Intelligent assessment method and system for community disease-sharing home medication

Abstract

The invention provides an intelligent evaluation method and an intelligent evaluation system for community common-disease home medication, which are suitable for monitoring the home medication safety of old people with common diseases. The method comprises the following steps: constructing a dynamic knowledge graph including disease combination nodes, drug information nodes, individual feature nodes and association relationships thereof; long-term medication data and temporary medication data of a user are obtained through a multi-mode interaction interface integrating optical character recognition, voice recognition and natural language processing technologies; performing triple conflict analysis on the medication data based on the dynamic knowledge graph, wherein the triple conflict analysis comprises drug interaction prediction, disease taboo matching and individual feature cross validation; and generating high, medium and low drug use risk levels according to a conflict analysis result, and triggering an emergency notification protocol when the risk is high.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical management technology, and specifically to a method and system for intelligent home medication safety assessment suitable for elderly people with comorbidities in the community, and more particularly to a home medication safety assessment technology based on a dynamic knowledge graph, multimodal interaction technology, and a triple analysis mechanism for medication risks. Background Art

[0002] Elderly people living in the community with comorbidities (i.e., those with two or more chronic diseases) face the complex challenges of multiple illnesses, multiple medications, and long-term medication use. The safety of their medications at home has become a key challenge in health management. Research on medication management robots, smart medicine boxes, and mobile apps has been conducted abroad. These tools primarily assist doctors in prescribing, providing medication reminders, tracking symptoms, and sharing data, some of which use gamification to improve medication compliance. In China, three types of recommendation systems have emerged in the field of medication safety: intelligent interactive medication reminder management systems, remote doctor-patient resource matching systems, and personalized medication plan recommendation systems that assist doctors (such as those used in Traditional Chinese Medicine diagnosis and treatment).

[0003] However, the existing technology has the following shortcomings: First, there is a lack of data on temporary medication for elderly people with comorbidities. The elderly in my country have poor medication compliance and the phenomenon of purchasing over-the-counter drugs on their own is prominent. There is a lack of early warning mechanism for the interaction between chronic disease prescription drugs and temporary medications (such as painkillers, sleeping pills, health products, traditional Chinese medicine, etc.), and there are significant safety risks in home self-use of drugs; second, data updates are lagging behind, and the existing system relies on manual updates of the knowledge base. The launch of new drugs and the latest research results cannot be linked in real time, and the risk of new drug interactions may be missed; third, there is a gap in home medication monitoring. The existing system mainly serves medical professionals, is highly professional and has complex terminology. Elderly users find it difficult to use it independently due to operational difficulties, resulting in a lack of effective management of out-of-hospital medication safety.

[0004] Therefore, in order to solve the problem of temporary medication safety management for elderly people with comorbidities in the community at home, there is an urgent need for an intelligent assessment method and system that can cover temporary medication data, dynamically update knowledge, and be easy to operate. Summary of the Invention

[0005] To address the deficiencies of existing technologies, the present invention proposes an intelligent evaluation method and system for home medication for comorbidities in the community. The technical solutions adopted by the present invention include:

[0006] In a first aspect, the present invention provides a method for intelligently assessing medication use at home for comorbidities in a community, the method comprising:

[0007] construct a dynamic knowledge graph, the dynamic knowledge graph comprising disease combination nodes, drug information nodes, individual characteristic nodes, and association relationships therebetween, wherein the drug information nodes comprise drug category attributes and side effect attributes, and the individual characteristic nodes comprise age and liver and kidney function indicators;

[0008] obtain long-term medication data and temporary medication data of a user through a multi-modal interactive interface integrating optical character recognition, speech recognition, and natural language processing technology;

[0009] perform triple conflict analysis on the long-term medication data and the temporary medication data based on the dynamic knowledge graph, the triple conflict analysis comprising drug interaction prediction, disease contraindication matching, and individual characteristic cross-validation;

[0010] generate a medication risk level according to the triple conflict analysis result, and trigger an emergency notification protocol when a high-risk conflict occurs.

[0011] Further, the construction of the dynamic knowledge graph specifically comprises:

[0012] extract disease combination, drug information, and individual characteristic entity information from prescription data and drug instruction manuals, and create nodes respectively to form a triple structure centered on the drug information nodes;

[0013] establish a treatment association link between the drug information nodes and the disease combination nodes, and a metabolic risk association link between the drug information nodes and the individual characteristic nodes;

[0014] store the triple structure into a graph database, and configure a multi-hop path query interface.

[0015] Further, the obtaining of the long-term medication data and the temporary medication data of the user through the multi-modal interactive interface comprises:

[0016] in response to a medicine box scanning operation, extract drug text information through optical character recognition, and perform standardized mapping of the text information based on a natural language processing model;

[0017] or

[0018] in response to a voice instruction, generate text through a speech recognition model, and associate the text to a generic name of a preset drug name library through a natural language processing model;

[0019] when the standardized mapping or the generic name association fails, trigger a dynamic knowledge graph updating process.

[0020] Further, the triple conflict analysis comprises:

[0021] query the chemical component overlap and interaction relationship between the temporary medication and the long-term medication in the dynamic knowledge graph;

[0022] Verify the contraindication state of the user's existing disease for the temporary medication;

[0023] Evaluate the drug metabolism risk in combination with the user's age, liver and kidney function indicators;

[0024] Integrate and analyze the results through a weighted scoring model to output high, medium and low risk conclusions.

[0025] Further, the dynamic knowledge graph updating process comprises:

[0026] When the standardized mapping or generic name association fails, generate a drug missing instruction;

[0027] In response to the drug missing instruction, guide the user to upload a drug instruction image;

[0028] Extract the drug components, indications and contraindications key fields from the instruction through the BioBERT model to construct temporary triples;

[0029] Perform logical consistency detection of the temporary triples and the dynamic knowledge graph using the SAT algorithm;

[0030] If a conflicting knowledge entry is detected, locate the conflict path and calculate the weight through the LBX algorithm, and delete the knowledge entry with a lower weight;

[0031] If no conflict is detected, merge the conflict-free temporary triples into the dynamic knowledge graph, and record the timestamp and version snapshot.

[0032] Further, locating the conflict path and calculating the weight through the LBX algorithm, and deleting the knowledge entry with a lower weight comprises:

[0033] Generate the dynamic knowledge graph entity vector Et and the temporary triple entity vector Ed using the word2vec model, and locate the conflict association path through the similarity network;

[0034] Calculate the existing knowledge entry Wt and the temporary triple Wd based on a preset weight calculation formula, and delete the knowledge entry with a lower weight.

[0035] Optionally, the trigger emergency notification protocol comprises:

[0036] Generate structured warning information containing conflict types, pathological mechanisms and contraindicated drug combinations;

[0037] Push the warning information to the preset emergency contact terminal through an encrypted channel.

[0038] In a second aspect, the present application also provides a community co-disease home medication intelligent evaluation system applying the method of the first aspect.

[0039] The system comprises a knowledge graph construction module for establishing and maintaining a dynamic knowledge graph comprising disease combinations, drug information and individual characteristics in the form of triples;

[0040] A multi-modal interaction module integrating optical character recognition, speech recognition and natural language processing technology is used for long-term medication data and temporary medication data of the user;

[0041] A conflict analysis module is used for performing drug interaction prediction, disease contraindication matching and individual characteristic cross-validation;

[0042] A risk response module is used for generating medication risk levels and managing emergency notification protocols.

[0043] In a third aspect, the present application further provides an electronic device comprising a processor and a memory;

[0044] The memory is used for storing operation instructions;

[0045] The processor is used for calling the operation instructions, wherein the processor implements the steps of the method of the first aspect when executing the program.

[0046] In a fourth aspect, the present application further provides a computer readable storage medium, wherein the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the method of the first aspect.

[0047] The technical solution provided by the present application has the following beneficial effects:

[0048] The scheme provided by the present application effectively solves the problems of temporary medication data missing, knowledge updating lag and home monitoring in the existing comorbidity medication system by constructing a dynamic knowledge graph comprising disease combinations, drug information and individual characteristics, combining multi-modal interaction technology and a three-fold conflict analysis mechanism. The dynamic knowledge graph significantly improves the recognition accuracy of complex comorbidity medication and temporary medication conflicts by adding drug categories and side effect attributes. The dynamic updating mechanism based on active learning shortens the knowledge updating time and ensures the real-time inclusion of new drugs and the latest research results through incremental updating and conflict detection technology. The multi-modal interaction interface integrating OCR, speech recognition and natural language processing enables the elderly to complete medication data acquisition by only scanning the medicine box or voice input, thereby completely solving the high use threshold problem caused by complex professional terms in traditional systems. The combination of three-fold conflict analysis (drug interaction prediction, disease contraindication matching and individual characteristic cross-validation) and a weighted scoring model realizes the graded early warning of medication risks. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present application, the drawings needed to be used in the description of the present application will be briefly introduced as follows.

[0050] Figure 1 A community co-morbidity home medication intelligent assessment system provided by the present application is shown in the architectural design diagram.

[0051] Figure 2 A community co-morbidity home medication intelligent assessment method provided by the present application is shown in the flowchart.

[0052] Figure 3 A knowledge graph triple architecture design diagram provided by the present application is shown in the architectural design diagram.

[0053] Figure 4 A community co-morbidity home medication intelligent assessment system structure diagram provided by the present application is shown in the architectural design diagram.

[0054] Figure 5 A structural schematic diagram of an exemplary electronic device provided by the present application is shown in the structural schematic diagram. DETAILED DESCRIPTION

[0055] The present application provides a community co-morbidity home medication intelligent assessment method and system, effectively solving the problems of temporary medication data missing, knowledge update lag and home monitoring in the existing co-morbidity medication system, significantly improving the recognition accuracy of complex co-morbidity medication and temporary medication conflicts, ensuring the real-time inclusion of new drugs and the latest research results, and avoiding the high use threshold problem caused by the complexity of professional terms in traditional systems. The exemplary embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments of the present application, and it should be understood that the present application is not limited by the exemplary embodiments described herein.

[0056] SUMMARY

[0057] The community co-morbidity home medication intelligent assessment system (hereinafter referred to as "system") of the present application adopts a four-layer architecture design, including a device and data layer, a service layer, a business application layer and a service layer (as shown in Figure 1 The front end of the system is a WeChat mini program, suitable for community co-morbidity home scenarios for the elderly.

[0058] Device and data layer: Real community hospital prescription data (such as Beijing Fengtai District Community Health Service Center 2024 January-December hypertension, diabetes co-morbidity patient prescription data), drug instruction manual (obtained by crawling public databases such as China Medicine Query Platform, Yiyuan Network, etc.) as the core data source, clean, de-duplicate, standardize the data to ensure data quality. For scanned version of drug instruction manual pictures, Tesseract OCR engine is used to extract text information.

[0059] Service layer: integrate optical character recognition (OCR), speech recognition, natural language processing (NLP, such as ERNIE 3.0 model) technology, support users to obtain drug information through scanning medicine box or voice input, and complete drug name standardization mapping (solve the corresponding problem of trade name, generic name, alias).

[0060] Business application layer: based on dynamic knowledge graph (stored by Neo4j graph database), through the double-engine driving of "static knowledge graph conflict detection + dynamic knowledge expansion", realize the triple conflict analysis of drug interaction prediction, disease contraindication matching and individual characteristic cross verification.

[0061] Service layer: provide Web, WeChat applet, PAD and other multi-terminal access, support users to query risk results, receive emergency notifications, and integrate medical guidance functions (such as push the contact information of nearby top three hospitals endocrinology department).

[0062] In view of the above technical problems, the present application provides the following specific embodiments:

[0063] Example One

[0064] This embodiment takes a 72-year-old patient with Zhangmian in a community in Haidian District of Beijing (suffering from hypertension, diabetes, long-term taking amlodipine besylate tablets and metformin hydrochloride sustained-release tablets) as an example, as shown in the following table, the specific implementation process is as follows: Figure 2

[0065] S1. User basic information input

[0066] The user logs in to the WeChat applet and enters the

basic information

[0067] S2. Temporary medication data acquisition

[0068] ​The user self-purchases eszopiclone tablets (trade name: sulepride) due to insomnia, and scans the medicine box through the "scan" function of the applet. The system triggers the OCR module to extract the text information of the medicine box (including "eszopiclone tablets", "indication: insomnia", "specification: 1mg / tablet"), and confirms that the temporary medication generic name is "eszopiclone" after the NLP model standardization mapping.

[0069] S3. Triple conflict analysis

[0070] The analysis layer calls the dynamic knowledge graph (as shown in Figure 3 ), and performs the following three analyses in parallel:

[0071] Drug interaction prediction: query the knowledge graph for the chemical composition overlap and interaction relationship of eszopiclone and amlodipine, metformin; eszopiclone (benzodiazepine) has no direct interaction with amlodipine (calcium channel blocker), but when combined with metformin (biguanide), it may enhance central inhibition.

[0072] Disease contraindication matching: verify the contraindication status of eszopiclone for the user's existing diseases (hypertension, diabetes). The graph associates the disease combination node with the drug information node, and displays the contraindication results of eszopiclone for hypertension and diabetes.

[0073] Individual characteristics cross verification: combined with the user's age (72 years old) and liver and kidney function indicators, assess the drug metabolism risk.

[0074] Integrate the analysis results through a weighted scoring model (weight distribution: drug interaction 40%, disease contraindication 30%, individual characteristics 30%), and finally output the "medium risk" conclusion (score 65, threshold: low risk ≤50, medium risk 51-80, high risk ≥81), which suggests that "eszopiclone combined with metformin may enhance central inhibition, and it is recommended to reduce the dose of metformin or use eszopiclone for a short period of time, and closely observe symptoms such as dizziness and fatigue."

[0075] S4. Risk response and emergency notification

[0076] Because the analysis result this time is medium risk, the system displays the risk conclusion and suggestions on the applet interface; if the analysis result is high risk (such as the user mistakenly takes warfarin combined with aspirin), the system will generate structured warning information including conflict type (enhanced risk of bleeding), pathological mechanism (synergistic effect of anti-platelet and anticoagulant drugs), and contraindicated drug combination, and push it to the user's pre-set emergency contact through an encrypted channel (HTTPS + AES-256 encryption).

[0077] In summary, the method and system for generating a full-scene test case set based on a message log provided by the embodiments of the present application have the following technical effects:​

[0078] The scheme provided by the application solves the problems of temporary medication data missing, knowledge updating lag and family monitoring in the existing comorbidity medication system by constructing a dynamic knowledge graph containing disease combinations, drug information and individual characteristics, combining multi-modal interaction technology and a triple conflict analysis mechanism. The dynamic knowledge graph significantly improves the recognition accuracy of complex comorbidity medication and temporary medication conflicts by adding drug categories and side effect attributes. Meanwhile, the multi-modal interaction interface integrating OCR, speech recognition and natural language processing enables the elderly to complete medication data acquisition by simply scanning the medicine box or voice input, thereby completely solving the high use threshold problem caused by complex professional terms in traditional systems. The combination of triple conflict analysis and a weighted scoring model also realizes the graded early warning of medication risks.

[0079] Example Two

[0080] Based on the same inventive concept as the method of generating a full-scene test case set based on a message log in the foregoing embodiments, this embodiment takes the first access of a new hypoglycemic drug "semeglu-tide tablets" (not initially included in the knowledge graph) by the system as an example to illustrate the updating process of the dynamic knowledge graph:

[0081] S11. Triggering update conditions

[0082] After the user scans the semeglu-tide tablet medicine box and the OCR extracts the text information, the NLP model fails to associate to the drug name library (because the generic name is not matched), and the system generates a "drug missing instruction".

[0083] S12. Upload supplementary data

[0084] The user is guided to upload the instruction manual image of the semeglu-tide tablet drug (containing indications, contraindications, drug interactions and other key information) through the applet.

[0085] S13. Knowledge extraction and temporary triple construction

[0086] The BioBERT model is used to extract drug ingredients, indications and contraindications from the drug instruction manual, and perform entity recognition and relationship extraction to obtain temporary triples.

[0087] S14. Conflict detection and knowledge graph updating

[0088] S14.1 Logical conflict detection

[0089] SAT algorithm is called to detect the logical consistency of the temporary triples and the dynamic knowledge graph;

[0090] There is a knowledge entry in the existing dynamic knowledge graph that GLP-1 receptor agonists (such as dulaglutide, liraglutide, etc.) have no drug interaction with amlodipine (calcium channel blocker drugs), which is derived from the "Guidelines for the Management of Hypertension with Diabetes"; while the temporary triplets extracted from the drug instructions show that semaglutide tablets combined with amlodipine may enhance the antihypertensive effect and need to monitor blood pressure.

[0091] The SAT solver verifies the existence of contradictory words "no interaction" and "interaction", and determines that it is a logical conflict, triggering the conflict resolution process.

[0092] S14.2 Conflict resolution and graph update

[0093] The LBX (local optimization resolution) algorithm is used to perform semantic similarity calculation and weight evaluation. The cosine similarity between the existing entity Et (such as "liraglutide") and the temporary entity Ed (such as "semaglutide tablets") is calculated to be 0.92 (both belong to GLP-1 class and have similar structures) by the pre-trained word2vec model, and the semantic similarity of the associated entities (such as "amlodipine" and "calcium channel blockers") is calculated to be 0.98.

[0094] The similarity network locates the conflict association path, i.e. semaglutide tablets→GLP-1 class→drug interaction→calcium channel blockers→amlodipine.

[0095] The weight calculation formula is set as W = α·C + β·T + γ·E (α, β, γ are weight coefficients), and the final weight level is calculated according to the source credibility, time decay, and evidence level of the three elements:

[0096]

[0097] Since Wd (0.99) > Wt (0.67), the existing entry of GLP-1 receptor agonists and amlodipine without drug interaction is deleted, the temporary triplets are retained, and the resolution record is generated: the conflict reason is that the evidence level of the new drug instructions is higher than that of the old guidelines, and the clinical data of semaglutide is updated.

[0098] S14.3 Logical conflict detection

[0099] The lightweight inference engine Jena is called to perform rule verification on the updated local subgraph, and SPARQL queries are used to check whether there are contradictory interaction descriptions for the same drug pair.

[0100] The updated graph is verified to only retain the semaglutide-related interaction entries, ensuring that the resolved knowledge logic is self-consistent and conflict-free.

[0101] S15. Knowledge graph update and version management

[0102] Merge the conflict-free temporary triple into the existing knowledge graph, and record the update timestamp and version snapshot.

[0103] Example Three

[0104] Based on the same inventive concept as the community comorbidity home medication intelligent evaluation method in the foregoing embodiment, the application also provides a community comorbidity home medication intelligent evaluation system to apply the above method, and the structure is as shown in the drawing. Figure 4 The system comprises:

[0105] A knowledge graph construction module 401 is configured to establish and maintain a dynamic knowledge graph of triples containing disease combinations, drug information, and individual characteristics.

[0106] A multi-modal interaction module 402 is configured to integrate OCR, speech recognition, and NLP technologies to obtain long-term medication data and temporary medication data of a user.

[0107] A conflict analysis module 403 is configured to perform drug interaction prediction, disease contraindication matching, and individual characteristic cross-validation.

[0108] A risk response module 404 is configured to generate a medication risk level and manage an emergency notification protocol.

[0109] Further, the knowledge graph construction module 401 is specifically configured to:

[0110] Using prescription data and drug instructions as data sources, extract disease combination, drug information, and individual characteristic entity information, and create nodes respectively to form a triple structure centered on the drug information node.

[0111] Establish a treatment association link between the drug information node and the disease combination node, and a metabolic risk association link between the drug information node and the individual characteristic node.

[0112] Store the triple structure in a graph database and configure a multi-hop path query interface.

[0113] Further, the multi-modal interaction module 402 is specifically configured to:

[0114] In response to a medicine box scanning operation, extract drug text information through optical character recognition, and perform standardized mapping of the text information based on a natural language processing model.

[0115] Or

[0116] In response to a voice instruction, generate text through a speech recognition model, and associate the text to a generic name in a preset drug name library through a natural language processing model.

[0117] When the standardized mapping or generic name association fails, a dynamic knowledge graph update process is triggered.

[0118] Further, the conflict analysis module 403 is specifically configured to:

[0119] Query the chemical component overlap and interaction relationship between the temporary medication and the long-term medication in the dynamic knowledge graph;

[0120] Verify the contraindication state of the temporary medication on the user's existing diseases;

[0121] Combine the user's age and liver and kidney function indicators to evaluate the drug metabolism risk;

[0122] Integrate and analyze the results through a weighted scoring model to output high, medium, and low risk conclusions.

[0123] Further, the system further comprises a knowledge graph update module 405, and the knowledge graph update module 405 is specifically configured to:

[0124] When the standardized mapping or generic name association fails, generate a drug missing instruction;

[0125] In response to the drug missing instruction, guide the user to upload a drug instruction image;

[0126] Extract the drug components, indications, and contraindications key fields from the instruction through a BioBERT model to construct temporary triples;

[0127] Perform logical consistency detection of the temporary triples and the dynamic knowledge graph using a SAT algorithm;

[0128] If a conflicting knowledge entry is detected, locate the conflict path and calculate the weight through an LBX algorithm, and delete the knowledge entry with a lower weight;

[0129] If no conflict is detected, merge the non-conflicting temporary triples into the dynamic knowledge graph, and record the timestamp and version snapshot;

[0130] Further, the knowledge graph update module 405 is further configured to:

[0131] Generate the dynamic knowledge graph entity vector Et and the temporary triple entity vector Ed using a word2vec model, and locate the conflict association path through a similarity network;

[0132] Calculate the existing knowledge entry Wt and the temporary triple Wd based on a preset weight calculation formula, and delete the knowledge entry with a lower weight.

[0133] Further, the risk response module 404 is specifically configured to:

[0134] Generate structured early warning information containing conflict types, pathological mechanisms, and contraindicated drug combinations.

[0135] Push the early warning information to the preset emergency contact terminal through an encrypted channel.

[0136] Through the foregoing detailed description of the community co-morbidity home medication intelligent evaluation method, those skilled in the art can clearly understand the implementation method of the community co-morbidity home medication intelligent evaluation system in the embodiment. For the sake of brevity of the description, no further details are given.

[0137] Exemplary Electronic Device and Computer-Readable Storage Medium

[0138] Figure 5 The structural schematic diagram of the electronic device according to the embodiment of the application is shown.

[0139] The electronic device 500 includes a processor 501 and a memory 503. The processor 501 and the memory 503 are connected, for example, through a bus 502. Optionally, the electronic device 500 can also include a transceiver 504. It should be noted that the transceiver 504 is not limited to one in actual application. It can be understood that the structure shown in the embodiment of the application does not constitute a specific limitation on the specific structure of the electronic device 500. In other embodiments of the application, the electronic device 500 can include more or fewer components than those shown, or combine certain components, or split certain components, or different arrangement of components. The components shown can be implemented in hardware, software or a combination of software and hardware. Optionally, the electronic device can also include a display screen 505 for displaying images or receiving user operation instructions when necessary.

[0140] Among them, the processor 501 is applied to the embodiments of the application, and is used to implement the method shown in the method embodiment. The transceiver 504 can include a receiver and a transmitter, and the transceiver 504 is applied to the embodiments of the application, and is used to implement the function of communication between the electronic device and other devices in the embodiments of the application.

[0141] The processor 501 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 501 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0142] The processor 501 may also include one or more processing units. For example, the processor 501 may include an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural network processing unit (NPU). The different processing units may be independent devices or integrated into one or more processors. The controller may serve as the nerve center and command center of the electronic device 500. The controller may generate operation control signals based on instruction opcodes and timing signals to control instruction fetching and execution. The processor 501 may also include memory for storing instructions and data. In some embodiments, the memory in the processor 501 is a high-speed cache memory. This memory can store instructions or data that have just been used or are being recycled by the processor 501. If the processor 501 needs to use the instruction or data again, it can directly retrieve it from the memory. This avoids duplicate accesses, reduces the processor 501's waiting time, and thus improves system efficiency.

[0143] The processor 501 can run an intelligent evaluation method for home medication for comorbidities in the community provided in an embodiment of the present application. The processor 501 may include different devices. For example, when integrating a CPU and a GPU, the CPU and the GPU can cooperate to execute an intelligent evaluation method for home medication for comorbidities in the community provided in an embodiment of the present application. Part of the algorithm is executed by the CPU, and the other part of the algorithm is executed by the GPU to obtain faster processing efficiency.

[0144] The bus 502 can include a path that transmits information between the above-mentioned components. The bus 502 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 502 can be divided into an address bus, a data bus, a control bus, and the like. For convenience of representation, Figure 5 In the figure, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0145] The memory 503 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), and can include a high-speed random access memory, and can include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, a universal flash storage (UFS), or other optical disk storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, and the like), magnetic disk storage medium, or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0146] Optionally, the memory 503 is configured to store application program code for implementing the scheme of the present application, and the processor 501 is configured to control the execution. The processor 501 is configured to execute the application program code stored in the memory 503 to implement the community co-disease home medication intelligent evaluation method provided in any of the embodiments of the present application.

[0147] The memory 503 can be configured to store computer-executable program codes including instructions. The processor 501 performs various functions of the electronic device 500 and processes data by running the instructions stored in the memory 503. The memory 503 can include a program area and a data area. The program area can store an operating system, codes of application programs, and the like. The data area can store data created during use of the electronic device 500 (for example, images, videos, and the like captured by a camera application), and the like.

[0148] The memory 503 can further store one or more computer programs corresponding to the community comorbidity home medication intelligent evaluation method provided in the embodiments of the present application. The one or more computer programs are stored in the memory 503 and configured to be executed by the one or more processors 501, and the one or more computer programs include instructions, which can be used to perform each step in the above-mentioned corresponding embodiments.

[0149] Of course, the code of the community comorbidity home medication intelligent evaluation method provided in the embodiments of the present application can also be stored in an external memory.

[0150] The display screen 505 includes a display panel. The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), a Miniled, a MicroLed, a Micro-oLed, a quantum dot light emitting diode (QLED), or the like. In some embodiments, the electronic device 500 can include one or N display screens 505, where N is a positive integer greater than 1. The display screen 505 can be configured to display information input by a user or information provided to the user, as well as various graphical user interfaces (GUIs). For example, the display screen 505 can display photos, videos, webpages, or files, and the like.

[0151] The electronic device provided in the embodiments of the present application is applicable to any of the above-mentioned methods, and thus the beneficial effects achieved thereby can refer to the beneficial effects provided in the above-mentioned corresponding methods, which will not be described herein again.

[0152] The embodiment of the present application further provides a computer readable storage medium capable of realizing all steps in the method in the above-mentioned embodiment, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to realize all steps in the method in the above-mentioned embodiment.

[0153] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take a form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. In addition, the present application can take a form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0154] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the 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 a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure One one or more flows and / or blocks Figure One a system for implementing the functions specified in one or more flows and / or blocks.

[0155] These computer program instructions can also be stored in a computer readable memory capable of directing the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer readable memory produce a product including an instruction system, which implements the functions specified in the flowcharts and / or block diagrams. Figure One one or more flows and / or blocks Figure One a system for implementing the functions specified in one or more flows and / or blocks.

[0156] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure One one or more flows and / or blocks Figure OneThe present invention is described in detail below. ...

[0157] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention are intended to be covered by the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. An intelligent evaluation method for home medication for comorbidities in the community, characterized by: include: Constructing a dynamic knowledge graph, wherein the dynamic knowledge graph includes disease combination nodes, drug information nodes, individual feature nodes and their associations, wherein the drug information nodes include drug category attributes and side effect attributes, and the individual feature nodes include age and liver and kidney function indicators; Acquiring the user's long-term medication data and temporary medication data through a multimodal interactive interface, wherein the multimodal interactive interface integrates optical character recognition, speech recognition, and natural language processing technologies; Performing a triple conflict analysis on the long-term medication data and the temporary medication data based on the dynamic knowledge graph, wherein the triple conflict analysis includes drug interaction prediction, disease contraindication matching, and individual feature cross-validation; Generate medication risk levels based on triple conflict analysis results and trigger emergency notification protocols in the event of high-risk conflicts.

2. The method according to claim 1, characterized in that The construction of the dynamic knowledge graph specifically includes: Using prescription data and drug instructions as data sources, disease combinations, drug information, and individual feature entity information are extracted and nodes are created for each, forming a triple structure centered on the drug information node. Establishing a treatment association link between the drug information node and the disease combination node, and establishing a metabolic risk association link between the drug information node and the individual feature node; The triple structure is stored in a graph database, and a multi-hop path query interface is configured.

3. The method according to claim 1, characterized in that The method of obtaining the user's long-term medication data and temporary medication data through the multimodal interactive interface includes: In response to the medicine box scanning operation, the drug text information is extracted through optical character recognition and the text information is standardized and mapped based on the natural language processing model; or In response to voice commands, the speech recognition model generates text, and the natural language processing model associates it with the generic name in the preset drug name library; When the standardized mapping or common name association fails, the dynamic knowledge graph update process is triggered.

4. The method according to claim 1, wherein The triple conflict analysis includes: Querying the chemical composition overlap and interaction relationship between temporary medications and long-term medications in the dynamic knowledge graph; Verify the contraindications of temporary medication for the user's existing disease; Assess drug metabolism risk based on user age and liver and kidney function indicators; The analysis results are integrated through a weighted scoring model to output three risk levels: high, medium, and low.

5. The method according to claim 3, characterized in that The dynamic knowledge graph update process includes: Generate a drug missing instruction when standardized mapping or generic name association fails; In response to the drug missing instruction, guiding the user to upload an image of the drug instructions; The BioBERT model is used to extract key fields such as drug ingredients, indications, and contraindications from the instructions and construct temporary triples. Using the SAT algorithm to perform logical consistency detection between the temporary triple and the dynamic knowledge graph; If conflicting knowledge items are detected, the conflicting paths are located and weights are calculated using the LBX algorithm, and the knowledge items with lower weights are deleted. If no conflict is detected, the conflict-free temporary triples are merged into the dynamic knowledge graph, and the timestamp and version snapshot are recorded.

6. The method according to claim 5, characterized in that The LBX algorithm is used to locate conflicting paths and calculate weights. Knowledge items with lower weights are deleted, including: The word2vec model is used to generate the dynamic knowledge graph entity vector Et and the temporary triple entity vector Ed, and the conflicting association paths are located through the similarity network; The existing knowledge items Wt and temporary triples Wd are calculated based on the preset weight calculation formula, and the knowledge items with lower weights are deleted.

7. The method according to claim 1, characterized in that The triggering emergency notification protocol includes: Generate structured warning information including conflict type, pathological mechanism, and contraindicated drug combinations; The warning information is pushed to the preset emergency contact terminal through an encrypted channel.

8. An intelligent evaluation system for home medication for comorbidities in a community, applied to the method according to any one of claims 1 to 7, characterized in that: include: The knowledge graph construction module is used to build and maintain a dynamic knowledge graph of triples containing disease combinations, drug information, and individual characteristics; A multimodal interaction module that integrates optical character recognition, speech recognition, and natural language processing technologies for users' long-term and temporary medication data; Conflict analysis module, used to perform drug interaction prediction, disease contraindication matching, and individual feature cross-validation; Risk Response Module, which generates medication risk ratings and manages emergency notification protocols.

9. An electronic device, characterized in that: including processor and memory; The memory is used to store operation instructions; The processor is configured to execute the method according to any one of claims 1 to 7 by calling the operation instruction.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images