Method and device for assessing medication risk during pregnancy, and electronic device

Abstract

This application provides a method, device, and electronic device for assessing medication risks during pregnancy. Through automated knowledge graph querying and risk calculation, a risk assessment report can be quickly generated, reducing the time and cognitive load required for doctors to manually review literature and memorize complex rules. Simultaneously, it can match the current stage of medication risk to the patient's specific gestational age and perform comprehensive, real-time risk assessment based on the pregnant patient's real-time physiological indicators. This avoids the shortcomings of risk assessments that fail to consider gestational changes and individual patient differences, improving the accuracy of risk assessment. Furthermore, by acquiring medication information and physiological indicators, it can continuously and in real-time calculate the risk of medication use based on the patient's actual medication behavior and changes in physiological state during medication. Based on the real-time risk assessment results, medication guidance can be provided to patients, enhancing their understanding of the medication regimen and reducing the risks of medication use during pregnancy.

Description

Technical Field

[0001] This application relates to the field of medication risk assessment during pregnancy, and in particular to a method, apparatus and electronic device for medication risk assessment during pregnancy. Background Technology

[0002] Medication safety during pregnancy is a significant clinical challenge in perinatal medicine. Current clinical practice involves a complex and dynamically changing system of drug contraindications, requiring a comprehensive assessment of medication risks during pregnancy based on specific gestational age and maternal physiological indicators. This increases the memory burden on clinicians. Furthermore, the fragmentation of key medical data makes it difficult to monitor and evaluate medication behavior and processes in conjunction with changes in pregnant patients' physiological indicators and individual differences. This further leads to errors in risk assessment of medication use during pregnancy, impacting the health of both the patient and the fetus. Summary of the Invention

[0003] To overcome the problems existing in related technologies, this application provides a method, device and electronic device for assessing the risk of medication use during pregnancy.

[0004] The method for assessing the risk of medication use during pregnancy provided in this application includes: Based on the medications currently to be used by the pregnant patient and the current stage of pregnancy, search the knowledge graph for possible complications that may be caused by the use of the medications by the pregnant patient. For the aforementioned complication, based on the baseline weight set for the complication and the current gestational age of the pregnant patient, a current weight value for the risk that the complication may cause is determined; wherein, the greater the deviation of the current gestational age of the pregnant patient from the set standard gestational age, the higher the current weight value for the risk that the complication may cause. If the number of medications currently to be used by the pregnant patient is greater than 1, then for the complication, the superposition risk coefficient of the complication when the medications are superimposed is determined; the superposition risk coefficient is determined based on the risk coefficient of each medication causing the complication, wherein the risk coefficient of any medication causing the complication is determined based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the medication, and the current physiological indicators of the pregnant patient. For the aforementioned complication, a risk assessment score is evaluated based on the superimposed risk coefficient, the current weight value of the risk that the complication may cause, the basic risk coefficient of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

[0005] In some embodiments, if the number of medications currently to be used by the pregnant patient is greater than 1, the complications that may arise from the use of the medications by the pregnant patient include: The potential complications arising from each medication currently being used by the pregnant patient; and the potential complications arising from the combination of medications currently being used by the pregnant patient.

[0006] In some embodiments, determining the current weight value for the potential risk of the complication includes: For the aforementioned complications, determine the difference between the current gestational age and the standard gestational age of the pregnant patient; Based on the difference and the base weight set for the complication, the current weight value of the risk that the complication may cause is determined.

[0007] In some embodiments, if the number of medications currently to be used by the pregnant patient is 1, the method further includes: For each complication, a risk assessment score is evaluated based on the current weight of the risk that the complication may cause, the baseline risk coefficient of the drug, and the current physiological indicators of the pregnant patient.

[0008] In some embodiments, determining the superimposed risk coefficient of the complication when drugs are combined includes: For the aforementioned complication, the risk factor for the corresponding drug is determined based on the current weight value of the risk that the complication may cause, the baseline risk factor of each drug that causes the complication, and the current physiological indicators of the pregnant patient. Based on the risk coefficients and drug action coefficients of each drug corresponding to the complication, the cumulative risk coefficient of the complication when drugs are combined is determined.

[0009] In some embodiments, the risk assessment score for evaluating the risk posed by the complication includes: Calculate the sum of the baseline risk coefficients of each drug that causes this complication; Based on the calculation results, the superimposed risk coefficient, the current weight value of the risk that the complication may cause, and the individualized values ​​corresponding to the current physiological indicators of the pregnant patient, the risk assessment score of the risk caused by the complication is evaluated.

[0010] In some embodiments, the method further includes: If the risk assessment score of any complication is greater than a set threshold, a medication alternative is generated; at least one medication in the medication alternative is different from any medication currently to be used by the pregnant patient.

[0011] In some embodiments, generating a medication alternative includes: For complications whose risk assessment scores exceed a set threshold, identify M drugs that could cause such complications from the medications currently being used by the pregnant patient. The medications in the M medications are replaced according to a replacement strategy from few to many. After each replacement, the steps for finding possible complications caused by the use of the medication by the pregnant patient are retrieved from the knowledge graph based on the medication currently to be used by the pregnant patient and the current stage of pregnancy. Replacement stops when the risk assessment score of the complication is less than the set threshold. The replaced medication has the same efficacy as the original medication, and the replaced medication is an available medication that the pregnant patient is allowed to use.

[0012] A second aspect of this application provides a device for assessing the risk of medication use during pregnancy, comprising: The search module is configured to search the knowledge graph for possible complications that may be caused by the use of the drug by the pregnant patient, based on the medication that the pregnant patient is currently using and the current stage of pregnancy of the pregnant patient. The first determining module is configured to, for the complication, determine the current weight value of the risk that the complication may cause based on the base weight set for the complication and the current gestational age of the pregnant patient; wherein, the more the current gestational age of the pregnant patient deviates from the set standard gestational age, the higher the current weight value of the risk that the complication may cause. The second determining module is configured to determine the superposition risk coefficient of the complication when the drugs are superimposed if the number of drugs to be used by the pregnant patient is greater than 1. The superposition risk coefficient is determined based on the risk coefficient of each drug causing the complication, wherein the risk coefficient of any drug causing the complication is determined based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the drug, and the current physiological indicators of the pregnant patient. The assessment module is configured to assess the risk assessment score of the complication based on the superimposed risk coefficient, the current weight value of the risk that the complication may cause, the baseline risk coefficient of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

[0013] A third aspect of this application provides an electronic device, comprising: Processor- and machine-readable storage media; The machine-readable storage medium stores machine-executable instructions that can be executed by the processor; The processor is configured to execute machine-executable instructions to implement the method according to any one of the first aspects of the claims.

[0014] The technical solutions provided by the embodiments of this application may include the following beneficial effects: This application's embodiments firstly utilize automated knowledge graph queries and risk calculations to rapidly generate risk assessment reports, reducing the time and cognitive load required for doctors to manually review literature and memorize complex rules. Simultaneously, it can match the current stage of medication risk to the patient's specific gestational age and comprehensively calculate risk assessment results based on real-time physiological indicators of pregnant patients. This avoids the shortcomings of risk assessments that fail to consider gestational changes and individual patient differences, improving the accuracy of risk assessment. Furthermore, by acquiring medication information and physiological indicators, it can continuously and in real-time calculate the risk of medication use based on the patient's actual medication behavior and changes in physiological state during medication. Based on the real-time risk assessment results, medication guidance can be provided to patients, enhancing their understanding of medication regimens and reducing the risks of medication use during pregnancy.

[0015] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this application, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] Figure 1 This is a flowchart illustrating a method for assessing the risk of medication use during pregnancy, as shown in some exemplary embodiments; Figure 2 These are flowcharts illustrating one method for determining the current weight value, as shown in some exemplary embodiments; Figure 3 These are flowcharts illustrating one method for determining a superimposed risk coefficient, as shown in some exemplary embodiments; Figure 4 These are flowcharts illustrating one method for generating alternative medications, as shown in some exemplary embodiments. Figure 5 This is a structural schematic diagram of a pregnancy medication risk assessment device illustrated by some exemplary embodiments; Figure 6 These are schematic diagrams illustrating the structure of an electronic device through some exemplary embodiments. Detailed Implementation

[0018] The technical solutions in the embodiments (or "implementations") of this application will be clearly and completely described herein with reference to the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0019] If the embodiments of this application contain terms relating to directional indications or positional relationships (such as up, down, left, right, front, back, inside, outside, top, bottom, center, vertical, horizontal, longitudinal, transverse, length, width, counterclockwise, clockwise, axial, radial, circumferential, etc.), such terms are only used to explain the relative positional relationships and movement of the components in a specific posture (as shown in the attached figures); if the specific posture changes, the directional indications or positional relationships will also change accordingly. Furthermore, the terms "first" and "second" used in the embodiments of this application are only for descriptive convenience and should not be construed as indicating or implying relative importance.

[0020] Medication safety during pregnancy is a significant clinical challenge in perinatal medicine. Current clinical practice involves a complex and dynamically changing system of drug contraindications, requiring a comprehensive assessment of medication risks during pregnancy based on specific gestational age and maternal physiological indicators. This increases the memory burden on clinicians. Furthermore, the fragmentation of key medical data hinders continuous monitoring and real-time alerts of medication use and changes in physiological indicators during treatment, further leading to errors in risk assessments of medication use during pregnancy and impacting the health of both the patient and the fetus.

[0021] To address the aforementioned problems, this application proposes a method, apparatus, and electronic device for assessing medication risks during pregnancy. The following embodiments are provided to further illustrate this application: Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for assessing the risk of medication use during pregnancy, as shown in some exemplary embodiments. The method may include the following steps: Step S110: Based on the medications currently to be used by the pregnant patient and the current stage of pregnancy, search the knowledge graph for possible complications caused by the use of medications by the pregnant patient.

[0022] This can be achieved by using a Hospital Information System (HIS) to obtain the current test reports, historical electronic prescriptions, and other test reports of pregnant patients in real time. Alternatively, it can combine data from manually entered electronic medical records by doctors, data automatically synchronized from testing devices (such as blood glucose meters and fetal heart monitors), data entered by the patient, the actual medication pickup time recorded by the smart pillbox, and photos of self-purchased medication uploaded via an app on the patient's mobile phone. OCR (Optical Character Recognition) technology is then used to identify information in the photos to obtain the current medication to be used by the pregnant patient, the current stage of pregnancy, and the patient's current physiological indicators. The current stage of pregnancy can be divided into early pregnancy, mid-pregnancy, and late pregnancy based on the gestational week.

[0023] Understandably, if a pregnant patient has more than one medication currently in use, the potential complications from these medications can include the complications that each medication could cause, as well as the potential complications from the combined effects of all medications. For example, if the medications currently in use include aspirin and warfarin, the potential complications could include placental abruption (caused by aspirin), neural tube defects (caused by warfarin), and bleeding (caused by the combined effects of aspirin and warfarin).

[0024] The knowledge graph contains sub-graphs corresponding to different stages of pregnancy, and also includes information on complications associated with different medications at different stages of pregnancy. The knowledge graph can extract data on drug-induced complications at different stages of pregnancy from authoritative medical literature, drug instructions, clinical guidelines, and structured pharmaceutical databases.

[0025] For example, the Cypher query language of the Neo4j graph database can be used to identify the patient's current pregnancy stage code. By utilizing the graph traversal capability of the graph database, all medical knowledge subgraphs directly associated with the pregnancy stage node in the knowledge graph can be automatically retrieved and activated.

[0026] It should be noted that this process may include preprocessing data such as the currently used medication and the current stage of pregnancy for pregnant patients. Specifically, this may include standardizing medication names, for example, unifying the chemical name of aspirin as "Aspirin". For instance, after obtaining the name of the currently used medication (such as "acetylsalicylic acid" in a doctor's handwritten prescription or "Aspirin" entered by the patient via an app), the pre-trained medical entity recognition model LinseerEX can be used for initial mapping. This model has been trained on medical texts (including drug instructions, electronic medical records, and academic literature) and can accurately identify the correspondence between drug aliases, brand names, chemical names, and generic names. For example, inputting "acetylsalicylic acid" allows the model to map it to the standard corresponding chemical name node using its built-in medical dictionary and semantic understanding capabilities. For rare aliases, spelling errors, or new drug names that the pre-trained model cannot fully cover, a self-built entity encoder is used for deep matching. Medication use time can be mapped to the pregnancy stage, such as the 10th week of pregnancy being the first trimester. Invalid and abnormal test data are filtered out. For example, if the output result is a blood glucose level >20 mmol / L due to sensor malfunction, the data can be ignored. It should be noted that threshold ranges can be set for each test data point; if the result exceeds the threshold range, the test data is considered invalid.

[0027] Step S120: For complications, based on the baseline weight set for the complication and the current gestational age of the pregnant patient, determine the current weight value of the risk that the complication may cause; wherein, the more the current gestational age of the pregnant patient deviates from the set standard gestational age, the higher the current weight value of the risk that the complication may cause.

[0028] In one embodiment, see Figure 2 , Figure 2 This is a flowchart illustrating, through some exemplary embodiments, a method for determining a current weight value for the potential risk posed by a complication. Determining the current weight value for the potential risk posed by the complication may include the following steps: Step S210: For complications, determine the difference between the current gestational age and the standard gestational age of the pregnant patient.

[0029] Here, "current gestational age" refers to the patient's actual gestational age at the time of conception, while "standard gestational age" is the predetermined gestational age. The difference between the current gestational age and the standard gestational age can be understood as the deviation between the patient's actual gestational age and the standard gestational age; the actual gestational age can be less than or greater than the standard gestational age.

[0030] Step S220: Based on the difference and the base weight set for the complication, determine the current weight value of the risk that the complication may cause.

[0031] The potential complications are assigned a base weight at different gestational weeks. For example, in the second trimester, the base weight for placental abruption is 0.5. The current weight value for the potential risk of this complication can be determined using the following formula: Current weight value = Base weight set for this complication × (1 + 0.2) (Current gestational age - Standard gestational age) Understandably, based on the current weight values, the potential complications of the medication to be used can be dynamically reflected at the current gestational age of the pregnant patient, thus improving the timeliness of risk assessment for the medication to be used by pregnant patients. It can reflect the true risk level of the pregnant patient at the current stage of pregnancy in real time, avoiding the shortcomings of failing to adapt to individual differences in gestational age. This ensures that the risk assessment results are always synchronized with the patient's actual physiological state.

[0032] Step S130: If the number of medications currently to be used by the pregnant patient is greater than 1, then for the complication, determine the superimposed risk coefficient of the complication when the medications are superimposed; the superimposed risk coefficient is determined based on the risk coefficient of each medication causing the complication, wherein the risk coefficient of any medication causing the complication is determined based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the medication, and the current physiological indicators of the pregnant patient.

[0033] In one feasible embodiment, please refer to Figure 3 , Figure 3 This is a flowchart illustrating one of the exemplary embodiments for determining a cumulative risk factor. Determining the cumulative risk factor for the complication occurring when drugs are combined may include the following steps: Step S310: For complications, determine the risk factor of the complication for the corresponding drug based on the current weight value of the risk that the complication may cause, the baseline risk factor of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

[0034] Each drug has a predefined baseline risk coefficient. For example, the baseline risk coefficient for warfarin can be set at 1.0, and for aspirin at 0.3. Understandably, the higher the baseline risk coefficient, the higher the risk assessment score, and the greater the risk of complications for pregnant patients taking the medication. The individualized values ​​corresponding to the current physiological indicators of the pregnant patient can be assessed by the physician based on these indicators. For instance, the physician can pre-determine the weights of various examination indicators regarding their clinical importance in relation to drug metabolism or risk during pregnancy. For example, liver function might have a weight of 0.3, kidney function 0.3, blood pressure 0.2, blood sugar 0.1, and fetal heart rate 0.1. If the pregnant patient has abnormal liver function but other physiological indicators are normal, the individualized value is 1.3; if all indicators are normal, the individualized value is 1.0.

[0035] Step S320: Based on the risk coefficients and drug action coefficients of each drug corresponding to the complication, determine the superimposed risk coefficient of the complication when drugs are combined.

[0036] The drug action coefficient is determined based on all drugs to be used. The drug action coefficient can be determined according to the therapeutic effects of the drugs. For example, if the therapeutic effects of the drugs are synergistic, the drug action coefficient can be less than 1; if the therapeutic effects of the drugs are antagonistic, the drug action coefficient can be greater than 4; if the therapeutic effects of the drugs are additive or unrelated, the drug action coefficient can be greater than 1 and less than 4. It is understandable that when the therapeutic effects of the drugs exhibit both synergistic and antagonistic effects, the drug action coefficient can be evaluated based on the actual situation.

[0037] For example, the superposition risk coefficient of this complication when drugs are combined can be determined using the following formula: The cumulative risk coefficient = Σ(risk coefficient of drug A × risk coefficient of drug B × drug effect coefficient) Among them, Drug A and Drug B represent all drugs currently to be used. The risk coefficients of Drug A and Drug B can both be calculated using the following formula: Risk coefficient = (current weight value + basic drug risk coefficient) × individualized value.

[0038] Understandably, by referencing the current weighting value, the drug's risk coefficient, and the patient's individualized physiological indicators, it is possible not only to assess the dynamic risk weight of the drug at a specific gestational week, but also to refer to the drug's inherent risk attributes. Furthermore, based on the patient's real-time physiological state and individualized values, risk assessment becomes more precise and individualized, improving its reliability and accuracy.

[0039] Step S140: For complications, assess the risk assessment score of the complication based on the superimposed risk coefficient, the current weight value of the risk that the complication may cause, the basic risk coefficient of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

[0040] In one feasible embodiment, assessing the risk assessment score of the complication may include: calculating the sum of the basic risk coefficients of each drug that causes the complication; and assessing the risk assessment score of the complication based on the calculation results, the superimposed risk coefficients, the current weight value of the risk that the complication may cause, and the individualized values ​​corresponding to the current physiological indicators of the pregnant patient.

[0041] Specifically, Apache Flink can be used to handle streaming risk assessment, and the risk assessment score can be calculated according to the following formula: Risk assessment score = (Current weight value + Basic risk coefficient of drug A + Basic risk coefficient of drug B + Superimposed risk coefficient) × Individualized value In another feasible embodiment, if the number of medications currently to be used by the pregnant patient is 1, the method may further include: for each complication, assessing the risk assessment score of the complication based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the medication, and the current physiological indicators of the pregnant patient.

[0042] Specifically, the risk assessment score for the risk of this complication can be calculated using the following formula when the quantity of the drug to be used is 1: Risk assessment score = (current weight value + basic risk coefficient of drug) × individualized value.

[0043] The method in this embodiment, through automated knowledge graph querying and risk calculation, can quickly generate risk assessment reports, reducing the time and cognitive load required for doctors to manually review literature and memorize complex contraindications. Simultaneously, it can match the current stage of medication risk to the patient's specific gestational age and perform comprehensive, real-time risk assessment based on the pregnant patient's real-time physiological indicators. This avoids the shortcomings of risk assessments that fail to consider gestational changes and individual patient differences, improving the accuracy of risk assessment. Furthermore, by acquiring medication information and physiological indicators, it can continuously and in real-time calculate the risk of medication use based on the patient's actual medication behavior and changes in physiological state during medication. Based on the real-time risk assessment results, medication guidance can be provided to patients, enhancing their understanding of the medication regimen and reducing the medication risks during pregnancy.

[0044] In one feasible embodiment, the method may further include: generating a medication alternative if the risk assessment score of any complication is greater than a set threshold; at least one medication in the medication alternative is different from any medication currently to be used by the pregnant patient.

[0045] The thresholds can be pre-configured based on clinical guidelines, historical safety data, and expert experience, and can be set in stages according to the specific strategies of medical institutions.

[0046] Specifically, please refer to Figure 4 , Figure 4 This is a flowchart illustrating one of the exemplary embodiments for generating a medication alternative. Generating a medication alternative may include the following steps: Step S410: For complications with risk assessment scores greater than a set threshold, identify M drugs that cause the complication from the medications currently being used by the pregnant patient.

[0047] Specifically, if the risk assessment score for any complication caused by a currently used medication exceeds a set threshold, then M medications that could cause this complication can be identified from the pregnant patient's current medication list. For example, if the medications to be used include aspirin, warfarin, and metformin, and the risk assessment score for neural tube defects in the fetus among the complications caused by these three medications exceeds a set threshold, then warfarin and metformin can be identified from the pregnant patient's current medication list as two medications that could cause this complication.

[0048] Step S420: Replace each of the M drugs according to the replacement strategy from few to many, and after each replacement, return to the knowledge graph based on the drug to be used by the pregnant patient and the current stage of pregnancy of the pregnant patient, and search for the steps that may cause complications caused by the use of the drug by the pregnant patient, until the risk assessment score of the complication is less than the set threshold, and stop the replacement; wherein, the replaced drug has the same efficacy as the original drug, and the replaced drug is an available drug that the pregnant patient is allowed to use.

[0049] The process involves using a search algorithm to find drugs with the same efficacy as the original drug. The algorithm attempts to replace each of the M drugs using a strategy of replacing as few as possible. This can be understood as replacing any one drug initially. After the replacement, based on the pregnant patient's current medication and stage of pregnancy, the algorithm searches the knowledge graph for steps that might cause complications from drug use during pregnancy. This yields a risk assessment score after replacing one drug. If the risk assessment score for complications caused by the replaced drug is less than a set threshold, the replaced medication regimen is used. If the risk assessment score for complications caused by the replaced drug is greater than the set threshold, then, in addition to replacing one drug, another drug from the M drugs (excluding the one replaced initially) is replaced, resulting in a replacement regimen that replaces two drugs. After the replacement, based on the pregnant patient's current medication and stage of pregnancy, the steps for potential complications caused by medication use during pregnancy are searched in the knowledge graph. The risk assessment score after replacing two medications is obtained. The process continues until the risk assessment score is less than a set threshold. If the risk assessment score is less than a set threshold, the medications in the M medications will not be replaced again.

[0050] For example, if the medications to be used include aspirin, warfarin, and metformin, and the risk assessment score for neural tube defects in the fetus caused by these three medications is greater than a set threshold, then warfarin and metformin, which are currently the three medications the pregnant patient is using, are identified as potentially causing this complication. First, aspirin, which has platelet aggregation inhibition effects, is replaced with indobufen, which also has platelet aggregation inhibition effects. After the replacement, based on the pregnant patient's current medications and stage of pregnancy, the steps for potential complications caused by medication use during pregnancy are searched in the knowledge graph. The risk assessment score after replacing one medication is obtained. If the calculated risk assessment score is less than the set threshold, indobufen, warfarin, and metformin are selected as alternative medications. If the calculated risk assessment score is greater than the set threshold, after replacing indobufen with aspirin, warfarin is then replaced with heparin, which has the same efficacy. After the replacement, the steps that may cause complications when the pregnant patient uses the medication are found in the knowledge graph based on the current medication to be used by the pregnant patient and the current stage of pregnancy. The risk assessment score after replacing the two medications is then obtained.

[0051] This can be understood as follows: before developing alternative treatments, medications that are not permitted for use in pregnant patients can be excluded based on contraindication information and using a Constraint Satisfaction Problem Solver (CSP). For example, if a pregnant patient has abnormal liver function, medications contraindicated due to liver function abnormalities can be excluded from the alternative treatment pool.

[0052] It should be noted that the calculated risk assessment scores can also be categorized into different levels, with different risk levels corresponding to different thresholds. For example, threshold judgments can be performed based on the Drools (Declarative Rule-Oriented Language System) rule engine. The risk assessment score is compared with preset risk level thresholds to generate different risk level results and respond to different actions. For instance: a risk assessment score less than 0.1 indicates low risk, and only a log is recorded by the system without proactive alerts; a risk assessment score greater than or equal to 0.3 and less than 0.7 indicates medium risk, generating a yellow alert and pushing the risk level result to the doctor's workstation; a risk assessment score greater than or equal to 0.7 generates a red alert, informing both the doctor and the patient.

[0053] The method described in this application, through a gradual drug replacement strategy, can improve the accuracy of medication risk management during pregnancy. It can achieve risk control goals with minimal intervention while retaining the effective components of the original treatment plan, ensuring the safety and continuity of efficacy of the alternative treatment and improving medication safety for pregnant patients.

[0054] For a second aspect of the embodiments of this application, please refer to Figure 5 , Figure 5 This is a structural schematic diagram illustrating a pregnancy medication risk assessment device according to some exemplary embodiments. A pregnancy medication risk assessment device is provided, comprising: The search module 510 is configured to search the knowledge graph for possible complications caused by the use of medications by pregnant patients, based on the medications currently to be used by the pregnant patients and the current stage of pregnancy of the pregnant patients. The first determining module 520 is configured to determine the current weight value of the risk that the complication may cause, based on the base weight set for the complication and the current gestational age of the pregnant patient; wherein, the more the current gestational age of the pregnant patient deviates from the set standard gestational age, the higher the current weight value of the risk that the complication may cause. The second determining module 530 is configured to determine the superimposed risk coefficient of a complication when the number of drugs to be used by the pregnant patient is greater than 1 if the number of drugs to be used by the pregnant patient is greater than 1. The superimposed risk coefficient is determined based on the risk coefficient of each drug causing the complication, wherein the risk coefficient of any drug causing the complication is determined based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the drug, and the current physiological indicators of the pregnant patient. Assessment module 540 is configured to assess the risk assessment score of a complication based on the superimposed risk coefficient, the current weight value of the risk that the complication may cause, the baseline risk coefficient of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

[0055] Optionally, if the number of medications currently intended for use by a pregnant patient is greater than 1, the possible complications arising from medication use during pregnancy include: Potential complications of each medication currently being used by a pregnant patient; and potential complications due to the combination of medications currently being used by a pregnant patient.

[0056] Optionally, determine the current weight value of the risk that the complication may cause, including: For complications, determine the difference between the current gestational age and the standard gestational age of the pregnant patient; Based on the difference and the base weight set for the complication, determine the current weight value of the risk that the complication may cause.

[0057] Optionally, if the number of medications currently needed by a pregnant patient is 1, the assessment module further includes: For each complication, a risk assessment score is evaluated based on the current weight of the risk that the complication may cause, the baseline risk coefficient of the drug, and the current physiological indicators of the pregnant patient.

[0058] Optionally, the cumulative risk factor for the complication when drugs are combined can be determined, including: For complications, the risk factor of the complication for the corresponding drug is determined based on the current weight value of the risk that the complication may cause, the baseline risk factor of each drug that causes the complication, and the current physiological indicators of the pregnant patient. Based on the risk coefficients and drug action coefficients of each drug corresponding to the complication, the cumulative risk coefficient of the complication when drugs are combined is determined.

[0059] Optionally, a risk assessment score is used to evaluate the risk posed by this complication, including: Calculate the sum of the baseline risk coefficients of each drug that causes this complication; Based on the calculation results, the superimposed risk coefficient, the current weight value of the risk that the complication may cause, and the individualized values ​​corresponding to the current physiological indicators of the pregnant patient, the risk assessment score of the risk caused by the complication is evaluated.

[0060] Optionally, the device further includes: The generation module is configured to generate a medication alternative if the risk assessment score of any complication is greater than a set threshold; at least one medication in the medication alternative is different from any medication currently to be used by the pregnant patient.

[0061] Optionally, generating alternative medication options includes: For complications whose risk assessment scores exceed a set threshold, identify M drugs that could cause such complications from the medications currently being used by the pregnant patient. The algorithm attempts to replace each of the M drugs using a replacement strategy that starts with fewer drugs and gradually increases the number of drugs. After each replacement, it returns to the knowledge graph based on the current drugs to be used by the pregnant patient and the current stage of pregnancy. It searches for possible complications caused by the use of drugs by pregnant patients until the risk assessment score of the complications is less than a set threshold. The replacement drugs have the same efficacy as the original drugs and are the available drugs that are permitted for use by pregnant patients.

[0062] The device in this embodiment, through automated knowledge graph querying and risk calculation, can quickly generate risk assessment reports, reducing the time and cognitive load required for doctors to manually review literature and memorize complex contraindication rules. Simultaneously, it can match the medication risk at the current stage based on the patient's specific gestational age and perform comprehensive, real-time risk assessment based on the pregnant patient's real-time physiological indicators. This avoids the shortcomings of risk assessments that fail to consider gestational changes and individual patient differences, improving the accuracy of risk assessment. Furthermore, by acquiring medication information and physiological indicators, it can continuously and in real-time calculate the risk of medication use based on the patient's actual medication behavior and changes in physiological state during medication. Based on the real-time risk assessment results, medication guidance can be provided to patients, enhancing their understanding of the medication regimen and reducing the medication risks during pregnancy.

[0063] For the third aspect of this application, please refer to Figure 6 , Figure 6 These are structural schematic diagrams of an electronic device illustrating some exemplary embodiments. An electronic device is provided, comprising: a processor and a machine-readable storage medium; the machine-readable storage medium storing machine-executable instructions executable by the processor; the processor executing the machine-executable instructions to implement any of the methods in the first aspect.

[0064] For example, the aforementioned machine-readable storage medium can be any electronic, magnetic, optical, or other physical storage device that can contain or store information such as executable instructions, data, etc. For instance, machine-readable storage media can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drives (such as hard disk drives), solid-state drives, any type of storage disk (such as optical discs, DVDs, etc.), or similar storage media, or combinations thereof.

[0065] The electronic device described in this application, through automated knowledge graph querying and risk calculation, can quickly generate risk assessment reports, reducing the time and cognitive load required for doctors to manually review literature and memorize complex contraindication rules. Simultaneously, it can match the current stage of medication risk to the patient's specific gestational age and perform comprehensive, real-time risk assessment based on the pregnant patient's real-time physiological indicators. This avoids the shortcomings of risk assessments that fail to consider gestational changes and individual patient differences, improving the accuracy of risk assessment. Furthermore, by acquiring medication information and physiological indicators, it can continuously and in real-time calculate the risk of medication use based on the patient's actual medication behavior and changes in physiological state during medication. Based on the real-time risk assessment results, it can provide medication guidance to patients, enhancing their understanding of the medication regimen and reducing the medication risks during pregnancy.

Claims

1. A method for assessing the risk of medication use during pregnancy, characterized in that, include: Based on the medications currently to be used by the pregnant patient and the current stage of pregnancy, search the knowledge graph for possible complications that may be caused by the use of the medications by the pregnant patient. For the aforementioned complication, based on the baseline weight set for the complication and the current gestational age of the pregnant patient, a current weight value for the risk that the complication may cause is determined; wherein, the greater the deviation of the current gestational age of the pregnant patient from the set standard gestational age, the higher the current weight value for the risk that the complication may cause. If the number of medications currently to be used by the pregnant patient is greater than 1, then for the complication, the superposition risk coefficient of the complication when the medications are superimposed is determined; the superposition risk coefficient is determined based on the risk coefficient of each medication causing the complication, wherein the risk coefficient of any medication causing the complication is determined based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the medication, and the current physiological indicators of the pregnant patient. For the aforementioned complication, a risk assessment score is evaluated based on the superimposed risk coefficient, the current weight value of the risk that the complication may cause, the basic risk coefficient of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

2. The method according to claim 1, characterized in that, If the number of medications currently to be used by the pregnant patient is greater than 1, the complications that may arise from the use of the medication by the pregnant patient include: The potential complications arising from each medication currently being used by the pregnant patient; and the potential complications arising from the combination of medications currently being used by the pregnant patient.

3. The method according to claim 1, characterized in that, The determination of the current weight value for the potential risk of this complication includes: For the aforementioned complications, determine the difference between the current gestational age and the standard gestational age of the pregnant patient; Based on the difference and the base weight set for the complication, the current weight value of the risk that the complication may cause is determined.

4. The method according to claim 1, characterized in that, If the number of medications currently to be used by the pregnant patient is 1, then the method further includes: For each complication, a risk assessment score is evaluated based on the current weight of the risk that the complication may cause, the baseline risk coefficient of the drug, and the current physiological indicators of the pregnant patient.

5. The method according to claim 1, characterized in that, Determining the cumulative risk coefficient of the complication when drugs are combined includes: For the aforementioned complication, the risk factor for the corresponding drug is determined based on the current weight value of the risk that the complication may cause, the baseline risk factor of each drug that causes the complication, and the current physiological indicators of the pregnant patient. Based on the risk coefficients and drug action coefficients of each drug corresponding to the complication, the cumulative risk coefficient of the complication when drugs are combined is determined.

6. The method according to claim 1, characterized in that, The risk assessment score for evaluating the risk posed by this complication includes: Calculate the sum of the baseline risk coefficients of each drug that causes this complication; Based on the calculation results, the superimposed risk coefficient, the current weight value of the risk that the complication may cause, and the individualized values ​​corresponding to the current physiological indicators of the pregnant patient, the risk assessment score of the risk caused by the complication is evaluated.

7. The method according to claim 1, characterized in that, The method further includes: If the risk assessment score of any complication is greater than a set threshold, a medication alternative is generated; at least one medication in the medication alternative is different from any medication currently to be used by the pregnant patient.

8. The method according to claim 7, characterized in that, The generated medication alternatives include: For complications whose risk assessment scores exceed a set threshold, identify M drugs that could cause such complications from the medications currently being used by the pregnant patient. The medications in the M medications are replaced according to a replacement strategy from few to many. After each replacement, the steps for finding possible complications caused by the use of the medication by the pregnant patient are retrieved from the knowledge graph based on the medication currently to be used by the pregnant patient and the current stage of pregnancy. Replacement stops when the risk assessment score of the complication is less than the set threshold. The replaced medication has the same efficacy as the original medication, and the replaced medication is an available medication that the pregnant patient is allowed to use.

9. A device for assessing the risk of medication use during pregnancy, characterized in that, include: The search module is configured to search the knowledge graph for possible complications that may be caused by the use of the drug by the pregnant patient, based on the medication that the pregnant patient is currently using and the current stage of pregnancy of the pregnant patient. The first determining module is configured to, for the complication, determine the current weight value of the risk that the complication may cause based on the base weight set for the complication and the current gestational age of the pregnant patient; wherein, the more the current gestational age of the pregnant patient deviates from the set standard gestational age, the higher the current weight value of the risk that the complication may cause. The second determining module is configured to determine the superposition risk coefficient of the complication when the drugs are superimposed if the number of drugs to be used by the pregnant patient is greater than 1. The superposition risk coefficient is determined based on the risk coefficient of each drug causing the complication, wherein the risk coefficient of any drug causing the complication is determined based on the current weight value of the risk that the complication may cause, the basic risk coefficient of the drug, and the current physiological indicators of the pregnant patient. The assessment module is configured to assess the risk assessment score of the complication based on the superimposed risk coefficient, the current weight value of the risk that the complication may cause, the baseline risk coefficient of each drug that causes the complication, and the current physiological indicators of the pregnant patient.

10. An electronic device, characterized in that, include: Processor- and machine-readable storage media; The machine-readable storage medium stores machine-executable instructions that can be executed by the processor; The processor is configured to execute machine-executable instructions to implement the method of any one of claims 1-8.

Images