field is achieved through a cross-scale data coupling interface. Specifically, this interface uses a hierarchical octree data structure to organize the physical field data. At each simulation time step, the stress field, temperature field, and substance concentration field calculated by the mechanical simulation unit are mapped to the leaf nodes of the octree. Subsequently, through a downsampling operation, these physical field data are aggregated to the root node of the corresponding microscopic simulation region as local boundary conditions for microscopic solutions. Furthermore, cell state transition indicators calculated by the microscopic cell and molecular dynamics simulation unit, such as cell viability, lactic acid production rate, and nitric oxide release, are upsampled through a homogenization algorithm.

[0063] It should be noted that the homogenization algorithm is based on the principle of local volume averaging, which transforms discrete events at the microscale into continuous field variables at the macroscale. These continuous field variables are used to update the material property parameters of tissue voxels in real time.

[0064] Then, the system receives the trainee's operation commands and compiles them into force boundary conditions, heat sources, and material sources applied to specific nodes of the physiological digital twin model. Simultaneously, it transmits the contact reaction forces calculated by the mechanical simulation unit back to the trainee. Specifically, the trainee's operating equipment includes a six-DOF force feedback operating device and a high-resolution stereoscopic display device. The six-DOF force feedback operating device captures the position and orientation of its end effector and the torque applied by the trainee in real time. The aforementioned data is transmitted and compiled. It should be noted that the compilation is performed in a compilation module, which can optionally be a compilation module publicly available in the prior art. In this embodiment, the module has a built-in virtual instrument mapper that converts the kinematic parameters of the physical instrument into the corresponding parameters of the virtual surgical instrument in the physiological digital twin model. When the virtual instrument comes into contact with the tissue surface in the model, the module determines the set of contact points based on the intersection test of the bounding box hierarchy and the precise triangular facets. Then, according to the interaction type, the contact event is compiled into different physical boundary conditions. For cutting operations, the compilation module applies a moving crack tip on the contact path to trigger the fracture criterion in the material point method. For electrocautery operations, a heat source term is applied to the contact area, the power density of which is proportional to the torque applied by the trainee. For suturing operations, a nonlinear spring connection is established between the puncture points to simulate the tension of the suture.

[0065] The contact reaction force calculated by the mechanical simulation unit is processed by the force feedback rendering algorithm and then transmitted back to the trainee through the motor of the force feedback operating device.

[0066] Next, after monitoring the trainee's operation, the physiological digital twin model exhibits multidimensional changes and performs online competency assessment. This online assessment synchronously records the trainee's operation sequence and key physiological indicator time-series data, and compares the operation sequence with standard surgical procedures based on a basic medical knowledge graph. When physiological indicators exceed safety thresholds, the most direct abnormal operation is identified and its consequences predicted. Specifically, this step synchronously records two data streams. The first data stream is the trainee's complete operation sequence, including the trajectory of the virtual instrument in the world coordinate system, the magnitude of the normal and tangential forces applied to the tissue, the operation duration, and the interaction type encoding. The second data stream is the key physiological indicator time-series data output by the physiological digital twin model, including heart rate and arterial blood pressure calculated by the cardiovascular sub-model, respiratory rate and tidal volume calculated by the respiratory sub-model, arterial blood oxygen saturation calculated by the blood oxygen sub-model, brain electrical power spectral density calculated by the brain electrical activity sub-model, and tissue blood flow and metabolite concentration in specific anatomical regions such as the liver or kidney calculated by the local perfusion sub-model.

[0067] Furthermore, the knowledge graph comprises four main entities: standard anatomical structures, such as the aorta and sciatic nerve; pathophysiological states, such as hypotension and metabolic acidosis; standard surgical procedures, such as "laparoscopic cholecystectomy"; and causal relationships, spatial topological relationships, and functional dependencies. First, this step uses a dynamic time warping algorithm to align the trainee's real-time operational sequence with the corresponding standard operating procedures in the knowledge graph, calculating an operational deviation metric. This deviation metric comprehensively considers trajectory deviation distance, force overload degree, and operational timing misalignment. When any monitored physiological indicator exceeds a preset safety threshold, such as blood oxygen saturation below 90%, a reverse lookup is immediately initiated. Specifically, using Spark query language, the algorithm starts from the abnormal physiological state entity in the knowledge graph, tracing upwards along the causal predicate until the most direct operational error node is located. For example, if blood pressure suddenly drops, the reasoning path might be "sudden drop in blood pressure" ← "massive bleeding" ← "accidental severing of the hepatic artery" ← "instrument trajectory deviates from the gallbladder triangle". Meanwhile, based on the predictive capabilities of the physiological digital twin model, the cascading pathophysiological consequences of this error within the next 30 seconds can be deduced, such as massive bleeding leading to insufficient tissue perfusion, which in turn causes lactic acid accumulation and metabolic acidosis.

[0068] Finally, based on the results of the online competency assessment, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structures on a stereoscopic display device, explaining the nature of the error, the microscopic damage mechanism, and the macroscopic physiological effects. The intervention instructions adjust the parameters of subsequent virtual experiments based on the learner's competency profile. Specifically, at the visual level, the stereoscopic display device renders the damaged key anatomical structures in a semi-transparent, high-brightness color, such as displaying the mistakenly severed hepatic artery with a red pulsating light effect, and overlaying a perspective view to show its spatial relationship with the surrounding bile ducts. In a fixed area of ​​the display interface, the nature of the error is precisely explained in text form, such as the instrument trajectory deviating from the safe zone of the gallbladder triangle, mistakenly severing a branch of the right hepatic artery. The microscopic mechanism is then explained: arterial rupture leads to loss of local perfusion pressure, causing hepatocytes to initiate apoptosis within 5 minutes due to hypoxia, and the macroscopic effects: an estimated 20% decrease in intracardiac output within 10 seconds, triggering systemic hypotension.

[0069] At the auditory level, the synthesized speech delivers the aforementioned key information in a calm and clear tone.

[0070] Optionally, after trainees have accumulated operational data from multiple experiments, a Bayesian network is used to construct a trainee competency profile. This profile includes quantitative evaluation values ​​across multiple dimensions, such as proficiency in instrument operation, anatomical spatial awareness, sensitivity to physiological indicator monitoring, and emergency decision-making response time.

[0071] Example 2

[0072] Based on the same inventive concept as the basic medical virtual experimental teaching method in Embodiment 1 above, such as Figure 2 As shown, this invention also provides a basic medical virtual experimental teaching system, including a physiological model construction module for data connection, a dynamic simulation engine module, a student interaction and operation module, and a student ability assessment and adaptive intervention module. All of the aforementioned modules are executed by a processor, which can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.; the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0073] Specifically,

[0074] The physiological model building module is used to build physiological digital twin models;

[0075] The dynamic simulation engine module, based on the physiological digital twin model, initiates dynamic simulation based on the dynamic simulation engine. The dynamic simulation engine includes at least a mechanical simulation unit, a microscopic cellular and molecular dynamics simulation unit, and a cross-scale data coupling interface.

[0076] The student interaction and operation module is used to receive operation instructions from the trainees and compile them into force boundary conditions, heat sources and material sources applied to specific nodes of the physiological digital twin model. At the same time, it transmits the contact reaction force calculated by the mechanical simulation unit back to the trainees.

[0077] The trainee competency assessment and adaptive intervention module monitors multidimensional changes in the physiological digital twin model after a trainee's operation and conducts online competency assessment. This online assessment synchronously records the trainee's operation sequence and key physiological indicator time-series data, and compares the operation sequence with standard surgical procedures based on a basic medical knowledge graph. When physiological indicators exceed safety thresholds, the module identifies the most direct abnormal operation and predicts its consequences. Based on the online competency assessment results, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structure on a stereoscopic display device and explains the nature of the error, the microscopic damage mechanism, and the macroscopic physiological impact. The intervention instructions adjust subsequent virtual experiment parameters according to the trainee's competency profile.

[0078] The physiological model construction module acquires image data through medical digital imaging and communication standard interfaces, and generates tetrahedral volume meshes using the surface element advancement algorithm and the constrained Delaunay triangulation algorithm.

[0079] The specific examples of the basic medical virtual experiment teaching method in the aforementioned Embodiment 1 are also applicable to the basic medical virtual experiment teaching system of this embodiment, so for the sake of brevity, they will not be described in detail here.

[0080] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0081] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A virtual experimental teaching method for basic medical sciences, characterized in that, include: Constructing a physiological digital twin model of the target organism; Based on the physiological digital twin model, a dynamic simulation based on a dynamic simulation engine is initiated. The dynamic simulation engine includes at least a mechanical simulation unit, a microscopic cellular and molecular dynamics simulation unit, and a cross-scale data coupling interface. The cross-scale data coupling interface enables information interaction between the physical field and the microscopic reaction field. The cross-scale data coupling interface uses the samples of the stress field, temperature field and material concentration field calculated by the mechanical simulation unit as the local boundary conditions of the microscopic cell and molecular dynamics simulation unit, and uses the samples of cell state transition and metabolite generation rate calculated by the microscopic cell and molecular dynamics simulation unit to update the material property parameters of the corresponding tissue voxels in the mechanical simulation unit. It receives the operation instructions from the trainee and compiles them into force boundary conditions, heat sources, and material sources applied to specific nodes of the physiological digital twin model. At the same time, it transmits the contact reaction force calculated by the mechanical simulation unit back to the trainee. After monitoring the trainees' operations, the physiological digital twin model showed multidimensional changes and conducted online competency assessment. The online competency assessment synchronously recorded the trainees' operation sequence and key physiological indicator time series data, and compared the operation sequence with standard surgical operation procedures based on basic medical knowledge graph. When the physiological indicators exceeded the safety threshold, the most direct abnormal operation was identified and the consequences were predicted. Based on the results of the online competency assessment, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structure on a stereoscopic display device and explains the nature of the error, the microscopic damage mechanism, and the macroscopic physiological effects. The intervention instructions adjust the parameters of subsequent virtual experiments according to the learner's competency profile.

2. The method according to claim 1, characterized in that, The process of constructing a physiological digital twin model of the target organism includes at least the following: Magnetic resonance imaging and computed tomography data were acquired, and three-dimensional surface reconstruction was performed using the surface element advancement algorithm. The constrained Delaunay triangulation algorithm was used to generate tetrahedral volume meshes, which served as the computational domain for mechanical simulation units. The tissue slice images are segmented and feature extracted to obtain cell density, collagen fiber orientation and matrix component ratio. Based on the cell density, collagen fiber orientation and matrix component ratio, a constitutive model characterizing the mechanical and physiological properties of the tissue is established. Based on metabolic pathways, signal transduction cascades, and pharmacokinetic parameters from public genomics and proteomics databases, a system of ordinary differential equations was constructed: Where C is the molecular concentration vector, P is the local physical environment parameter vector, E is the external intervention event encoding vector, and f is the nonlinear biochemical reaction kinetic function; By integrating tetrahedral volume meshes, constitutive relation models, and ordinary differential equations, and through data interfaces and time synchronization mechanisms, a physiological digital twin model is constructed.

3. The method according to claim 2, characterized in that, The microscopic cellular and molecular dynamics simulation unit uses the Hodgkin-Huxley model to simulate the electrical activity of nerve cells and employs a multi-compartment pharmacokinetic model to calculate the distribution and metabolism of drugs in different tissues.

4. The method according to claim 3, characterized in that, The cross-scale data coupling interface uses a hierarchical octree data structure to organize physical field data, aggregates the physical field data into a microscopic simulation region as boundary conditions, and uses a homogenization algorithm to transform microscopic discrete events into macroscopic continuous field variables to update material property parameters.

5. The method according to claim 4, characterized in that, The process of receiving the trainee's operation commands, compiling them into force boundary conditions, heat sources, and material sources applied to specific nodes of the physiological digital twin model, and simultaneously transmitting the contact reaction forces calculated by the mechanical simulation unit back to the trainee includes: The system acquires the position, posture, and torque information of the force feedback operation device of the trainee, and converts it into motion parameters of the virtual surgical instrument in the model based on the virtual instrument mapper. Contact detection is performed based on the intersection test of bounding box hierarchy and triangular facets, and contact events are compiled into different physical boundary conditions according to the interaction type. The contact reaction force is transmitted back to the force feedback operation device operated by the trainee through a force feedback rendering algorithm.

6. The method according to claim 1, characterized in that, The operation sequence includes instrument trajectory, force magnitude, duration of action, and interaction type encoding; the key physiological indicators time series data include heart rate, blood pressure, blood oxygen saturation, respiratory rate, EEG power spectral density, and regional tissue perfusion and metabolite concentration.

7. The method according to claim 1, characterized in that, The basic medical knowledge graph is constructed based on a resource description framework and includes at least standard anatomical structure entities, pathophysiological state entities, standard surgical procedure entities, causal relationships, spatial topological relationships, and functional dependencies.

8. A basic medical virtual experimental teaching system, characterized in that, include: The physiological model building module is used to build physiological digital twin models; The dynamic simulation engine module, based on the physiological digital twin model, initiates dynamic simulation based on the dynamic simulation engine. The dynamic simulation engine includes at least a mechanical simulation unit, a microscopic cellular and molecular dynamics simulation unit, and a cross-scale data coupling interface. The student interaction and operation module is used to receive operation instructions from the trainees and compile them into force boundary conditions, heat sources and material sources applied to specific nodes of the physiological digital twin model. At the same time, it transmits the contact reaction force calculated by the mechanical simulation unit back to the trainees. The trainee competency assessment and adaptive intervention module monitors multidimensional changes in the physiological digital twin model after a trainee's operation and conducts online competency assessment. This online assessment synchronously records the trainee's operation sequence and key physiological indicator time-series data, and compares the operation sequence with standard surgical procedures based on a basic medical knowledge graph. When physiological indicators exceed safety thresholds, the module identifies the most direct abnormal operation and predicts its consequences. Based on the online competency assessment results, personalized teaching feedback and intervention instructions are generated. The personalized teaching feedback visualizes the damaged anatomical structure on a stereoscopic display device and explains the nature of the error, the microscopic damage mechanism, and the macroscopic physiological impact. The intervention instructions adjust subsequent virtual experiment parameters according to the trainee's competency profile.

9. The system according to claim 8, characterized in that, The physiological model construction module acquires image data through a medical digital imaging and communication standard interface, and generates a tetrahedral volume mesh using a surface element advancement algorithm and a constrained Delaunay triangulation algorithm.

Images