Diabetes medication scheme evaluation method and system

Abstract

The invention relates to the technical field of diabetes mellitus medical informatics, and provides a diabetes mellitus medication scheme evaluation method and system.The diabetes mellitus medication scheme evaluation method comprises the steps that a database is constructed, and comprehensive data of multiple diabetics is input into the database; obtaining current multi-dimensional detection data and current medication scheme data of the target patient; performing medication adaptability evaluation on the current medication scheme data based on the current multi-dimensional detection data in combination with a database to obtain a first evaluation result; setting an online evaluation window, and acquiring multi-party medication adaptability evaluation data of a doctor side from the current medication scheme data based on the current multi-dimensional detection data; analyzing the multi-party medication adaptability evaluation data to obtain a second evaluation result; and fusing the first evaluation result and the second evaluation result, and determining a medication scheme evaluation result of the target patient. According to the method and the system, the adaptability and the accuracy of a diabetes medication scheme are improved through a scientific evaluation means, so that the treatment effect and confidence of a patient are improved, and the medication decision process of a doctor is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of diabetes medical informatics, and in particular to a diabetes medication regimen evaluation method and system. Background Art

[0002] Diabetes is a complex chronic disease with diverse pathogenesis and significant individual differences. In clinical practice, the medication regimen for diabetic patients depends on the physician's clinical experience and judgment. However, the existing technology has a series of problems in the management of diabetes medication, resulting in differences in the effects of medication among patients.

[0003] Individual differences in impact: Individual differences among diabetic patients are reflected in many aspects, including age, weight, coexisting diseases, medication history, lifestyle habits, etc. This makes the efficacy and side effects of the same hypoglycemic drug in different patients vary greatly. In addition, the patient's drug response is also closely related to factors such as their metabolic capacity and drug interactions.

[0004] Drug-related differences: Similar hypoglycemic drugs produced by different manufacturers may differ in their ingredients, dosage forms, formulations, and bioavailability. This may result in widely different efficacy and side effects even for patients with the same disease state using the same drug regimen, thereby affecting the reliability of the treatment effect.

[0005] Limitations of physician decision-making: In the current clinical environment, physicians usually develop medication plans based on their own experience and understanding of the patient, and rely on the patient's regular review indicators to adjust the medication plan; this traditional model often lacks sufficient scientific data support, especially when dealing with complex cases. It may be difficult for doctors to accurately determine the best medication plan, which not only leads to uncertainty in treatment effects, but may also cause delays in patients' condition. Summary of the invention

[0006] To this end, the technical problem to be solved by the present invention is to overcome the uncertainty in the efficacy of diabetes medication regimens in the traditional model caused by individual differences among different patients, differences in drug properties and limitations in doctors' decision-making in the prior art, and to provide a diabetes medication regimen evaluation method and system that comprehensively considers individual differences among patients and differences in drug properties, and improves the adaptability and accuracy of diabetes medication regimens through scientific evaluation methods, thereby enhancing patients' treatment effects and confidence and optimizing doctors' medication decision-making process.

[0007] In a first aspect, in order to solve the above technical problems, the present invention provides a method for evaluating a diabetes medication regimen, comprising: Constructing a database, and entering the comprehensive data of multiple diabetic patients into the database; wherein each piece of the comprehensive data includes multi-dimensional detection data and medication regimen data of the corresponding patient; Obtain the current multi-dimensional test data and current medication regimen data of the target patient; In combination with the database, based on the current multi-dimensional detection data, the current medication regimen data is evaluated for medication adaptability to obtain a first evaluation result; Setting an online evaluation window, obtaining multi-party medication adaptability evaluation data on the doctor's side based on the current multi-dimensional detection data and the current medication regimen data; analyzing the multi-party medication adaptability evaluation data to obtain a second evaluation result; The first evaluation result and the second evaluation result are integrated to determine the evaluation result of the medication regimen for the target patient.

[0008] In one embodiment of the present invention, comprehensive data of multiple diabetic patients are entered into the database, including: Create a hierarchical data storage structure, which includes a primary layer, a secondary layer, a tertiary layer, and a quaternary layer; Among them, patient identity information is stored based on the first-level layer; data type information is stored based on the second-level layer; timestamps of various data under each data type are stored according to time series based on the third-level layer; and data information under each data type is stored based on the fourth-level layer.

[0009] In one embodiment of the present invention, in combination with the database, the current medication regimen data is evaluated for medication adaptability based on the current multi-dimensional detection data to obtain a first evaluation result, including: Extracting data from the database using the current medication regimen data as a search target to obtain a sample data set; Performing individual feature recognition on the sample data set to obtain an individual feature sample set; Performing sugar index identification on the sample data set to obtain a sugar index sample set; Performing drug feature recognition on the sample data set to obtain a drug feature sample set; Performing blood sugar lowering index identification on the sample data set to obtain a blood sugar lowering index sample set; Constructing a convolutional neural network model, using the individual feature sample set, the sugar index sample set and the drug feature sample set as input data, and using the blood sugar lowering index sample set as output data to train the convolutional neural network model until a convergence condition is met to obtain a blood sugar lowering index predictor; The current multi-dimensional detection data and the current medication regimen data are input into the blood sugar lowering index predictor to obtain the first evaluation result.

[0010] In one embodiment of the present invention, the multiple medication adaptability evaluation data is analyzed to obtain a second evaluation result, including: determining a treatment goal for the target patient; Determine a multivariate evaluation dimension according to the treatment goal; wherein the multivariate evaluation dimension at least includes a drug efficacy evaluation dimension, a drug safety evaluation dimension, a drug tolerance evaluation dimension, an individualized adaptation evaluation dimension, and an economic evaluation dimension; A multidimensional evaluation space is established according to the multivariate evaluation dimensions; wherein the drug efficacy evaluation dimension, the drug safety evaluation dimension, the drug tolerance evaluation dimension, the individualized adaptation evaluation dimension and the economic evaluation dimension each constitute a dimension of the multidimensional evaluation space; Mapping the multiple medication adaptability evaluation data to the multi-dimensional evaluation space to obtain multiple sets of spatial evaluation data; wherein the multiple sets of spatial evaluation data correspond one-to-one to the doctor end; The second evaluation result is determined by analyzing based on the multiple sets of spatial evaluation data.

[0011] In one embodiment of the present invention, analyzing and determining the second evaluation result based on the multiple sets of spatial evaluation data includes: Extracting spatial evaluation data of a first evaluation dimension from the multiple groups of spatial evaluation data, performing clustering on the spatial evaluation data of the first evaluation dimension, and obtaining a first cluster center; Calculating the distance ED between each of the spatial evaluation data of the first evaluation dimension and the first cluster center respectively, and generating a first dimension evaluation weight distribution of the spatial evaluation data of the first evaluation dimension according to the distance ED; The plurality of groups of spatial evaluation data are weighted according to the first dimensional evaluation weight distribution to determine the second evaluation result.

[0012] In one embodiment of the present invention, the smaller the distance ED is, the greater the weight of the corresponding space evaluation data is; conversely, the larger the distance ED is, the smaller the weight of the corresponding space evaluation data is.

[0013] In one embodiment of the present invention, the distance ED is calculated as follows: ; Where n is the coordinate dimension; is the position of the cluster center in the kth dimension; Evaluate the position of a data point in the kth dimension for one of the spaces; k is the dimension number of the multidimensional space, and its value range is 1, 2, 3, ...n.

[0014] In one embodiment of the present invention, fusing the first evaluation result and the second evaluation result to determine the evaluation result of the medication regimen of the target patient includes: Performing single matching mapping on the first evaluation result and the second evaluation result to obtain multiple mapping chains; Traversing the multiple mapping chains for synchronous proofreading to obtain N synchronous mapping chains and M asynchronous mapping chains; Performing intra-chain mean calculation on the M asynchronous mapping chains to obtain M valid asynchronous data; The N synchronous mapping chains and the M valid asynchronous data are combined to determine the evaluation result of the medication regimen of the target patient.

[0015] In one embodiment of the present invention, after obtaining the multi-drug suitability evaluation data on the doctor's side, it also includes: Identify the personal privacy data based on the online evaluation window, and encode and encrypt the multi-party medication adaptability evaluation data on the doctor's side to obtain the multi-party encrypted medication adaptability evaluation data; The multi-party encrypted medication suitability assessment data is analyzed to obtain a second assessment result.

[0016] In a second aspect, based on the same inventive concept, the present invention also provides a diabetes medication method evaluation system, comprising: A database for collecting comprehensive data of multiple diabetic patients; wherein each piece of the comprehensive data includes multi-dimensional test data and medication regimen data of the corresponding patient; A data acquisition module, which is used to obtain the current multi-dimensional detection data and current medication regimen data of the target patient; A first medication adaptability evaluation module, which is used to combine the database and perform medication adaptability evaluation on the current medication regimen data based on the current multi-dimensional detection data to obtain a first evaluation result; An online evaluation window module, which obtains the multi-party medication adaptability evaluation data of the current medication regimen data from the doctor side by setting an online evaluation window; A second medication adaptability assessment module, which is used to analyze the multiple medication adaptability assessment data to obtain a second assessment result; A fusion module is used to fuse the first evaluation result and the second evaluation result to determine the evaluation result of the medication regimen of the target patient.

[0017] The above technical solution of the present invention has the following beneficial effects compared with the prior art: The diabetes medication regimen evaluation method and system described in the present invention centrally manages the multi-dimensional detection data and medication regimen data of an area or region in a database, which is convenient for subsequent analysis and evaluation; the multi-dimensional detection data of the patient is combined with the medication regimen for evaluation, which helps to fully understand the impact of the drug on the patient, especially the differences between different individuals; this data-based evaluation method can improve the personalization and pertinence of the medication regimen; by combining the information in the database to evaluate the suitability of the medication, and providing an online evaluation window to obtain the multi-party evaluation data of the doctor, it can effectively combine clinical experience with data analysis, and improve the scientificity and feasibility of the medication regimen.

[0018] The above takes into account the individual differences of patients and the differences in drug characteristics, and improves the adaptability and accuracy of diabetes medication plans through scientific evaluation methods, thereby enhancing patients' treatment effects and confidence and optimizing doctors' medication decision-making process. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart of a method for evaluating a diabetes medication regimen in a preferred embodiment of the present invention; Figure 2 A flowchart of obtaining a first evaluation result in combination with a database in a preferred embodiment of the present invention; Figure 3 A flow chart of obtaining a second evaluation result in a preferred embodiment of the present invention; Figure 4 A flowchart of a preferred embodiment of the present invention that combines the first evaluation result and the second evaluation result; Figure 5 It is a structural block diagram of the diabetes medication method evaluation system in the preferred embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention. Example 1

[0021] Reference Figure 1 As shown, the embodiment of the present invention discloses a method for evaluating a diabetes medication regimen, comprising: S100, constructing a database, and entering comprehensive data of multiple diabetic patients into the database; wherein each piece of the comprehensive data includes multi-dimensional detection data and medication regimen data of the corresponding patient; In specific application scenarios, the database collects and centrally manages multidimensional detection data and medication regimen data of a district or region; each piece of the comprehensive data includes multidimensional detection data and medication regimen data of the corresponding patient. Specifically, the multidimensional detection data includes physiological detection data: including blood sugar level, glycosylated hemoglobin (HbA1c), weight, BMI, blood pressure, liver and kidney function, lipid indicators, urine microalbumin, etc.; genetic detection data: including drug metabolism genes (such as CYP450 genes, the variation of which may affect the patient's metabolic rate of certain drugs), genetic susceptibility genes, etc. (such as genes related to insulin resistance and diabetes development); lifestyle-related data: eating habits, exercise, sleep quality, smoking and drinking, etc.; medical history data: diabetes type, complications, other diseases, etc.; clinical imaging and diagnostic data: imaging examinations (such as fundus examination results of diabetic retinopathy, imaging data of heart disease, kidney examinations, etc.), other diagnostic results (such as diabetic foot examinations, neuropathy assessments, etc.); social factors: economic status and social support. Medication regimen data include: drug selection information: drug type, drug manufacturer, main ingredients of drugs, drug type, etc.; drug dosage and usage: drug dosage, medication frequency, medication time, dosage form, treatment course length, etc.; drug adverse reactions and side effects: drug side effect records, drug reaction records, etc.; pharmacoeconomic data: drug costs and drug cost-effectiveness analysis, etc. Each piece of comprehensive data contains the corresponding patient's multi-dimensional test data and medication regimen data, ensuring the comprehensiveness of the data and supporting personalized medication evaluation.

[0022] S200, obtaining current multi-dimensional detection data and current medication regimen data of the target patient; In a specific application scenario, the current multi-dimensional test data is directly obtained by combining the current test results and historical test results of the target patient. The current medical institution of the target patient prescribes medication regimen data based on the current multi-dimensional test data. The form of the medication regimen data is the medication regimen data included in the database as described above.

[0023] S300, combining the database, and performing medication adaptability evaluation on the current medication regimen data based on the current multi-dimensional detection data to obtain a first evaluation result; In specific application scenarios, data-driven model analysis is performed on the current medication regimen data based on the historical comprehensive data of multiple diabetic patients included in the database and the current multi-dimensional test data of the target patients, effectively combining historical data with individual differences to provide data support for subsequent doctor-side evaluations.

[0024] S400, setting an online evaluation window, obtaining multi-party medication adaptability evaluation data on the doctor's side based on the current multi-dimensional detection data and the current medication scheme data; analyzing the multi-party medication adaptability evaluation data to obtain a second evaluation result; In specific application scenarios, by setting up an online evaluation window, the doctor can provide an online medication suitability assessment based on the patient's current multi-dimensional test data. This process increases the expert's judgment and experience, strengthens the evaluation model that combines artificial intelligence with human experts, and through the evaluation results of different doctors on the same patient, it can improve the accuracy of diagnosis and reduce the deviation of individual doctors.

[0025] S500: Fusion the first evaluation result and the second evaluation result to determine an evaluation result of the medication regimen for the target patient.

[0026] In specific application scenarios, the first evaluation result based on data is integrated with the second evaluation result based on the doctors' multi-party experience to ensure the scientificity and feasibility of the medication plan.

[0027] The diabetes medication regimen evaluation method described in the present invention centrally manages the multidimensional detection data and medication regimen data of a district or region in the database, which is convenient for subsequent analysis and evaluation; the multidimensional detection data of the patient is combined with the medication regimen for evaluation, which helps to fully understand the impact of the drug on the patient, especially the differences between different individuals; this data-based evaluation method can improve the personalization and pertinence of the medication regimen; by combining the information in the database to evaluate the adaptability of the medication, and providing an online evaluation window to obtain the multi-party evaluation data of the doctor, it can effectively combine clinical experience with data analysis to improve the scientificity and feasibility of the medication regimen. Taking into account the individual differences of patients and the differences in drug characteristics, the adaptability and accuracy of the diabetes medication regimen are improved through scientific evaluation methods, thereby improving the treatment effect and confidence of patients and optimizing the doctor's medication decision-making process.

[0028] In a further improvement of an embodiment of the present invention, comprehensive data of multiple diabetic patients are entered into the database, including creating a hierarchical data storage structure, which includes a primary layer, a secondary layer, a tertiary layer and a fourth layer; wherein patient identity information is stored based on the primary layer; data type information is stored based on the secondary layer; timestamps of various data under each of the data types are stored according to a time series based on the third layer; and data information under each of the data types is stored based on the fourth layer.

[0029] In specific application scenarios, the database structure adopts a hierarchical data storage structure, which aims to provide efficient data storage, management and query capabilities. Each level has specific storage tasks, which can ensure the structured and flexible access of patient data as a whole, and support future expansion and real-time updates.

[0030] Specifically, the first layer is used to store the patient's unique identity and related basic information, such as: patient ID; personal information: such as name, gender, date of birth, contact information, address, etc.; basic medical history information: such as diabetes type (type 1, type 2, gestational diabetes, etc.), diagnosis date, medical history, etc. The information stored in the first layer is a fixed structure, and a relational database (such as MySQL, PostgreSQL) is used for easy query and management.

[0031] The secondary layer is used to store data type information, including clinical test data, medication regimens, lifestyle habits and complication information. The information stored in the secondary layer is non-relational data and uses a document-based database (such as MongoDB), which is suitable for storing data types with flexible structures and diverse levels.

[0032] The third level organizes data by time series, so that all data under each data type contains timestamp information. For patient test data and treatment records, time series can help track their changes. Use a time series database (such as InfluxDB or TimescaleDB) to store this data. The time series database is specially optimized for the storage and query of time series data and can efficiently process a large amount of timestamp-related data.

[0033] The fourth level stores specific data information related to data types, including the specific value of each data item and its additional information (such as test value, drug dosage, etc.). It can be stored in a combination of relational database and document database. For structured numerical data (such as blood sugar, weight, etc.), a relational database is used, and for irregular, time-related or multi-data type data, a document database is used.

[0034] In a further improvement of the embodiment of the present invention, reference is made to Figure 2 As shown, in combination with the database, the current medication regimen data is evaluated for medication adaptability based on the current multi-dimensional detection data to obtain a first evaluation result, including: Step 1: extracting data from the database using the current medication regimen data as a search target to obtain a sample data set; Step 2: performing individual feature recognition on the sample data set to obtain an individual feature sample set; Performing sugar index identification on the sample data set to obtain a sugar index sample set; Performing drug feature recognition on the sample data set to obtain a drug feature sample set; Performing blood sugar lowering index identification on the sample data set to obtain a blood sugar lowering index sample set; Step 3: constructing a convolutional neural network model, taking the individual feature sample set, the sugar index sample set and the drug feature sample set as input data, and taking the blood sugar lowering index sample set as output data to train the convolutional neural network model until a convergence condition is met to obtain a blood sugar lowering index predictor; Step 4: Input the current multi-dimensional detection data and the current medication regimen data into the blood sugar lowering index predictor to obtain the first evaluation result.

[0035] In specific application scenarios, the goal of the glucose-lowering index predictor is to predict the glucose-lowering index of the target patient based on the current medication regimen based on the historical data of a batch of patients collected in the database (including individual characteristics of the patient, glucose index, medication information and glucose-lowering index). A deep learning model (such as long short-term memory network LSTM, deep neural network, convolutional neural network, etc.) is used to perform deep learning on the historical comprehensive data of a batch of patients to obtain a glucose-lowering index predictor. The trained and verified effective glucose-lowering index predictor is put into actual use, and the current multidimensional detection data and current medication regimen data of the target patient are input into the glucose-lowering index predictor to obtain the glucose-lowering index.

[0036] In a further improvement of the embodiment of the present invention, reference is made to Figure 3 As shown, the multi-drug suitability assessment data is analyzed to obtain a second assessment result, including: determining a treatment goal for the target patient; Determine a multivariate evaluation dimension according to the treatment goal; wherein the multivariate evaluation dimension at least includes a drug efficacy evaluation dimension, a drug safety evaluation dimension, a drug tolerance evaluation dimension, an individualized adaptation evaluation dimension, and an economic evaluation dimension; A multidimensional evaluation space is established according to the multivariate evaluation dimensions; wherein the drug efficacy evaluation dimension, the drug safety evaluation dimension, the drug tolerance evaluation dimension, the individualized adaptation evaluation dimension and the economic evaluation dimension each constitute a dimension of the multidimensional evaluation space; Mapping the multiple medication adaptability evaluation data to the multi-dimensional evaluation space to obtain multiple sets of spatial evaluation data; wherein the multiple sets of spatial evaluation data correspond one-to-one to the doctor end; The second evaluation result is determined by analyzing based on the multiple sets of spatial evaluation data.

[0037] In specific application scenarios, based on the online evaluation window, multiple evaluation dimensions can be determined, such as drug efficacy evaluation dimension, drug safety evaluation dimension, drug tolerance evaluation dimension, individualized adaptation evaluation dimension and economic evaluation dimension, etc. Among them, drug efficacy evaluation includes blood sugar control, blood sugar volatility and the effect of drugs on diabetic complications; drug safety evaluation includes adverse reactions, effects on liver and kidney function and drug interactions; drug tolerance evaluation includes patient compliance, adverse reaction tolerance and convenience of medication; individualized adaptation evaluation includes age, gender differences, weight and BMI, comorbidities and underlying diseases and genetic factors, etc.; economic evaluation includes drug costs and medical insurance coverage, etc. Of course, according to needs, it can also include long-term effect and prognosis evaluation dimensions, multidisciplinary collaborative evaluation dimensions from endocrinology, cardiovascular, ophthalmology, nephrology and other specialties, and individualized evaluation dimensions of blood sugar control targets, etc. The multiple evaluation dimensions determined above can fully reflect the different aspects of the adaptability of the medication regimen and provide a more comprehensive evaluation perspective for the medication regimen. A multidimensional evaluation space is established based on the multi-dimensional evaluation dimensions, which is used to represent the combination relationship of different evaluation dimensions. The multiple medication suitability assessment data are mapped to the corresponding positions in the multidimensional assessment space, and the assessment results are displayed in a visual way. One set of spatial assessment data corresponds to the assessment data of one doctor. Multiple sets of spatial assessment data are jointly analyzed to determine the second assessment result. By determining the multi-evaluation dimensions based on the online assessment window, establishing a multi-dimensional assessment space, and determining the second assessment structure based on the joint analysis of multiple sets of spatial assessment data, this method can better reflect the opinions and views of different doctors on the medication plan, and provide valuable feedback for the evaluation, optimization and improvement of the medication plan.

[0038] Further, refer to Figure 3 As shown, according to the plurality of sets of spatial evaluation data, the second evaluation result is analyzed and determined, including: Extracting spatial evaluation data of a first evaluation dimension from the multiple groups of spatial evaluation data, performing clustering on the spatial evaluation data of the first evaluation dimension, and obtaining a first cluster center; Calculating the distance ED between each of the spatial evaluation data of the first evaluation dimension and the first cluster center respectively, and generating a first dimension evaluation weight distribution of the spatial evaluation data of the first evaluation dimension according to the distance ED; The plurality of groups of spatial evaluation data are weighted according to the first dimensional evaluation weight distribution to determine the second evaluation result.

[0039] Specifically, the distance ED is calculated as follows: ; Where n is the coordinate dimension; is the position of the cluster center in the kth dimension; Evaluate the position of a data point in the kth dimension for one of the spaces; k is the dimension number of the multidimensional space, and its value range is 1, 2, 3, ...n.

[0040] In a specific application scenario, the spatial evaluation data of the first evaluation dimension is extracted from the multiple groups of spatial evaluation data. The spatial evaluation data of the first evaluation dimension here is any one of the multiple dimensions, such as extracting the spatial evaluation data of the drug efficacy evaluation dimension from the multiple groups of spatial evaluation data. At this time, the method focuses on the evaluation data of the drug efficacy evaluation dimension, ensuring that the analysis focuses on the key evaluation dimension and providing accurate data input for subsequent cluster analysis. Cluster analysis is performed on the extracted spatial evaluation data of the first evaluation dimension, with the purpose of dividing the data into several groups so as to better identify different evaluation opinions. Common clustering methods include K-means, DBSCAN, etc. Through clustering, the characteristics of different patterns in the first evaluation dimension data can be identified, for example, which data points show similar treatment effects or reactions, so as to facilitate targeted evaluation. For each data point after clustering, the distance ED between it and the cluster center in the clustering result is calculated. The distance ED reflects the similarity between each evaluation data point and the cluster center. The smaller the distance, the closer the data point is to the cluster center. By calculating the distance, the degree of deviation of each data point from the cluster center is quantified, and then its weight in the evaluation dimension can be evaluated, which provides a basis for subsequent weighted evaluation. The first dimension evaluation weight distribution is generated according to the distance ED between each spatial evaluation data and the center of the first cluster. The smaller the distance, the greater the evaluation weight; conversely, the evaluation data points with a longer distance will be assigned a lower weight. The weight distribution can be generated by normalization or standardization. By adjusting the weight of each data point, the data points closer to the cluster center occupy a more important position in the evaluation, thereby improving the accuracy and reliability of the evaluation results. Using the generated first dimension evaluation weight distribution, all multiple groups of spatial evaluation data are weighted. The weighted value of each group of data will be associated with the evaluation weight of the first dimension in which it is located. The weighted evaluation results obtained by weighted calculation reflect the contribution of different dimensions to the overall evaluation results; through weighted calculation, it can be ensured that the importance of each dimension is reasonably reflected, making the evaluation results more accurate and reasonable. Through the comprehensive evaluation of multi-dimensional and multi-group coordinate data, more accurate medication regimen evaluation results can be provided, further improving the personalization and accuracy of medication regimen evaluation.

[0041] In a further improvement of the embodiment of the present invention, reference is made to Figure 4 As shown, fusing the first evaluation result and the second evaluation result to determine the evaluation result of the medication regimen for the target patient includes: Performing single matching mapping on the first evaluation result and the second evaluation result to obtain multiple mapping chains; Traversing the multiple mapping chains for synchronous proofreading to obtain N synchronous mapping chains and M asynchronous mapping chains; Performing intra-chain mean calculation on the M asynchronous mapping chains to obtain M valid asynchronous data; The N synchronous mapping chains and the M valid asynchronous data are combined to determine the evaluation result of the medication regimen of the target patient.

[0042] In a specific application scenario, a single matching mapping is performed on the first evaluation result and the second evaluation result to obtain multiple mapping chains; each mapping chain will record the association relationship between the corresponding data points in the first evaluation result and the second evaluation result. This mapping can help further identify the similarities and differences between the two evaluation results. Multiple mapping chains are synchronously proofread. The synchronous mapping chain refers to the part with smaller differences in the two evaluation results, indicating that the evaluation trend of this part of the data is relatively consistent; through proofreading, the common characteristics of these data are found and gathered together to form N synchronous mapping chains. Asynchronous mapping chains refer to data points with large differences in the two evaluation results. The differences may be caused by individual differences, drug reactions and other factors. These asynchronous data are separated to form M asynchronous mapping chains; through synchronous and asynchronous proofreading, while ensuring data consistency, it can be clarified which parts of the data have evaluation consistency and which data show large fluctuations and uncertainties, which is convenient for subsequent processing and analysis. The mean value within the chain is calculated for the M asynchronous mapping chains. The purpose of this step is to reduce the outliers or noise in the data and obtain representative "central values". For each asynchronous mapping chain, their mean is calculated (which may be the mean of some specific indicators) to generate M valid asynchronous data. These valid asynchronous data help to better understand the actual treatment effect of patients in the case of large differences; through mean calculation, the volatility in the data can be reduced, making the final evaluation results more stable, avoiding excessive interference to the evaluation results due to a single abnormal data point. Combining N synchronous mapping chains and M valid asynchronous data, comprehensively considering the contribution of these two types of data to the evaluation results, these two types of data can be fused through weighted averaging, synthetic functions, etc., to obtain the final evaluation results of the medication regimen; by fusing synchronous and valid asynchronous data, comprehensively considering the factors of stability and uncertainty, the evaluation results obtained are more comprehensive, scientific and accurate.

[0043] In a further improvement of an embodiment of the present invention, after obtaining the multi-party medication adaptability assessment data on the doctor's side, it also includes identifying personal privacy data based on the online assessment window, encoding and encrypting the multi-party medication adaptability assessment data on the doctor's side to obtain the multi-party encrypted medication adaptability assessment data; analyzing the multi-party encrypted medication adaptability assessment data to obtain a second assessment result.

[0044] In specific application scenarios, it is necessary to identify personal privacy information in the multi-party medication adaptability assessment data from the doctor's side. This privacy information may include sensitive information such as the patient's name, ID number, contact information, medical history, test data, etc. Through data identifiers or rules, sensitive data that needs to be encrypted is automatically identified; ensure that sensitive data can be accurately identified and isolated, thereby reducing the risk of privacy leakage during the analysis process. Once the privacy data that needs to be protected is identified, the data will be encoded and encrypted. Symmetric encryption (such as AES) or asymmetric encryption (such as RSA) algorithms can be used to encrypt the data to ensure that the data will not be accessed by unauthorized personnel during transmission and storage. The encrypted data is merged into multi-party encrypted medication adaptability assessment data. The evaluation data of each doctor is uniformly summarized after encryption to ensure that these encrypted data can be used for further analysis at the same time without exposing the patient's personal privacy information; by aggregating encrypted data, it is ensured that the evaluation data of different doctors can still maintain the security and integrity of the data when used in a multi-party environment, providing protection for further analysis. Example 2

[0045] Based on the same inventive concept, Figure 5 As shown, the present invention also provides a diabetes medication method evaluation system, comprising: A database for collecting comprehensive data of multiple diabetic patients; wherein each piece of the comprehensive data includes multi-dimensional test data and medication regimen data of the corresponding patient; A data acquisition module, which is used to obtain the current multi-dimensional detection data and current medication regimen data of the target patient; A first medication adaptability evaluation module, which is used to combine the database and perform medication adaptability evaluation on the current medication regimen data based on the current multi-dimensional detection data to obtain a first evaluation result; An online evaluation window module, which obtains the multi-party medication adaptability evaluation data of the current medication regimen data from the doctor side by setting an online evaluation window; A second medication adaptability assessment module, which is used to analyze the multiple medication adaptability assessment data to obtain a second assessment result; A fusion module is used to fuse the first evaluation result and the second evaluation result to determine the evaluation result of the medication regimen of the target patient.

[0046] The diabetes medication method evaluation system of the embodiment of the present invention is used to execute the above-mentioned diabetes medication regimen evaluation method, and has the same technical effect in the process of solving technical problems, which will not be repeated here.

[0047] In summary, the diabetes medication regimen evaluation method and system described in the present invention centrally manages the multi-dimensional detection data and medication regimen data of diabetes patients in a district, a region or the whole country in the database, which is convenient for subsequent analysis and evaluation; the multi-dimensional detection data of patients is combined with the medication regimen for evaluation, which helps to fully understand the impact of drugs on patients, especially the differences between different individuals; this data-based evaluation method can improve the personalization and pertinence of medication regimens; by combining the information in the database to evaluate the adaptability of medication, and providing an online evaluation window to obtain the multi-party evaluation data of doctors, it can effectively combine clinical experience with data analysis to improve the scientificity and feasibility of medication regimens. Taking into account the individual differences of patients and the differences in drug characteristics, the adaptability and accuracy of diabetes medication regimens are improved through scientific evaluation methods, thereby improving the treatment effect and confidence of patients and optimizing the medication decision-making process of doctors.

[0048] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0049] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0050] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0051] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0052] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A method for evaluating a diabetes medication regimen, characterized in that: include, Constructing a database, and entering the comprehensive data of multiple diabetic patients into the database; wherein each piece of the comprehensive data includes multi-dimensional detection data and medication regimen data of the corresponding patient; Obtain the current multi-dimensional test data and current medication regimen data of the target patient; In combination with the database, based on the current multi-dimensional detection data, the current medication regimen data is evaluated for medication adaptability to obtain a first evaluation result; Setting an online evaluation window, obtaining multi-party medication adaptability evaluation data on the doctor's side based on the current multi-dimensional detection data and the current medication regimen data; analyzing the multi-party medication adaptability evaluation data to obtain a second evaluation result; The first evaluation result and the second evaluation result are integrated to determine the evaluation result of the medication regimen for the target patient.

2. The method for evaluating a diabetes medication regimen according to claim 1, characterized in that: The comprehensive data of multiple diabetic patients are entered into the database, including, Create a hierarchical data storage structure, which includes a primary layer, a secondary layer, a tertiary layer, and a quaternary layer; Among them, patient identity information is stored based on the first-level layer; data type information is stored based on the second-level layer; timestamps of various data under each data type are stored according to time series based on the third-level layer; and data information under each data type is stored based on the fourth-level layer.

3. The method for evaluating a diabetes medication regimen according to claim 1 or 2, characterized in that: In combination with the database, the current medication regimen data is evaluated for medication adaptability based on the current multi-dimensional detection data to obtain a first evaluation result, including: Extracting data from the database using the current medication regimen data as a search target to obtain a sample data set; Performing individual feature recognition on the sample data set to obtain an individual feature sample set; Performing sugar index identification on the sample data set to obtain a sugar index sample set; Performing drug feature recognition on the sample data set to obtain a drug feature sample set; Performing blood sugar lowering index identification on the sample data set to obtain a blood sugar lowering index sample set; Constructing a convolutional neural network model, using the individual feature sample set, the sugar index sample set and the drug feature sample set as input data, and using the blood sugar lowering index sample set as output data to train the convolutional neural network model until a convergence condition is met to obtain a blood sugar lowering index predictor; The current multi-dimensional detection data and the current medication regimen data are input into the blood sugar lowering index predictor to obtain the first evaluation result.

4. The method for evaluating a diabetes medication regimen according to claim 1, characterized in that: Analyzing the multiple medication suitability assessment data to obtain a second assessment result, including: determining a treatment goal for the target patient; Determine a multivariate evaluation dimension according to the treatment goal; wherein the multivariate evaluation dimension at least includes a drug efficacy evaluation dimension, a drug safety evaluation dimension, a drug tolerance evaluation dimension, an individualized adaptation evaluation dimension, and an economic evaluation dimension; A multidimensional evaluation space is established according to the multivariate evaluation dimensions; wherein the drug efficacy evaluation dimension, the drug safety evaluation dimension, the drug tolerance evaluation dimension, the individualized adaptation evaluation dimension and the economic evaluation dimension each constitute a dimension of the multidimensional evaluation space; Mapping the multiple medication adaptability evaluation data to the multi-dimensional evaluation space to obtain multiple sets of spatial evaluation data; wherein the multiple sets of spatial evaluation data correspond one-to-one to the doctor end; The second evaluation result is determined by analyzing based on the multiple sets of spatial evaluation data.

5. The method for evaluating a diabetes medication regimen according to claim 4, characterized in that: Analyzing and determining the second evaluation result according to the plurality of sets of spatial evaluation data includes: Extracting spatial evaluation data of a first evaluation dimension from the multiple groups of spatial evaluation data, performing clustering on the spatial evaluation data of the first evaluation dimension, and obtaining a first cluster center; Calculating the distance ED between each of the spatial evaluation data of the first evaluation dimension and the first cluster center respectively, and generating a first dimension evaluation weight distribution of the spatial evaluation data of the first evaluation dimension according to the distance ED; The plurality of groups of spatial evaluation data are weighted according to the first dimensional evaluation weight distribution to determine the second evaluation result.

6. The method for evaluating a diabetes medication regimen according to claim 5, characterized in that: The smaller the distance ED is, the greater the weight of the corresponding space evaluation data is; conversely, the larger the distance ED is, the smaller the weight of the corresponding space evaluation data is.

7. The method for evaluating a diabetes medication regimen according to claim 5 or 6, characterized in that: The distance ED is calculated as follows: ； Where n is the coordinate dimension; is the position of the cluster center in the kth dimension; Evaluate the position of a data point in the kth dimension for one of the spaces; k is the dimension number of the multidimensional space, and its value range is 1, 2, 3, ...n.

8. The method for evaluating a diabetes medication regimen according to claim 1, characterized in that: The first evaluation result and the second evaluation result are integrated to determine the evaluation result of the medication regimen of the target patient, including: Performing single matching mapping on the first evaluation result and the second evaluation result to obtain multiple mapping chains; Traversing the multiple mapping chains for synchronous proofreading to obtain N synchronous mapping chains and M asynchronous mapping chains; Performing intra-chain mean calculation on the M asynchronous mapping chains to obtain M valid asynchronous data; The N synchronous mapping chains and the M valid asynchronous data are combined to determine the evaluation result of the medication regimen of the target patient.

9. The method for evaluating a diabetes medication regimen according to claim 1, characterized in that: After obtaining the multi-drug suitability assessment data from the doctor side, it also includes: Identify the personal privacy data based on the online evaluation window, and encode and encrypt the multi-party medication adaptability evaluation data on the doctor's side to obtain the multi-party encrypted medication adaptability evaluation data; The multi-party encrypted medication suitability assessment data is analyzed to obtain a second assessment result.

10. A diabetes medication method evaluation system, characterized by: include, A database for collecting comprehensive data of multiple diabetic patients; wherein each piece of the comprehensive data includes multi-dimensional test data and medication regimen data of the corresponding patient; A data acquisition module, which is used to obtain the current multi-dimensional detection data and current medication regimen data of the target patient; A first medication adaptability evaluation module, which is used to combine the database and perform medication adaptability evaluation on the current medication regimen data based on the current multi-dimensional detection data to obtain a first evaluation result; An online evaluation window module, which obtains the multi-party medication adaptability evaluation data of the current medication regimen data from the doctor side by setting an online evaluation window; A second medication adaptability assessment module, which is used to analyze the multiple medication adaptability assessment data to obtain a second assessment result; A fusion module is used to fuse the first evaluation result and the second evaluation result to determine the evaluation result of the medication regimen of the target patient.

Images