An electronic information management system for medical and nursing archives

Abstract

The present invention relates to the field of medical care information technology, and specifically to an electronic information management system for medical and nursing archives, comprising: a first processing end, obtaining the similarity of work skills between target objects, performing a first clustering using the similarity of work skills, determining a plurality of first clustering clusters, obtaining the similarity of capabilities between target objects in each first clustering cluster, and then performing a second clustering of the target objects in each first clustering cluster using the similarity of capabilities, and determining a plurality of second clustering clusters. By clustering the personal information archive data of medical and nursing personnel according to the similarity of capabilities, and determining the medical and nursing archive word segmentation score value of each target object in each of the second clustering clusters, the personalized training word segmentation of medical and nursing personnel in each cluster is calculated, and then a target training method is formulated for the medical and nursing personnel in each cluster according to the personalized training word segmentation of medical and nursing personnel in each cluster, so as to eliminate the isolation between the archive data of medical and nursing personnel and improve management efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of medical care information, and in particular to an electronic information management system for medical care archives. Background Art

[0002] Medical and nursing archives work is a basic work of medical archives. It refers to the archives used in medical institutions to record and manage medical staff's personal information, reports during treatment or nursing, and other related materials. The electronic information management system for medical and nursing archives refers to an information system that uses computer software, hardware and network technology to collect, store, manage and apply medical staff's personal information, professional skills, work experience, scientific research results and other data. It is an important part of the information construction of the medical industry and is of great significance for standardizing medical and nursing management, optimizing resource allocation, and improving the level of medical services.

[0003] The existing electronic information management system for medical and nursing archives is mainly used in scenarios such as personnel file management and training management. In addition, the system can also provide a reference for talent training and provide data decision-making for talent support strategies. However, when conducting talent training analysis based on the electronic information management system for medical and nursing archives, the data between the business systems are isolated, that is, the data between the business systems are not connected, which affects management efficiency and the personalized training of medical staff. Summary of the invention

[0004] In order to solve the technical problem that data silos exist between business systems of the electronic information management system for medical and nursing archives, which affects the talent training analysis of medical staff, the purpose of the present invention is to provide an electronic information management system for medical and nursing archives, which adds clustering of the similarity of work skills of medical staff and performs multiple clustering analyses, obtains the medical and nursing archive word segmentation score of the medical staff based on multiple clustering analyses, determines the target training method, and eliminates the silos between the archive data of medical staff to effectively carry out personalized training of medical staff.

[0005] The technical solution adopted is specifically as follows: a medical and nursing file electronic information management system is provided, which includes: a first processing end, obtaining the work skill similarity between target objects, using the work skill similarity to perform a first clustering to determine multiple first clustering clusters, obtaining the ability similarity between the target objects in each of the first clustering clusters, and then using the ability similarity to cluster the target objects in each of the first clustering clusters again to determine multiple second clustering clusters; a second processing end, obtaining the medical and nursing file word segmentation score value of the target object in each of the second clustering clusters, using the medical and nursing file word segmentation score value to determine the target training method of the target object in each of the second clustering clusters, and then using the target training method to train the target object.

[0006] In one embodiment of the present invention, the first processing end includes: a first similarity determination module, which obtains basic information of the target object, uses the basic information to establish a basic information feature set, and then determines the work skill similarity between different target objects; a first clustering module, which uses the work skill similarity to perform a first clustering of the target objects to obtain a plurality of first clustering clusters; a second similarity determination module, which obtains the prescription ability similarity between different target objects in each of the first clustering clusters, and obtains the professional skill similarity between different target objects, and uses the prescription ability similarity and the professional skill similarity to determine the ability similarity between different target objects in the first clustering cluster; a second clustering module, which uses the ability similarity to cluster the target objects in the first clustering cluster again to obtain a plurality of second clustering clusters.

[0007] In one embodiment of the present invention, the basic information of the target object is obtained, and a basic information feature set is established using the basic information to further determine the similarity of work skills between different target objects, including: obtaining the basic information corresponding to each target object, and then establishing a basic information feature set corresponding to each target object using the basic information; wherein the basic information at least includes practice scope data, practice time data, and technical certificate data; extracting a first basic information feature set of any one of the target objects, and a second basic information feature set of any other target object; and using the first basic information feature set and the second basic information feature set to determine the similarity of work skills between two different target objects.

[0008] In one embodiment of the present invention, the second similarity determination module includes: a first sub-similarity determination module, which obtains the prescription right category and the number of prescription rights of the target object in the first cluster, and uses the prescription right category and the prescription right number to determine the prescription ability similarity between different target objects in each of the first cluster; a second sub-similarity determination module, which obtains the operation level and the operation number of the target object in the first cluster, and uses the operation level and the operation number to determine the professional skill similarity between different target objects in each of the first cluster; a third sub-similarity determination module, which uses the prescription ability similarity and the professional skill similarity to determine the ability similarity between different target objects in each of the first cluster.

[0009] In one embodiment of the present invention, the step of obtaining the prescription right category and the number of prescription rights of the target object in the first cluster, and using the prescription right category and the number of prescription rights to determine the prescription ability similarity between different target objects in each of the first clusters, includes: obtaining the first prescription right category and the first prescription right number of any one of the target objects in the first cluster, and obtaining the second prescription right category and the second prescription right number of another target object in the first cluster; and using the first prescription right category, the first prescription right number, the second prescription right category and the second prescription right number to determine the prescription ability similarity between different target objects in each of the first clusters.

[0010] In one embodiment of the present invention, the step of obtaining the operation level and the number of operations of the target object in the first cluster, and determining the similarity of professional skills between different target objects in each of the first clusters by using the operation level and the number of operations, includes: obtaining the first operation level and the first operation number of any one of the target objects in the first cluster, and obtaining the second operation level and the second operation number of another target object in the first cluster; and determining the similarity of professional skills between different target objects in each of the first clusters by using the first operation level, the first operation number, the second operation level and the second operation number.

[0011] In one embodiment of the present invention, the second processing end includes: a score determination module, which obtains the medical and nursing file segmentation score value of each target object in each of the second clusters; a training method determination module, which uses the medical and nursing file segmentation score value to determine the target training method of the target object in each of the second clusters, and then uses the target training method to train the target object.

[0012] In one embodiment of the present invention, the step of obtaining the medical file segmentation score value of each target object in each of the second clustering clusters includes: obtaining the score value of the target segmentation in the medical file corresponding to each of the target objects in each of the second clustering clusters, and obtaining the first occurrence number of the target segmentation; and determining the medical file segmentation score value of each of the target objects in each of the second clustering clusters using the score value and the first occurrence number.

[0013] In one embodiment of the present invention, the use of the medical file segmentation score value to determine the target training method of the target object in each of the second clustering clusters includes: sorting the medical file segmentation score values, and determining the daily experience segmentation array of each of the target objects based on the sorting result; obtaining the total number of the target objects in the second clustering cluster, and obtaining the second occurrence number of the target daily experience segmentation in the daily experience segmentation array of the target object; using the total number and the second occurrence number to determine the possibility of personalized training segmentation; using the personalized training segmentation possibility to determine the target training method of the target object in each of the second clustering clusters.

[0014] In one embodiment of the present invention, the use of the target training method to train the target object includes: using prescription rights to divide the target object into a first target object and a second target object; using the target training method to train the first target object; using the prescription rights to determine the second training method, and then using the second training method to train the second target object.

[0015] The beneficial effects of the present invention are as follows: the electronic information management system for medical and nursing archives of the present invention comprises: a first processing end, obtaining the similarity of work skills between target objects, performing a first clustering using the similarity of work skills, determining a plurality of first clustering clusters, obtaining the similarity of ability between the target objects in each of the first clustering clusters, and then performing a second clustering of the target objects in each of the first clustering clusters using the similarity of ability, and determining a plurality of second clustering clusters; a second processing end, obtaining the medical and nursing archive word segmentation score value of the target object in each of the second clustering clusters, and determining a plurality of second clustering clusters using the medical and nursing archive word segmentation score value. - A target training method for the target object in the second cluster, and then using the target training method to perform training on the target object; that is, the present application clusters the personal information file data of medical staff in the medical file electronic information management system according to ability similarity, and determines the medical file segmentation score value of each target object in each second cluster, thereby calculating the personalized training segmentation of the medical staff in each cluster, and then formulating a target training method for the medical staff in each cluster based on the personalized training segmentation of the medical staff in each cluster, so as to eliminate the isolation between the medical staff file data and improve management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are 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.

[0017] Figure 1 A schematic diagram of the structure of a medical and nursing archive electronic information management system provided by the first embodiment of the present invention.

[0018] Figure 2 A schematic diagram of the structure of a medical record electronic information management system provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a medical and nursing record electronic information management system proposed by the present invention in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0020] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0021] The specific scheme of the electronic information management system for medical records provided by the present invention is described in detail below with reference to the accompanying drawings.

[0022] See also Figure 1 , which shows a schematic diagram of the structure of a medical record electronic information management system provided by an embodiment of the present invention.

[0023] like Figure 1 As shown, a medical record electronic information management system 10 includes: a first processing end 100 and a second processing end 200.

[0024] The first processing end 100 obtains the similarity of work skills between target objects, performs the first clustering using the similarity of work skills, determines a plurality of first clustering clusters, obtains the similarity of capabilities between target objects in each first clustering cluster, and then uses the capability similarity to cluster the target objects in each first clustering cluster again, and determines a plurality of second clustering clusters; the second processing end 200 obtains the medical and nursing file segmentation score value of the target object in each second clustering cluster, uses the medical and nursing file segmentation score value to determine the target training method of the target object in each second clustering cluster, and then uses the target training method to train the target object.

[0025] Among them, the electronic information management system is a system that uses computer software, hardware and network technology to collect, store, manage and apply the archival information of medical staff. Through such a system, the system aims to improve the efficiency and security of archival management, while helping to analyze the capabilities of medical staff and provide personalized training suggestions. The target objects refer to medical staff, including doctors and nurses.

[0026] Medical and nursing records are records of medical staff's personal information, including personal information files, prescription rights, professional and technical medical and nursing record files, historical treatment report files, and scientific research results files. The focus of each type of file is different. Personal information files include name, gender, department, graduation school, major, qualification certificate number, scope of practice, technical certificate, work experience, etc. Prescription rights are used to record the prescription rights of medical staff, such as general prescription rights, anesthesia prescription rights, antibiotic prescription rights, acquisition time, etc. Antibiotic prescription rights are divided into different levels, such as: non-restricted level, restricted level, special practical level. Professional and technical conditions include medical staff's surgical ability, technical expertise, professional group, etc. Surgical projects are divided into three different levels: primary, secondary and tertiary.

[0027] The collection of medical records in the embodiments of the present invention is authorized by relevant users, and the collection process does not violate relevant laws and regulations and does not violate public order and good customs.

[0028] In addition, work skill similarity refers to the similarity of work skills among different medical staff, which is used for the first clustering grouping. Ability similarity is to further evaluate the similarity of medical staff in ability on the basis of the first clustering, which is used for a more refined second clustering. The medical and nursing file segmentation score refers to the personalized training of medical staff through the medical and nursing file segmentation score when processing and analyzing medical and nursing files. The first clustering cluster divides medical staff into multiple preliminary clusters based on the similarity of work skills. The second clustering cluster is based on the first clustering, and each cluster is further subdivided according to the ability similarity to form multiple more detailed clusters. The medical and nursing file segmentation score is the score obtained after text analysis of medical and nursing files, which is used to evaluate the relevance of the medical staff's file information in a specific field. The target training method is a personalized training plan or method formulated based on the second clustering cluster where the medical staff is located.

[0029] Specifically, the first processing end is responsible for obtaining the similarity of work skills between target objects (i.e., medical staff), and based on this similarity, the first clustering is performed to form multiple preliminary first clusters; then, for each target object in the first cluster, its ability similarity is further obtained, and these ability similarities are used to perform secondary clustering on the target objects in each first cluster to determine multiple more detailed second clusters. The second processing end is used to obtain the medical and nursing file segmentation score value of the target object in each second cluster, and the personalized training method of the target object in each second cluster is determined based on these score values, and specific training operations are performed on the target object accordingly.

[0030] In this embodiment, the first clustering cluster is determined by roughly grouping according to the similarity of work skills, and the second clustering cluster further optimizes the classification of the first clustering cluster through more detailed capability analysis. Then, the medical and nursing archive word segmentation score value of each target object in each second clustering cluster is determined by the second processing end, so as to calculate the personalized training word segmentation of medical and nursing staff in each cluster, and then formulate a targeted training method according to the personalized training word segmentation of medical and nursing staff in each cluster, so as to eliminate the isolation between the archive data of medical and nursing staff and improve the pertinence and effect of improving the ability of medical and nursing staff, and improve the efficiency of archive management, and promote the personalized career development of medical and nursing staff.

[0031] See also Figure 2 , Figure 2 It is a structural diagram of a medical record electronic information management system provided in the second embodiment of the present application.

[0032] The first processing end 100 may include:

[0033] The first similarity determination module 110 obtains basic information of the target object, establishes a basic information feature set using the basic information, and further determines the similarity of work skills between different target objects.

[0034] Among them, obtaining the basic information of the target object refers to extracting basic information from the files of each target object, which at least includes the scope of practice, practice time, other relevant technical certificates, prescription rights, and professional and technical columns. These basic information comprehensively reflects the work skills of medical staff. Among them, prescription rights and professional and technical columns can specifically show the doctor's expertise and work proficiency. The similarity of work skills can accurately reflect the degree of overlap between different target objects in skills and professional fields. This process lays the foundation for subsequent cluster analysis and personalized training.

[0035] Specifically, the first similarity determination module 110 obtains basic information of the target object, establishes a basic information feature set using the basic information, and then determines the similarity of work skills between different target objects, as follows:

[0036] First, basic information corresponding to each target object is obtained, and then a basic information feature set corresponding to each target object is established using the basic information.

[0037] Among them, basic information at least includes scope of practice, practice time, other relevant technical certificates, prescription rights, and professional and technical columns, etc. The basic information feature set may include practice scope data, practice time data, and technical certificate data.

[0038] Secondly, a first basic information feature set of any target object and a second basic information feature set of any other target object are extracted.

[0039] Among them, the first basic information feature set and the second basic information feature set both include at least practice scope data, practice time data and technical certificate data.

[0040] Specifically, taking the personal information file of a doctor as an example, the indicator data of "scope of practice", "practice time" and "other relevant technical certificates" in the personal basic information column can effectively reflect the doctor's work skills. At the same time, the data in the "prescription rights" and "professional skills" columns can show the doctor's areas of expertise and work proficiency. Therefore, when analyzing the relevance of the basic information of different doctors, it is necessary to quantify it based on these indicators. Specifically, for the personal basic information column in the personal information file of each doctor, the data items of "scope of practice", "practice time" and "other relevant technical certificates" can be used to form a personal basic information feature set. , for example, , the first data in this set indicates that the doctor's practice scope is internal medicine, and the second data indicates that the doctor's practice time is June 15, 2011. The third data indicates that the doctor has also obtained the attending physician qualification certificate. Through the quantitative analysis of these characteristic data, the working skills of doctors can be more accurately evaluated and compared.

[0041] Finally, the first basic information feature set and the second basic information feature set are used to determine the similarity of work skills between two different target objects.

[0042] Based on the first basic information feature set, the second basic information feature set and the following formula, the similarity of work skills between two different target objects is determined:

[0043]

[0044] In the formula, Represents the similarity of job skills between the p-th target object and the q-th target object. It represents the calculation of Jaccard correlation coefficient. Jaccard correlation coefficient (Jaccard Similarity Coefficient) is used to measure the similarity between two target objects. It is a well-known technology in the art and will not be described in detail. Represents the first basic information feature set of the p-th target object. Represents the second basic information feature set of the qth target object. The larger the value of , the higher the similarity of the work skills between the p-th target object and the q-th target object.

[0045] The first clustering module 120 performs a first clustering on the target objects by using the similarity of work skills to obtain a plurality of first clusters.

[0046] Among them, the first clustering module performs the first clustering of the target objects, and groups the target objects by using the similarity of work skills to obtain multiple first clustering clusters. Specifically, this process is based on the similarity of work skills and uses the K-means method for clustering. K-means is a commonly used clustering algorithm, whose goal is to divide the work skill similarity data into K clusters, so that the work skill similarity within the cluster is as similar as possible, while the work skill similarity between clusters is as different as possible.

[0047] Specifically, the K-means method is used to perform the first clustering of all target objects. Here, the distance between different target objects is quantified based on the similarity of work skills calculated in advance. The number K of the first clustering clusters is determined by the silhouette coefficient method (Silhouette Score), which can evaluate the clustering effect under different K values ​​and help select the optimal number of clusters. Therefore, all target objects will be divided into multiple first clustering clusters, and the target objects in each first clustering cluster have a high similarity in basic work skills. Among them, the silhouette coefficient method (Silhouette Score) is well known to those skilled in the art, and the detailed process of the first clustering of all target objects will not be described in detail.

[0048] Furthermore, the second similarity determination module 130 obtains the prescription ability similarity between different target objects in each first cluster, and obtains the professional skill similarity between different target objects, and determines the ability similarity between different target objects in the first cluster using the prescription ability similarity and the professional skill similarity.

[0049] Among them, the similarity of prescription ability is used to measure the similarity of prescription ability among different target objects (medical staff). Specifically, the similarity of prescription ability evaluates the ability of medical staff in prescribing, including their proficiency in handling various prescriptions, the level of prescription authority (such as general prescriptions, anesthetic prescriptions, antibiotic prescriptions, etc.), and the complexity and diversity of prescriptions. A high similarity of prescription ability indicates that the medical staff's ability in the field of prescription is relatively close. The similarity of professional skills is used to measure the similarity of professional skills among different target objects. The similarity of professional skills evaluates the skill level and expertise of medical staff in their professional field, including the types of surgeries that can be performed, technical expertise, and professional groups involved. A high similarity of professional skills indicates that medical staff are relatively similar in professional skills and technical levels. Ability similarity is used to combine the similarity of prescription ability and professional skills to reflect the similarity of overall ability among different target objects. Ability similarity combines the similarity of medical staff in prescription ability and professional skills to provide a comprehensive comparison of the target objects' abilities. A high similarity of ability indicates that these target objects have a high degree of similarity in the overall level of their work ability.

[0050] That is, the second similarity determination module can more accurately evaluate the overall similarity in ability of these target objects by determining the similarity in prescription ability and professional skills of the target objects in each first cluster, thereby providing basic data for the subsequent second clustering and personalized training.

[0051] Further, see Figure 2, the second similarity determination module 130 may include: a first sub-similarity determination module, a second sub-similarity determination module and a second sub-similarity determination module.

[0052] The first sub-similarity determination module obtains the prescription right category and the number of prescription rights of the target object in the first cluster, and determines the prescription ability similarity between different target objects in each first cluster by using the prescription right category and the number of prescription rights.

[0053] Among them, the prescription right category can be divided into general prescription right and antimicrobial prescription right. The scope of capabilities involved in general prescription right and antimicrobial prescription right is different; general prescription right usually refers to the ability to prescribe general drugs, not special controlled drugs. For antimicrobial prescription right, specific levels need to be considered to more accurately evaluate the similarities between them.

[0054] In addition, the right to prescribe antibiotics can be divided into three levels: unrestricted level, restricted level and special use level. The unrestricted level means that common antibiotics can be prescribed, but there are no additional restrictions on their use. The restricted level refers to specific restrictions on the use of antibiotics, such as special approval or conditions. The special use level refers to antibiotics used in special circumstances, usually for specific diseases or with special requirements.

[0055] That is, the first sub-similarity determination module is used to calculate the similarity according to the prescription right category and quantity of the target object in each first cluster. For example, if two objects have the same category of prescription rights and a similar number, their prescription ability similarity will be high.

[0056] Specifically, first, a first prescription right category and a first prescription right quantity of any target object in the first cluster are obtained, and a second prescription right category and a second prescription right quantity of another target object in the first cluster are obtained.

[0057] Among them, the prescription rights referred to in the first prescription right and the second prescription right have been described in detail above and will not be repeated here.

[0058] Secondly, the first prescription right category, the first prescription right quantity, the second prescription right category and the second prescription right quantity are used to determine the prescription ability similarity between different target objects in each first cluster.

[0059] Among them, prescription ability similarity is used to measure the similarity of the target object's ability or proficiency in prescription.

[0060] The calculation process for determining the prescription ability similarity between different target objects in each first cluster is as follows:

[0061]

[0062] In the formula, represents the prescription ability similarity between the p-th target object and the q-th target object. Indicates the number of first prescription rights possessed by the p-th target object. Indicates the number of second prescription rights possessed by the qth target object.

[0063] Indicates Jaccard correlation coefficient calculation. Indicates the number of categories contained in the first prescription right of the pth target object. Indicates the number of categories included in the second prescription right of the qth target object. is to avoid the denominator being 0. The smaller the value of is, the more similar the number of prescription rights of the pth target object and the qth target object is. The larger the value of is, the closer the number of categories contained in the prescription rights of the p-th target object and the q-th target object are. The larger the value of , the greater the similarity of prescription ability between the p-th target object and the q-th target object.

[0064] Furthermore, the second sub-similarity determination module obtains the operation level and the operation quantity of the target objects in the first clusters, and determines the professional skill similarity between different target objects in each first cluster using the operation level and the operation quantity.

[0065] The operation level refers to the level of surgery or operation that the target object can perform. It is usually divided into multiple levels. The operation level represents the professional skill level of the physician. The higher the level, the more complex and difficult the physician can undertake. For example, level one, level two, or level three. The number of operations refers to the number of surgeries or operations performed by the target object at different operation levels. Professional skill similarity is used to measure the similarity of the target objects in terms of professional skills.

[0066] Specifically, first obtain the first operation level and the first operation quantity of any target object in the first cluster, and obtain the second operation level and the second operation quantity of another target object in the first cluster; secondly, use the first operation level, the first operation quantity, the second operation level and the second operation quantity to determine the similarity of professional skills between different target objects in each first cluster.

[0067] Among them, the target objects with the same or similar operation levels have higher similarity in professional skills. The first operation level refers to the operation level of the first target object in the first cluster. The second operation level refers to the operation level of another target object in the first cluster. The first operation quantity refers to the first operation level quantity of the first target object in the first cluster. It may represent the frequency or total amount of the first operation level performed by the target object in the first cluster, reflecting the degree of accumulation of its experience. The first operation quantity refers to the first operation level quantity of the first target object in the first cluster. It may represent the frequency or total amount of the first operation level performed by the target object in the first cluster, reflecting the degree of accumulation of its experience. The second operation quantity refers to the second operation level quantity of another target object in the first cluster. It may represent the frequency or total amount of the second operation level performed by the target object, reflecting the degree of accumulation of its experience.

[0068] That is, by comparing the operation levels and operation quantities of the target objects in each first cluster, we can derive their similarity in professional skills, which is used to measure the similarity of the target objects in skill level and experience, and further helps to understand the skill backgrounds of different target objects for more effective clustering and analysis.

[0069] The similarity of professional skills between different target objects in each first cluster can be calculated by the following formula:

[0070]

[0071] In the formula, Represents the professional skills similarity between the p-th target object and the q-th target object. Indicates the number of operation levels. Indicates the number of first operations of the p-th target object at the i-th first operation level. Indicates the number of second operations of the qth target object at the ith second operation level. +1 is to avoid the denominator being 0, where a larger value of i indicates a higher operation level. The smaller the value of , the more similar the number of operation levels that the p-th target object and the q-th target object can perform in the i-th operation level.

[0072] In addition, if a target object can perform high-level operations, it means that it can also perform low-level operations, because high-level operations usually require the target object to have more comprehensive knowledge, superb skills and rich clinical experience. Therefore, the target object who can complete high-level operations usually has the ability and qualifications to handle low-level operations. Therefore, the higher the weight of the professional skills similarity between the p-th target object and the q-th target object in the high-level operation, the more To quantify. In addition, The larger the value of , the higher the similarity of professional skills between the p-th target object and the q-th target object.

[0073] The third sub-similarity determination module determines the ability similarity between different target objects in each first cluster using the prescription ability similarity and the professional skill similarity.

[0074] Among them, ability similarity refers to the similarity of the comprehensive abilities between different target objects within a cluster.

[0075] Specifically, based on the prescription ability similarity and the professional skill similarity, the ability similarity between different target objects in each first cluster can be calculated by the following formula:

[0076]

[0077] In the formula, Represents the ability similarity between the p-th target object and the q-th target object. represents the prescription ability similarity between the p-th target object and the q-th target object. Represents the professional skills similarity between the p-th target object and the q-th target object. The larger the value of , the higher the ability similarity between the p-th target object and the q-th target object.

[0078] The second clustering module 140 clusters the target objects in the first clusters again by using the capability similarity to obtain a plurality of second clusters.

[0079] The target objects in the first cluster can be clustered again using the K-means method, which is a common clustering algorithm that performs clustering by allocating data to a predetermined number of clusters to minimize the distance from the points within the cluster to the cluster center.

[0080] In addition, the silhouette coefficient method can be used to determine the number of the second cluster (i.e., the K value). The silhouette coefficient method is a method for evaluating clustering quality, which helps determine the optimal number of clusters by calculating the silhouette coefficient of each data point. The silhouette coefficient reflects the difference between the similarity between the data point and its cluster and the similarity between other clusters.

[0081] After the second clustering, the target objects in each second cluster have higher ability similarity. This means that the target objects within each second cluster are relatively close in comprehensive ability and have high internal consistency. The number of second clusters finally obtained is N, and the target objects in each cluster are relatively similar in ability. In this way, through hierarchical clustering, the target objects can be better organized at a detailed level, providing a more accurate basis for subsequent analysis and decision-making.

[0082] Specifically, in the first cluster, the target object is clustered for the second time based on the ability similarity, and each cluster is further refined to analyze the similarity of the target object in more detail.

[0083] Continue reading Figure 2 , Figure 2 It is a structural diagram of a medical record electronic information management system provided in the second embodiment of the present application.

[0084] The second processing end 200 may include:

[0085] The score determination module 210 obtains the medical record word segmentation score of each target object in each second cluster.

[0086] Among them, medical and nursing records usually refer to the historical treatment report files and scientific research results files of the patients treated by the target object. Among them, the historical treatment report files are systematic documents that record the patient's health status, medical history, treatment process, nursing measures and other information.

[0087] Specifically, obtaining the medical file word segmentation score value of each target object in each second cluster may include the following contents:

[0088] First, the score value of the target word in the medical file corresponding to each target object in each second cluster is obtained, and the first occurrence number of the target word is obtained.

[0089] Target segmentation usually refers to some key words or terms in historical treatment report archives and scientific research results archives, which are essential for understanding the archive content and conducting analysis. For example, treatment methods, drug names or treatment processes in historical treatment report archives, such as treatment processes such as "drug treatment" and "surgical treatment"; drug names such as "amoxicillin" and "reserpine"; treatments such as "chemotherapy" and "physical therapy".

[0090] Secondly, the medical file word segmentation score of each target object in each second cluster is determined by using the score value and the first occurrence number.

[0091] Specifically, the medical record word segmentation score of each target object in each second cluster is calculated as follows:

[0092]

[0093] In the formula, Represents the score value of the tth target word of the fth target object. Indicates the number of times the tth participle of the fth target object appears in all participles. Represents the number of all segmented words in all segmented word arrays in the historical treatment report files and scientific research results files of the target object. The larger the value of , the larger the score value of the tth word of the fth target object.

[0094] That is, for the target objects in the nth second cluster, assuming there are F in total, the JieBa word segmentation tool can be used to segment the historical treatment report files and scientific research results files of each target object, and the historical treatment report file word segmentation array and the scientific research results file word segmentation array can be obtained. Then, the frequency of occurrence of each word in the word segmentation array of all historical treatment report files of each doctor is counted to determine the score value of each word.

[0095] The training method determination module 220 determines the target training method of the target object in each second cluster using the medical file word segmentation score value, and then trains the target object using the target training method.

[0096] Among them, the target training method refers to the personalized training strategy for the target object.

[0097] Specifically, using the medical and nursing archive word segmentation score value, determining the target training method of the target object in each second cluster may include the following contents:

[0098] First, the medical and nursing file word segmentation score values ​​are sorted, and the daily experience word segmentation array of each target object is determined based on the sorting results.

[0099] Among them, the sorting method is to sort all medical and nursing file word segmentation score values ​​in order from large to small.

[0100] Specifically, the medical and nursing archive word segmentation scores in each second cluster can be sorted and ranked first. The segmented words are recorded as the daily experience segmented word array of the f-th target object. Similarly, the daily experience segmented word arrays of the other F-1 target objects in each second cluster can be calculated.

[0101] Secondly, the total number of target objects in the second cluster is obtained, and the second occurrence number of the target daily experience segmentation word in the daily experience segmentation word array of the target object is obtained.

[0102] Next, the total number and the second occurrence number are used to determine the possibility of personalized training segmentation; and the target training method of the target object in each second cluster is determined using the personalized training segmentation possibility.

[0103] The probability of personalized training segmentation is calculated by the following formula:

[0104]

[0105] In the formula, Indicates the possibility that the vth daily experience segmentation is a personalized training segmentation. Represents the number of times the vth daily experience word appears in the daily experience word array of all F target objects. Represents the number of all target objects in the cluster. The larger the value of , the more times the v-th daily experience segmentation word appears in the daily experience segmentation word array of all F target objects, indicating that the segmentation word will be used in the work of multiple target objects, and the less necessary it is in the subsequent personalized training, and the smaller the possibility that the v-th daily experience segmentation word is a personalized training segmentation word.

[0106] In addition, the probability of each daily experience segmentation being a personalized training segmentation is calculated, and the probability values ​​of all segmentations in daily experience are sorted. The segmentations with a value greater than 0.6 are regarded as personalized training segmentations, and these personalized training segmentations are considered to have high relevance and application value in personalized training. Based on the screened personalized training segmentations, managers formulate corresponding personalized training plans for each cluster (i.e., each group of target objects). According to the formulated personalized training plan, targeted training is carried out for the target objects. The training content should be combined with the actual needs and problems represented by the personalized training segmentations to improve the effectiveness and pertinence of the training. For personnel in other clusters, the above method can also be used to screen personalized segmentations and formulate training plans to ensure that the target objects of each cluster can obtain personalized training that suits their needs. Through this process, managers can effectively formulate targeted training plans for target objects of different clusters, thereby improving the accuracy and effectiveness of overall training.

[0107] Furthermore, the target object is trained using the target training method, including: using prescription rights to divide the target object into a first target object and a second target object; using the target training method to train the first target object; using prescription rights to determine the second training method, and then using the second training method to train the second target object.

[0108] Among them, the first target object refers to doctors, and the second target object refers to nurses.

[0109] Specifically, since the nurse does not have the right to prescribe, the nurse's prescription ability is not calculated, and the prescription right is used to divide the target object into a first target object and a second target object. Adopting a personalized target training method is to use the target training method to train the first target object. Here, since the nurse does not have the right to prescribe, the influence of the prescription right is excluded, and the nurse's prescription ability is not calculated. Other calculation methods are the same and will not be repeated here to determine the second training method, and use the second training method to train the second target object. This classification and training method ensures that doctors and nurses respectively receive training programs suitable for their roles, thereby improving the overall training effect.

[0110] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A medical and nursing archive electronic information management system, characterized in that: The electronic information management system for medical and nursing records includes: The first processing end obtains the work skill similarity between the target objects, performs a first clustering using the work skill similarity to determine a plurality of first clusters, obtains the ability similarity between the target objects in each of the first clusters, and then performs a second clustering on the target objects in each of the first clusters using the ability similarity to determine a plurality of second clusters; The second processing end obtains the medical and nursing file segmentation score of the target object in each of the second clusters, determines the target training method of the target object in each of the second clusters by using the medical and nursing file segmentation score, and then trains the target object by using the target training method; The first processing end comprises: A first similarity determination module is configured to obtain basic information of the target object, establish a basic information feature set using the basic information, and further determine the similarity of work skills between different target objects; A first clustering module, using the job skill similarity to perform a first clustering on the target objects to obtain a plurality of first clusters; a second similarity determination module, which obtains the prescription ability similarity between different target objects in each of the first clusters, and obtains the professional skill similarity between different target objects, and determines the ability similarity between different target objects in the first cluster by using the prescription ability similarity and the professional skill similarity; The second clustering module clusters the target objects in the first clusters again by using the capability similarity to obtain a plurality of second clusters.

2. The electronic information management system for medical records according to claim 1, characterized in that: The acquiring basic information of the target object, using the basic information to establish a basic information feature set, and then determining the similarity of work skills between different target objects, includes: Obtaining basic information corresponding to each of the target objects, and then using the basic information to establish a basic information feature set corresponding to each of the target objects; wherein the basic information at least includes practice scope data, practice time data, and technical certificate data; Extracting a first basic information feature set of any one of the target objects and a second basic information feature set of any other of the target objects; The similarity of the work skills between two different target objects is determined by using the first basic information feature set and the second basic information feature set.

3. The medical and nursing records electronic information management system according to claim 1, characterized in that: The second similarity determination module comprises: A first sub-similarity determination module is configured to obtain the prescription right category and the number of prescription rights of the target object in the first cluster, and determine the prescription ability similarity between different target objects in each of the first clusters by using the prescription right category and the prescription right number; A second sub-similarity determination module is configured to obtain an operation level and an operation quantity of the target object in the first cluster, and determine the professional skill similarity between different target objects in each of the first clusters by using the operation level and the operation quantity; The third sub-similarity determination module determines the ability similarity between different target objects in each of the first clusters by using the prescription ability similarity and the professional skill similarity.

4. The electronic information management system for medical records according to claim 3 is characterized in that: The obtaining of the prescription right category and the number of prescription rights of the target object in the first cluster, and determining the prescription ability similarity between different target objects in each of the first clusters by using the prescription right category and the number of prescription rights, includes: Acquire a first prescription right category and a first prescription right quantity of any one of the target objects in the first cluster, and acquire a second prescription right category and a second prescription right quantity of another target object in the first cluster; The prescription ability similarity between different target objects in each of the first clusters is determined by using the first prescription right category, the first prescription right quantity, the second prescription right category, and the second prescription right quantity.

5. The medical and nursing records electronic information management system according to claim 3, characterized in that: The obtaining of the operation level and the number of operations of the target object in the first cluster, and determining the professional skill similarity between different target objects in each of the first clusters by using the operation level and the number of operations, includes: Acquire a first operation level and a first operation quantity of any one of the target objects in the first cluster, and acquire a second operation level and a second operation quantity of another target object in the first cluster; The professional skill similarity between different target objects in each of the first clusters is determined by using the first operation level, the first operation quantity, the second operation level, and the second operation quantity.

6. The medical and nursing records electronic information management system according to claim 1, characterized in that: The second processing end includes: A score determination module is used to obtain a medical record word segmentation score value of each target object in each of the second clusters; The training method determination module uses the medical file segmentation score value to determine the target training method of the target object in each of the second clusters, and then uses the target training method to train the target object.

7. The medical and nursing records electronic information management system according to claim 6, characterized in that: The step of obtaining the medical file word segmentation score of each target object in each of the second clusters includes: Obtaining the score value of the target word in the medical file corresponding to each of the target objects in each of the second clusters, and obtaining the first occurrence number of the target word; The medical record word segmentation score of each target object in each of the second clusters is determined by using the score value and the first occurrence number.

8. The medical and nursing records electronic information management system according to claim 6, characterized in that: The step of using the medical file word segmentation score to determine the target training method of the target object in each of the second clusters includes: Sorting the medical and nursing file word segmentation score values, and determining a daily experience word segmentation array for each target object according to the sorting result; Acquire the total number of the target objects in the second cluster, and acquire the second occurrence number of the target daily experience segmentation word in the daily experience segmentation word array of the target object; Determine the possibility of personalized training word segmentation by using the total number and the second number of occurrences; The target training mode of the target object in each of the second clusters is determined by using the personalized training segmentation possibility.

9. The medical and nursing records electronic information management system according to claim 1, characterized in that: The using the target training method to train the target object includes: Using prescription rights to divide the target object into a first target object and a second target object; Performing training on the first target object using the target training method; The prescription right is used to determine a second training method, and then the second training method is used to perform training on the second target object.

Images