Medical health data intelligent management method and system

Abstract

The invention discloses an intelligent management method and system for medical and health data, and relates to the technical field of medical data management, and the method comprises the steps: obtaining the medical and health data correspondingly collected by each medical object, marking the part of medical and health data with the discrete degree not meeting the clustering expectation after the outcluster clustering analysis, and obtaining the part of medical and health data; the method comprises the following steps: acquiring part of medical and health data, taking the part of medical and health data as abnormal data, when abnormal data is monitored in a medical object, generating an early warning prompt, sending the early warning prompt to a data operation and maintenance worker corresponding to the corresponding medical object, and starting data correction processing on the abnormal data by the data operation and maintenance worker. The method comprises the following steps: extracting information features of all medical and health data of each medical object to construct a corresponding medical data graph, and performing periodic information maintenance on the medical and health data through the medical data graph; and according to the medical data atlas, generating inquiry auxiliary reference information of the current doctor seeing patient and pushing the inquiry auxiliary reference information.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data management, and in particular to a medical and health data intelligent management method and system. Background Art

[0002] With the continuous development of the medical industry, a large amount of medical and health data generated by various hospitals, medical institutions, community health service stations and other medical objects in the process of medical and health services is difficult to be effectively managed, and the management of medical and health data is crucial.

[0003] How to quickly locate abnormal data in the medical and health data of hospitals, medical institutions and community health service stations, and issue corresponding warnings and handle the abnormal data; how to establish a medical data map for the large amount of medical and health data corresponding to each medical object to facilitate the subsequent maintenance and management of the corresponding medical and health data.

[0004] When consulting patients in hospitals, medical institutions and community health service stations, some medical data information in the medical data map that can serve as a reference for the current patient's condition is pushed to the doctor. These are the problems currently faced by various hospitals, medical institutions and community health service stations in medical and health data management. Summary of the invention

[0005] In order to solve the above problems, the purpose of the present invention is to provide a medical and health data intelligent management method and system.

[0006] The purpose of the present invention can be achieved by the following technical solution: A medical and health data intelligent management method, comprising the following steps:

[0007] Step S1: Obtain the medical and health data collected corresponding to each medical object, and after performing outlier cluster analysis on the medical and health data, mark out some medical and health data whose discreteness does not meet the clustering expectation, and regard this part of the medical and health data as abnormal data;

[0008] Step S2: When abnormal data is detected in a medical object, an early warning is generated and sent to the data operation and maintenance personnel corresponding to the medical object, and the data operation and maintenance personnel initiates data correction processing for the abnormal data;

[0009] Step S3: extracting information features of all medical and health data of each medical object, constructing a medical data map of the corresponding medical object according to the information features, and performing periodic information maintenance on the medical and health data through the medical data map;

[0010] Step S4: When the doctor is interviewing the patient under the corresponding medical object, the consultation auxiliary reference information of the current patient is generated according to the medical data map, and the consultation auxiliary reference information is pushed to the doctor.

[0011] Furthermore, the process of obtaining the medical and health data collected corresponding to each medical object includes:

[0012] The types of medical objects include hospitals, medical institutions and community health service stations;

[0013] Each medical object is numbered and denoted as i, then i=1, 2, 3, ..., n, where n is a natural number greater than 0. A corresponding medical data storage terminal is set at each medical object, and a data collection period and a data collection frequency are set for each medical object. During the data collection period, each medical object transmits the corresponding medical and health data obtained by itself to its own medical data storage terminal according to the corresponding data collection frequency for storage, and the medical and health data corresponding to the medical object numbered i is denoted as Data[i].

[0014] Furthermore, after performing outlier cluster analysis on the medical and health data, a portion of the medical and health data whose discreteness does not meet the clustering expectation is marked, and the process of treating the portion of the medical and health data as abnormal data includes:

[0015] The medical and health data corresponding to each medical object is divided into several data clusters, a data monitoring node is set for each data cluster, the data cluster is converted into a text document and entered into the corresponding data monitoring node, and the data monitoring node obtains the representation vectors of several words and sentences included in each data cluster;

[0016] Obtain the cosine similarity value between the representation vector corresponding to each word and sentence and the preset reference vector, and record the cosine similarity value as COSθ, set a first determination threshold corresponding to the cosine similarity value COSθ, and record the first determination threshold as μ, record the number of words and sentences corresponding to COSθ≤μ in each data cluster as N1, and record the number of words and sentences corresponding to COSθ>μ in each data cluster as N2;

[0017] Generate a discrete coefficient corresponding to each data cluster according to N1 and N2, record the discrete coefficient corresponding to the data cluster numbered i as τ[i], then τ[i]=N1 / (N1+N2), set a second determination threshold corresponding to the discrete coefficient, and record the second determination threshold as η;

[0018] When τ[i]≥η, some medical and health data represented by the data cluster corresponding to the current discrete coefficient τ[i] are marked as having a discrete degree that does not meet the clustering expectations, and some medical data that do not meet the clustering expectations are regarded as abnormal data. When τ[i]<η, it means that the discrete degree of the medical and health data represented by the data cluster corresponding to the current discrete coefficient meets the clustering expectations, and no operation is performed.

[0019] Furthermore, when abnormal data is detected in a medical object, an early warning is generated and sent to the data operation and maintenance personnel corresponding to the corresponding medical object. The data operation and maintenance personnel initiates the process of data correction processing of the abnormal data, including:

[0020] The data synchronization monitoring is performed according to the preset monitoring frequency and monitoring steps, wherein the monitoring step is composed of a number of sub-execution steps. When each sub-execution step is performed according to the preset execution order, it indicates that the data synchronization monitoring is in a correct execution state and no operation is performed. When any sub-execution step in the monitoring step is not performed according to the preset execution order, it is marked that the data synchronization monitoring is in an abnormal execution state.

[0021] For data synchronization monitoring in an abnormal execution state, the monitoring frequency and monitoring steps are reset, and then the data synchronization monitoring is re-executed. When abnormal data is detected in the corresponding medical object, an early warning prompt corresponding to the abnormal data is generated, and the early warning prompt is sent to the data operation and maintenance personnel corresponding to the medical object represented by the current abnormal data, and the data operation and maintenance personnel determine whether the abnormal data can be corrected;

[0022] If so, the abnormal data is input into a preset parameter library, and the abnormal correction parameters corresponding to the abnormal data are matched from the parameter library, and then the abnormal data is converted into corresponding normal medical and health data through the abnormal correction parameters. If not, the abnormal data is directly discarded.

[0023] Furthermore, the process of extracting information features of all medical and health data of each medical object and constructing a medical data map of the corresponding medical object according to the information features includes:

[0024] When all medical and health data of each medical object do not have abnormalities, an information extraction period is set for each medical object, and an initial blank data map is established. During the information extraction period, all medical and health data of each medical object are processed by big data technology to obtain information features corresponding to each medical object. The information features include different disease types, disease examination items, disease diagnosis and treatment drugs and disease treatment measures corresponding to different disease types, and also include the disease progression trend of the corresponding disease after the corresponding disease examination items, disease diagnosis and treatment drugs and disease treatment measures of different disease types are completed;

[0025] Among them, different disease types are used as the main index data of the atlas, the disease examination items, disease diagnosis and treatment drugs and disease treatment measures corresponding to different disease types are used as the atlas branch extension data, and the disease progression trend of different diseases after the execution of the corresponding disease examination items, disease diagnosis and treatment drugs and disease treatment measures is used as the atlas branch convergence data. A number of path nodes are set in the blank data atlas, and the main index data of the atlas, the atlas branch extension data and the atlas branch convergence data corresponding to each medical object are assigned to corresponding path nodes in the blank data atlas, and then the diagnostic atlas path corresponding to each medical object is constructed, and the diagnostic atlas paths corresponding to all medical objects are integrated to generate the final medical data atlas.

[0026] Furthermore, the process of periodic information maintenance of medical and health data through the medical data graph includes:

[0027] Each diagnostic map path included in the medical data map is regarded as an information maintenance object, and a corresponding maintenance cycle is set for each information maintenance object. Each information maintenance object performs corresponding periodic information maintenance under its own maintenance cycle. The maintenance content of the periodic information maintenance includes updating the newly generated medical and health data corresponding to each diagnostic map path in the medical data map, deleting the expired medical and health data in each diagnostic map path, and modifying the erroneous medical and health data in each diagnostic map path.

[0028] Furthermore, when a doctor is consulting a patient under a corresponding medical object, the process of generating consultation auxiliary reference information of the current patient according to the medical data map and pushing the consultation auxiliary reference information to the doctor includes:

[0029] Each medical object is assigned a corresponding doctor, who is responsible for consulting all patients under the corresponding medical object. When the doctor consults the patients he is responsible for, he obtains the real-time consultation information of the patients. The real-time consultation information is obtained by performing a series of medical examinations on the current patients and asking about the patients' condition.

[0030] The multiple diagnosis graph paths included in the medical data graph corresponding to the current medical object are respectively regarded as a matching entity, and the real-time consultation information of each patient is regarded as the element to be matched. The matching entity corresponding to each element to be matched is matched from the medical data graph, and all the medical and health data included in the diagnosis graph path corresponding to the matching entity is regarded as the auxiliary reference information of the consultation of the current patient;

[0031] An electronic health record system, mobile medical applications and communication tools are arranged in the doctor's work scenario for consultation, and the generated consultation auxiliary reference information of the patient is pushed to the doctor for display through the electronic health record system, mobile medical applications and communication tools.

[0032] Furthermore, a medical and health data intelligent management system includes:

[0033] The medical and health data analysis module is used to obtain the medical and health data collected corresponding to each medical object, and after performing outlier cluster analysis on the medical and health data, mark out some medical and health data whose discrete degree does not meet the clustering expectation, and regard this part of the medical and health data as abnormal data;

[0034] The medical and health data correction module generates an early warning when abnormal data is detected in a medical object and sends it to the data operation and maintenance personnel corresponding to the medical object, who then initiates the data correction process for the abnormal data.

[0035] The medical data map module is used to extract the information characteristics of all medical and health data of each medical object, construct the medical data map of the corresponding medical object according to the information characteristics, and perform periodic information maintenance on the medical and health data through the medical data map;

[0036] The medical consultation auxiliary module generates consultation auxiliary reference information of the current patient based on the medical data map when the doctor is consulting the patient under the corresponding medical object, and pushes the consultation auxiliary reference information to the doctor.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. By acquiring the medical and health data collected from each medical object, and marking the part of the medical and health data whose discrete degree does not meet the clustering expectations as abnormal data after outlier cluster analysis, when abnormal data is monitored in the medical object, an early warning prompt is generated and sent to the data operation and maintenance personnel corresponding to the corresponding medical object. The data operation and maintenance personnel initiate the data correction processing of the abnormal data, efficiently locate the abnormal medical and health data, and make early warnings and processing in time, ensuring the security of medical and health data.

[0039] 2. Extract the information features of all medical and health data of each medical object, construct a medical data map of the corresponding medical object based on the information features, and perform periodic information maintenance on the medical and health data through the medical data map. When the doctor interviews the patients under the corresponding medical object, the auxiliary reference information of the current patient is generated according to the medical data map, and the auxiliary reference information is pushed to the doctor. The established medical data map facilitates the subsequent maintenance and management of medical and health data, and can push auxiliary reference information that is helpful for diagnosing the patient's condition when the doctor interviews the patient, thereby improving the efficiency of the consultation to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0041] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0042] like Figure 1 As shown, a medical and health data intelligent management method includes the following steps:

[0043] Step S1: Obtain the medical and health data collected corresponding to each medical object, and after performing outlier cluster analysis on the medical and health data, mark out some medical and health data whose discreteness does not meet the clustering expectation, and regard this part of the medical and health data as abnormal data;

[0044] Step S2: When abnormal data is detected in a medical object, an early warning is generated and sent to the data operation and maintenance personnel corresponding to the medical object, and the data operation and maintenance personnel initiates data correction processing for the abnormal data;

[0045] Step S3: extracting information features of all medical and health data of each medical object, constructing a medical data map of the corresponding medical object according to the information features, and performing periodic information maintenance on the medical and health data through the medical data map;

[0046] Step S4: When the doctor is interviewing the patient under the corresponding medical object, the consultation auxiliary reference information of the current patient is generated according to the medical data map, and the consultation auxiliary reference information is pushed to the doctor.

[0047] It should be further explained that, in the specific implementation process, the process of obtaining the medical and health data collected corresponding to each medical object includes:

[0048] The types of medical objects include hospitals, medical institutions and community health service stations;

[0049] Each medical object is numbered and the number is recorded as i, then i=1, 2, 3, ..., n, where n is a natural number greater than 0, a corresponding medical data storage terminal is set at each medical object, and a data collection period and a data collection frequency are set for each medical object. During the data collection period, each medical object transmits the corresponding acquired medical and health data to its own medical data storage terminal according to the corresponding data collection frequency for storage;

[0050] Among them, the medical and health data corresponding to the medical object numbered i is recorded as Data[i].

[0051] It should be further explained that, in the specific implementation process, the process of marking some medical and health data whose discrete degree does not meet the clustering expectation after outlier cluster analysis on medical and health data and treating this part of medical and health data as abnormal data includes:

[0052] The medical and health data corresponding to each medical object is divided into several data clusters, and the outlier cluster analysis is performed on the several data clusters included in the medical and health data under each medical object by using a clustering algorithm, specifically: a data monitoring node is set for each data cluster, the data cluster is converted into a text document and entered into the corresponding data monitoring node, and the data monitoring node obtains the representation vectors of the several words and sentences included in each data cluster;

[0053] Obtain the cosine similarity value between the representation vector corresponding to each word and sentence and the preset reference vector, specifically as follows: set a reference vector, and record the reference vector as P, record the representation vector corresponding to each word and sentence as Q[ID], where ID is the binary code of each corresponding word and sentence in the data cluster;

[0054] The cosine similarity between the reference vector P and the representation vector Q[ID] is expressed as follows:

[0055] The cosine similarity value is recorded as COSθ, then Among them, θ is the angle between the reference vector P and the representation vector Q[ID], ||P|| is the modulus of the reference vector, and ||Q[ID]|| is the modulus corresponding to the representation vector;

[0056] Set the first judgment threshold corresponding to the cosine similarity value COSθ, and record the first judgment threshold as μ, record the number of words and sentences corresponding to COSθ≤μ in each data cluster as N1, and record the number of words and sentences corresponding to COSθ>μ in each data cluster as N2;

[0057] Generate a discrete coefficient corresponding to each data cluster according to N1 and N2, record the discrete coefficient corresponding to the data cluster numbered i as τ[i], then τ[i]=N1 / (N1+N2), set a second determination threshold corresponding to the discrete coefficient, and record the second determination threshold as η;

[0058] When τ[i]≥η, some medical and health data represented by the data cluster corresponding to the current discrete coefficient τ[i] are marked as having a discrete degree that does not meet the clustering expectations, and some medical data that do not meet the clustering expectations are regarded as abnormal data. When τ[i]<η, it means that the discrete degree of the medical and health data represented by the data cluster corresponding to the current discrete coefficient meets the clustering expectations, and no operation is performed.

[0059] It should be further explained that, in the specific implementation process, when abnormal data is detected in a medical object, an early warning prompt is generated and sent to the data operation and maintenance personnel corresponding to the corresponding medical object. The data operation and maintenance personnel initiate the data correction process for the abnormal data, which includes:

[0060] Perform real-time data synchronization monitoring on each medical object, and the data synchronization monitoring is performed according to a preset monitoring frequency and monitoring steps, wherein the monitoring step is composed of a number of sub-execution steps. When each sub-execution step is performed according to a preset execution order, it indicates that the data synchronization monitoring is in a correct execution state and no operation is performed. When any sub-execution step in the monitoring step is not performed according to the preset execution order, it is marked that the data synchronization monitoring is in an abnormal execution state.

[0061] For data synchronization monitoring in an abnormal execution state, the monitoring frequency and monitoring steps are reset, and then the data synchronization monitoring is re-executed. When abnormal data is detected in the corresponding medical object, an early warning prompt corresponding to the abnormal data is generated, and the early warning prompt is sent to the data operation and maintenance personnel corresponding to the medical object represented by the current abnormal data;

[0062] The data operation and maintenance personnel shall determine whether the abnormal data can be processed by data correction. If so, the abnormal data shall be input into the preset parameter library, and the abnormal correction parameters corresponding to the abnormal data shall be matched from the parameter library, and then the abnormal data shall be converted into corresponding normal medical and health data through the abnormal correction parameters. If not, the abnormal data shall be directly eliminated.

[0063] It should be further explained that, in the specific implementation process, the process of extracting the information features of all medical and health data of each medical object and constructing the medical data map of the corresponding medical object according to the information features includes:

[0064] When all medical and health data of each medical object are normal, an information extraction period is set for each medical object, and the information extraction period is recorded as T 提取 , then T 提取 =[t 始 , t 终 ], where t 始 is the start time of the information extraction period, t 终 The end time of the information extraction period;

[0065] Establish an initial blank data map, and during the information extraction period, perform feature processing on all medical and health data of each medical object through big data technology, and then obtain the information features corresponding to each medical object, the information features include different disease types, disease examination items, disease diagnosis and treatment drugs and disease treatment measures corresponding to different disease types, and also include the disease progression trend of the corresponding disease after the corresponding disease examination items, disease diagnosis and treatment drugs and disease treatment measures of different disease types are completed;

[0066] Among them, different disease types are used as the main index data of the atlas, the disease examination items, disease diagnosis and treatment drugs and disease treatment measures corresponding to different disease types are used as the atlas branch extension data, and the disease progression trends of different diseases after the corresponding disease examination items, disease diagnosis and treatment drugs and disease treatment measures are completed are used as the atlas branch convergence data;

[0067] The blank data map is provided with a plurality of path nodes, and the map main index data, map branch extension data and map branch convergence data corresponding to each medical object are assigned to corresponding path nodes in the blank data map, thereby constructing the diagnostic map path corresponding to each medical object, integrating the diagnostic map paths corresponding to all medical objects, and generating the final medical data map.

[0068] It should be further explained that, in the specific implementation process, the process of periodic information maintenance of medical and health data through medical data graph includes:

[0069] Each diagnostic graph path included in the medical data graph is regarded as an information maintenance object, and a maintenance cycle corresponding to each information maintenance object is set. Each information maintenance object performs corresponding periodic information maintenance under its own maintenance cycle;

[0070] The maintenance content of the periodic information maintenance includes updating the newly generated medical and health data corresponding to each diagnostic map path in the medical data map, deleting expired medical and health data in each diagnostic map path, and modifying erroneous medical and health data in each diagnostic map path.

[0071] It should be further explained that, in the specific implementation process, when a doctor interviews a patient under the corresponding medical object, the process of generating the current patient's interview auxiliary reference information based on the medical data map and pushing the interview auxiliary reference information to the doctor includes:

[0072] Each medical object is equipped with a corresponding doctor, who is responsible for consulting all patients under the corresponding medical object. When the doctor consults the patients for whom he is responsible, he obtains the real-time consultation information corresponding to the patients. The real-time consultation information is obtained by performing a series of medical examinations on the current patients and inquiring about the patients' conditions.

[0073] The multiple diagnosis graph paths included in the medical data graph corresponding to the current medical object are respectively regarded as a matching entity, and the real-time consultation information of each patient is regarded as the element to be matched. The matching entity corresponding to each element to be matched is matched from the medical data graph, and all the medical and health data included in the diagnosis graph path corresponding to the matching entity is regarded as the auxiliary reference information of the consultation of the current patient;

[0074] An electronic health record system, mobile medical applications and communication tools are arranged in the doctor's work scenario for consultation. The generated consultation auxiliary reference information of the patient is pushed to the doctor through the electronic health record system, mobile medical applications and communication tools, and displayed at the doctor in the form of pop-up windows, lists and charts.

[0075] It should be noted that the consultation auxiliary reference information is used to provide data information similar to the condition of the current patient. The doctor corresponding to the patient analyzes the consultation auxiliary reference information and selects some data information that is helpful for diagnosing the patient's condition, thereby improving the consultation efficiency to a certain extent.

[0076] The present invention also provides a medical and health data intelligent management system, the system comprising:

[0077] The medical and health data analysis module is used to obtain the medical and health data collected corresponding to each medical object, and after performing outlier cluster analysis on the medical and health data, mark out some medical and health data whose discrete degree does not meet the clustering expectation, and regard this part of the medical and health data as abnormal data;

[0078] The medical and health data correction module generates an early warning when abnormal data is detected in a medical object and sends it to the data operation and maintenance personnel corresponding to the medical object, who then initiates the data correction process for the abnormal data.

[0079] The medical data map module is used to extract the information characteristics of all medical and health data of each medical object, construct the medical data map of the corresponding medical object according to the information characteristics, and perform periodic information maintenance on the medical and health data through the medical data map;

[0080] The medical consultation auxiliary module generates consultation auxiliary reference information of the current patient based on the medical data map when the doctor is consulting the patient under the corresponding medical object, and pushes the consultation auxiliary reference information to the doctor.

[0081] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A medical and health data intelligent management method, characterized in that: The following steps are involved: Step S1: Obtain the medical and health data collected corresponding to each medical object, and after performing outlier cluster analysis on the medical and health data, mark out some medical and health data whose discreteness does not meet the clustering expectation, and regard this part of the medical and health data as abnormal data; Step S2: When abnormal data is detected in a medical object, an early warning is generated and sent to the data operation and maintenance personnel corresponding to the medical object, and the data operation and maintenance personnel initiates data correction processing for the abnormal data; Step S3: extracting information features of all medical and health data of each medical object, constructing a medical data map of the corresponding medical object according to the information features, and performing periodic information maintenance on the medical and health data through the medical data map; Step S4: When the doctor is interviewing the patient under the corresponding medical object, the consultation auxiliary reference information of the current patient is generated according to the medical data map, and the consultation auxiliary reference information is pushed to the doctor.

2. A medical and health data intelligent management method according to claim 1, characterized in that: The process of obtaining the medical and health data collected corresponding to each medical object includes: The types of medical objects include hospitals, medical institutions and community health service stations; Each medical object is numbered and denoted as i, then i=1, 2, 3, ..., n, where n is a natural number greater than 0. A corresponding medical data storage terminal is set at each medical object, and a data collection period and a data collection frequency are set for each medical object. During the data collection period, each medical object transmits the corresponding medical and health data obtained by itself to its own medical data storage terminal according to the corresponding data collection frequency for storage, and the medical and health data corresponding to the medical object numbered i is denoted as Data[i].

3. A medical and health data intelligent management method according to claim 2, characterized in that: The process of marking some medical and health data whose discrete degree does not meet the clustering expectation after outlier cluster analysis on medical and health data and treating this part of medical and health data as abnormal data includes: The medical and health data corresponding to each medical object is divided into several data clusters, a data monitoring node is set for each data cluster, the data cluster is converted into a text document and entered into the corresponding data monitoring node, and the data monitoring node obtains the representation vectors of several words and sentences included in each data cluster; Obtain the cosine similarity value between the representation vector corresponding to each word and sentence and the preset reference vector, and record the cosine similarity value as COSθ, set a first determination threshold corresponding to the cosine similarity value COSθ, and record the first determination threshold as μ, record the number of words and sentences corresponding to COSθ≤μ in each data cluster as N1, and record the number of words and sentences corresponding to COSθ>μ in each data cluster as N2; Generate a discrete coefficient corresponding to each data cluster according to N1 and N2, record the discrete coefficient corresponding to the data cluster numbered i as τ[i], then τ[i]=N1 / (N1+N2), set a second determination threshold corresponding to the discrete coefficient, and record the second determination threshold as η; When τ[i]≥η, some medical and health data represented by the data cluster corresponding to the current discrete coefficient τ[i] are marked as having a discrete degree that does not meet the clustering expectations, and some medical data that do not meet the clustering expectations are regarded as abnormal data. When τ[i]<η, it means that the discrete degree of the medical and health data represented by the data cluster corresponding to the current discrete coefficient meets the clustering expectations, and no operation is performed.

4. A medical and health data intelligent management method according to claim 3, characterized in that: When abnormal data is detected in a medical object, an early warning is generated and sent to the data operation and maintenance personnel corresponding to the medical object. The data operation and maintenance personnel initiate the data correction process for the abnormal data, including: The data synchronization monitoring is performed according to the preset monitoring frequency and monitoring steps, wherein the monitoring step is composed of a number of sub-execution steps. When each sub-execution step is performed according to the preset execution order, it indicates that the data synchronization monitoring is in a correct execution state and no operation is performed. When any sub-execution step in the monitoring step is not performed according to the preset execution order, it is marked that the data synchronization monitoring is in an abnormal execution state. For data synchronization monitoring in an abnormal execution state, the monitoring frequency and monitoring steps are reset, and then the data synchronization monitoring is re-executed. When abnormal data is detected in the corresponding medical object, an early warning prompt corresponding to the abnormal data is generated, and the early warning prompt is sent to the data operation and maintenance personnel corresponding to the medical object represented by the current abnormal data, and the data operation and maintenance personnel determine whether the abnormal data can be corrected; If so, the abnormal data is input into a preset parameter library, and the abnormal correction parameters corresponding to the abnormal data are matched from the parameter library, and then the abnormal data is converted into corresponding normal medical and health data through the abnormal correction parameters. If not, the abnormal data is directly discarded.

5. A medical and health data intelligent management method according to claim 4, characterized in that: The process of extracting the information features of all medical and health data of each medical object and constructing the medical data map of the corresponding medical object according to the information features includes: When all medical and health data of each medical object do not have abnormalities, an information extraction period is set for each medical object, and an initial blank data map is established. During the information extraction period, all medical and health data of each medical object are processed by big data technology to obtain information features corresponding to each medical object. The information features include different disease types, disease examination items, disease diagnosis and treatment drugs and disease treatment measures corresponding to different disease types, and also include the disease progression trend of the corresponding disease after the corresponding disease examination items, disease diagnosis and treatment drugs and disease treatment measures of different disease types are completed; Among them, different disease types are used as the main index data of the atlas, the disease examination items, disease diagnosis and treatment drugs and disease treatment measures corresponding to different disease types are used as the atlas branch extension data, and the disease progression trend of different diseases after the execution of the corresponding disease examination items, disease diagnosis and treatment drugs and disease treatment measures is used as the atlas branch convergence data. A number of path nodes are set in the blank data atlas, and the main index data of the atlas, the atlas branch extension data and the atlas branch convergence data corresponding to each medical object are assigned to corresponding path nodes in the blank data atlas, and then the diagnostic atlas path corresponding to each medical object is constructed, and the diagnostic atlas paths corresponding to all medical objects are integrated to generate the final medical data atlas.

6. A medical and health data intelligent management method according to claim 5, characterized in that: The process of periodic information maintenance of medical and health data through medical data graphs includes: Each diagnostic map path included in the medical data map is regarded as an information maintenance object, and a corresponding maintenance cycle is set for each information maintenance object. Each information maintenance object performs corresponding periodic information maintenance under its own maintenance cycle. The maintenance content of the periodic information maintenance includes updating the newly generated medical and health data corresponding to each diagnostic map path in the medical data map, deleting the expired medical and health data in each diagnostic map path, and modifying the erroneous medical and health data in each diagnostic map path.

7. A medical and health data intelligent management method according to claim 6, characterized in that: When a doctor is consulting a patient under a corresponding medical object, the process of generating auxiliary reference information of the current patient's consultation based on the medical data map and pushing the auxiliary reference information to the doctor includes: Each medical object is assigned a corresponding doctor, who is responsible for consulting all patients under the corresponding medical object. When the doctor consults the patients he is responsible for, he obtains the real-time consultation information of the patients. The real-time consultation information is obtained by performing a series of medical examinations on the current patients and asking about the patients' condition. The multiple diagnosis graph paths included in the medical data graph corresponding to the current medical object are respectively regarded as a matching entity, and the real-time consultation information of each patient is regarded as the element to be matched. The matching entity corresponding to each element to be matched is matched from the medical data graph, and all the medical and health data included in the diagnosis graph path corresponding to the matching entity is regarded as the auxiliary reference information of the consultation of the current patient; An electronic health record system, mobile medical applications and communication tools are arranged in the doctor's work scenario for consultation, and the generated consultation auxiliary reference information of the patient is pushed to the doctor for display through the electronic health record system, mobile medical applications and communication tools.

8. A medical and health data intelligent management system, used to implement the data intelligent management method according to any one of claims 1 to 7, characterized in that: The system includes: The medical and health data analysis module is used to obtain the medical and health data collected corresponding to each medical object, and after performing outlier cluster analysis on the medical and health data, mark out some medical and health data whose discrete degree does not meet the clustering expectation, and regard this part of the medical and health data as abnormal data; The medical and health data correction module generates an early warning when abnormal data is detected in a medical object and sends it to the data operation and maintenance personnel corresponding to the medical object, who then initiates the data correction process for the abnormal data. The medical data map module is used to extract the information characteristics of all medical and health data of each medical object, construct the medical data map of the corresponding medical object according to the information characteristics, and perform periodic information maintenance on the medical and health data through the medical data map; The medical consultation auxiliary module generates consultation auxiliary reference information of the current patient based on the medical data map when the doctor is consulting the patient under the corresponding medical object, and pushes the consultation auxiliary reference information to the doctor.

Images