Information processing method and device, electronic equipment and storage medium

Abstract

The invention provides an information processing method and device, electronic equipment and a storage medium. According to the specific implementation scheme, the method comprises the steps of obtaining an input query question related to an illness state; cue words are generated based on the query question and input into the large model, the large model outputs a first recommendation answer based on the cue words, and the first recommendation answer at least comprises a first medicine combination; acquiring medication evidence information of medicines in the first medicine combination; and according to the medication evidence information and the query problem, determining an identification result of the first drug combination, the identification result at least comprising description information of the drug and a matching condition of the drug and an illness state. Therefore, the recommendation answer is automatically generated through the large model, the accuracy of medicine use can be improved, meanwhile, the recognition result of the first medicine combination is determined through the medicine evidence information, verification of the first medicine combination can be achieved, and through the verification mechanism, the accuracy of medicine use can be improved. The medication safety can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, specifically to the field of artificial intelligence technology such as intelligent search, big data and deep learning, and in particular to an information processing method, device, electronic device and storage medium. Background Art

[0002] With the rapid development of artificial intelligence technology, large language models are increasingly used in the field of medical health. Especially in the medication recommendation system, how to ensure the accuracy and safety of the recommendation results has become the key to the development of the industry. However, traditional medication recommendation systems often rely on fixed algorithms and rules, which are difficult to cope with complex and changing clinical situations, and there are problems with improper recommendations or medication safety. Summary of the invention

[0003] The present disclosure provides an information processing method, device, electronic device and storage medium.

[0004] According to a first aspect of the present disclosure, there is provided an information processing method, comprising: obtaining an input query question related to a disease condition; generating a prompt word based on the query question and inputting the prompt word into a large model, wherein the large model outputs a first recommended answer based on the prompt word, wherein the first recommended answer includes at least a first drug combination; obtaining medication evidence information of the drugs in the first drug combination; and determining an identification result of the first drug combination based on the medication evidence information and the query question, wherein the identification result includes at least description information of the drugs and a matching situation between the drugs and the disease condition.

[0005] According to a second aspect of the present disclosure, there is provided an information processing device, comprising: an input module for acquiring an input query question related to a disease condition; an output module for generating a prompt word based on the query question and inputting it into a large model, wherein the large model outputs a first recommended answer based on the prompt word, wherein the first recommended answer includes at least a first drug combination; an acquisition module for acquiring medication evidence information of drugs in the first drug combination; and a determination module for determining a recognition result of the first drug combination based on the medication evidence information and the query question, wherein the recognition result includes at least description information of the drugs and a matching situation between the drugs and the disease condition.

[0006] According to a third aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the information processing method described in the above-mentioned one aspect embodiment.

[0007] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, on which a computer program / instructions are stored, and the computer instructions are used to enable the computer to execute the information processing method described in the above-mentioned embodiment.

[0008] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the information processing method described in the above-mentioned first embodiment is implemented.

[0009] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description.

[0010] Beneficial effect: Thus, by automatically generating recommended answers through a large model, the accuracy of medication can be improved. At the same time, the identification result of the first drug combination is determined through drug evidence information, and the verification of the first drug combination can be realized. Through this verification mechanism, the safety of medication can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0012] Figure 1 A flowchart of an information processing method provided by an embodiment of the present disclosure;

[0013] Figure 2 A flowchart of another information processing method provided by an embodiment of the present disclosure;

[0014] Figure 3 A flowchart of another information processing method provided by an embodiment of the present disclosure;

[0015] Figure 4 A flowchart of another information processing method provided by an embodiment of the present disclosure;

[0016] Figure 5 A flowchart of another information processing method provided by an embodiment of the present disclosure;

[0017] Figure 6 A flowchart of another information processing method provided by an embodiment of the present disclosure;

[0018] Figure 7 A schematic diagram of the overall process of an information processing device provided by an embodiment of the present disclosure;

[0019] Figure 8 A schematic diagram of the overall process of another information processing device provided by an embodiment of the present disclosure;

[0020] Fig. 9 A schematic diagram of the structure of an information processing device provided by an embodiment of the present disclosure;

[0021] Fig.10 A block diagram of an electronic device according to an information processing method of an embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0023] The information processing method, apparatus, and electronic device according to the embodiments of the present disclosure are described below with reference to the accompanying drawings.

[0024] Artificial Intelligence (AI) is a discipline that studies how computers can simulate certain thought processes and intelligent behaviors of human beings (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level technologies and software-level technologies. AI hardware technologies generally include computer vision technology, speech recognition technology, natural language processing technology, as well as learning / deep learning, big data processing technology, knowledge graph technology, and other aspects.

[0025] Natural Language Processing (NLP) is an important field in the fields of computer science and artificial intelligence. It studies various theories and methods that can achieve effective communication between humans and computers using natural language. Natural Language Processing is a science that integrates linguistics, computer science, and mathematics. Natural Language Processing is mainly used in machine translation, public opinion monitoring, automatic summarization, opinion extraction, text classification, question answering, text semantic comparison, speech recognition, etc.

[0026] Deep learning (DL) is a new research direction in the field of machine learning (ML). It is introduced into machine learning to make it closer to its original goal - artificial intelligence. Deep learning is the inherent laws and representation levels of learning sample data. The information obtained in the learning process is of great help in the interpretation of data such as text, images and sounds. Its ultimate goal is to enable machines to have analytical learning capabilities like humans and to recognize data such as text, images and sounds. Deep learning is a complex machine learning algorithm that has achieved results in speech and image recognition that far exceed previous related technologies.

[0027] Smart search is a new generation of search engine that combines artificial intelligence technology. In addition to providing traditional fast retrieval and relevance sorting functions, it can also provide user role registration, automatic identification of user interests, semantic understanding of content, intelligent information filtering and push, etc.

[0028] Figure 1 A flowchart of an information processing method provided by an embodiment of the present disclosure.

[0029] like Figure 1 As shown, the information processing method may include:

[0030] S101, obtaining an input query question related to the disease condition.

[0031] It should be noted that the query question may be manually input, or may be obtained through information from other channels, for example, through the inquiry records of the patient by medical staff, or may be obtained through the patient's medical records.

[0032] In one possible implementation, after obtaining data related to the query question, the data can be preprocessed to obtain query questions related to the condition. There can be many methods for preprocessing, which are not limited here. For example, preprocessing may include one or more of data cleaning, data conversion, data standardization, etc. By preprocessing the data, the query question can be accurately obtained, and the efficiency of subsequent data processing can be improved.

[0033] S102, generating prompt words based on the query question and inputting them into the big model, and the big model outputs a first recommended answer based on the prompt words, and the first recommended answer at least includes a first drug combination.

[0034] It should be noted that the prompt words are keywords in the query question. For example, the query question may be "What are the Chinese patent medicines for treating insomnia and night sweats?", and the corresponding prompt words may be "insomnia", "night sweats", and "Chinese patent medicines".

[0035] In the embodiments of the present disclosure, there may be many methods for generating prompt words based on the query question, which are not limited here.

[0036] In one possible implementation, the query question may be input into the prompt word generation model, so that the prompt word generation model outputs the prompt word corresponding to the query question. The prompt word generation model is a pre-trained model and may exist in the storage space of the electronic device so that it can be retrieved and used when needed.

[0037] In another possible implementation, prompt words may be extracted from the query question based on a preset prompt word extraction rule to obtain prompt words. The prompt word extraction rule is designed in advance and may be changed according to actual design needs, and is not limited here. For example, the prompt word extraction rule may be: extracting all nouns and verbs appearing in the query question, or extracting nouns, verbs, and phrases consisting of nouns and nearby verbs appearing in the query question as prompt words, etc.

[0038] It should be noted that the large model in the embodiment of the present disclosure is pre-trained and has the ability to recommend medications. In one possible implementation, a medication recommendation question-answer pair <query, answer> can be first constructed based on the open source LLM, and a set of instructions can be further constructed. Then, instruction fine-tuning training is performed based on the medical pre-trained base model.

[0039] It should be noted that in order to solve the problem of speech style in recommended answers, a preference data set is constructed on the basis of this model to complete data processing (DPO) optimization, and finally the large model with the best effect is retained.

[0040] It should be noted that the drug combination in the embodiments of the present disclosure refers to the combined use of one or more drugs to achieve the purpose of coordinated treatment. Different drug combinations are used independently of each other, and the same drug combination is used in combination according to the doctor's advice.

[0041] S103, obtaining medication evidence information of the drugs in the first drug combination.

[0042] It should be noted that the medication evidence information may include various information, which is not limited here. For example, the medication evidence information may include information such as drug instructions, indications, main functions, and adverse reactions.

[0043] In the disclosed embodiments, there are multiple ways to obtain medication evidence information. For example, medication evidence information can be obtained through the instructions of the medication itself, or by searching medical books.

[0044] S104, determining the recognition result of the first drug combination according to the medication evidence information and the query question, the recognition result at least including the description information of the drug and the matching status of the drug and the disease condition.

[0045] In actual operation, since there may be side effects when the first drug combination is used simultaneously, or the components in the first drug combination may worsen the patient's condition, it is necessary to further analyze the first drug combination based on the medication evidence information and the query questions.

[0046] It should be noted that there are many methods for determining the identification result of the first drug combination based on the medication evidence information and the query question, which are not limited here. For example, the medication evidence information, the query question and the first drug combination can be input into the analysis model to output the identification result of the first drug combination.

[0047] In the disclosed embodiment, firstly, an input query question related to the condition is obtained, then a prompt word is generated based on the query question and input into the large model, and the large model outputs a first recommended answer based on the prompt word, and the first recommended answer includes at least a first drug combination, and then the medication evidence information of the drugs in the first drug combination is obtained, and finally, the recognition result of the first drug combination is determined based on the medication evidence information and the query question, and the recognition result includes at least the description information of the drug and the matching of the drug with the condition. Thus, by automatically generating recommended answers through the large model, the accuracy of medication can be improved, and at the same time, by determining the recognition result of the first drug combination through the drug evidence information, the verification of the first drug combination can be realized. Through this verification mechanism, the safety of medication can be improved.

[0048] In the above embodiment, the medication evidence information of the drugs in the first drug combination can also be obtained by Figure 2 Explaining further, the method includes:

[0049] S201, performing natural language processing on the drugs in the first drug combination to obtain drug names related to the drugs.

[0050] S202, generating a search task according to the drug name, and executing the search task through a search engine to obtain drug use evidence information.

[0051] In the disclosed embodiment, a search engine is used to perform a retrieval task to obtain drug use evidence information. The retrieval priorities of different data sources may be first determined, and then the retrieval engine may be called. The search engine may then perform an orderly search of different data sources based on the retrieval task and the retrieval priority to obtain drug use evidence information.

[0052] In the disclosed embodiment, natural language processing is first performed on the drugs in the first drug combination to obtain drug names related to the drugs, and then a search task is generated based on the drug names, and the search task is executed through a search engine to obtain drug use evidence information. Thus, through natural language processing, a search task can be automatically established, and drug use evidence information can be retrieved, thereby improving the efficiency of obtaining drug use evidence information.

[0053] In the above embodiment, the search engine is used to perform the search task to obtain the drug use evidence information, and the drug use evidence information can also be obtained through Figure 3 Explaining further, the method includes:

[0054] S301, determining the search priorities of different data sources.

[0055] The data source may include many types, which are not limited here. For example, the data source may include drug instructions, medical classics, refined editions, encyclopedias, etc.

[0056] It should be noted that there are many schools of medicine, such as traditional Chinese medicine, Western medicine, etc., and the corresponding search results may be different for different data sources. Therefore, when conducting a search, different search priorities can be set according to design needs or the searcher's habits and requirements. This can improve the final search effect and the practicality of this solution.

[0057] In the disclosed embodiment, different data sources may correspond to different search priorities. For example, the search priority of drug instructions is higher than that of medical classics and higher than that of condensed knowledge.

[0058] S302, calling a search engine, and the search engine searches different data sources in an orderly manner based on the search task and according to the search priority to obtain drug use evidence information.

[0059] In the above embodiment, different data sources are searched in an orderly manner to obtain medication evidence information of the drug. In response to retrieving first medication evidence information from two or more data sources, a first confidence level of the first medication evidence information is obtained, and then the medication evidence information of the drug is determined from the first medication evidence information based on the first confidence level.

[0060] It should be noted that the higher the search priority, the higher the corresponding first confidence, and the more reliable the search results obtained by the corresponding search.

[0061] It should be noted that the search result with the largest first confidence can be used as the final drug use evidence information. Therefore, when multiple drug use evidence information is retrieved, the drug use evidence information can be determined from multiple search results according to the first confidence set in advance, which can avoid the problem that the drug use evidence information cannot be determined when there are differences between different data sources.

[0062] In the disclosed embodiment, the search priorities of different data sources are first determined, and then the search engine is called, and the search engine searches the different data sources in an orderly manner based on the search task and the search priority to obtain the drug use evidence information. Therefore, when searching, different search priorities can be set according to the design needs or the habits and requirements of the searcher, so as to improve the final search effect and the practicality of the solution.

[0063] In one possible implementation, in response to the search engine failing to obtain medication evidence information of a drug from different data sources, candidate medication evidence information satisfying set conditions is determined according to a query question as the medication evidence information of the drug.

[0064] It should be noted that the setting conditions are designed in advance and can be changed according to actual design needs, and no limitation is made here.

[0065] It should be noted that, according to the query question, the candidate medication evidence information that meets the set conditions can be determined. First, the candidate questions similar to the query question can be obtained from the target knowledge base, wherein the target knowledge base includes question-answer pairs, and then the answer information of the target question-answer pair associated with the candidate question is obtained as the second medication evidence information, and finally the second confidence of the second medication evidence information is obtained, and the candidate medication evidence information is determined from the second medication evidence information according to the second confidence. Therefore, in the case where valid information cannot be retrieved through the query question, the medication evidence information of the drug can be determined by setting preset conditions, which can avoid the situation where the query question is inaccurate or the medication evidence information of the drug cannot be retrieved according to the query question, thereby improving the practicality of this solution.

[0066] It should be noted that the target knowledge base is designed in advance and is a database that stores query questions and their corresponding answer information. It should be noted that the query questions in the target knowledge base may correspond to multiple answer information, and the second confidence level of each answer information may be different. The second confidence level is designed in advance and can be set according to design needs or the habits and requirements of the searcher.

[0067] In one possible implementation of the present disclosure, the data in the target knowledge base may come from drug instructions, medical classics, refined editions, encyclopedias and other knowledge, and the above knowledge is structured and parsed to build an es search service. The second confidence level of answer information from different sources may be different. For example, drug instructions are higher than medical classics and higher than refined editions.

[0068] In the above embodiment, the identification result of the first drug combination is determined based on the medication evidence information and the query question, and the Figure 4 Explaining further, the method includes:

[0069] S401, judging whether there are abnormal drugs in the first drug combination according to the drug use evidence information.

[0070] It should be noted that there are many methods for determining whether there are abnormal drugs in the first drug combination, which are not limited here. For example, it can be determined whether there are side effects when the drugs in the first drug combination are used simultaneously, or whether the simultaneous use of the drugs will reduce the efficacy of the drugs.

[0071] S402, in response to the presence of abnormal drugs in the first drug combination, generating an identification result of the first drug combination according to the drug use evidence information and the abnormal drugs, the query question and the first recommended answer, wherein the identification result also includes the abnormal drugs and the abnormal reasons of the abnormal drugs.

[0072] In the disclosed embodiment, firstly, it is determined whether there are abnormal drugs in the first drug combination according to the medication evidence information. In response to the presence of abnormal drugs in the first drug combination, an identification result of the first drug combination is generated according to the medication evidence information and the abnormal drugs, the query question and the first recommended answer, wherein the identification result also includes the abnormal drugs and the abnormal reasons of the abnormal drugs. Thus, it is determined whether there are abnormal drugs in the first drug combination according to the medication evidence information, so as to correct the first recommended answer. Through this verification mechanism, the safety of medication can be improved.

[0073] In the disclosed embodiment, after obtaining the recognition result, the first recommended answer can be rewritten according to the recognition result to obtain a second recommended answer, and the second recommended answer at least includes the second drug combination. For example, the first recommended answer may include the first drug combination A, B, and C, and the recognition result is that A is abnormal, then the second recommended answer may only include the second drug combination B and C.

[0074] It should be noted that, in order to rewrite the first recommended answer according to the recognition result to obtain the second recommended answer, the relevant description of the query question in the first recommended answer can be retained first, and then the key description information of each drug in the second drug combination can be determined, and finally the second recommended answer can be generated according to the relevant description and the key description information of the drug. Figure 5 As shown. Figure 5 It can be seen that Jinkui Shenqi Pills and Yougui Pills are not abnormal drugs, so the conclusions of the second recommended answer are the same as those of the first recommended answer.

[0075] In another possible implementation, in response to the identification result indicating that there are abnormal drugs in the first drug combination, the abnormal drugs in the first drug combination are removed, and the non-abnormal drugs in the first drug combination are retained, and the second recommended answer is generated in combination with the query question. For example, Figure 6As shown. Figure 6 It can be seen that "Thinking process": According to the reference evidence, Yupingfeng granules have the effects of invigorating qi, strengthening the exterior, and stopping sweating, and have a certain effect on improving the symptoms of anhidrosis. However, this customer wants to solve the problem of swollen feet, which is contrary to his need to increase sweating and improve the symptoms of anhidrosis. Therefore, ephedrine is not suitable for this customer's symptoms. "Conclusion": ""Problematic drug": "ephedrine", so ephedrine is removed from the second recommended answer, and only Yupingfeng granules are retained.

[0076] In another possible implementation, in response to the recognition result indicating that there are no abnormal drugs in the first drug combination, the first drug combination is determined to be a second drug combination, and a second recommended answer is generated in combination with a relevant description of the query question.

[0077] It should be noted that the second recommended answers finally generated in the above three embodiments are different and can be personalized for different groups of people.

[0078] Therefore, by setting different rewriting rules, the first recommended answer can be personalized according to the habits of different groups of people, thereby improving the patient's user experience.

[0079] Figure 7 FIG. 1 is a schematic diagram of an overall flow of an information processing method according to an embodiment of the present disclosure. Figure 7 As shown, after obtaining the input query question related to the disease condition, a prompt word can be generated based on the query question and input into the big model, and the big model outputs a first recommended answer based on the prompt word, and then natural language processing is performed on the drugs in the first drug combination to obtain drug-related drug names, and then a retrieval task is generated based on the drug name, and the retrieval task is executed through a search engine to obtain drug use evidence information, and then based on the medication evidence information, it is determined whether there are abnormal drugs in the first drug combination. In response to the presence of abnormal drugs in the first drug combination, an identification result of the first drug combination is generated based on the medication evidence information and the abnormal drugs, the query question and the first recommended answer. If there are no abnormal drugs in the first drug combination, the identification result does not include the abnormal drugs and the abnormal reasons for the abnormal drugs.

[0080] Figure 8 FIG. 1 is a schematic diagram of the overall flow of another information processing method in an embodiment of the present disclosure. Figure 8As shown, firstly, a query question is obtained, and then a query instruction is generated based on the query question, the query instruction is used to call the big model and input the query question into the big model for query, wherein the big model is trained in advance, and the big model outputs a first recommended answer, the first recommended answer at least includes a first drug combination, and the drug name is obtained by natural language processing of the first drug combination, and then the drug name is input into the search engine to determine the medication evidence information through the search engine search, if the search is empty, the medication evidence information is obtained from the target knowledge base, the first recommended answer is identified, in response to the identification result indicating that there are abnormal drugs in the first drug combination, the abnormal drugs in the first drug combination are removed, and the first drug combination without abnormal drugs is retained, combined with the query question, a second recommended answer is generated, in response to the identification result being normal, the first drug combination is determined to be the second drug combination, and in combination with the relevant description of the query question, a second recommended answer is generated.

[0081] Corresponding to the information processing methods provided in the above-mentioned embodiments, an embodiment of the present disclosure further provides an information processing device. Since the information processing device provided in the embodiment of the present disclosure corresponds to the information processing methods provided in the above-mentioned embodiments, the implementation methods of the above-mentioned information processing methods are also applicable to the information processing device provided in the embodiment of the present disclosure and will not be described in detail in the following embodiments.

[0082] Fig. 9 The schematic diagram of the structure of an information processing device provided by the embodiment of the present disclosure includes: an input module 910 , an output module 920 , an acquisition module 930 and a determination module 940 .

[0083] The input module 910 is used to obtain input query questions related to the disease condition.

[0084] The output module 920 is used to generate prompt words based on the query question and input them into the big model, and the big model outputs a first recommended answer based on the prompt words, and the first recommended answer at least includes a first drug combination.

[0085] The acquisition module 930 is used to acquire medication evidence information of the drugs in the first drug combination.

[0086] The determination module 940 is used to determine the recognition result of the first drug combination according to the medication evidence information and the query question, and the recognition result at least includes the description information of the drug and the matching status of the drug and the disease condition.

[0087] In the disclosed embodiment, the acquisition module 930 is also used to: perform natural language processing on the drugs in the first drug combination to obtain drug names related to the drugs; generate a retrieval task based on the drug names, and execute the retrieval task through a search engine to obtain drug use evidence information.

[0088] In the disclosed embodiment, the acquisition module 930 is also used to: determine the retrieval priority of different data sources; call the retrieval engine, and the search engine performs an orderly search of different data sources based on the retrieval task and the retrieval priority to obtain the drug use evidence information.

[0089] In the embodiment of the present disclosure, the acquisition module 930 is also used to: in response to retrieving the first medication evidence information from two or more data sources, obtain the first confidence of the first medication evidence information; and determine the medication evidence information of the drug from the first medication evidence information according to the first confidence.

[0090] In the disclosed embodiment, the acquisition module 930 is also used to: in response to the retrieval engine failing to obtain medication evidence information of the drug from different data sources, determine candidate medication evidence information that meets the set conditions according to the query question as the medication evidence information of the drug.

[0091] In the disclosed embodiment, the acquisition module 930 is also used to: obtain candidate questions similar to the query question from the target knowledge base, wherein the target knowledge base includes question-answer pairs; obtain answer information of the target question-answer pair associated with the candidate question as second medication evidence information; obtain a second confidence level of the second medication evidence information, and determine candidate medication evidence information from the second medication evidence information based on the second confidence level.

[0092] In the embodiment of the present disclosure, the determination module 940 is also used to: determine whether there are abnormal drugs in the first drug combination based on the medication evidence information; in response to the existence of abnormal drugs in the first drug combination, generate an identification result of the first drug combination based on the medication evidence information and the abnormal drugs, the query question and the first recommended answer; wherein the identification result is also used for the abnormal drugs and the abnormal reasons for the abnormal drugs.

[0093] In the disclosed embodiment, if there is no abnormal drug in the first drug combination, the identification result does not include the abnormal drug and the abnormal reason of the abnormal drug.

[0094] In the disclosed embodiment, the device is further used to: rewrite the first recommended answer according to the recognition result to obtain a second recommended answer, and the second recommended answer is also used for at least the second drug combination.

[0095] In the disclosed embodiment, the determination module 940 is further used to: retain the relevant description of the query question in the first recommended answer; determine the key description information of each drug in the second drug combination; and generate a second recommended answer based on the relevant description and the key description information of the drug.

[0096] In the embodiment of the present disclosure, the determination module 940 is also used to: in response to the identification result indicating that there are abnormal drugs in the first drug combination, remove the abnormal drugs in the first drug combination, retain the drugs without abnormalities in the first drug combination, and generate a second recommended answer in combination with the query question.

[0097] In the embodiment of the present disclosure, the determination module 940 is further used to: in response to the identification result indicating that there are no abnormal drugs in the first drug combination, determine that the first drug combination is a second drug combination, and generate a second recommended answer in combination with the relevant description of the query question.

[0098] Therefore, by automatically generating recommended answers through a large model, the accuracy of medication can be improved. At the same time, the identification result of the first drug combination is determined through drug evidence information, and the first drug combination can be verified. Through this verification mechanism, the safety of medication can be improved.

[0099] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0100] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0101] Fig.10 A schematic block diagram of an example electronic device 1000 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0102] like Fig.10 As shown, the device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program / instruction stored in a read-only memory (ROM) 1002 or a computer program / instruction loaded from a storage unit 1006 to a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0103] A number of components in the device 1000 are connected to the I / O interface 1005, including: an input unit 1006 such as a keyboard, a mouse, etc.; an output unit 1007 such as various types of displays, speakers, etc.; a storage unit 1008 such as a disk, an optical disk, etc.; and a communication unit 1009 such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0104] The computing unit 1001 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as information processing methods. For example, in some embodiments, the information processing method may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 1006. In some embodiments, part or all of the computer program / instructions may be loaded and / or installed on the device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program / instructions are loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the information processing method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute the information processing method in any other appropriate manner (eg, by means of firmware).

[0105] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs / instructions, which may be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general programmable processor, which may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0106] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0107] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0108] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0109] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0110] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs / instructions running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0111] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in the disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in the disclosure can be achieved, and this document does not limit them here.

[0112] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. An information processing method, It is characterized in that The method comprises: Obtaining input query questions related to the disease condition; Generate a prompt word based on the query question and input it into a large model, and the large model outputs a first recommended answer based on the prompt word, wherein the first recommended answer at least includes a first drug combination; Obtaining medication evidence information of the drugs in the first drug combination; According to the medication evidence information and the query question, an identification result of the first drug combination is determined, and the identification result at least includes description information of the drug and matching status of the drug and the condition.

2. The method according to claim 1, in, The obtaining of medication evidence information of the drugs in the first drug combination includes: Performing natural language processing on the drugs in the first drug combination to obtain drug names related to the drugs; A search task is generated according to the drug name, and the search task is executed through a search engine to obtain medication evidence information of the drug.

3. The method according to claim 2, in, The step of performing the retrieval task through a search engine to obtain the medication evidence information of the drug includes: Determine the search priorities of different data sources; The search engine is called, and the search engine performs an orderly search on the different data sources based on the search task and according to the search priority to obtain the medication evidence information of the drug.

4. The method according to claim 3, in, The orderly searching of the different data sources to obtain the medication evidence information of the drug includes: In response to retrieving first medication evidence information from two or more data sources, obtaining a first confidence level of the first medication evidence information; According to the first confidence level, medication evidence information of the drug is determined from the first medication evidence information.

5. The method according to claim 3, in, The method further comprises: In response to the search engine failing to obtain medication evidence information of the drug from the different data sources, candidate medication evidence information satisfying set conditions is determined according to the query question as the medication evidence information of the drug.

6. The method according to claim 5, in, The step of determining candidate medication evidence information that meets set conditions according to the query question includes: Acquire candidate questions similar to the query question from a target knowledge base, wherein the target knowledge base includes question-answer pairs; Obtaining answer information of a target question-answer pair associated with the candidate question as second medication evidence information; A second confidence level of the second medication use evidence information is obtained, and the candidate medication use evidence information is determined from the second medication use evidence information according to the second confidence level.

7. The method according to any one of claims 1 to 6, in, The step of determining the identification result of the first drug combination according to the medication evidence information and the query question includes: Determining whether there is an abnormal drug in the first drug combination according to the medication evidence information; In response to the presence of an abnormal drug in the first drug combination, generating an identification result of the first drug combination according to the drug use evidence information and the abnormal drug, the query question and the first recommended answer; The identification result also includes the abnormal drug and the abnormal reason of the abnormal drug.

8. The method according to claim 7, in, If there is no abnormal drug in the first drug combination, the identification result does not include the abnormal drug and the abnormal reason of the abnormal drug.

9. The method according to claim 1, in, The method further comprises: The first recommended answer is rewritten according to the recognition result to obtain the second recommended answer, where the second recommended answer at least includes a second drug combination.

10. The method according to claim 9, in, The rewriting the first recommended answer according to the recognition result to obtain the second recommended answer includes: Retaining the relevant description of the query question in the first recommended answer; Determining key descriptive information of each drug in the second drug combination; The second recommended answer is generated according to the relevant description and the key description information of the drug.

11. The method according to claim 9, in, The rewriting the first recommended answer according to the recognition result to obtain the second recommended answer includes: In response to the identification result indicating that there are abnormal drugs in the first drug combination, the abnormal drugs in the first drug combination are removed, and the non-abnormal drugs in the first drug combination are retained, and the second recommended answer is generated in combination with the query question.

12. The method according to claim 9, in, The rewriting the first recommended answer according to the recognition result to obtain the second recommended answer includes: In response to the identification result indicating that there are no abnormal drugs in the first drug combination, the first drug combination is determined to be a second drug combination, and the second recommended answer is generated in combination with the relevant description of the query question.

13. An information processing device, It is characterized in that The method comprises: An input module, used to obtain input query questions related to the disease condition; An output module, configured to generate prompt words based on the query question and input them into a large model, and the large model outputs a first recommended answer based on the prompt words, wherein the first recommended answer at least includes a first drug combination; An acquisition module, used for acquiring medication evidence information of the drugs in the first drug combination; A determination module is used to determine the recognition result of the first drug combination according to the medication evidence information and the query question, and the recognition result at least includes the description information of the drug and the matching status of the drug and the condition.

14. The device according to claim 13, in, The acquisition module is further used for: Performing natural language processing on the drugs in the first drug combination to obtain drug names related to the drugs; A search task is generated according to the drug name, and the search task is executed through a search engine to obtain medication evidence information of the drug.

15. The device according to claim 14, in, The acquisition module is further used for: Determine the search priorities of different data sources; The search engine is called, and the search engine performs an orderly search on the different data sources based on the search task and according to the search priority to obtain the medication evidence information of the drug.

16. The device according to claim 15, in, The acquisition module is further used for: In response to retrieving first medication evidence information from two or more data sources, obtaining a first confidence level of the first medication evidence information; According to the first confidence level, medication evidence information of the drug is determined from the first medication evidence information.

17. The device according to claim 15, in, The acquisition module is further used for: In response to the search engine failing to obtain medication evidence information of the drug from the different data sources, candidate medication evidence information satisfying set conditions is determined according to the query question as the medication evidence information of the drug.

18. The device according to claim 17, in, The acquisition module is further used for: Acquire candidate questions similar to the query question from a target knowledge base, wherein the target knowledge base includes question-answer pairs; Obtaining answer information of a target question-answer pair associated with the candidate question as second medication evidence information; A second confidence level of the second medication use evidence information is obtained, and the candidate medication use evidence information is determined from the second medication use evidence information according to the second confidence level.

19. The device according to any one of claims 13 to 18, in, The determining module is further used for: Determining whether there is an abnormal drug in the first drug combination according to the medication evidence information; In response to the presence of an abnormal drug in the first drug combination, generating an identification result of the first drug combination according to the drug use evidence information and the abnormal drug, the query question and the first recommended answer; The identification result is also used for the abnormal drug and the abnormal cause of the abnormal drug.

20. The device according to claim 19, in, If there is no abnormal drug in the first drug combination, the identification result does not include the abnormal drug and the abnormal reason of the abnormal drug.

21. The device according to claim 1, in, The device is also used for: The first recommended answer is rewritten according to the recognition result to obtain the second recommended answer, and the second recommended answer is also used for at least a second drug combination.

22. The device according to claim 21, in, The determining module is further used for: Retaining the relevant description of the query question in the first recommended answer; Determining key descriptive information of each drug in the second drug combination; The second recommended answer is generated according to the relevant description and the key description information of the drug.

23. The device according to claim 21, in, The determining module is further used for: In response to the identification result indicating that there are abnormal drugs in the first drug combination, the abnormal drugs in the first drug combination are removed, and the non-abnormal drugs in the first drug combination are retained, and the second recommended answer is generated in combination with the query question.

24. The device according to claim 9, in, The determining module is further used for: In response to the identification result indicating that there are no abnormal drugs in the first drug combination, the first drug combination is determined to be a second drug combination, and the second recommended answer is generated in combination with the relevant description of the query question.

25. An electronic device, include: at least one processor; as well as A memory communicatively connected to the at least one processor; characterized in that The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

26. A non-transitory computer-readable storage medium storing computer instructions, It is characterized in that The computer instructions are used to enable the computer to execute the information processing method according to any one of claims 1-12.

27. A computer program product comprising a computer program / instructions, It is characterized in that When the computer program / instructions are executed by a processor, the information processing method according to any one of claims 1 to 12 is implemented.

Images