A multi-source clinical data aided decision method and system for childhood leukemia

Abstract

The application discloses a multi-source clinical data aided decision method and system for childhood leukemia, and belongs to the field of medical informatization, and comprises the following steps: acquiring a clinical data set for representing disease characteristics, treatment characteristics and physiological characteristics of a target object; performing similarity retrieval in a case database based on the clinical data set to obtain a comparable case set of the target object; and generating medication aided information for the target object based on the clinical data set and the comparable case set. The application realizes fine support for the medication aided decision process of childhood leukemia, improves the accuracy of comparable case screening, enhances the reliability of medication aided information, and reduces decision deviation caused by insufficient data utilization.

Description

Technical Field

[0001] This application belongs to the field of medical informatics, specifically relating to a multi-source clinical data-assisted decision-making method, system, device, storage medium, and computer program product for childhood leukemia. Background Technology

[0002] Childhood leukemia is a malignant hematological disease with complex pathogenesis and rapid progression. Its clinical diagnosis and treatment typically require comprehensive consideration of multiple sources of clinical information, including disease subtype, molecular biological characteristics, previous treatment regimens, and physical indicators, to develop an individualized, holistic treatment plan. With the development of medical informatization, hospitals have gradually accumulated a large amount of case data related to childhood leukemia, including disease characteristics, treatment features, physiological characteristics, and medication records, providing a data foundation for supporting clinical decision-making.

[0003] In existing technologies, clinical data such as electronic medical records, laboratory test results, and past medication records are typically structured and used to provide medication references for clinicians based on preset rules or statistical models. For example, several cases with similar characteristics can be selected from a case database for clinicians to refer to their treatment plans and medication records.

[0004] However, due to the complexity of the data, existing case comparison processes are often based on simple statistics of medication records of similar cases. This makes it impossible for clinicians to effectively extract valid comparable cases from the case database as a valid reference for the formulation of treatment plans. Summary of the Invention

[0005] This application aims to provide a method, system, device, storage medium, and computer program product for multi-source clinical data-assisted decision-making in childhood leukemia, at least addressing the problem of insufficient refinement in the decision-making process for medication assistance in childhood leukemia.

[0006] In a first aspect, embodiments of this application disclose a multi-source clinical data-assisted decision-making method for childhood leukemia, including:

[0007] Obtain clinical datasets to characterize the disease features, treatment features, and physiological features of the target object;

[0008] Based on the clinical dataset, a similarity search is performed in the case database to obtain a comparable case set for the target object; the comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object;

[0009] Based on the clinical dataset and the comparable case set, medication support information for the target subject is generated.

[0010] Secondly, embodiments of this application also disclose a multi-source clinical data-assisted decision-making system for childhood leukemia, comprising:

[0011] The data acquisition module is used to acquire clinical datasets that characterize the disease features, treatment features, and physiological features of the target object.

[0012] The case comparison module is used to perform similarity retrieval in the case database based on the clinical dataset to obtain a comparable case set for the target object; the comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object;

[0013] The medication assistance module is used to generate medication assistance information for the target object based on the clinical dataset and the comparable case set.

[0014] Thirdly, embodiments of this application also disclose an electronic device, including a processor and a memory, wherein the memory stores a program or instructions that can run on the processor, and the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0015] Fourthly, embodiments of this application also disclose a readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0016] Fifthly, embodiments of this application also disclose a computer program product, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps described in the first aspect.

[0017] In summary, in this embodiment, by acquiring a clinical dataset that characterizes the disease, treatment, and physiological features of the target object, the system can comprehensively grasp the multidimensional clinical information of the target object before case retrieval, improving the feature coverage of subsequent case matching and laying a data foundation for accurately identifying comparable cases. Then, based on the clinical dataset, a similarity search is performed in the case database to obtain a comparable case set containing multiple target case data. This allows the case screening process to simultaneously consider multi-source information such as disease features, treatment features, physiological features, and medication record features, enhancing the refinement of case matching and reducing matching bias caused by missing data or one-sided features, thereby improving the effectiveness and reference value of the comparable case set. Finally, medication assistance information for the target object is generated based on the clinical dataset and the comparable case set. This allows the system to provide more targeted medication references for the target object based on a comprehensive analysis of multi-source clinical features and historical medication experience. This not only improves the individualization of medication assistance information but also reduces the inadequacy of reference caused by relying on a single statistical method, making the generated medication assistance information closer to the actual clinical characteristics of the target object. Therefore, the method based on the embodiments of this application provides refined support for the decision-making process of medication for childhood leukemia, improves the accuracy of comparable case screening, enhances the reliability of medication support information, and reduces decision bias caused by insufficient data utilization. Attached Figure Description

[0018] In the attached diagram:

[0019] Figure 1 This is a flowchart illustrating the steps of a multi-source clinical data-assisted decision-making method for childhood leukemia, as provided in an embodiment of this application.

[0020] Figure 2 This is a flowchart illustrating the steps of another multi-source clinical data-assisted decision-making method for childhood leukemia provided in this application embodiment;

[0021] Figure 3 This is a block diagram of a multi-source clinical data-assisted decision-making system for childhood leukemia, provided in an embodiment of this application.

[0022] Figure 4 This is a block diagram of an electronic device provided in one embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, in this application, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0025] like Figure 1 The image shows a multi-source clinical data-assisted decision-making method for childhood leukemia provided in an embodiment of this application.

[0026] The method may include the following steps:

[0027] Step 101: Obtain a clinical dataset to characterize the disease features, treatment features, and physiological features of the target object.

[0028] In some embodiments of this application, since complete information on the target subject's disease status, past treatments, and basic physiological conditions is required before conducting case searches, a clinical dataset reflecting these aspects is needed. Specifically, this can be achieved by extracting disease features, treatment features, and physiological features from hospital information systems or relevant data sources, and then organizing this information into a clinical dataset suitable for analysis. Disease features typically describe the pathological type and its progression, treatment features reflect past treatment regimens and their responses, and physiological features present the patient's basic physiological state. In this way, the system can comprehensively grasp the multi-source clinical information of the target subject, providing the necessary data foundation for subsequent screening of comparable cases.

[0029] In a specific example, the target subject is a child leukemia patient undergoing treatment evaluation. The system retrieves the patient's genetic testing results, immunophenotyping records, previous chemotherapy regimens, past drug responses, age, weight, complete blood count results, and liver and kidney function test results from the hospital information system, and compiles this information into a clinical dataset. In this way, the system obtains multi-source clinical data reflecting the patient's disease characteristics, treatment history, and physiological state, providing a sufficient information foundation for subsequent retrieval of similar cases from the case database.

[0030] Step 102: Perform a similarity search in the case database based on the clinical dataset to obtain a set of comparable cases for the target object.

[0031] The comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object.

[0032] In some embodiments of this application, after organizing the clinical dataset, it is necessary to find cases with similar characteristics to the target object from the case database to provide comparable references for subsequent generation of medication assistance information. Therefore, a similarity search is required in the case database based on the clinical dataset. Specifically, this process involves comparing the target object's disease characteristics, treatment characteristics, and physiological characteristics with the corresponding characteristics of each case in the case database, and selecting several target case data based on the comparison results to form a comparable case set. The comparable case set typically presents a collection of cases that have a high degree of similarity to the target object in multi-source features, and includes the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics of these cases. After completing this step, the system obtains a set of cases with clinical characteristics relatively close to the target object, providing a necessary reference basis for subsequent generation of more targeted medication assistance information.

[0033] In a specific example, the system has already obtained the genetic testing results, previous chemotherapy regimens, age, blood routine test results, and liver and kidney function test results of a child with leukemia. The system then compares these characteristics one by one with historical cases stored in the case database, calculates the similarity between each case and the target subject, and selects several cases that are relatively similar in terms of disease type, previous treatment response, and physiological status. In this way, the system obtains a set of comparable cases, whose medication records and treatment results can provide a reference for subsequently generating medication support information.

[0034] Step 103: Based on the clinical dataset and comparable case set, generate medication support information for the target subjects.

[0035] In some embodiments of this application, after obtaining a comparable case set, it is necessary to generate medication support information for reference based on the medication records contained in these cases and the clinical characteristics of the target subject. Therefore, corresponding analysis based on the clinical dataset and the comparable case set is required. Specifically, this process involves organizing and aggregating the medication records in the comparable case set, and combining this with the target subject's disease characteristics, treatment characteristics, and physiological characteristics to deduce different drug combinations and their dosage ranges, forming medication support information suitable for the target subject. This medication support information typically presents drug combinations that may be suitable for the target subject, dosage ranges, and related risk warnings. Ultimately, this allows the system to provide more targeted medication references for the target subject based on multi-source clinical data and historical medication experience.

[0036] In a specific example, the system has obtained a clinical dataset of a child with leukemia and has selected several cases from the case database that are similar to this patient in terms of disease subtype, previous treatment response, and physiological status. The system then organizes the medication records of these cases, such as statistically analyzing common drug combinations, analyzing drug responses at different dosages, and combining the target patient's genetic testing results, blood routine test results, and liver and kidney function results to deduce several drug combinations and their dosage ranges that may be suitable for this patient. Finally, it generates a medication support information document containing drug combinations, dosage ranges, and related risk warnings, which can be used by clinicians when developing treatment plans.

[0037] In summary, in this embodiment, by acquiring a clinical dataset that characterizes the disease, treatment, and physiological features of the target object, the system can comprehensively grasp the multidimensional clinical information of the target object before case retrieval, improving the feature coverage of subsequent case matching and laying a data foundation for accurately identifying comparable cases. Then, based on the clinical dataset, a similarity search is performed in the case database to obtain a comparable case set containing multiple target case data. This allows the case screening process to simultaneously consider multi-source information such as disease features, treatment features, physiological features, and medication record features, enhancing the refinement of case matching and reducing matching bias caused by missing data or one-sided features, thereby improving the effectiveness and reference value of the comparable case set. Finally, medication assistance information for the target object is generated based on the clinical dataset and the comparable case set. This allows the system to provide more targeted medication references for the target object based on a comprehensive analysis of multi-source clinical features and historical medication experience. This not only improves the individualization of medication assistance information but also reduces the inadequacy of reference caused by relying on a single statistical method, making the generated medication assistance information closer to the actual clinical characteristics of the target object. Therefore, the method based on the embodiments of this application provides refined support for the decision-making process of medication for childhood leukemia, improves the accuracy of comparable case screening, enhances the reliability of medication support information, and reduces decision bias caused by insufficient data utilization.

[0038] Figure 2 This application provides another multi-source clinical data-assisted decision-making method for childhood leukemia.

[0039] The method may include the following steps:

[0040] Step 201: Obtain a clinical dataset to characterize the disease features, treatment features, and physiological features of the target object.

[0041] The method shown in this step has been explained in step 101 and will not be repeated here.

[0042] Step 202: Perform a similarity search in the case database based on the clinical dataset to obtain a comparable case set for the target object.

[0043] The comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object.

[0044] The method shown in this step has been explained in step 102 and will not be repeated here.

[0045] Optionally, the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics of each case are stored in the case database through corresponding case characteristics. Step 202 includes the following sub-steps:

[0046] Sub-step 2021: Determine the clinical characteristics of the target subjects based on the clinical dataset.

[0047] In some embodiments of this application, since it is necessary to extract core features that can be used for matching calculations from the multi-source clinical data of the target object before conducting similarity retrieval, it is necessary to determine the clinical features of the target object. Specifically, this can be achieved by organizing, filtering, and structuring the disease features, treatment features, and physiological features contained in the clinical dataset to form a feature set that can be used for subsequent similarity calculations. Clinical features are typically used to present key descriptions of the target object's disease state, past treatment, and basic physiological condition, and are important inputs for subsequent comparison of case features. In this way, the system can obtain a set of structured clinical features, providing the necessary basic information for subsequent calculation of the similarity between the target object and case objects in the database.

[0048] In a specific example, the system has acquired the genetic testing results, immunophenotyping data, previous chemotherapy regimens, past drug responses, age, blood routine test results, and liver and kidney function test results of a child leukemia patient. The system then processes this data, for example, converting the genetic testing results and blood routine test results into comparable feature codes, extracting the previous chemotherapy regimens as treatment feature vectors, and summarizing the liver and kidney function indicators as physiological feature parameters. In this way, the system obtains a set of clinical features that reflect the patient's disease characteristics, treatment history, and physiological state, providing a clear input basis for subsequent similarity calculations and screening of comparable cases from the case database.

[0049] Sub-step 2022 involves identifying multiple target case data from the case database based on the similarity score between clinical characteristics and case characteristics of each case object stored in the case database, in order to obtain a comparable case set.

[0050] In some embodiments of this application, after determining the clinical characteristics of the target object, it is necessary to screen cases with similar characteristics from the case database to construct a comparable case set for subsequent analysis. Therefore, it is necessary to determine the target case data based on the similarity score between the clinical characteristics and the case characteristics. Specifically, the clinical characteristics of the target object can be compared with the case characteristics of each case object in the database, a similarity score can be calculated based on the comparison results, and a number of case objects can be selected as target case data according to the score. The similarity score is usually used to reflect the degree of similarity between the target object and each case object in terms of disease characteristics, treatment characteristics, and physiological characteristics, and is an important basis for forming a comparable case set. In this way, the system can screen out cases with relatively similar clinical characteristics to the target object from a large number of cases, providing more targeted reference data for the subsequent generation of medication assistance information.

[0051] In a specific example, the system has obtained the clinical characteristics of a child with leukemia, such as gene testing results, immunophenotyping, previous chemotherapy regimens, blood routine test results, and liver and kidney function indicators. The system then compares these characteristics with the characteristics of each historical case stored in the case database, such as whether the gene mutation types are consistent, whether the previous treatment regimens are similar, and whether the physiological indicators are within a similar range, and calculates a similarity score based on these comparison results. In this way, the system selects several cases from the database that are relatively close to the target patient in terms of disease type, treatment response, and physiological status, and forms these cases into a comparable case set, providing a clear reference basis for subsequent generation of medication support information.

[0052] In the aforementioned sub-steps 2021 and 2022, mature models widely used in the field of medical data analysis can be employed to implement the corresponding analysis and computation functions. These include Gradient Boosting Decision Tree (GBDT) models, Random Forest (RF) models, Multilayer Perceptron (MLP) models, or Cosine Similarity Model (CSM) models based on vector retrieval. These models can jointly model disease features, treatment features, physiological features, and medication record features to support tasks such as feature extraction, similarity calculation, case screening, or risk analysis. The training dataset for these models can originate from multi-source clinical data accumulated in hospital information systems, including a large number of historical case samples with disease features, treatment features, physiological features, and medication record features. This data can be combined with manually labeled case similarity tags, medication response tags, or treatment outcome tags for training, enabling the model to learn the relationships between different clinical features. It should be noted that the specific training methods and implementation details of the corresponding models have been fully disclosed and applied in existing technologies, and will not be elaborated upon in this application.

[0053] Step 203: Based on the clinical dataset and comparable case set, generate medication support information for the target subjects.

[0054] The method shown in this step has been explained in step 103 and will not be repeated here.

[0055] Optionally, step 203 includes the following sub-steps:

[0056] Sub-step 2031: Aggregate multiple medication record features in the comparable case set to obtain medication aggregation features, and determine multiple prescription data for the target object based on the medication aggregation features.

[0057] The prescription data includes at least one combination of prescription data for the target object, and at least one dose estimation data corresponding to each combination of prescription data.

[0058] In some embodiments of this application, after obtaining a comparable case set, it is necessary to extract medication-related statistical information from these cases to deduce potentially applicable drug combinations and dosage ranges for the target population. Therefore, it is necessary to aggregate multiple medication record features from the comparable case set. Specifically, this can be achieved by organizing and statistically analyzing information such as drug names, drug combinations, dosages, and dosing frequencies recorded in the comparable case set, forming aggregated medication features that reflect overall medication trends. Aggregated medication features are typically used to present the frequency of occurrence, dosage distribution, and correspondence with clinical characteristics of different drug combinations in comparable cases, serving as a crucial basis for deriving prescription data. In this way, the system can determine multiple prescription data based on these aggregated features. Each prescription data includes at least one drug combination data and corresponding dosage estimation data, providing foundational information for subsequent risk assessment and decision support.

[0059] In a specific example, the system has already screened several cases from the case database that are relatively similar to the target subject in terms of disease type, previous treatment response, and physiological state. The system then organizes the medication records of these cases, such as counting the frequency of different drug combinations in these cases, analyzing the dosage range of each drug in different cases, and recording the use of these combinations during treatment. The system then obtains a set of medication aggregation characteristics, such as a certain drug combination being used in most cases, or a certain dosage range being common in multiple cases. Based on these aggregation characteristics, the system further derives multiple prescription data, each containing a single drug combination and the corresponding dosage estimate, providing a clear input basis for subsequent risk assessment.

[0060] Optionally, sub-step 2031 includes the following sub-steps:

[0061] Sub-step 20311: Based on the frequency aggregation of multiple medication record features, obtain medication aggregation features, and determine at least one prescription combination data based on the medication aggregation features.

[0062] In some embodiments of this application, after obtaining a comparable case set, it is necessary to identify representative drug combinations from the medication records of these cases to provide a basis for subsequent dosage analysis and prescription construction. Therefore, it is necessary to aggregate the frequency of multiple medication record features. Specifically, this can be done by statistically analyzing the frequency of different drug combinations in the comparable case set, analyzing the usage patterns of these combinations in historical cases, and extracting drug combinations with high frequency or clinical significance as medication aggregation features. Medication aggregation features are typically used to present the co-occurrence of different drug combinations in comparable cases and are an important basis for determining drug combination data. In this way, at least one drug combination can be determined based on these aggregation features, providing a clear input basis for subsequent dose interval estimation and prescription data construction.

[0063] In a specific example, the system has already screened several cases from the case database that are relatively similar to the target subject in terms of disease type, past treatment response, and physiological state. The system then statistically analyzes the medication records of these cases, such as recording the frequency of different drug combinations in these cases, identifying which drug combinations are used in most cases, or which combinations have a high frequency of use in specific treatment phases. After processing in this way, the system obtains a set of medication aggregation characteristics, such as the high frequency of a certain drug combination in comparable cases. Based on these aggregation characteristics, the system further identifies at least one drug prescription combination, providing a foundation for subsequent dose distribution analysis and prescription data generation.

[0064] Sub-step 20312: Based on the dose distribution analysis of each drug combination data, determine the dose interval estimation data corresponding to each drug combination data, and based on each dose interval estimation data, at least one dose estimation data.

[0065] In some embodiments of this application, after determining multiple drug combination data, it is necessary to further analyze the dosage usage of these drug combinations in comparable cases to provide more valuable dosage information for subsequent prescription data construction. Therefore, dose distribution analysis is required for each drug combination data. Specifically, this can be achieved by statistically analyzing the dosages corresponding to each drug combination in the comparable case set, analyzing its distribution range, common dose intervals, and correspondence with clinical characteristics in different cases, thereby determining the dose interval estimation data corresponding to each drug combination data. The dose interval estimation data is typically used to present the more common or safer dose range of a certain drug combination in historical cases. Based on this, the system can also derive at least one dose estimation data from the dose interval estimation data, which can be used as input for subsequent prescription data construction. This will enable the generation of corresponding dose interval estimates for each drug combination data, providing a clearer dosage basis for subsequent risk assessment and prescription selection.

[0066] In a specific example, the system has identified several drug combination data from the comparable case set, such as a combination of a chemotherapy drug and supportive medication. The system then statistically analyzes the dosage records of these drug combinations in the comparable cases, for example, analyzing whether the dosage of a particular drug in different cases is concentrated within a certain range, or whether it exhibits different dosage distributions at different treatment stages. After processing in this way, the system obtains a set of dose interval estimates, such as the dosage of a particular drug in most cases being concentrated within a relatively narrow range. Based on these dose interval estimates, the system further derives at least one dose estimate, providing a clear dosage reference for subsequent prescription data construction.

[0067] Sub-step 20313: Determine each corresponding combination of drug prescription data and dosage estimation data as a set of prescription data.

[0068] In some embodiments of this application, after determining the drug combination data and deriving the corresponding dose ranges and dose estimates, it is necessary to integrate these contents into a structured result that can be directly used for subsequent risk assessment and decision support. Therefore, each drug combination data and its corresponding dose estimate data need to be determined as a set of prescription data. Specifically, this can be achieved by pairing the drug combination data obtained in the preceding steps with its corresponding dose estimate data and organizing it according to a preset data structure to form prescription data that reflects the drug combination and its dosage recommendations. It should be noted that in actual clinical scenarios, the dosage ratio of the same drug combination may differ in different cases, and these differences often lead to different medication effects or risk levels. Therefore, the same drug combination may correspond to different prescription data under different dosage ratios. This application retains this dose-level difference when constructing prescription data, so that the generated prescription data can more accurately reflect the actual medication patterns in comparable cases. In this way, multiple sets of prescription data can be obtained, providing a clear basis for subsequent risk analysis and prescription screening.

[0069] In a specific example, the system has identified several drug prescription combinations based on a comparable case set, such as a combination of a chemotherapy drug and supportive medication, and obtained corresponding dose estimates through dose distribution analysis. The system then pairs each drug prescription combination with its corresponding dose estimate; for example, it associates the combination of "drug A + drug B" with its derived different dose ratios. Since there may be two different usage patterns in comparable cases—"drug A at a higher dose + drug B at a standard dose" and "drug A at a standard dose + drug B at a lower dose"—the system organizes these different dose ratios into separate prescription data sets. After processing in this way, the system obtains multiple sets of prescription data, each containing a drug prescription combination and its corresponding dose estimate, providing a structured and more clinically relevant input for subsequent medication risk assessment.

[0070] Sub-step 2032: Determine the medication risk assessment data for each prescription.

[0071] In some embodiments of this application, after obtaining multiple prescription data, it is necessary to analyze the potential medication risks of each prescription data to screen out inappropriate prescription data in subsequent steps. Therefore, it is necessary to determine the medication risk assessment data for each prescription data separately. Specifically, this can be achieved by comparing the drug combination data and its dosage estimation data in the prescription data with the clinical characteristics of the target subject, and combining this with the medication response, dose sensitivity, and related adverse events recorded in previous cases to assess the risk level of different drug combinations at the current dosage. Medication risk assessment data is typically used to present the potential risks of drug combinations to the target subject, including risk warnings such as excessive dosage, drug interactions, or mismatch with the target subject's physiological state. Finally, the system generates corresponding risk assessment results for each prescription data, providing a clear basis for subsequent prescription screening.

[0072] In a specific example, the system has derived several prescription data sets based on a comparable case set. Each prescription data set includes a drug combination and a corresponding dose estimate. The system then compares these prescription data sets with the clinical characteristics of the target subjects. For example, it checks whether there is a risk of overdose of a certain drug in the target subject's liver and kidney function status, or analyzes whether the drug combination includes any adverse reaction records related to specific gene mutations in previous cases. In this way, the system generates corresponding medication risk assessment data for each prescription data set. For example, it may mark a combination as having a dose sensitivity risk, or indicate that a combination has caused adverse reactions in similar cases, thus providing clear reference information for subsequent screening of inappropriate prescription data.

[0073] Optionally, sub-step 2032 includes the following sub-steps:

[0074] Sub-step 20321: Determine the risk characteristic data of each drug in each prescription data under the corresponding dose estimation data.

[0075] Among them, risk characteristic data is used to characterize each drug response under the corresponding dose estimation data, as well as the incidence of each drug response.

[0076] In some embodiments of this application, after obtaining multiple prescription data, it is necessary to further analyze the possible medication reactions of each drug in the prescription data at corresponding doses to provide a basis for subsequent risk stratification and prescription screening. Therefore, it is necessary to determine the risk characteristic data of each drug in each prescription data under the corresponding dose estimation data. Specifically, this can be achieved by comparing the dose estimation data of the drug with the medication reaction information recorded in comparable case sets, statistically analyzing the different types of medication reactions that may occur within similar dose ranges, and calculating the incidence rate of each type of medication reaction, thereby forming risk characteristic data. Risk characteristic data is typically used to present the potential reaction patterns of a drug at a specific dose, including common reactions, rare reactions, and dose-related sensitivity changes. In this way, the system can generate corresponding risk characteristic data for each drug in each prescription data, providing clear input for subsequent risk stratification matching.

[0077] In a specific example, the system has already determined the prescription combination of "Drug A + Drug B" and the corresponding dose estimation data for a certain prescription. The system then extracts drug reaction records from a comparable case set for drugs A and B within similar dose ranges. For example, it analyzes whether drug A at that dose is likely to cause gastrointestinal discomfort, skin reactions, or mild blood cell changes, and calculates the incidence of these reactions in historical cases. Similarly, the system also compiles the reaction types and incidence rates of drug B at the corresponding doses. In this way, the system obtains a set of risk characteristic data to characterize the potential drug reactions and their probabilities for each drug under the current dose estimation data, providing basic information for subsequent risk stratification and prescription screening.

[0078] Sub-step 20322: Match the risk characteristic data of each drug in the prescription data with the preset risk classification conditions to obtain the medication risk assessment data of the prescription data.

[0079] In some embodiments of this application, since different drugs may exhibit various types of medication reactions at different doses, and these reactions may differ in incidence, severity, duration, and correlation with the target subject's clinical status, after obtaining the risk characteristic data for each drug, it is necessary to further match these risk characteristics with preset risk grading conditions item by item to form more detailed medication risk assessment data. Specifically, this can be achieved by comparing the risk characteristic data of each drug in multiple dimensions, such as determining whether the incidence range of different reaction types exceeds the corresponding threshold, whether the dose sensitivity triggers specific risk rules, and whether the potential impact of the reaction conflicts with the target subject's physiological indicators or past medication history. The system will comprehensively analyze the prescription data based on these multi-dimensional matching results to generate structured medication risk assessment data. This assessment data may include risk levels, triggered risk factors, corresponding reaction types, incidence ranges, and matching with the target subject's clinical status. This will provide more detailed and traceable risk assessment results for each prescription, providing sufficient basis for subsequent prescription screening.

[0080] In a specific example, the system has generated risk characteristic data for drug A and drug B in a prescription dataset. For instance, drug A may cause mild gastrointestinal reactions (incidence rate approximately 12%–18%) and occasional skin reactions (incidence rate approximately 3%–5%) at the current dose, while drug B may cause a decrease in blood cell counts within a similar dose range (incidence rate approximately 8%–10%), and this reaction exhibits dose sensitivity in some cases. The system then compares these risk characteristics item by item with preset risk grading conditions, such as determining whether the incidence of gastrointestinal reactions exceeds the threshold for mild reactions, whether skin reactions are considered negligible risks, whether a decrease in blood cell counts triggers a dose sensitivity rule, or whether the reaction potentially conflicts with the target subject's current blood cell count. The system then integrates these matching results to form medication risk assessment data for the prescription dataset, including flagged risk factors, corresponding reaction types, incidence ranges, and matching with the target subject's clinical status, thus providing more detailed reference information for subsequent prescription screening.

[0081] Optionally, based on sub-steps 2031 and 2032, the multi-source clinical data-assisted decision-making method for childhood leukemia in this application further includes the following sub-steps:

[0082] Sub-step 2033 involves removing the target prescription data from multiple prescription datasets based on medication risk assessment data and clinical datasets.

[0083] Among them, the target prescription data creates a pre-defined medication risk for the target population.

[0084] In some embodiments of this application, after completing the risk assessment of multiple prescription data, it is necessary to further screen the prescription data based on the clinical status of the target patient to ensure that the final retained prescription data meets clinical needs in terms of both safety and applicability. Therefore, it is necessary to remove target prescription data based on medication risk assessment data and clinical datasets. This is to take into account that some patients may have further medication needs, such as allergy risk to specific drugs, or greater sensitivity to dosage when certain physiological indicators deviate from the normal range. When determining whether prescription data triggers preset risk conditions, these individualized risk factors are included in the analysis. If a prescription data does not match the clinical status of the target patient in terms of dosage, drug combination, or individualized risk, it is removed as target prescription data. This will retain prescription data with lower risk and more consistent with the clinical characteristics of the target patient, providing more reliable candidate solutions for subsequent medication support decisions.

[0085] In a specific example, the system has generated several prescription data sets for a child with leukemia and calculated corresponding medication risk assessment data for each prescription. The system then compares these risk assessment results with the patient's clinical data, such as checking whether a particular drug combination has a history of frequent adverse reactions related to specific gene mutations, or whether a certain dosage exceeds the safe range for the patient under their current liver and kidney function. Considering the patient's history of allergy to a certain type of drug, the system uses this allergy risk as an additional screening criterion during the analysis, highlighting prescription data containing that drug. After processing in this way, the system identifies several prescription data sets that may pose a pre-set medication risk to the patient and removes these from the candidate set, thus retaining prescription data more suitable for the patient and providing a safer reference for subsequent medication support decisions.

[0086] Step 204: Determine the prescription records for medication support information and the target case data of the target subjects under the prescription records, and update the case database according to the corresponding prescription records and target case data.

[0087] In some embodiments of this application, considering that after generating medication support information, the content related to this information can be incorporated into the case database in a structured manner, so as to continuously accumulate referable data in subsequent case searches. Specifically, this can be achieved by first identifying the prescription record corresponding to the medication support information, and then updating the database by combining the case data of the target object under that prescription record. The prescription record is usually a record form formed by clinicians based on the medication support information, after screening, selection, or adjustment according to the actual situation, and is used to reflect the finally adopted medication plan; while the medication support information is a reference result generated by the system based on multi-source clinical data, which may contain multiple candidate combinations or dosage ranges. In this way, the system writes the prescription record and its corresponding case data that have been manually confirmed into the case database, ensuring that the database content is valid and authentic, and providing a more reliable data source for subsequent case searches and decision support.

[0088] In a specific example, the system has generated medication support information for a child with leukemia, including multiple drug combinations, dosage ranges, and risk warnings. After referring to this information, clinicians, considering the patient's liver and kidney function, blood routine test results, past medication responses, and current treatment goals, selected one drug combination from multiple candidate options and adjusted the dosage range appropriately, forming the final prescription record. The system then collects relevant clinical data for the patient under this prescription record, such as medication responses, laboratory test results, or observation records during treatment, and organizes this content into target case data. In this way, the system writes the prescription record and target case data into the case database, adding a new case record related to this patient, providing more clinically relevant reference data for future similar case searches.

[0089] Optionally, in embodiments of this application, disease features are used to reflect the core state of the target object at the disease biological level, serving as an important foundation for the system to perform case comparison and risk analysis. These features typically reveal the genetic, immunological, and cellular characteristics of the disease from different detection dimensions. For example, gene testing data can be used to identify driver mutations or drug resistance-related variants, immunophenotyping data can be used to distinguish different disease subtypes, bone marrow morphology data can reflect changes in cell morphology, minimal residual disease monitoring data can be used to determine the disease burden level, and chromosome karyotype, fusion gene, and molecular biology testing data can provide deeper molecular typing information. By integrating at least one of the above disease features, the system can obtain a more discriminative input basis for case screening, similarity calculation, and subsequent medication analysis.

[0090] Optionally, in embodiments of this application, treatment features are used to present key information from the target subject's past treatment processes, serving as an important basis for the system to determine treatment response patterns, dose sensitivity, and potential risks. Previous chemotherapy regimen data reflects the treatment pathways received by the target subject; previous medication records can be used to identify drug combinations and dosage patterns; previous medication response data and adverse reaction records can reveal the target subject's sensitivity or tolerance to specific drugs; and previous treatment course assessment data and risk stratification data can provide an overall trend in disease progression and treatment effectiveness. By integrating at least one of the above treatment features, the system can more accurately identify the target subject's individualized treatment background during prescription generation and risk assessment.

[0091] Optionally, in embodiments of this application, physiological characteristics are used to reflect the target subject's basic physiological state and organ function, serving as an important reference for the system to make individualized adjustments during dose estimation and risk assessment. Age and weight data can be used to estimate basic dose requirements, vital sign monitoring data can reflect current physiological stability, while liver function, kidney function, and cardiac function test data directly affect drug metabolism, clearance, and tolerability. By integrating at least one of the above physiological characteristics, the system can more accurately assess the target subject's drug tolerance during dose range estimation, risk characteristic analysis, and prescription screening.

[0092] Furthermore, it should be noted that the specific content of the aforementioned disease characteristics, treatment characteristics, and physiological characteristics can be dynamically determined based on the actual situation of the target population. Different types of data may differ in terms of availability, completeness, and update frequency. For example, for newly diagnosed patients who have not yet received systemic treatment, some "previous" treatment characteristics may be absent; for patients who have not completed all molecular tests, some disease characteristics may be pending supplementation; for patients undergoing dynamic monitoring, certain physiological characteristics may be temporarily missing or delayed in updating due to different testing cycles. When acquiring these data, they can be labeled according to their actual availability, including different states such as null values, pending updates, partially missing, or only having interval information. Subsequently, during case comparison, feature analysis, and risk assessment, corresponding processing strategies can be further adopted, such as skipping corresponding feature dimensions, reducing the weight of partially missing features, using structured default rules, or performing local comparisons based on known information, to ensure that different data states do not affect the overall execution logic of the method in this application, thereby ensuring the stability and feasibility of the method while fully respecting the actual clinical data collection conditions.

[0093] Furthermore, it is important to emphasize that the acquisition, processing, and storage of the aforementioned disease characteristics, treatment characteristics, and physiological characteristics are all conducted in compliance with relevant laws, regulations, and industry standards. When establishing and updating the case database, the system will rely on the authorization and permission of the patient or their guardian, and will adhere to necessary privacy protection requirements during data collection, transmission, and storage. For content involving personal information, the system will process it according to applicable data anonymization standards to ensure that data analysis and model building are completed without exposing identifiable information. It should be noted that the relevant authorization, anonymization, and compliance processing are only used to meet the legality requirements of data use and will not affect the execution flow, functional implementation, or technical effect of the technical solution of this application, nor will they change the overall structure and implementation method of this application.

[0094] This application aims to provide clinicians with medication support information based on multi-source clinical data to support their more comprehensive judgment during diagnosis and treatment, rather than directly generating prescriptions or replacing prescription rights. Any medication suggestions, risk warnings, or dosage estimates output by the system are auxiliary analysis results, and their use should be conducted under the supervision of qualified professionals who combine clinical experience, the patient's actual condition, and relevant regulatory requirements to make the final prescription decision. The technical solution of this application does not change the legal nature of prescription rights, nor does it weaken the independent judgment responsibility of clinicians in the treatment process.

[0095] In summary, in this embodiment, by acquiring a clinical dataset that characterizes the disease, treatment, and physiological features of the target object, the system can comprehensively grasp the multidimensional clinical information of the target object before case retrieval, improving the feature coverage of subsequent case matching and laying a data foundation for accurately identifying comparable cases. Then, based on the clinical dataset, a similarity search is performed in the case database to obtain a comparable case set containing multiple target case data. This allows the case screening process to simultaneously consider multi-source information such as disease features, treatment features, physiological features, and medication record features, enhancing the refinement of case matching and reducing matching bias caused by missing data or one-sided features, thereby improving the effectiveness and reference value of the comparable case set. Finally, medication assistance information for the target object is generated based on the clinical dataset and the comparable case set. This allows the system to provide more targeted medication references for the target object based on a comprehensive analysis of multi-source clinical features and historical medication experience. This not only improves the individualization of medication assistance information but also reduces the inadequacy of reference caused by relying on a single statistical method, making the generated medication assistance information closer to the actual clinical characteristics of the target object. Therefore, the method based on the embodiments of this application provides refined support for the decision-making process of medication for childhood leukemia, improves the accuracy of comparable case screening, enhances the reliability of medication support information, and reduces decision bias caused by insufficient data utilization.

[0096] refer to Figure 3 This application illustrates a multi-source clinical data-assisted decision-making system 30 for childhood leukemia, as provided in an embodiment of this application, comprising:

[0097] The data acquisition module 301 is used to acquire clinical datasets that characterize the disease features, treatment features, and physiological features of the target object.

[0098] The case comparison module 302 is used to perform similarity retrieval in the case database based on the clinical dataset to obtain a comparable case set for the target object; the comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object;

[0099] Medication support module 303 is used to generate medication support information for target subjects based on clinical datasets and comparable case sets.

[0100] Optionally, the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics of each case are stored in the case database through corresponding case characteristics. The case comparison module 302 includes:

[0101] The clinical features submodule is used to determine the clinical features of the target object based on the clinical dataset.

[0102] The similarity retrieval submodule is used to identify multiple target case data from the case database based on the similarity score between clinical features and case features of each case object stored in the case database, in order to obtain a comparable case set.

[0103] Optionally, the medication support module 303 includes:

[0104] The prescription generation submodule is used to aggregate multiple medication record features from a comparable case set to obtain medication aggregation features, and to determine multiple prescription data for the target object based on the medication aggregation features; the prescription data includes at least one combination of prescriptions for the target object, and at least one dose estimation data corresponding to each combination of prescriptions.

[0105] The risk assessment submodule is used to determine the medication risk assessment data for each prescription.

[0106] Optionally, the medication support module 303 includes:

[0107] The prescription correction submodule is used to remove target prescription data from multiple prescription datasets based on medication risk assessment data and clinical datasets; the target prescription data poses a preset medication risk to the target subject.

[0108] Optionally, the prescription generation submodule includes:

[0109] A prescription unit is used to aggregate the frequency of combinations of multiple medication record features to obtain medication aggregation features, and to determine at least one prescription combination data based on the medication aggregation features.

[0110] A dose unit is used to determine dose interval estimation data corresponding to each drug combination data based on dose distribution analysis of each drug combination data, and to provide at least one dose estimation data based on each dose interval estimation data.

[0111] The prescription unit is used to determine a set of prescription data by combining each corresponding drug prescription combination data and dosage estimation data.

[0112] Optional, the risk assessment submodule includes:

[0113] The risk characteristic unit is used to determine the risk characteristic data of each drug in each prescription data under the corresponding dose estimation data; the risk characteristic data is used to characterize each drug response under the corresponding dose estimation data, as well as the incidence of each drug response;

[0114] The risk assessment unit is used to match the risk characteristics data of each drug in the prescription data with preset risk classification conditions to obtain medication risk assessment data of the prescription data.

[0115] Optionally, the multi-source clinical data-assisted decision-making system 30 for childhood leukemia also includes:

[0116] The data update module is used to determine the prescription records for medication support information, as well as the target case data of the target object under the prescription records, and to update the case database according to the corresponding prescription records and target case data.

[0117] In summary, in this embodiment, by acquiring a clinical dataset that characterizes the disease, treatment, and physiological features of the target object, the system can comprehensively grasp the multidimensional clinical information of the target object before case retrieval, improving the feature coverage of subsequent case matching and laying a data foundation for accurately identifying comparable cases. Then, based on the clinical dataset, a similarity search is performed in the case database to obtain a comparable case set containing multiple target case data. This allows the case screening process to simultaneously consider multi-source information such as disease features, treatment features, physiological features, and medication record features, enhancing the refinement of case matching and reducing matching bias caused by missing data or one-sided features, thereby improving the effectiveness and reference value of the comparable case set. Finally, medication assistance information for the target object is generated based on the clinical dataset and the comparable case set. This allows the system to provide more targeted medication references for the target object based on a comprehensive analysis of multi-source clinical features and historical medication experience. This not only improves the individualization of medication assistance information but also reduces the inadequacy of reference caused by relying on a single statistical method, making the generated medication assistance information closer to the actual clinical characteristics of the target object. Therefore, the method based on the embodiments of this application provides refined support for the decision-making process of medication for childhood leukemia, improves the accuracy of comparable case screening, enhances the reliability of medication support information, and reduces decision bias caused by insufficient data utilization.

[0118] Reference Figure 4 The electronic device 500 may include one or more of the following components: processing component 502, memory 504, power supply component 506, multimedia component 508, audio component 510, input / output (I / O) interface 512, sensor component 514, and communication component 516.

[0119] Processing component 502 typically controls the overall operation of electronic device 500, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 502 may include one or more processors 520 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 502 may include one or more modules to facilitate interaction between processing component 502 and other components. For example, processing component 502 may include a multimedia module to facilitate interaction between multimedia component 508 and processing component 502.

[0120] Memory 504 is used to store various types of data to support the operation of electronic device 500. Examples of this data include instructions for any application or method operating on electronic device 500, contact data, phonebook data, messages, pictures, multimedia, etc. Memory 504 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0121] Power supply component 506 provides power to various components of electronic device 500. Power supply component 506 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 500.

[0122] Multimedia component 508 includes an interface that provides an output interface between electronic device 500 and user. In some embodiments, the interface may include a liquid crystal display (LCD) and a touch panel (TP). If the interface includes a touch panel, the interface may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may not only sense the boundaries of touch or swipe actions but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 508 includes a front-facing camera and / or a rear-facing camera. When electronic device 500 is in an operating mode, such as shooting mode or multimedia mode, the front-facing camera and / or rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0123] Audio component 510 is used to output and / or input audio signals. For example, audio component 510 includes a microphone (MIC) used to receive external audio signals when electronic device 500 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 504 or transmitted via communication component 516. In some embodiments, audio component 510 also includes a speaker for outputting audio signals.

[0124] Input / output (I / O) interface 512 provides an interface between processing component 502 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0125] Sensor assembly 514 includes one or more sensors for providing state assessments of various aspects of electronic device 500. For example, sensor assembly 514 may detect the on / off state of electronic device 500, the relative positioning of components such as the display and keypad of electronic device 500, changes in position of electronic device 500 or a component of electronic device 500, the presence or absence of user contact with electronic device 500, orientation or acceleration / deceleration of electronic device 500, and temperature changes of electronic device 500. Sensor assembly 514 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 514 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 514 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0126] Communication component 516 facilitates wired or wireless communication between electronic device 500 and other devices. Electronic device 500 can access wireless networks based on communication standards, such as WiFi, carrier networks (such as 2G, 3G, 4G, or 5G), or combinations thereof. In one exemplary embodiment, communication component 516 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 516 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0127] In an exemplary embodiment, the electronic device 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to implement the methods provided in the embodiments of this application.

[0128] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 504 including instructions, which can be executed by a processor 520 of an electronic device 500 to perform the above-described method. For example, the non-transitory storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0129] In an exemplary embodiment, the electronic device 500 may also be provided as a server, including a processing component 502, which further includes one or more processors, and memory resources represented by memory 504 for storing instructions, such as applications, that can be executed by the processing component 502. The applications stored in memory 504 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 502 is configured to execute instructions to perform the methods provided in the embodiments of this application.

[0130] Electronic device 500 may also include a power supply component 506 configured to perform power management of electronic device 500, a wired or wireless communication component 516 configured to connect electronic device 500 to a network, and an input / output (I / O) interface 512. Electronic device 500 may operate on an operating system stored in memory 504, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or similar.

[0131] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the above methods. This application can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0132] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the multi-source clinical data-assisted decision-making method for childhood leukemia as described in this application.

[0133] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0134] It should be noted that, for the sake of simplicity, the method embodiments of this application are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of this application.

[0135] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0136] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the claimed rights.

Claims

1. A multi-source clinical data-assisted decision-making method for childhood leukemia, characterized in that, include: Obtain clinical datasets to characterize the disease features, treatment features, and physiological features of the target object; Based on the clinical dataset, a similarity search is performed in the case database to obtain a comparable case set for the target object; the comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object; Based on the clinical dataset and the comparable case set, medication support information for the target subjects is generated; The step of generating medication support information for the target subject based on the clinical dataset and the comparable case set includes: Multiple medication record features in the comparable case set are aggregated to obtain medication aggregation features, and multiple prescription data for the target object are determined based on the medication aggregation features; the prescription data includes at least one combination of prescriptions for the target object, and at least one dose estimation data corresponding to each combination of prescriptions; the medication aggregation features are obtained based on the aggregation of combination frequencies of multiple medication record features; Determine the medication risk assessment data for each of the aforementioned prescription data.

2. The multi-source clinical data-assisted decision-making method for childhood leukemia as described in claim 1, characterized in that, The disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics of each case are stored in the case database through corresponding case characteristics. The step of performing a similarity search in the case database based on the clinical dataset to obtain a comparable case set for the target object includes: The clinical characteristics of the target subjects are determined based on the clinical dataset; Based on the similarity score between the clinical features and the case features of each case object stored in the case database, multiple target case data are determined from the case database to obtain the comparable case set.

3. The multi-source clinical data-assisted decision-making method for childhood leukemia as described in claim 1, characterized in that, The aggregation of multiple medication record features in the comparable case set to obtain aggregated medication features, and the determination of multiple prescription data for the target object based on the aggregated medication features, includes: Based on the frequency aggregation of multiple medication record features, the medication aggregation feature is obtained, and at least one prescription combination data is determined based on the medication aggregation feature; Based on the dose distribution analysis of each drug combination data, dose interval estimation data corresponding to each drug combination data is determined, and at least one dose estimation data is obtained based on each dose interval estimation data; Each corresponding combination of drug prescription data and the dose estimation data are determined as a set of prescription data.

4. The multi-source clinical data-assisted decision-making method for childhood leukemia as described in claim 1, characterized in that, The determination of medication risk assessment data for each prescription data includes: Determine the risk characteristic data for each drug in each prescription data under the corresponding dose estimation data; the risk characteristic data is used to characterize each drug response under the corresponding dose estimation data, and the incidence of each drug response; The risk characteristic data of each drug in the prescription data is matched with preset risk classification conditions to obtain the medication risk assessment data of the prescription data.

5. The multi-source clinical data-assisted decision-making method for childhood leukemia as described in claim 1, characterized in that, The multi-source clinical data-assisted decision-making method for childhood leukemia also includes: The prescription record for the medication support information is determined, as well as the target case data of the target object under the prescription record, and the case database is updated according to the corresponding prescription record and target case data.

6. A multi-source clinical data-assisted decision-making system for childhood leukemia, characterized in that, include: The data acquisition module is used to acquire clinical datasets that characterize the disease features, treatment features, and physiological features of the target object. The case comparison module is used to perform similarity retrieval in the case database based on the clinical dataset to obtain a comparable case set for the target object; the comparable case set contains multiple target case data corresponding to the clinical dataset; each target case data is used to record the disease characteristics, treatment characteristics, physiological characteristics, and medication record characteristics corresponding to the corresponding case object; The medication assistance module is used to generate medication assistance information for the target object based on the clinical dataset and the comparable case set. The drug adjuvant module includes: A prescription generation submodule is used to aggregate multiple medication record features in the comparable case set to obtain medication aggregation features, and to determine multiple prescription data for the target object based on the medication aggregation features; the prescription data includes at least one combination of prescriptions for the target object, and at least one dose estimation data corresponding to each combination of prescriptions; the medication aggregation features are obtained based on the aggregation of combination frequencies of multiple medication record features; The risk assessment submodule is used to determine the medication risk assessment data for each of the prescription data.

7. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the multi-source clinical data-assisted decision-making method for childhood leukemia as described in any one of claims 1 to 5.

8. An electronic device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the multi-source clinical data-assisted decision-making method for childhood leukemia as described in any one of claims 1 to 5.

9. A computer program product, characterized in that, The computer program product stores a computer program that, when executed by a processor, implements the steps of the multi-source clinical data-assisted decision-making method for childhood leukemia as described in any one of claims 1 to 5.

Images