Medical management system and method for whole course of disease integration
By using medical treatment labels and semantic similarity analysis in the medical management system, integrating the patient's diagnostic information outside the hospital and considering the diagnostic information of immediate relatives when necessary, the problem that the existing system cannot fully grasp the patient's entire course information is improved The accuracy and personalization of diagnosis and treatment have been achieved, reducing medical risks and costs.
Patent Information
- Application Number
- CN202411068520.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-12-03
AI Technical Summary
The existing medical management system only realizes the integrated management of internal hospital information, and cannot include the information outside the hospital information into the information integration platform, making it difficult for doctors to fully grasp the patient's entire course of the disease, affecting the diagnosis and treatment effect.
By setting up medical treatment tags, sending request instructions to each medical information system to obtain similar diagnostic information of patients, conducting semantic similarity analysis and integration, formulating diagnostic plans, and introducing diagnostic information from immediate relatives when necessary to evaluate disease similarity, Personalized treatment plan.
It realizes information sharing between different medical institutions, reduces information mismatch and omission, improves diagnosis accuracy and treatment effect, and reduces medical risks and costs.
Smart Images

Figure CN119069061A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of medical management for full-course integration, and specifically relates to a medical management system and method for full-course integration. Background Art
[0002] In today's era of rapid development of information technology, hospital information integration platforms used to realize medical informatization have gradually entered major hospitals, paved the way for information highways within the hospital, and become the core of hospital information interaction.
[0003] The invention patent application with publication number CN116580827A discloses a medical full course management system, including an evaluation module, an online medical module, a follow-up module and a storage module; the evaluation module is used to evaluate the scale and obtain evaluation information; the online medical module is used to make appointments for consultations, appointments for examinations and instant messaging, and generate online medical information; the follow-up module is used to store and view follow-up information; the storage module is used to store full course information, wherein the full course information includes: at least one of the evaluation information, the follow-up information, the scale and the online medical information.
[0004] The existing technology only realizes the integrated management of information within the hospital, and cannot incorporate the patient's information outside the hospital into the information integration platform. The data is relatively scattered, which makes it difficult for doctors to fully grasp the patient's entire disease course information, affecting the diagnosis and treatment effect. Therefore, the present invention provides a medical management system and method for full disease course integration. Summary of the invention
[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a medical management system and method for full-course integration, which is used to solve the technical problem that the prior art only realizes information integration management within the hospital, but cannot incorporate the patient's information outside the hospital into the information integration platform, and the data is relatively scattered, which makes it difficult for doctors to fully grasp the patient's full-course information, affecting the diagnosis and treatment effect.
[0006] To achieve the above object, the first aspect of the present invention provides a medical management method for full-course integration, comprising the following steps:
[0007] Step 1: Set a medical tag for the patient; the medical tag is the unique identification of the patient;
[0008] Step 2: After the patient agrees, a request instruction including the medical treatment tag and the target diagnosis information is sent to each medical information system, and each medical information system feeds back similar diagnosis information of the patient; the target diagnosis information is the various examination information of the patient in the target medical institution; the diagnosis information includes age, symptoms, genes, diagnosis results and various examination reports; the diagnosis results include common diseases and complex diseases;
[0009] Step 3: Integrate the target diagnostic information with similar diagnostic information to obtain comprehensive diagnostic information and formulate a corresponding diagnosis and treatment plan.
[0010] Preferably, each medical information system feeds back similar diagnosis information of the patients, including:
[0011] When each medical information system receives the request instruction, it checks whether the medical treatment tags are consistent; if yes, it will filter out all the diagnosis information of the patients, mark it as non-target diagnosis information, and sort it by time; if no, there is no diagnosis information of the patients; among which, non-target diagnosis information is information obtained from examinations not conducted in the target hospital;
[0012] Diagnostic information with the same diagnostic result is extracted from non-target diagnostic information and target diagnostic information to obtain similar diagnostic information.
[0013] By checking the consistency of the medical treatment labels, the present invention can ensure that the screened diagnostic information is for the correct patient, avoiding misdiagnosis or treatment errors caused by information mismatch; at the same time, it avoids unnecessary search and screening work, saving time and human resources; it realizes information sharing between different medical institutions, and promotes the rational allocation and efficient use of medical resources. At the same time, this also helps to strengthen the collaboration between medical institutions and improve the overall level of medical services.
[0014] Preferably, extracting diagnostic information with the same diagnostic result from the non-target diagnostic information and the target diagnostic information includes:
[0015] The diagnostic results of the target diagnostic information are compared with the diagnostic results of the non-target diagnostic information in turn for semantic similarity, and the non-target diagnostic information that is greater than the semantic similarity threshold is extracted and marked as similar diagnostic information, and fed back to the medical information system of the target medical institution; among which, the semantic similarity analysis can be implemented based on the method of the vector space model.
[0016] The present invention can accurately screen out content similar to target diagnostic information from a large amount of non-target diagnostic information through semantic similarity analysis; it helps doctors quickly understand similar diagnoses that patients have received in other medical institutions, so as to have a more comprehensive understanding of the patient's condition; through automated semantic similarity comparison, it reduces the workload of doctors in manually searching and comparing diagnostic information, thereby improving work efficiency; and by reducing medical risks caused by inconsistent or missing information, it enhances patients' trust in medical services.
[0017] Preferably, the target diagnostic information is integrated with similar diagnostic information to obtain comprehensive diagnostic information, including:
[0018] The examination reports in the similar diagnostic information are classified according to their types; the keywords of the examination reports in the target diagnostic information and the similar diagnostic information are extracted, the keywords in the examination reports of the same type are subjected to semantic similarity analysis, and the examination reports with a semantic similarity greater than a threshold are marked as different reports; and similar diagnostic information not contained in the target diagnostic information is retained; wherein the different reports are verified by the doctor according to the diagnosis and treatment situation or further diagnosis, and the examination reports that meet the current symptoms are retained.
[0019] The present invention extracts keywords from target diagnostic information and similar diagnostic information, accurately identifies examination reports that are different from similar diagnostic information and target diagnostic information, helps doctors quickly discover possible abnormalities or missed information, reduces medical risks caused by inconsistent or missed information, and improves diagnostic accuracy; comprehensively understands the patient's medical history and diagnostic information, provides patients with more personalized and accurate treatment plans, and thus optimizes the patient's medical experience.
[0020] Preferably, if the diagnosis result is a complex disease, similar diagnosis information of the immediate family members is obtained based on the target diagnosis information of the patient and the medical treatment labels of the immediate family members; the similarity between the comprehensive diagnosis information and the similar diagnosis information of the immediate family members is analyzed to obtain the disease similarity.
[0021] Preferably, the analysis of the similarity between the comprehensive diagnosis information and similar diagnosis information of the immediate family members to obtain the disease similarity includes:
[0022] The ages, genes and symptoms of the patients and their immediate family members are extracted respectively, input into the similarity assessment model, and the disease similarity is output; the properties of complex diseases are analyzed based on the similarity; wherein, the similarity assessment model is trained by an artificial intelligence model.
[0023] The present invention aims at complex diseases. By introducing similar diagnostic information of immediate family members, the similarity of diseases between patients and immediate family members is evaluated, and the family characteristics and genetic background of the disease are fully understood. Potential risks of disease can be discovered at an early stage, so that preventive measures can be taken in time to reduce the probability of disease occurrence or slow down the progression of the disease. This helps doctors to more accurately judge the nature, development stage and possible complications of the disease, and formulate more effective treatment plans for patients.
[0024] Preferably, the similarity-based analysis of the properties of complex diseases includes:
[0025] Determine whether the disease similarity is greater than the disease similarity threshold; if yes, the complex disease is a genetic disease; if no, the complex disease is not a genetic disease.
[0026] Preferably, the similarity evaluation model is obtained by training an artificial intelligence model, including:
[0027] Extract the age, gene and symptoms of several patients and their immediate family members from various medical information systems and integrate them into standard input data; mark the disease similarity as standard output data;
[0028] The artificial intelligence model is trained using standard input data and standard output data to obtain a similarity evaluation model.
[0029] Preferably, formulating a corresponding diagnosis and treatment plan based on the comprehensive diagnostic information includes:
[0030] If the diagnosis is a common disease, medication and regular check-ups will be used based on the comprehensive diagnostic information;
[0031] If the complex disease is a genetic disease, a treatment plan for the target disease will be developed based on the treatment plan of the patient's immediate family members;
[0032] If the complex disease is not a genetic disease, the patient's living conditions will be analyzed to find the cause; living conditions include living habits and living conditions.
[0033] The present invention formulates personalized treatment plans according to different diagnostic results, thereby ensuring that the treatment measures match the patient's specific condition, thereby improving the treatment effect; at the same time, the use of targeted treatment methods can avoid unnecessary waste of medical resources; for conditions that do not require complex treatment, excessive examinations and unnecessary operations can be avoided, thereby reducing medical costs; this not only helps to improve economic benefits, but also enables a more reasonable allocation and utilization of medical resources.
[0034] Preferably, a medical management system for a medical management method for integrating the entire course of disease is characterized by comprising an information collection module, an information integration module, and a diagnosis and evaluation module;
[0035] The information collection module collects the target diagnosis information and similar diagnosis information of the patient through the medical treatment label, as well as the similar diagnosis information of his / her immediate family members;
[0036] The information integration module integrates the target diagnostic information with similar diagnostic information to obtain comprehensive diagnostic information;
[0037] The diagnostic evaluation module analyzes the similarity between the comprehensive diagnostic information and similar diagnostic information of the patient's immediate family members to obtain the disease similarity; based on the disease similarity, it determines whether the patient's disease is a genetic disease; if so, a treatment plan for the target disease is formulated; if not, the cause of the disease is analyzed.
[0038] Compared with the prior art, the present invention has the following beneficial effects:
[0039] 1. The present invention obtains non-target diagnostic information of patients through medical treatment labels, and performs similarity analysis between the non-target diagnostic information and the target diagnostic information to obtain similar diagnostic information; based on the similar diagnostic information, different reports therein and examination reports not included in the target diagnostic information are extracted to obtain comprehensive diagnostic information; this helps doctors to have a more comprehensive understanding of the patient's condition, reduces medical risks caused by inconsistent or missing information, and improves the accuracy of diagnosis;
[0040] 2. If the diagnosis result of the present invention is a complex disease, similar diagnostic information between the immediate family members and the patient will be analyzed and extracted, and the similar diagnostic information will be input into the similarity evaluation model to output the disease similarity. If the disease similarity is greater than the disease similarity threshold, it is a genetic disease. After being diagnosed with a genetic disease, the doctor can formulate a more accurate and personalized treatment plan based on the genetic characteristics. This can not only improve the treatment effect, but also avoid unnecessary examinations and treatments, optimize the entire diagnosis and treatment process, and provide patients with a better medical experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 is a flow chart of the method of the present invention;
[0043] Figure 2 The analysis process of complex diseases of the present invention;
[0044] Figure 3 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION
[0045] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0046] See also Figure 1 The first aspect of the present invention provides a medical management method for full-course integration, comprising the following steps:
[0047] Step 1: Set a medical tag for the patient; the medical tag is the unique identification of the patient;
[0048] Step 2: After the patient agrees, a request instruction including the medical treatment tag and the target diagnosis information is sent to each medical information system, and each medical information system feeds back similar diagnosis information of the patient; the target diagnosis information is the various examination information of the patient in the target medical institution; the diagnosis information includes age, symptoms, genes, diagnosis results and various examination reports; the diagnosis results include common diseases and complex diseases:
[0049] Specifically, when each medical information system receives the request instruction, it checks whether the medical treatment tags are consistent; if yes, it filters out all the diagnosis information of the patients, marks them as non-target diagnosis information, and sorts them by time; if no, there is no diagnosis information of the patients; among which, non-target diagnosis information is obtained by examinations not conducted in the target hospital;
[0050] The diagnostic results of the target diagnostic information are compared with the diagnostic results of the non-target diagnostic information in turn for semantic similarity, and the non-target diagnostic information that is greater than the semantic similarity threshold is extracted and marked as similar diagnostic information, and fed back to the medical information system of the target medical institution; among which, the semantic similarity analysis can be implemented based on the method of the vector space model.
[0051] It should be noted that complex diseases are those that require the cooperation of at least two departments to treat; other disease types are common diseases.
[0052] Among them, the medical treatment label includes the patient's name, ID number, medical card number and other unique identification information. The medical information system refers to a system that uses modern means such as computer software and hardware technology and network communication technology to comprehensively manage the information of medical patients.
[0053] For example, when patient A goes to hospital X for treatment, hospital X will send patient A's treatment tag and diagnosis results to the medical information systems of other hospitals. First, each medical information system will check patient A's treatment tag. If the treatment tags are consistent, each medical information system will filter out the diagnosis information of patient A in all non-target hospitals. This information may include past disease diagnoses, examination results, medication records, and various examination reports. This information is marked as non-target diagnosis information and sorted in chronological order so that doctors can clearly understand the historical development of patient A's condition. If the treatment tags are inconsistent, it means that the system may not be able to find a patient record that matches the current treatment request, so the system determines that there is no diagnosis information for the patient.
[0054] The system will compare the semantic similarity between the diagnosis result of patient A in hospital X and the diagnosis result of the non-target diagnosis information screened out above; this comparison can be achieved based on methods such as vector space models to find non-target diagnosis information with similar diagnosis results, which is of reference value to the current condition of patient A.
[0055] Among them, the Vector Space Model (VSM) is an algebraic model used for information filtering, information retrieval, indexing, and evaluating relevance. The basic principle of the vector space model is to define the features of the text as vectors in a multidimensional space, define the same or similar features as close vectors, and define different or unrelated features as far away vectors, so as to detect the correlation between data. In the vector space model, documents are represented as vectors in the document space, so that we can measure the similarity between documents by calculating the similarity between vectors.
[0056] Step 3: Integrate the target diagnosis information with similar diagnosis information to obtain comprehensive diagnosis information and formulate a corresponding diagnosis and treatment plan:
[0057] Specifically, the inspection reports in the similar diagnostic information are classified according to their types; the keywords of the inspection reports in the target diagnostic information and the similar diagnostic information are extracted, the keywords in the inspection reports of the same type are subjected to semantic similarity analysis to obtain a semantic similarity value, and it is determined whether the semantic similarity value is greater than a semantic similarity threshold; if so, the inspection report of the non-target diagnostic information is not retained; if not, the inspection report is marked as a different report; and similar diagnostic information not contained in the target diagnostic information is retained.
[0058] The doctor shall verify the different reports based on the treatment or further diagnosis, and retain the examination reports that are consistent with the current symptoms.
[0059] For example, among Zhang San's similar diagnosis information, the system found 10 examination reports, of which 5 were electrocardiogram reports, 3 were blood routine reports, and 2 were urine routine reports. For each examination report, the system used natural language processing (NLP) technology to extract keywords. These keywords can represent the main content of the report and help with subsequent semantic similarity analysis. For example, in Zhang San's electrocardiogram report, the system extracted keywords such as "arrhythmia" and "myocardial ischemia". In the electrocardiogram reports of similar diagnosis information, the system also extracted corresponding keywords. The semantic similarity between keywords was analyzed to obtain similar diagnosis information.
[0060] See also Figure 2 ,If the diagnosis result is a complex disease, similar diagnosis information of the immediate family members is obtained based on the target diagnosis information of the patient and the medical treatment labels of the immediate family members;
[0061] The similarity between the comprehensive diagnosis information and the similar diagnosis information of their immediate family members was analyzed to obtain the disease similarity. The analysis process is as follows:
[0062] The ages, genes and symptoms of the patients and their immediate family members are extracted respectively, input into the similarity assessment model, and the disease similarity is output; it is determined whether the disease similarity is greater than the disease similarity threshold; if yes, the complex disease is a genetic disease; if no, the complex disease is not a genetic disease; wherein the similarity assessment model is trained by an artificial intelligence model.
[0063] The training process of the similarity evaluation model is as follows:
[0064] Extract the age, gene and symptoms of several patients and their immediate family members from various medical information systems and integrate them into standard input data; mark the disease similarity as standard output data;
[0065] The artificial intelligence model is trained using standard input data and standard output data to obtain a similarity evaluation model.
[0066] If the diagnosis is a common disease, medication and regular check-ups will be used based on the comprehensive diagnostic information;
[0067] If the complex disease is a genetic disease, a treatment plan for the target disease will be developed based on the treatment plan of the patient's immediate family members;
[0068] If the complex disease is not a genetic disease, the patient's living conditions will be analyzed to find the cause; living conditions include living habits and living conditions.
[0069] See also Figure 3, the second aspect of the present invention provides a medical management system for a medical management method for full-course integration, including an information collection module, an information integration module, and a diagnosis and evaluation module;
[0070] The information collection module collects the target diagnosis information and similar diagnosis information of the patient through the medical treatment label, as well as the similar diagnosis information of his / her immediate family members;
[0071] The information integration module integrates the target diagnostic information with similar diagnostic information to obtain comprehensive diagnostic information;
[0072] The diagnostic evaluation module analyzes the similarity between the comprehensive diagnostic information and similar diagnostic information of the patient's immediate family members to obtain the disease similarity; based on the disease similarity, it determines whether the patient's disease is a genetic disease; if so, a treatment plan for the target disease is formulated; if not, the cause of the disease is analyzed.
[0073] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.
[0074] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A medical management method for the integration of the entire course of disease, characterized in that: The following steps are involved: Step 1: Set a medical tag for the patient; the medical tag is the unique identification of the patient; Step 2: After the patient agrees, a request instruction including the patient's visit label and target diagnosis information is sent to each medical information system, and each medical information system feeds back similar diagnosis information of the patient; the target diagnosis information is the various examination information of the patient in the target medical institution; the diagnosis information includes age, symptoms, genes, diagnosis results and various examination reports; the diagnosis results include common diseases and complex diseases; Step 3: Integrate the target diagnostic information with similar diagnostic information to obtain comprehensive diagnostic information; based on the comprehensive diagnostic information, formulate a corresponding diagnosis and treatment plan.
2. A medical management method for full course of disease integration according to claim 1, characterized in that: Each medical information system feeds back similar diagnosis information of the patient, including: When each medical information system receives the request instruction, it checks whether the medical treatment tags are consistent; if yes, it will filter out all the diagnosis information of the patients, mark it as non-target diagnosis information, and sort it by time; if no, there is no diagnosis information of the patients; among which, non-target diagnosis information is information obtained from examinations not conducted in the target hospital; Diagnostic information with the same diagnostic result is extracted from non-target diagnostic information and target diagnostic information to obtain similar diagnostic information.
3. A medical management method for full course of disease integration according to claim 2, characterized in that: The extracting diagnostic information with the same diagnostic result from the non-target diagnostic information and the target diagnostic information includes: The diagnostic results of the target diagnostic information are compared with the diagnostic results of the non-target diagnostic information in turn for semantic similarity, and the non-target diagnostic information that is greater than the semantic similarity threshold is extracted and marked as similar diagnostic information, and fed back to the medical information system of the target medical institution; among which, the semantic similarity analysis can be implemented based on the method of the vector space model.
4. A medical management method for full course of disease integration according to claim 1, characterized in that: The target diagnostic information is integrated with similar diagnostic information to obtain comprehensive diagnostic information, including: The examination reports in the similar diagnostic information are classified according to their types; the keywords of the examination reports in the target diagnostic information and the similar diagnostic information are extracted, the keywords in the examination reports of the same type are subjected to semantic similarity analysis, and the examination reports with a semantic similarity greater than a threshold are marked as different reports; and similar diagnostic information not contained in the target diagnostic information is retained; wherein the different reports are verified by the doctor according to the diagnosis and treatment situation or further diagnosis, and the examination reports that meet the current symptoms are retained.
5. A medical management method for full-course integration according to claim 1, characterized in that: If the diagnosis result is a complex disease, similar diagnosis information of the immediate family members is obtained based on the target diagnosis information of the patient and the medical treatment labels of the immediate family members; the similarity between the comprehensive diagnosis information and the similar diagnosis information of the immediate family members is analyzed to obtain the disease similarity.
6. A medical management method for full course integration according to claim 5, characterized in that: The analysis of the similarity between the comprehensive diagnosis information and the similar diagnosis information of the immediate family members to obtain the disease similarity includes: The ages, genes and symptoms of the patients and their immediate family members are extracted respectively, input into the similarity assessment model, and the disease similarity is output; the properties of complex diseases are analyzed based on the similarity; wherein, the similarity assessment model is trained by an artificial intelligence model.
7. A medical management method for full-course integration according to claim 6, characterized in that: The attributes of complex diseases analyzed based on similarity include: Determine whether the disease similarity is greater than the disease similarity threshold; if yes, the complex disease is a genetic disease; if no, the complex disease is not a genetic disease.
8. A medical management method for full-course integration according to claim 6, characterized in that: The similarity evaluation model is obtained by training an artificial intelligence model, and includes: Extract the age, gene and symptoms of several patients and their immediate family members from various medical information systems and integrate them into standard input data; mark the disease similarity as standard output data; The artificial intelligence model is trained using standard input data and standard output data to obtain a similarity evaluation model.
9. A medical management method for full-course integration according to claim 1, characterized in that: Based on the comprehensive diagnostic information, a corresponding diagnosis and treatment plan is formulated, including: If the diagnosis is a common disease, medication and regular check-ups will be used based on the comprehensive diagnostic information; If the complex disease is a genetic disease, a treatment plan for the target disease will be developed based on the treatment plan of the patient's immediate family members; If the complex disease is not a genetic disease, the patient's living conditions will be analyzed to find the cause; living conditions include living habits and living conditions.
10. A medical management system for a medical management method for integrating the entire course of disease, characterized in that: It includes information collection module, information integration module and diagnosis and evaluation module; The information collection module collects the target diagnosis information and similar diagnosis information of the patient through the medical treatment label, as well as the similar diagnosis information of his / her immediate family members; The information integration module integrates the target diagnostic information with similar diagnostic information to obtain comprehensive diagnostic information; The diagnostic assessment module analyzes the similarity between the comprehensive diagnostic information and similar diagnostic information of the patient's immediate family members to obtain the disease similarity; based on the disease similarity, it determines whether the patient's disease is a genetic disease; if so, a treatment plan for the target disease is formulated; If not, analyze the cause.
