A Medical Care Data Sharing Method and System

Abstract

The present invention relates to the technical field of medical data processing, and provides a medical care data sharing method and system, including: obtaining the medical care data of a number of patients and the relevant data of each medical node; classifying the diagnosis and treatment data of the same type for different patients, and quantifying the necessity of analyzing the diagnosis and treatment data of each type for each patient; determining the analysis timeliness of the diagnosis and treatment data of each type for each patient for each medical node according to the time series change of the diagnosis and treatment data of each type for the patient; analyzing the corresponding relationship between each medical node and the diagnosis and treatment data of each type for the patient, combining the necessity of analysis and the distribution of medical nodes, and limiting by the analysis timeliness, to determine the sharing necessity of the diagnosis and treatment data of each type for the patient for each medical node; and sharing the medical care data of a number of patients accordingly. The present invention aims to solve the problem that the sharing and calling between medical institutions are affected due to the timeliness difference of medical care data.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and particularly to a method and system for sharing medical care data. Background Art

[0002] Medical care data includes patients' medical data and nursing data. Medical data mainly includes various types of physiological data covered by patients' medical records and diagnosis and treatment information, while nursing data includes various types of physiological data during the nursing process after patient treatment and relevant data on some patients' living habits; through the sharing of medical care data, multi-departments or multi-medical institutions analyze the medical data and nursing data of a large number of patients, which helps the communication between hospital departments and the assessment of disease risks, and then formulate relevant treatment plans and subsequent nursing plans for different patients.

[0003] During the process of sharing medical care data, patients' medical care data is usually continuous monitoring of various types of physiological data, and its data volume is huge. It is necessary to selectively distribute and store the medical care data according to the needs of each node (department, medical institution) for medical care data, so as to improve the calling efficiency of medical care data during the analysis process, and then improve the sharing efficiency of medical care data; however, the medical care data required by different nodes corresponding to departments or medical institutions is different, that is, different types of medical care data are required. At the same time, different types of medical care data have timeliness differences based on patients' diseases, so it is necessary to avoid the impact of the timeliness differences of medical care data on the calling of various types of medical care data by different nodes, and then reduce the sharing efficiency of medical care data. Summary of the Invention

[0004] The present invention provides a method and system for sharing medical care data to solve the problem that the existing medical care data affects the sharing and calling between medical institutions due to timeliness differences. The specific technical solutions adopted are as follows:

[0005] The present invention proposes a method for sharing medical care data, which includes the following steps:

[0006] Obtain the medical care data of several patients and the relevant data of each medical node, and the medical care data of the patients includes several types of diagnosis and treatment data;

[0007] Classify the diagnosis and treatment data of the same type of different patients, analyze the differences between the diagnosis and treatment data of each type of the same patient in the classification results, and quantify the necessity of analyzing the diagnosis and treatment data of each patient for each type; according to the time-series changes of the diagnosis and treatment data of each patient for each type, combined with the relevant data of each medical node, determine the analysis timeliness of the diagnosis and treatment data of each patient for each medical node;

[0008] Analyze the correspondence between the relevant data of each medical node and the diagnostic and treatment data of each type of patient, combine the necessity of analyzing the diagnostic and treatment data of each type of patient and the distribution of medical nodes, and limit the analysis timeliness of each type of diagnostic and treatment data for each medical node to determine the necessity of sharing the diagnostic and treatment data of each type of patient for each medical node;

[0009] Share the medical care data of several patients based on the sharing necessity.

[0010] Optionally, the specific method for classifying the diagnostic and treatment data of the same type for different patients includes:

[0011] For the diagnostic and treatment data corresponding to the physiological data obtained by one-time detection, perform numerical processing on the diagnostic and treatment data of this type for all patients, and use it as the representative data of the diagnostic and treatment data of each patient of this type;

[0012] For the diagnostic and treatment data corresponding to the physiological data obtained by continuous monitoring, obtain the median of the standard range of the diagnostic and treatment data of this type and use it as the standard value of the diagnostic and treatment data of this type. Based on the deviation value between the diagnostic and treatment data of this type of any patient and the standard value, through numerical processing and select the numerical processing result corresponding to the largest absolute deviation value, as the representative data of the diagnostic and treatment data of this patient of this type;

[0013] Perform density clustering on the representative data of the diagnostic and treatment data of any type of all patients, use the absolute value of the difference between the representative data of patients as the distance metric, and obtain several clusters of the diagnostic and treatment data of this type; The cluster with the largest number of representative data among all clusters is used as the standard cluster of the diagnostic and treatment data of this type.

[0014] Optionally, the specific method for analyzing the differences between the classification results of the diagnostic and treatment data of each type of the same patient and quantifying the necessity of analyzing the diagnostic and treatment data of each type of each patient includes:

[0015]

[0016] Among them, represents the necessity of analyzing the diagnostic and treatment data of the th type of the th patient, represents the number of types of the diagnostic and treatment data of the th patient, represents the absolute value of the difference between the representative data of the th type of the th patient and the centroid of the cluster to which it belongs in the cluster, represents the th patient, The absolute value of the difference between the representative data of each type of diagnosis and treatment data and the centroid of the cluster to which it belongs within the cluster denotes the th patient's th type of diagnosis and treatment data's representative data and the th patient's th type of diagnosis and treatment data's mean value of all representative data in the standard cluster, denotes the th patient's th type of diagnosis and treatment data's representative data and the th patient's th type of diagnosis and treatment data's mean value of all representative data in the standard cluster, denotes the linear normalization function.

[0017] Optionally, for the analysis timeliness of each type of diagnosis and treatment data of the patient for each medical node, the specific acquisition method is as follows:

[0018] Take the mean value of the duration from the visit to the discharge of several patients collected by the th medical node for the medical care data as the retention time of the medical care data under the th medical node;

[0019] Based on the time series changes of each type of diagnosis and treatment data of the patient within the retention time of the medical care data under each medical node, obtain the analysis timeliness of each type of diagnosis and treatment data of the patient for each medical node.

[0020] Optionally, the method for obtaining the analysis timeliness of each type of diagnosis and treatment data of the patient for each medical node based on the time series changes of each type of diagnosis and treatment data of the patient within the retention time of the medical care data under each medical node includes the following specific methods:

[0021] Starting from the visit of the th patient, until the end of the retention time of the medical care data under the th medical node, record the th patient's th type of diagnosis and treatment data during this period as the diagnosis and treatment sequence of the th patient's th type of diagnosis and treatment data under the th medical node;

[0022] Count the number of data points in the diagnosis and treatment sequence that exceed the th patient's The quantity of the medical data within the standard range of each type of medical data, and using the last medical data in the medical sequence as the right boundary of the window, intercept a sequence with a reference window as the medical reference sequence of the th patient's th type of medical data at the th medical node;

[0023] Based on the retention time of the medical care data at the th medical node, combined with the proportion of the medical data that exceeds the standard range of the th patient's th type of medical data, and the fluctuation of the medical data in the medical reference sequence, obtain the analysis timeliness of the th patient's th type of medical data for the th medical node; The analysis timeliness has a positive correlation with the retention time, the proportion, and the fluctuation.

[0024] Optionally, analyze the corresponding relationship between the relevant data of each medical node and the medical data of each type of the patient, combine the analysis necessity of the medical data of each type of each patient and the distribution of the medical nodes, and determine the sharing necessity of the medical data of each type of the patient for each medical node by limiting through the analysis timeliness of each type of medical data for each medical node. The specific methods included are:

[0025] Based on the corresponding relationship between the relevant data of each medical node and the medical data of each type of the patient and its analysis necessity, obtain the storage necessity of the medical data of each type of the patient for each medical node;

[0026] According to the differences between the analysis timeliness of each type of medical data for different medical nodes, combined with the distribution of the medical nodes, quantify the sharing factor between different medical nodes for each type of medical data;

[0027] On the basis of the storage necessity, analyze the sharing factor to determine the sharing necessity of the medical data of each type of the patient for each medical node.

[0028] Optionally, the specific methods included in obtaining the storage necessity of the medical data of each type of the patient for each medical node are:

[0029] Based on the inclusion relationship of the types of medical treatment data of patients on medical nodes, and the proportion of the types of medical treatment data included in both patients and medical nodes in the total amount of medical treatment data types of patients and medical nodes, combined with the analysis necessity of various types of medical treatment data of patients, the storage necessity of various types of medical treatment data of patients for each medical node is obtained; the storage necessity is positively correlated with both the proportion and the analysis necessity.

[0030] Optionally, the specific method for quantifying the sharing factors of various types of medical treatment data for different medical nodes includes:

[0031] Based on the inclusion relationship of the same type of medical treatment data of patients for two medical nodes and the difference in the analysis necessity of this type of medical treatment data for the two medical nodes, combined with the spatial distance between the two medical nodes and the proportion of the quantity of the types of medical treatment data they jointly contain, the sharing factor of the same type of medical treatment data of patients for the two medical nodes is obtained; the sharing factor is negatively correlated with the difference in the analysis necessity and the spatial distance, and the sharing factor is positively correlated with the proportion of the quantity.

[0032] Optionally, the specific method for obtaining the sharing necessity of various types of medical treatment data of patients for each medical node is:

[0033] Taking the th patient's th type of medical treatment data as an example, for several medical nodes among all medical nodes where the storage necessity of this type of medical treatment data is not equal to 0, they are arranged in ascending order of storage necessity to obtain the node storage sequence of this type of medical treatment data. The sharing necessity of this type of medical treatment data for the first medical node in its node storage sequence is calculated as follows:

[0034]

[0035] Among them, represents the storage necessity of the th patient's th type of medical treatment data for the first medical node in its node storage sequence, represents the number of medical nodes in the node storage sequence of the th patient's th type of medical treatment data, represents the sharing factor of the th patient's th type of medical treatment data for the first medical node and the th medical node in its node storage sequence, represents the The type of diagnosis and treatment data for the th medical node in its node storage sequence; represents a weight normalization function;

[0036] After obtaining the corresponding sharing necessity for the first medical node in the node storage sequence, calculate the sharing necessity corresponding to the second medical node, where the medical nodes whose storage necessity has been updated to sharing necessity during the calculation process participate in the calculation with the sharing necessity;

[0037] And so on, to obtain the th patient's type of diagnosis and treatment data for each medical node in its node storage sequence.

[0038] The present invention also proposes a medical care data sharing system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0039] The beneficial effects of the present invention are as follows: First, through the deviation analysis of the same type of diagnosis and treatment data, the characteristic performance of the diagnosis and treatment data for the patient's disease is reflected, and the analysis necessity of the diagnosis and treatment data is quantified; at the same time, the temporal change of the diagnosis and treatment data and its retention time at the corresponding medical node are analyzed to determine the analysis timeliness of the diagnosis and treatment data for each medical node, providing a judgment basis for the subsequent sharing of the diagnosis and treatment data between medical nodes; through the analysis of the corresponding relationship between each type of diagnosis and treatment data of the patient and the relevant data of the medical node, and the characteristic performance of the diagnosis and treatment data reflected by the analysis necessity, the storage relationship of the diagnosis and treatment data in a single medical node is quantified, and then by analyzing the distribution of different medical nodes and combining the differences in their analysis timeliness for the diagnosis and treatment data, the transmission relationship of the diagnosis and treatment data between different medical nodes is determined, so as to adjust the storage relationship of the diagnosis and treatment data in the medical node and obtain the sharing necessity, so as to improve the utilization efficiency of medical care data sharing; and then the shared storage of the diagnosis and treatment data in the patient's medical care data is carried out, effectively improving the communication between medical nodes and the utilization of their storage space, and at the same time giving full play to the shared transmission relationship between medical care data, realizing the efficient and effective sharing of medical care data. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] To more clearly illustrate the technical solutions 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 drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 Schematic flowchart of a medical care data sharing method provided by an embodiment of the present invention. Specific implementation manners

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0043] Please refer to Figure 1 , which shows a flowchart of a medical care data sharing method provided by an embodiment of the present invention. The method includes the following steps:

[0044] Step S001: Obtain the medical care data of several patients and the relevant data of each medical node.

[0045] The purpose of this embodiment is that during the storage process of medical care data, different medical nodes (departments or medical institutions) have different requirements for different types of diagnosis and treatment data of patients. At the same time, during the continuous monitoring of various types of diagnosis and treatment data, there are differences in the analysis timeliness for each medical node. Therefore, it is necessary to obtain the medical care data of patients, including the continuous monitoring of various types of physiological data in the patient's medical records and treatment records, as well as various types of care data composed of living habits during the subsequent care process of the patient; at the same time, it is necessary to record in which department the patient's medical care data is collected, and the physiological data and care data required by each department (medical institution).

[0046] Specifically, after a patient is admitted to the hospital for treatment, various types of data collected from their medical records, treatment records, and post-treatment care process together constitute the patient's medical care data. Among them, the personal information in the medical records includes multiple types of data, which are directly used as the patient's multiple types of diagnosis and treatment data. Since personal information does not change, in subsequent processing, the retention time of the multiple types of diagnosis and treatment data corresponding to personal information is processed based on the maximum retention time. In the medical records and treatment records, the patient is required to detect multiple types of physiological data in the corresponding department. The detection frequency is obtained according to the collection frequency of the corresponding type of physiological data, and the multiple types of physiological data are used as the patient's multiple types of diagnosis and treatment data. In addition to the various types of physiological data continuously monitored in the medical records and treatment records during the post-treatment care process, it also includes the patient's diet type (such as liquid diet, infusion, normal diet, dietary restrictions, etc., presented as data through classification and labeling), toilet-related records, etc., all of which are used as the patient's multiple types of diagnosis and treatment data, thus obtaining the multiple types of diagnosis and treatment data in the patient's medical care data.

[0047] Furthermore, multiple hospitals and medical institutions jointly share medical care data. Based on medical care big data analysis and joint consultation, each department in each hospital and each medical institution is regarded as a medical node. Based on the departments and medical institutions corresponding to each medical node, the types of diagnosis and treatment data it needs are obtained, that is, the types of the patient's diagnosis and treatment data required in the diagnosis, treatment, and care process of the department or medical institution, and these are used as the relevant data of each medical node.

[0048] It should be noted that there are differences in the retention time of the same type of diagnosis and treatment data in different departments, which are affected by the corresponding relationship between the signs reflected by each type of diagnosis and treatment data in each department and the department. For example, the emergency department needs to monitor physiological data that can more intuitively reflect the patient's physical condition, such as blood pressure, blood oxygen, body temperature, etc., compared with ordinary departments. Different departments need to monitor more professional physiological data on the basis of conventional physiological data. At the same time, the retention time of the patient's medical care data in the emergency department is relatively short compared with that in ordinary departments, while ordinary departments need long-term monitoring. Thus, the effective sharing of medical care data is carried out to improve the effective utilization of the patient's medical care data and the calling efficiency of each medical node for medical care data.

[0049] Step S002: Classify the same type of diagnosis and treatment data of different patients, analyze the differences between the various types of diagnosis and treatment data of the same patient among the classification results, and quantify the necessity of analyzing the various types of diagnosis and treatment data of each patient. Based on the time-series changes of the various types of diagnosis and treatment data of the patient, combined with the relevant data of each medical node, determine the analysis timeliness of the various types of diagnosis and treatment data of the patient for each medical node.

[0050] Preferably, in an embodiment of the present invention, the diagnostic and treatment data of the same type for different patients are classified, the differences between the diagnostic and treatment data of each type for the same patient among the classification results are analyzed, and the necessity of analyzing the diagnostic and treatment data of each patient and each type is quantified. The specific methods include:

[0051] It should be noted that usually, the values of the same physiological index are close to each other under normal circumstances. The greater the deviation of the physiological index, the more it can reflect certain physiological characteristics or disease characteristics of the patient. Therefore, by clustering the diagnostic and treatment data (physiological indices) of the same type and based on the clustering results, combined with the changes between the cluster categories to which the patient belongs in different types of diagnostic and treatment data, the greater the deviation from the normal cluster category (the cluster category with the largest number of data points in the same type of diagnostic and treatment data), and the greater the change in the cluster categories to which different types of diagnostic and treatment data of the same patient belong, the more likely the corresponding type of diagnostic and treatment data is the manifestation characteristic of certain disease symptoms, and accordingly, the more necessary it is to analyze.

[0052] Specifically, among the diagnostic and treatment data of each type of the patient, it is divided into physiological data obtained by one-time detection and physiological data obtained by continuous monitoring; for the diagnostic and treatment data corresponding to the physiological data obtained by one-time detection, the diagnostic and treatment data of this type for all patients are numerically processed and used as the representative data of the diagnostic and treatment data of each patient for this type; for the diagnostic and treatment data corresponding to the physiological data obtained by continuous monitoring, the median of the standard range of the diagnostic and treatment data of this type is obtained and used as the standard value of the diagnostic and treatment data of this type. The diagnostic and treatment data of any one patient for this type are respectively subtracted from the standard value to obtain several deviation values of the diagnostic and treatment data of this patient for this type (the difference obtained by subtracting the standard value from the patient's data). The deviation values of the diagnostic and treatment data of this type for all patients are numerically processed, and the numerically processed result corresponding to the deviation value with the largest absolute value among the several deviation values of the diagnostic and treatment data of this patient for this type is used as the representative data of the diagnostic and treatment data of this patient for this type.

[0053] Furthermore, taking the diagnostic and treatment data of any one type as an example, DBSCAN clustering is performed on the representative data of the diagnostic and treatment data of this type for all patients (if a patient does not have the diagnostic and treatment data of this type, clustering is not performed). The distance metric uses the absolute value of the difference between the representative data of the patients to obtain several clusters of the diagnostic and treatment data of this type; the cluster with the largest number of representative data among all clusters is used as the standard cluster of the diagnostic and treatment data of this type.

[0054] Furthermore, for the th patient's th type of diagnostic and treatment data, the necessity of its analysis is calculated as follows:

[0055]

[0056] Among them, represents the number of types of diagnosis and treatment data of the th patient, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the centroid of the cluster to which it belongs in the cluster, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the centroid of the cluster to which it belongs in the cluster, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the mean value of all representative data in the standard cluster of the th type of diagnosis and treatment data of the th patient, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the mean value of all representative data in the standard cluster of the th type of diagnosis and treatment data of the th patient, represents a linear normalization function, and the normalization object is the th patient's diagnosis and treatment data of all types except the th type .

[0057] It should be noted that the greater the difference between the representative data and the standard cluster, and the greater the change in the difference between the representative data of other types and the standard cluster, the more the diagnosis and treatment data of this type can reflect the disease symptoms of the patient, and the more necessary it is for analysis; at the same time, taking the difference in the degree of outlier of the representative data of different types of diagnosis and treatment data in their respective clusters as the weight, the greater the degree of outlier of this type compared to the degree of outlier of other types, the higher the credibility of the deviation between the corresponding representative data and the standard cluster, and vice versa, if it is smaller or even smaller than the degree of outlier of other types, then the deviation between the corresponding representative data and the standard cluster reflects the characteristics of the disease symptoms weaker, and the corresponding weight is smaller.

[0058] Preferably, in an embodiment of the present invention, according to the time series change of the diagnosis and treatment data of each type of the patient, combined with the relevant data of each medical node, the analysis timeliness of the diagnosis and treatment data of each type of the patient for each medical node is determined. The specific method included is:

[0059] It should be noted that the diagnosis and treatment data of each type of patient is time-sensitive for each medical node, that is, each medical node sets different retention times for the diagnosis and treatment data of each type of patient, depending on the corresponding relationship between the type of diagnosis and treatment data and the relevant data of the medical node. The more the corresponding type of diagnosis and treatment data can reflect the disease characteristics covered by the relevant medical node, the stronger the corresponding relationship, and the greater the impact on timeliness. At the same time, for the diagnosis and treatment data corresponding to the continuously monitored physiological data, the temporal variation of the diagnosis and treatment data itself also has a certain timeliness. When the deviation of the diagnosis and treatment data exceeds the standard range of the corresponding type of diagnosis and treatment data and the fluctuation of the diagnosis and treatment data is large, its analyzability is strong. When the fluctuation gradually decreases and the deviation is within the standard range, after a period of time, the corresponding type of diagnosis and treatment data tends to be normal and its analyzability weakens. Combining the corresponding relationship between the type of diagnosis and treatment data and the medical node, the timeliness is quantitatively analyzed.

[0060] Specifically, taking the th medical node as an example, the mean value of the length of stay from the visit to discharge of several patients collected by the th medical node, that is, several patients who seek medical treatment in the department or medical structure corresponding to the th medical node, is used as the retention time of the medical care data under the th medical node.

[0061] Furthermore, a reference window is preset. In this embodiment, the reference window is described by taking the length of 10 pieces of diagnosis and treatment data as an example; taking the th patient's th type of diagnosis and treatment data as an example, starting from the visit of the th patient and ending at the retention time of the medical care data under the th medical node (a period of time is obtained from the start time and a duration), the th patient's th type of diagnosis and treatment data during this period is recorded as the diagnosis and treatment sequence of the th patient's th type of diagnosis and treatment data under the th medical node; it should be noted that the th patient's th type of diagnosis and treatment data is of the continuously monitored physiological data type; count the number of pieces of diagnosis and treatment data in the diagnosis and treatment sequence that exceed the standard range of the th patient's th type of diagnosis and treatment data. At the same time, taking the last piece of diagnosis and treatment data in the diagnosis and treatment sequence as the right boundary of the window, a sequence is intercepted with the reference window as the th patient's th type of diagnosis and treatment data under the The diagnosis and treatment reference sequence under each medical node.

[0062] Furthermore, based on The retention time of medical care data under each medical node, combined with the retention time of medical care data exceeding the first Patient No. The proportion of the standard range of the diagnosis and treatment data of each type, as well as the fluctuation of the diagnosis and treatment data in the diagnosis and treatment reference sequence, are obtained. The patient's Types of medical data for The analysis timeliness of each medical node.

[0063] As an example, The patient's Types of medical data for Analysis timeliness of medical nodes The calculation method is:

[0064]

[0065] in, Indicates The retention time of medical care data under each medical node, Indicates The patient's Types of medical data are in The number of diagnosis and treatment data in the diagnosis and treatment sequence under the medical node, Indicates that the diagnosis and treatment sequence exceeds the Patient No. The number of medical data within the standard range for each type of medical data, Indicates The patient's Types of medical data are in The variance of all diagnosis and treatment data in the diagnosis and treatment reference sequence under the medical node; represents an exponential function with a natural constant as the base. In this embodiment, Model to present inverse proportional relationship and normalization processing, As the input of the model, the implementer can set the inverse proportional function and normalization function according to the actual situation; is a linear normalization function, and the normalization object is the retention time of medical care data under each medical node; in particular, if any type of diagnosis and treatment data of a patient is obtained once, then its analysis timeliness for the corresponding medical node is the normalized result of the retention time of the medical care data under the corresponding node, with the first For example, a medical node .

[0066] It should be noted that in the quantification process of analyzing timeliness, the longer the retention time of the medical node itself, the greater the deviation of the patient's diagnosis and treatment data during the retention time, and the longer the duration under abnormal data monitoring, the more valuable the corresponding type of diagnosis and treatment data for analysis, the greater the analysis timeliness. At the same time, if the last segment of the diagnosis and treatment data in the diagnosis and treatment sequence still fluctuates greatly, that is, the variance is larger, it will still improve the analysis value of the corresponding type of diagnosis and treatment data, and the analysis timeliness will be greater.

[0067] So far, first, through the deviation analysis of the same type of diagnosis and treatment data, the characteristic performance of the diagnosis and treatment data for the patient's disease is reflected, and the necessity of analyzing the diagnosis and treatment data is quantified; at the same time, the time series change of the diagnosis and treatment data and its retention time at the corresponding medical node are analyzed to determine the analysis timeliness of the diagnosis and treatment data for each medical node, providing a judgment basis for the sharing of diagnosis and treatment data between subsequent medical nodes.

[0068] Step S003: Analyze the corresponding relationship between the relevant data of each medical node and the patient's various types of diagnosis and treatment data, combine the necessity of analyzing the patient's various types of diagnosis and treatment data and the distribution of medical nodes, and limit through the analysis timeliness of various types of diagnosis and treatment data for each medical node to determine the sharing necessity of the patient's various types of diagnosis and treatment data for each medical node.

[0069] Preferably, in an embodiment of the present invention, the specific method included in this step is:

[0070] Based on the corresponding relationship between the relevant data of each medical node and the patient's various types of diagnosis and treatment data and its analysis necessity, obtain the storage necessity of the patient's various types of diagnosis and treatment data for each medical node;

[0071] According to the differences in the analysis timeliness of various types of diagnosis and treatment data for different medical nodes, combined with the distribution of medical nodes, quantify the sharing factors of various types of diagnosis and treatment data between different medical nodes;

[0072] On the basis of the storage necessity, analyze the sharing factors to determine the sharing necessity of the patient's various types of diagnosis and treatment data for each medical node.

[0073] As an example, based on the corresponding relationship between the relevant data of each medical node and the patient's various types of diagnosis and treatment data and its analysis necessity, obtaining the storage necessity of the patient's various types of diagnosis and treatment data for each medical node, the specific method included is:

[0074] It should be noted that the necessity of analyzing various types of patient diagnosis and treatment data itself reflects the characteristics of the corresponding types of diagnosis and treatment data for the patient's disease symptoms. On this basis, if the relevant data of each medical node contains the corresponding type of diagnosis and treatment data, and the other types of patient diagnosis and treatment data are more similar to the relevant data of the corresponding medical node, the medical node is more required to store the corresponding type of patient diagnosis and treatment data for the analysis of relevant medical care data of the patient, enriching the patient database samples of medical institutions or departments, and thus obtaining the storage necessity.

[0075] Specifically, based on the inclusion relationship of the patient's diagnosis and treatment data types on the medical node, and the proportion of the diagnosis and treatment data types included in both the patient and the medical node in the total amount of the patient's and medical node's diagnosis and treatment data types, combined with the necessity of analyzing various types of patient diagnosis and treatment data, the storage necessity of various types of patient diagnosis and treatment data for each medical node is obtained.

[0076] As an example, for the th patient's th type of diagnosis and treatment data and the th medical node, the medical care data of the th patient contains types of diagnosis and treatment data, and the relevant data of the th medical node includes a total of types of diagnosis and treatment data. Then, the storage necessity of the th patient's th type of diagnosis and treatment data for the th medical node is calculated as follows:

[0077]

[0078] Among them, represents the necessity of analyzing the th patient's th type of diagnosis and treatment data. represents the parameter of the th medical node including the th patient's th type of diagnosis and treatment data. If it includes, then , otherwise if it does not include, then ; represents the number of types of diagnosis and treatment data included in both the th patient and the th medical node. represents the number of types of diagnosis and treatment data included in the relevant data of the th medical node. represents the The number of types of diagnosis and treatment data included in the medical care data of a patient.

[0079] It should be noted that during the storage necessity analysis process, if the diagnosis and treatment data of this type of the patient is not included in the corresponding medical node, the storage necessity is directly 0, that is, there is no need for storage analysis; and the closer the diagnosis and treatment data included in the patient's medical care data is to the diagnosis and treatment data included in the relevant data of the medical node as a whole, that is, the larger the proportion of those included in both, and at the same time the greater the analysis necessity of the diagnosis and treatment data itself, the greater the corresponding storage necessity, and the more the corresponding node needs to store the diagnosis and treatment data for analysis.

[0080] It should be noted that the storage necessity targets the characteristics of whether to store each type of diagnosis and treatment data of a single medical node, while the sharing of medical care data can ensure data transmission between multiple medical nodes. Then, it is necessary to analyze the distribution between medical nodes and the similarity degree between relevant data, which can reflect the data transmission relationship between different medical nodes; at the same time, there are also differences in the analysis timeliness of the same type of diagnosis and treatment data of the same patient by different medical nodes. The greater the difference in analysis timeliness, it will interfere with the normal data transmission relationship, that is, the corresponding type of diagnosis and treatment data in a medical node is close to losing timeliness, and the role of sharing this type of diagnosis and treatment data by other medical nodes for analysis will become worse, and the sharing factor will decrease based on the transmission relationship.

[0081] As an example, based on the differences in the analysis timeliness of each type of diagnosis and treatment data for different medical nodes, combined with the distribution of medical nodes, to quantify the sharing factor of each type of diagnosis and treatment data between different medical nodes, the specific method included is:

[0082] Specifically, based on the inclusion relationship of the same type of diagnosis and treatment data of a patient by two medical nodes and the difference in the analysis necessity of this type of diagnosis and treatment data for the two medical nodes, combined with the spatial distance between the two medical nodes and the proportion of the number of the same type of diagnosis and treatment data they commonly contain, the sharing factor of the same type of diagnosis and treatment data of the patient between the two medical nodes is obtained.

[0083] As an example, for the th medical node and the th medical node, and the st patient's rd type of diagnosis and treatment data, the spatial distance (actual geographical location distance) between the th medical node and the th medical node is obtained. Then, the st patient's rd type of diagnosis and treatment data for the th medical node and the Sharing factor between medical nodes The calculation method is as follows:

[0084]

[0085] Among them, represents that the th medical node and the th medical node both include the parameters of the th type of diagnosis and treatment data of the th patient. If both include, then , otherwise if there is a medical node that does not include, then ; represents the spatial distance between the th medical node and the th medical node; To avoid the problem that the value of the exponential function is too small, this embodiment uses for description; represents the exponential function with the natural constant as the base. This embodiment uses model to present the inverse proportional relationship and normalization processing. is the input of the model, and the implementer can set the inverse proportional function and normalization function according to the actual situation; represents the number of types of diagnosis and treatment data included in the relevant data of the th medical node. represents the number of types of diagnosis and treatment data included in the relevant data of the th medical node. represents the number of types of diagnosis and treatment data that are both included in the relevant data of the th medical node and the th medical node; represents the absolute value function. and respectively represent the analysis timeliness of the th type of diagnosis and treatment data of the th patient for the th medical node and the th medical node.

[0086] It should be noted that if the inclusion relationships of two medical nodes with respect to the same type of diagnosis and treatment data are different, then the two nodes do not have a sharing relationship for this type of diagnosis and treatment data, and the sharing factor is 0; the smaller the spatial distance between two medical nodes, the more convenient the data transmission between the two medical nodes, which is conducive to the sharing of diagnosis and treatment data; and the more types of diagnosis and treatment data jointly included by the two medical nodes, and the larger the proportion in the total number of types of diagnosis and treatment data of the two medical nodes as a whole, the closer to 1, the more similar the diseases treated by the two medical nodes themselves, the more conducive to the communication between medical nodes, and the corresponding sharing factor is larger; at the same time, the greater the difference in the analysis timeliness between medical nodes, the more unfavorable it is to the sharing of medical care data, and it is necessary to store them separately in a timely manner to ensure the timeliness of diagnosis and treatment data.

[0087] As an example, on the basis of the storage necessity, analyze the sharing factor to determine the sharing necessity of each type of diagnosis and treatment data of the patient for each medical node. The specific method includes:

[0088] It should be noted that the sharing factor reflects the data transmission relationship between different medical nodes. Then, based on the sharing factor of the transmission relationship between a medical node and other medical nodes, the storage necessity of this medical node can be adjusted. If the sharing factors with multiple medical nodes are all large, and the storage necessities of other medical nodes for the corresponding type of diagnosis and treatment data of the corresponding patient are also large, then the storage necessity of the corresponding medical node can be appropriately reduced to combine the sharing relationship of medical care data and improve the utilization efficiency of data transmission between medical nodes, and finally obtain the sharing necessity of each type of diagnosis and treatment data of the patient for each medical node.

[0089] Specifically, taking the th patient's th type of diagnosis and treatment data as an example, arrange the several medical nodes with non-zero storage necessities of this type of diagnosis and treatment data for all medical nodes in ascending order of storage necessity to obtain the node storage sequence of this type of diagnosis and treatment data. Then the sharing necessity of this type of diagnosis and treatment data for the first medical node in its node storage sequence is calculated as follows:

[0090]

[0091] Among them, represents the storage necessity of the th patient's th type of diagnosis and treatment data for the first medical node in its node storage sequence, represents the number of medical nodes in the node storage sequence of the th patient's th type of diagnosis and treatment data, Indicates the th patient's th type of diagnosis and treatment data for the sharing factor between the first medical node and the th medical node in its node storage sequence; Indicates the th patient's th type of diagnosis and treatment data for the storage necessity of the th medical node in its node storage sequence; Indicates a weight normalization function, and the normalization object is the th patient's th type of diagnosis and treatment data for the sharing factors between the first medical node and each medical node in its node storage sequence; during the traversal process Indicates starting the traversal from the first medical node but not traversing the first medical node, that is, skipping the first medical node and directly starting the traversal from the second medical node.

[0092] Further, after obtaining the corresponding sharing necessity for the first medical node in the node storage sequence, calculate the sharing necessity corresponding to the second medical node. Among them, the medical nodes whose storage necessity has been updated to sharing necessity during the calculation process participate in the calculation with the sharing necessity. At the same time, during the traversal process of other medical nodes in the node storage sequence, the medical node being calculated itself is not traversed, that is, the second medical node is not traversed during the calculation of the sharing necessity of the second medical node; and so on, to obtain the th patient's th type of diagnosis and treatment data for the sharing necessity of each medical node in its node storage sequence, and for the medical nodes with a storage necessity of 0, the sharing necessity is also 0.

[0093] Further, obtain the sharing necessity of the diagnosis and treatment data of each patient and each type for each medical node according to the above method.

[0094] It should be noted that based on the storage necessity, using the sharing factor as the weight, quantify the impact of the storage necessity of other medical nodes on this medical node. The larger the sharing factor and the greater the storage necessity of the corresponding medical node, the more this medical node can analyze the corresponding type of diagnosis and treatment data through sharing. Then, it is necessary to adjust the storage necessity smaller, and use the mean value of the difference obtained from the storage necessity and the 1 - sharing relationship as the corresponding sharing necessity.

[0095] So far, through the analysis of the correspondence between the diagnosis and treatment data of various types of patients and the relevant data of medical nodes, as well as the characteristic manifestations of the diagnosis and treatment data reflected by the analysis necessity, the storage relationship of the diagnosis and treatment data in a single medical node is quantified. Then, by analyzing the distribution of different medical nodes and combining the differences in the analysis timeliness of the diagnosis and treatment data, the transmission relationship of the diagnosis and treatment data for different medical nodes is determined. Based on this, the storage relationship of the diagnosis and treatment data in the medical nodes is adjusted to obtain the sharing necessity, so as to improve the utilization efficiency of the sharing of medical care data.

[0096] Step S004: Share the medical care data of several patients based on the sharing necessity.

[0097] It should be noted that the sharing necessity is to judge whether each medical node needs to share each type of diagnosis and treatment data. If there is a relatively high demand for a certain type of diagnosis and treatment data among medical nodes (departments), and there are significant differences in the retention time (analysis timeliness) of this type of diagnosis and treatment data in different medical nodes, then it is necessary to share and store it in multiple medical nodes in a timely manner. Based on this, the storage of each type of diagnosis and treatment data in medical nodes is judged according to the sharing necessity, and the storage under multiple medical nodes realizes the sharing of the corresponding medical care data.

[0098] Specifically, a sharing threshold is preset. In this embodiment, the sharing threshold is described as 0.7. For any type of diagnosis and treatment data of any patient, if the sharing necessity of this type of diagnosis and treatment data for any medical node is greater than or equal to the sharing threshold, then this type of diagnosis and treatment data needs to be shared to this medical node. During the monitoring of the medical care data of this patient, the corresponding type of diagnosis and treatment data is transmitted to this medical node and stored, thus realizing the sharing of the medical care data of the patient. It should be noted that if the sharing necessity is less than the threshold but the medical care data of this patient is obtained by collecting in this medical node, it is not necessary to delete any type of diagnosis and treatment data from the medical care data of this patient in this medical node, that is, during the sharing process of medical care data, it is necessary to ensure that the corresponding data source medical node still stores the corresponding medical care data.

[0099] So far, by analyzing various types of diagnosis and treatment data in the medical care data of patients, combining the relevant data of each medical node, quantifying the characteristic manifestations of the diagnosis and treatment data itself for the patient's condition, and determining its timeliness performance for different medical nodes, the storage and sharing of the diagnosis and treatment data for each medical node are judged. Finally, the sharing necessity of the diagnosis and treatment data for the medical node is obtained, and based on this, the sharing and storage of the diagnosis and treatment data in the medical care data of patients are carried out, effectively improving the communication between medical nodes and the utilization of their storage space, and at the same time giving full play to the sharing and transmission relationship between medical care data, realizing the efficient and effective sharing of medical care data.

[0100] Another embodiment of the present invention provides a medical care data sharing system, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the above method steps S001 to S004 are implemented.

[0101] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A medical care data sharing method, characterized in that, The method includes the following steps: Obtain the medical care data of a number of patients and the relevant data of each medical node, where the medical care data of the patients includes several types of diagnosis and treatment data; Classify the diagnosis and treatment data of the same type for different patients, analyze the differences between the diagnosis and treatment data of each type for the same patient among the classification results, and quantify the necessity of analyzing the diagnosis and treatment data of each type for each patient. The specific methods included are: Among them, represents the necessity for analyzing the th type of diagnosis and treatment data of the th patient, represents the number of types of diagnosis and treatment data of the th patient, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the centroid of the cluster to which it belongs, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the centroid of the cluster to which it belongs, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the mean value of all representative data in the standard cluster of the th type of diagnosis and treatment data of the th patient, represents the absolute value of the difference between the representative data of the th type of diagnosis and treatment data of the th patient and the mean value of all representative data in the standard cluster of the th type of diagnosis and treatment data of the th patient, represents the linear normalization function; Based on the temporal changes of the diagnosis and treatment data of each type for the patient, and in combination with the relevant data of each medical node, determine the analysis timeliness of the diagnosis and treatment data of each type for the patient for each medical node. The specific acquisition method is: Regarding the medical care data of several patients collected by the th medical node, the mean value of the length of time from seeing a doctor to discharge for each patient is used as the retention time of the medical care data under the th medical node; Starting from the visit of the th patient, until the retention time of the medical care data under the th medical node ends, the th patient's th type of diagnosis and treatment data during this period is recorded as the th patient's th type of diagnosis and treatment data under the th medical node; Count the number of diagnostic data in the diagnostic sequence that exceed the standard range of the diagnostic data of the -th patient of the -th type, and use the last diagnostic data in the diagnostic sequence as the right boundary of the window, and intercept a sequence with a reference window as the diagnostic reference sequence of the -th patient's -th type of diagnostic data under the -th medical node; Based on the retention time of medical care data under the th medical node, combined with the proportion of medical treatment data that exceeds the standard range of the th type of medical treatment data of the th patient, and the fluctuation of medical treatment data in the medical treatment reference sequence, obtain the analysis timeliness of the th type of medical treatment data of the th patient for the th medical node; the analysis timeliness has a positive correlation with the retention time, the proportion, and the fluctuation; Analyze the corresponding relationship between the relevant data of each medical node and the diagnosis and treatment data of each type for the patient, combine the necessity of analyzing the diagnosis and treatment data of each type for each patient and the distribution of medical nodes, and determine the sharing necessity of the diagnosis and treatment data of each type for the patient for each medical node by restricting through the analysis timeliness of the diagnosis and treatment data of each type for each medical node. The specific methods included are: Based on the corresponding relationship between the relevant data of each medical node and the diagnosis and treatment data of each type for the patient and its analysis necessity, obtain the storage necessity of the diagnosis and treatment data of each type for the patient for each medical node. The specific methods included are: The acquisition method of the storage necessity includes: Based on the inclusion relationship of the diagnosis and treatment data types of the patient on the medical node and the proportion of the diagnosis and treatment data types included in both the patient and the medical node in the total amount of diagnosis and treatment data types of the patient and the medical node, and in combination with the analysis necessity of the diagnosis and treatment data of each type for the patient, obtain the storage necessity of the diagnosis and treatment data of each type for the patient for each medical node; The storage necessity is positively correlated with both the proportion and the analysis necessity; Based on the differences in the analysis timeliness of the diagnosis and treatment data of each type for different medical nodes, and in combination with the distribution of medical nodes, quantify the sharing factor of the diagnosis and treatment data of each type between different medical nodes. The specific methods included are: The acquisition method of the sharing factor includes: Based on the inclusion relationship of the same type of diagnosis and treatment data of the patient for two medical nodes and the difference in the analysis necessity of this type of diagnosis and treatment data for the two medical nodes, and in combination with the spatial distance between the two medical nodes and the proportion of the quantity of the diagnosis and treatment data of the types they commonly contain, obtain the sharing factor of the same type of diagnosis and treatment data of the patient between the two medical nodes; The sharing factor is negatively correlated with the difference in the analysis necessity and the spatial distance, and the sharing factor is positively correlated with the proportion of the quantity; Based on the storage necessity, analyze the sharing factor to determine the sharing necessity of the diagnosis and treatment data of each type for the patient for each medical node. The specific acquisition method is: Taking the th patient's type of diagnosis and treatment data as an example, for several medical nodes with non-zero storage necessity among all medical nodes for this type of diagnosis and treatment data, arrange them in ascending order of storage necessity to obtain the node storage sequence of this type of diagnosis and treatment data. The sharing necessity of this type of diagnosis and treatment data for the first medical node in its node storage sequence is calculated as follows: Among them, represents the storage necessity of the th type of diagnosis and treatment data of the th patient for the first medical node in its node storage sequence, represents the number of medical nodes in the node storage sequence of the th type of diagnosis and treatment data of the th patient, represents the sharing factor of the th type of diagnosis and treatment data of the th patient for the first medical node and the th medical node in its node storage sequence, represents the storage necessity of the th type of diagnosis and treatment data of the th patient for the th medical node in its node storage sequence; represents the weight normalization function; After obtaining the corresponding sharing necessity for the first medical node in the node storage sequence, calculate the sharing necessity corresponding to the second medical node, where the medical nodes for which the storage necessity has been updated to the sharing necessity during the calculation process participate in the calculation with the sharing necessity; And so on, to obtain the sharing necessity of the diagnostic and treatment data of the -th patient for each medical node in its node storage sequence with respect to the -th type; Share the medical care data of a number of patients based on the sharing necessity.

2. The method for sharing medical care data according to claim 1, wherein The specific methods included in the classification of the diagnosis and treatment data of the same type for different patients are: For the medical treatment data corresponding to the type of physiological data obtained by one-time detection, numerical processing is performed on the medical treatment data of this type for all patients, and it is used as the representative data of the medical treatment data of this type for each patient; For the medical treatment data corresponding to the type of physiological data obtained by continuous monitoring, the median of the standard range of the medical treatment data of this type is obtained and used as the standard value of the medical treatment data of this type. Based on the deviation value between the medical treatment data of this type of any patient and the standard value, through numerical processing and selecting the numerical processing result corresponding to the deviation value with the largest absolute value, it is used as the representative data of the medical treatment data of this type for this patient; Density clustering is performed on the representative data of the medical treatment data of any type for all patients, and the absolute value of the difference between the representative data of the patients is used as the distance metric to obtain several clusters of the medical treatment data of this type; The cluster with the largest number of representative data among all clusters is used as the standard cluster of the medical treatment data of this type.

3. A medical care data sharing system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of a medical care data sharing method as described in any one of claims 1-2.

Images