Intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching

Abstract

The invention discloses an intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching, and relates to the field of intelligent medical treatment, and the system comprises a cloud database which is used for uploading patient diagnosis and treatment cases and storing the patient diagnosis and treatment cases; the selection module is used for selecting a patient diagnosis and treatment case in the cloud database and taking the selected diagnosis and treatment case as a to-be-simulated diagnosis and treatment case; the editing module is used for obtaining the to-be-simulated diagnosis and treatment case selected by the selection module, reading and manually editing diagnosis logic and medication details on the basis of the to-be-simulated diagnosis and treatment case, and massive real and high-quality learning resources are provided for medical teaching by collecting numerous standard cases of successful rehabilitation of patients and performing screening and classified storage; a user can select from the cases for simulation diagnosis and treatment and manually edit diagnosis logic and medication details, and understanding and application of medical knowledge are deepened in practice.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical technology, and particularly to an intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching. Background Art

[0002] Medical teaching simulated diagnosis and treatment uses simulation technology to construct a simulated clinical environment, enabling medical students to practice diagnosis and treatment operations in a risk-free scenario. Through forms such as simulated patients, virtual cases, and standardized patients, it covers the full process training of medical history taking, physical examination, diagnosis, and treatment, helping students master clinical thinking and skills, improve communication abilities, and make up for the lack of real clinical practice opportunities. It is an important teaching method for integrating theory and practice in medical education.

[0003] The invention patent application with the application number 202311136970.X discloses an intelligent analysis system for electronic medical record data based on artificial intelligence, including: an electronic medical record data acquisition module for acquiring structured and unstructured data of each electronic medical record; an electronic medical record screening module for analyzing the similarity of diseases to which each electronic medical record belongs, and further screening to obtain each electronic medical record corresponding to newly emerged diseases; an electronic medical record classification module for analyzing the similarity of the same category of diseases to which each electronic medical record corresponding to newly emerged diseases belongs to other electronic medical records, and further classifying to obtain each electronic medical record corresponding to various newly emerged diseases: a disease risk parameter acquisition module for acquiring risk parameters of various newly emerged diseases; a disease risk parameter analysis module for analyzing the risk assessment coefficients of various newly emerged diseases; a disease transmission data acquisition module for acquiring the number of patients with various newly emerged diseases and the incubation periods of various newly emerged diseases within a set monitoring time period; a disease transmission data analysis module for analyzing the transmission speed of various newly emerged diseases: a disease data analysis module for analyzing the transmission risk coefficients of various newly emerged diseases according to the risk assessment coefficients and transmission speeds of various newly emerged diseases, further obtaining the prevention levels of various newly emerged diseases, and giving feedback. This application aims to solve the problem that "due to the lack of reference historical cases corresponding to uncommon diseases, it has a negative impact on the determination of their treatment plans, and the current analysis of electronic medical record data also ignores the analysis of the transmission risks of uncommon diseases, so the transmission risk levels of uncommon diseases cannot be obtained, and thus targeted prevention measures and screening suggestions cannot be provided, nor can people be timely reminded to take safety protection measures, which is not conducive to people's health and safety".

[0004] However, in the medical teaching scenario, most of the time, experienced physicians take intern physicians on rounds to accumulate diagnosis and treatment experience. This method is limited by factors such as the energy of experienced physicians and cannot provide conditions for intern physicians to learn independently and accumulate diagnosis and treatment experience.

[0005] For this reason, an intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching is proposed. Summary of the Invention

[0006] In view of the above-mentioned disadvantages of the prior art, the present invention provides an intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching, which can effectively solve the problems of the prior art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0008] The present invention discloses an intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching, including:

[0009] A cloud database for uploading patient diagnosis and treatment cases and storing the patient diagnosis and treatment cases; a selection module for selecting patient diagnosis and treatment cases in the cloud database and taking the selected diagnosis and treatment cases as the to-be-simulated diagnosis and treatment cases; an editing module for obtaining the to-be-simulated diagnosis and treatment cases selected by the operation of the selection module, and manually editing the diagnosis logic and medication details based on the reading of the to-be-simulated diagnosis and treatment cases; an evaluation module for receiving the diagnosis logic and medication details edited in the editing module, and comparing the received diagnosis logic and medication details with the diagnosis logic and medication details included in the to-be-simulated diagnosis and treatment cases to evaluate the simulated diagnosis and treatment score of the user from whom the diagnosis logic and medication details are edited; a classification unit for obtaining the historical evaluation results of the user's simulated diagnosis and treatment scores in the evaluation module and classifying the user based on the historical evaluation results. Further, the patient diagnosis and treatment cases uploaded in the cloud database are from any system-authorized end-users, and the patient diagnosis and treatment cases are all standard cases where the patients have successfully recovered. A filtering unit and a management unit are provided under the cloud database. The filtering unit is used to traverse the patient diagnosis and treatment cases uploaded in the cloud database and clean the low-value cases among the uploaded patient diagnosis and treatment cases. The management unit is used to receive the patient diagnosis and treatment cases that have been cleaned and distinguish the patient diagnosis and treatment cases based on the departments to which the patient diagnosis and treatment cases belong. The cloud database further stores the patient diagnosis and treatment cases separately based on the distinction results of the patient diagnosis and treatment cases;

[0010] Among them, the format of the patient diagnosis and treatment cases is text and images, and the content of the patient diagnosis and treatment cases includes: condition description, diagnosis logic, and medication details.

[0011] Further, the determination of low-value cases for patient diagnosis and treatment cases in the filtering unit follows:

[0012]

[0013] Where: SIMM(A,B) is the similarity between patient diagnosis and treatment case A and patient diagnosis and treatment case B;

[0014] Reconstruc tionCost(GA , G B ) is the semantic reconstruction cost; ω is the weight; is the average similarity of the image data contained in patient diagnosis and treatment case A and patient diagnosis and treatment case B respectively;

[0015] Among them, the similarity between each patient diagnosis and treatment case is calculated based on the above formula. The system-end user defines the threshold for picking up the cleaning target of the patient diagnosis and treatment case, and deletes any one of the two patient diagnosis and treatment cases that are greater than or equal to the threshold for picking up the cleaning target of the patient diagnosis and treatment case as the deletion target to complete the cleaning of the patient diagnosis and treatment case, where ω ∈ (0, 1).

[0016] Furthermore, when any one of the two patient diagnosis and treatment cases for which the similarity is calculated does not contain image data, the calculation formula of SIMM(A, B) is deformed into

[0017] That is, only the semantic similarity is considered in the similarity calculation process.

[0018] Furthermore, the value-taking logic of the semantic reconstruction cost ReconstructionCost(G A , G B ) is as follows:

[0019] Let the semantic topology graphs corresponding to patient diagnosis and treatment case A and patient diagnosis and treatment case B be G A (V A , E A ), G B (V B , E B ), where V is the set of semantic nodes and E is the set of semantic relationship edges. G A (V A , E A ), G B (V B , E B ) are abbreviated as G A , G B ; The logic for constructing the semantic topology graph is as follows: Using dependency syntax analysis and semantic role annotation, the patient diagnosis and treatment case is split into semantic atomic units as nodes V, and directed edges E are constructed according to the semantic relationship type, and the relationship weight is marked for each edge;

[0020] Find the optimal matching between the nodes of G A , G B through the Hungarian algorithm, minimize the semantic difference of the nodes, and then for the matched node pairs, calculate the relationship difference of the corresponding edges;

[0021]

[0022] where: e A and e B are the edges in the semantic topology graph G A corresponding to text A, and the edges in the semantic topology graph G B corresponding to text B; Type(e A ), Type(e B ) are the semantic relation types of the edges e A and e B ;

[0023] Similarity(Type(e A ), Type(e B )) is the similarity between Type(e A ) and Type(e B );

[0024] Finally, calculate ReconstructionCost(G A , G B ):

[0025]

[0026] where: (vA, vB) is the element composed of the matching nodes v A and v B in G A and G B ;

[0027] Matchednodes is the optimal matching set between the nodes of G A and G B found by the Hungarian algorithm;

[0028] NodeCost(v A , v B ) is the semantic difference cost between the nodes v A and v B , calculated based on the method of WordNet synset distance; (e A , e B ) is the element composed of the edges respectively related to the matching nodes v A and v B in G A and G B ; MatchedEdges is the set of edges corresponding to the matching node pairs; EdgeCost(e A , e B ) is the relation difference cost between the edges e A and e B , obtained based on the edge mapping cost calculation method.

[0029] Furthermore, during the operation stage of the selection module, the system user obtains permission to access the cloud database in the selection module. First, based on the selection of the patient diagnosis and treatment case type by differentiating the storage range in the cloud database, and then further select the patient diagnosis and treatment cases in the selected differentiated storage range.

[0030] Furthermore, a locking unit is provided inside the selection module. The locking unit is used to lock the diagnostic logic and medication details included in the content of each patient diagnosis and treatment case in the differentiated storage range when the system user selects the patient diagnosis and treatment cases in the differentiated storage range, making the locked part of the patient diagnosis and treatment case content unreadable;

[0031] Among them, after the system user selects the patient diagnosis and treatment cases, the readable part of the patient diagnosis and treatment cases is read synchronously, that is, the content of the condition description is read, and then the editing module is used to complete the editing of the diagnostic logic and medication details. After the diagnostic logic and medication details are completed, the locking module reopens the reading permission of the locked diagnostic logic and medication details.

[0032] Furthermore, a recording unit is provided at the lower level of the evaluation module. The recording unit is used to record the user's historical simulation diagnosis and treatment scores; the logic for evaluating the simulation diagnosis and treatment scores by comparing the diagnostic logic and medication details in the evaluation module is as follows:

[0033] Taking the diagnostic logic and medication details as the comparison targets respectively, the results are obtained based on the similarity calculation logic of the patient diagnosis and treatment cases, and the simulation diagnosis and treatment scores are comprehensively evaluated and output;

[0034] F = [SIMM logic (a, b) × λ + SIMM take (a, b) × (1 - λ)] × ε;

[0035] In the formula: F is the simulation diagnosis and treatment score; SIMM logic (a, b) is the similarity between the diagnostic logic edited by the user and the diagnostic logic in the case to be simulated; SIMM take (a, b) is the similarity between the medication details edited by the user and the medication details in the case to be simulated; λ is the weight coefficient; ε is the conversion factor of the simulation diagnosis and treatment score;

[0036] Among them, the simulation diagnosis and treatment score conversion factor ε > 0, and the weight coefficient λ satisfies λ < 1 - λ.

[0037] Furthermore, during the operation stage of the recording unit, the simulated diagnosis and treatment scores are differentiated and recorded based on the sources of the simulated diagnosis and treatment scores of the users. When the classification unit obtains the historical evaluation results of the users' simulated diagnosis and treatment scores, it obtains them from the recording unit. During the operation stage of the classification unit, the passing judgment threshold of the simulated diagnosis and treatment scores is synchronously set. Then, the latest simulated diagnosis and treatment scores of each user are obtained from the differentiated and recorded simulated diagnosis and treatment scores of the users, and compared with the passing judgment threshold of the simulated diagnosis and treatment scores. Based on the comparison result with the passing judgment threshold of the simulated diagnosis and treatment scores, the simulated diagnosis and treatment scores are differentiated, that is, the users are classified.

[0038] Furthermore, a filtering unit and a management unit are interconnected through wireless network interaction under the cloud database. The cloud database is interconnected with a selection module through wireless network interaction. Inside the selection module, it is interconnected with a locking unit through wireless network. The selection module is interconnected with an editing module and an evaluation module through wireless network. Under the evaluation module, a recording unit is interconnected through wireless network. The evaluation module is interconnected with the classification unit through wireless network. The classification unit is interconnected with the recording unit through wireless network.

[0039] Adopting the technical solution provided by the present invention, compared with the known prior art, it has the following beneficial effects:

[0040] The present invention provides an intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching. During the operation of the system, by collecting numerous standard cases where patients have successfully recovered, after screening, classification, and storage, it provides a large amount of real and high-quality learning resources for medical teaching. Users can select from these cases for simulated diagnosis and treatment, manually edit the diagnosis logic and medication details, and deepen the understanding and application of medical knowledge in practice. The system will compare the edited content with real cases and give simulated diagnosis and treatment scores to help users identify their own deficiencies. At the same time, the system will also record and classify the historical scores to facilitate understanding the progress trajectory of learners. According to the set passing judgment threshold, the learning achievements of learners can be differentiated. This not only improves the pertinence of medical teaching, enabling teachers to teach students in accordance with their aptitudes, but also effectively exercises the diagnosis and treatment thinking and practical abilities of learners, greatly improving the quality and efficiency of medical teaching and contributing to the cultivation of medical talents. Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0042] Figure 1It is a schematic structural diagram of an intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching. Detailed implementation manners

[0043] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. 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.

[0044] The present invention will be further described below with reference to the embodiments.

[0045] Embodiment:

[0046] An intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching in this embodiment, as Figure 1 shown, includes: a cloud database for uploading patient diagnosis and treatment cases and storing the patient diagnosis and treatment cases;

[0047] The patient diagnosis and treatment cases uploaded in the cloud database are from any system-authorized end users, and the patient diagnosis and treatment cases are all standard cases where the patients have successfully recovered. A filtering unit and a management unit are arranged at the lower level of the cloud database. The filtering unit is used to traverse the patient diagnosis and treatment cases uploaded in the cloud database and clean the low-value cases among the uploaded patient diagnosis and treatment cases. The management unit is used to receive the patient diagnosis and treatment cases that have been cleaned and distinguish the patient diagnosis and treatment cases based on the departments to which the patient diagnosis and treatment cases belong. The cloud database further stores the patient diagnosis and treatment cases separately based on the distinction results of the patient diagnosis and treatment cases; wherein, the format of the patient diagnosis and treatment cases is text and images, and the content of the patient diagnosis and treatment cases includes: disease description, diagnosis logic, and medication details;

[0048] For the determination of low-value cases of patient diagnosis and treatment cases in the filtering unit, it follows that:

[0049]

[0050] In the formula: SIMM(A, B) is the similarity between patient diagnosis and treatment case A and patient diagnosis and treatment case B;

[0051] ReconstructionCost(G A ,G B ) is the semantic reconstruction cost; ω is the weight; is the average value of the similarity of the image data included in patient diagnosis and treatment case A and patient diagnosis and treatment case B respectively;

[0052] Among them, the similarity between each patient diagnosis and treatment case is calculated based on the above formula. The system-end user defines the pickup threshold for cleaning the patient diagnosis and treatment case. For any one of the two patient diagnosis and treatment cases whose similarity is greater than or equal to the pickup threshold for cleaning the patient diagnosis and treatment case, a deletion operation is performed as the deletion target to complete the cleaning of the patient diagnosis and treatment case, where ω ∈ (0, 1);

[0053] When any one of the two patient diagnosis and treatment cases for which the similarity calculation is performed does not contain image data,

[0054] The calculation formula of SIMM(A, B) is deformed to

[0055] That is, only semantic similarity is considered in the similarity calculation process;

[0056] The value-taking logic of the semantic reconstruction cost ReconstructionCost(G A , G B ) is as follows:

[0057] Let the semantic topology graphs corresponding to patient diagnosis and treatment case A and patient diagnosis and treatment case B be G A (V A , E A ), G B (V B , E B ), respectively, where V is the set of semantic nodes and E is the set of semantic relationship edges. G A (V A , E A ), G B (V B , E B ) are abbreviated as G A , G B ; The semantic topology graph construction logic is as follows: Using dependency syntax analysis and semantic role annotation, the patient diagnosis and treatment case is split into semantic atomic units as nodes V, and directed edges E are constructed according to the semantic relationship type, and each edge is marked with a relationship weight;

[0058] Find the optimal matching between the nodes of G A , G B through the Hungarian algorithm, minimize the semantic difference between the nodes, and then for the matched node pairs, calculate the relationship difference of the corresponding edges;

[0059]

[0060] In the formula: e A , e B are the edges in the semantic topology graph G A corresponding to text A and the edges in the semantic topology graph G B corresponding to text B; Type(eA )、Type(e B ) is the semantic relationship type of edge e A and e B ;

[0061] Similarity(Type(e A ), Type(e B )) is the similarity between Type(e A ) and Type(e B );

[0062] Finally, calculate ReconstructionCost(G A , G B ):

[0063]

[0064] Where: (vA, vB) is the element composed of the mutually matching nodes v A and v B in G A and v B ;

[0065] Matchednodes is the optimal matching set between the nodes of G A and G B found by the Hungarian algorithm;

[0066] NodeCost(v A , v B ) is the semantic difference cost between nodes v A and v B , calculated based on the method of WordNet synset distance; (e A , e B ) is the element composed of the edges respectively related to the mutually matching nodes v A and v B in G A and G B ; MatchedEdges is the set of edges corresponding to the matching node pairs; EdgeCost(e A , e B ) is the relationship difference cost between edges e A and e B , obtained based on the edge mapping cost calculation method;

[0067] Through the above logical and formula limitations, it is ensured that the filtering unit stably completes the cleaning of the patient diagnosis and treatment cases, making the patient diagnosis and treatment cases retained in the cloud database more valuable for system users.

[0068] A selection module for selecting patient diagnosis and treatment cases in a cloud database and using the selected diagnosis and treatment cases as the cases to be simulated for diagnosis and treatment;

[0069] During the operation stage of the selection module, the system user obtains permission to access the cloud database in the selection module, first selects the type of patient diagnosis and treatment cases based on the differentiated storage range in the cloud database, and then further selects patient diagnosis and treatment cases in the selected differentiated storage range;

[0070] A locking unit is internally set in the selection module. The locking unit is used to lock the diagnostic logic and medication details included in the content of each patient diagnosis and treatment case in the differentiated storage range during the stage when the system user selects patient diagnosis and treatment cases in the differentiated storage range, making the locked part of the patient diagnosis and treatment case content unreadable;

[0071] Among them, after the system user selects a patient diagnosis and treatment case, the readable part of the patient diagnosis and treatment case is read synchronously, that is, the description of the condition is read, and then the editing module is used to complete the editing of the diagnostic logic and medication details. After the diagnostic logic and medication details are edited, the locking module reopens the reading permission for the locked diagnostic logic and medication details;

[0072] An editing module for obtaining the cases to be simulated for diagnosis and treatment selected by the operation of the selection module and manually editing the diagnostic logic and medication details based on the reading of the cases to be simulated for diagnosis and treatment;

[0073] An evaluation module for receiving the diagnostic logic and medication details edited in the editing module and evaluating the simulation diagnosis and treatment score of the source user of the edited diagnostic logic and medication details based on the comparison between the received diagnostic logic and medication details and the diagnostic logic and medication details included in the cases to be simulated for diagnosis and treatment;

[0074] A recording unit is set at the lower level of the evaluation module. The recording unit is used to record the historical simulation diagnosis and treatment scores of users;

[0075] The logic for evaluating the simulation diagnosis and treatment score in the evaluation module through the comparison of diagnostic logic and medication details is as follows:

[0076] Taking the diagnostic logic and medication details as the comparison targets respectively, the results are obtained based on the similarity calculation logic of the patient diagnosis and treatment cases, and the simulation diagnosis and treatment score is comprehensively evaluated and output;

[0077] F = [SIMM logic (a, b) × λ + SIMM take (a, b) × (1 - λ)] × ε;

[0078] In the formula: F is the simulation diagnosis and treatment score; SIMM logic (a, b) is the similarity between the diagnostic logic edited by the user and the diagnostic logic in the cases to be simulated for diagnosis and treatment; SIMM take(a, b) is the similarity between the medication details edited by the user and the medication details in the case to be simulated for diagnosis and treatment; λ is the weight coefficient; ε is the conversion factor for the simulated diagnosis and treatment score;

[0079] Among them, the conversion factor ε for the simulated diagnosis and treatment score is ε > 0, and the weight coefficient λ satisfies λ < 1 - λ;

[0080] Through the above logical formula, the simulated diagnosis and treatment score of the user is evaluated to provide support for the further operation of the modules in this embodiment of the system.

[0081] During the operation stage of the recording unit, each simulated diagnosis and treatment score from the user is distinguished and recorded based on the simulated diagnosis and treatment score. When the classification unit obtains the historical evaluation result of the user's simulated diagnosis and treatment score, it obtains it from the recording unit. During the operation stage of the classification unit, the passing judgment threshold for the simulated diagnosis and treatment result is set synchronously. Then, the latest simulated diagnosis and treatment score of each user is obtained from the distinguished and recorded user simulated diagnosis and treatment scores, and compared with the passing judgment threshold for the simulated diagnosis and treatment result. Based on the comparison result with the passing judgment threshold for the simulated diagnosis and treatment result, the simulated diagnosis and treatment scores are distinguished, that is, the users are classified;

[0082] The classification unit is used to obtain the historical evaluation result of the user's simulated diagnosis and treatment score in the evaluation module and classify the user based on the historical evaluation result;

[0083] The lower level of the cloud database is interconnected with a filtering unit and a management unit through wireless network interaction. The cloud database is interconnected with a selection module through wireless network interaction. Inside the selection module, it is interconnected with a locking unit through wireless network. The selection module is interconnected with an editing module and an evaluation module through wireless network. The lower level of the evaluation module is interconnected with a recording unit through wireless network. The evaluation module is interconnected with a classification unit through wireless network. The classification unit is interconnected with the recording unit through wireless network. In this embodiment, the cloud database runs to upload patient diagnosis and treatment cases, stores the patient diagnosis and treatment cases, the filtering unit synchronously traverses the patient diagnosis and treatment cases uploaded in the cloud database, cleans the low-value cases in the uploaded patient diagnosis and treatment cases, the management unit receives the cleaned patient diagnosis and treatment cases in real time, differentiates the patient diagnosis and treatment cases based on the departments to which the patient diagnosis and treatment cases belong, the cloud database further differentiates and stores the patient diagnosis and treatment cases based on the differentiation results of the patient diagnosis and treatment cases. The selection module runs later to select patient diagnosis and treatment cases in the cloud database, and takes the selected diagnosis and treatment cases as the to-be-simulated diagnosis and treatment cases. The locking unit synchronously locks the diagnosis logic and medication details included in the content of each patient diagnosis and treatment case in the differentiated storage range during the stage when the system user selects patient diagnosis and treatment cases in the differentiated storage range, making the locked part of the patient diagnosis and treatment case content unreadable. Then, the editing module obtains the to-be-simulated diagnosis and treatment cases selected by the operation of the selection module, manually edits the diagnosis logic and medication details based on the reading of the to-be-simulated diagnosis and treatment cases, and receives the edited diagnosis logic and medication details in the editing module through the evaluation module. Based on the comparison between the received diagnosis logic and medication details and the diagnosis logic and medication details included in the to-be-simulated diagnosis and treatment cases, it evaluates the simulation diagnosis score of the source user of the edited diagnosis logic and medication details. The recording unit synchronously records the user's historical simulation diagnosis score. Finally, the classification unit obtains the historical evaluation results of the user's simulation diagnosis score in the evaluation module, and classifies the user based on the historical evaluation results.

[0084] Through the system in the above embodiment, a way for intern doctors to accumulate diagnosis and treatment experience online is provided, so that intern doctors can accumulate diagnosis and treatment experience faster and put it into the work of seeing patients.

[0085] In summary, during the operation of the system in the above embodiments, by collecting numerous standard cases where patients have successfully recovered, through screening, classification, and storage, a large amount of real and high-quality learning resources are provided for medical teaching. Users can select from these cases for simulated diagnosis and treatment, manually edit the diagnosis logic and medication details, and deepen their understanding and application of medical knowledge in practice. The system will compare the edited content with real cases and give a score for the simulated diagnosis and treatment to help users identify their own deficiencies. At the same time, the system will also record and classify the historical scores to facilitate understanding of the learners' progress trajectory. According to the set passing judgment threshold, the learning achievements of learners can be distinguished. This not only improves the pertinence of medical teaching, enabling teachers to teach students in accordance with their aptitudes, but also effectively exercises the diagnosis and treatment thinking and practical abilities of learners, greatly improving the quality and efficiency of medical teaching and contributing to the cultivation of medical talents.

[0086] The above embodiments are only used to illustrate the technical solutions of the present invention, not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching, characterized in that, Including: A cloud database for uploading patient diagnosis and treatment cases and storing the patient diagnosis and treatment cases. A selection module for selecting patient diagnosis and treatment cases in the cloud database and using the selected diagnosis and treatment cases as the to-be-simulated diagnosis and treatment cases. An editing module for obtaining the to-be-simulated diagnosis and treatment cases selected by the operation of the selection module, and manually editing the diagnosis logic and medication details based on the to-be-simulated diagnosis and treatment cases read. An evaluation module for receiving the diagnosis logic and medication details edited in the editing module, comparing the received diagnosis logic and medication details with the diagnosis logic and medication details included in the to-be-simulated diagnosis and treatment cases, and evaluating the simulation diagnosis and treatment scores of the users who edited the diagnosis logic and medication details. A classification unit for obtaining the historical evaluation results of the users' simulation diagnosis and treatment scores in the evaluation module and classifying the users based on the historical evaluation results.

2. The intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching according to claim 1, wherein The patient diagnosis and treatment cases uploaded in the cloud database are from the authorized end-users of any system, and the patient diagnosis and treatment cases are all standard cases where the patients have recovered successfully. A filtering unit and a management unit are set under the cloud database. The filtering unit is used to traverse the patient diagnosis and treatment cases uploaded in the cloud database and clean the low-value cases among the uploaded patient diagnosis and treatment cases. The management unit is used to receive the patient diagnosis and treatment cases that have been cleaned, distinguish the patient diagnosis and treatment cases based on the departments to which the patient diagnosis and treatment cases belong, and the cloud database further stores the patient diagnosis and treatment cases separately based on the distinction results of the patient diagnosis and treatment cases. Among them, the format of the patient diagnosis and treatment cases is text and images, and the content of the patient diagnosis and treatment cases includes: condition description, diagnosis logic, and medication details.

3. The intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching according to claim 2, characterized in that, The determination of low-value cases for patient diagnosis and treatment cases in the filtering unit follows: Where: SIMM(A,B) is the similarity between patient diagnosis and treatment case A and patient diagnosis and treatment case B. ReconstructionCost(G A ,G B ) is the semantic reconstruction cost; ω is the weight; is the average similarity of the image data contained in patient diagnosis and treatment case A and patient diagnosis and treatment case B respectively; Among them, the similarity between each patient diagnosis and treatment case is calculated based on the above formula. The system end-user customizes the pick-up threshold for the cleaning target of the patient diagnosis and treatment cases, and deletes any one of the two patient diagnosis and treatment cases that is greater than or equal to the pick-up threshold for the cleaning target of the patient diagnosis and treatment cases as the deletion target to complete the cleaning of the patient diagnosis and treatment cases, ω∈(0,1).

4. An intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching according to claim 3, characterized in that, When any one of the two patient diagnosis and treatment cases for which the similarity is calculated does not contain image data, the calculation formula of SIMM(A,B) is deformed into That is, the similarity calculation process only considers semantic similarity.

5. The intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching according to claim 3, characterized in that The value-taking logic of the semantic reconstruction cost ReconstructionCost(G A , G B ) is as follows: Let the semantic topology graphs corresponding to patient diagnosis and treatment case A and patient diagnosis and treatment case B be G A (V A , E A ), G B (V B , E B ), where V is the set of semantic nodes and E is the set of semantic relation edges. G A (V A , E A ), G B (V B , E B ) are briefly referred to as G A , G B ; The construction logic of the semantic topology graph is as follows: Using dependency syntactic analysis and semantic role annotation, the patient diagnosis and treatment case is split into semantic atomic units as nodes V, and directed edges E are constructed according to the semantic relation types, and the relation weights are marked for each edge; The optimal matching between the nodes of G A , G B is found through the Hungarian algorithm to minimize the semantic differences between the nodes, and then for the matched node pairs, the relation differences of the corresponding edges are calculated; Where: e A and e B are the edges in the semantic topology graph G A corresponding to text A, and the edges in the semantic topology graph G B corresponding to text B; Type(e A ), Type(e B ) are the semantic relationship types of the edges e A and e B ; Similarity(Type(e A ),Type(e B )) is the similarity between Type(e A ) and Type(e B ); Final calculation of ReconstructionCost(G A ,G B ): where: (v A , v B ) is the element composed of the mutually matching nodes v A , G B in G A , v B ; Matched nodes are the optimal matching set between G A and G B nodes found by the Hungarian algorithm; NodeCost(v A ,v B ) is the semantic difference cost between nodes v A and v B , calculated based on the WordNet synset distance method; (e A ,e B ) is the element composed of the edges respectively related to the mutually matching nodes v A and v B in G A and G B ; MatchedEdges is the set of edges corresponding to the matching node pairs; EdgeCost(e A ,e B ) is the relationship difference cost between edges e A and e B , obtained based on the edge mapping cost calculation method.

6. The intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching according to claim 1, wherein During the operation stage of the selection module, the system user obtains the permission to access the cloud database in the selection module, first selects the type of patient diagnosis and treatment cases based on the distinguished storage range in the cloud database, and then further selects the patient diagnosis and treatment cases in the selected distinguished storage range.

7. An intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching according to claim 1, characterized in that A locking unit is provided inside the selection module. The locking unit is used to lock the diagnosis logic and medication details included in the content of each patient diagnosis and treatment case in the distinguished storage range during the stage when the system user selects the patient diagnosis and treatment cases in the distinguished storage range, so that the locked part of the patient diagnosis and treatment case content is unreadable. Among them, after the system user selects a patient diagnosis and treatment case, the readable part of the patient diagnosis and treatment case is read synchronously, that is, the content of the condition description is read, and then the editing module is used to complete the editing of the diagnosis logic and medication details. After the diagnosis logic and medication details are edited, the locking module re-opens the reading permissions of the locked diagnosis logic and medication details.

8. An intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching according to claim 1, characterized in that, A recording unit is set at a lower level of the evaluation module, and the recording unit is used to record the user's historical simulation diagnosis and treatment scores; In the evaluation module, the logic for evaluating the simulation diagnosis and treatment score by comparing the diagnosis logic and medication details is as follows: Using the diagnosis logic and medication details as comparison targets respectively, the results are obtained based on the similarity calculation logic of the patient diagnosis and treatment case, and the simulation diagnosis and treatment score is output through comprehensive evaluation; F = [SIMM logic (a, b) × λ + SIMM take (a, b) × (1 - λ)] × ε; where: F is the simulated diagnosis and treatment score; SIMM logic (a, b) is the similarity between the diagnosis logic edited by the user and the diagnosis logic in the case to be simulated for diagnosis and treatment; SIMM take (a, b) is the similarity between the medication details edited by the user and the medication details in the case to be simulated for diagnosis and treatment; λ is the weight coefficient; ε is the conversion factor for the simulated diagnosis and treatment score; Among them, the simulation diagnosis and treatment score conversion factor ε > 0, and the weight coefficient λ satisfies λ < 1 - λ.

9. An intelligent medical record analysis and simulated diagnosis and treatment system for medical teaching according to claim 8, characterized in that, During the operation stage of the recording unit, each simulation diagnosis and treatment score is distinguished and recorded based on the source user of the simulation diagnosis and treatment score. When the classification unit obtains the historical evaluation result of the user's simulation diagnosis and treatment score, it obtains it from the recording unit. During the operation stage of the classification unit, the passing judgment threshold of the simulation diagnosis and treatment result is set synchronously, and then the latest simulation diagnosis and treatment score of each user is obtained from the distinguished and recorded user simulation diagnosis and treatment scores, and compared with the passing judgment threshold of the simulation diagnosis and treatment result. Based on the comparison result with the passing judgment threshold of the simulation diagnosis and treatment result, the simulation diagnosis and treatment scores are distinguished, that is, the users are classified.

10. An intelligent medical record analysis and simulation diagnosis and treatment system for medical teaching according to claim 1, characterized in that, A filtering unit and a management unit are connected to the lower level of the cloud database through wireless network interaction. The cloud database is connected to a selection module through wireless network interaction. The selection module is internally connected to a locking unit through wireless network. The selection module is connected to an editing module and an evaluation module through wireless network interaction. A recording unit is connected to the lower level of the evaluation module through wireless network interaction. The evaluation module is connected to a classification unit through wireless network. The classification unit is connected to the recording unit through wireless network.

Images