Wearable VR system for health education

Abstract

The invention discloses a wearable VR system for health education, and relates to the technical field of virtual reality medical education, and the system comprises a knowledge modeling unit for constructing a medical knowledge graph, a resource preprocessing unit for executing rendering resource pre-distribution, an attention tracking unit for obtaining a sight focus, and a motion prediction unit for collecting a head pose. And the intelligent adaptation unit realizes content scheduling. Teaching content is organized by constructing the medical knowledge graph, resource optimization allocation is performed in combination with user sight focus and head motion prediction, and intelligent presentation of a teaching scene and dynamic content scheduling are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality medical education, and in particular to a wearable VR system for health education. Background Art

[0002] The application of virtual reality technology in the field of medical health education has brought innovative breakthroughs to the traditional teaching model. The current VR medical education system mainly adopts static resource preloading and fixed rendering strategies, and displays medical knowledge content through three-dimensional modeling and scene arrangement. Such systems generally use knowledge organization methods based on keyword matching to construct teaching content, use simple rendering optimization strategies based on frustum culling, and obtain user attention information through traditional eye tracking algorithms. In terms of head movement prediction, it mainly relies on classic algorithms such as linear Kalman filtering for state estimation. The scheduling of teaching content usually adopts a preset branch selection structure, which lacks the ability to adapt to user learning behavior in real time.

[0003] However, these technical solutions have obvious shortcomings in practical applications. First, the knowledge organization method based on keyword matching is difficult to accurately express the complex logical relationship between clinical diagnosis and treatment knowledge, which affects the coherence and professionalism of teaching content. Secondly, simple rendering optimization strategies cannot effectively cope with the frequent changes in perspective and content switching in VR teaching scenes, resulting in inefficient allocation of rendering resources. Thirdly, traditional eye tracking algorithms are easily affected by pupil imaging distortion and gaze edge area positioning deviation in VR head-mounted display environments, which reduces the accuracy of user attention analysis. In addition, classic motion prediction algorithms have difficulty handling the nonlinear characteristics of head motion and sensor noise, which affects the prediction accuracy of rendering optimization. Finally, the preset content scheduling structure lacks real-time perception and intelligent adaptation capabilities of the user's learning status, and cannot provide a personalized learning experience. Summary of the invention

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

[0005] Therefore, the present invention provides a wearable VR system for health education, which can solve the problems mentioned in the background technology.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a wearable VR system for health education, comprising: a knowledge modeling unit, used to construct a medical knowledge graph; the knowledge modeling unit is connected to a resource preprocessing unit and an intelligent adaptation unit, used to transmit the medical knowledge graph to the resource preprocessing unit, and respond to the teaching content retrieval request of the intelligent adaptation unit; the resource preprocessing unit is used to pre-allocate rendering resources for teaching scenes according to the hierarchical association between pathological knowledge nodes and clinical diagnosis and treatment relationship edges in the medical knowledge graph; the resource preprocessing unit is connected to the intelligent adaptation unit, used to receive rendering strategy optimization instructions from the intelligent adaptation unit; an attention tracking unit is used to obtain Take the user's line of sight and generate an attention heat map; the attention tracking unit is connected to the intelligent adaptation unit to transmit the attention heat map to the intelligent adaptation unit; a motion prediction unit is used to collect the user's head posture data and generate a head motion prediction trajectory; the motion prediction unit is connected to the intelligent adaptation unit to transmit the head motion prediction trajectory to the intelligent adaptation unit; the intelligent adaptation unit is used to locate the pathological knowledge node of current concern in the medical knowledge graph according to the attention heat map, dynamically optimize the pre-allocation of rendering resources in the teaching scene based on the head motion prediction trajectory, and select the next stage of teaching content according to the hierarchical association of the pathological knowledge nodes.

[0007] As a preferred solution of the wearable VR system for health education described in the present invention, the knowledge modeling unit includes a knowledge extraction module, a relationship construction module and a graph storage module; the knowledge extraction module is used to extract pathological knowledge nodes from medical textbooks; the relationship construction module is used to establish the clinical diagnosis and treatment relationship between pathological knowledge nodes, specifically: using a multi-layer classification method to divide the node hierarchy of pathological knowledge nodes; using a text semantic analysis method to extract contextual association features between pathological knowledge nodes; calculating the clinical correlation score between pathological knowledge nodes based on a deep neural network; the deep neural network adopts a bidirectional long short-term memory network structure, specifically inputting the association feature vectors of two nodes into the deep neural network, and analyzing the temporal dependency The clinical logical relationship between features is extracted, and the network output is normalized to a clinical relevance score in the range of 0 to 1 using a sigmoid activation function; a segmented mapping method is used to assign weight coefficients to connections between pathological knowledge nodes; the segmented mapping method is based on the clinical relevance score, specifically, when the clinical relevance score is less than a first preset threshold, a weight coefficient of A is assigned, when the clinical relevance score is between the first preset threshold and the second preset threshold, the weight coefficient is proportional to the score, and when the clinical relevance score is greater than the second preset threshold, a weight coefficient of B is assigned; a multi-level knowledge structure is constructed using a graph structure quantization method; the graph storage module is connected to the relationship construction module and stores the constructed medical knowledge graph.

[0008] As a preferred solution of the wearable VR system for health education described in the present invention, wherein: the resource preprocessing unit includes a hierarchical analysis module, a resource allocation module and a cache management module; the hierarchical analysis module performs quantitative calculations on the hierarchical associations between the pathological knowledge nodes, specifically: calculating the out-degree and in-degree of the pathological knowledge nodes in the medical knowledge graph; the out-degree represents the number of clinical diagnosis and treatment relationship edges pointing from the pathological knowledge node to other pathological knowledge nodes, and the in-degree represents the number of clinical diagnosis and treatment relationship edges pointing to the pathological knowledge node; constructing the association propagation matrix of the pathological knowledge node, including the following steps: establishing an N×N square matrix; wherein N is the medical knowledge The total number of pathological knowledge nodes in the graph, the rows and columns of the square matrix correspond to different pathological knowledge nodes respectively; traverse each of the clinical diagnosis and treatment relationship edges in the medical knowledge graph, take the starting node of the clinical diagnosis and treatment relationship edge as the row index of the square matrix, and the ending node as the column index, fill in the weight coefficient in the corresponding position, and construct an initial association propagation matrix; wherein, if there is a direct connection between the pathological knowledge nodes, the value of the corresponding position of the association propagation matrix is ​​the weight coefficient determined by the relationship construction module through the segmented mapping method, and if there is no direct connection, it is zero; normalize each row of the initial association propagation matrix so that the sum of the elements in each row is 1, and obtain the normalized association propagation matrix M ; Generate the hierarchical importance score of the pathological knowledge node through iterative calculation. After the iteration converges, the corresponding value in the importance vector obtained during the iteration converges as the hierarchical importance score, and make the following adjustments based on the difference between the out-degree and the in-degree of each pathological knowledge node: if the difference between the out-degree and the in-degree of the pathological knowledge node is greater than half of the total number of connections of the pathological knowledge node, adjust the hierarchical importance score to the square of the current value; otherwise, keep the hierarchical importance score unchanged; the resource allocation module determines the priority of the rendering resource pre-allocation based on the edge computing device according to the hierarchical association, including the following steps: calculate the resource demand of the pathological knowledge node, and Determine the resource allocation priority based on the hierarchical importance score; wherein, if the hierarchical importance score of the pathological knowledge node exceeds the average value of the level where the pathological knowledge node is located, then increase the resource allocation priority of the pathological knowledge node; evaluate the computing power of the edge computing device, and obtain the available resources of the edge computing device; formulate the rendering resource pre-allocation strategy based on the resource allocation priority and the available resources; wherein, when the resource demand exceeds the available resources of the edge computing device, split the rendering task with lower priority into multiple subtasks according to the task dependency; the cache management module is connected to the resource allocation module and manages the storage space of the edge computing device.

[0009] As a preferred solution of the wearable VR system for health education described in the present invention, the attention tracking unit includes an image acquisition module, a pupil positioning module and a heat map generation module; the image acquisition module collects a user's eye image sequence at a preset frame rate through a binocular camera, and performs denoising and illumination compensation processing on the user's eye image sequence to obtain a sight line image; the pupil positioning module performs pupil detection and corneal reflection point positioning on the sight line image, and calculates the screen coordinates of the sight line focus based on the relative position of the pupil center and the corneal reflection point; wherein, if the pupil contour obtained by the pupil detection meets the ellipticity requirement and the brightness value of the corneal reflection point is greater than the average brightness value of the sight line image, the position relationship between the pupil center and the corneal reflection point is used to calculate the screen coordinates of the sight line focus; otherwise, the current sight line image is marked as an invalid frame and returned for re-acquisition; wherein, the judgment process of whether the pupil contour meets the ellipticity requirement includes: extracting the contour point set of the pupil contour, calculating the minimum circumscribed rectangle of the contour point set, constructing a standard ellipse based on the major axis and minor axis of the minimum circumscribed rectangle, and calculating the average distance from the point in the contour point set to the standard ellipse. Distance deviation, the ratio of the average distance deviation to the circumference of the standard ellipse is defined as the ellipticity fitting error, and it is determined whether the ellipticity fitting error is less than the ratio of the area of ​​the pupil contour to the circumference; the process of calculating the screen coordinates of the sight focus using the positional relationship between the pupil center and the corneal reflection point includes: establishing a polar coordinate system with the pupil center as the origin, calculating the polar coordinate parameters of the corneal reflection point relative to the pupil center, constructing a sight vector according to the polar coordinate parameters, and taking the intersection of the sight vector and the display plane as the screen coordinates of the sight focus; the heat map generation module performs spatiotemporal clustering analysis on the sight focus, and generates an attention heat map according to the residence time and spatial distribution characteristics of the sight focus; wherein, if the hierarchical importance score of the pathological knowledge node corresponding to the sight focus is higher than the average value of the nodes at the same level, and the residence time of the sight focus in a certain area exceeds the average residence time of the adjacent areas, then in the attention heat map, the heat value of a certain area is set to the product of the sight focus residence time and the hierarchical importance score; otherwise, the heat value of the area is set to the sight focus residence time value.

[0010] As a preferred solution of the wearable VR system for health education described in the present invention, the motion prediction unit includes a data acquisition module, a state estimation module and a trajectory prediction module; the data acquisition module obtains head posture sensor data, performs low-pass filtering preprocessing on the head posture sensor data, and constructs a time series data sequence according to the sampling frequency of the sensor; the head posture sensor data includes three-axis acceleration, three-axis angular velocity and three-axis magnetic field intensity; the state estimation module estimates the current motion state based on the recursive Bayesian prediction method, specifically constructing a 16-dimensional state vector containing position, velocity, acceleration and posture quaternion, establishing a nonlinear state transfer equation considering sensor noise, and using particle physics to predict the current motion state. The sub-filter algorithm is used for state estimation, and the abnormal state is corrected by Kalman smoothing; the abnormal state judgment includes state mutation judgment, physical constraint violation judgment and sensor abnormality judgment; the state mutation judgment is realized by calculating the change of the state estimation value at adjacent moments and comparing it with the respective relevant thresholds, including position mutation, speed mutation, acceleration mutation and posture angle mutation; the physical constraint violation judgment includes speed over-limit, acceleration over-limit and angular velocity over-limit; the sensor abnormality is judged by observing the residual; when the abnormal state is detected, the fixed interval Kalman smoother is used to correct the abnormal state; the trajectory prediction module is connected with the state estimation module and generates the head movement prediction trajectory.

[0011] As a preferred solution of the wearable VR system for health education described in the present invention, the execution process of the particle filter algorithm includes: sampling N particles from the state prior distribution, calculating the importance weight of each particle based on the observation equation, normalizing the importance weight, calculating the effective particle number, resampling when the effective particle number is less than the effective particle number threshold, and finally obtaining the posterior distribution estimate of the state vector.

[0012] As a preferred solution of the wearable VR system for health education described in the present invention, wherein: the intelligent adaptation unit includes a node positioning module, an optimization strategy module and a content scheduling module; the node positioning module determines the currently concerned pathological knowledge node in the medical knowledge graph; the optimization strategy module generates a dynamic optimization plan for pre-allocation of rendering resources based on a deep reinforcement learning model; the content scheduling module executes the dynamic optimization plan and updates the teaching content of the next stage according to the hierarchical association of the pathological knowledge nodes.

[0013] To further solve the above technical problems, the present invention provides the following technical solutions: a method for a wearable VR system for health education, comprising: constructing a medical knowledge graph including pathological knowledge nodes and clinical diagnosis and treatment relationship edges; the medical knowledge graph constructs the relationship between nodes through concept mapping; based on the edge computing device, rendering resources for the teaching scene are pre-allocated according to the hierarchical association between the pathological knowledge nodes and the clinical diagnosis and treatment relationship edges; the user's line of sight is obtained through a binocular camera and an attention heat map is generated; the attention heat map represents the distribution of the user's attention to different areas in the teaching scene; the user's head posture data is collected based on a head posture sensor and a head movement prediction trajectory is generated through recursive Bayesian prediction; the head movement prediction trajectory is used to estimate the user's field of view at the next moment; the pathological knowledge node currently concerned is located in the medical knowledge graph according to the attention heat map, the rendering resource pre-allocation in the teaching scene is dynamically optimized based on the head movement prediction trajectory, and the teaching content of the next stage is selected according to the hierarchical association of the pathological knowledge nodes; the dynamic optimization adopts a deep reinforcement learning model, and the attention heat map and the head movement prediction trajectory are used as state inputs.

[0014] A computer device includes a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the wearable VR system for health education as described above when executing the computer program.

[0015] 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 wearable VR system for health education as described above are implemented.

[0016] Beneficial effects of the present invention: The present invention adopts a multi-head self-attention mechanism and a bidirectional long short-term memory network to construct a medical knowledge graph through a knowledge modeling unit, solving the problem that it is difficult to accurately grasp the logical relationship of clinical diagnosis and treatment knowledge based on traditional word frequency statistics; the resource preprocessing unit pre-allocates rendering resources based on graph structure features and hierarchical importance scores, overcoming the limitation that traditional resource scheduling methods are difficult to adapt to the dynamic switching requirements of VR teaching scenes; the attention tracking unit adopts pupil contour verification based on standard ellipse fitting error and line of sight vector mapping in polar coordinate system, effectively solving the problems of pupil imaging distortion and edge area gaze coordinate deviation in VR head display environment; the motion prediction unit combines recursive Bayesian framework and particle filter algorithm to predict head motion, improving the accuracy of nonlinear motion state estimation and abnormal state correction; the intelligent adaptation unit coordinately optimizes rendering resource allocation and teaching content scheduling through a deep reinforcement learning model, realizing scene adaptive presentation based on user attention and head motion. The system organically combines the structured expression of medical knowledge with the intelligent optimization of VR interactive experience, providing a new teaching solution for health education. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0018] Figure 1 This is a schematic diagram of the overall structure of a wearable VR system for health education proposed by the present invention; Figure 2 This is an overall flow chart of the method of using a wearable VR system for health education proposed by the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and 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.

[0021] Example 1, reference Figure 1 , as an embodiment of the present invention, provides a wearable VR system for health education.

[0022] Figure 1 The overall structure diagram of a wearable VR system for health education is shown, including the following contents: The knowledge modeling unit is used to construct a medical knowledge graph. The knowledge modeling unit is connected to the resource preprocessing unit and the intelligent adaptation unit, and is used to transmit the medical knowledge graph to the resource preprocessing unit and respond to the teaching content retrieval request of the intelligent adaptation unit.

[0023] The resource preprocessing unit is used to pre-allocate rendering resources for teaching scenarios based on the hierarchical association between pathological knowledge nodes and clinical diagnosis and treatment relationship edges in the medical knowledge graph based on edge computing devices. The resource preprocessing unit is connected to the intelligent adaptation unit to receive rendering strategy optimization instructions from the intelligent adaptation unit.

[0024] The attention tracking unit is used to obtain the user's visual focus through the binocular camera and generate an attention heat map. The attention tracking unit is connected to the intelligent adaptation unit to transmit the attention heat map to the intelligent adaptation unit.

[0025] The motion prediction unit is used to collect the user's head posture data based on the head posture sensor and generate the head motion prediction trajectory through recursive Bayesian prediction. The motion prediction unit is connected to the intelligent adaptation unit and is used to transmit the head motion prediction trajectory to the intelligent adaptation unit.

[0026] The intelligent adaptation unit is used to locate the currently focused pathological knowledge nodes in the medical knowledge graph according to the attention heat map, dynamically optimize the pre-allocation of rendering resources in the teaching scene based on the predicted trajectory of head movement, and select the next stage of teaching content according to the hierarchical association of pathological knowledge nodes.

[0027] Specifically, Figure 2 As shown, it is a module structure diagram of each unit in a wearable VR system for health education provided by the present invention.

[0028] The knowledge modeling unit includes a knowledge extraction module, a relationship construction module, and a graph storage module. The knowledge extraction module is used to extract pathological knowledge nodes from medical textbooks, the relationship construction module is used to establish clinical diagnosis and treatment relationships between pathological knowledge nodes, and the graph storage module is connected to the relationship construction module and stores the constructed medical knowledge graph.

[0029] The process of extracting pathological knowledge nodes from medical textbooks by the knowledge extraction module includes the following steps: Step 1: Scan the medical textbook text using a rule-based medical terminology recognition method; Step 2: Use domain dictionary matching method to perform standardized mapping of medical terms; Step three, use attribute labeling method to add labels to pathological knowledge nodes.

[0030] Among them, the medical terminology recognition method is based on the expert annotation rule library, which is used to identify four types of key medical terms: disease name, symptom description, diagnosis method, and treatment plan. The domain dictionary matching method is based on character sequence matching, which specifically includes: establishing a medical terminology standard dictionary containing standard terms and their synonyms and synonym comparison tables, matching the identified medical terms with the medical terminology standard dictionary, and generating a unique node identification code for the matched standard terms; attribute labels include node type, node description, and node source.

[0031] The process of establishing clinical diagnosis and treatment relationships between pathology knowledge nodes in the relationship building module includes the following steps: Step 1: Use a multi-layer classification method to divide the node levels of pathological knowledge nodes; Step 2: Use text semantic analysis method to extract contextual association features between pathological knowledge nodes; Step 3: Calculate the clinical correlation scores between pathological knowledge nodes based on deep neural network; Step 4: Use segmented mapping method to assign weight coefficients to the connections between pathological knowledge nodes; Step 5: Use graph structure quantification methods to build a multi-level knowledge structure.

[0032] Among them, the multi-layer classification method is based on node attribute labels, dividing the pathological knowledge nodes into four basic levels: disease layer, symptom layer, diagnosis layer, and treatment layer; the text semantic analysis method is based on the attention mechanism, specifically including: extracting a context window containing 10 sentences before and after the target node, using the multi-head self-attention mechanism to calculate the semantic relevance of each sentence in the window with the target node, and extracting the sentence with the highest attention weight as the correlation feature between the nodes. The deep neural network adopts a bidirectional long short-term memory network structure, specifically including: inputting the correlation feature vector of the two nodes into the deep neural network, extracting the clinical logical relationship between the features through temporal dependency analysis, and using the sigmoid activation function to normalize the network output to a clinical relevance score in the range of 0 to 1. The segmented mapping method is based on the clinical relevance score, specifically including: assigning a weight coefficient to A when the clinical relevance score is less than the first preset threshold, when the clinical relevance score is between the first preset threshold and the second preset threshold, the weight coefficient is proportional to the score, and when the clinical relevance score is greater than the second preset threshold, the weight coefficient of B is assigned. The graph structure quantization method is based on weighted path analysis, specifically including: constructing an adjacency matrix according to the weight coefficients between nodes, calculating the shortest weighted path between nodes based on the adjacency matrix, and determining the hierarchical position relationship of the nodes according to the weighted path, which is used to guide the resource preprocessing unit to perform priority allocation of rendering resources. In this embodiment, the first preset threshold is set to 0.3, A is 0.1; the second preset threshold is set to 0.7, and B is 1.0.

[0033] Preferably, compared with the prior art, the relationship construction module of the present invention improves the construction quality of the medical knowledge graph by integrating the multi-head self-attention mechanism and the bidirectional long short-term memory network. Traditional methods usually rely on simple word frequency statistics or fixed window co-occurrence analysis to establish node associations, which makes it difficult to accurately grasp the logical relationship of clinical diagnosis and treatment knowledge; the attention mechanism introduced by the present invention can accurately capture contextual semantic features, and the bidirectional LSTM structure can simultaneously consider forward and backward clinical logic dependencies, so that the establishment of node associations is more in line with medical professional rules. Especially in the weight allocation link, the three-stage mapping strategy based on dynamic thresholds designed by the present invention can quantify the node association strength more finely than the linear mapping or fixed threshold classification method commonly used in the prior art. Experiments show that this multi-level relationship construction method improves the relationship accuracy of the knowledge graph by about 30%, and the automatic association accuracy of the newly added nodes reaches more than 85%, providing a more reliable decision-making basis for the subsequent dynamic scheduling of VR teaching resources.

[0034] The process of storing the constructed medical knowledge graph by the graph storage module includes the following steps: Step 1: Use graph database storage method to store multi-level knowledge structure; Step 2: Use dynamic cache warm-up method to manage knowledge node access; Among them, the graph database storage method is based on the index establishment mechanism, including: establishing node index and relationship index, and building a query interface based on node attributes, relationship type, and weight coefficient. The dynamic cache warm-up method is based on access frequency statistics, including: counting the number of node visits in the last hour, loading the node and its associated nodes into memory when the number of node visits exceeds 10 times per hour, and updating the cache content every 10 minutes to improve the data response speed when switching teaching scenes.

[0035] It should be noted that after the medical knowledge graph is constructed, the knowledge modeling unit needs to transmit the medical knowledge graph to the resource processing unit through the data interaction interface. The specific process is as follows: first, the knowledge modeling unit serializes the constructed medical knowledge graph into a standard JSON format, which contains node data, relationship data and weight information; then, the serialized data is transmitted in batches to the data receiving module of the resource preprocessing unit through the message queue mechanism; wherein, the message queue adopts the publish-subscribe mode to ensure the reliability of data transmission; then, the data receiving module of the resource preprocessing unit deserializes the received JSON data and restores it to a graph structure; finally, based on the weight information in the graph structure, the resource preprocessing unit performs initial allocation of rendering resources and returns confirmation information to the knowledge modeling unit after the data transmission is completed. This message queue-based data transmission mechanism not only ensures the reliable transmission of large-scale knowledge graph data, but also supports incremental updates of data.

[0036] Preferably, the knowledge modeling unit of the present invention provides intelligent knowledge support for the wearable VR health education system by organically combining the structured expression of medical textbook knowledge with VR teaching resource management. Compared with traditional knowledge management methods, this unit adopts semantic analysis based on attention mechanism and multi-level weight quantization strategy, which not only improves the accuracy of medical knowledge graph construction, but more importantly, through deep collaboration with resource preprocessing unit and intelligent adaptation unit, it realizes accurate push of teaching content and dynamic optimization of rendering resources. In the scenario of large-scale dynamic update of medical knowledge, the knowledge modeling unit uses the data interaction mechanism of message queue and the cache preheating strategy of graph database, so that the system can efficiently process real-time data streams and ensure smooth switching of VR teaching scenes.

[0037] The resource preprocessing unit includes a hierarchical analysis module, a resource allocation module and a cache management module. The hierarchical analysis module performs quantitative calculations on the hierarchical associations between pathological knowledge nodes, the resource allocation module determines the priority of rendering resource pre-allocation based on the hierarchical associations of the edge computing device, and the cache management module is connected to the resource allocation module and manages the storage space of the edge computing device.

[0038] Specifically, the hierarchical analysis module performs quantitative calculation on the hierarchical associations between pathological knowledge nodes, specifically including the following steps: Step 1: Calculate the out-degree and in-degree of the pathology knowledge node in the medical knowledge graph; Step 2: construct the association propagation matrix of pathological knowledge nodes; Step 3: Generate hierarchical importance scores for pathological knowledge nodes.

[0039] Among them, the out-degree represents the number of clinical diagnosis and treatment relationship edges pointing from the pathology knowledge node to other pathology knowledge nodes, and the in-degree represents the number of clinical diagnosis and treatment relationship edges pointing to the pathology knowledge node.

[0040] In this embodiment, the association propagation matrix is ​​constructed in the following manner: first, an N×N square matrix is ​​established, where N is the total number of pathological knowledge nodes in the medical knowledge graph, and the rows and columns of the square matrix correspond to different pathological knowledge nodes; then, each clinical diagnosis and treatment relationship edge in the medical knowledge graph is traversed, and the starting node of the clinical diagnosis and treatment relationship edge is used as the row index of the square matrix, and the ending node is used as the column index, and the weight coefficient is filled in the corresponding position to construct the initial association propagation matrix; finally, each row of the initial association propagation matrix is ​​normalized so that the sum of the elements in each row is 1, and the normalized association propagation matrix M is obtained.

[0041] Among them, filling in the weight coefficient at the corresponding position includes: if there is a direct connection between the pathological knowledge nodes, the value of the corresponding position of the association propagation matrix is ​​the weight coefficient determined by the relationship construction module through the segmented mapping method, and if there is no direct connection, it is zero.

[0042] The hierarchical importance score is obtained by iterative calculation. The iterative calculation process of the hierarchical importance score is as follows: first, each pathological knowledge node is assigned the same initial importance value of 1.0 and recorded as a vector ; Then in each iteration k, the new importance vector is calculated by the following formula: ; Among them, M is the normalized correlation propagation matrix; β is the importance attenuation factor, which is 0.85; k is the iteration round; R k is the node importance vector after the kth iteration.

[0043] When the L1 norm of the importance vector difference of two consecutive iterations meets the following conditions: ; It is considered that the iteration converges, where is the convergence threshold, and its value is 0.001.

[0044] After the iteration converges, the corresponding value in the importance vector obtained during the iteration convergence is used as the hierarchical importance score, and the following adjustments are made based on the difference between the out-degree and in-degree of each pathological knowledge node: If the difference between the out-degree and in-degree of the pathological knowledge node is greater than half of the total number of connections of the pathological knowledge node, the hierarchical importance score is adjusted to the square of the current value; otherwise, the hierarchical importance score remains unchanged.

[0045] The process of determining the priority of rendering resource pre-allocation based on the edge computing device according to the hierarchical association by the resource allocation module includes the following steps: Step 1: Calculate the resource demand of the pathology knowledge node and determine the resource allocation priority based on the hierarchical importance score; if the hierarchical importance score of the pathology knowledge node exceeds the average value of the hierarchical level of the pathology knowledge node, then increase the resource allocation priority of the pathology knowledge node; Step 2: Evaluate the computing power of the edge computing device and obtain the available resources of the edge computing device; Step three: formulate a rendering resource pre-allocation strategy based on resource allocation priority and available resources; when the resource demand exceeds the available resources of the edge computing device, the rendering task with lower priority is split into multiple subtasks according to task dependencies.

[0046] Among them, the resource demand is calculated based on the hierarchical importance score and the rendering complexity of the pathological knowledge node, specifically: The hierarchical importance scores of pathological knowledge nodes are mapped to the [0,1] interval through minimum-maximum normalization, denoted as , represents the normalized value of node importance; the rendering complexity parameters of the pathological knowledge node (including the sum of the quotient obtained by dividing the number of 3D model faces by 10000 and the quotient obtained by dividing the texture resolution by 1024) are normalized and recorded as , represents the normalized value of rendering complexity; finally, the resource demand D is obtained by weighted summation: ; in, is the importance weight coefficient, and its value is 0.6.

[0047] The task dependency in the rendering resource pre-allocation strategy is achieved by constructing a directed acyclic graph: taking the pathology knowledge nodes as vertices in the graph, taking the direction of the clinical diagnosis and treatment relationship edge as the dependency direction of task execution, and determining the execution order of rendering tasks through topological sorting; when it is necessary to split the rendering tasks, priority is given to ensuring the rendering integrity of the pathology knowledge nodes on the critical path, and the pathology knowledge nodes on the non-critical path can reduce the rendering accuracy or delay loading. This sophisticated resource scheduling mechanism ensures that the system can always maintain the optimal rendering effect and response speed when the resources of the edge computing device are limited.

[0048] The process of managing the storage space of the edge computing device by the cache management module includes the following steps: Step 1: Divide the storage space level of edge computing devices; Step 2: Execute cache replacement operation; Step three: Maintain cache consistency.

[0049] Among them, the storage space is divided into a cache area, a main memory area and an external memory area; cache replacement is based on the least recently used principle. If the usage rate of the cache area exceeds the third preset threshold and the hierarchical importance score of the pathological knowledge node to be cached is higher than the average value of the cached nodes, the cache replacement operation is triggered; consistency maintenance adopts a write-back strategy to mark the updated pathological knowledge nodes.

[0050] It should be noted that, in this embodiment, the cache replacement operation includes the following processing procedures: first, the usage frequency score of each cached node in the cache area is calculated, and the calculation of the usage frequency score is comprehensively evaluated through the following multiple dimensions: (1) Time dimension: considering the time interval between the last access time of the node and the current time, as well as the access pattern and frequency distribution of the node in the most recent time window; (2) Access dimension: including statistical features such as the historical cumulative number of accesses to the node, the length of the continuous access sequence, and the variance of the access interval; (3) Relevance dimension: evaluating the degree of association between the node and the current teaching scene, including features such as topological distance and semantic similarity in the knowledge graph; (4) Resource dimension: considering resource occupancy features such as the data size and rendering complexity of the node. The above multi-dimensional features are used to obtain the final usage frequency score through a machine learning model or a weighted calculation method; then, the cached nodes are sorted according to the usage frequency score, and the node with the lowest score is selected for elimination, and its data is migrated to the main memory area; finally, the data of the pathological knowledge node to be cached is loaded from the main memory area to the available space in the cache area.

[0051] In response to the cache management needs in VR medical education scenarios, the third preset threshold is a key parameter determined by analyzing the multi-level storage structure of edge computing devices and the switching characteristics of medical teaching content. Since the knowledge nodes in the medical teaching scenario show obvious local access characteristics, that is, a group of associated nodes will be repeatedly accessed within a specific time period, the setting of the third preset threshold needs to strike a balance between cache utilization and replacement overhead. Statistical analysis based on large-scale teaching data shows that when the cache capacity is set to 25% of the total memory of the edge computing device, setting the third preset threshold to 75% can effectively reduce the probability of cache thrashing while maintaining a high cache hit rate. In addition, considering the real-time requirements of scene switching in VR teaching, the setting of the third preset threshold also reserves sufficient cache space to cope with sudden high-frequency access requests, so that the system has sufficient dynamic adaptability while maintaining stable performance.

[0052] Preferably, the present invention realizes the pre-allocation of rendering resources for teaching scenarios through a resource allocation mechanism based on graph structure features combined with edge computing devices. By analyzing the hierarchical association characteristics of pathological knowledge nodes, the pre-allocation of rendering resources is optimized, thereby improving the rendering efficiency and response speed of the system.

[0053] The attention tracking unit includes an image acquisition module, a pupil positioning module and a heat map generation module.

[0054] Specifically, the image acquisition module acquires a sequence of user eye images at a preset frame rate through a binocular camera, and performs denoising and illumination compensation processing on the sequence of user eye images to obtain a sight line image.

[0055] The pupil positioning module performs pupil detection and corneal reflection point positioning on the sight line image, and calculates the screen coordinates of the sight line focus based on the relative position of the pupil center and the corneal reflection point. If the pupil contour obtained by the pupil detection meets the ellipticity requirements and the brightness value of the corneal reflection point is greater than the average brightness value of the sight line image, the positional relationship between the pupil center and the corneal reflection point is used to calculate the screen coordinates of the sight line focus. Otherwise, the current sight line image is marked as an invalid frame and returned for re-capture.

[0056] The process of determining whether the pupil contour meets the ellipticity requirement includes: extracting the contour point set of the pupil contour, calculating the minimum circumscribed rectangle of the contour point set, constructing a standard ellipse based on the major axis and the minor axis of the minimum circumscribed rectangle, calculating the average distance deviation from the points in the contour point set to the standard ellipse, defining the ratio of the average distance deviation to the perimeter of the standard ellipse as the ellipticity fitting error, and determining whether the ellipticity fitting error is less than the ratio of the area to the perimeter of the pupil contour; The process of calculating the screen coordinates of the line of sight focus using the positional relationship between the pupil center and the corneal reflection point includes: establishing a polar coordinate system with the pupil center as the origin, calculating the polar coordinate parameters of the corneal reflection point relative to the pupil center, constructing a line of sight vector according to the polar coordinate parameters, and taking the intersection of the line of sight vector and the display plane as the screen coordinates of the line of sight focus, wherein the direction angle of the line of sight vector is a function of the azimuth angle of the polar coordinate parameters, and the elevation angle of the line of sight vector is a function of the radial distance of the polar coordinate parameters.

[0057] It should be noted that based on the use scenario of the wearable VR system for health education of the present invention, the pupil positioning module effectively solves the problem of pupil imaging distortion caused by learners frequently turning their heads to watch virtual teaching content during VR health education by introducing a pupil contour verification method based on standard ellipse fitting error. This method uses the area-to-perimeter ratio of the pupil contour as an adaptive reference standard, overcoming the technical defect that the traditional fixed threshold method is difficult to adapt to the pupil characteristics of different learners when wearing a VR head display, and improves the accuracy of pupil detection during the interaction of health education content.

[0058] Preferably, by using the sight vector mapping method in the polar coordinate system to calculate the screen coordinates of the sight focus, the pupil positioning module of the present invention solves the coordinate deviation problem caused by the traditional linear mapping method when looking at the edge area of ​​the virtual scene in the VR display environment. This method uses the polar coordinate relationship between the pupil center and the corneal reflection point to construct the sight vector. Compared with the plane projection method in the prior art, it can more accurately track the actual gaze position of the learner when watching health education content, and provide reliable sight data support for the intelligent adaptation of subsequent teaching content.

[0059] Furthermore, the heat map generation module performs spatiotemporal clustering analysis on the gaze focus and generates an attention heat map based on the dwell time and spatial distribution characteristics of the gaze focus.

[0060] If the hierarchical importance score of the pathological knowledge node corresponding to the gaze focus is higher than the average value of the nodes at the same level, and the gaze focus stays in this area for longer than the average stay in adjacent areas, then the heat value of the area in the attention heat map is set to the product of the gaze focus stay time and the hierarchical importance score; otherwise, the heat value of the area is set to the gaze focus stay time value.

[0061] Specifically, the spatiotemporal clustering analysis process of sight focus includes: constructing a time series of sight focus, calculating the Euclidean distance and time interval of adjacent sight focus, and assigning sight focus with a Euclidean distance less than a spatial threshold and a time interval less than a time threshold to the same cluster. The spatial threshold is the minimum size of the display area boundary of the pathological knowledge node, and the time threshold is a multiple of the sampling period of the sight focus time series.

[0062] The process of generating an attention heat map based on the dwell time and spatial distribution characteristics of the gaze focus includes: taking the center point of the spatiotemporal cluster as the center of the heat distribution, mapping the number of gaze focuses in each cluster to the variance of the Gaussian distribution, mapping the sum of the time intervals of the gaze focuses within the cluster to the peak of the Gaussian distribution, and superimposing the Gaussian distributions of all clusters to obtain the attention heat map, where the variance of the Gaussian distribution is proportional to the cluster range, and the peak is proportional to the dwell time.

[0063] Preferably, the attention tracking unit of the present invention combines spatiotemporal clustering analysis with attention heat map by real-time analysis of learners' sight behavior characteristics in VR health education scenes, effectively solving the technical problem that traditional VR teaching systems are difficult to accurately capture the distribution of learners' attention to health knowledge content. The unit is based on the temporal clustering method of sight focus, which can adaptively adjust the clustering parameters, so that the system can more accurately identify the learners' stay mode of key knowledge points in the virtual teaching environment, and provide a quantitative basis for the dynamic optimization of teaching content. In addition, the attention tracking unit uses a heat map generation method based on Gaussian distribution superposition to visualize the learners' attention distribution in VR health education content, overcoming the attention assessment bias caused by simply accumulating the stay time in the prior art. By using the heat calculation method that combines the sight stay time with the knowledge node importance score, the system can more objectively reflect the learners' cognitive investment in health education content, and provide a decision-making basis that is more in line with teaching practice for the personalized push and difficulty adjustment of subsequent teaching content.

[0064] The motion prediction unit includes a data acquisition module, a state estimation module and a trajectory prediction module. The data acquisition module acquires the head posture sensor data, the state estimation module estimates the current motion state based on the recursive Bayesian prediction method, and the trajectory prediction module is connected to the state estimation module and generates a head motion prediction trajectory.

[0065] Specifically, the head posture sensor data includes three-axis acceleration, three-axis angular velocity and three-axis magnetic field strength; the data acquisition module performs low-pass filtering preprocessing on the head posture sensor data and constructs a time series data sequence according to the sampling frequency of the sensor.

[0066] The recursive Bayesian prediction method includes: constructing a ,speed , acceleration and attitude quaternion The 16-dimensional state vector is used to establish the nonlinear state transfer equation considering the sensor noise. , the particle filter algorithm is used for state estimation, and the abnormal state is corrected by Kalman smoothing; in, is the three-dimensional position coordinate of the head in the world coordinate system, in meters; are the velocity components of the head in three directions, in meters per second; are the acceleration components of the head in three directions, in meters per second squared; is the unit quaternion describing the head posture, where is the scalar part, is the vector part; is the system process noise, which obeys a Gaussian distribution with a mean of 0 and a covariance matrix of Q.

[0067] Specifically, the abnormal state determination includes state mutation determination, physical constraint violation determination and sensor abnormality determination; the state mutation determination is realized by calculating the change of the state estimation value at adjacent moments and comparing it with the respective relevant thresholds, including position mutation , speed mutation , acceleration mutation and attitude angle mutation ; Physical restraint violation determination including speed limit , Acceleration exceeds limit and angular velocity exceeds the limit ; Sensor anomalies are detected by observing residuals Make a judgment; when an abnormal state is detected, a fixed interval Kalman smoother is used to correct the abnormal state.

[0068] in, is the position mutation threshold, set to 0.1m; is the velocity mutation threshold, set to 2.0 m / s; is the acceleration mutation threshold, set to 20.0m / s²; is the attitude angle mutation threshold, set to 45°; is the maximum permissible speed, set to 3.0m / s; is the maximum permissible acceleration, set to 30.0m / s²; is the maximum permissible angular velocity, set to 720° / s; is the sensor observation residual threshold, which is set according to the sensor characteristics.

[0069] Among them, the nonlinear state transfer equation expresses the temporal change relationship of position, velocity, acceleration and attitude quaternion as follows: Position update , speed update , acceleration update , posture update ; is the system sampling time interval, set to 0.005 seconds (200Hz sampling rate); is a random disturbance of acceleration, which obeys a Gaussian distribution with a mean of 0 and a standard deviation of 0.1 m / s²; is the angular velocity vector at the current moment, which is directly measured by the gyroscope; is the unit vector of the rotation axis, obtained by normalizing the angular velocity vector; is the quaternion multiplication operator.

[0070] The execution process of the particle filter algorithm includes: sampling N particles from the state prior distribution, calculating the importance weight of each particle based on the observation equation, normalizing the weight, calculating the number of effective particles, resampling when the number of effective particles is less than the effective particle number threshold, and finally obtaining the posterior distribution estimate of the state vector. The effective particle number threshold is set to 50% of the total number of particles.

[0071] The trajectory prediction module is connected to the state estimation module and is used to generate a head motion prediction trajectory based on the state estimation result; wherein the trajectory prediction process includes: establishing a trajectory prediction model based on a long short-term memory network, extracting the time-varying characteristics of the head motion, constructing a prediction strategy according to the periodicity and continuity constraints of the motion characteristics, and using a sliding time window method to gradually predict the head position and posture parameters at future moments; Among them, the periodicity and continuity constraints of motion features include: extracting the main frequency components of head motion as network input features through Fourier analysis, introducing a velocity smoothness loss function to ensure trajectory continuity, setting the constraint range of acceleration and angular velocity according to the physiological limits of human head motion, and dynamically smoothing the predicted trajectory.

[0072] Preferably, the motion prediction unit of the present invention realizes accurate prediction of the head motion of the VR head display user by combining the recursive Bayesian framework and the deep learning method. Compared with the traditional prediction method, the unit has obvious advantages in modeling the nonlinear characteristics of the motion state and processing sensor noise, and provides reliable motion prediction support for the rendering optimization of the teaching scene. Experiments show that the average position error of the prediction method is controlled within 5mm, and the posture error is controlled within 2°, which can meet the real-time interaction requirements of VR health education scenes.

[0073] The intelligent adaptation unit includes a node positioning module, an optimization strategy module, and a content scheduling module. The node positioning module determines the pathological knowledge node of current concern in the medical knowledge graph, the optimization strategy module generates a dynamic optimization plan for pre-allocation of rendering resources based on the deep reinforcement learning model, and the content scheduling module executes the dynamic optimization plan and updates the teaching content of the next stage according to the hierarchical association of the pathological knowledge nodes.

[0074] Specifically, the node positioning module is responsible for locating the pathological knowledge node of current concern in the medical knowledge graph, extracting the user's visual focus area by receiving the attention heat map output by the attention tracking unit, determining the focus area by using the set heat map threshold, and mapping it to the three-dimensional model surface in the scene to obtain the corresponding model part identification, and finally retrieving the relevant pathological knowledge node in the medical knowledge graph constructed by the knowledge modeling unit according to the model part identification. The optimization strategy module generates a rendering resource pre-allocation scheme based on the head motion prediction trajectory output by the motion prediction unit, specifically including: converting the predicted trajectory into a sequence of key path points containing position and posture information, calculating the view cone and visible area corresponding to each path point, and assigning rendering priority to the models and textures in the visible area based on the preset resource allocation strategy, where the main focus area (the current node related part) is given the highest priority and complete details, the secondary focus area (the adjacent node related part) is given a medium priority and reduced details, and the background area is given the lowest priority and the lowest details. The content scheduling module is responsible for executing the resource pre-allocation scheme and updating the teaching content, including adjusting the model detail level LOD and texture resolution according to the priority output by the optimization strategy module, preloading the resources that may be used in the next stage (including the detailed content of the current node, the content of the adjacent nodes directly connected to the current node, and the basic background resources), and selecting the candidate teaching content for the next stage according to the outgoing edge relationship of the nodes in the knowledge graph. The key parameter settings in this embodiment include: the heat map threshold is 0.7 (normalized heat value), the cone parameter FOV=90 ​​degrees, the near plane=0.1 meters, the far plane=100 meters, the LOD level is 3 levels (high / medium / low precision), the texture resolution is original / half / quarter, and the preload buffer size is 100MB. The intelligent adaptation unit of this embodiment realizes the intelligent pre-allocation of teaching scene rendering resources by combining the user's attention distribution and head movement prediction. The unit can accurately track the user's learning focus, load the content that is about to enter the field of view in advance, and dynamically adjust the rendering details according to the degree of attention, thereby improving the rendering efficiency while ensuring the visual quality. In addition, the content organization method based on knowledge graph makes the connection between teaching content more natural and smooth.

[0075] In summary, the present invention uses a multi-head self-attention mechanism and a bidirectional long short-term memory network to construct a medical knowledge graph through a knowledge modeling unit, solving the problem that it is difficult to accurately grasp the logical relationship of clinical diagnosis and treatment knowledge based on traditional word frequency statistics; the resource preprocessing unit pre-allocates rendering resources based on graph structure features and hierarchical importance scores, overcoming the limitation that traditional resource scheduling methods are difficult to adapt to the dynamic switching requirements of VR teaching scenes; the attention tracking unit uses pupil contour verification based on standard ellipse fitting error and line of sight vector mapping in polar coordinates to effectively solve the problems of pupil imaging distortion and edge area gaze coordinate deviation in VR head-mounted display environment; the motion prediction unit combines the recursive Bayesian framework and particle filter algorithm to predict head motion, improving the accuracy of nonlinear motion state estimation and abnormal state correction; the intelligent adaptation unit coordinately optimizes rendering resource allocation and teaching content scheduling through a deep reinforcement learning model, realizing scene adaptive presentation based on user attention and head motion. The system organically combines the structured expression of medical knowledge with the intelligent optimization of VR interactive experience, providing a new teaching solution for health education.

[0076] Example 2, reference Figure 2 , which is an embodiment of the present invention, based on the first embodiment, this embodiment further provides a method of using a wearable VR system for health education.

[0077] Figure 2 The overall process diagram of the method includes: S1: Construct a medical knowledge graph containing pathological knowledge nodes and clinical diagnosis and treatment relationship edges; the medical knowledge graph constructs the relationship between nodes through concept mapping; S2: Pre-allocate rendering resources for teaching scenarios based on edge computing devices according to the hierarchical association between pathological knowledge nodes and clinical diagnosis and treatment relationship edges; S3: Obtain the user's sight focus through the binocular camera and generate an attention heat map; the attention heat map represents the distribution of the user's attention to different areas in the teaching scene; S4: Collecting user head posture data based on the head posture sensor and generating head motion prediction trajectory through recursive Bayesian prediction; the head motion prediction trajectory is used to estimate the user's field of view at the next moment; S5: Locate the currently focused pathological knowledge nodes in the medical knowledge graph based on the attention heat map, dynamically optimize the pre-allocation of rendering resources in the teaching scene based on the head motion prediction trajectory, and select the next stage of teaching content according to the hierarchical association of the pathological knowledge nodes; the dynamic optimization uses a deep reinforcement learning model, taking the attention heat map and head motion prediction trajectory as state input.

[0078] Embodiment 3 is an embodiment of the present invention, which is different from the previous embodiment in that: if the function is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0079] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0080] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0081] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. 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 wearable VR system for health education, characterized in that: include: Knowledge modeling unit, used to build medical knowledge graph; The knowledge modeling unit is connected to the resource preprocessing unit and the intelligent adaptation unit, and is used to transmit the medical knowledge graph to the resource preprocessing unit and respond to the teaching content retrieval request of the intelligent adaptation unit; A resource preprocessing unit, used for pre-allocating rendering resources for teaching scenes according to the hierarchical association between pathological knowledge nodes and clinical diagnosis and treatment relationship edges in the medical knowledge graph; The resource preprocessing unit is connected to the intelligent adaptation unit and is used to receive a rendering strategy optimization instruction from the intelligent adaptation unit; Attention tracking unit, used to obtain the user's visual focus and generate an attention heat map; The attention tracking unit is connected to the intelligent adaptation unit and is used to transmit the attention heat map to the intelligent adaptation unit; A motion prediction unit, used to collect user head posture data and generate a head motion prediction trajectory; the motion prediction unit is connected to the intelligent adaptation unit and is used to transmit the head motion prediction trajectory to the intelligent adaptation unit; An intelligent adaptation unit is used to locate the pathological knowledge node of current concern in the medical knowledge graph according to the attention heat map, dynamically optimize the pre-allocation of rendering resources in the teaching scene based on the predicted head movement trajectory, and select the teaching content of the next stage according to the hierarchical association of the pathological knowledge nodes.

2. The wearable VR system for health education according to claim 1, characterized in that: The knowledge modeling unit includes a knowledge extraction module, a relationship construction module and a graph storage module; The knowledge extraction module is used to extract pathological knowledge nodes from medical textbooks; The relationship building module is used to establish the clinical diagnosis and treatment relationship between pathological knowledge nodes, specifically: A multi-layer classification method is used to divide the node levels of pathological knowledge nodes; The text semantic analysis method is used to extract the contextual association features between pathological knowledge nodes; Calculating clinical correlation scores between pathological knowledge nodes based on a deep neural network; the deep neural network adopts a bidirectional long short-term memory network structure, specifically inputting the associated feature vectors of two nodes into the deep neural network, extracting the clinical logical relationship between the features through temporal dependency analysis, and using a sigmoid activation function to normalize the network output to a clinical correlation score in the range of 0 to 1; A segmented mapping method is used to assign weight coefficients to connections between pathological knowledge nodes; the segmented mapping method is based on a clinical relevance score, specifically, when the clinical relevance score is less than a first preset threshold, a weight coefficient is assigned as A, when the clinical relevance score is between the first preset threshold and the second preset threshold, the weight coefficient is proportional to the score, and when the clinical relevance score is greater than the second preset threshold, a weight coefficient is assigned as B; Use graph structure quantification methods to build multi-level knowledge structures; The graph storage module is connected to the relationship construction module and stores the constructed medical knowledge graph.

3. The wearable VR system for health education according to claim 2, characterized in that: The resource preprocessing unit includes a hierarchical analysis module, a resource allocation module and a cache management module; The hierarchical analysis module performs quantitative calculation on the hierarchical associations between the pathological knowledge nodes, specifically: Calculate the out-degree and in-degree of the pathological knowledge node in the medical knowledge graph; the out-degree represents the number of clinical diagnosis and treatment relationship edges pointing from the pathological knowledge node to other pathological knowledge nodes, and the in-degree represents the number of clinical diagnosis and treatment relationship edges pointing to the pathological knowledge node; Constructing the association propagation matrix of the pathological knowledge node includes the following steps: Establish an N×N square matrix; wherein N is the total number of the pathological knowledge nodes in the medical knowledge graph, and the rows and columns of the square matrix correspond to different pathological knowledge nodes respectively; Traverse each of the clinical diagnosis and treatment relationship edges in the medical knowledge graph, use the starting node of the clinical diagnosis and treatment relationship edge as the row index of the square matrix, and the ending node as the column index, fill in the weight coefficient at the corresponding position, and construct an initial association propagation matrix; wherein, if there is a direct connection between the pathological knowledge nodes, the value of the corresponding position of the association propagation matrix is ​​the weight coefficient determined by the relationship construction module through the segmented mapping method, and if there is no direct connection, it is zero; Normalizing each row of the initial correlation propagation matrix so that the sum of the elements in each row is 1, thereby obtaining the normalized correlation propagation matrix M; Generate the hierarchical importance score of the pathological knowledge node through iterative calculation. After the iteration converges, use the corresponding value in the importance vector obtained when the iteration converges as the hierarchical importance score, and make the following adjustments according to the difference between the out-degree and the in-degree of each pathological knowledge node: if the difference between the out-degree and the in-degree of the pathological knowledge node is greater than half of the total number of connections of the pathological knowledge node, adjust the hierarchical importance score to the square of the current value; otherwise, keep the hierarchical importance score unchanged; The resource allocation module determines the priority of the rendering resource pre-allocation based on the edge computing device according to the hierarchical association, including the following steps: Calculating the resource demand of the pathology knowledge node, and determining the resource allocation priority based on the hierarchical importance score; wherein, if the hierarchical importance score of the pathology knowledge node exceeds the average value of the hierarchical level of the pathology knowledge node, then increasing the resource allocation priority of the pathology knowledge node; Evaluate the computing capability of the edge computing device and obtain the available resources of the edge computing device; Formulate the rendering resource pre-allocation strategy based on the resource allocation priority and the available resource amount; wherein, when the resource demand exceeds the available resource amount of the edge computing device, split the rendering task with a lower priority into multiple subtasks according to the task dependency; The cache management module is connected to the resource allocation module and manages the storage space of the edge computing device.

4. The wearable VR system for health education according to claim 3, characterized in that: The attention tracking unit includes an image acquisition module, a pupil positioning module and a heat map generation module; The image acquisition module acquires a user eye image sequence at a preset frame rate through a binocular camera, and performs denoising and illumination compensation processing on the user eye image sequence to obtain a sight line image; The pupil positioning module performs pupil detection and corneal reflection point positioning on the sight line image, and calculates the screen coordinates of the sight line focus based on the relative position of the pupil center and the corneal reflection point; wherein, if the pupil contour obtained by the pupil detection meets the ellipticity requirement and the brightness value of the corneal reflection point is greater than the average brightness value of the sight line image, the screen coordinates of the sight line focus are calculated using the positional relationship between the pupil center and the corneal reflection point; otherwise, the current sight line image is marked as an invalid frame and returned for re-collection; The process of determining whether the pupil contour meets the ellipticity requirement includes: extracting a contour point set of the pupil contour, calculating a minimum circumscribed rectangle of the contour point set, constructing a standard ellipse based on the major axis and the minor axis of the minimum circumscribed rectangle, calculating an average distance deviation from points in the contour point set to the standard ellipse, defining a ratio of the average distance deviation to the perimeter of the standard ellipse as an ellipticity fitting error, and determining whether the ellipticity fitting error is less than a ratio of the area to the perimeter of the pupil contour; The process of calculating the screen coordinates of the sight focus using the positional relationship between the pupil center and the corneal reflection point includes: establishing a polar coordinate system with the pupil center as the origin, calculating the polar coordinate parameters of the corneal reflection point relative to the pupil center, constructing a sight vector according to the polar coordinate parameters, and taking the intersection of the sight vector and the display plane as the screen coordinates of the sight focus; The heat map generation module performs spatiotemporal clustering analysis on the gaze focus, and generates an attention heat map according to the residence time and spatial distribution characteristics of the gaze focus; wherein, if the hierarchical importance score of the pathological knowledge node corresponding to the gaze focus is higher than the average value of the nodes at the same level, and the residence time of the gaze focus in a certain area exceeds the average residence time of adjacent areas, then in the attention heat map, the heat value of a certain area is set to the product of the gaze focus residence time and the hierarchical importance score; otherwise, the heat value of the area is set to the gaze focus residence time value.

5. The wearable VR system for health education according to claim 4, characterized in that: The motion prediction unit includes a data acquisition module, a state estimation module and a trajectory prediction module; The data acquisition module acquires head posture sensor data, performs low-pass filtering preprocessing on the head posture sensor data, and constructs a time series data sequence according to the sampling frequency of the sensor; the head posture sensor data includes three-axis acceleration, three-axis angular velocity and three-axis magnetic field intensity; The state estimation module estimates the current motion state based on the recursive Bayesian prediction method, specifically constructing a 16-dimensional state vector including position, velocity, acceleration and attitude quaternion, establishing a nonlinear state transfer equation considering sensor noise, using a particle filter algorithm for state estimation, and correcting abnormal states through Kalman smoothing; the determination of the abnormal state includes state mutation determination, physical constraint violation determination and sensor abnormality determination; the state mutation determination is realized by calculating the change in the state estimation value at adjacent moments and comparing it with the respective relevant thresholds, including position mutation, velocity mutation, acceleration mutation and attitude angle mutation; the physical constraint violation determination includes speed overrun, acceleration overrun and angular velocity overrun; the sensor abnormality is judged by observing the residual; when an abnormal state is detected, a fixed interval Kalman smoother is used to correct the abnormal state; The trajectory prediction module is connected to the state estimation module and generates the head movement prediction trajectory.

6. The wearable VR system for health education according to claim 5, characterized in that: The execution process of the particle filter algorithm includes: sampling N particles from the state prior distribution, calculating the importance weight of each particle based on the observation equation, normalizing the importance weight, calculating the effective particle number, resampling when the effective particle number is less than the effective particle number threshold, and finally obtaining the posterior distribution estimate of the state vector.

7. The wearable VR system for health education according to claim 6, characterized in that: The intelligent adaptation unit includes a node positioning module, an optimization strategy module and a content scheduling module; The node positioning module determines the pathology knowledge node of current concern in the medical knowledge graph; The optimization strategy module generates a dynamic optimization scheme for pre-allocation of rendering resources based on a deep reinforcement learning model; The content scheduling module executes the dynamic optimization scheme and updates the teaching content of the next stage according to the hierarchical association of the pathological knowledge nodes.

8. A method using the wearable VR system for health education as claimed in any one of claims 1 to 7, characterized in that: include, Constructing a medical knowledge graph including pathological knowledge nodes and clinical diagnosis and treatment relationship edges; the medical knowledge graph constructs the relationship between nodes through concept mapping; Pre-allocating rendering resources for teaching scenes based on edge computing devices according to the hierarchical association between the pathological knowledge nodes and the clinical diagnosis and treatment relationship edges; Use binocular cameras to obtain the user's visual focus and generate an attention heat map; The attention heat map represents the distribution of users' attention to different areas in the teaching scene; The head posture data of the user is collected based on the head posture sensor and a head movement prediction trajectory is generated through recursive Bayesian prediction; the head movement prediction trajectory is used to estimate the field of view range of the user at the next moment; Locating the pathological knowledge node of current concern in the medical knowledge graph according to the attention heat map, dynamically optimizing the pre-allocation of rendering resources in the teaching scene based on the head movement prediction trajectory, and selecting the teaching content of the next stage according to the hierarchical association of the pathological knowledge nodes; The dynamic optimization adopts a deep reinforcement learning model and takes the attention heat map and the head movement prediction trajectory as state inputs.

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 wearable VR system for health education described in any one of claims 1 to 7 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 wearable VR system for health education described in any one of claims 1 to 7 are implemented.

Images