Medical education and research integrated system based on common bottom layer

Abstract

The application relates to the technical field of medical system, in particular to a medical teaching and research integrated system based on a common bottom layer, aiming to break the information silos among medical treatment, teaching and research, and realize full-process closed-loop linkage and dynamic knowledge transformation. The system comprises a diagnosis and treatment decision subsystem, a research cohort management subsystem and a teaching subsystem, integrates data through a trinity data engine, realizes information sharing and collaborative work. The diagnosis and treatment decision subsystem is based on expert thinking chain and provides diagnosis and treatment support for critical diseases. The research cohort management subsystem is responsible for screening cases, sorting data, implementing quality control and statistical analysis, and adjusting the expert thinking chain according to the research results. The teaching subsystem uses a standardized case presentation method for teaching. Through the feedback closed loop of clinical discovery, research verification, knowledge update and teaching feedback, the application promotes the rapid dissemination and application of medical knowledge, and improves the quality of medical services and the level of medical education.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical systems, and particularly relates to a medical education and research integrated system based on a common bottom layer. BACKGROUND

[0002] With the rapid development of medical technology and the continuous advancement of medical informatization construction, medical institutions have accumulated a large amount of data in clinical diagnosis and treatment, medical education and scientific research innovation. However, at present, these data are scattered in different systems, lacking effective integration and sharing mechanism, resulting in obvious disconnection between medical, teaching and research work, which seriously affects the improvement of medical service quality and medical education and research level. SUMMARY

[0003] Therefore, the embodiments of the present application are committed to providing a medical education and research integrated system based on a common bottom layer to integrate and share data between medical, teaching and research work.

[0004] The present application provides a medical education and research integrated system based on a common bottom layer, comprising:

[0005] A diagnosis and treatment decision subsystem is configured to provide diagnosis and treatment decision support for preset diseases and syndromes based on expert thinking chains, specifically including: for each case, based on the collected information, judging each core index related to the disease condition, guiding the user to carry out information sorting and disease analysis, automatically generating disease analysis and treatment suggestions, recording treatment choices and effect feedback, automatically reminding whether to include the case in the research queue when the case meets the preset research queue conditions, prompting the treatment node, and visualizing the disease development process.

[0006] The expert thinking chain is a set of medical processes and medical schemes constructed based on historical cases and expert suggestions; the core index is an index necessary for judging the disease condition under the existing diagnosis and treatment scheme.

[0007] The scientific research cohort management subsystem is used to obtain medical information, to include cases meeting preset requirements in the medical information into a scientific research cohort, to sort each case in the scientific research cohort based on a preset structured data set, and to perform desensitization processing on the information of the cases; wherein the structured data set is a general data set or an individualized data set obtained by adjusting based on user input information; to assist scientific researchers in generating treatment suggestions and sending the treatment suggestions to corresponding clinicians to obtain actual treatment choices and effect feedback of the clinicians; to exclude cases not meeting requirements in the scientific research cohort; to implement quality control and statistical analysis on data of the scientific research cohort, to access an AI tool to query related background information, to sort a statistical result framework, and to generate a data result analysis report; to adjust an expert thinking chain based on actual scientific research results of the scientific research cohort; and the medical information includes patient medical records, examination and test results, and treatment plans in a clinical diagnosis and treatment process;

[0008] The teaching subsystem is used to set a preset case presentation mode based on the expert thinking chain and a preset cognitive architecture to guide a user to form a disease analysis thinking habit and to perform structured thinking, to generate teaching materials based on a material set and a case set obtained by the diagnosis and treatment decision subsystem and the scientific research cohort management subsystem after standardization and expert-level quality control, and to generate a skill training system and an ability improvement training system in combination with a simulation person;

[0009] The standardization and expert-level quality control include: being converted into the preset case presentation mode and being audited by a preset expert;

[0010] The simulation person is a device used to simulate a patient.

[0011] In some embodiments, the diagnosis and treatment decision subsystem includes:

[0012] A core index preset module is used to preset core indexes required for judging diseases and syndromes based on expert experience, and to display the core indexes in a table form on a system interface for medical staff;

[0013] The core indexes are indexes required for judging a disease under an existing diagnosis and treatment scheme;

[0014] An information collection and analysis module is used to support manual input or automatic collection of core information related to disease judgment, to automatically generate disease analysis after completion of the collection, and to give a treatment suggestion;

[0015] The core information is information required for judging a disease under an existing diagnosis and treatment scheme;

[0016] A record and feedback module is used to record and feedback treatment schemes and treatment effects of medical staff, to prompt a treatment node, and to visually display a disease development process;

[0017] A scientific research cohort reminding module is configured to remind whether to include a case into a scientific research cohort when the case meets preset scientific research cohort conditions.

[0018] In some embodiments, the scientific research cohort management subsystem comprises:

[0019] A platform design and dataset editing module is configured to provide a general structured dataset or a general scientific research process platform, or to generate an individualized structured dataset or an individualized scientific research process platform according to user configurations;

[0020] A treatment suggestion and feedback module is configured to generate and display a treatment suggestion for a case for scientific research to a scientific researcher, to obtain a scientific research treatment suggestion determined by the scientific researcher, to send the scientific research treatment suggestion to a clinician, to obtain feedback of the clinician on whether to adopt the scientific research treatment suggestion and record clinical behaviors and corresponding treatment effects, and to exclude the current case if the current case does not meet current scientific research requirements.

[0021] A data quality control and analysis report generation module is configured to implement online quality control and statistical analysis on scientific research cohort data, access AI tools to query relevant background information, sort out a statistical result framework, and generate a data result analysis report.

[0022] An expert thinking chain adjustment module is configured to regularly evaluate and optimize diagnosis and treatment processes and schemes in an expert thinking chain according to scientific research results and expert opinions.

[0023] In some embodiments, the teaching subsystem comprises:

[0024] A standardized presentation setting module is configured to set a preset case presentation mode based on an expert thinking chain and a preset cognitive architecture.

[0025] A material set and case set management module is configured to store a preset number of material sets and case sets generated by a diagnosis and treatment decision subsystem and a scientific research cohort management subsystem, and to use the material sets and case sets as teaching materials after standardization and expert-level quality control.

[0026] A training system generation module is configured to generate a skill training system and a capability improvement training system in combination with a simulation person.

[0027] An AI automatic identification and quality control platform generation module is configured to label a preset number of medical images and text materials, and train an AI model to realize automatic identification and quality control.

[0028] In some embodiments, a trinity data engine is further included.

[0029] The trinity data engine uses optical character recognition technology and natural language processing technology to recognize and analyze medical record texts, extract key information such as disease names and symptom descriptions, store expert thinking chains in the form of data structures in the system database for calling by the diagnosis and treatment decision subsystem and the research cohort management subsystem, and is also used to bind clinical cases with expert thinking chains and research conclusions to generate diagnosis and treatment guidelines and teaching cases, provide real-time patient condition data and expert recommendations for the diagnosis and treatment decision subsystem, provide structured research data sets for the research cohort management subsystem, and provide teaching cases and materials for the teaching subsystem.

[0030] In some embodiments, the trinity data engine is also used to update the diagnosis and treatment processes and schemes in the expert thinking chain in real time according to the treatment effect data in the clinical feedback and the statistical analysis data in the research results, according to a preset update rule and algorithm, and adjust the diagnosis and treatment guidelines and teaching cases in combination with expert opinions.

[0031] In some embodiments, the diagnosis and treatment decision subsystem further comprises a behavior guidance and compliance monitoring module.

[0032] The behavior guidance and compliance monitoring module is used to compare the deviation of the doctor's behavior from the expert thinking chain, and if the deviation is greater than a preset value, a warning is triggered, and the use of data and the ethical approval status are recorded in real time through the system's built-in permission management and approval process monitoring module.

[0033] In some embodiments, the case presentation mode comprises:

[0034] The time axis is integrated into the patient's full-cycle diagnosis and treatment track, combining the 7 key steps of patient admission / department admission, initial examination and diagnosis after information collection, treatment plan development, treatment process observation, disease condition change evaluation and feedback, treatment plan adjustment, and patient discharge / department discharge, sorting out the disease development process, stating the content at each time point based on a preset format, showing the trend of changes in core indicators in the case over time, and reflecting the disease trend and treatment direction.

[0035] In some embodiments, the research cohort management subsystem is also used to automatically analyze medical data using data mining and machine learning algorithms, identify abnormal data patterns and potential correlation relationships therein, propose research hypotheses based on the characteristics and distribution of abnormal data, and generate corresponding research project proposals in combination with medical knowledge and expert experience.

[0036] The application provides a medical education and research integration system based on a common bottom layer, a diagnosis and treatment decision subsystem, which is used to provide diagnosis and treatment decision support for preset diseases and syndromes based on expert thinking chains, specifically including: for each case, based on the collected information, judging each index related to the disease condition, guiding the user to carry out information sorting and disease analysis, automatically generating disease analysis and treatment suggestions, recording treatment choices and effect feedback, and when the case meets the preset research queue conditions, automatically reminding whether to include the case in the research queue; wherein the expert thinking chain is a set of medical processes and medical schemes constructed based on historical cases and expert suggestions; the research queue management subsystem is used to obtain medical information, include cases that meet the preset requirements in the medical information in the research queue; based on the preset structured data set, sort each case in the research queue, and desensitize the information of the case; wherein the structured data set is a general data set or an individualized data set obtained by adjusting the user input information; assist researchers in generating treatment suggestions and send them to corresponding clinicians to obtain actual treatment choices and effect feedback of the clinicians; exclude cases that do not meet the requirements of the research queue; implement quality control and statistical analysis on the data of the research queue, and access AI tools to query related background information, sort statistical result framework, and generate data result analysis report; based on the actual research results of the research queue, adjust the expert thinking chain; the medical information includes patient medical records, examination results and treatment schemes in the clinical diagnosis and treatment process, and a teaching subsystem for setting a preset case presentation mode based on the expert thinking chain and the preset cognitive architecture, guiding the user to analyze the disease condition, thinking habit and structured thinking; based on the material set and case set obtained by the diagnosis and treatment decision subsystem and the research queue management subsystem, the teaching material is generated after standardization and expert-level quality control, and the skill training system and the ability improvement training system are generated by combining with the simulation person. By setting in this way, the application provides a medical education and research integration system based on a common bottom layer, which integrates diagnosis and treatment decision, research queue management and teaching functions, realizes the deep integration and collaborative work of medical treatment, teaching and research, and has the following remarkable beneficial effects: improving diagnosis and treatment efficiency and quality: the diagnosis and treatment decision subsystem can quickly judge the index and generate treatment suggestions based on the expert thinking chain, standardize the clinical behavior, reduce the risk of misdiagnosis and mistreatment, and improve the accuracy and timeliness of the diagnosis and treatment of severe diseases. Promote scientific research innovation and transformation: the research queue management subsystem realizes automatic screening, data sorting and quality control analysis of cases, improves the efficiency and quality of scientific research. The research results are fed back to the diagnosis and treatment decision system to optimize the expert thinking chain and accelerate the transformation and application of research achievements to clinical practice. Optimize medical education resources and effects: the teaching subsystem uses standardized cases and expert-level quality control materials, combines with simulation person technology, constructs an interactive teaching environment, cultivates the clinical thinking and practical ability of medical students, improves the quality and effect of medical education, and realizes the dynamic update of teaching content.Breaking information island, realizing data sharing and cooperation: the system is based on common underlying architecture, integrating medical, teaching and scientific research data, breaking the information barriers between various fields, realizing seamless sharing and interaction of data, providing strong data support for medical education and research integration, and promoting the coordinated development of medical services and medical education and research. BRIEF DESCRIPTION OF DRAWINGS

[0037] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description of embodiments of the present application, taken in conjunction with the accompanying drawings. The drawings provided in the present application are used to provide further understanding of the embodiments of the present application, and constitute a part of the specification, and are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0038] Figure 1 is a structure diagram of a medical education and research integration system based on common underlying architecture provided by an embodiment of the present application. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0040] Figure 1 is a structure diagram of a medical education and research integration system based on common underlying architecture provided by an embodiment of the present application. As shown in Figure 1 The medical education and research integration system based on common underlying architecture comprises:

[0041] A diagnosis and treatment decision subsystem 1 is configured to provide diagnosis and treatment decision support for preset diseases and syndromes (the preset diseases and syndromes are generally common critical diseases and syndromes recognized by the medical community) based on expert thinking chains. Specifically, for each case, based on the collected information, the system judges various indicators related to the disease condition, guides the user to sort and analyze the information and the disease condition, automatically generates disease condition analysis and treatment suggestions, records treatment options and effect feedback, and automatically reminds whether to include the case in the research cohort when the case meets the preset research cohort conditions.

[0042] Among them, the expert thinking chain is a collection of medical processes and medical schemes based on historical cases and expert suggestions;

[0043] The scientific research queue management subsystem 2 is used to obtain medical information, to include cases meeting preset requirements in the medical information into a scientific research queue, to sort each case in the scientific research queue based on a preset structured data set, and to perform desensitization processing on the information of the case; wherein the structured data set is a general data set or an individualized data set obtained by adjusting based on user input information; to assist scientific researchers in generating treatment suggestions and sending them to corresponding clinicians to obtain actual treatment choices and effect feedback of the clinicians; to exclude cases not meeting requirements in the scientific research queue; to implement quality control and statistical analysis on data of the scientific research queue, and to access AI tools to query related background information, to sort a statistical result framework, and to generate a data result analysis report; to adjust the expert thinking chain based on actual scientific research results of the scientific research queue; and the medical information includes patient medical records, examination results and treatment plans in the clinical diagnosis and treatment process.

[0044] The teaching subsystem 3 is used to set a preset case presentation mode based on the expert thinking chain and a preset cognitive architecture, to guide users to form disease analysis thinking habits and to perform structured thinking, to generate teaching materials after standardization and expert-level quality control based on the material set and the case set obtained by the diagnosis and treatment decision subsystem and the scientific research queue management subsystem, and to combine the teaching materials with a simulation person to generate a skill training system and an ability improvement training system.

[0045] The standardization and expert-level quality control include: being converted into the preset case presentation mode, and being audited by a preset expert;

[0046] The preset case presentation mode is a standardized and structured case display form designed based on the expert thinking chain (a medical process and a scheme set constructed from historical cases and expert suggestions) and the preset cognitive architecture.

[0047] The specific presentation logic is: integrating a whole-cycle diagnosis and treatment track of a patient according to a time axis, covering 7 key steps of patient admission / department admission, preliminary examination and diagnosis after collecting information, treatment plan formulation, treatment process observation, disease condition change evaluation and feedback, treatment plan adjustment, and patient discharge / department discharge.

[0048] The presentation requirement is: structuring, standardizing and stating content at each time point according to a preset format, focusing on displaying trends of core indicators (such as vital signs, laboratory examination results, etc.) with time, and intuitively reflecting disease condition changes and treatment directions.

[0049] The auditing by the preset expert: the auditing subject is a system preset expert team composed of medical experts with rich clinical experience, teaching experience and research background.

[0050] The review content includes whether the presentation of the case after conversion meets the preset standards, whether the core information in the case (such as disease analysis, treatment recommendations, data logic, etc.) is accurate, complete, and consistent with the latest medical guidelines and expert consensus.

[0051] The purpose of the review is to ensure the scientificity, standardization, and practicality of the teaching materials, and to provide high-quality case bases for medical teaching.

[0052] Simulators are professional medical devices used to simulate patients, with highly simulated physiological and pathological state simulation functions. In the teaching subsystem, simulators are combined with standardized teaching materials to generate skill training systems and ability enhancement training systems. Learners can perform operation training on simulators that closely resemble real clinical scenarios, such as physical examination, puncture, first aid treatment, etc., to enhance their clinical practice ability.

[0053] Simulators can be combined with AI, VR / MR, etc. to achieve interactive and virtual reality training effects, enhancing the immersion and practicality of teaching.

[0054] The diagnosis and treatment decision subsystem is a core component of the medical-teaching-research integrated system based on a common bottom layer, aiming to provide diagnosis and treatment decision support for diseases and syndromes. This system integrates expert experience and the latest medical knowledge to build expert thinking chains and provide comprehensive diagnosis and treatment assistance for medical personnel. Its main functions include: core indicator judgment: for each case, based on the collected information, judge each core indicator related to the disease. Information sorting and disease analysis: guide the user to sort information and analyze the disease. Automatically generate disease analysis and treatment recommendations: automatically generate disease analysis and treatment recommendations according to preset rules and expert thinking chains. Record treatment options and effect feedback: record the treatment plan of medical personnel and the treatment effect of patients. Research cohort reminder: when the case meets the preset research cohort conditions, automatically remind whether to include the case in the research cohort.

[0055] Specifically, the core indicators usually include but are not limited to vital signs (such as body temperature, blood pressure, heart rate, etc.), laboratory test results (such as blood routine, biochemical indicators, etc.), imaging examination results (such as X-ray, CT, MRI, etc.), and other clinical manifestations closely related to the disease. For example, the oxygen metabolism indicators (such as Lac, ScvO2) and perfusion indicators (such as mottling, PI) of shock patients. These indicators are determined based on expert experience by collecting and analyzing a large number of historical case data and expert clinical experience, and are strictly reviewed and verified by an expert team to ensure their scientificity and practicality. The system will display these core indicators in table form in a prominent position on the interface, and the table will not only list the indicator names, but also may include normal ranges, abnormal prompts, etc. information, to facilitate medical personnel to quickly view, refer to and judge the disease.

[0056] The diagnosis and treatment decision subsystem 1 comprises:

[0057] A core index preset module 11 is configured to preset core indexes required for judging the conditions of different diseases and syndromes according to expert experience, and display the core indexes in the form of a table on the system interface for medical personnel.

[0058] The core index refers to a medical parameter that must be focused on when diagnosing and treating a specific disease or syndrome. These indexes generally include, but are not limited to, vital signs (such as body temperature, blood pressure, heart rate, etc.), laboratory test results (such as blood routine, biochemical indexes, etc.), imaging examination results (such as X-ray, CT, MRI, etc.), and other clinical manifestations closely related to the condition. By collecting and analyzing a large number of historical case data and expert clinical experience, the core indexes for each disease or syndrome are determined. These indexes are strictly reviewed and verified by an expert team to ensure their scientificity and practicality. The system displays these core indexes in the form of a table in a prominent position on the interface, making it easy for medical personnel to quickly view and refer to them. The table not only lists the index names, but also may include normal ranges, abnormal prompts, and other information to help medical personnel better understand and apply these indexes.

[0059] By clearly defining the core indexes, medical personnel can more accurately identify and assess the condition, reducing the risk of misdiagnosis and missed diagnosis. Ensuring that medical personnel follow standardized procedures during diagnosis improves the efficiency and quality of diagnosis and treatment. By solidifying expert experience in the form of core indexes in the system, it is easy for young doctors to learn and draw on, promoting the overall level of the medical team.

[0060] An information collection and analysis module 12 is configured to support manual input or automatic collection of core information related to condition judgment, automatically generate condition analysis after collection, and provide treatment recommendations according to preset criteria.

[0061] The core information is information that is clearly defined as necessary for judging the condition under existing diagnosis and treatment plans, and is a key basis for diagnosis and treatment decisions and condition analysis, with the following specific characteristics:

[0062] It is directly related to the "core index" (key medical parameters for judging the condition, such as vital signs, laboratory test results, etc.), and is various information used to accurately judge the condition around the core index.

[0063] The core information includes, but is not limited to, the patient's symptoms, medical history (such as past medical history, allergy history), detailed description of examination and test results, treatment response, etc. For example, for a patient with shock, in addition to core indexes such as oxygen metabolism indexes (Lac, ScvO2) and perfusion indexes (mottling, PI), information such as the shock trigger, duration, and response to initial treatment are also core information.

[0064] Core information acquisition method: Core information can be acquired through manual input by medical staff (such as entering patient symptoms, medical history) or automatic collection by the system (such as synchronizing laboratory reports, imaging results from the hospital information system).

[0065] Core role of core information: The key input of the system's "information collection and analysis module", the module automatically generates disease analysis and treatment recommendations based on core information combined with expert thinking chains, providing a basis for clinical diagnosis and treatment decisions.

[0066] The system supports multiple information collection methods, including manual input and automatic collection. Manual input allows medical staff to enter patient information such as medical history, symptoms, and test results according to actual conditions; automatic collection automatically acquires real-time data of patients through integration with hospital information systems (such as electronic medical record systems, laboratory information systems, imaging systems, etc.), reducing the workload of medical staff. The system analyzes and processes the collected information based on pre-set rules and expert thinking chains. First, evaluate the core indicators to determine whether they are within the normal range; then, combine the patient's medical history, symptoms, and other related factors to generate a disease analysis report. Based on the disease analysis results, the system further provides treatment recommendations, including drug treatment, surgical treatment, and rehabilitation treatment. The pre-set rules are a series of logical judgment rules based on expert experience and medical guidelines, used to guide the system on how to analyze the disease and generate treatment recommendations. Expert thinking chains are a more complex decision-making model that simulates the thinking process of experts in the diagnosis and treatment process, considering the interaction and influence of various factors.

[0067] Automatic information collection and analysis functions greatly reduce the time for medical staff to manually process data, improving diagnosis and treatment efficiency. Disease analysis and treatment recommendations based on expert thinking chains can provide more scientific and comprehensive diagnosis and treatment plans, helping to improve treatment effectiveness. It provides strong decision support for medical staff, especially when facing complex diseases or difficult cases, helping them make more accurate judgments quickly.

[0068] The record and feedback module 13 is used to record and feedback the treatment plan and treatment effect to the medical staff; the prompt treatment node can visually display the disease development process;

[0069] The system records the treatment plan developed by medical staff for patients in detail, including drug names, doses, administration routes, treatment cycles, and other information, as well as specific surgical treatment plans and steps. Medical staff can feed back the treatment effect of patients to the system, including the improvement of patient symptoms, changes in laboratory test results, and comparisons of imaging test results. The system can also automatically collect patient follow-up data to objectively evaluate the treatment effect.

[0070] By recording and analyzing the treatment effect, medical staff can timely evaluate the effectiveness of the treatment plan and adjust the treatment strategy to improve the treatment effect. The recording and feedback function of the system provides data support for medical quality evaluation, which helps medical institutions continuously optimize medical service processes and improve overall medical service quality.

[0071] The system prompts the medical staff of the key treatment nodes to ensure the standardization and timeliness of the treatment process. According to the characteristics of the disease and the treatment plan, the key treatment nodes such as drug adjustment time, review time, and operation time are preset. The system automatically sends a reminder to the medical staff when approaching the preset treatment node to ensure the timely implementation of the treatment measures. According to the treatment effect and the change of the patient's condition, the treatment node is dynamically adjusted to ensure the flexibility and adaptability of the treatment plan.

[0072] The system displays the development process of the disease through visual means to help medical staff and patients better understand the changes in the disease. The patient's condition changes are displayed in the form of a timeline, including symptom onset time, examination time, treatment time, etc., so that medical staff can clearly understand the development context of the disease. Combined with laboratory test results, imaging test results, and other multi-dimensional data, the system displays the disease change trend through charts and graphs, such as blood pressure change curve and tumor volume change graph. It provides data comparison and analysis at different time points to help medical staff evaluate the treatment effect and timely adjust the treatment plan.

[0073] The scientific research cohort reminder module 14 is used to automatically remind whether to include the case in the scientific research cohort when there is a case that meets the preset scientific research cohort conditions.

[0074] The scientific research cohort conditions are preset according to the research purpose and requirements of the scientific research project, usually including patient's disease type, disease severity, treatment method, treatment effect, and other multi-dimensional conditions. These conditions are stored in the system in the form of rules for screening cases that meet the requirements. The system monitors patient case information in real time and automatically sends a reminder to medical staff when a case that meets the scientific research cohort conditions is found. The reminder can be sent through system interface pop-up windows, SMS, email, and other ways to ensure that medical staff can receive the reminder information in time. After receiving the reminder, medical staff can decide whether to include the case in the scientific research cohort according to the actual situation. The cases included in the scientific research cohort will be the research object of the scientific research project and provide data support for scientific research.

[0075] The automated case screening and reminder function greatly improves the efficiency of research cohort construction, reduces the workload of researchers manually screening cases, and accelerates the progress of research projects. Cases screened by the system meet the pre-set research cohort conditions, ensuring the quality and consistency of research data and improving the reliability of research results. The research cohort reminder module not only provides convenience for research work, but also brings feedback and guidance to clinical practice, promoting the coordinated development of clinical and research, and promoting medical progress. The diagnosis and treatment decision subsystem provides comprehensive diagnosis and treatment decision support for medical staff through core indicator presetting, information collection and analysis, recording and feedback, and research cohort reminders. It not only improves the efficiency and quality of diagnosis and treatment, promotes the close integration of clinical and research, but also provides strong support for the improvement of medical service quality and the optimization of medical education. The research cohort management subsystem is an important part of the common underlying medical education and research integrated system, aiming to efficiently manage and utilize medical information to support the smooth progress of research projects. The system achieves comprehensive management of research cohorts through the following functional modules: case inclusion and sorting: obtaining medical information, screening cases that meet the pre-set requirements for inclusion in the research cohort, and sorting and de-identifying based on structured data sets. Treatment recommendations and feedback: assisting researchers in generating treatment recommendations, sending them to clinicians, and obtaining actual treatment choices and effectiveness feedback. Case exclusion: excluding cases that do not meet research requirements. Data quality control and analysis: implementing quality control and statistical analysis, accessing AI tools to query relevant background information, and generating data result analysis reports. Expert thought chain adjustment: adjusting the expert thought chain based on research results and expert opinions.

[0076] The research cohort management subsystem 2 includes a platform design and data set editing module 21 for providing a general structured data set or a general research process platform, or generating a personalized structured data set or a personalized research process platform according to user configuration; detailed as follows:

[0077] General structured data set: a pre-set standardized data set covering the basic data structure of common diseases and syndromes, suitable for most research projects.

[0078] Personalized data set generation: adjusting and customizing the general data set according to user input specific research requirements, such as studying the effect of a specific disease subtype or a specific treatment regimen, to generate a personalized structured data set.

[0079] General research process platform: provides a standardized research process framework, including case screening, data collection, quality control analysis, etc., to ensure the standardization and consistency of research projects.

[0080] Personalized research process platform: allows users to customize and optimize the general research process according to the special requirements of specific research projects, improving research efficiency and quality.

[0081] In this way, the individual needs of different scientific research projects for data and processes are met, and the flexibility and adaptability of scientific research are improved. By providing standardized data sets and process frameworks, the normative and scientific nature of scientific research projects is ensured. Strong tool support is provided to researchers to promote scientific innovation and breakthroughs.

[0082] The treatment recommendation and feedback module 22 is used to generate and display treatment recommendations for scientific research for the case to the researchers; obtains the scientific research treatment recommendations determined by the researchers, sends the scientific research treatment recommendations to the clinicians; obtains the feedback of the clinicians on whether to adopt and record the clinical behavior and corresponding treatment effect; if the current case does not meet the current scientific research needs, the current case is excluded;

[0083] Based on the preset expert thinking chain and scientific research goals, targeted treatment recommendations are generated for each case included in the scientific research cohort. These recommendations may include drug treatment plans, surgical plans, rehabilitation plans, etc. Researchers make decisions based on treatment recommendations and send the determined treatment recommendations to clinicians. Clinicians choose whether to adopt and record the actual treatment choices and treatment effect feedback. If a case is found not to meet the scientific research needs during the scientific research process, such as a treatment plan that does not meet the research goals or severe data loss, the system will automatically prompt the researchers to exclude the case, ensuring the quality and consistency of the scientific research cohort.

[0084] Specific scientific research cohort intelligent management includes:

[0085] Dynamic enrollment: automatically screen patients who meet the conditions according to the scientific research goals (such as "the value of lung ultrasound in analyzing and predicting common causes of acute respiratory failure patients");

[0086] The system presets the scientific research cohort enrollment conditions, and when the patient conditions meet the clinical diagnosis and treatment, the system automatically prompts whether to include in the corresponding scientific research cohort, and reminds and screens the corresponding exclusion criteria; during the scientific research cohort process, if the exclusion conditions are met, real-time prompts are given on whether to exclude the cohort, and finally the platform user determines whether to enroll in the cohort or exclude, and records the user behavior and the involved exclusion conditions.

[0087] Data enrichment: correlate patient genomic data, follow-up records, image features, and generate scientific research-ready data sets;

[0088] Manually enter / automatically capture patient history descriptions, laboratory tests, imaging tests, and other indicators during treatment,

[0089] Risk monitoring: real-time detection of data loss or abnormalities (such as follow-up loss), triggering automatic reminders.

[0090] The system sets mandatory and optional items, and if the mandatory items are not filled in, an automatic reminder will be sent. After filling in all the items, the next step can be entered. At the same time, if the filling value exceeds the preset range during the actual filling process, an automatic reminder will be triggered to check the normal range.

[0091] Conversion update: update the expert thinking chain according to the results of the scientific research team, and further generate new scientific research team suggestions to verify their superiority.

[0092] In this way, the workload of scientific researchers and clinicians is reduced through automated processes, and the efficiency of scientific research is improved. Real-time acquisition and recording of treatment effect feedback ensure the accuracy and integrity of scientific research data. Strengthen the communication and cooperation between scientific researchers and clinicians, and promote the clinical application of scientific research results.

[0093] Data quality control and analysis report generation module 23, used for online quality control and statistical analysis of scientific research team data, access to AI tools to query related background information, comb statistical result framework, generate data result analysis report;

[0094] Real-time monitoring and quality control of data in the scientific research team to ensure the accuracy, integrity and consistency of the data. Quality control includes data format checking, outlier detection, data integrity verification, etc. Statistical methods are used to analyze scientific data, including descriptive statistics, correlation analysis, hypothesis testing, etc. to reveal the rules and trends in the data. Access to natural language processing (NLP), machine learning and other AI tools to query related background information such as literature materials, guideline recommendations, etc. to provide more comprehensive references for data analysis. According to the statistical analysis results, generate structured data result analysis report, including charts, text descriptions, etc. to facilitate scientific researchers to quickly understand and apply.

[0095] In this way, the accuracy and reliability of data analysis are improved through AI tools and professional statistical methods. Automatic generation of analysis reports reduces the time spent by scientific researchers in manually organizing and analyzing data, accelerating the scientific research process. Present the analysis results in the form of a report to facilitate internal sharing and communication among the scientific research team, promoting the dissemination and application of knowledge.

[0096] Expert thinking chain adjustment module 24, used to evaluate and optimize the diagnosis and treatment process and scheme in the expert thinking chain according to the scientific research results and expert opinions. Regularly collect and analyze the data results of the scientific research team to evaluate the effectiveness and limitations of the existing expert thinking chain. Organize expert teams to discuss and analyze the scientific research results and make optimization suggestions. Adjust and optimize the diagnosis and treatment process and scheme in the expert thinking chain according to the scientific research results and expert opinions to ensure that it always reflects the latest medical knowledge and clinical practice.

[0097] In this way, the scientificity and accuracy of diagnosis and treatment decisions are improved by continuously optimizing the expert thought chain. The conversion of research results into clinical practice is facilitated, promoting the updating and dissemination of medical knowledge. The application of research results in clinical diagnosis and treatment is accelerated, and the efficiency of research conversion is improved.

[0098] The research cohort management subsystem provides comprehensive support for research projects through platform design and data set editing, treatment recommendations and feedback, data quality control and analysis report generation, and expert thought chain adjustment. It not only improves research efficiency and quality, but also promotes the close integration of clinical and research, and provides strong support for the updating and dissemination of medical knowledge.

[0099] The teaching subsystem 3 is an important part of the medical education and research integrated system based on the common bottom layer, aiming to provide high-quality medical education resources by integrating expert experience and clinical practice, and cultivating the clinical thinking and practical ability of medical personnel. The system realizes comprehensive support for medical education through the following functional modules: Specifically, based on the expert thought chain and the preset cognitive architecture, the standardized and structured case presentation mode is set to guide users to form the habit of disease analysis thinking and structured thinking. The material set and case set obtained by the diagnosis and treatment decision subsystem and the research cohort management subsystem are standardized and expert-level quality controlled to generate teaching materials. Combined with the mannequin, skill training system and ability improvement training system are generated. A preset number of medical images and text materials are labeled to train AI models to realize automatic recognition and quality control, improving the quality and accuracy of teaching materials.

[0100] The standardized presentation setting module 31 sets the preset case presentation mode based on the expert thought chain and the preset cognitive architecture;

[0101] The expert thought chain is a collection of medical processes and medical solutions based on historical cases and expert recommendations. It simulates the thinking process of experts in the diagnosis and treatment process, taking into account the interaction and influence of various factors. The preset cognitive architecture is designed according to the goals and requirements of medical education, aiming to guide learners to form a systematic clinical thinking mode. It includes the steps of disease analysis, identification of core indicators, and methods of diagnostic reasoning.

[0102] The system presents case information in a standardized and structured form (i.e., the preset case presentation mode), such as through flowcharts, tables, timelines, etc., so that learners can clearly understand and master the logic and methods of disease analysis.

[0103] Through standardized case presentation, the learner is guided to form a systematic clinical thinking mode, and the ability of disease analysis and diagnosis reasoning is improved. The structured presentation method enables the learner to quickly grasp the key information and improve the learning efficiency. The standardized presentation method facilitates the dissemination and sharing of knowledge, and promotes the standardization and standardization of medical education.

[0104] Expert thinking chain driven behavior guidance is as follows:

[0105] Real-time diagnosis and treatment guidance: the system presets the diagnosis and treatment framework according to different clinical scenarios and clinical problems, including information to be collected, information interpretation results and diagnosis and treatment suggestions, guiding the user to collect and sort information, information interpretation thinking path, and giving corresponding treatment decision suggestions, especially key treatment.

[0106] For example, for a shock patient, the system automatically suggests immediately adding vasoactive drugs to maintain vital signs, emergency blood gas analysis to determine whether there are factors that may cause the condition to worsen or even cardiac arrest in a short period of time, and simultaneous rapid examination by methods such as critical ultrasound to determine the cause of shock for further treatment; after stabilizing the vital signs, further detailed assessment and optimization of subsequent treatment are performed, such as completing the assessment and treatment of the patient's oxygen metabolism, perfusion, flow, and anterior and posterior obstruction, body response, and primary disease. For example, if an acute brain injury patient has unequal pupils, it indicates brain herniation, and the first thing to do is to establish a rapid artificial airway, perform emergency osmotic therapy, and simultaneously call a specialist for surgical treatment; if it is simply intracranial hypertension, it needs to be evaluated and treated from the aspects of systemic circulation and perfusion, local brain circulation and perfusion, brain oxygen metabolism, and brain electrical function.

[0107] For example, a shock patient needs to show oxygen metabolism indicators such as Lac, ScvO2, perfusion indicators such as mottling, PI, flow such as CO, GAP, forward obstruction such as nasopharyngeal artery spectrum, RI, backward regurgitant resistance such as hepatic vein spectrum, portal vein spectrum, body response indicators such as inflammation indicators, coagulation indicators, ultrasound-driven heart and lung conditions, cardiac stress globoid, and possible primary disease types such as infection, fever, and pyogenic sputum, CT consolidation often indicates the presence of pulmonary infection.

[0108] For example, if there is oxygen metabolism imbalance, low perfusion, insufficient flow, low volume, high power, high drive, and decreased hemoglobin, it is recommended to supplement the volume and blood transfusion, and to find the bleeding site for hemostasis.

[0109] Material set and case set management module 32: stores the preset number of material sets and case sets generated by the diagnosis and treatment decision subsystem and the research team management subsystem, and after standardization and expert-level quality control, they are used as teaching materials;

[0110] Training system generation module 33, combined with a mannequin, generates a skill training system and an ability enhancement training system;

[0111] Simulated human is a highly simulated medical training device that can simulate the physiological and pathological states of real patients. By combining with simulated human, learners can train clinical skills in a nearly real environment. The system combines standardized cases and teaching materials with simulated human technology to generate an interactive skill training system. Learners can perform actual operations on simulated human according to case prompts, such as physical examination, puncture operation, surgery simulation, etc. In addition to skill training, the system also provides ability enhancement training, including clinical decision-making ability, team cooperation ability, communication ability, etc.

[0112] By simulating real clinical environment, the clinical practice ability and operation skills of learners are improved. Interactive training method enhances the participation and experience of learners, improves learning interest and effect. In addition to clinical skills, the comprehensive ability of learners is also focused on, such as clinical decision-making ability, team cooperation ability, etc.

[0113] AI automatic identification and quality control platform generation module 34: label a preset number of medical images and text materials, train AI model to realize automatic identification and quality control, improve the quality and accuracy of teaching materials.

[0114] Label a preset number of medical images and text materials, including disease name, symptom description, examination result, etc. Key information. Use these labeled data to train AI model to automatically identify and extract key information. The trained AI model can automatically identify and extract key information from medical images and text, improving the efficiency and accuracy of data processing. AI model can also control the quality of teaching materials, check the integrity and accuracy of data, and correct errors in time.

[0115] Automatic identification and processing function greatly improves the efficiency of data processing, reduces the workload of manual labeling and auditing. Through the quality control of AI model, the accuracy and reliability of teaching materials are ensured, and the teaching quality is improved. Intelligent technology is introduced into medical education, promoting the modernization and intelligent development of medical education.

[0116] Teaching subsystem provides comprehensive support for medical education through standardized presentation settings, material set and case set management, skill training system generation, and AI automatic identification and quality control function modules. It not only enriches medical education resources, improves teaching quality and learning efficiency, but also cultivates the clinical thinking and practical ability of medical staff, promotes the standardization and intelligent development of medical education.

[0117] In some embodiments, based on the common bottom medical education research integrated system, it also includes a trinity data engine 4;

[0118] The trinity data engine uses optical character recognition (OCR) technology and natural language processing (NLP) technology to recognize and analyze medical record texts, extract key information such as disease names and symptom descriptions, and store expert thinking chains in the form of data structures in the system database for the diagnosis and treatment decision subsystem and the research cohort management subsystem to call. It is also used to bind clinical cases with expert thinking chains and research conclusions to generate diagnosis and treatment guidelines and teaching cases, provide real-time patient condition data and expert recommendations for the diagnosis and treatment decision subsystem, provide structured research data sets for the research cohort management subsystem, and provide teaching cases and materials for the teaching subsystem.

[0119] The trinity data engine is a core technology module of the medical-teaching-research integrated system based on a common bottom layer, aiming to integrate optical character recognition (OCR) technology and natural language processing (NLP) technology to achieve efficient recognition and deep analysis of medical record texts. The engine not only extracts key information, but also stores data structures and expert thinking chains to provide strong data support and intelligent decision-making basis for various subsystems of the system.

[0120] Further,

[0121] Optical character recognition (OCR) technology recognizes medical record texts and converts the text content in the image into editable and searchable electronic text. Detailed description as follows: Denoising, binarization, and tilt correction are performed on medical record images to improve recognition accuracy. Advanced OCR algorithms are used to recognize the text content in the image, supporting multiple fonts and formats. Spelling check and context analysis are performed on the recognition results to further improve the accuracy and readability of the text.

[0122] Natural language processing (NLP) technology: Deep analysis of the electronic text recognized by OCR to extract key information such as disease names and symptom descriptions. Detailed description as follows: Preprocessing operations such as word segmentation, part-of-speech tagging, and named entity recognition are performed on the electronic text. Machine learning and deep learning algorithms are used to recognize and extract key information such as disease names, symptom descriptions, and test results from the text. Context analysis and semantic understanding are used to ensure the accuracy of the extracted information.

[0123] Data structuring and storage store the extracted key information in the form of data structures in the system database for various subsystems to call. Detailed description as follows: Design a reasonable data structure, such as a relational database table structure, to store the extracted key information. Store structured data in the system database to ensure data security and accessibility. Establish data indexes to improve data query and retrieval efficiency.

[0124] Expert thinking chain storage and calling refers to storing expert thinking chains in the form of data structures in the system database for the diagnosis and treatment decision subsystem and the research team management subsystem to call. The details are as follows: Convert the diagnosis and treatment experience and thinking process of experts into executable rules and models. Store the expert thinking chain model in the database to ensure its accessibility and scalability. The diagnosis and treatment decision subsystem and the research team management subsystem can call the expert thinking chain in real time to obtain diagnosis and treatment suggestions and research guidance.

[0125] Bind clinical cases with expert thinking chains and research conclusions to generate diagnosis and treatment guidelines and teaching cases. Specifically, associate and bind the key information in the clinical case with the expert thinking chain and the research conclusion. According to the bound data, generate personalized diagnosis and treatment guidelines to support clinical decision-making. Convert the bound data into teaching cases to provide rich teaching materials for the teaching subsystem.

[0126] Specifically, provide real-time patient condition data and expert suggestions for the diagnosis and treatment decision subsystem. Push the patient's real-time condition data to the diagnosis and treatment decision subsystem to ensure the timeliness and accuracy of the information. According to the expert thinking chain, provide scientific diagnosis and treatment suggestions for the diagnosis and treatment decision subsystem to assist medical personnel in making more accurate decisions.

[0127] Provide structured research data sets for the research team management subsystem. Organize the extracted and structured data into research data sets to support the smooth progress of research projects. Perform quality control on the research data sets to ensure the accuracy and completeness of the data.

[0128] Provide teaching cases and materials for the teaching subsystem. Convert clinical cases into teaching cases to enrich the teaching resources of the teaching subsystem. Provide high-quality medical images and text materials to support skill training and capability improvement of the teaching subsystem.

[0129] In this way, data processing efficiency can be improved, and through OCR and NLP technology, efficient recognition and deep analysis of medical records text can be achieved, reducing the workload of manual processing. Through data structuring and expert thinking chain storage, the accuracy and reliability of the data are ensured, and the overall performance of the system is improved. Promote the integration of medical education and research, provide strong data support for diagnosis and treatment decision-making, research management, and teaching training, and promote the deep integration of medical treatment, teaching, and research. Promote intelligent medical treatment, introduce advanced technology, promote the intelligent development of the medical system, and improve the quality and efficiency of medical services.

[0130] The integrated data engine combines OCR and NLP technologies to achieve efficient recognition and in-depth analysis of medical record texts, providing powerful data support and intelligent decision-making basis for the integrated medical, educational, and research system. It not only improves the efficiency and quality of data processing but also promotes the deep integration of medical care, teaching, and research, driving the intelligent development of the medical system.

[0131] Furthermore, the three-in-one data engine is also used to update the diagnosis and treatment process and plan in the expert thought chain in real time based on the treatment effect data in clinical feedback and the statistical analysis data in scientific research results, according to the preset update rules and algorithms. At the same time, it adjusts the diagnosis and treatment guidelines and teaching cases in combination with expert opinions.

[0132] The integrated data engine not only handles data collection, processing, and distribution, but also possesses dynamic update capabilities. This function ensures that the system can adjust and optimize the diagnostic and treatment processes and plans within the expert thought process in real time based on the latest clinical feedback and research results, while also optimizing and adjusting treatment guidelines and teaching cases in conjunction with expert opinions. This dynamic update mechanism helps the system always maintain the latest medical knowledge and best practices, improving the accuracy of diagnostic and treatment decisions and the timeliness of teaching content.

[0133] The three-in-one data engine first automatically collects treatment effect data from clinical feedback (such as improvement in patient symptoms, changes in test results, etc.) and statistical analysis results from the research cohort management subsystem (such as research conclusions, data analysis reports, etc.). It then analyzes this data using preset algorithms to identify content that deviates from existing treatment guidelines, teaching cases, or expert thought processes (such as a treatment plan not being as effective as expected, or research conclusions conflicting with existing guidelines), and triggers an alert function.

[0134] When the engine identifies content requiring adjustment (especially information that significantly deviates from the existing architecture), it compiles the relevant data and anomalies and submits them to a pre-designated expert team. This expert team, composed of medical experts with extensive clinical, teaching, and research experience, conducts focused discussions on these issues, combining their professional experience with the latest medical knowledge to assess whether adjustments are necessary and, if so, in which direction.

[0135] Regarding treatment guidelines: Based on the discussion results, the expert team determines whether to update the treatment procedures, medication recommendations, and indicator judgment criteria in the guidelines. If adjustments are needed, the engine transforms the expert consensus into structured rules and embeds them into the treatment guidelines to ensure that the guidelines are consistent with the latest clinical practices and research findings.

[0136] For teaching cases: The expert team reviews the presentation format, core knowledge points, and clinical reasoning guidance logic of existing teaching cases, and proposes modification suggestions based on new treatment guidelines and research conclusions (such as adding new treatment plans, updating the focus of case analysis, etc.). Based on expert opinions, the engine updates the standardized presentation content, interactive modules, and assessment points of the teaching cases to ensure the scientific validity and timeliness of the teaching materials.

[0137] The revised treatment guidelines and teaching cases will be synchronized to all subsystems of the system (treatment decision-making subsystem and teaching subsystem), and further feedback will be collected through clinical application and teaching practice. The engine continuously monitors this feedback data, and if new deviations are found, the above process will be restarted to form a closed loop of "data identification - expert evaluation - optimization and adjustment - practice verification" to ensure that the treatment guidelines and teaching cases are always consistent with expert consensus and the latest medical advancements.

[0138] The system acquires patient treatment outcome data from the treatment decision-making subsystem and clinical information system, including symptom improvement, changes in laboratory test results, and comparisons of imaging results. A combination of automated collection and manual input ensures data comprehensiveness and accuracy. It also acquires statistical analysis results from the research cohort management subsystem, including research conclusions and data analysis reports. A combination of automated extraction and manual upload by researchers ensures data timeliness and reliability. Based on preset update rules and algorithms, the system updates treatment processes and protocols in the expert thought chain in real time. A series of preset logical judgment rules are used to assess the impact of clinical feedback and research results on existing treatment processes and protocols. Machine learning and data mining algorithms are used to analyze clinical feedback and research results data and generate update suggestions. The system automatically adjusts treatment processes and protocols in the expert thought chain based on the evaluation results of the update rules and algorithms, ensuring they always reflect the latest medical knowledge and best practices. Treatment guidelines and teaching cases are optimized and adjusted based on expert opinions. Senior medical experts review and evaluate the updated treatment processes and protocols. Expert opinions and suggestions on the updates are collected through expert meetings and online discussions. Based on expert opinions, the treatment guidelines and teaching cases are further optimized and adjusted to ensure their scientific validity and practicality. The system automatically collects treatment effect data from clinical feedback and statistical analysis data from research results. According to preset update rules and algorithms, the collected data is evaluated, and update suggestions are generated. These suggestions are submitted to the expert team for review and evaluation, and expert opinions are collected. Based on expert opinions, the treatment processes and plans in the expert thought process are optimized and adjusted. The updated treatment guidelines and teaching cases are published to the system for use by various subsystems. The system continuously monitors clinical feedback and research results to ensure the continuous and effective operation of the update mechanism.

[0139] By updating expert thought processes in real time, the system ensures that diagnostic and treatment decisions are always based on the latest medical knowledge and clinical practice, thereby improving the accuracy and effectiveness of diagnosis and treatment. The system optimizes and adjusts treatment guidelines and teaching cases based on expert opinions, ensuring the timely updating and effective dissemination of medical knowledge. Optimized teaching cases better reflect actual clinical needs, improving the quality and effectiveness of medical education. The dynamic update mechanism enables the system to quickly adapt to changes and developments in the medical field, maintaining its advanced nature and practicality.

[0140] The dynamic update function of the three-in-one data engine collects clinical feedback and research results data, combines them with preset update rules and algorithms, and incorporates expert opinions to update and optimize the diagnosis and treatment processes and plans in the expert thought process in real time. This function not only improves the accuracy of diagnosis and treatment decisions and the timeliness of teaching content, but also promotes the updating and dissemination of medical knowledge, enhances the system's adaptability and flexibility, and provides strong support for the continuous optimization of the integrated medical, educational, and research system.

[0141] Treatment decision-making subsystem 1 also includes a behavior guidance and compliance monitoring module;

[0142] The behavior guidance and compliance monitoring module is used to compare the deviation between the doctor's behavior and the expert's thought process. If the deviation is greater than a preset value, an early warning is triggered. Through the system's built-in permission management and approval process monitoring module, the usage of data and the ethical approval status are recorded in real time to ensure the compliance and legality of data use.

[0143] The behavior guidance and compliance monitoring module is a crucial component of the treatment decision-making subsystem. It aims to ensure that healthcare professionals' treatment behaviors align with expert thought processes and pre-defined treatment guidelines, while simultaneously guaranteeing the compliance and legality of data usage. Through real-time monitoring and early warning mechanisms, this module helps healthcare professionals standardize their treatment practices, reduce medical risks, and ensure both medical quality and data security.

[0144] The behavior guidance function is as follows: it compares the deviation between the doctor's behavior and the expert's thought process. If the deviation is greater than the preset value, an alert is triggered.

[0145] The expert thought chain is a set of medical processes and plans built upon historical cases and expert recommendations, simulating the thought processes of experts during diagnosis and treatment. The system monitors the behavior of medical staff in real time during the diagnosis and treatment process, including diagnostic decisions, treatment plan selection, and requests for examinations and tests. Through preset evaluation rules and algorithms, it calculates the deviation between the medical staff's behavior and the expert thought chain. The deviation assessment considers multiple factors, such as diagnostic accuracy, the rationality of the treatment plan, and the necessity of examinations and tests. When the deviation exceeds a preset threshold, the system automatically triggers an alert, reminding medical staff to reassess and adjust their diagnostic and treatment behaviors.

[0146] The compliance monitoring function refers to the real-time recording of data usage and ethical approval status through the system's built-in permission management and approval process monitoring modules, ensuring the compliance and legality of data use.

[0147] The system has a built-in access control module that assigns different data access and operation permissions to users with different roles (such as doctors, nurses, and researchers) to ensure data security and confidentiality. The system monitors the data usage approval process to ensure that all data use undergoes necessary ethical approval and compliance review. The approval process includes data usage application, approval opinion recording, and approval result feedback. The system records data usage in real time, including data access time, access content, and operation records, facilitating subsequent auditing and traceability. The system conducts regular compliance checks to ensure that data use complies with relevant laws, regulations, and ethical requirements. Checks include the legality of data sources, the rationality of data usage purposes, and the compliance of data processing procedures.

[0148] Through real-time monitoring and early warning mechanisms, we help medical staff standardize their diagnostic and treatment practices, reduce the risk of misdiagnosis and mistreatment, and improve the quality of care. Through access control and approval process monitoring, we ensure the compliance and legality of data use, protecting patient privacy and data security. Combined with expert guidance, we help medical staff make more scientific and rational treatment decisions, improving the quality of medical services. Through early warning mechanisms and compliance monitoring, we promptly identify and correct potential medical risks, reducing medical disputes and legal risks.

[0149] The behavior guidance and compliance monitoring module compares the deviation between doctors' behavior and experts' thought processes, triggering real-time alerts. Combined with the system's built-in access control and approval process monitoring, it ensures the compliance and legality of data use. This module not only standardizes the diagnostic and treatment behaviors of medical staff and improves the quality of care, but also guarantees the security and legality of data, providing strong support for the efficient operation and risk management of the medical system.

[0150] The case presentation method includes: integrating the patient's entire treatment trajectory along a timeline, combining the seven key steps of patient admission / department admission, preliminary examination and diagnosis after information collection, treatment plan formulation, observation of treatment process, assessment and feedback of changes in condition, adjustment of treatment plan, and patient discharge / department discharge, sorting out the disease development process, presenting the content of each time point in a structured and standardized manner, showing the trend of changes in core indicators in important cases over time, reflecting the trend of disease changes and treatment direction.

[0151] The case presentation method is an important component of the teaching subsystem, designed to help learners better understand and master the entire course of patient care through structured and standardized presentations. This method integrates key steps from admission to discharge along a timeline, demonstrating the progression of the disease and treatment outcomes, thus helping learners develop systematic clinical thinking and analytical skills.

[0152] This system integrates the entire patient's treatment journey along a timeline, encompassing seven key steps: admission / departmental admission, initial examination and diagnosis after information collection, treatment plan development, observation during treatment, assessment and feedback on changes in the patient's condition, adjustment of the treatment plan, and discharge / departmental discharge. The timeline presents the specific details and time points of each key step, enabling learners to clearly understand the chronological order and critical nodes in the patient's entire treatment process. The content at each time point is presented in a structured and standardized manner (i.e., presented based on a pre-defined format). The content at each time point is categorized into different types, such as symptom description, examination results, treatment measures, and changes in the patient's condition. A unified format and template are used to present the content, ensuring accuracy and consistency. For example, visualization tools such as tables, flowcharts, and timelines are used to display key information. Core indicators in important cases, such as vital signs, laboratory test results, and imaging results, are highlighted, and their trends over time are shown. The changes in core indicators in important cases over time are demonstrated, reflecting the trend of disease progression and treatment direction. The charts (such as line graphs and bar charts) are used to display the changing trends of core indicators, helping learners to intuitively understand the changes in their condition. The treatment effectiveness is evaluated by combining the time points of treatment measures, demonstrating the impact of these measures on changes in the condition. Based on changes in the condition and treatment effectiveness, the treatment plan is dynamically adjusted, and the time points and specific details of these adjustments are recorded on a timeline.

[0153] Through structured and standardized presentations, learners can quickly understand and master the entire patient care process, improving learning efficiency. Learners are guided to develop a systematic clinical thinking model, learning to analyze and assess patients' conditions holistically, thus improving diagnostic and treatment abilities. Rich visualization tools and dynamic presentation methods make the teaching content more vivid and intuitive, enhancing teaching effectiveness. Standardized case presentation facilitates knowledge dissemination and sharing, promoting the standardization and normalization of medical education.

[0154] The case presentation method integrates the patient's entire treatment trajectory along a timeline, combining structured and standardized presentations to demonstrate the disease progression and treatment outcomes. This approach not only helps learners better understand and master the patient's treatment process but also cultivates their clinical thinking and analytical skills, thereby improving the quality and effectiveness of medical education.

[0155] In some embodiments, the research cohort management subsystem is also used to automatically analyze medical data using data mining and machine learning algorithms, identify abnormal data patterns and potential correlations, and, based on the characteristics and distribution patterns of abnormal data, combine medical knowledge and expert experience to propose research hypotheses with research value and generate corresponding research project proposals.

[0156] Automated analysis of medical data is performed to identify anomalous data patterns and potential correlations. Medical data in research cohorts is cleaned, standardized, and feature-extracted to ensure data quality and usability. Various data mining and machine learning algorithms are employed, such as cluster analysis, association rule mining, decision trees, random forests, and neural networks, with appropriate algorithms selected based on specific research needs. Historical data is used to train the selected algorithms, optimizing model parameters and improving model accuracy and generalization ability. Anomalous patterns in the data are identified through the model, such as abnormally high test results or unusual treatment responses. Potential correlations in the data are discovered, such as associations between certain symptoms and specific diseases, or comparisons of the effects of different treatment regimens. Based on the characteristics and distribution patterns of anomalous data, combined with medical knowledge and expert experience, valuable research hypotheses are proposed. Feature analysis is performed on identified anomalous data to extract key features and patterns. Anomalous data is interpreted by combining medical literature, guidelines, and expert experience to find possible medical explanations. Based on feature analysis and medical knowledge, valuable research hypotheses are proposed. For example, if a drug is found to have significant efficacy in a specific patient group, a hypothesis for further research into the drug's mechanism of action is proposed. Generate corresponding research project proposals. Design standardized research project proposal templates, including sections on research background, research objectives, research methods, expected results, and timeline. Automatically populate the relevant content in the proposal template based on the generated research hypotheses, ensuring the completeness and scientific rigor of the proposal. Submit the generated proposals to an expert team for review and evaluation, and revise and improve them based on expert feedback. Submit the revised research project proposals to research management departments or funding agencies to apply for project approval and funding.

[0157] Through automated data mining and machine learning algorithms, it quickly identifies anomalous data patterns and potential correlations, reducing the workload of researchers manually analyzing data and improving research efficiency. Combining medical knowledge and expert experience, it proposes research hypotheses with research value, helping to discover new directions in medical research and groundbreaking results. The generated research project proposals undergo expert review and evaluation to ensure the scientific validity and feasibility of research projects, improving research quality. It provides researchers with powerful tools to stimulate innovative thinking and promote the development of medical research.

[0158] The research cohort management subsystem employs data mining and machine learning algorithms to automatically analyze medical data, identify anomalous data patterns and potential correlations, and, combined with medical knowledge and expert experience, propose research hypotheses with research value and generate corresponding research project proposals. This function not only improves the efficiency and quality of scientific research but also provides new directions and ideas for medical research, promoting scientific innovation and medical progress.

[0159] Step 3: Generation and Feedback of Teaching Case Library

[0160] Case construction: The entire diagnosis and treatment process (data + decision chain + outcome) of typical cases is encapsulated into an interactive teaching module;

[0161] The large number of real clinical cases generated in steps 2 and 3, based on standardized and structured expression of expert thinking framework, are desensitized. The resulting large amount of data, decision chains, and treatment results are then packaged into interactive teaching modules according to different needs.

[0162] The generated large amount of traditional data indicators and text content are combined with medical history information and structured for teaching purposes, which can be used as teaching content for clinical diagnosis. The modular organization of different systems can be used as teaching content for evaluating different systems.

[0163] The generated large amount of image data, such as ultrasound images and other pictures and videos, is automatically identified and labeled by the AI ​​in Patent 2, and undergoes expert-level quality control. Individual pictures and videos can be used for image quality control and judgment teaching content, while grouped image data combined with clinical information can be used as phenotypic teaching content, thereby realizing visual accumulation in the learning process.

[0164] The generated data, decision chains, treatment outcome feedback, and other information are combined in a time sequence to form a complete disease diagnosis and treatment process. The structured presentation is based on a framework pre-set by expert experience and can be used as teaching content for classic cases, simulations, and interactive exercises.

[0165] By combining the clinical analysis, visual accumulation, and case-based teaching content generated above with the mannequin training system, a training system for basic clinical skills, image acquisition skills, image interpretation skills, clinical decision-making skills, and specialized competencies can be formed. Furthermore, AI, VR / MR, and other technologies can be integrated to achieve interactive and virtual reality effects.

[0166] Each of the aforementioned teaching modules will be marked with its key points, highlights, and teaching value based on existing expert prior standards. The corresponding learning modules will be linked to relevant professional knowledge, guidelines, and literature through AI big data models to achieve learning goals from specific points to a broader perspective.

[0167] (2) Dynamic updates: When the research conclusions of the case are published or the guidelines are updated, the teaching case will automatically synchronize with the latest knowledge;

[0168] The underlying framework of the aforementioned learning module was set up after joint review by the expert team based on expert experience and the latest understanding. The system regularly and automatically searches for and updates relevant content and synchronizes relevant knowledge points. In addition, if content that is found to be significantly different from the existing architecture is found, a reminder function is set up and the information is compiled. The expert team discusses this and adjusts or maintains the architecture based on the discussion results.

[0169] (3) Ability assessment: The trainees’ ability to apply expert thinking chain is assessed through simulated diagnosis and treatment (such as virtual patient consultation).

[0170] The system includes an assessment and feedback module. It sets up post-lesson quizzes for the key points of each learning module and collects user feedback. The system regularly collects, organizes, and analyzes the assessment results and feedback, and pushes them to administrators. It also provides suggestions based on AI, and the administrators organize meetings to decide whether to adopt them.

[0171] A dedicated assessment module is set up to conduct batch assessments of the learned content, including basic clinical skills, image acquisition skills, image interpretation skills, clinical decision-making skills, and competence in the management of specific diseases / syndromes. The assessment can be conducted through direct answering, comparative answering, and simulated diagnosis and treatment to evaluate the trainees' mastery of the above content. The system regularly collects, organizes, and analyzes the assessment results and evaluation feedback, and pushes them to the management personnel. It also provides suggestions in conjunction with AI, and the management organizes a meeting to discuss whether to adopt them. The system is continuously updated and iterated.

[0172] The following is an example of generating a research hypothesis:

[0173] Based on abnormal patterns in clinical data (such as an increased rate of adverse reactions to a certain drug), research hypotheses are automatically proposed; and by linking historical studies, similar research design and analysis methods are recommended.

[0174] Based on this application, the following is an example of a closed-loop feedback mechanism for the collaboration between teaching, clinical practice, and scientific research:

[0175] 1. Clinical findings: The efficacy of a certain antihypertensive drug decreased in obese patients;

[0176] Specifically, the clinical feedback data collected through the treatment decision-making subsystem revealed that a certain antihypertensive drug was less effective than expected in obese patients. When healthcare professionals observed poor blood pressure control in obese patients while using this drug, they recorded this phenomenon and reported it to the system.

[0177] 2. Scientific validation: An automated cohort of obese hypertensive patients was constructed, and the correlation between treatment efficacy and BMI was analyzed;

[0178] An automated cohort of obese hypertensive patients was constructed to analyze the correlation between treatment efficacy and BMI. The research cohort management subsystem automatically selected an obese hypertensive patient cohort, ensuring that patients within the cohort had similar BMI indices. Treatment efficacy analysis was performed on patients within the cohort, statistically analyzing the responses to antihypertensive drugs across different BMI ranges and assessing the correlation between efficacy and BMI. Statistical analysis and machine learning algorithms were used to verify the hypothesis that the efficacy of antihypertensive drugs decreases in obese patients.

[0179] 3. Knowledge update: Transform the conclusions into new expert rules ("BMI≥30→Prefer drug X");

[0180] Based on research validation results and expert opinions, new expert rules are generated, clearly defining drug X as the preferred choice for patients with a BMI ≥ 30. These new expert rules are stored in the system database for use by the treatment decision-making subsystem. Treatment guidelines are updated to incorporate these new expert rules, ensuring that clinical decisions align with the latest medical knowledge.

[0181] 4. Teaching feedback: Generate interactive teaching cases on "Choosing a Blood Pressure Lowering Plan for Obese Patients" and push them to doctors.

[0182] The teaching subsystem generates interactive teaching cases based on new expert rules and research findings, showcasing the process of selecting antihypertensive treatment plans for obese patients. These cases include basic patient information, a description of the patient's condition, the basis for treatment plan selection, and evaluation of treatment effectiveness. Through interactive Q&A and simulations, learners master correct diagnostic and treatment methods. The generated teaching cases are then pushed to doctors for study and reference, improving their clinical decision-making abilities and treatment levels.

[0183] The closed-loop feedback mechanism linking teaching, clinical practice, and scientific research achieves rapid knowledge updates and dissemination through four key steps: clinical discovery, scientific verification, knowledge updating, and teaching feedback. This improves the quality of medical services and the level of medical education. This mechanism not only promotes close integration of research and clinical practice but also enhances the professional capabilities of medical staff and drives the development of the medical field.

[0184] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the integrated medical, educational, and research system based on a common underlying layer according to various embodiments of this application as described in the foregoing relevant parts of this specification.

[0185] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0186] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A medical-education-research integrated system based on a common underlying layer, characterized in that: include: The diagnosis and treatment decision-making subsystem is used to provide diagnosis and treatment decision support for preset diseases and syndromes based on expert thinking chains. Specifically, it includes: for each case, based on the collected information, judging the core indicators related to the condition, guiding users to sort out information and analyze the condition, automatically generating condition analysis and treatment suggestions, recording treatment choices and effect feedback, automatically reminding whether to include the case in the research queue when the case meets the preset research queue conditions, prompting treatment nodes, and visually displaying the disease development process. Among them, the expert thought chain is a set of medical processes and medical plans built based on historical cases and expert advice; and the core indicators are the indicators necessary to judge the condition under the existing diagnosis and treatment plan. The research cohort management subsystem is used to acquire medical information, include cases that meet preset requirements into the research cohort, organize the cases in the research cohort based on a preset structured dataset, and anonymize the case information; the structured dataset can be a general dataset or a personalized dataset adjusted based on user input information; assist researchers in generating treatment suggestions and sending them to corresponding clinicians to obtain actual treatment choices and effect feedback from clinicians; remove cases that do not meet the requirements of the research cohort; perform quality control and statistical analysis on the data of the research cohort, and connect AI tools to query relevant background information, sort out the statistical result framework, and generate data result analysis reports; adjust the expert thought process based on the actual research results of the research cohort; the medical information includes: patient medical records, examination and test results, and treatment plans during the clinical diagnosis and treatment process; The teaching subsystem is used to set up preset case presentation methods based on expert thinking chains and preset cognitive frameworks to guide users to form disease analysis thinking habits and conduct structured thinking; based on the material sets and case sets obtained by the diagnosis and treatment decision-making subsystem and the scientific research cohort management subsystem, teaching materials are generated after standardization and expert-level quality control, and combined with the simulator to generate skills training system and capacity building training system; The standardization and expert-level quality control include: converting the cases into a preset case presentation format and reviewing them by preset experts; The simulator is a device used to simulate patients; A three-in-one data engine; The aforementioned three-in-one data engine employs optical character recognition (OCR) and natural language processing (NLP) technologies to identify and analyze medical record texts, extracting key information such as disease names and symptom descriptions. It stores expert thought chains in the system database in the form of data structures for use by the treatment decision-making subsystem and the research cohort management subsystem. Furthermore, it binds clinical cases with expert thought chains and research conclusions to generate treatment guidelines and teaching cases. This provides the treatment decision-making subsystem with real-time patient condition data and expert suggestions, the research cohort management subsystem with structured research datasets, and the teaching subsystem with teaching cases and materials.

2. The integrated medical, educational, and research system based on a common underlying layer as described in claim 1, characterized in that, The diagnostic and treatment decision-making subsystem includes: The core indicator preset module is used to pre-set the core indicators required for the diagnosis of different diseases and syndromes based on expert experience, and display them to medical staff in tabular form on the system interface. The information collection and analysis module is used to support manual input or automatic collection of core information related to the diagnosis of the disease. After the collection is completed, the disease analysis is automatically generated and treatment suggestions are given. The core information mentioned above is the information necessary to determine the condition under the existing treatment plan; The recording and feedback module is used to record and provide feedback on the treatment plans and effects of medical staff; it also highlights treatment milestones and visualizes the disease progression. The research queue reminder module is used to automatically remind users whether to include a case in the research queue when a case meets the preset research queue criteria.

3. The integrated medical, educational, and research system based on a common underlying layer as described in claim 1, characterized in that, The scientific research queue management subsystem includes: The platform design and dataset editing module is used to provide general structured datasets or general research process platforms, or to generate personalized structured datasets or personalized research process platforms according to user configurations. The treatment suggestion and feedback module is used to generate and display research-oriented treatment suggestions for cases to researchers; obtain research-oriented treatment suggestions determined by researchers and send them to clinicians; obtain feedback from clinicians on whether to adopt the suggestions and record clinical behavior and corresponding treatment effects; if the current case does not meet the current research needs, the current case is removed. The data quality control and analysis report generation module is used to perform online quality control and statistical analysis on scientific research cohort data, connect to AI tools to query relevant background information, organize the statistical results framework, and generate data result analysis reports. The expert thinking chain adjustment module is used to periodically evaluate and optimize the diagnosis and treatment processes and plans in the expert thinking chain based on research results and expert opinions.

4. The integrated medical, educational, and research system based on a common underlying layer as described in claim 1, characterized in that, The teaching subsystem includes: Standardized presentation settings module: Based on expert thought processes and preset cognitive frameworks, preset case presentation methods are set; Material and Case Set Management Module: Stores a preset number of material and case sets generated by the diagnosis and treatment decision-making subsystem and the research cohort management subsystem, and uses them as teaching materials after standardization and expert-level quality control; Training system generation module: Combines with simulators to generate skills training systems and capacity-building training systems; The AI-powered automatic recognition and quality control platform generation module annotates a preset number of medical images and text materials to train an AI model for automatic recognition and quality control.

5. The integrated medical, educational, and research system based on a common underlying layer as described in claim 1, characterized in that, The three-in-one data engine is also used to update the diagnosis and treatment process and plan in the expert thought chain in real time according to the treatment effect data in clinical feedback and the statistical analysis data in scientific research results, and to adjust the diagnosis and treatment guidelines and teaching cases in combination with expert opinions.

6. The integrated medical, educational, and research system based on a common underlying layer as described in claim 1, characterized in that, The treatment decision-making subsystem also includes a behavior guidance and compliance monitoring module; The behavior guidance and compliance monitoring module is used to compare the deviation between the doctor's behavior and the expert's thought process. If the deviation is greater than a preset value, an early warning is triggered. The system's built-in permission management and approval process monitoring module records the data usage and ethical approval status in real time.

7. The integrated medical, educational, and research system based on a common underlying layer as described in claim 2, characterized in that, The case presentation methods include: The data is integrated into a timeline to trace the patient's entire treatment cycle. It combines seven key steps: patient admission / departmental admission, preliminary examination and diagnosis after information collection, treatment plan formulation, observation during treatment, assessment and feedback on changes in the condition, adjustment of the treatment plan, and patient discharge / departmental discharge. The data outlines the development of the disease, presents the content of each time point in a pre-set format, and shows the trend of changes in core indicators over time, reflecting the trend of disease changes and treatment direction.

8. The integrated medical, educational, and research system based on a common underlying layer as described in claim 1, characterized in that, The research cohort management subsystem is also used to automatically analyze medical data using data mining and machine learning algorithms, identify abnormal data patterns and potential correlations, and propose research hypotheses and generate corresponding research project proposals based on the characteristics and distribution patterns of abnormal data, combined with medical knowledge and expert experience.

Images