Medication management method and system serving home health management user
Through the analysis of HIS, user and drug information of home health management users, and the decision tree model and random forest algorithm are used to predict and process drug feedback information, the problems of decreased drug compliance and adverse drug reactions are solved, and the safety and management of drug use have been improved.
Patent Information
- Application Number
- CN202510407923.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-02
- Publication Date
- 2025-07-29
AI Technical Summary
Due to the lack of professional supervision, home health management users have decreased drug compliance, and the risk of recurrence of the disease and increased adverse drug reactions. The existing technology lacks an effective mechanism for synchronizing and feedback of drug information.
By obtaining the user's HIS information, user information and drug information, using pre-trained multiple decision tree models and random forest algorithms, predicting drug feedback information, and combining actual feedback information, correcting feature information, establishing a synchronization channel between health managers and users, and providing drug reminders and drug side effects comfort suggestions.
Improve drug compliance, promptly detect and deal with adverse drug reactions, reduce recurrence, ensure drug safety, and enhance drug management and medical safety.
Smart Images
Figure CN120388759A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of Internet technology for medical and health services. Specifically, it relates to a medication management method and system for users of home health management. Background Art
[0002] When users are managing their health at home, due to environmental changes and lack of professional supervision, they often experience a decline in medication compliance, resulting in recurrence of the disease, poor treatment effects, and even an increased risk of relapse.
[0003] Adverse drug reactions refer to harmful reactions that occur with qualified drugs under normal usage and dosage and are unrelated to the purpose of medication. These reactions vary due to factors such as drug type, usage method, and individual differences, causing adverse effects to users. When users are out of the medication monitoring of health managers, the side effects after their medication need to be promptly reported to the health managers so that the health managers can adjust the medication plan based on the user's feedback. Therefore, a channel for synchronizing the medication information between health managers and users needs to be established to ensure that both parties can obtain each other's medication information and feedback in real time.
[0004] With the deepening application of intelligent algorithms, artificial intelligence and machine learning algorithms will play a greater role in medication reminders. By analyzing a large amount of medical data of users, including medical history, genetic information, medication reactions, living habits, etc., it is possible to more accurately predict the reactions of users to different drugs and possible adverse drug reactions, so as to timely predict the adverse drug reactions of users' medication and provide intelligent medication guidance. Summary of the Invention
[0005] Aiming at the deficiencies in the prior art, the purpose of the present disclosure is to provide a medication management method for users of home health management.
[0006] To achieve the above purpose, according to one aspect of the present disclosure, a medication management method for users is provided, including:
[0007] Obtain the HIS information, user information, and drug information of the user;
[0008] Perform feature extraction processing on the HIS information, user information, and drug information of the user to determine the feature information, where the feature information includes the user's gender, age, drug, medical diagnosis, and side effects;
[0009] According to the feature information and multiple pre-trained decision tree models, use the random forest algorithm to determine the predicted medication feedback information of the user;
[0010] Obtain the actual medication feedback information of the user;
[0011] Determine the cause of the medication reaction based on the predicted medication feedback information and the actual medication feedback information of the user;
[0012] Modify the feature nodes of the feature information according to the cause of the medication reaction to determine new feature information.
[0013] Optionally, the method further includes:
[0014] Reach out to the user with medication reminders, which include medication reminders, discontinuation reminders, and reminder for refilling prescriptions for follow-up consultations. The ways of reaching out with medication reminders include text message reminders, voice reminders, and synchronized reminders to related relatives.
[0015] Optionally, the method further includes:
[0016] Send information on soothing suggestions for drug side effects to the user according to the user's actual medication feedback information and the side effect characteristics in the drug knowledge base.
[0017] Optionally, the method for determining the pre-trained multiple decision tree models includes:
[0018] Train a preset decision tree model with similar training samples to determine multiple decision tree models. The similar training samples include the feature information and adverse reaction information of verification samples similar to the user.
[0019] Perform verification and optimization processing on the multiple decision tree models with similar verification samples and the random forest algorithm to determine the pre-trained multiple decision tree models. The similar verification samples include the feature information and adverse reaction information of verification samples similar to the user.
[0020] Optionally, the step of determining the cause of the medication reaction based on the predicted medication feedback information and the actual medication feedback information of the user includes:
[0021] Compare the predicted medication feedback information and the actual medication feedback information of the user to determine error information;
[0022] Conduct artificial follow-up based on the error information to determine the cause of the medication reaction.
[0023] Optionally, the method further includes:
[0024] Iterate the new feature information into the pre-trained multiple decision tree models to determine the user's new predicted medication feedback information;
[0025] Determine a new cause of the medication reaction based on the user's new predicted medication feedback information and the user's actual medication feedback information;
[0026] According to the new cause of the medication reaction, modify the feature nodes of the feature information to determine the new feature information.
[0027] Optionally, the HIS information of the user includes electronic medical records, examination and inspection information, medication records, surgical records, adverse reaction records, and summaries;
[0028] The user information includes user basic information, medication history, and family history;
[0029] The drug information includes usage and dosage, contraindications, and side effects.
[0030] According to a second aspect of the present disclosure, there is provided a medication management system for home health management users, including:
[0031] A first acquisition module for acquiring the HIS information, user information, and drug information of the user;
[0032] A feature extraction module for performing feature extraction processing on the HIS information, basic information of the user, and drug information to determine feature information, where the feature information includes the gender, age, drug, medical diagnosis, and side effects of the user;
[0033] A medication feedback prediction module for determining the predicted medication feedback information of the user by using the random forest algorithm according to the feature information and multiple pre-trained decision tree models;
[0034] A second acquisition module for acquiring the actual medication feedback information of the user;
[0035] A medication reaction analysis module for determining the cause of the medication reaction according to the predicted medication feedback information and the actual medication feedback information of the user;
[0036] A medication correction module for correcting the feature information according to the cause of the medication reaction.
[0037] According to a third aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method provided in the first aspect of the present disclosure are implemented.
[0038] According to a fourth aspect of the present disclosure, there is provided an electronic device, including:
[0039] A memory, on which a computer program is stored;
[0040] A processor for executing the computer program in the memory to implement the steps of the method provided in the first aspect of the present disclosure.
[0041] Compared with the prior art, the embodiments of the present disclosure have at least one of the following beneficial effects:
[0042] Through the above technical solutions, by analyzing a large amount of medical data such as the user's HIS information, user information, and drug information, multiple pre-trained decision tree models are used to accurately predict the user's predicted medication feedback information, predict potential drug risks, and combine the user's actual medication feedback information to timely discover and handle drug adverse reactions and adjust the medication, reduce the recurrence of the user's condition due to adverse reactions, establish a channel for synchronizing the medication information of the primary diagnosis health manager and the user, effectively enhance the user's medication management, ensure the user's medication safety, and improve medical safety.
[0043] In the embodiments of the present disclosure, by reaching the user with medication reminders, the user's medication compliance is improved, the medication reminder channels are enriched, and the medication timeliness is improved.
[0044] In the embodiments of the present disclosure, based on the user's actual medication feedback information and the side effect characteristics in the drug knowledge base, information on soothing suggestions for drug side effects is sent to the user, providing intelligent medication guidance for predicting the adverse reactions of the user's medication and improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] By reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings, other features, objects, and advantages of the present disclosure will become more apparent:
[0046] Figure 1 is a flowchart of a medication management method for serving home health management users according to an exemplary embodiment.
[0047] Figure 2 is a schematic structural diagram of a medication feedback for serving home health management users according to an exemplary embodiment.
[0048] Figure 3 is a block diagram of a medication management system for serving home health management users according to an exemplary embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0049] The present disclosure will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present disclosure, but do not limit the present disclosure in any form. It should be noted that those of ordinary skill in the art can make several modifications and improvements without departing from the concept of the present disclosure. These all fall within the protection scope of the present disclosure.
[0050] Figure 1 is a flowchart of a medication management method for serving home health management users according to an exemplary embodiment. Figure 2It is a schematic diagram of an architecture for providing medication feedback to users of home health management according to an exemplary embodiment.
[0051] As Figure 1 , Figure 2 shown, the present disclosure provides a medication management method for users of home health management, including S11 to S16.
[0052] S11, obtain the user's HIS information, user information, and medication information.
[0053] The user's HIS information includes electronic medical records, examination and inspection information, medication records, surgical records, adverse reaction records, and summaries.
[0054] The user information includes user basic information, medication history, and family history. Among them, the user basic information may include information such as gender, age, height, weight, HIS information, etc.
[0055] The medication information includes usage and dosage, contraindications, and side effects. Among them, the medication information is prescription medication information.
[0056] S12, perform feature extraction processing on the user's HIS information, user information, and medication information to determine feature information.
[0057] Among them, the feature extraction processing can adopt the keyword extraction method or the named entity recognition (NER) method of natural language processing technology.
[0058] For simple information, the keyword extraction method can be adopted. With the assistance of a medical dictionary and glossary, keywords are determined, and features are extracted by identifying the keywords in the user's HIS information, user information, and medication information. Among them, the keywords are usually related to disease prevention, symptoms, and treatment methods.
[0059] For complex information, the named entity recognition (NER) method of natural language processing technology can be adopted to identify medical entities in the text of the user's HIS information, user information, and medication information, such as disease names, drug names, body parts, examination and inspection items, etc. Among them, the named entity recognition (NER) method of natural language processing technology determines the boundaries and categories of entities according to the context in the text through a trained model using machine learning or deep learning algorithms.
[0060] Among them, the feature information includes the user's gender, age, drugs, medical diagnosis, and side effects.
[0061] S13, according to the feature information and multiple pre-trained decision tree models, use the random forest algorithm to determine the user's predicted medication feedback information.
[0062] Among them, multiple pre-trained decision tree models are all trained using data of other users of the same type as the user to be predicted. When predicting the predicted medication feedback information of different users, the multiple pre-trained decision tree models need to be retrained.
[0063] S14. Obtain the actual medication feedback information of the user.
[0064] Among them, the user can be reached through multiple channels, such as online channels, offline channels, and phone and text message channels, to prompt the user to fill out a medication feedback questionnaire to obtain the actual medication feedback information of the user.
[0065] S15. Determine the cause of the medication reaction based on the predicted medication feedback information of the user and the actual medication feedback information of the user.
[0066] S16. Modify the feature nodes of the feature information according to the cause of the medication reaction to determine the new feature information.
[0067] Through the above technical solutions, by analyzing a large amount of medical data such as the user's HIS information, user information, and drug information, multiple pre-trained decision tree models are used to accurately predict the predicted medication feedback information of the user, predict potential drug risks, and combine the actual medication feedback information of the user to timely discover and handle adverse drug reactions and adjust the medication, reduce the recurrence of the user's condition due to adverse reactions, establish a channel for synchronizing the medication information between the primary diagnosis health manager and the user, effectively enhance the user's medication management, ensure the user's medication safety, and improve medical safety.
[0068] In a possible embodiment, the method for determining multiple pre-trained decision tree models includes S101 to S102.
[0069] S101. Train the preset decision tree model using the same-type training samples to determine multiple decision tree models.
[0070] Among them, the same-type training samples include the feature information and adverse reaction information of the training samples of the same type as the user.
[0071] S102. Perform verification and optimization processing on the multiple decision tree models using the same-type verification samples and the random forest algorithm to determine the multiple pre-trained decision tree models.
[0072] Among them, the same-type verification samples include the feature information and adverse reaction information of the samples of the same type as the user.
[0073] As an example, the homogeneous verification samples include the feature information and adverse reactions of the verification samples of the same type as the user. The homogeneous verification samples can be obtained from the hospital information system (HIS) of the hospital. The home health management data (including HIS information, user information, and drug information) of other users of the same type as the user is collected to form the homogeneous verification samples.
[0074] According to the decision tree construction algorithm (CART), a decision tree can be constructed. 70% of the samples can be used as training samples, and 30% as verification samples. The samples are used to train the preset decision tree model.
[0075] The homogeneous training samples are randomly split into 10 groups to train the preset decision tree model, forming 10 decision tree models.
[0076] The 10 decision tree models are verified and optimized using the homogeneous verification samples and the random forest algorithm. The performance of the 10 decision tree models is evaluated to determine the 10 pre-trained decision tree models.
[0077] In a possible embodiment, S13. According to the feature information and multiple pre-trained decision tree models, the random forest algorithm is used to determine the predicted medication feedback information of the user, which may include:
[0078] The feature information can be input into the multiple pre-trained decision tree models, and the random forest algorithm is used to analyze and predict the possible adverse reactions of the user's reuse of prescription drugs, forming the predicted medication feedback information of the user.
[0079] Among them, each decision tree model represents a decision tree. Each decision tree will output a classification prediction result (yes or no) on whether adverse reactions will occur or the severity of the adverse reactions (numerical prediction if the adverse reaction classification result is yes) for the input feature information. When splitting each node, a part of the features is randomly selected from all the feature information. For example, m features are selected from a total of p features, where m < p, and then the optimal feature and split point are selected from these m features for the growth of the tree. Different decision trees will be generated according to the different selected partial features.
[0080] Based on the classification prediction result (yes or no) of the adverse reactions output by the decision tree, random forest ensemble learning is performed based on the random forest algorithm, and the final prediction result is determined by voting. For example, if 60 out of 100 decision trees predict that adverse reactions will occur and 40 predict that adverse reactions will not occur, then the random forest predicts that the user will have adverse reactions, forming the predicted medication feedback information of the user.
[0081] In a possible embodiment, based on the predicted medication feedback information of the user determined by multiple pre-trained decision tree models, a list of adverse reactions of the current user taking prescription drugs is listed as the basis for the medication feedback questionnaire to form a medication feedback questionnaire.
[0082] In a possible embodiment, S14, obtaining the actual medication feedback information of the user may include:
[0083] Based on the above-formed medication feedback questionnaire, the user is reminded to fill out the medication feedback questionnaire through multiple channels to obtain the actual medication feedback information of the user.
[0084] Exemplarily, online channels such as official websites, social media, emails, mobile applications, and online forums or communities remind the user to fill out the medication feedback questionnaire; offline channels such as hospitals or clinics, pharmacies, user communication meetings or lectures remind the user to fill out the medication feedback questionnaire; phone and SMS channels such as phone calls and SMS links remind the user to fill out the medication feedback questionnaire; the medication feedback questionnaire can also be filled out through third-party health service platforms and user self-service terminals.
[0085] In a possible embodiment, S15, determining the cause of the medication reaction based on the predicted medication feedback information and the actual medication feedback information of the user may include S201 to S202.
[0086] S201, comparing the predicted medication feedback information and the actual medication feedback information of the user to determine the error information.
[0087] Among them, the error information is the information that is inconsistent between the actual medication feedback information and the predicted medication feedback information of the user.
[0088] S202, conducting manual follow-up based on the error information to determine the cause of the medication reaction.
[0089] Among them, the error information can be provided to the primary diagnosis health manager corresponding to the user, and the primary diagnosis health manager conducts manual follow-up and analyzes the cause of the error information, so as to determine and feedback the cause of the medication reaction.
[0090] In a possible embodiment, S16, modifying the feature nodes of the feature information according to the cause of the medication reaction to determine the new feature information may include:
[0091] Based on the cause of the medication reaction analyzed and determined by the primary diagnosis health manager, the feature nodes in the feature information are modified, which may include modifying the feature nodes, adding or reducing the feature nodes, that is, modifying, adding or reducing the feature values, such as modifying, adding, reducing the diagnosis and drug information, so as to optimize the medication advice for the same drug for different diseases and different users.
[0092] In a possible embodiment, a method for managing a user's medication may further include S17 to S18.
[0093] S17, iterating new feature information to a plurality of pre-trained decision tree models to determine new predicted medication feedback information of the user.
[0094] S18, determining a new cause of medication reaction based on the user's new predicted medication feedback information and the user's actual medication feedback information.
[0095] S19, modifying the feature nodes of the feature information according to the new cause of medication reaction to determine new feature information.
[0096] Repeat the above steps S17 to S19, continuously iterate the new feature information to a plurality of pre-trained decision tree models, thereby continuously modifying the feature information, optimizing the medication advice for the same drug for different diseases and different users, and improving the accuracy of the decision tree model in predicting adverse drug reactions.
[0097] In a possible embodiment, a method for managing a user's medication may further include S20.
[0098] S20, reaching the user with medication reminders.
[0099] Reaching the user with medication reminders includes medication reminders, discontinuation reminders, and reminder for follow-up consultation and prescribing.
[0100] The ways of reaching the user with medication reminders include SMS reminders, voice reminders, and synchronous reminders associated with relatives.
[0101] Among them, the communication devices or health devices of the user and the communication devices or health devices of the user's family members can be associated for data authorization, so as to execute reaching the user with medication reminders.
[0102] Exemplarily, by sending mobile phone SMS reminders, IVR intelligent voice reminders, WeChat message notifications, health watch / band vibration reminders, and synchronous reminders associated with relatives to the user, the user's medication compliance is improved, the channels for medication reminders are enriched, and the timeliness of medication is improved.
[0103] In a possible embodiment, a method for managing a user's medication may further include S21.
[0104] S21, sending information on soothing advice for drug side effects to the user according to the user's actual medication feedback information and the side effect characteristics in the drug knowledge base.
[0105] Among them, the actual medication feedback information of users can be collected according to a preset frequency, and associated with the side effect characteristics corresponding to the drugs in the drug knowledge base, automatically reminding users of the common side effects of the drugs used, and sending drug side effect soothing suggestion information, providing intelligent medication guidance for predicting the adverse reactions of users' medication, and improving the medical experience of users.
[0106] A medication management method for users of the present disclosure provides a medication feedback channel, thereby providing a channel for synchronizing the medication information between a health manager and users, and being able to adjust the medication in a timely manner according to the users' medication feedback; it also provides a medication reminder channel, timely reminding users of medication, discontinuation of medication, and matters related to prescribing medication for follow-up visits, enriching the medication reminder channel, improving the timeliness of medication, and improving the medication compliance of users.
[0107] A medication management method provided by the present disclosure can be applied to the scenarios of medication feedback and medication reminder for users' home health management. Based on the HIS information and subsection information collected by the hospital information system (HIS) of the hospital, according to the HIS data of similar users in the hospital, the adverse reactions of users are predicted through a decision tree model, effectively enhancing the medication management of users, discovering potential drug risks in advance, and being able to optimize the medication combination for users with multi-drug combination therapy by analyzing the adverse reactions of drug combinations in historical data, and timely discovering the adverse reactions of users' medication through an automated follow-up medication feedback questionnaire, reducing the recurrence of the disease caused by adverse reactions of users.
[0108] A medication management method provided by the present disclosure can be applied to a health management system and a drug recommendation system.
[0109] Figure 3 It is a block diagram of a medication management system serving users of home health management shown according to an exemplary embodiment.
[0110] Based on the same concept, the present disclosure further provides a medication management system 100 serving users of home health management, including: a first acquisition module 110, a feature extraction module 120, a medication feedback prediction module 130, a second acquisition module 140, a medication reaction analysis module 150, and a medication correction module 160.
[0111] The first acquisition module 110 is used to acquire the HIS information, user information, and drug information of users;
[0112] The feature extraction module 120 is used to perform feature extraction processing on the HIS information, basic information of users, and drug information of users to determine feature information, and the feature information includes the gender, age, drugs, medical diagnosis, and side effects of the users;
[0113] The medication feedback prediction module 130 is used to determine the predicted medication feedback information of the user according to the feature information and multiple pre-trained decision tree models by using the random forest algorithm;
[0114] The second acquisition module 140 is used to acquire the actual medication feedback information of the user;
[0115] The medication reaction analysis module 150 is used to determine the cause of the medication reaction according to the predicted medication feedback information of the user and the actual medication feedback information of the user;
[0116] The medication correction module 160 is used to correct the feature information according to the cause of the medication reaction.
[0117] Through the above technical solution, by analyzing a large amount of medical data such as the user's HIS information, user information, and drug information, multiple pre-trained decision tree models are used to accurately predict the predicted medication feedback information of the user, predict potential drug risks, and combine the actual medication feedback information of the user to timely discover and handle adverse drug reactions and adjust the medication, reduce the recurrence of the user's condition due to adverse reactions, establish a channel for synchronizing the medication information between the primary diagnosis health manager and the user, effectively enhance the user's medication management, ensure the user's medication safety, and improve medical safety.
[0118] Optionally, a user's medication management system 100 further includes:
[0119] The medication reminder module is used to reach the user with medication reminders. The medication reminder reach includes medication reminders, discontinuation reminders, and follow-up visit prescription reminders. The ways of reaching the medication reminder include text message reminders, voice reminders, and associated relative synchronization reminders.
[0120] Optionally, a user's medication management system 100 further includes:
[0121] The side effect soothing advice module: is used to send drug side effect soothing advice information to the user according to the actual medication feedback information of the user and the side effect characteristics in the drug knowledge base.
[0122] Optionally, a user's medication management system 100 further includes:
[0123] The decision tree model pre-training module is used to train the preset decision tree model with similar training samples to determine multiple decision tree models. The similar training samples include the feature information and adverse reaction information of the training samples similar to the user; use similar verification samples and the random forest algorithm to perform verification and optimization processing on the multiple decision tree models to determine the multiple pre-trained decision tree models. The verification samples include the feature information and adverse reaction information of the verification samples similar to the user.
[0124] Optionally, the medication response analysis module 150 includes:
[0125] A comparison sub-module for comparing the predicted medication feedback information of the user and the actual medication feedback information of the user to determine error information;
[0126] An artificial follow-up sub-module for performing artificial follow-up based on the error information to determine the cause of the medication response.
[0127] Optionally, the medication feedback prediction module 130 is further configured to iterate the new feature information to a plurality of pre-trained decision tree models to determine the new predicted medication feedback information of the user;
[0128] The medication response analysis module 150 is further configured to determine the new cause of the medication response according to the new predicted medication feedback information of the user and the actual medication feedback information of the user;
[0129] The medication correction module 160 is further configured to correct the feature nodes of the feature information according to the new cause of the medication response to determine the new feature information.
[0130] Optionally, the HIS information of the user includes electronic medical records, inspection and examination information, medication records, surgical records, adverse reaction records, and summaries; the user information includes user basic information, medication history, and family history; the drug information includes usage and dosage, contraindications, and side effects.
[0131] Regarding the embodiments of the above system, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated herein.
[0132] Based on the same inventive concept, in another embodiment of the present disclosure, an electronic device is further provided, including a memory, a processor, and a computer program stored on the memory and capable of running on the processor, where the processor is configured to execute a medication management method for serving users of home health management when executing the program.
[0133] Optionally, a memory for storing programs; the memory may include volatile memory (e.g., random-access memory, such as static random-access memory (SRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDR SDRAM), etc.); the memory may also include non-volatile memory, such as flash memory. The memory is used to store computer programs (such as application programs and functional modules for implementing the above methods), computer instructions, etc. The above computer programs, computer instructions, etc. can be partitioned and stored in one or more memories. And the above computer programs, computer instructions, data, etc. can be called by the processor.
[0134] The above computer programs, computer instructions, etc. can be partitioned and stored in one or more memories. And the above computer programs, computer instructions, data, etc. can be called by the processor.
[0135] A processor for executing the computer programs stored in the memory to implement the steps in the methods related to the above embodiments. For specific details, reference can be made to the relevant descriptions in the previous method embodiments.
[0136] The processor and the memory can be of independent structure or integrated into an integrated structure. When the processor and the memory are of independent structure, the memory and the processor can be coupled and connected through a bus.
[0137] In the embodiments of the present disclosure, a non-transitory computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of a method for medication management serving users of home health management in any of the above embodiments.
[0138] Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.
[0139] This disclosure is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the disclosure. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce means for implementing the functions specified in the process Figure 1 one process or multiple processes and / or blocks Figure 1 means for implementing the functions specified in one block or multiple blocks.
[0140] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in the process Figure 1 one process or multiple processes and / or blocks Figure 1 means for implementing the functions specified in one block or multiple blocks.
[0141] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the process Figure 1 one process or multiple processes and / or blocks Figure 1 means for implementing the functions specified in one block or multiple blocks.
[0142] Although the preferred embodiments of the present disclosure have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the present disclosure.
[0143] Obviously, those skilled in the art can make various changes and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalent technologies, the present disclosure is also intended to include these modifications and variations.
Claims
1. A medication management method for home health management users, characterized in that, Including: Obtain the user's HIS information, user information, and drug information; Perform feature extraction processing on the user's HIS information, the user information, and the drug information to determine feature information, where the feature information includes the user's gender, age, drug, medical diagnosis, and side effects; According to the feature information and multiple pre-trained decision tree models, use the random forest algorithm to determine the predicted medication feedback information of the user; Obtain the actual medication feedback information of the user; According to the predicted medication feedback information of the user and the actual medication feedback information of the user, determine the reason for the medication reaction; According to the reason for the medication reaction, correct the feature nodes of the feature information to determine new feature information.
2. The method according to claim 1, characterized in that, The method further includes: Send medication reminders to the user, where the medication reminders include medication reminders, discontinuation reminders, and follow-up prescription reminders, and the ways of sending the medication reminders include text message reminders, voice reminders, and associated relative synchronization reminders.
3. The method according to claim 1, characterized in that, The method further includes: According to the actual medication feedback information of the user and the side effect characteristics in the drug knowledge base, send drug side effect soothing advice information to the user.
4. The method according to claim 1, wherein The method for determining the multiple pre-trained decision tree models includes: Use the same type of training samples to train a preset decision tree model to determine multiple decision tree models, where the same type of training samples include the feature information and adverse reaction information of the training samples of the same type as the user; Use the same type of verification samples and the random forest algorithm to perform verification and optimization processing on the multiple decision tree models to determine the multiple pre-trained decision tree models, where the same type of verification samples include the feature information and adverse reaction information of the verification samples of the same type as the user.
5. The method according to claim 1, wherein The step of determining the reason for the medication reaction according to the predicted medication feedback information of the user and the actual medication feedback information of the user includes: Compare the predicted medication feedback information of the user with the actual medication feedback information of the user to determine error information; Perform manual follow-up according to the error information to determine the reason for the medication reaction.
6. The method according to claim 1, characterized in that The method further includes: Iterate the new feature information to the multiple pre-trained decision tree models to determine the new predicted medication feedback information of the user; According to the new predicted medication feedback information of the user and the actual medication feedback information of the user, determine the new reason for the medication reaction; According to the new reason for the medication reaction, correct the feature nodes of the feature information to determine new feature information.
7. The method according to claim 1, wherein The user's HIS information includes electronic medical records, inspection and examination information, medication records, surgical records, adverse reaction records, and summaries; The user information includes user basic information, medication history, and family history; The drug information includes usage and dosage, contraindications, and side effects.
8. A medication management system for home health management users, characterized in that, Including: The first acquisition module is used to obtain the user's HIS information, user information, and drug information; The feature extraction module is used to perform feature extraction processing on the user's HIS information, the user's basic information, and the drug information to determine feature information, where the feature information includes the user's gender, age, drug, medical diagnosis, and side effects; A medication feedback prediction module, configured to determine the predicted medication feedback information of the user by using a random forest algorithm based on the feature information and a plurality of pre-trained decision tree models; A second acquisition module, configured to acquire the actual medication feedback information of the user; A medication reaction analysis module, configured to determine the cause of the medication reaction according to the predicted medication feedback information of the user and the actual medication feedback information of the user; A medication correction module, configured to correct the feature information according to the cause of the medication reaction; 9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, the steps of the method according to any one of claims 1-7 are implemented.
10. An electronic device, characterized in that, Comprising: A memory, on which a computer program is stored; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1-7.