Clinical medication data recording method and system

Abstract

The invention relates to the field of medication information recording, in particular to a clinical medication data recording method and system. The method comprises the following steps: firstly, acquiring clinical medication information of each patient in different dimensions and disease time of the patient, quantifying the clinical medication information, acquiring a medication information coding value of each patient in each dimension, acquiring disease similarity of each patient according to the difference of the medication information coding values of each patient in the same dimension, and determining the disease similarity of each patient according to the disease similarity. Clustering the patients based on the difference of the disease similarity of the patients, and obtaining patient record priorities according to the difference of the disease similarity of the patients in a cluster, the overall level of the disease similarity, the number of the patients and the difference of disease time; according to the distribution of the medicine use information coding values of the same dimension of all the patients, dimension record priorities are obtained; and in combination with the patient record priority and the dimension record priority, recording clinical medication information of the patient. The efficiency of recording clinical medication data of a large number of patients can be improved.

Description

Technical Field

[0001] The present invention relates to the field of medication information recording, and in particular to a clinical medication data recording method and system. Background Art

[0002] An epidemic is a highly contagious disease with a high incidence rate. Collecting and recording various clinical medication data information of patient groups can provide valuable data resources for drug research and epidemiological research, helping researchers understand important data information such as the long-term effects, adverse reactions and medication trends of drugs, thereby helping to formulate more accurate treatment strategies and preventive measures.

[0003] In the related art, the clinical medication information of a patient group is usually stored and recorded with the same priority. However, since there are a large number of non-epidemic patients in the entire patient group, and the clinical medication information of the patients usually contains information in multiple dimensions, there is a lot of non-critical redundant information in the multi-dimensional clinical medication information, resulting in low efficiency in recording the clinical medication information of a large number of patients using existing methods. Summary of the invention

[0004] In order to solve the technical problem of low efficiency in recording clinical medication information of a large number of patients through existing methods, the purpose of the present invention is to provide a clinical medication data recording method and system, and the technical scheme adopted is as follows: The present invention proposes a method for recording clinical medication data, the method comprising: Obtain clinical medication information of different dimensions for each patient and the duration of illness of each patient, perform quantitative preprocessing on the clinical medication information, and obtain a medication information coding value of each dimension for each patient; According to the difference in the medication information coding values ​​of the same dimension between the patients, the symptom similarity of each patient is obtained; based on the difference in the symptom similarity between the patients, all patients are clustered to obtain multiple clusters; according to the difference in the symptom similarity between the patients in each cluster, the overall level of the symptom similarity of all patients in each cluster, the number of patients in each cluster, and the difference in the illness time between the patients in each cluster, the patient record priority of each patient is obtained; According to the distribution of the medication information coding values ​​of the same dimension of all patients, the dimensional record priority of each dimension is obtained; the information record priority of each dimension of each patient is obtained by combining the patient record priority and the dimensional record priority; Based on the information recording priority, the clinical medication information of each dimension of each patient is recorded with different priorities.

[0005] Furthermore, obtaining the similarity of symptoms of each patient includes: Taking any dimension as the target dimension, taking the average of the medication information coding values ​​of the target dimension of all patients as the overall coding value of the target dimension; Taking any patient as a target patient, negatively correlating the absolute value of the difference between the medication information coding value of the target dimension of the target patient and the overall coding value of the target dimension, and obtaining the degree of convergence of the target dimension of the target patient; The accumulated values ​​of the approximation degrees of all dimensions of the target patient are normalized to obtain the symptom similarity of the target patient.

[0006] Furthermore, obtaining a plurality of clusters includes: The absolute value of the difference between the similarities of the symptoms of any two patients is used as the distance measure between any two patients; Using the K-means clustering algorithm, all patients are clustered based on the distance metric between any two patients to obtain multiple clusters.

[0007] Further, obtaining the patient record priority of each patient includes: Taking any cluster as a target cluster, and taking the average of the disease similarities of all patients in the target cluster as the cluster center value of the target cluster; The absolute value of the difference between the disease similarity of each patient in the target cluster and the cluster center value of the target cluster is used as the disease similarity deviation of each patient in the target cluster, and a negative correlation mapping is performed on the average value of the disease similarity deviation of all patients in the target cluster to obtain a first clustering degree of the target cluster; According to the difference in illness time between the patients in the target cluster, obtaining a second clustering degree of the target cluster; The first clustering degree, the second clustering degree, the cluster center value and the number of patients in the target clustering cluster are integrated and normalized to obtain the patient record priority of each patient in the target clustering cluster.

[0008] Further, obtaining the second aggregation degree of the target cluster includes: The average of the illness time of all patients in the target cluster is taken as the overall illness time of the target cluster; The absolute value of the difference between the illness time of each patient in the target cluster and the overall illness time of the target cluster is used as the illness time deviation of each patient in the target cluster, and a negative correlation mapping is performed on the average value of the illness time deviation of all patients in the target cluster to obtain the second clustering degree of the target cluster.

[0009] Furthermore, obtaining the dimension record priority of each dimension includes: The standard deviation of the medication information coding value of the target dimension of all patients is negatively correlated and normalized to obtain the dimension record priority of the target dimension.

[0010] Furthermore, obtaining the information record priority of each dimension of each patient includes: The product value of the patient record priority of each patient and the dimension record priority of each dimension is normalized to obtain the information record priority of each dimension of each patient.

[0011] Furthermore, recording the clinical medication information of each dimension of each patient with different priorities includes: The clinical medication information of the patient dimension corresponding to the information recording priority is recorded in descending order of the information recording priority.

[0012] Furthermore, obtaining the medication information coding value of each dimension of each patient includes: Using the Jieba word segmentation tool, the clinical medication information of each dimension of each patient is segmented to obtain multiple phrases in each dimension of each patient; Use the one-hot encoding algorithm to encode multiple phrases in each dimension of each patient to obtain the one-hot encoding value of each phrase; The average of the one-hot encoding values ​​of all phrases in each dimension of each patient is used as the encoding value of the medication information of each dimension of each patient.

[0013] The present invention also proposes a clinical medication data recording system, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of a clinical medication data recording method when executing the computer program.

[0014] The present invention has the following beneficial effects: The present invention takes into account that the existing methods are inefficient in recording clinical medication information of a large number of patients. Therefore, the clinical medication information of different dimensions and the illness time of each patient are first obtained. Considering that the clinical medication information is usually presented in the form of text, it is necessary to quantify it to obtain the medication information coding value of each dimension of each patient, so as to provide a data basis for subsequent analysis. Since the present invention needs to give priority to recording the clinical medication information of epidemic patients, so as to realize epidemiological research and analysis, and there are some non-epidemic patients in a large number of patient groups, the recording priority of clinical medication information of non-epidemic patients should be appropriately lowered to improve the efficiency of information recording. The symptoms of epidemic patients are relatively similar, and their clinical medication information of the same dimension is also relatively close. Therefore, the similarity of the symptoms between the target patient and other patients can be reflected by the similarity of the symptoms, and then the patients with similar symptoms are divided into the same cluster. Considering that the more concentrated the similarity of the symptoms of each patient in the cluster, the lower the probability of death. In addition, the greater the similarity of the symptoms of each patient, the greater the possibility that the patients in the cluster suffer from epidemic diseases. At the same time, epidemics are highly contagious and break out in a short period of time, resulting in a large number of epidemic patients and a close illness time. Therefore, analysis can be conducted based on this feature, and the patient record priority can be used to reflect the priority of clinical medication data recording for each patient. Considering that there is redundant information in some non-critical dimensions in the multi-dimensional clinical medication information of patients, the information recording priority of these dimensions should be lower to improve the efficiency of clinical medication data recording. The clinical medication information of these redundant dimensions has poor consistency in distribution in the patient population, which is specifically manifested in that the distribution of the medication information coding values ​​of the redundant dimensions in the patient population is relatively discrete. Therefore, the priority of information recording for each dimension can be reflected through the dimension record priority, and then the clinical medication information of each dimension of each patient can be recorded with different priorities through the information record priority, thereby improving the efficiency of clinical medication data recording for a large number of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] 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.

[0016] Figure 1 A flow chart of a method for recording clinical medication data provided by one embodiment of the present invention; Figure 2 A flow chart of a method for obtaining the patient record priority of each patient provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0017] 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 a clinical medication data recording method and system proposed according to the present invention, its specific implementation, structure, features and effects, 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.

[0018] 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.

[0019] The following is a detailed description of a clinical medication data recording method and system provided by the present invention in conjunction with the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of a clinical medication data recording method provided by an embodiment of the present invention, the method comprising: Step S1: Obtain clinical medication information of different dimensions for each patient and the duration of illness of each patient, perform quantitative preprocessing on the clinical medication information, and obtain a medication information coding value of each dimension for each patient.

[0021] When medical staff in hospitals use drugs to treat patients' diseases, they usually store information in multiple dimensions, such as the name, dosage, usage, formula, dosage and medication time of the drugs used for patients in the hospital information management system. At the same time, medical staff will also ask the patient about the duration of the illness and store the patient's illness time along with the multiple dimensions of information about the drugs in the hospital information management system.

[0022] Therefore, the embodiment of the present invention first obtains the clinical medication information of different dimensions of each patient and the illness time of each patient from the hospital information management system. Since the clinical medication information is usually presented in the form of text, the text information cannot be used for subsequent data analysis. Therefore, it is necessary to quantify the clinical medication information and preprocess it to obtain the medication information coding value of each dimension of each patient. The medication information coding value of each dimension of each patient can represent the clinical medication information of the patient in that dimension, thereby providing a data basis for subsequent data analysis.

[0023] Preferably, in one embodiment of the present invention, the method for obtaining the medication information coding value of each dimension of each patient specifically includes: First, the Jieba word segmentation tool is used to segment the clinical medication information of each dimension of each patient to obtain multiple phrases in each dimension of each patient. Then, the one-hot encoding algorithm is used to encode the multiple phrases in each dimension of each patient to obtain the one-hot encoding value of each phrase. Then, the average of the one-hot encoding values ​​of all phrases in each dimension of each patient is used as the encoding value of the medication information of each dimension of each patient.

[0024] Among them, the jieba word segmentation tool and the one-hot encoding algorithm are both technical means well known to those skilled in the art and will not be elaborated here.

[0025] Step S2: Take any patient as the target patient, and obtain the target patient's symptom similarity based on the difference in the medication information coding value of the same dimension between the target patient and all patients; cluster all patients based on the difference in symptom similarity between patients to obtain multiple clusters; obtain the patient record priority of each patient based on the difference in symptom similarity between patients in each cluster, the overall level of symptom similarity of all patients in each cluster, the number of patients in each cluster, and the difference in illness time between patients in each cluster.

[0026] The embodiment of the present invention needs to give priority to recording the clinical medication information of epidemic patients, so as to realize the research and analysis of epidemiology. There are some non-epidemic patients in a large number of patient groups. The recording priority of clinical medication information of non-epidemic patients should be appropriately lowered to improve the efficiency of information recording. The symptoms of epidemic patients are relatively similar, and their clinical medication information and medication information coding values ​​of the same dimension are also relatively close. Therefore, the differences in medication information coding values ​​of the same dimension between each patient can be analyzed. The obtained symptom similarity reflects the similarity of the symptoms suffered by the target patient and other patients. The greater the symptom similarity, the closer the medication information coding value of each patient is to the overall level of the medication information coding values ​​of all patients, which further indicates that the patient is more likely to suffer from an epidemic.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining the symptom similarity of each patient specifically includes: First, take any dimension as the target dimension, and take the average of the medication information coding values ​​of the target dimension of all patients as the overall coding value of the target dimension. The overall coding value is used to reflect the overall level of the medication information coding values ​​of the target dimension of all patients. The overall coding value of each dimension can be obtained by the same method.

[0028] Then, any patient is taken as the target patient, and the absolute value of the difference between the target patient's target dimension medication information coding value and the overall coding value of the target dimension is negatively correlated to obtain the degree of convergence of the target patient's target dimension. The greater the degree of convergence, the closer the target patient's medication coding value is to the overall level of the medication coding values ​​of all patients in the target dimension, which further indicates that the target patient is more likely to show epidemic characteristics in the target dimension.

[0029] The same method as above can be used to obtain the degree of convergence of each dimension of the target patient, and then the accumulated value of the degree of convergence of all dimensions of the target patient can be normalized, and the calculation result can be limited to range, thereby obtaining the similarity of the symptoms of the target patients.

[0030] In one embodiment of the present invention, the normalization processing can be specifically, for example, maximum and minimum value normalization processing, and the normalization in subsequent steps can all adopt maximum and minimum value normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of numerical values, which will not be described in detail.

[0031] As an example, in one embodiment of the present invention, the expression of the target patient's symptom similarity may be specifically, for example, as follows: in, Indicates the similarity of the target patient's symptoms; The target patient The coding value of medication information in each dimension; Indicates The overall coding value of the dimension; The target patient The degree of convergence of dimensions; Indicates the number of dimensions; represents the normalization function; Indicates the preset first adjustment parameter, which is used to prevent the denominator from being 0. The value range is In one embodiment of the present invention, Set to 0.01, The specific value of can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0032] It should be noted that in other embodiments of the present invention, negative correlation mapping may be achieved through other basic mathematical operations, which will not be elaborated herein.

[0033] The similarity of symptoms of each patient can be obtained by the same method as above.

[0034] Since the symptoms of patients suffering from epidemic diseases are highly similar and close to each other, all patients can be clustered based on the differences in the similarities of symptoms between patients to obtain multiple clusters. The clusters can then be analyzed to improve the accuracy of the analysis of the possibility of patients suffering from epidemic diseases, and then accurately analyze the priority of recording the patients' clinical medication information.

[0035] Preferably, in one embodiment of the present invention, the method for obtaining multiple clusters specifically includes: The absolute value of the difference between the similarities of symptoms of any two patients is used as the distance metric between any two patients, and then the K-means clustering algorithm is used to cluster all patients based on the distance metric between any two patients to obtain multiple clusters, wherein the number of clusters can be determined by the existing elbow method. In other embodiments of the present invention, other clustering algorithms based on distance metrics can also be used to implement clustering operations, which is not limited here.

[0036] Since the more concentrated the similarity of symptoms of each patient in the cluster is and the greater the similarity of symptoms of each patient is, the greater the possibility that the patients in the cluster suffer from epidemic diseases. At the same time, epidemics are highly contagious and break out in a short period of time, resulting in a large number of epidemic patients and a relatively close illness time. Therefore, the difference in symptom similarity between patients in each cluster, the overall level of symptom similarity of all patients in each cluster, the number of patients in each cluster, and the difference in illness time between patients in each cluster can be analyzed. The priority of the obtained patient record reflects the priority of recording the clinical medication data information of each patient. Subsequently, the patient's clinical medication information can be recorded with different priorities based on the patient record priority to improve the efficiency of information recording.

[0037] Preferably, in one embodiment of the present invention, the method for obtaining the patient record priority of each patient specifically includes: See also Figure 2 , which shows a flow chart of a method for obtaining the patient record priority of each patient provided by an embodiment of the present invention.

[0038] Step S201: taking any cluster as a target cluster, and taking the average value of the disease similarities of all patients in the target cluster as the cluster center value of the target cluster.

[0039] First, any cluster is analyzed and any cluster is taken as the target cluster. From the above analysis, it can be seen that the greater the similarity of the patients' symptoms, the greater the possibility that the patients suffer from an epidemic. Therefore, the average value of the symptom similarity of all patients in the target cluster can be used as the cluster center value of the target cluster. The larger the cluster center value, the greater the overall level of symptom similarity of all patients in the target cluster, which further indicates that the patients in the target cluster are more likely to suffer from an epidemic, and the greater the priority of recording the clinical medication information of the patients in the target cluster.

[0040] Step S202: The absolute value of the difference between the symptom similarity of each patient in the target cluster and the cluster center value of the target cluster is used as the symptom similarity deviation of each patient in the target cluster, and a negative correlation mapping is performed on the average of the symptom similarity deviations of all patients in the target cluster to obtain the first clustering degree of the target cluster.

[0041] From the above analysis, it can be seen that the more concentrated or close the similarities of the symptoms of each patient in the cluster are, the greater the possibility that the patients in the cluster suffer from epidemic diseases. Therefore, the absolute value of the difference between the symptom similarity of each patient in the target cluster and the cluster center value of the target cluster can be used as the symptom similarity deviation of each patient in the target cluster. The smaller the symptom similarity deviation, the closer the symptom similarity of each patient in the target cluster is to the cluster center value. Then, the average value of the symptom similarity deviation of all patients in the target cluster can be negatively correlated to obtain the first clustering degree of the target cluster. The larger the first clustering degree, the more concentrated the symptom similarity of each patient in the target cluster is at the cluster center value, which means that the possibility that the patients in the target cluster suffer from epidemic diseases is greater, and the priority of recording the clinical medication information of the patients in the target cluster is greater.

[0042] As an example, in one embodiment of the present invention, the expression of the first clustering degree of the target cluster may be specifically, for example, as follows: in, Indicates the first clustering degree of the target cluster; Indicates the first cluster in the target cluster Similarity of symptoms among patients; Indicates the cluster center value of the target cluster; Indicates the first The similarity deviation of symptoms among patients; Represents the number of patients in the target cluster; Indicates the preset second adjustment parameter, which is used to prevent the denominator from being 0. The value range is In one embodiment of the present invention, Set to 0.01, The specific value of can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0043] It should be noted that in other embodiments of the present invention, negative correlation mapping may be achieved through other basic mathematical operations, which will not be elaborated herein.

[0044] Step S203: obtaining a second clustering degree of the target cluster according to the difference in illness duration between the patients in the target cluster.

[0045] The smaller the difference in illness time between patients in the target cluster, the more likely it is that the patients in the target cluster have the characteristics of an epidemic outbreak in a short period of time. Therefore, the second clustering degree of the target cluster can be obtained according to the difference in illness time between patients in the target cluster. The larger the second clustering degree, the more concentrated the illness time of patients in the target cluster, which further indicates that the possibility of patients in the target cluster suffering from an epidemic is greater, and the higher the priority of recording the clinical medication information of patients in the target cluster.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the second clustering degree of the target cluster specifically includes: The average of the illness time of all patients in the target cluster is taken as the overall illness time of the target cluster, and then the absolute value of the difference between the illness time of each patient in the target cluster and the overall illness time of the target cluster is taken as the deviation of the illness time of each patient in the target cluster. A smaller deviation of the illness time indicates that the illness time of each patient in the target cluster is more concentrated in the overall illness time. Then, a negative correlation mapping can be performed on the average of the illness time deviations of all patients in the target cluster to obtain the second clustering degree of the target cluster.

[0047] As an example, in one embodiment of the present invention, the expression of the second clustering degree of the target cluster may be specifically, for example, as follows: in, Indicates the second clustering degree of the target cluster; Indicates the first cluster in the target cluster The duration of illness of each patient; Represents the overall illness time of the target cluster; Indicates the first The deviation of the illness time of each patient; Represents the number of patients in the target cluster; Indicates the preset third adjustment parameter, which is used to prevent the denominator from being 0. The value range is In one embodiment of the present invention, Set to 0.01, The specific value of can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0048] It should be noted that in other embodiments of the present invention, negative correlation mapping may be achieved through other basic mathematical operations, which will not be elaborated herein.

[0049] Step S204: The first clustering degree, the second clustering degree, the cluster center value and the number of patients in the target cluster are integrated and normalized to obtain the patient record priority of each patient in the target cluster.

[0050] The larger the first clustering degree, the second clustering degree and the cluster center value of the target cluster, the greater the possibility that the patients in the target cluster suffer from an epidemic. The more contagious an epidemic is, the more patients are infected with the epidemic. Therefore, the more patients there are in the target cluster, the more likely the patient is to suffer from an epidemic. Therefore, the first clustering degree, the second clustering degree, the cluster center value and the number of patients in the target cluster can be combined and normalized, and the calculation result can be limited to The patient record priority of each patient in the target cluster is obtained. It should be noted that the patient record priority of each patient in the target cluster is the same.

[0051] In an embodiment of the present invention, the integration of the four can be achieved by calculating the sum or product of the first clustering degree, the second clustering degree, the cluster center value and the number of patients in the target cluster, which is not limited here.

[0052] As an example, in one embodiment of the present invention, the expression of the patient record priority of each patient in the target cluster may be specifically, for example, as follows: in, Represents the patient record priority of each patient in the target cluster; Indicates the first clustering degree of the target cluster; Indicates the second clustering degree of the target cluster; Indicates the cluster center value of the target cluster; Represents the number of patients in the target cluster; Represents the normalization function.

[0053] The patient record priority of each patient in each cluster can be obtained by the same method as above, thereby obtaining the patient record priority of each patient.

[0054] Step S3: According to the distribution of the medication information coding values ​​of the same dimension of all patients, the dimension record priority of each dimension is obtained; the information record priority of each dimension of each patient is obtained by combining the patient record priority and the dimension record priority.

[0055] In the multi-dimensional clinical medication information of patients, there is no shortage of redundant information in some non-critical dimensions. The information recording priority of these dimensions should be lower to improve the efficiency of recording clinical medication data. However, the consistency of the distribution of clinical medication information in these redundant dimensions in the patient population is poor, which is specifically manifested in that the distribution of the medication information coding values ​​of the redundant dimensions in the patient population is relatively discrete. On the contrary, for some more important dimensions, the distribution of their clinical medication information in the patient population is relatively consistent. Therefore, the distribution of the medication information coding values ​​of the same dimension of all patients can be analyzed, and the priority of information recording in each dimension can be reflected by the obtained dimension recording priority. Subsequently, based on the dimension recording priority and combined with the above-obtained patient record priority, the recording priority of each dimension of clinical medication information of each patient is accurately analyzed, thereby improving the efficiency of clinical medication data recording.

[0056] Preferably, in one embodiment of the present invention, the method for obtaining the dimensional record priority of each dimension specifically includes: The smaller the standard deviation of the target dimension medication information coding value of the patient group, the more consistent the distribution of clinical medication information in the target dimension. Therefore, the standard deviation of the target dimension medication information coding value of all patients is negatively correlated and normalized, and the calculation result is limited to range, thereby obtaining the dimension record priority of the target dimension.

[0057] As an example, in one embodiment of the present invention, the expression of the dimension record priority of the target dimension may be specifically, for example: in, Indicates the dimension record priority of the target dimension; represents the standard deviation of the medication information coding values ​​of the target dimension for all patients; represents the normalization function; Indicates the preset fourth adjustment parameter, which is used to prevent the denominator from being 0. The value range is In one embodiment of the present invention, Set to 0.01, The specific value of can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0058] Since the priorities for recording clinical medication information of different patients are different, and there are also differences in the priorities for recording clinical medication information of different dimensions for the same patient, the patient record priority and dimension record priority can be combined to obtain the information recording priority of each dimension for each patient. The greater the information recording priority, the greater the priority for recording the clinical medication information of this dimension for the patient. Subsequently, based on the information recording priority, the clinical medication information of different dimensions of different patients can be recorded according to different priorities to improve the efficiency of recording clinical medication data.

[0059] Preferably, in one embodiment of the present invention, the method for obtaining the information record priority of each dimension of each patient specifically includes: Normalize the product of each patient's patient record priority and each dimension's dimension record priority, and limit the calculation result to range, thereby obtaining the information record priority of each dimension for each patient.

[0060] As an example, in one embodiment of the present invention, the expression of the information record priority of each dimension of each patient may be specifically, for example, as follows: in, Indicates The patient's The information recording priority of each dimension; Indicates The patient record priority for each patient; Indicates The dimension record priority of each dimension; Represents the normalization function.

[0061] In other embodiments of the present invention, the sum of the patient record priority of each patient and the dimension record priority of each dimension may be normalized to obtain the information record priority of each dimension of each patient, which is not limited here.

[0062] Step S4: Based on the information recording priority, the clinical medication information of each dimension of each patient is recorded with different priorities.

[0063] The greater the information recording priority of a certain dimension of a patient, the more priority is required to record the clinical medication information of that dimension of the patient. Therefore, based on the information recording priority, the clinical medication information of each dimension of each patient can be recorded with different priorities, thereby improving the efficiency of clinical medication data recording for a large number of patient groups.

[0064] Preferably, in one embodiment of the present invention, the method for recording clinical medication information of each dimension of each patient with different priorities specifically includes: The clinical medication information of the patient dimension corresponding to the information recording priority is recorded in descending order of the information recording priority.

[0065] In other embodiments of the present invention, the information recording priority of each dimension of each patient can also be divided into a range, thereby dividing into multiple small ranges, and according to the small range to which the information recording priority belongs, the clinical medication information of the corresponding dimension of the corresponding patient is recorded with different priorities. The specific process is: since the value range of the information recording priority is , at this time you can It is evenly divided into five small ranges of the same length. The priorities of these five small ranges gradually increase from left to right. Then, according to the priority of the small range to which the information recording priority belongs, the clinical medication information of the corresponding dimensions of the corresponding patients is recorded with different priorities.

[0066] An embodiment of the present invention provides a clinical medication data recording system, which includes a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement the method described in steps S1 to S4.

[0067] In summary, the embodiment of the present invention first obtains the clinical medication information of different dimensions of each patient and the illness time of each patient, performs quantitative preprocessing on the clinical medication information, and obtains the medication information coding value of each dimension of each patient; obtains the symptom similarity of each patient based on the difference in the medication information coding values ​​of the same dimension between each patient; clusters all patients based on the difference in symptom similarity between each patient to obtain multiple clusters; obtains the patient record priority of each patient based on the difference in symptom similarity between each patient in each cluster, the overall level of symptom similarity of all patients in each cluster, the number of patients in each cluster, and the difference in illness time between each patient in each cluster; obtains the dimensional record priority of each dimension based on the distribution of the medication information coding values ​​of the same dimension of all patients; obtains the information record priority of each dimension of each patient by combining the patient record priority and the dimensional record priority; and records the clinical medication information of each dimension of each patient with different priorities based on the information record priority.

[0068] 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.

[0069] 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 method for recording clinical medication data, characterized in that: The method comprises: Obtain clinical medication information of different dimensions for each patient and the duration of illness of each patient, perform quantitative preprocessing on the clinical medication information, and obtain a medication information coding value of each dimension for each patient; According to the difference in the medication information coding values ​​of the same dimension between the patients, the symptom similarity of each patient is obtained; based on the difference in the symptom similarity between the patients, all patients are clustered to obtain multiple clusters; according to the difference in the symptom similarity between the patients in each cluster, the overall level of the symptom similarity of all patients in each cluster, the number of patients in each cluster, and the difference in the illness time between the patients in each cluster, the patient record priority of each patient is obtained; According to the distribution of the medication information coding values ​​of the same dimension of all patients, the dimensional record priority of each dimension is obtained; the information record priority of each dimension of each patient is obtained by combining the patient record priority and the dimensional record priority; Based on the information recording priority, the clinical medication information of each dimension of each patient is recorded with different priorities.

2. A clinical medication data recording method according to claim 1, characterized in that: The obtaining of the similarity of symptoms of each patient comprises: Taking any dimension as the target dimension, taking the average of the medication information coding values ​​of the target dimension of all patients as the overall coding value of the target dimension; Taking any patient as a target patient, negatively correlating the absolute value of the difference between the medication information coding value of the target dimension of the target patient and the overall coding value of the target dimension, and obtaining the degree of convergence of the target dimension of the target patient; The accumulated values ​​of the approximation degrees of all dimensions of the target patient are normalized to obtain the symptom similarity of the target patient.

3. A clinical medication data recording method according to claim 1, characterized in that: The obtaining of multiple clusters comprises: The absolute value of the difference between the similarities of the symptoms of any two patients is used as the distance measure between any two patients; Using the K-means clustering algorithm, all patients are clustered based on the distance metric between any two patients to obtain multiple clusters.

4. A clinical medication data recording method according to claim 1, characterized in that: The obtaining of each patient's patient record priority includes: Taking any cluster as a target cluster, and taking the average of the disease similarities of all patients in the target cluster as the cluster center value of the target cluster; The absolute value of the difference between the disease similarity of each patient in the target cluster and the cluster center value of the target cluster is used as the disease similarity deviation of each patient in the target cluster, and a negative correlation mapping is performed on the average value of the disease similarity deviation of all patients in the target cluster to obtain a first clustering degree of the target cluster; According to the difference in illness time between the patients in the target cluster, obtaining a second clustering degree of the target cluster; The first clustering degree, the second clustering degree, the cluster center value and the number of patients in the target clustering cluster are integrated and normalized to obtain the patient record priority of each patient in the target clustering cluster.

5. A clinical medication data recording method according to claim 4, characterized in that: The obtaining of the second clustering degree of the target cluster comprises: The average of the illness time of all patients in the target cluster is taken as the overall illness time of the target cluster; The absolute value of the difference between the illness time of each patient in the target cluster and the overall illness time of the target cluster is used as the illness time deviation of each patient in the target cluster, and a negative correlation mapping is performed on the average value of the illness time deviation of all patients in the target cluster to obtain the second clustering degree of the target cluster.

6. A clinical medication data recording method according to claim 2, characterized in that: The step of obtaining the dimension record priority of each dimension includes: The standard deviation of the medication information coding value of the target dimension of all patients is negatively correlated and normalized to obtain the dimension record priority of the target dimension.

7. A clinical medication data recording method according to claim 1, characterized in that: The obtaining of the information record priority of each dimension of each patient includes: The product value of the patient record priority of each patient and the dimension record priority of each dimension is normalized to obtain the information record priority of each dimension of each patient.

8. A clinical medication data recording method according to claim 1, characterized in that: The recording of the clinical medication information of each dimension of each patient with different priorities includes: The clinical medication information of the patient dimension corresponding to the information recording priority is recorded in descending order of the information recording priority.

9. A clinical medication data recording method according to claim 1, characterized in that: The step of obtaining the medication information coding value of each dimension of each patient includes: Using the Jieba word segmentation tool, the clinical medication information of each dimension of each patient is segmented to obtain multiple phrases in each dimension of each patient; Use the one-hot encoding algorithm to encode multiple phrases in each dimension of each patient to obtain the one-hot encoding value of each phrase; The average of the one-hot encoding values ​​of all phrases in each dimension of each patient is used as the encoding value of the medication information of each dimension of each patient.

10. A clinical medication data recording system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images