Intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning

Abstract

The invention relates to the technical field of diagnosis and treatment path planning, in particular to an intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning, which comprises a symptom specialized mapping module, a holographic panoramic navigation module, a path optimization core module and a real-time feedback module, wherein the symptom specialized mapping module generates a dynamic medical resource recommendation parameter set by fusing a symptom ontology feature vector, medical institution treatment efficiency data and a specialized association rule base, and dynamic medical resource recommendation parameters comprise a multi-dimensional matching degree tensor and a treatment efficiency quantification matrix; the holographic panoramic navigation module receives the multi-dimensional matching degree tensor and constructs a three-dimensional medical resource topological graph, wherein the three-dimensional medical resource topological graph comprises hospital nodes and department connecting lines; the path optimization core module analyzes hospital nodes in the three-dimensional medical resource topological graph, and generates a path with optimal cost performance based on a multi-objective optimization algorithm of a treatment efficiency quantification matrix; and the real-time feedback module acquires and updates the treatment efficiency quantization matrix through a back propagation algorithm to form a closed-loop optimization system.

Description

Technical Field

[0001] The present invention relates to the technical field of diagnosis and treatment path planning, specifically an intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning. Background Art

[0002] Currently, medical resource recommendation systems generally suffer from problems such as multi-dimensional data fragmentation, insufficient dynamic adaptability, and weak visualization decision support. Traditional systems mostly rely on historical cure rates of departments or expert rule bases for static recommendations, lacking comprehensive analysis of symptom characteristics, real-time treatment efficacy, and spatial constraints. For example, a patient with chest pain may be mechanically assigned to the cardiology department without considering the response speed of the emergency department in different hospitals or the availability of interventional treatment equipment.

[0003] In addition, in the prior art, the matching between symptoms and departments is mostly based on a two-dimensional matrix, which cannot express associations above three dimensions such as the geographical location of the hospital and the cooperation relationship between departments; the update of treatment effect data lags behind, making it difficult to timely reflect the efficacy changes after the upgrade of department equipment or the application of new therapies; the path planning is mainly based on a single goal (such as the lowest cost), ignoring the multi-goal game of cure rate, time, distance, etc.; the visualization interface is limited to two-dimensional charts, making it difficult for doctors and patients to intuitively understand the spatial distribution and optimization logic of cross-institutional resources. These problems lead to the deviation of recommendation results from actual needs, an increase in the risk of misdiagnosis, and the inability to adapt to the dynamic changes of medical resources. Summary of the Invention

[0004] Aiming at the three core problems in medical resource recommendation: single dimension matching between symptoms and departments, inefficient multi-goal path planning under spatio-temporal constraints, and insufficient system dynamic adaptability; firstly, traditional methods cannot quantify the multi-dimensional associations between symptom characteristics and department expertise; secondly, the cross-institutional referral path planning lacks comprehensive optimization of time, space, and efficacy (cure rate / cost); finally, the static rule base is difficult to timely correct the recommendation deviation caused by the development of medical technology or resource changes. These problems result in low resource matching accuracy, extended treatment cycles, and long-term system performance degradation.

[0005] To achieve the above object, the present invention provides an intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning, which includes a symptom-specialty mapping module, a holographic panoramic navigation module, a path optimization core module, and a real-time feedback module, wherein:

[0006] The symptom-specialty mapping module generates a dynamic medical resource recommendation parameter set by fusing symptom ontology feature vectors, medical institution treatment efficacy data, and specialty association rule bases. The dynamic medical resource recommendation parameters include a multi-dimensional matching degree tensor and a treatment efficacy quantification matrix;

[0007] The holographic panoramic navigation module receives the multi-dimensional matching degree tensor and constructs a three-dimensional medical resource topology map, where the spatial coordinates of hospital nodes map to actual geographical locations, and the optical attributes of department connection lines are dynamically rendered using the ratio of treatment efficacy to cost in the treatment efficacy quantization matrix;

[0008] The path optimization core module analyzes the spatio-temporal constraint parameters of hospital nodes in the three-dimensional medical resource topology map and generates the most cost-effective path based on the multi-objective optimization algorithm of the treatment efficacy quantization matrix;

[0009] The real-time feedback module collects the actual execution data of the most cost-effective path and the heat key hospital nodes identified according to the optical attributes of department connection lines, and updates the treatment efficacy quantization matrix through the backpropagation algorithm to form a closed-loop optimization system.

[0010] As a further improvement of this technical solution, by fusing symptom ontology feature vectors, medical institution treatment efficacy data, and specialty association rule libraries, a dynamic medical resource recommendation parameter set containing multi-dimensional matching degree tensors is generated. The multi-dimensional matching degree tensor is used to quantify the comprehensive matching degree among symptoms, medical institutions, and departments. The calculation formula for each element Mijk in the multi-dimensional matching degree tensor M is as follows:

[0011] Mijk = w1·similarity(Si, Rjk) + w2·Tjk, where Mijk represents the matching degree between the i-th symptom and the k-th department of the j-th medical institution; Si represents the i-th symptom feature vector; Rjk represents the association rule between the k-th department of the j-th medical institution and the symptom; Ejk represents the treatment effect index of the k-th department of the j-th medical institution; similarity(Si, Rjk) represents the similarity between the symptom feature vector Si and the association rule Rjk; w1 and w2 are weight coefficients used to adjust the importance of different factors;

[0012] Traditional two-dimensional matrices (such as symptom-department matching tables) cannot express the association between spatial distribution and dynamic efficacy (real-time cure rate of departments), while the multi-dimensional matching degree tensor M is used to balance symptom adaptability and treatment efficacy, avoiding recommending departments that are far from patients or have insufficient efficacy.

[0013] As a further improvement of this technical solution, the three-dimensional medical resource topology map includes hospital nodes and department connections, and the specific construction is as follows:

[0014] Each hospital is represented as a node in the three-dimensional topology map. As a hospital node, the spatial coordinates of each hospital node correspond to its actual geographical location;

[0015] In a three-dimensional topological graph, different hospital nodes and different departments within a hospital are connected by connecting lines. The optical properties of the department connecting lines are dynamically rendered using the ratio of treatment efficacy to cost in the treatment efficacy quantification matrix. The dynamic rendering includes color depth, thickness, and transparency. By normalizing the ratio of treatment efficacy to cost in the treatment efficacy quantification matrix and making a positive correlation adjustment to the color depth, thickness, and transparency according to the normalized ratio of treatment efficacy to cost, the positive correlation adjustment means that the higher the ratio of treatment efficacy to cost, the higher the color depth, thickness, and transparency;

[0016] Since two-dimensional charts cannot intuitively express spatial accessibility and multi-objective conflicts, high-cost-effective departments and high-cost-effective departments are quickly identified through a three-dimensional medical resource topological graph.

[0017] As a further improvement of this technical solution, the identification process of the thermal key hospital nodes specifically includes:

[0018] The thermal key hospital nodes identified according to the optical properties of the department connecting lines are obtained by mapping and converting the optical properties of the department connecting lines into thermal values, and generating a radial thermal field centered on each hospital node, and iteratively searching along the gradient direction of the thermal field to find the local maximum point as the thermal key hospital node;

[0019] The real-time feedback module includes an update unit, and the update unit is used to update the treatment efficacy quantification matrix in the symptom-specialty mapping module, specifically including:

[0020] By minimizing the deviation between the optimal cost-effective path and the thermal key hospital nodes, the path deviation loss and the treatment efficacy error are obtained, and the cure rate Curability(Ejk) in the treatment efficacy quantification matrix is updated. The update formula is as follows:

[0021] Where Lp represents the path deviation loss, and Le represents the treatment efficacy error; η represents the learning rate, which is used to control the step size of parameter update; represents the partial derivative of the path deviation loss and the treatment efficacy error with respect to the cure rate Curability(Ejk);

[0022] Curability′(Ejk) = Curability(Ejk) - ΔCurability(Ejk), where Curability′(Ejk) represents the cure rate in the updated treatment efficacy quantification matrix;

[0023] According to the cure rate in the updated treatment efficacy quantification matrix, the ratio of treatment efficacy to cost in the treatment efficacy quantification matrix E is recalculated to form a closed-loop optimization system until the maximum element change amount of the treatment efficacy quantification matrix E is less than the preset safety threshold;

[0024] Since the static rule library cannot adapt to the progress of medical technology (such as the popularization of new therapies), through closed-loop optimization, the parameters are dynamically corrected, that is, the cure rate in the treatment effect quantification matrix. For example, after the department equipment is upgraded, the actual cure rate data triggers real-time updates.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] By fusing the symptom ontology feature vector, the medical institution treatment efficiency data, and the specialty association rule library through the multi-dimensional matching degree tensor, the accuracy and scenario adaptability of medical resource matching are significantly improved; the three-dimensional medical resource topology map dynamically renders the ratio of treatment effect to cost, intuitively presenting the spatial distribution and resource status of high-cost-effective departments, and enhancing the decision-making efficiency of cross-institutional collaboration; the path optimization module combines spatio-temporal constraints and multi-objective game theory to balance the cure rate, cost, and accessibility, generating the optimal cost-effective path; the real-time feedback mechanism dynamically corrects the treatment efficiency parameters, that is, the cure rate in the treatment effect quantification matrix, through heat key node recognition and backpropagation update, forming a closed-loop optimization, enabling the system to continuously adapt to the development of medical technology and resource changes, and finally achieving the coordinated improvement of precise recommendation, efficient resource utilization, and system self-evolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the overall module of the present invention;

[0028] Figure 2 It is a schematic diagram of the unit of the real-time feedback module of the present invention;

[0029] Figure 3 It is a schematic diagram of the overall process of the planning system of the present invention;

[0030] In the figure: 100, symptom specialty mapping module; 200, holographic panoramic navigation module; 300, path optimization core module; 400, real-time feedback module; 401, key node recognition unit; 402, update unit; 500, personal health record system; 600, symptom ontology input system; 700, intelligent medical resource service system. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0032] Next, please refer to Figure 1, the present invention provides a technical solution: an intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning, including a symptom specialty mapping module 100, a holographic panoramic navigation module 200, a path optimization core module 300, and a real-time feedback module 400.

[0033] The symptom specialty mapping module 100 generates a dynamic medical resource recommendation parameter set by fusing symptom ontology feature vectors, medical institution treatment efficacy data, and specialty association rule bases, where:

[0034] The symptom ontology feature vector represents a set of data used to describe the specific symptoms of a patient, constructed based on a medical knowledge graph. Each symptom has its corresponding feature vector. The symptom feature vector includes, but is not limited to, the frequency of symptom occurrence, severity, duration, etc. Relevant information is extracted from the patient's medical record or self-report using natural language processing technology and converted into a form that can be used for calculation;

[0035] The medical institution treatment efficacy data reflects the performance of different hospitals and treatment departments in dealing with specific types of diseases, and is used to extract the treatment effect indicators of each medical institution and its treatment department, including the cure rate and treatment cost. By performing big data analysis on historical diagnosis and treatment records, the treatment effect indicators of each medical institution and its treatment department are obtained;

[0036] The specialty association rule base represents a predefined knowledge base used to extract the association rules between departments and symptoms, and contains the relationship models between various symptoms and the corresponding best treatment departments. For example, for a specific heart problem, there are multiple treatment methods available (such as drug treatment, surgical intervention, etc.), and each method is suitable for different stages of the disease development. This rule base will help determine which types of experts or departments are most suitable for dealing with a given combination of symptoms.

[0037] By fusing the symptom ontology feature vector, medical institution treatment efficacy data, and specialty association rule base, a dynamic medical resource recommendation parameter set including a multi-dimensional matching degree tensor and a treatment efficacy quantification matrix is generated. Among them, the treatment efficacy quantification matrix is a data structure containing multiple treatment effect indicators, used to quantify the treatment effect indicators of different medical institutions and their departments. Each element Ejk in the treatment efficacy quantification matrix E represents the treatment effect indicator of the kth department of the jth medical institution;

[0038] Among them, the multi-dimensional matching degree tensor is used to quantify the comprehensive matching degree between symptoms, medical institutions, and departments. Its design purpose is to achieve precise resource matching through multi-dimensional cross-analysis in complex medical scenarios. The calculation formula for each element Mijk in the multi-dimensional matching degree tensor M is as follows:

[0039] $M_{ijk} = w_1 \cdot similarity(S_i, R_{jk}) + w_2 \cdot T_{jk}$, where $M_{ijk}$ represents the matching degree between the $i$-th symptom and the $k$-th department of the $j$-th medical institution; $S_i$ represents the $i$-th symptom feature vector; $R_{jk}$ represents the association rule between the $k$-th department of the $j$-th medical institution and the symptom; $E_{jk}$ represents the treatment effect index of the $k$-th department of the $j$-th medical institution; $similarity(S_i, R_{jk})$ represents the similarity between the symptom feature vector $S_i$ and the association rule $R_{jk}$; $w_1$ and $w_2$ are weight coefficients used to adjust the importance of different factors.

[0040] Combine the multi-dimensional matching degree tensor $M$ and the treatment effect quantification matrix $E$ into a dynamic medical resource recommendation parameter set $P$.

[0041] The calculation formula for the curability-cost ratio in the treatment effect quantification matrix $E$ is as follows:

[0042] $T_{jk}$ represents the curability-cost ratio of the $k$-th department of the $j$-th medical institution; $Curability(E_{jk})$ represents the cure rate of the $k$-th department of the $j$-th medical institution; $Cost(E_{jk})$ represents the treatment cost of the $k$-th department of the $j$-th medical institution.

[0043] The holographic panoramic navigation module 200 receives the multi-dimensional matching degree tensor and constructs a three-dimensional medical resource topology graph. The three-dimensional medical resource topology graph includes hospital nodes and department connection lines. The specific construction is as follows:

[0044] Each hospital is represented as a node in the three-dimensional topology graph, serving as a hospital node. The spatial coordinates of each hospital node correspond to its actual geographical location. For example, the actual location of the hospital is mapped to the corresponding location in the three-dimensional space using longitude and latitude data;

[0045] In the three-dimensional topology graph, different hospital nodes and different departments within the hospital are connected by connection lines. The optical properties of the department connection lines are dynamically rendered using the curability-cost ratio in the treatment effect quantification matrix. The dynamic rendering includes color depth, thickness, and transparency. By normalizing the curability-cost ratio in the treatment effect quantification matrix and making a positive correlation adjustment to the color depth, thickness, and transparency according to the normalized curability-cost ratio, the positive correlation adjustment means that the higher the curability-cost ratio, the higher the color depth, thickness, and transparency.

[0046] The path optimization core module 300 analyzes the spatio-temporal constraint parameters of the hospital nodes in the three-dimensional medical resource topology graph and generates the optimal cost-effective path based on the multi-objective optimization algorithm of the treatment effect quantification matrix. Specifically, it includes:

[0047] The spatio-temporal constraint parameters of hospital nodes in the three-dimensional medical resource topology map include time constraints and space constraints. The time constraints include appointment waiting time, treatment time, transportation time, etc.; the space constraints include the actual geographical location of the hospital, the physical distance between departments, etc.;

[0048] Construct a multi-objective function based on the treatment efficacy quantification matrix. The multi-objective function includes maximizing the cure rate, minimizing the treatment cost, shortest time, and shortest space distance;

[0049] Select a multi-objective optimization algorithm, including but not limited to genetic algorithm and particle swarm optimization, etc.; Through multiple iterations, determine the optimal cost-effective path. The optimal cost-effective path includes all intermediate hospital nodes and their department connection lines from the initial hospital to the final treatment hospital.

[0050] Please refer to Figure 2 , the key node recognition unit 401 in the real-time feedback module 400 collects the actual execution data of the optimal cost-effective path, and the thermal key hospital nodes identified according to the optical properties of the department connection lines, specifically including:

[0051] According to the hospital nodes and department connection lines in the three-dimensional medical resource topology map, use 3D modeling tools to model the internal environment of the hospital and construct a virtual hospital environment; Visualize the optimal cost-effective path in the virtual hospital environment. The user can see the virtual path through the mixed reality headset and interact to collect the actual execution data of the optimal cost-effective path;

[0052] The thermal key hospital nodes identified according to the optical properties of the department connection lines are obtained by mapping and converting the optical properties of the department connection lines into thermal values, generating a radial thermal field centered on each hospital node, and iteratively searching along the gradient direction of the thermal field to find the local maximum point as the thermal key hospital node.

[0053] The update unit 402 in the real-time feedback module 400 updates the treatment efficacy quantification matrix in the symptom-specialty mapping module 100 through the backpropagation algorithm to form a closed-loop optimization system, specifically including:

[0054] Update the treatment efficacy quantification matrix through the backpropagation algorithm. Specifically, by minimizing the deviation between the optimal cost-effective path and the thermal key hospital nodes, obtain the path deviation loss and treatment efficacy error, and update the cure rate Curability(Ejk) in the treatment efficacy quantification matrix. The update formula is as follows:

[0055] Where Lp represents the path deviation loss, and Le represents the treatment efficacy error; η represents the learning rate (usually set to 0.01), which is used to control the step size of parameter update to prevent update failure caused by too large (oscillation) or too small (slow convergence) step size; Denotes the partial derivative of the path deviation loss and the treatment efficacy error with respect to the curability (Curability(Ejk));

[0056] Curability′(Ejk) = Curability(Ejk) - ΔCurability(Ejk), where Curability′(Ejk) represents the curability in the updated treatment efficacy quantization matrix;

[0057] According to the curability in the updated treatment efficacy quantization matrix, recalculate the efficacy - cost ratio in the treatment efficacy quantization matrix E to form a closed - loop optimization system until the maximum element change of the treatment efficacy quantization matrix E is less than the preset safety threshold.

[0058] Please refer to Figure 3 , To improve the path of the overall intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning, the system starts with the input of the personal health record system 500, the symptom ontology input system 600, and the intelligent medical resource service system 700. Through multimodal intelligent diagnosis and treatment, it generates a fusion of symptom ontology feature vectors, medical institution treatment efficacy data, and specialty association rule bases. Combining with the symptom - specialty mapping module 100, it calculates the multi - dimensional matching degree tensor and the treatment efficacy quantization matrix, and maps the recommended hospital specialties to a three - dimensional topological space. The holographic panoramic navigation module 200 constructs a visual medical network based on the optical properties (color, thickness dynamically reflecting the efficacy - cost ratio) of the department connection lines. The path optimization core module 300 then uses a multi - objective algorithm to balance curability, cost, and spatio - temporal constraints to generate the optimal diagnosis and treatment path. During the execution process, the data security system ensures the flow of medical data throughout the process through encryption and permission control. The real - time feedback module 400 continuously collects the actual execution data of the path with the best cost - performance ratio, driving the model to dynamically update the treatment efficacy quantization matrix in the symptom - specialty mapping module through backpropagation, forming a "diagnosis and treatment - feedback - learning" closed loop.

[0059] Ultimately, the intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning realizes progressive optimization based on data accumulation and model self - training, making the recommended path always fit the actual medical resource efficacy and patient needs.

[0060] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above - mentioned embodiments. The above - mentioned embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. Intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning, characterized by: It includes a symptom specialty mapping module (100), a holographic panoramic navigation module (200), a path optimization core module (300), and a real-time feedback module (400), where: The symptom specialty mapping module (100) generates a dynamic medical resource recommendation parameter set by integrating symptom ontology feature vectors, medical institution treatment efficacy data, and a specialty association rule base. The dynamic medical resource recommendation parameters include a multi-dimensional matching degree tensor and a treatment efficacy quantification matrix; The holographic panoramic navigation module (200) receives the multi-dimensional matching degree tensor and constructs a three-dimensional medical resource topology map, where the spatial coordinates of hospital nodes map to actual geographical locations, and the optical properties of department connection lines are dynamically rendered using the efficacy-cost ratio in the treatment efficacy quantification matrix; The path optimization core module (300) analyzes the spatio-temporal constraint parameters of hospital nodes in the three-dimensional medical resource topology map and generates the most cost-effective path based on the multi-objective optimization algorithm of the treatment efficacy quantification matrix; The real-time feedback module (400) collects the actual execution data of the most cost-effective path and the thermal key hospital nodes identified according to the optical properties of department connection lines, and updates the treatment efficacy quantification matrix through the backpropagation algorithm to form a closed-loop optimization system.

2. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1 is characterized in that: The treatment efficacy quantification matrix includes a data structure of multiple treatment effect indicators for quantifying the treatment effect indicators of different medical institutions and their departments. Each element Ejk in the treatment efficacy quantification matrix E represents the treatment effect indicator of the k-th department of the j-th medical institution.

3. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 2, characterized in that The treatment effect indicators in the treatment efficacy quantification matrix include the cure rate and the treatment cost.

4. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1, wherein The multi-dimensional matching degree tensor is used to quantify the comprehensive matching degree among symptoms, medical institutions, and departments. The calculation formula for each element Mijk in the multi-dimensional matching degree tensor M is as follows: Mijk = w1·similarity(Si, Rjk) + w2·Tjk, where Mijk represents the matching degree between the i-th symptom and the k-th department of the j-th medical institution; Si represents the i-th symptom feature vector; Rjk represents the association rule between the k-th department of the j-th medical institution and the symptom; Ejk represents the treatment effect indicator of the k-th department of the j-th medical institution; similarity(Si, Rjk) represents the similarity between the symptom feature vector Si and the association rule Rjk; w1 and w2 are weight coefficients used to adjust the importance of different factors.

5. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1 is characterized in that: The calculation formula for the efficacy-cost ratio in the treatment efficacy quantification matrix is as follows: Tjk represents the efficacy-cost ratio of the kth department of the jth medical institution; Curability(Ejk) represents the cure rate of the kth department of the jth medical institution; Cost(Ejk) represents the treatment cost of the kth department of the jth medical institution.

6. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1 is characterized in that: The three-dimensional medical resource topology map includes hospital nodes and department connections, and the specific construction is as follows: Each hospital is represented as a node in the three-dimensional topology map, serving as a hospital node. The spatial coordinates of each hospital node correspond to its actual geographical location; In the three-dimensional topology map, different hospital nodes and different departments within the hospital are connected by connecting lines. The optical properties of the department connecting lines are dynamically rendered using the therapeutic effect-cost ratio in the therapeutic effect quantification matrix, where the dynamic rendering includes color depth, coarseness, and transparency. The therapeutic effect-cost ratio in the therapeutic effect quantification matrix is normalized, and the color depth, coarseness, and transparency are positively correlated and adjusted based on the normalized therapeutic effect-cost ratio. The positive correlation adjustment means that the higher the therapeutic effect-cost ratio, the higher the color depth, coarseness, and transparency.

7. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1 is characterized in that: The process of generating the optimal cost-effective path specifically includes: The spatiotemporal constraint parameters of hospital nodes in the three-dimensional medical resource topology graph include time constraints and space constraints; A multi-objective function is constructed based on the treatment efficacy quantification matrix, which includes maximizing cure rate, minimizing treatment cost, minimizing time and minimizing spatial distance; Using a multi-objective optimization algorithm, the cost-effective path is determined through multiple iterations. The cost-effective path includes all intermediate hospital nodes and their department connection lines from the initial hospital to the final treatment hospital.

8. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1, wherein, The real-time feedback module (400) includes a key node identification unit (401), which is used to collect actual execution data and identify thermal key hospital nodes, wherein the process of collecting actual execution data specifically includes: Based on the hospital nodes and department connection lines in the three-dimensional medical resource topology map, 3D modeling tools are used to model the internal environment of the hospital and construct a virtual hospital environment. The most cost-effective path is visualized in the virtual hospital environment. Users can see the virtual path through a mixed reality headset and interact with it to collect actual execution data of the most cost-effective path.

9. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 8, wherein The identification process of the thermal key hospital node specifically includes: The thermally critical hospital nodes are identified based on the optical properties of the department connection lines. The optical property mapping of the department connection lines is converted into thermal values. A radial thermal field is generated with each hospital node as the center. An iterative search is performed along the gradient direction of the thermal field to find the local maximum point, which is used as the thermally critical hospital node.

10. The intelligent diagnosis and treatment path dynamic planning system based on symptom ontology reasoning according to claim 1, wherein The real-time feedback module (400) includes an updating unit (402), which is used to update the treatment efficacy quantization matrix in the symptom-specialty mapping module (100), specifically including: By minimizing the deviation between the cost-effective optimal path and the key thermal hospital node, the path deviation loss and treatment efficacy error are obtained, and the cure rate Curability (Ejk) in the treatment efficacy quantification matrix is updated. The update formula is as follows: Where Lp represents the path deviation loss, Le represents the treatment efficacy error; η represents the learning rate, which is used to control the step size of parameter update; represents the partial derivatives of path deviation loss and treatment efficacy error with respect to cure rate Curability(Ejk); Curability′(Ejk)=Curability(Ejk)-ΔCurability(Ejk), where Curability′(Ejk) represents the cure rate in the updated treatment efficacy quantification matrix; According to the cure rate in the updated treatment efficacy quantization matrix, the efficacy-cost ratio in the treatment efficacy quantization matrix E is recalculated to form a closed-loop optimization system until the maximum element change in the treatment efficacy quantization matrix E is less than the preset safety threshold.

Images