Untoward effect monitoring method and device and electronic equipment

Abstract

The invention relates to the technical field of medical data analysis, in particular to an adverse reaction monitoring method and device and electronic equipment, and the method comprises the steps: obtaining data information related to an adverse reaction record, and building a knowledge graph between a drug and an adverse reaction event by adopting an ontology building method based on the data information; acquiring physiological function data and symptom description voice of a target patient; based on the knowledge graph, the physiological function data and the symptom description voice, the adverse reaction of the target patient is monitored, a monitoring result is obtained, based on the monitoring result and the knowledge graph, a judgment basis is provided for diagnosis of medical staff, through the knowledge graph, the adverse reaction of the medicine can be efficiently monitored, and according to the patient information, the diagnosis accuracy is improved. Real-time alarm and medication guidance can be realized, and doctors can be helped to make more scientific decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data analysis, and in particular to a method, device and electronic equipment for monitoring adverse reactions. Background Art

[0002] With the improvement of modern medical level, drugs are more and more widely used in the treatment of diseases. However, the adverse reactions of drugs are also increasing. How to effectively monitor and manage these adverse reactions has become an important issue in the medical field. At present, the traditional monitoring method relies on manual recording and reporting, which is inefficient and easy to miss. Moreover, new adverse reactions cannot be identified in a timely and effective manner.

[0003] Therefore, how to efficiently monitor the adverse reactions of drugs is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above problems, the present invention provides a method, device and electronic device for monitoring adverse reactions that overcome the above problems or at least partially solve the above problems.

[0005] In a first aspect, the present invention provides a method for monitoring adverse reactions, comprising: Obtain data information related to the record of adverse reactions; Based on the data information, an ontology construction method is used to establish a knowledge graph between drugs and adverse reaction events; Acquire physiological function data and symptom description voice of target patients; Based on the knowledge graph, the physiological function data and the symptom description voice, the adverse reactions of the target patient are monitored to obtain monitoring results; Based on the monitoring results and the knowledge graph, a judgment basis is provided for the diagnosis of medical staff.

[0006] Preferably, the obtaining of data information related to the record of adverse reactions includes: Obtain data and information related to adverse reactions from drug instructions, academic papers, electronic medical records, and adverse event reports.

[0007] Preferably, based on the data information, an ontology construction method is used to establish a knowledge graph between drugs and adverse reaction events, including: Based on the data information, determine patient characteristics, drug names, adverse events, symptoms, and diseases; With patient characteristics, drug names, drug ingredients, adverse events, symptoms, and diseases as entity nodes, and with the first pointing relationship between patient characteristics and drug names, the second pointing relationship between patient characteristics and diseases, the third pointing relationship between drug names and adverse events, the fourth pointing relationship between adverse events and symptoms, the fifth pointing relationship between drug names and diseases, and the sixth pointing relationship between drug names and drug ingredients, the connection relationship between entity nodes is established to construct a knowledge graph between drugs and adverse events.

[0008] Preferably, based on the knowledge graph, physiological function data and symptom description voice, the adverse reactions of the target patient are monitored to obtain monitoring results, including: Text parsing, feature extraction, and semantic analysis are used on the target patient’s symptom description speech to determine the target symptoms; The target symptoms and the physiological function data are compared with each entity node in the knowledge graph, and it is determined whether there is a preset pointing relationship with the adverse reaction event to obtain the monitoring result.

[0009] Preferably, based on the knowledge graph, physiological function data and symptom description voice, the adverse reactions of the target patient are monitored to obtain monitoring results, including: Obtaining an adverse reaction event prediction model, wherein the adverse reaction event prediction model is determined based on historical adverse reaction events, and the historical adverse reaction events are determined based on the knowledge graph; Based on the adverse reaction event prediction model, physiological function data and symptom description voice, the adverse reactions of the target patients are monitored to obtain monitoring results.

[0010] Preferably, after monitoring the adverse reactions of the target patient based on the adverse reaction event prediction model, physiological function data and symptom description voice and obtaining the monitoring results, the method further includes: Based on the monitoring results, determine new adverse events and / or new drugs; Based on the newly added adverse reaction events and / or newly added drugs, the knowledge graph is optimized and updated to obtain a new knowledge graph.

[0011] Preferably, based on the newly added adverse reaction events and / or newly added drugs, the knowledge graph is optimized and updated to obtain a new knowledge graph, including: Based on the newly added adverse reaction events and / or newly added drugs, determining target entity nodes related to the newly added adverse reaction events and / or newly added drugs; The newly added adverse reaction events and / or newly added drugs are respectively linked to corresponding target entity nodes to obtain a new knowledge graph.

[0012] Preferably, based on the monitoring results and the knowledge graph, a judgment basis is provided for the diagnosis of medical staff, including: Based on the new knowledge graph, the characteristics of target patients with the same adverse reaction events are analyzed to provide a basis for diagnosis by medical staff.

[0013] In a second aspect, the present invention further provides a device for monitoring adverse reactions, comprising: A first acquisition module is used to acquire data information related to the record of adverse reactions; Establishing a module for establishing a knowledge graph between drugs and adverse events based on the data information by adopting an ontology construction method; The second acquisition module is used to acquire the physiological function data and symptom description voice of the target patient; An obtaining module is used to monitor adverse reactions of target patients based on the knowledge graph, the physiological function data and the symptom description voice to obtain monitoring results; A module is provided for providing a judgment basis for the diagnosis of medical staff based on the monitoring results and the knowledge graph.

[0014] In a third aspect, the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0016] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages: The present invention provides a method for monitoring adverse reactions, including: acquiring data information related to records of adverse reactions, and based on the data information, using an ontology construction method to establish a knowledge graph between drugs and adverse reaction events; acquiring physiological function data and symptom description speech of target patients; based on the knowledge graph, physiological function data and symptom description speech, monitoring the adverse reactions of target patients to obtain monitoring results, and providing a judgment basis for the diagnosis of medical staff based on the monitoring results and the knowledge graph, so as to efficiently monitor the adverse reactions of drugs through the knowledge graph, and provide real-time alarms and medication guidance based on patient information, to help doctors make more scientific decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Also, throughout the accompanying drawings, the same reference figures are used to represent the same components. In the drawings: Figure 1 A schematic diagram showing the steps of the method for monitoring adverse reactions in an embodiment of the present invention is shown; Figure 2 A schematic diagram showing the structure of a monitoring device for adverse reactions in an embodiment of the present invention is shown; Figure 3 A schematic diagram of the structure of a computer device for implementing a method for monitoring adverse reactions in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0018] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0019] Embodiment 1: The embodiment of the present invention provides a method for monitoring adverse reactions, such as Figure 1 As shown, including: S101, obtaining data information related to the record of adverse reactions; S102, based on data information, using ontology construction method to establish a knowledge graph between drugs and adverse events; S103, obtaining physiological function data and symptom description voice of the target patient; S104, monitoring the adverse reactions of the target patients based on the knowledge graph, physiological function data and symptom description voice, and obtaining monitoring results; S105, based on the monitoring results and knowledge graph, provides a basis for medical staff's diagnosis.

[0020] First, S101, data information related to the record of adverse reactions is obtained, specifically through API interfaces, web crawlers, manual entry, etc. Specifically, data information related to adverse reactions can be obtained from drug instructions, academic papers, electronic medical records, and adverse event reports.

[0021] Among them, for clinical data: By integrating with the electronic medical record systems of hospitals or medical institutions, medication information and adverse records of patients are obtained through APIs or database interfaces. Data on drug use and related reactions in laboratory reports are collected through interfaces. Moreover, the database of the National Adverse Drug Reaction Monitoring Center is also integrated to obtain adverse drug reaction reports in real time.

[0022] For drug information: By integrating drug databases through API interfaces, basic information about drugs is obtained, including drug names, dosages, indications, adverse reactions, and so on.

[0023] For literature and reports: The Scrapy or Beautifulsoup libraries in Python are used to detect relevant research reports, pathological analyses, etc. from public medical literature databases, and to detect and extract drug and adverse reaction information from them.

[0024] After obtaining these data and information, the data is cleaned and preprocessed. The data and information are subjected to data cleaning, duplicate removal, and normalization to ensure data quality.

[0025] Specific cleaning methods include cleaning with regular expressions, stop word removal, and stemming, etc.

[0026] Among them, regular expressions are specifically used to extract data in specific formats, such as drug dosage data, administration timing data, and so on. Stop word removal is specifically to remove words that frequently appear in the text but have little impact on the text theme and meaning, such as "of", "in", etc. These words are often considered stop words. The existence of stop words will increase the complexity of text processing, and removing them will simplify the text and improve the efficiency of text processing. Stemming is specifically to simplify the inflected forms of words into a so-called "stem" or root form, which is achieved by deleting the suffixes of words, so as to group words with the same stem together and simplify the process of text analysis and information retrieval.

[0027] Next, execute S102. Based on the data and information, a knowledge graph between drugs and adverse reaction events is established using the ontology construction method.

[0028] Ontology construction mainly includes: obtaining information sources, concept classification and semantic relationship sorting, formal encoding, ontology evaluation, and reasoning verification. Specifically, a series of principles are followed to ensure that the constructed ontology has high efficiency and adaptability.

[0029] Specifically, based on the data information, determine patient characteristics, drug names, drug ingredients, adverse events, symptoms and diseases; take patient characteristics, drug names, adverse events, symptoms and diseases as entity nodes, and use the first pointing relationship between patient characteristics and drug names, the second pointing relationship between patient characteristics and diseases, the third pointing relationship between drug names and adverse reactions, the fourth pointing relationship between adverse events and symptoms, the fifth pointing relationship between drug names and diseases, and the sixth pointing relationship between drug names and drug ingredients to establish connection relationships between entity nodes and construct a knowledge graph between drugs and adverse events.

[0030] When constructing a knowledge graph, it is necessary to first extract entity node information related to adverse reactions. The extraction process here can be specifically achieved by training a classification model. First, through a large number of historical adverse reaction data files, the entities related to adverse reactions are labeled to train the classification model. Then, the classification model is applied, so that the entity node information related to adverse reactions in the data information can be obtained through the classification model.

[0031] In the present invention, patient characteristics, drug names, drug ingredients, adverse events, symptoms and disease entity nodes are used. Among them, the patient characteristics can specifically be the patient's age, gender, preferences and other characteristics. Adverse events are allergic reactions, toxic reactions, dose-related reactions, gastrointestinal reactions, nervous system symptoms, cardiovascular symptoms, blood system abnormalities, drug dependence reactions, antagonistic reactions, drug interactions, cumulative reactions, drug transport reactions, pseudoreactions, etc. Symptoms include: abnormal physiological function data, such as increased blood pressure, increased blood sugar, etc., and various perceptual states that cause discomfort to patients, such as headaches, dizziness, blurred vision, etc. The disease can be the disease that the patient originally suffered from, or it can be a new disease generated after taking the drug. The drug name is the name of the drug that often appears on the market, and the corresponding drug ingredient is the ingredient contained in the drug. Of course, different drug names will also have the same drug ingredient.

[0032] Next, establish connections between entities, such as the first pointing relationship between patient characteristics and drug names, for example, patients with a certain disease, the corresponding relationship between the drugs they take and the contraindications of taking drugs. The second pointing relationship between patient characteristics and diseases, the relationship between patients with some hobbies, such as smoking, drinking, etc., and heart and lung diseases, etc. The third pointing relationship between drug names and adverse events, for example, the existing drugs clearly have certain adverse events, and the connection relationship can be directly marked. The fourth pointing relationship between adverse events and symptoms, for example, the symptoms corresponding to allergic reactions may be rashes, itching, etc., establish a connection relationship between allergic reactions and corresponding symptoms, and mark the relationship type. The fifth pointing relationship between drug names and diseases, for example, long-term use of certain drugs may affect certain parts of the body, and the diseases involved are associated, and the sixth pointing relationship between drug names and drug ingredients, for example, the drug ingredients included in certain drugs have a subordinate relationship, and of course, drug ingredients also have the same type or mutual restraint relationship, etc. Based on the above relationship, the connection relationship between entity nodes is established to construct a knowledge graph between drugs and adverse events.

[0033] After obtaining the knowledge graph, execute S103 to obtain the physiological function data and symptom description voice of the target patient.

[0034] This step is to collect the target patient's physiological function data and symptom description voice in real time. The physiological function data can be monitored by a blood pressure meter, a blood glucose meter, or other types of monitoring equipment. The symptom description voice can be collected by a voice collection module.

[0035] Next, execute S104 to monitor the adverse reactions of the target patient based on the knowledge graph, physiological function data and symptom description voice to obtain the monitoring results.

[0036] Specifically, text parsing, feature extraction and semantic analysis are used on the target patient's symptom description voice to determine the target symptoms; the target symptoms and physiological function data are compared with each entity node in the knowledge graph, and it is determined whether there is a preset pointing relationship with the adverse reaction event to obtain the monitoring results.

[0037] In the process of obtaining monitoring results, the currently collected target patient's symptom description voice and physiological function data can be compared with each entity node in the knowledge graph to make a judgment. This kind of monitoring is to find out the existing bad records.

[0038] Specifically, for symptom description speech, the speech content can be generated into text information, and the physiological function data and the text information can be integrated into the electronic medical record. Then, text parsing is performed, including: using technologies such as word segmentation, part-of-speech tagging, and named entity recognition to perform preliminary parsing of the content in the text information to obtain parsing results. Then, based on the parsing results, the text features are extracted using technologies such as term frequency-inverse document frequency (TF-IDF) and word vectors (Word2Vec, GloVe). The extracted text features are subjected to semantic analysis. Specifically, the long short-term memory model (LSTM) is used for semantic recognition, so as to identify the target symptoms related to the adverse reactions of the drug.

[0039] Finally, the obtained target symptoms and physiological function data are compared with each entity node in the knowledge graph to obtain the adverse reaction results related to these entities.

[0040] In the specific comparison process, the graph traversal algorithm of the graph database (such as breadth-first search and depth-first search) can be used to analyze the association between drugs and adverse reactions and discover potential adverse reaction patterns. It is also possible to use the inference engine to automatically deduce the possible adverse drug reaction risks based on the existing data.

[0041] There are two types of monitoring results mentioned above, one is that the corresponding adverse reaction result is found, and the other is that the corresponding adverse reaction result is not found.

[0042] Among them, in the case where no corresponding adverse reaction results are found, further judgment can be made in the following manner.

[0043] Obtain an adverse reaction event prediction model, which is determined based on historical adverse reaction events, which are determined based on a knowledge graph; monitor the adverse reactions of target patients based on the adverse reaction event prediction model, physiological function data, and symptom description voice to obtain monitoring results.

[0044] First, obtain cases of drugs and adverse reaction events as historical adverse reaction events, where the adverse reaction events of the drug are determined by the content of the knowledge graph. Divide the historical adverse reaction events into training sets and test sets in a ratio of 8:2, and use any of the following models for model training: decision tree, random forest, support vector machine, neural network. Then, use linear regression or logistic regression methods to predict the probability of adverse drug reactions, thereby obtaining an adverse reaction event prediction model.

[0045] Finally, machine learning libraries such as scikit-learn and XGBoost were used to train the adverse event prediction model, and its hyperparameters were tuned to improve the prediction accuracy of the model.

[0046] Next, the physiological function data and voice symptom data of the target patient collected in S103 are input into the optimized adverse reaction event prediction model to obtain the monitoring result, that is, the adverse reaction result of the target patient.

[0047] This can identify adverse reactions of patients that are not recorded in the knowledge graph. Model prediction can help discover common adverse reaction combinations and potential high-risk drugs.

[0048] In addition, based on the adverse reaction event prediction model, physiological function data and symptom description voice, the adverse reactions of the target patients are monitored. After obtaining the monitoring results, it also includes: Based on the monitoring results, the knowledge graph is optimized and updated to obtain a new knowledge graph.

[0049] The optimization and updating process is to determine the target entity nodes related to the newly added adverse reaction events and / or the newly added drugs based on the newly added adverse reaction events and / or the newly added drugs; Link the newly added adverse reaction events and / or newly added drugs to the corresponding target entity nodes to obtain a new knowledge graph.

[0050] For adverse reaction records that do not exist in the original knowledge graph, the adverse reaction event prediction model can be used to predict them, thereby obtaining monitoring results. There are three types of monitoring results: The first one is to determine that the new drug has adverse reactions, which are recorded in the knowledge graph, but the new drug is not recorded in the original knowledge graph; The second type is to determine that the drug has a new adverse reaction, which is not recorded in the original knowledge graph, while the drug is already recorded in the knowledge graph; The third type is to determine that a new drug has new adverse reactions, and the new drug and the new adverse reactions are not recorded in the original knowledge graph.

[0051] Based on the newly added adverse reaction events and / or the newly added drugs, target entity nodes related to the newly added adverse reaction events and / or the newly added drugs are determined.

[0052] For the first and second cases, the relevant target entity nodes can be clearly obtained. For the third case, an associated query can be performed through patient characteristics, symptoms, diseases, and drug ingredients to obtain the target entity node with the highest relevance.

[0053] Next, the target entity points corresponding to the above three situations are linked to the newly added adverse reaction events and / or new drugs. Of course, the link relationship also needs to be labeled, so as to achieve the purpose of optimizing and updating the knowledge graph and obtain a new knowledge graph.

[0054] Finally, execute S105 to provide a judgment basis for the diagnosis of medical staff based on the monitoring results and the knowledge graph.

[0055] Specifically, based on the new knowledge graph, the characteristics of target patients with the same adverse reaction events are analyzed to provide a basis for medical staff's diagnosis.

[0056] Through the new knowledge graph or the original knowledge graph, the characteristics of the target patients with the same adverse reaction events can be analyzed. By summarizing and generalizing the rules with universal significance, it can provide more intuitive judgment basis for medical staff, so that medical staff can provide diagnosis and prompts for users with the same patient characteristics based on the characteristics of the target patients, and then provide real-time alarms and medication guidance. The safety of drug use is improved.

[0057] For example, it was found that patients with the same adverse reaction event all had heart disease, etc.

[0058] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages: The present invention provides a method for monitoring adverse reactions, including: acquiring data information related to records of adverse reactions, and based on the data information, using an ontology construction method to establish a knowledge graph between drugs and adverse reaction events; acquiring physiological function data and symptom description speech of target patients; based on the knowledge graph, physiological function data and symptom description speech, monitoring the adverse reactions of target patients to obtain monitoring results, and providing a judgment basis for the diagnosis of medical staff based on the monitoring results and the knowledge graph, so as to efficiently monitor the adverse reactions of drugs through the knowledge graph, and provide real-time alarms and medication guidance based on patient information, to help doctors make more scientific decisions.

[0059] Embodiment 2: Based on the same inventive concept, the embodiment of the present invention also provides a monitoring device for adverse reactions, such as Figure 2 As shown, including: The first acquisition module 201 is used to acquire data information related to the record of adverse reactions; Establishing module 202, for establishing a knowledge graph between drugs and adverse events based on the data information by adopting an ontology building method; The second acquisition module 203 is used to acquire physiological function data and symptom description voice of the target patient; Obtaining module 204, for monitoring adverse reactions of target patients based on the knowledge graph, the physiological function data and the symptom description voice, and obtaining monitoring results; A module 205 is provided for providing a judgment basis for the diagnosis of medical personnel based on the monitoring results and the knowledge graph.

[0060] In an optional implementation, the first acquisition module 201 is used to: Obtain data and information related to adverse reactions from drug instructions, academic papers, electronic medical records, and adverse event reports.

[0061] In an optional implementation, a module 202 is established for: Based on the data information, determine patient characteristics, drug names, adverse events, symptoms, and diseases; With patient characteristics, drug names, drug ingredients, adverse events, symptoms, and diseases as entity nodes, and with the first pointing relationship between patient characteristics and drug names, the second pointing relationship between patient characteristics and diseases, the third pointing relationship between drug names and adverse events, the fourth pointing relationship between adverse events and symptoms, the fifth pointing relationship between drug names and diseases, and the sixth pointing relationship between drug names and drug ingredients, the connection relationship between entity nodes is established to construct a knowledge graph between drugs and adverse events.

[0062] In an optional implementation, the module 204 is obtained for: Text parsing, feature extraction, and semantic analysis are used on the target patient’s symptom description speech to determine the target symptoms; The target symptoms and the physiological function data are compared with each entity node in the knowledge graph, and it is determined whether there is a preset pointing relationship with the adverse reaction event to obtain the monitoring result.

[0063] In an optional implementation, the module 204 is obtained for: Obtaining an adverse reaction event prediction model, wherein the adverse reaction event prediction model is determined based on historical adverse reaction events, and the historical adverse reaction events are determined based on the knowledge graph; Based on the adverse reaction event prediction model, physiological function data and symptom description voice, the adverse reactions of the target patients are monitored to obtain monitoring results.

[0064] In an optional implementation, the system further includes: an optimization and updating module, configured to: Based on the monitoring results, determine new adverse events and / or new drugs; Based on the newly added adverse reaction events and / or newly added drugs, the knowledge graph is optimized and updated to obtain a new knowledge graph.

[0065] In an optional implementation, the optimization update module is used to: Based on the newly added adverse reaction events and / or newly added drugs, determining target entity nodes related to the newly added adverse reaction events and / or newly added drugs; The newly added adverse reaction events and / or newly added drugs are respectively linked to corresponding target entity nodes to obtain a new knowledge graph.

[0066] In an optional embodiment, a module 205 is provided for: Based on the new knowledge graph, the characteristics of target patients with the same adverse reaction events are analyzed to provide a basis for diagnosis by medical staff.

[0067] Embodiment three: Based on the same inventive concept, an embodiment of the present invention provides a computer device, such as Figure 3 As shown, it includes a memory 304, a processor 302, and a computer program stored in the memory 304 and executable on the processor 302. When the processor 302 executes the program, the steps of the above-mentioned adverse reaction monitoring method are implemented.

[0068] Among them, Figure 3 In the embodiment of the present invention, a bus architecture (represented by bus 300) is shown, and bus 300 may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 306 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, namely a transceiver, providing a unit for communicating with various other devices on a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 may be used to store data used by processor 302 when performing operations.

[0069] Embodiment 4: Based on the same inventive concept, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned adverse reaction monitoring method when executed by a processor.

[0070] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific languages ​​is for disclosing the best mode of the present invention.

[0071] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0072] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recorded in each embodiment. More specifically, as reflected in each embodiment, the inventive aspects lie in less than all the features of the single embodiment disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0073] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and further may be divided into a plurality of submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device so disclosed may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature that provides the same, equivalent or similar purpose.

[0074] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in a specific embodiment, any one of the claimed embodiments may be used in any combination.

[0075] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all functions of some or all components of the monitoring device for adverse reactions and computer equipment according to an embodiment of the present invention. The present invention may also be implemented as a device or device program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0076] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

Claims

1. A method for monitoring adverse reactions, characterized in that: include: Obtain data information related to the record of adverse reactions; Based on the data information, an ontology construction method is used to establish a knowledge graph between drugs and adverse reaction events; Acquire physiological function data and symptom description voice of target patients; Based on the knowledge graph, the physiological function data and the symptom description voice, the adverse reactions of the target patient are monitored to obtain monitoring results; Based on the monitoring results and the knowledge graph, a judgment basis is provided for the diagnosis of medical staff.

2. The method according to claim 1, characterized in that The obtaining of data information related to the record of adverse reactions includes: Obtain data and information related to adverse reactions from drug instructions, academic papers, electronic medical records, and adverse event reports.

3. The method according to claim 1, characterized in that Based on the data information, an ontology construction method is used to establish a knowledge graph between drugs and adverse reaction events, including: Based on the data information, determine patient characteristics, drug names, adverse events, symptoms, and diseases; With patient characteristics, drug names, drug ingredients, adverse events, symptoms, and diseases as entity nodes, and with the first pointing relationship between patient characteristics and drug names, the second pointing relationship between patient characteristics and diseases, the third pointing relationship between drug names and adverse events, the fourth pointing relationship between adverse events and symptoms, the fifth pointing relationship between drug names and diseases, and the sixth pointing relationship between drug names and drug ingredients, the connection relationship between entity nodes is established to construct a knowledge graph between drugs and adverse events.

4. The method according to claim 1, characterized in that Based on the knowledge graph, physiological function data and symptom description voice, the adverse reactions of the target patient are monitored to obtain monitoring results, including: Text parsing, feature extraction, and semantic analysis are used on the target patient’s symptom description speech to determine the target symptoms; The target symptoms and the physiological function data are compared with each entity node in the knowledge graph, and it is determined whether there is a preset pointing relationship with the adverse reaction event to obtain the monitoring result.

5. The method according to claim 3, characterized in that Based on the knowledge graph, physiological function data and symptom description voice, the adverse reactions of the target patient are monitored to obtain monitoring results, including: Obtaining an adverse reaction event prediction model, wherein the adverse reaction event prediction model is determined based on historical adverse reaction events, and the historical adverse reaction events are determined based on the knowledge graph; Based on the adverse reaction event prediction model, physiological function data and symptom description voice, the adverse reactions of the target patients are monitored to obtain monitoring results.

6. The method according to claim 5, characterized in that After monitoring the adverse reactions of the target patient based on the adverse reaction event prediction model, physiological function data and symptom description voice and obtaining the monitoring results, the method further includes: Based on the monitoring results, determine new adverse events and / or new drugs; Based on the newly added adverse reaction events and / or newly added drugs, the knowledge graph is optimized and updated to obtain a new knowledge graph.

7. The method according to claim 6, characterized in that Based on the newly added adverse reaction events and / or newly added drugs, the knowledge graph is optimized and updated to obtain a new knowledge graph, including: Based on the newly added adverse reaction events and / or newly added drugs, determining target entity nodes related to the newly added adverse reaction events and / or newly added drugs; The newly added adverse reaction events and / or newly added drugs are respectively linked to corresponding target entity nodes to obtain a new knowledge graph.

8. The method according to claim 6, characterized in that Based on the monitoring results and knowledge graph, a judgment basis is provided for the diagnosis of medical staff, including: Based on the new knowledge graph, the characteristics of target patients with the same adverse reaction events are analyzed to provide a basis for diagnosis by medical staff.

9. A device for monitoring adverse reactions, characterized in that: include: A first acquisition module is used to acquire data information related to the record of adverse reactions; Establishing a module for establishing a knowledge graph between drugs and adverse events based on the data information by adopting an ontology construction method; The second acquisition module is used to acquire the physiological function data and symptom description voice of the target patient; An obtaining module is used to monitor adverse reactions of target patients based on the knowledge graph, the physiological function data and the symptom description voice to obtain monitoring results; A module is provided for providing a judgment basis for the diagnosis of medical staff based on the monitoring results and the knowledge graph.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

Images