A diabetes screening data processing management system and method thereof

Abstract

The present invention discloses a diabetes screening data processing management system and a method thereof, which belong to the technical field of service management; constructs a first entity set and a first relationship set between a patient and clinical indicator data; constructs a second entity set and a second relationship set between a patient and medication information data to form a personalized knowledge graph for the patient; constructs a multidimensional data set for the patient, and calculates a diabetes risk assessment value; calculates the intervention measures that are most suitable for the patient; evaluates the effectiveness of the intervention measures in reducing the risk of diabetes; and designs a dynamic adjustment formula for the intervention measures; the present invention constructs a diabetes risk assessment model to accurately quantify the patient's diabetes risk, combines a personalized intervention matching model to formulate the most suitable intervention measures for the patient, and dynamically adjusts the intervention plan according to the patient's actual response, ultimately providing the patient with an accurate, intelligent, and personalized diabetes risk management and health intervention plan; and can effectively improve screening accuracy and optimize intervention effects.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing management, and specifically to a diabetes screening data processing management system and method thereof. Background Art

[0002] As a chronic disease worldwide, the incidence of diabetes has increased significantly in recent years. With the aging of the population, changes in lifestyle and unreasonable dietary structure, the number of diabetic patients continues to grow. Existing methods for the diagnosis and management of diabetes usually rely on clinicians to judge the test results of patients' blood sugar, glycosylated hemoglobin (HbA1c) and other physiological indicators, and formulate treatment plans based on personal experience. However, since each diabetic patient has different physical constitutions, medication reactions and lifestyle habits, it is difficult for existing treatment plans to achieve refined management for individuals. In recent years, with the rapid development of big data technology, artificial intelligence and knowledge graph technology, these technologies have been gradually applied to the medical field, especially in the screening and management of chronic diseases such as diabetes, in order to improve the accuracy of diabetes risk assessment and optimize the formulation of personalized treatment plans.

[0003] However, the existing diabetes screening and management systems still have many shortcomings. First, many systems rely only on limited patient data (such as blood sugar levels) to assess diabetes risk, while ignoring other important clinical information of the patient, such as medication status, lifestyle, etc. In addition, traditional screening methods are usually static and lack the tracking and management of dynamic changes in the patient's condition. This leads to a lag in the assessment of diabetes risk and the inability to capture the deterioration or improvement of the patient's condition in a timely manner. In addition, in the design of diabetes intervention programs, current methods usually adopt universal recommendations and fail to fully consider individual differences and real-time feedback, which may lead to inefficient or even counterproductive intervention effects. Summary of the invention

[0004] The object of the present invention is to provide a diabetes screening data processing management system and method thereof to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A diabetes screening data processing and management method, the method comprising the following steps: obtaining clinical indicator data and medication information data of a patient, and constructing a clinical indicator data set and a medication information data set respectively; based on the clinical indicator data set and the patient database, extracting entities and relationships between the patient and the clinical indicator data, constructing a first entity set and a first relationship set between the patient and the clinical indicator data, and constructing an initial knowledge graph; constructing a second entity set and a second relationship set between the patient and the medication information data, and adding them to the initial knowledge graph to form a final patient personalized knowledge graph; constructing a multidimensional data set for the patient, and calculating a diabetes risk assessment value; based on the diabetes risk assessment value and the multidimensional data set, calculating the most suitable intervention measures for the patient; collecting data after the patient executes the intervention measures, constructing an intervention measures evaluation model, and evaluating the effectiveness of the intervention measures in reducing the diabetes risk; and designing a dynamic adjustment formula for the intervention measures based on the effectiveness and the intervention measures.

[0007] As a preferred solution of the diabetes screening data processing and management method of the present invention, a patient-based database is constructed, denoted as ,in, represents the ath patient, A represents the total number of patients; get the patient Clinical indicator data and medication information data are used to construct clinical indicator data sets. and medication information dataset ,in, represents the clinical index data of the ath patient, Represents the medication information data of the ath patient.

[0008] The clinical indicator data include patients The blood sugar level, glycosylated hemoglobin and blood pressure of the patient; the medication information data includes The type, dosage and duration of medication.

[0009] As a preferred solution of the diabetes screening data processing and management method of the present invention, based on the clinical indicator data set and the patient database, entities and relationships between patients and clinical indicator data are extracted, and the entities include patients , blood sugar level data , Glycated hemoglobin data and blood pressure data ,in, represents the i-th blood sugar level data, represents the bth glycated hemoglobin data, represents the cth type of blood pressure data; the relationship includes measurement, inspection and detection; the measurement represents the patient With blood sugar level data The measurement relationship between the patient and HbA1c data The relationship between the examinations, the examinations represent the patient With blood pressure data The detection relationship between them.

[0010] Construct the first entity set and the first relationship set between patients and clinical indicator data, respectively denoted as the first entity set ST = {patient , blood sugar level data , Glycated hemoglobin data , blood pressure data } and the first relationship set GX={measurement, inspection, detection}, based on the first entity set ST and the first relationship set GX, construct an initial knowledge graph, denoted as G={ST, GX}.

[0011] As a preferred solution of the diabetes screening data processing and management method of the present invention, based on the medication information data set and the patient database, entities and relationships between patients and medication information data are extracted, and the entities include patients , Drug Type Data , dosage data and medication time data ,in, represents the data of the fth drug type, represents the hth dose data, represents the nth medication time data; the relationship includes prescribing, taking and frequency; the prescribing represents the patient Drug type data The relationship between the prescription and the administration of the drug indicates that the patient With dosage data The relationship between the taking of Medication time data The frequency relationship between them.

[0012] Construct the second entity set and the second relationship set between the patient and medication information data, respectively denoted as the second entity set SST = {patient , Drug Type Data , dosage data , medication time data } and the second relationship set GGX={prescription, medication, frequency}, add the second entity set and the second relationship set to the initial knowledge graph G={ST, GX} to form the final patient personalized knowledge graph, denoted as EG={ST∪SST, GX∪GGX}.

[0013] Constructing patients The multidimensional dataset is denoted by , the multidimensional dataset As input variables, a diabetes risk assessment model was constructed as follows:

[0014] ,

[0015] in, represents the diabetes risk assessment value, , and Represents the regression coefficient of the preset clinical indicator data, , and Represents the regression coefficient of the preset medication information, Indicates an additional item.

[0016] Formula used to calculate diabetes risk assessment value , assessing a patient's risk of diabetes by integrating their clinical data (such as blood glucose, glycosylated hemoglobin, and blood pressure) and medication information (such as drug type, dosage, and medication duration); Represents the patient's i-th blood sugar level data. Blood sugar is an important indicator for the diagnosis and management of diabetes. Continuous high blood sugar may indicate that the patient has an increased risk of diabetes. Represents the patient's glycosylated hemoglobin data. Glycated hemoglobin (HbA1c) can reflect the patient's average blood sugar level in the past 2 to 3 months and is an important indicator of long-term control of diabetes; Represents the patient's blood pressure data. Hypertension is closely related to the risk of diabetes, so blood pressure data plays an important role in diabetes risk assessment. Data on the types of medications that patients take. Different types of medications may have different effects on diabetes control, for example, some medications may promote or suppress blood sugar levels. Represents the patient's drug dosage data. Changes in drug dosage directly affect its efficacy and side effects. Too high or too low a dosage will affect the effect of diabetes management; Represents the patient's medication time data. The frequency and timing of medication are crucial to controlling blood sugar levels; the main function of this formula is to calculate each patient's diabetes risk score By combining the patient's clinical data and medication information, the formula provides a comprehensive diabetes risk assessment model that can help doctors or health management systems identify high-risk patients and take preventive or intervention measures in advance. This risk assessment model can automatically process large amounts of patient data and provide support for personalized health management.

[0017] Design of intervention recommendation matching model based on diabetes risk assessment value And cubes , matching the most suitable patient The matching formula for intervention measures is as follows:

[0018] ,

[0019] in, The most suitable for patients intervention measures, , and Represents the preset weight coefficient, , and Represents the parameter coefficient of each clinical index, , and Represents the parameter coefficients of each drug data.

[0020] In the present invention, the formula generates a matching value by weighted summing up the multi-dimensional health data of the patient. , thereby selecting the most suitable personalized intervention plan for the patient; by combining the patient's clinical indicators and medication information, the intervention recommendations are automatically matched to the patient, so that the intervention measures can be customized; for example, if a patient has a high glycated hemoglobin and poor medication compliance, the system can recommend strengthening drug management or adjusting the dosage. This automated matching process reduces the error of human decision-making, improves the accuracy and efficiency of personalized treatment, and optimizes the overall effect of diabetes management.

[0021] As a preferred solution of the diabetes screening data processing and management method of the present invention, the data after the patient performs the intervention measures are collected to form a new data set. ,in, Indicates patient The ith physiological index value after intervention, N is the total number of physiological indicators.

[0022] Constructing intervention evaluation models to leverage newly collected data The effectiveness of the intervention in reducing the risk of diabetes was evaluated by comparing it with the previous risk assessment model parameters. The model calculations were as follows:

[0023] ,

[0024] in, Indicates effectiveness, It represents the diabetes risk assessment value after the patient implements the intervention plan. is the patient's baseline or target value, The patient's compliance coefficient reflects the degree to which the patient implements the intervention program (such as compliance with medication and diet plans), which can be obtained through data collection and doctor observation and evaluation. , and is the weight coefficient, which indicates the importance of different dimensions in the evaluation of intervention effects.

[0025] In the present invention, the formula is used to evaluate the effectiveness of intervention measures; after the intervention measures are implemented, the effect of the intervention is quantified by comparing the patient's diabetes risk score before and after the intervention, as well as the difference between the physiological indicators after the intervention and the baseline value; the intervention effect can be quantified for each patient, providing data support for subsequent treatment decisions; if the intervention effect is significant, it indicates that the current treatment plan is effective, otherwise the intervention strategy may need to be adjusted; this data-based evaluation system can dynamically track the patient's health changes and ensure the accuracy and effectiveness of the intervention measures.

[0026] Based on effectiveness and interventions , design a dynamic adjustment formula for intervention measures, the calculation formula is as follows:

[0027] ,

[0028] in, Dynamically adjusted intervention measures.

[0029] Interventions based on dynamic adjustments , build an evaluation cycle in months, automatically adjust the intervention plan based on the patient health data in each evaluation cycle, and update the patient's personalized knowledge graph EG. If some of the pre-effects are not ideal, re-optimize the intervention strategy based on the matching function.

[0030] In the present invention, by multiplication, the formula outputs an adjusted intervention value If the intervention effect is not good, the adjusted value will prompt the system to optimize and adjust the intervention plan; if the intervention effect is good, the system will recommend maintaining or slightly fine-tuning the current intervention measures; the design of this formula ensures that the system can dynamically adjust the intervention measures according to the individual differences of patients and real-time data feedback; through this dynamic optimization, the system can continuously optimize the patient's intervention plan in each evaluation cycle, thereby improving the intervention effect and preventing delays in treatment or adverse reactions due to long-term use of inappropriate plans; this is particularly important in the management of chronic diseases such as diabetes, because the patient's health status will change over time, and intervention measures also need to be adjusted accordingly.

[0031] A diabetes screening data processing and management system comprises: a patient database and data set construction module, a clinical indicator knowledge graph construction module, a medication information and risk assessment module and an intervention measure assessment and adjustment module; the patient database and data set construction module is used to obtain the patient's clinical indicator data and medication information data, and respectively construct a clinical indicator data set and a medication information data set; the clinical indicator knowledge graph construction module is used to extract entities and relationships between the patient and the clinical indicator data based on the clinical indicator data set and the patient database, construct a first entity set and a first relationship set between the patient and the clinical indicator data, and construct an initial knowledge graph; the medication information and risk assessment module is used to construct a second entity set and a second relationship set between the patient and the medication information data, and add them to the initial knowledge graph to form a final patient personalized knowledge graph; a multidimensional data set of the patient is constructed to calculate a diabetes risk assessment value; based on the diabetes risk assessment value and the multidimensional data set, the most suitable intervention measure for the patient is calculated; the intervention measure assessment and adjustment module is used to collect data after the patient executes the intervention measure, construct an intervention measure assessment model, and evaluate the effectiveness of the intervention measure in reducing the diabetes risk; based on the effectiveness and the intervention measure, a dynamic adjustment formula for the intervention measure is designed.

[0032] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: by constructing a patient-based database, the clinical indicator data (such as blood sugar, glycosylated hemoglobin, blood pressure) and medication information of the patient are comprehensively obtained, the accuracy and systematic management of the data are ensured, and the foundation for screening and analysis is laid; subsequently, the system constructs a knowledge graph by extracting entities and relationships between the patient and the clinical indicators and medication information, realizes the structuring and association analysis of the patient's health data, and provides intelligent support for subsequent risk assessment; through multidimensional data sets and regression coefficients, the system constructs a diabetes risk assessment model to accurately quantify the patient's diabetes risk, and combines a personalized intervention matching model to formulate the most suitable intervention measures for the patient; after collecting the data after the intervention, the system evaluates the intervention effect, and dynamically adjusts the intervention plan according to the patient's actual response to form a feedback closed loop; by continuously updating the patient's personalized knowledge graph, the system realizes the continuous optimization of the intervention strategy, and ultimately provides the patient with an accurate, intelligent, and personalized diabetes risk management and health intervention plan; this method can effectively improve the accuracy of screening, optimize the intervention effect, and achieve the purpose of long-term health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0034] Figure 1 It is a schematic diagram of the steps of a diabetes screening data processing and management method of the present invention.

[0035] Figure 2 It is a structural schematic diagram of a diabetes screening data processing and management system of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Example 1, please refer to Figure 1 In the first embodiment, a diabetes screening data processing and management method is provided, the method comprising the following steps:

[0038] Step S1: Acquire the patient's clinical indicator data and medication information data, and construct a clinical indicator data set and a medication information data set respectively.

[0039] Specifically, a patient-based database is constructed, denoted as ,in, represents the ath patient, A represents the total number of patients; get the patient Clinical indicator data and medication information data are used to construct clinical indicator data sets. and medication information dataset ,in, represents the clinical index data of the ath patient, Represents the medication information data of the ath patient.

[0040] Furthermore, the clinical indicator data include patients The blood sugar level, glycosylated hemoglobin and blood pressure of the patient; the medication information data includes The type, dosage and duration of medication.

[0041] Step S2: Based on the clinical indicator data set and the patient database, extract the entities and relationships between the patient and the clinical indicator data, construct a first entity set and a first relationship set between the patient and the clinical indicator data, and construct an initial knowledge graph.

[0042] Specifically, based on the clinical indicator data set and the patient database, entities and relationships between patients and clinical indicator data are extracted, wherein the entities include patients , blood sugar level data , Glycated hemoglobin data and blood pressure data ,in, represents the i-th blood sugar level data, represents the bth glycated hemoglobin data, represents the cth type of blood pressure data; the relationship includes measurement, inspection and detection; the measurement represents the patient With blood sugar level data The measurement relationship between the patient and HbA1c data The relationship between the examinations, the examinations represent the patient With blood pressure data The detection relationship between them.

[0043] Furthermore, the first entity set and the first relationship set between the patient and the clinical indicator data are constructed, which are respectively denoted as the first entity set ST={patient , blood sugar level data , Glycated hemoglobin data , blood pressure data } and the first relationship set GX={measurement, inspection, detection}, based on the first entity set ST and the first relationship set GX, construct an initial knowledge graph, denoted as G={ST, GX}.

[0044] Step S3: Construct a second entity set and a second relationship set between the patient and medication information data, and add them to the initial knowledge graph to form the final patient personalized knowledge graph; construct a multidimensional data set for the patient and calculate the diabetes risk assessment value; based on the diabetes risk assessment value and the multidimensional data set, calculate the intervention measures that are most suitable for the patient.

[0045] Specifically, based on the medication information data set and the patient database, entities and relationships between patients and medication information data are extracted, wherein the entities include patients , Drug Type Data , dosage data and medication time data ,in, represents the data of the fth drug type, represents the hth dose data, represents the nth medication time data; the relationship includes prescribing, taking and frequency; the prescribing represents the patient Drug type data The relationship between the prescription and the administration of the drug indicates that the patient With dosage data The relationship between the taking of Medication time data The frequency relationship between them.

[0046] Furthermore, a second entity set and a second relationship set between the patient and medication information data are constructed, which are respectively recorded as the second entity set SST={patient , Drug Type Data , dosage data , medication time data } and the second relationship set GGX={prescription, medication, frequency}, add the second entity set and the second relationship set to the initial knowledge graph G={ST, GX} to form the final patient personalized knowledge graph, denoted as EG={ST∪SST, GX∪GGX}.

[0047] Constructing patients The multidimensional dataset is denoted by , the multidimensional dataset As input variables, a diabetes risk assessment model was constructed as follows:

[0048] ,

[0049] in, represents the diabetes risk assessment value, , and Represents the regression coefficient of the preset clinical indicator data, , and Represents the regression coefficient of the preset medication information, Indicates an additional item.

[0050] Furthermore, a matching model for intervention recommendations was designed based on diabetes risk assessment values. And cubes , matching the most suitable patient The matching formula for intervention measures is as follows:

[0051] ,

[0052] in, The most suitable for patients intervention measures, , and Represents the preset weight coefficient, , and Represents the parameter coefficient of each clinical index, , and Represents the parameter coefficients of each drug data.

[0053] Step S4: Collect data after the patient implements the intervention measures, construct an intervention measure evaluation model, and evaluate the effectiveness of the intervention measures in reducing the risk of diabetes; based on the effectiveness and intervention measures, design a dynamic adjustment formula for the intervention measures.

[0054] Specifically, data after the patient performs the intervention measures are collected to form a new data set ,in, Indicates patient The ith physiological index value after intervention, N is the total number of physiological indicators.

[0055] Further, we will build an intervention evaluation model to utilize the newly collected data The effectiveness of the intervention in reducing the risk of diabetes was evaluated by comparing it with the previous risk assessment model parameters. The model calculations were as follows:

[0056] ,

[0057] in, Indicates effectiveness, It represents the diabetes risk assessment value after the patient implements the intervention plan. is the patient's baseline or target value, is the patient's compliance coefficient, , and is the weight coefficient, which indicates the importance of different dimensions in the evaluation of intervention effects.

[0058] Furthermore, based on the effectiveness and interventions , design a dynamic adjustment formula for intervention measures, the calculation formula is as follows:

[0059] ,

[0060] in, Dynamically adjusted intervention measures.

[0061] Interventions based on dynamic adjustments , build an evaluation cycle in months, automatically adjust the intervention plan based on the patient health data in each evaluation cycle, and update the patient's personalized knowledge graph EG. If some of the pre-effects are not ideal, re-optimize the intervention strategy based on the matching function.

[0062] Example 2, please refer to Figure 2In this embodiment, a diabetes screening data processing and management system is provided, which includes: a patient database and data set construction module, a clinical indicator knowledge graph construction module, a medication information and risk assessment module, and an intervention measure evaluation and adjustment module; the patient database and data set construction module: obtains the patient's clinical indicator data and medication information data, and respectively constructs a clinical indicator data set and a medication information data set; the clinical indicator knowledge graph construction module: based on the clinical indicator data set and the patient database, extracts the entities and relationships between the patient and the clinical indicator data, constructs a first entity set and a first relationship set between the patient and the clinical indicator data, and constructs an initial knowledge graph; the medication information and risk assessment module: constructs a second entity set and a second relationship set between the patient and the medication information data, and adds them to the initial knowledge graph to form a final patient personalized knowledge graph; constructs a multidimensional data set for the patient, and calculates a diabetes risk assessment value; based on the diabetes risk assessment value and the multidimensional data set, calculates the most suitable intervention measure for the patient; the intervention measure evaluation and adjustment module: collects data after the patient performs the intervention measure, constructs an intervention measure evaluation model, and evaluates the effectiveness of the intervention measure in reducing the diabetes risk; based on the effectiveness and the intervention measure, designs a dynamic adjustment formula for the intervention measure.

[0063] Embodiment 3: In this embodiment 3, a diabetes screening data processing and management method is provided. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through simulation experiments.

[0064] The experimental team conducted a 6-month clinical trial. This trial selected 50 patients with type 2 diabetes (experimental group) and 50 patients without diabetes but with high-risk factors (control group) for data collection and analysis; each patient's data included their blood sugar level, glycosylated hemoglobin, body mass index (BMI), blood pressure, drug type, dosage, medication time and other clinical indicators and medication data.

[0065] The specific contents of the test are as follows:

[0066] According to the patients' medical history and risk factors, 100 patients were randomly divided into two groups. The patients in the experimental group had been diagnosed with diabetes and received standardized drug treatment; the patients in the control group had not been diagnosed with diabetes but had risk factors for high blood sugar, such as obesity, hypertension, and family history of diabetes.

[0067] A detailed clinical data file is established for each patient, including: blood glucose level (fasting and 2 hours after meal, unit: mmol / L), glycosylated hemoglobin (HbA1c, unit: %), blood pressure (systolic and diastolic, unit: mmHg), body mass index (BMI, unit: kg / m²), type of medication (drug name, dosage), medication time and medication frequency (number of medications per day, unit: times / day).

[0068] A diabetes risk assessment model for patients is established through a multidimensional data model based on the patient's clinical indicators and medication information. For each patient, the initial values ​​of each clinical indicator are recorded, and the data is updated after 6 months to observe the changing trends of the patient's clinical indicators.

[0069] For patients in the experimental group, their medication regimen is adjusted based on the risk assessment results of the model. For example, for high-risk patients, the dosage of hypoglycemic drugs is adjusted and the frequency of medication is increased. Patients in the control group do not receive any intervention and only undergo routine health monitoring.

[0070] Collect data after the trial and compare the initial data with the final data. Focus on analyzing changes in key indicators such as blood sugar levels, HbA1c, and BMI to evaluate the effectiveness of intervention measures.

[0071] Table 1 Comparison of clinical indicators and drug dosages of patients

[0072] parameter Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Fasting blood glucose (mmol / L) 6.5 7.1 5.9 6.2 7.8 6.9 Postprandial blood glucose (mmol / L) 8.3 9.0 7.2 7.8 9.5 8.1 Glycated hemoglobin (HbA1c, %) 6.8 7.5 6.2 6.4 7.8 7.1 Body mass index (BMI, kg / m²) 28.5 29.2 26.8 27.5 30.1 28.9 Systolic blood pressure (mmHg) 130 140 125 135 145 138 Diastolic blood pressure (mmHg) 85 90 80 88 95 92 Drug dosage (mg / day) 50 75 50 60 80 70

[0073] As can be seen from Table 1, the key clinical indicators of the six patients in the trial group changed significantly after receiving treatment adjusted based on clinical indicators and medication information:

[0074] Judging from the initial fasting blood glucose and 2-hour postprandial blood glucose levels, the patients' fasting blood glucose values ​​were between 6.5 and 7.8 mmol / L, and their postprandial blood glucose values ​​were between 7.2 and 9.5 mmol / L. After 6 months of intervention, the patients' fasting blood glucose levels generally decreased by 0.3 to 0.7 mmol / L, and their postprandial blood glucose levels also decreased by an average of about 0.5 mmol / L, indicating that the patients' blood glucose control has been significantly improved.

[0075] HbA1c is an important indicator for long-term blood sugar control in diabetic patients. At the beginning of the trial, the patients' HbA1c values ​​were between 6.2% and 7.8%. After 6 months of medication adjustment, the HbA1c values ​​of all patients decreased by 0.3% to 0.7%, indicating that the long-term blood sugar control effect is significant.

[0076] Weight management also plays a vital role in diabetes control. At the beginning of the trial, the patients' BMI ranged from 26.8kg / m² to 30.1 kg / m². After 6 months, the patients' BMI decreased by an average of 0.5 to 1 kg / m², indicating that the patients' weight also decreased under the intervention measures.

[0077] Hypertension is closely related to diabetic complications. The patients' initial systolic blood pressure was between 125 mmHg and 145 mmHg, and their diastolic blood pressure was between 80 mmHg and 95 mmHg. After intervention, the patients' blood pressure levels dropped by about 5 mmHg, indicating that drug adjustment also played a certain role in controlling hypertension.

[0078] Based on the risk assessment results of each patient, the drug dosage and frequency of medication were adjusted. By comparing the patients' medication data, it can be seen that after increasing the drug dosage or adjusting the medication frequency, the patient's blood sugar, blood pressure and other indicators improved, which shows that the medication adjustment predicted by the model has a certain degree of accuracy and effectiveness.

[0079] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0080] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A diabetes screening data processing and management method, characterized in that: The method comprises the following steps: S1: Obtain the patient's clinical indicator data and medication information data, and construct a clinical indicator dataset and a medication information dataset respectively; S2: Based on the clinical indicator data set and the patient database, extract entities and relationships between patients and clinical indicator data, construct a first entity set and a first relationship set between patients and clinical indicator data, and construct an initial knowledge graph; S3: Construct the second entity set and the second relationship set between the patient and medication information data, and add them to the initial knowledge graph to form the final patient personalized knowledge graph; construct a multidimensional data set for the patient and calculate the diabetes risk assessment value; based on the diabetes risk assessment value and the multidimensional data set, calculate the most suitable intervention measures for the patient; S4: Collect data from patients after implementing intervention measures, build an intervention evaluation model, and evaluate the effectiveness of the intervention measures in reducing the risk of diabetes; based on the effectiveness and intervention measures, design a dynamic adjustment formula for the intervention measures; The respectively constructing of the clinical indicator data set and the medication information data set comprises: Construct a patient-based database, denoted as ,in, represents the ath patient, A represents the total number of patients; get the patient Clinical indicator data and medication information data are used to construct clinical indicator data sets. and medication information dataset ,in, represents the clinical index data of the ath patient, Represents the medication information data of the ath patient; The clinical indicator data include patients The blood sugar level, glycosylated hemoglobin and blood pressure of the patient; the medication information data includes the type, dosage and duration of medication; The process of step S2 includes: Based on the clinical indicator data set and the patient database, entities and relationships between patients and clinical indicator data are extracted, wherein the entities include patients , blood sugar level data , Glycated hemoglobin data and blood pressure data ,in, represents the i-th blood sugar level data, represents the bth glycated hemoglobin data, represents the cth type of blood pressure data; the relationship includes measurement, inspection and detection; measurement represents the patient With blood sugar level data The relationship between the measurements, examination indicates that the patient and HbA1c data The relationship between the examinations, the detection indicates that the patient With blood pressure data The detection relationship between Construct the first entity set and the first relationship set between patients and clinical indicator data, respectively denoted as the first entity set ST = {patient , blood sugar level data , Glycated hemoglobin data , blood pressure data } and the first relationship set GX={measurement, inspection, detection}, based on the first entity set ST and the first relationship set GX, construct an initial knowledge graph, denoted as G={ST, GX}; The process of step S3 includes: Based on the medication information data set and the patient database, entities and relationships between patients and medication information data are extracted, wherein the entities include patients , Drug Type Data , dosage data and medication time data ,in, represents the data of the fth drug type, represents the hth dose data, Indicates the nth medication time data; Said relationships include prescribing, administration, and frequency; The prescription indicates that the patient Drug type data The issuing relationship between them; The administration indicates that the patient With dosage data The relationship between the use of The frequency indicates the patient Medication time data The frequency relationship between Construct the second entity set and the second relationship set between the patient and medication information data, respectively denoted as the second entity set SST = {patient , Drug Type Data , dosage data , medication time data } and the second relation set GGX = {prescription, administration, frequency}; Add the second entity set and the second relationship set to the initial knowledge graph G={ST, GX} to form the final patient personalized knowledge graph, denoted as EG={ST∪SST, GX∪GGX}; The process of step S3 also includes: Constructing patients The multidimensional dataset is denoted as: ； The cube As input variables, a diabetes risk assessment model was constructed as follows: ； in, represents the diabetes risk assessment value, , and Represents the regression coefficient of the preset clinical indicator data, , and Represents the regression coefficient of the preset medication information, Indicates an additional item; Design of intervention recommendation matching model based on diabetes risk assessment value And cubes , matching the most suitable patient The matching formula for intervention measures is as follows: ； in, The most suitable for patients intervention measures, , and Represents the preset weight coefficient, , and Represents the parameter coefficient of each clinical index, , and Represents the parameter coefficients of each drug data; The process of step S4 includes: Collect data from patients after they have performed the intervention to form a new data set ,in, Indicates patient The value of the ith physiological index after the intervention, N is the total number of physiological indexes; Constructing intervention evaluation models to leverage newly collected data The effectiveness of the intervention in reducing the risk of diabetes was evaluated by comparing it with the previous risk assessment model parameters. The model calculations were as follows: ； in, Indicates effectiveness, It represents the diabetes risk assessment value after the patient implements the intervention plan. is the patient's baseline or target value, is the patient's compliance coefficient, , and is the weight coefficient, which indicates the importance of different dimensions in the evaluation of intervention effects; The process of step S4 also includes: Based on effectiveness and interventions , design a dynamic adjustment formula for intervention measures, the calculation formula is as follows: ； in, for dynamically adjusted interventions; Interventions based on dynamic adjustments , build an evaluation cycle in months, automatically adjust the intervention plan based on the patient health data in each evaluation cycle, and update the patient's personalized knowledge graph EG. If some of the pre-effects are not ideal, re-optimize the intervention strategy based on the matching function.

2. A diabetes screening data processing and management system, based on the diabetes screening data processing and management method described in claim 1, characterized in that: Also includes: Patient database and data set construction module: obtain the patient's clinical indicator data and medication information data, and construct the clinical indicator data set and medication information data set respectively; Clinical indicator knowledge graph construction module: based on the clinical indicator data set and the patient database, extract the entities and relationships between the patient and the clinical indicator data, construct the first entity set and the first relationship set between the patient and the clinical indicator data, and construct the initial knowledge graph; Medication information and risk assessment module: construct the second entity set and the second relationship set between the patient and medication information data, and add them to the initial knowledge graph to form the final patient personalized knowledge graph; construct the patient's multidimensional data set and calculate the diabetes risk assessment value; Calculate the most appropriate intervention for the patient based on diabetes risk assessment and multidimensional datasets; Intervention evaluation and adjustment module: collect data after patients implement intervention measures, build an intervention evaluation model, and evaluate the effectiveness of the intervention measures in reducing the risk of diabetes; Based on effectiveness and intervention measures, a formula for dynamic adjustment of intervention measures is designed.

3. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the diabetes screening data processing and management method described in claim 1 are implemented.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a diabetes screening data processing and management method described in claim 1 are implemented.

Images