An intelligent management method for nursing information in neurology department

Abstract

The present invention relates to the field of nursing information management, and particularly to an intelligent management method for nursing information in the department of neurology. The method first obtains the electronic medical records of each historical patient in the department of neurology, different types of nursing plans, and the electronic medical record of the patient to be tested, performs word segmentation processing on each electronic medical record and nursing plan to obtain the word segmentation of the electronic medical record and nursing plan, clusters all historical patients based on the content information differences of the same type of nursing plans among the historical patients, and obtains the degree of association between the patient to be tested and each cluster according to the number of times the word segmentation of the electronic medical record of the patient to be tested appears in the electronic medical records of each historical patient in each cluster and the number of identical word segmentations, and recommends the nursing plans of the historical patients in the cluster to the patient to be tested. The present invention can provide targeted and personalized recommendations for the nursing plan of the patient to be tested, and improve the effect of nursing information management in the department of neurology.

Description

Technical Field

[0001] The present invention relates to the field of nursing information management, and particularly to an intelligent management method for neurology nursing information. Background Art

[0002] Neurology nursing is a professional nursing work for patients with nervous system diseases. During the process of nursing neurology patients, making targeted and personalized recommendations for the nursing plans of patients helps improve the treatment effect, enhance patient satisfaction, optimize resource utilization, and promote the rehabilitation of patients.

[0003] In related technologies, usually the nursing plans of other patients with similar conditions or symptoms to the patient to be tested are recommended to the patient to be tested. However, due to factors such as the physical constitution and living habits of different patients being different, even for patients with similar conditions or symptoms, there are certain differences in the nursing plans they need, resulting in the inability to make targeted and personalized recommendations for the nursing plans of the patient to be tested through existing algorithms, reducing the effect of nursing information management. Summary of the Invention

[0004] In order to solve the technical problem that existing algorithms cannot make targeted and personalized recommendations for the nursing plans of patients to be tested, reducing the effect of nursing information management, the purpose of the present invention is to provide an intelligent management method for neurology nursing information, and the specific technical solution adopted is as follows:

[0005] The present invention proposes an intelligent management method for neurology nursing information, and the method includes:

[0006] Obtain the electronic medical records of each historical patient in neurology, different types of nursing plans, and the electronic medical record of the patient to be tested, and perform word segmentation processing on the electronic medical records and nursing plans of the patient to be tested and each historical patient respectively to obtain the word segmentation of the electronic medical records and nursing plans;

[0007] Take any type of nursing plan of any historical patient as the target nursing plan, perform sentiment analysis on the target nursing plan to obtain the sentiment score value of each word segmentation in the target nursing plan; extract keywords from the target nursing plan to obtain the keywords of the target nursing plan; according to the sentiment score values of each word segmentation in the preset neighborhood of each keyword in the target nursing plan, obtain the description feature value of each keyword of the target nursing plan; according to the difference in the description feature values of the same keywords and the number of the same keywords between the same type of nursing plans of any two historical patients, obtain the nursing information difference degree between any two historical patients; based on the nursing information difference, cluster all historical patients to obtain multiple clustering clusters;

[0008] Take any one of the clustering clusters as the target clustering cluster, and obtain the degree of association between the patient to be tested and the target clustering cluster according to the number of times each word segment in the electronic medical record of the patient to be tested appears in the electronic medical records of each historical patient in the target clustering cluster, and the number of identical word segments in the electronic medical records between the patient to be tested and each historical patient in the target clustering cluster;

[0009] Based on the degree of association between the patient to be tested and each clustering cluster, recommend the nursing plans of the historical patients in each clustering cluster to the patient to be tested according to different priorities.

[0010] Further, the description feature values of each keyword for obtaining the target nursing plan include:

[0011] Take the average value of the sentiment score values of all word segments in the preset neighborhood of each keyword in the target nursing plan as the description feature value of each keyword in the target nursing plan.

[0012] Further, the obtaining of the difference degree of nursing information between any two historical patients includes:

[0013] Take any one type as the target type, and take the set composed of all keywords in the nursing plan of the target type of each historical patient as the keyword set of each historical patient regarding the target type;

[0014] Arbitrarily select two patients as the first historical patient and the second historical patient, perform a negative correlation mapping on the Jaccard correlation coefficient between the keyword set of the first historical patient regarding the target type and the keyword set of the second historical patient regarding the target type, and obtain the first information difference coefficient between the first historical patient and the second historical patient regarding the target type;

[0015] Take the average value of the absolute value of the difference between the description feature values of all identical keywords between the nursing plan of the target type of the first historical patient and the nursing plan of the target type of the second historical patient as the second information difference coefficient between the first historical patient and the second historical patient regarding the target type;

[0016] Integrate the first information difference coefficient and the second information difference coefficient to obtain the comprehensive information difference coefficient between the first historical patient and the second historical patient regarding the target type;

[0017] Take the average value of the comprehensive information difference coefficients between the first historical patient and the second historical patient regarding all types as the difference degree of nursing information between the first historical patient and the second historical patient.

[0018] Further, the obtaining of multiple clustering clusters includes:

[0019] Take the difference degree of the nursing information between any two historical patients as the distance metric between any two historical patients. Use the K-means clustering algorithm and cluster all historical patients based on the distance metric between any two historical patients to obtain multiple clusters.

[0020] Further, obtaining the degree of association between the patient to be tested and the target cluster includes:

[0021] Take any word segment in the electronic medical record of the patient to be tested as the target word segment, and select the reference historical patients of the target word segment from all historical patients in the target cluster, where the target word segment appears at least once in the electronic medical record of the reference historical patients;

[0022] According to the number of reference historical patients of the target word segment and the number of times the target word segment appears in the electronic medical record of each reference historical patient, obtain the comprehensive distribution breadth of the target word segment with respect to the target cluster; take the average value of the comprehensive distribution breadths of all word segments in the electronic medical record of the patient to be tested with respect to the target cluster as the overall distribution breadth of the patient to be tested with respect to the target cluster;

[0023] Take the set composed of all word segments in the electronic medical record of the patient to be tested as the first word segment set of the patient to be tested, take any historical patient in the target cluster as the target historical patient, and take the set composed of all word segments in the electronic medical record of the target historical patient as the second word segment set of the target historical patient;

[0024] Take the Jaccard correlation coefficient between the first word segment set of the patient to be tested and the second word segment set of the target historical patient as the disease similarity between the patient to be tested and the target historical patient; take the average value of the disease similarities between the patient to be tested and all historical patients in the target cluster as the overall similarity between the patient to be tested and the target cluster;

[0025] After comprehensively processing the overall similarity between the patient to be tested and the target cluster and the overall distribution breadth of the patient to be tested with respect to the target cluster and performing normalization, obtain the degree of association between the patient to be tested and the target cluster.

[0026] Further, obtaining the comprehensive distribution breadth of the target word segment with respect to the target cluster includes:

[0027] Take the number of all reference historical patients of the target word segment as the numerator and the number of all historical patients in the target cluster as the denominator, and take the ratio as the first distribution breadth of the target word segment in the target cluster;

[0028] The average number of occurrences of the target word segment in the electronic medical records of all reference historical patients is used as the second distribution breadth of the target word segment in the target clustering cluster;

[0029] The first distribution breadth and the second distribution breadth are synthesized to obtain the comprehensive distribution breadth of the target word segment with respect to the target clustering cluster.

[0030] Further, the step of recommending the nursing plans of the historical patients in each clustering cluster to the patient to be tested according to different priorities includes:

[0031] The nursing plans of the historical patients in each clustering cluster are recommended to the patient to be tested in the order from largest to smallest of the degree of association between the patient to be tested and each clustering cluster.

[0032] Further, the step of obtaining the word segments of the electronic medical record and the nursing plan includes:

[0033] The Jieba word segmentation tool is used to perform word segmentation processing on the electronic medical records and nursing plans of the patient to be tested and each historical patient to obtain the word segments of the electronic medical records and nursing plans.

[0034] Further, the step of obtaining the sentiment score value of each word segment in the target nursing plan includes:

[0035] The NTUSD sentiment dictionary is used to process the target nursing plan to obtain the sentiment score value of each word segment in the target nursing plan.

[0036] Further, the step of obtaining the keywords of the target nursing plan includes:

[0037] The knowledge library is used to extract professional vocabulary from the target nursing plan as the keywords of the target nursing plan.

[0038] The present invention has the following beneficial effects:

[0039] Considering that the existing algorithms cannot provide targeted and personalized recommendations for the care plans of patients to be tested, reducing the effectiveness of nursing information management, this invention first obtains the electronic medical records of each historical patient in the department of neurology, different types of care plans, and the electronic medical records of the patients to be tested, and performs word segmentation on each electronic medical record and care plan to obtain the word segmentation of the electronic medical records and care plans. Subsequently, based on the word segmentation, the differences in electronic medical records and care plans among patients can be accurately analyzed. Considering that the physical constitutions and living habits of different patients are different, even for patients with similar diseases or symptoms, there are certain differences in the care plans they need. Therefore, first, sentiment analysis is performed on the target care plan to reflect the emotional semantic features of each word segmentation in the target care plan through the sentiment score value, and the keywords in the target care plan are extracted. The emotional semantic features of each word segmentation within the local range of each keyword are reflected through the description feature value. Then, the difference in the content information of the care plans among historical patients is reflected through the nursing information difference degree, and then the historical patients with similar care plans are classified into the same clustering cluster, eliminating the interference of factors such as physical constitution and living habits among patients on the subsequent personalized recommendation of care plans. At the same time, considering that the similarity of the disease manifestations between the patients to be tested and the historical patients in different clustering clusters is different, and in the process of recommending care plans, it is necessary to recommend the care plans of historical patients with similar diseases to the patients to be tested. Therefore, the similarity and relevance of the disease manifestations between the patients to be tested and the historical patients in the target clustering cluster are reflected through the association degree, and then the care plans of the historical patients in each clustering cluster are recommended to the patients to be tested according to different priorities, to improve the effectiveness of nursing information management. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0041] Figure 1 It is a flowchart of an intelligent management method for nursing information in the department of neurology provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the attached drawings and preferred embodiments, details a smart management method for neurology nursing information proposed according to the present invention, including its specific implementation manner, structure, features and effects, as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this invention belongs.

[0044] The following specifically describes the specific solution of a smart management method for neurology nursing information provided by the present invention in combination with the attached drawings.

[0045] Please refer to Figure 1 , which shows a flowchart of a smart management method for neurology nursing information provided by an embodiment of the present invention. The method includes:

[0046] Step S1: Obtain the electronic medical records of each historical patient in the neurology department, different types of nursing plans, and the electronic medical record of the patient to be tested. Perform word segmentation processing on the electronic medical records and nursing plans of the patient to be tested and each historical patient respectively to obtain the word segmentation of the electronic medical records and nursing plans.

[0047] As one of the management methods for neurology nursing information, the recommendation of neurology nursing plans helps to improve the treatment effect, enhance patient satisfaction, optimize resource utilization, and promote the rehabilitation of patients. During the process of nursing neurology patients, medical staff will conduct various inquiries and examinations on patients, such as inquiries about the condition, CT or ultrasound imaging examinations, etc., so as to integrate the information about the patient's physical condition collected into the patient's electronic medical record. Then, based on the patient's electronic medical record, various types of nursing plans are formulated for the patient to promote the patient's rehabilitation, and the patient's electronic medical record is stored together with the nursing plan in the hospital's terminal database for subsequent use and analysis.

[0048] Therefore, in the embodiment of the present invention, first, the electronic medical records of each historical patient in the neurology department, different types of nursing plans, and the electronic medical record of the patient to be tested are obtained from the hospital's terminal database. Among them, different types of nursing plans mainly include, for example, a drug nursing plan, which contains drug information provided for promoting rehabilitation, a rehabilitation nursing plan, which includes guidance information such as limb function exercise, language rehabilitation training, and cognitive rehabilitation training provided for promoting rehabilitation, and a psychological nursing plan, which includes psychological counseling information provided for promoting rehabilitation, etc.

[0049] It should be noted that the nursing plans for historical patients are known, while the nursing plan for the patient to be tested is unknown. Subsequently, the nursing plans of historical patients need to be recommended to the patient to be tested.

[0050] Since the electronic medical records and nursing plans of historical patients and the patient to be tested usually appear in text form, the embodiments of the present invention also need to perform word segmentation on the electronic medical records and nursing plans of the patient to be tested and each historical patient respectively to obtain the word segmentation of the electronic medical records and nursing plans, which is convenient for subsequent analysis of the differences in electronic medical records and nursing plans between patients based on the word segmentation. After word segmentation, each electronic medical record has multiple word segments, and each nursing plan also has multiple word segments.

[0051] Preferably, in an embodiment of the present invention, an existing Jieba word segmentation tool is used to perform word segmentation on the electronic medical records and nursing plans of the patient to be tested and each historical patient to obtain the word segmentation of the electronic medical records and nursing plans. In other embodiments of the present invention, other word segmentation methods can also be used for word segmentation, which is not limited herein.

[0052] Step S2: Take any type of nursing plan of any historical patient as the target nursing plan, perform sentiment analysis on the target nursing plan to obtain the sentiment score value of each word segment in the target nursing plan; extract keywords from the target nursing plan to obtain the keywords of the target nursing plan; obtain the description feature value of each keyword in the target nursing plan according to the sentiment score values of the word segments in the preset neighborhood of each keyword in the target nursing plan; obtain the nursing information difference degree between any two historical patients according to the difference in the description feature values of the same keywords and the number of the same keywords between the same type of nursing plans of any two historical patients; cluster all historical patients based on the nursing information difference to obtain multiple clustering clusters.

[0053] Due to factors such as the physical constitution and living habits of different patients being different, that is, there are individual differences between patients. Even for patients with similar conditions or symptoms, there are still certain differences in the nursing plans they need. If the nursing plan of a historical patient whose electronic medical record is similar to that of the patient to be tested is directly recommended to the patient to be tested, the final recommendation effect will be reduced. Therefore, the embodiments of the present invention eliminate the interference of individual differences by dividing historical patients with similar nursing plans into the same clustering cluster, so as to provide targeted and personalized recommendations for the nursing plan of the patient to be tested, improving the recommendation effect and the effect of nursing information management.

[0054] Before clustering historical patients, it is necessary to analyze the information differences in the care plans among historical patients. In the embodiments of the present invention, first, any care plan of any type of a historical patient is used as the target care plan. Since the care plan is provided by medical staff, the content information in the care plan contains degree words expressing the emotions of medical staff. The differences in the emotional semantic features of the content between care plans can reflect the information differences between care plans. Therefore, emotional analysis can be performed on the target care plan to obtain the emotional score value of each participle in the target care plan, and the emotional semantic features of each participle in the target care plan are reflected through the obtained emotional score values. Subsequently, based on the differences in the emotional score values of each participle between care plans, the information differences between care plans are accurately analyzed.

[0055] Preferably, in an embodiment of the present invention, the NTUSD emotion dictionary is used to process the target care plan to obtain the emotional score value of each participle in the target care plan. Among them, the NTUSD emotion dictionary is a well-known technical means for those skilled in the art and will not be elaborated here.

[0056] Then, in the embodiments of the present invention, keyword extraction is performed on the target care plan to obtain the keywords of the target care plan, and the description feature value of each keyword of the target care plan is obtained according to the emotional score values of each participle in the preset neighborhood of each keyword in the target care plan. The overall level of the emotional semantic features of each participle within the local range of each keyword in the target care plan is reflected through the description feature value, which is convenient for accurately analyzing the differences in the care plan information among historical patients based on the differences in the emotional score values of the same keywords between the same-type care plans of each historical patient, and improving the clustering effect. Among them, the preset neighborhood of a certain keyword in the target care plan includes the keyword and a preset number of the closest participles to the keyword. The preset number is set to 10. That is to say, a total of 11 participles are included in the preset neighborhood. The specific value of the preset number can also be set by the implementer according to the specific implementation scenario and will not be limited here.

[0057] Preferably, in an embodiment of the present invention, a knowledge library can be used to extract professional vocabulary from the target care plan as the keywords of the target care plan. For example, if the target care plan is a drug care plan, a drug knowledge library can be used to extract various professional vocabulary about drugs from the target care plan as the keywords of the target care plan. Among them, the method of extracting keywords using a knowledge library is a well-known technical means for those skilled in the art and will not be elaborated here.

[0058] Preferably, in an embodiment of the present invention, the method for obtaining the description feature value of each keyword of the target care plan specifically includes:

[0059] The average value of the sentiment score values of all word segments in the preset neighborhood of each keyword in the target care plan is used as the description feature value of each keyword in the target care plan.

[0060] By the same method as above, the description feature values of each keyword in each type of care plan for each historical patient can be obtained.

[0061] Since each historical patient has multiple types of care plans, in the process of analyzing the information differences between the care plans of historical patients, it is necessary to analyze the care plans of the same type among historical patients. The greater the difference in the sentiment score values of the same keywords between the care plans of the same type among historical patients, and the fewer the number of the same keywords between the care plans of the same type among historical patients, it indicates that the information differences between the care plans of historical patients are greater. Therefore, the differences in the sentiment score values of the same keywords and the number of the same keywords between the care plans of the same type of any two historical patients can be analyzed, and the content information differences between the care plans of any two historical patients are reflected through the obtained care information difference degree. Subsequently, clustering operations on historical patients can be achieved based on the care information difference degree.

[0062] Preferably, in an embodiment of the present invention, the method for obtaining the care information difference degree between any two historical patients specifically includes:

[0063] First, any one type of the care plan is used as the target type, and the set composed of all keywords in the care plan of the target type of each historical patient is used as the keyword set of each historical patient regarding the target type;

[0064] Then, any two patients are randomly selected as the first historical patient and the second historical patient, and the Jaccard correlation coefficient between the keyword set of the first historical patient regarding the target type and the keyword set of the second historical patient regarding the target type is negatively correlated and mapped to obtain the first information difference coefficient between the first historical patient and the second historical patient regarding the target type. The greater the first information difference coefficient, the greater the content information difference between the care plans of the target type between the first historical patient and the second historical patient.

[0065] The average value of the absolute value of the difference between the description feature values of all the same keywords between the care plan of the target type of the first historical patient and the care plan of the target type of the second historical patient is used as the second information difference coefficient between the first historical patient and the second historical patient regarding the target type. The greater the second information difference coefficient, the greater the difference in the emotional semantic features of the keywords in the care plans of the target type between the first historical patient and the second historical patient, and further indicates that the content information difference between the care plans of the target type between the first historical patient and the second historical patient is greater.

[0066] Furthermore, the first information difference coefficient and the second information difference coefficient are synthesized to obtain a comprehensive information difference coefficient regarding the target type between the first historical patient and the second historical patient. The larger the comprehensive information difference coefficient is, the greater the information difference of the nursing plans for the target type between the first historical patient and the second historical patient is.

[0067] In an embodiment of the present invention, the sum value or product value of the first information difference coefficient and the second information difference coefficient can be used as the comprehensive information difference coefficient regarding the target type between the first historical patient and the second historical patient to achieve the synthesis of the two, and there is no limitation here.

[0068] Through the same method as above, the comprehensive information difference coefficient regarding each type between the first historical patient and the second historical patient can be obtained, and the average value of the comprehensive information difference coefficients regarding all types between the first historical patient and the second historical patient is used as the nursing information difference degree between the first historical patient and the second historical patient.

[0069] As an example, in an embodiment of the present invention, the expression of the nursing information difference degree between the first historical patient and the second historical patient can be specifically, for example:

[0070]

[0071]

[0072]

[0073] Among them, represents the nursing information difference degree between the first historical patient and the second historical patient; represents the first information difference coefficient regarding the th type between the first historical patient and the second historical patient; represents the second information difference coefficient regarding the th type between the first historical patient and the second historical patient; represents the comprehensive information difference coefficient regarding the th type between the first historical patient and the second historical patient; represents the number of types of nursing plans; represents the first information difference coefficient regarding the target type between the first historical patient and the second historical patient; represents the Jaccard correlation coefficient between the keyword set of the first historical patient regarding the target type and the keyword set of the second historical patient regarding the target type; represents the second information difference coefficient regarding the target type between the first historical patient and the second historical patient; The descriptive feature value of the nth keyword in the care plan representing the target type of the first historical patient; The descriptive feature value of the nth keyword in the care plan representing the target type of the second historical patient, where the nth keyword in the care plan representing the target type of the first historical patient and the nth keyword in the care plan representing the target type of the second historical patient are the same; Represents the number of all identical keywords between the care plan of the first historical patient and the care plan of the second historical patient; Represents a preset adjustment parameter used to prevent the denominator from being zero, whose value range is , and in an embodiment of the present invention, is set to 0.01, and the specific value of

[0074] can also be set by the implementer according to the specific implementation scenario and is not limited herein.

[0075] Preferably, in an embodiment of the present invention, the method for obtaining multiple clustering clusters specifically includes:

[0076] Taking the care information difference degree between any two historical patients as the distance metric between any two historical patients, using the K-means clustering algorithm, and clustering all historical patients based on the distance metric between any two historical patients to obtain multiple clustering clusters, where the number of clustering clusters can be determined by the existing elbow method or silhouette coefficient method.

[0077] Thus, multiple clustering clusters are obtained.

[0078] Step S3: Taking any one of the clustering clusters as the target clustering cluster, obtaining the degree of association between the patient to be tested and the target clustering cluster according to the number of occurrences of each word segment in the electronic medical record of the patient to be tested in the electronic medical records of each historical patient in the target clustering cluster, and the number of identical word segments in the electronic medical records between the patient to be tested and each historical patient in the target clustering cluster.

[0079] Due to the different similarities in the disease manifestations between the patient to be tested and the historical patients in different clustering clusters, and during the process of recommending a nursing plan, it is necessary to recommend the nursing plan of the historical patients with similar diseases to the patient to be tested. Therefore, in the embodiments of the present invention, first, any clustering cluster is analyzed, and any clustering cluster is used as the target clustering cluster. The more times each word segment in the electronic medical record of the patient to be tested appears in the electronic medical records of each historical patient in the target clustering cluster, and the more the number of identical word segments in the electronic medical records between the patient to be tested and each historical patient in the target clustering cluster, it indicates that the content information of the electronic medical record of the patient to be tested and the electronic medical records of each historical patient in the target clustering cluster is more similar, and further indicates that the disease manifestations between the patient to be tested and each historical patient in the target clustering cluster are more similar. Therefore, the number of times each word segment in the electronic medical record of the patient to be tested appears in the electronic medical records of each historical patient in the target clustering cluster, and the number of identical word segments in the electronic medical records between the patient to be tested and each historical patient in the target clustering cluster can be analyzed, and the obtained degree of association reflects the similarity and relevance of the disease manifestations between the patient to be tested and the historical patients in the target clustering cluster. Subsequently, based on the degree of association, a targeted and personalized recommendation for the nursing plan of the patient to be tested can be made, thereby improving the effect of neurological nursing information management.

[0080] Preferably, in an embodiment of the present invention, the method for obtaining the degree of association between the patient to be tested and the target clustering cluster specifically includes:

[0081] First, any word segment in the electronic medical record of the patient to be tested is used as the target word segment, and the reference historical patients of the target word segment are selected from all the historical patients in the target clustering cluster, and the target word segment appears at least once in the electronic medical record of the reference historical patients;

[0082] The more the number of reference historical patients of the target word segment, and the more the number of times the target word segment appears in the electronic medical record of each reference historical patient, it indicates that the target word segment is more widely distributed in the electronic medical records of each historical patient in the target clustering cluster. According to the number of reference historical patients of the target word segment and the number of times the target word segment appears in the electronic medical record of each reference historical patient, the comprehensive distribution breadth of the target word segment with respect to the target clustering cluster is obtained. By the same method as above, the comprehensive distribution breadth of each word segment in the electronic medical record of the patient to be tested with respect to the target clustering cluster can be obtained. Furthermore, the average value of the comprehensive distribution breadths of all the word segments in the electronic medical record of the patient to be tested with respect to the target clustering cluster is used as the overall distribution breadth of the patient to be tested with respect to the target clustering cluster. The larger the overall distribution breadth, it indicates that each word segment in the electronic medical record of the patient to be tested is more widely and extensively distributed in the electronic medical records of the historical patients in the target clustering cluster, and further indicates that the disease manifestations between the patient to be tested and the historical patients in the target clustering cluster are more similar.

[0083] Preferably, in an embodiment of the present invention, the method for obtaining the comprehensive distribution breadth of the target word segmentation with respect to the target clustering cluster specifically includes:

[0084] Taking the number of all reference historical patients of the target word segmentation as the numerator, taking the number of all historical patients in the target clustering cluster as the denominator, taking the ratio as the first distribution breadth of the target word segmentation in the target clustering cluster, and taking the average value of the number of times the target word segmentation appears in the electronic medical records of all reference historical patients as the second distribution breadth of the target word segmentation in the target clustering cluster, and comprehensively combining the first distribution breadth and the second distribution breadth to obtain the comprehensive distribution breadth of the target word segmentation with respect to the target clustering cluster.

[0085] In an embodiment of the present invention, the sum value or product value of the first distribution breadth and the second distribution breadth can be used as the comprehensive distribution breadth of the target word segmentation with respect to the target clustering cluster to achieve the combination of the two, and there is no limitation here.

[0086] As an example, in an embodiment of the present invention, the expression of the comprehensive distribution breadth of the target word segmentation with respect to the target clustering cluster can be specifically, for example:

[0087]

[0088] Wherein, represents the comprehensive distribution breadth of the target word segmentation with respect to the target clustering cluster; represents the number of all reference historical patients of the target word segmentation; represents the number of all historical patients in the target clustering cluster; represents the first distribution breadth of the target word segmentation in the target clustering cluster; represents the number of times the target word segmentation appears in the electronic medical record of the th reference historical patient; represents the second distribution breadth of the target word segmentation in the target clustering cluster.

[0089] Then, taking the set composed of all word segmentations of the electronic medical record of the patient to be tested as the first word segmentation set of the patient to be tested, taking any one historical patient in the target clustering cluster as the target historical patient, and taking the set composed of all word segmentations of the electronic medical record of the target historical patient as the second word segmentation set of the target historical patient;

[0090] The Jaccard correlation coefficient between the first word segmentation set of the patient to be tested and the second word segmentation set of the target historical patient is used as the disease similarity between the patient to be tested and the target historical patient. The greater the disease similarity, the more similar the disease manifestations between the patient to be tested and the target historical patient. By the same method as above, the disease similarity between the patient to be tested and each historical patient in the target clustering cluster can be obtained. Furthermore, the average value of the disease similarities between the patient to be tested and all historical patients in the target clustering cluster can be used as the overall similarity between the patient to be tested and the target clustering cluster. The greater the overall similarity, the more similar the disease manifestations between the patient to be tested and each historical patient in the target clustering cluster.

[0091] Furthermore, after comprehensively considering the overall similarity between the patient to be tested and the target clustering cluster and the overall distribution breadth of the patient to be tested with respect to the target clustering cluster, and performing normalization processing, the calculation result is limited to a certain range to obtain the degree of association between the patient to be tested and the target clustering cluster.

[0092] In an embodiment of the present invention, the comprehensive consideration of the two can be achieved by calculating the sum value or product value of the overall similarity between the patient to be tested and the target clustering cluster and the overall distribution breadth of the patient to be tested with respect to the target clustering cluster, and no limitation is made here.

[0093] In an embodiment of the present invention, the normalization processing can be specifically, for example, the maximum-minimum normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of the numerical values, and details are not described herein.

[0094] As an example, in an embodiment of the present invention, the expression of the degree of association between the patient to be tested and the target clustering cluster can be specifically, for example:

[0095]

[0096]

[0097] Among them, represents the degree of association between the patient to be tested and the target clustering cluster; represents the overall distribution breadth of the patient to be tested with respect to the target clustering cluster; represents the overall similarity between the patient to be tested and the target clustering cluster; represents the disease similarity between the patient to be tested and the th historical patient in the target clustering cluster; represents the number of all historical patients in the target clustering cluster; represents the normalization function.

[0098] The degree of association between the patient to be measured and each clustering cluster can be obtained by the same method described above.

[0099] Step S4: Based on the degree of association between the patient to be measured and each clustering cluster, recommend the nursing plans of the historical patients in each clustering cluster to the patient to be measured according to different priorities.

[0100] The greater the degree of association between the patient to be measured and each clustering cluster, the greater the similarity and correlation of the disease manifestations between the patient to be measured and the historical patients in each clustering cluster. Therefore, it is necessary to preferentially recommend the nursing plans of the historical patients in the clustering clusters with a greater degree of association to the patient to be measured, so as to achieve targeted and personalized recommendations for the nursing plan of the patient to be measured and improve the effect of nursing information management.

[0101] Preferably, in an embodiment of the present invention, recommending the nursing plans of the historical patients in each clustering cluster to the patient to be measured according to different priorities includes:

[0102] Recommend the nursing plans of the historical patients in each clustering cluster to the patient to be measured in the order from largest to smallest of the degree of association between the patient to be measured and each clustering cluster.

[0103] It should be noted that the above sequence of 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.

[0104] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. An intelligent management method for neurology nursing information, characterized in that, The method includes: Obtaining the electronic medical records of each historical patient in the department of neurology, different types of nursing plans, and the electronic medical record of the patient to be tested, and performing word segmentation processing on the electronic medical records and nursing plans of the patient to be tested and each historical patient respectively to obtain the word segmentation of the electronic medical records and nursing plans; Taking any one type of nursing plan of any one historical patient as the target nursing plan, performing sentiment analysis on the target nursing plan to obtain the sentiment score value of each word segmentation in the target nursing plan, where the sentiment score value is used to reflect the sentiment semantic characteristics of each word segmentation in the target nursing plan; extracting keywords from the target nursing plan to obtain the keywords of the target nursing plan; obtaining the description feature value of each keyword of the target nursing plan according to the sentiment score values of the word segmentations in the preset neighborhood of each keyword in the target nursing plan, where the description feature value is used to reflect the sentiment semantic characteristics of the word segmentations within the local range of each keyword; obtaining the nursing information difference degree between any two historical patients according to the difference in the description feature values of the same keywords between the same type of nursing plans of any two historical patients and the number of the same keywords, where the nursing information difference degree is used to reflect the difference in the content information of the nursing plans between any two historical patients. Among them, the nursing information difference degree between the first historical patient and the second historical patient is the average value of the comprehensive information difference coefficients of all types between the first historical patient and the second historical patient. The comprehensive information difference coefficient regarding the target type is used to reflect the information difference of the nursing plan of the target type between the first historical patient and the second historical patient. The comprehensive information difference coefficient regarding the target type is obtained by synthesizing the first information difference coefficient and the second information difference coefficient. The first information difference coefficient is used to reflect the content information difference of the nursing plan of the target type between the first historical patient and the second historical patient. The first information difference coefficient is obtained by performing a negative correlation mapping on the Jaccard correlation coefficient between the keyword set of the target type of the first historical patient and the keyword set of the target type of the second historical patient. The second information difference coefficient is used to reflect the difference in the sentiment semantic characteristics of the keywords of the nursing plan of the target type between the first historical patient and the second historical patient. The second information difference coefficient is the average value of the absolute values of the differences in the description feature values of all the same keywords between the nursing plan of the target type of the first historical patient and the nursing plan of the target type of the second historical patient. The keyword set of each historical patient regarding the target type is the set composed of all the keywords of the nursing plan of the target type of each historical patient, and the target type is any type; clustering all the historical patients based on the nursing information difference to obtain multiple clustering clusters; Take any one of the clustering clusters as the target clustering cluster. According to the number of occurrences of each word segment in the electronic medical record of the patient to be tested in the electronic medical records of each historical patient in the target clustering cluster, and the number of identical word segments in the electronic medical records between the patient to be tested and each historical patient in the target clustering cluster, obtain the degree of association between the patient to be tested and the target clustering cluster. The degree of association is used to reflect the similarity and relevance of the disease manifestations between the patient to be tested and the historical patients in the target clustering cluster. The degree of association is obtained by comprehensively considering the overall similarity between the patient to be tested and the target clustering cluster and the overall distribution breadth of the patient to be tested with respect to the target clustering cluster and then performing normalization processing; the overall similarity is the average value of the disease similarities between the patient to be tested and all historical patients in the target clustering cluster. The disease similarity between the patient to be tested and the target historical patient is the Jaccard correlation coefficient between the first word segment set of the patient to be tested and the second word segment set of the target historical patient; the first word segment set is the set composed of all word segments in the electronic medical record of the patient to be tested, the target historical patient is any one of the historical patients in the target clustering cluster, and the second word segment set is the set composed of all word segments in the electronic medical record of the target historical patient; the overall distribution breadth is the average value of the comprehensive distribution breadths of all word segments in the electronic medical record of the patient to be tested with respect to the target clustering cluster. The comprehensive distribution breadth of the target word segment with respect to the target clustering cluster is obtained based on the number of reference historical patients of the target word segment and the number of occurrences of the target word segment in the electronic medical records of each reference historical patient. The target word segment is any word segment in the electronic medical record of the patient to be tested, and the reference historical patients are selected from all historical patients in the target clustering cluster, and the target word segment appears at least once in the electronic medical records of the reference historical patients. Based on the degree of association between the patient to be tested and each clustering cluster, recommend the nursing plans of the historical patients in each clustering cluster to the patient to be tested according to different priorities.

2. The intelligent management method for neurology nursing information according to claim 1, wherein The description feature values of each keyword for obtaining the target nursing plan include: Take the average value of the sentiment score values of all word segments in the preset neighborhood of each keyword in the target nursing plan as the description feature value of each keyword in the target nursing plan.

3. The intelligent management method for neurology nursing information according to claim 1, characterized in that, The obtaining of multiple clustering clusters includes: Take the difference degree of the nursing information between any two historical patients as the distance metric between any two historical patients. Use the K-means clustering algorithm and, based on the distance metric between any two historical patients, cluster all historical patients to obtain multiple clustering clusters.

4. The intelligent management method for neurology nursing information according to claim 1, wherein The method further includes: Take the number of all reference historical patients of the target word segment as the numerator and the number of all historical patients in the target clustering cluster as the denominator, and take the ratio as the first distribution breadth of the target word segment in the target clustering cluster; Take the average value of the number of occurrences of the target word segment in the electronic medical records of all reference historical patients as the second distribution breadth of the target word segment in the target clustering cluster; Comprehensively consider the first distribution breadth and the second distribution breadth to obtain the comprehensive distribution breadth of the target word segment with respect to the target clustering cluster.

5. An intelligent management method for neurology nursing information according to claim 1, characterized in that, Recommending the care plans of historical patients in each clustering cluster to the patient to be tested according to different priorities includes: Recommending the care plans of historical patients in each clustering cluster to the patient to be tested in the order from largest to smallest of the degree of association between the patient to be tested and each clustering cluster.

6. The intelligent management method for neurology nursing information according to claim 1, characterized in that, The word segmentation of obtaining the electronic medical records and care plans includes: Using the Jieba word segmentation tool, performing word segmentation processing on the electronic medical records and care plans of the patient to be tested and each historical patient to obtain the word segmentation of the electronic medical records and care plans.

7. The intelligent management method for neurology nursing information according to claim 1, characterized in that, Obtaining the sentiment score value of each word segment in the target care plan includes: Using the NTUSD sentiment dictionary to process the target care plan to obtain the sentiment score value of each word segment in the target care plan.

8. The intelligent management method for neurology nursing information according to claim 1, characterized in that Obtaining the keywords of the target care plan includes: Utilizing the knowledge word bank to extract professional vocabulary from the target care plan as the keywords of the target care plan.

Images