Intelligent agent-based interrogation method and apparatus, electronic device, and storage medium

Abstract

The disclosure provides an intelligent agent-based consultation method and device, electronic equipment and storage medium, relating to the technical field of artificial intelligence, and specifically relating to the fields of medical health and technology. The specific implementation scheme is: based on the consultation intelligent agent, a plurality of rounds of consultation guidance are performed to obtain target symptom image data; in the consultation process of the consultation intelligent agent, based on the search engine, the content obtained by the consultation guidance is searched in the target knowledge base to obtain target knowledge information; based on the diagnosis intelligent agent, diagnosis is performed according to the target symptom image data and the target knowledge information to obtain diagnosis reference information; in response to the diagnosis reference information indicating that the treatment condition is met, the decision intelligent agent is called to determine the treatment medical resource according to the diagnosis reference information and user information.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, specifically to the field of medical and health technology, and in particular to a diagnostic method, apparatus, electronic device, and storage medium based on intelligent agents. Background Technology

[0002] In recent years, medical consultation systems based on natural language processing have gradually emerged, evolving from early rule-based question-answering systems to machine learning-based intelligent consultations, and now to dialogue systems based on large language models. Currently, AI-based online consultation systems suffer from problems such as insufficient consultation expertise, lack of systematic guidance, and poor specialist compatibility. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for agent-based medical consultation.

[0004] According to one aspect of this disclosure, a consultation method based on an intelligent agent is provided, comprising: conducting multiple rounds of consultation guidance based on a consultation intelligent agent to obtain target symptom profile data; during the consultation process of the consultation intelligent agent, retrieving target knowledge information from a target knowledge base based on the content obtained through consultation guidance using a retrieval engine; performing a diagnosis based on the target symptom profile data and the target knowledge information using a diagnostic intelligent agent to obtain diagnostic reference information; and responding to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, invoking a decision intelligent agent to determine medical resources for seeking medical treatment based on the diagnostic reference information and user information.

[0005] According to another aspect of this disclosure, a medical consultation device based on an intelligent agent is provided, comprising: a consultation guidance module, used to conduct multiple rounds of consultation guidance based on a consultation intelligent agent to obtain target symptom profile data; a retrieval module, used to retrieve target knowledge information from a target knowledge base based on the content obtained through consultation guidance using a retrieval engine during the consultation process of the consultation intelligent agent; a diagnosis module, used to perform a diagnosis based on the target symptom profile data and the target knowledge information using a diagnostic intelligent agent to obtain diagnostic reference information; and a determination module, used to respond to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, and to invoke a decision intelligent agent to determine the medical resources for seeking medical treatment based on the diagnostic reference information and user information.

[0006] According to another aspect of this 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, the instructions being executed by the at least one processor to enable the at least one processor to perform the agent-based consultation method described in one aspect of the above-described embodiment.

[0007] According to another aspect of this disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein computer programs / instructions are stored thereon, the computer instructions being used to cause the computer to execute the agent-based diagnostic method described in one aspect of the above-described embodiment.

[0008] According to another aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the agent-based diagnostic method described in one aspect of the above-described embodiments.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0011] Figure 1 A flowchart illustrating an agent-based diagnostic method provided in this embodiment of the present disclosure; Figure 2 This is a schematic diagram of the structure of the intelligent diagnostic agent provided in the embodiments of this disclosure; Figure 3 A schematic diagram of the multi-round consultation guidance provided in the embodiments of this disclosure; Figure 4 A schematic diagram illustrating the process of retrieving a target knowledge base as provided in an embodiment of this disclosure; Figure 5 A flowchart illustrating another agent-based diagnostic method provided in this embodiment of the present disclosure; Figure 6 A schematic diagram illustrating the process of determining medical resources for treatment provided in an embodiment of this disclosure; Figure 7 A flowchart illustrating another agent-based diagnostic method provided in this embodiment of the present disclosure; Figure 8 A schematic diagram of the online consultation process provided in this embodiment of the disclosure; Figure 9 This is a schematic diagram of the structure of the consultation system provided in the embodiments of this disclosure; Figure 10 This is a schematic diagram of the structure of a medical consultation device based on an intelligent agent, provided in an embodiment of the present disclosure. Figure 11 This is a block diagram of an electronic device used to implement the agent-based medical consultation method of the present disclosure. Detailed Implementation

[0012] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0013] The following description, with reference to the accompanying drawings, outlines an agent-based diagnostic method, apparatus, electronic device, and storage medium.

[0014] Artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It involves both hardware and software technologies. AI hardware technologies generally include computer vision, speech recognition, natural language processing, and related technologies such as deep learning, big data processing, and knowledge graphs.

[0015] Figure 1 This is a flowchart illustrating a diagnostic method based on an intelligent agent, as provided in an embodiment of this disclosure.

[0016] like Figure 1 As shown, this agent-based diagnostic method may include: S101 uses a diagnostic intelligence agent to conduct multiple rounds of consultation guidance in order to obtain target symptom profile data.

[0017] It should be noted that the executing entity of the intelligent agent-based consultation method in this embodiment can be a hardware device with data processing capabilities and / or the necessary software to drive the hardware device. Optionally, the executing entity may include a server, a user terminal, and other intelligent devices. Optionally, the user terminal includes, but is not limited to, mobile phones, computers, and intelligent voice interaction devices. Optionally, the server includes, but is not limited to, a network server, an application server, or a server of a distributed system, or a server combined with blockchain, etc. This embodiment does not impose specific limitations.

[0018] In some embodiments, before conducting multiple rounds of consultation guidance based on the consultation agent, the consultation agent can first determine the user's consultation intent, and then determine the questions or scripts that need to be conducted in multiple rounds of consultation based on the consultation intent.

[0019] In some embodiments, the user's intent to seek medical advice can be obtained from the user's input information. This intent may include the user's symptom information, such as the type of disease.

[0020] In some embodiments, the consultation agent can consult on a specific disease or on multiple diseases. That is, the consultation agent can be a targeted agent or a comprehensive agent with a broader scope.

[0021] In some embodiments, the consultation agent can analyze the user's input information to determine whether the user has the intention to seek medical advice. If the user has the intention, it can generate guidance questions for the first round of consultation. By receiving the user's response information, it can generate guidance questions for the second round of consultation based on the response information, and so on, thereby realizing multi-round consultation guidance.

[0022] It should be noted that multi-round consultation guidance based on the consultation intelligence agent does not mean continuously conducting multiple rounds of consultation. Multi-round consultation guidance can be conducted according to a set number of rounds. For example, it can be pre-set to conduct n consultation guidance rounds for each user, where n is a natural number greater than 1.

[0023] In some embodiments, different rounds of consultation guidance can be provided to different users to improve diagnostic accuracy. The number of rounds of consultation guidance for different users can be based on the consultation guidance termination condition; that is, during the consultation guidance process, the consultation will stop only if the consultation guidance termination condition is met.

[0024] Different users can meet the end conditions of the consultation guidance at different times during the consultation guidance process, thus enabling different rounds of consultation guidance for different users. To prevent the same user from having unlimited consultations, a limit value for multiple rounds of consultation guidance can be set. When the number of consultation guidance rounds reaches the limit value, the multiple rounds of consultation guidance will end.

[0025] In other words, the termination condition for guided consultation can be that the number of rounds of guided consultation reaches a certain limit. Alternatively, the consultation can end when the intelligent agent obtains information sufficient to diagnose the user.

[0026] In some embodiments, during multi-round guided consultations based on a consultation agent, the consultation guide can use guiding questions to lead users to describe their symptoms in order to obtain the user's response information. The user's response information includes a description of their symptoms.

[0027] In some embodiments, the diagnostic agent can generate target symptom profile data based on symptom description information. This target symptom profile data may include the user's symptom information. For example, keywords can be extracted from the symptom description information and then structured and assembled to obtain the target symptom profile data.

[0028] In some embodiments, the intelligent agent can obtain multiple rounds of user responses by conducting multi-round consultations. It can obtain symptom profile data for each round from the responses in each round, and update the symptom profile data of the previous round using the responses of the next round until the multi-round consultation guidance ends, the symptom profile data is updated, and the target symptom profile data is obtained.

[0029] S102, during the consultation process of the consultation agent, the target knowledge base is searched based on the content obtained by the consultation guidance in the search engine to obtain the target knowledge information.

[0030] In some embodiments, the content obtained in any round according to the consultation guidance can be obtained during the consultation process of the consultation agent, and the content obtained in multiple rounds of consultation guidance can also be used as the content obtained according to the consultation guidance during the consultation process of the consultation agent.

[0031] In some embodiments, the content obtained by the consultation guidance may be the user's response information, and the target knowledge base may be the content of databases such as disease knowledge base, medical knowledge base and professional medical database.

[0032] In some embodiments, the disease knowledge base includes information such as the symptom characteristics, diagnostic criteria, and treatment recommendations for common diseases; the pharmaceutical knowledge base includes information such as the efficacy, usage, dosage, and precautions of drugs; and the professional database includes information such as examination data for licensed physicians and pharmacists and professional documents.

[0033] In some embodiments, the target knowledge base is a vector database, which can vectorize the response information to obtain query vectors, and input the query vectors into the retrieval engine so that the retrieval engine can perform retrieval in the target knowledge base based on the query vectors to obtain target knowledge information.

[0034] The target knowledge information may include, but is not limited to, information such as disease knowledge, medical knowledge, and professional medical data.

[0035] In some embodiments, retrieval in the target knowledge base based on the query vector can be performed using vector similarity. This involves calculating the similarity between the query vector and vectors in the target knowledge base, sorting them based on similarity, and selecting the top k vectors from the sorted results as the retrieval outcome. Here, k is a natural number greater than 1.

[0036] In some embodiments, after obtaining the search results, the contextual information of the search results can be determined and input into the large language model, which then generates target knowledge information based on the contextual information.

[0037] S103, Based on the diagnostic agent, diagnosis is performed according to the target symptom profile data and target knowledge information to obtain diagnostic reference information.

[0038] In some embodiments, by inputting target symptom profile data and target knowledge information into a diagnostic agent, the diagnostic agent performs a diagnosis to obtain diagnostic reference information including possible disease types, diagnostic basis, treatment suggestions, etc.

[0039] In some embodiments, the diagnostic agent can understand target symptom profile data and target knowledge information by invoking a terminology database, thereby unifying the target symptom profile data and target knowledge information into a terminological representation. For example, target symptom profile data may include chest pain, and target knowledge information may include chest pain; both can be unified as chest pain using the terminology database.

[0040] In some embodiments, the diagnostic agent can perform inference and diagnosis based on the unified target symptom profile data and target knowledge information to obtain diagnostic reference information.

[0041] For example, the symptom information contained in the target symptom profile data is matched with the target knowledge information to determine the disease knowledge, medical knowledge and professional medical information that match the symptom information as the matching result, and the disease type, diagnostic basis and treatment suggestions are determined from the matching result as diagnostic reference information.

[0042] In some embodiments, when matching disease knowledge, medical knowledge, and professional medical information, the matching value between symptom information and the matched disease knowledge, medical knowledge, and professional medical information can also be determined. When outputting diagnostic reference information, the matching value can also be output, thereby quantifying the diagnostic reference information and improving the reference value of the diagnostic information.

[0043] For example, diagnostic reference information could be disease type 1 (81%), disease type 2 (68%), etc.

[0044] S104, in response to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, the decision-making agent is invoked to determine the medical resources for seeking medical treatment based on the diagnostic reference information and user information.

[0045] In some embodiments, diagnostic reference information indicating that medical treatment criteria are met can be determined based on treatment recommendations within the diagnostic reference information. For example, a treatment recommendation of seeking medical attention as soon as possible can be considered as meeting the medical treatment criteria. The determination of whether diagnostic reference information indicates that medical treatment criteria are met can also be based on the disease type within the diagnostic reference information. For example, a target disease type for meeting medical treatment criteria can be pre-defined; when the disease type in the diagnostic reference information matches the target disease type, the medical treatment criteria are determined to be met.

[0046] In some embodiments, when the diagnostic reference information indicates that the conditions for seeking medical treatment are met, a decision-making agent can be invoked to determine the medical resources available for treatment based on the diagnostic reference information and user information. These medical resources may include the hospital, doctor, etc., to be treated.

[0047] In some embodiments, the decision-making agent may first determine candidate medical resources based on diagnostic reference information, and then determine the medical resource from the candidate medical resources based on user information.

[0048] For example, you can first identify general hospitals or specialized hospitals based on diagnostic reference information, and then select the target hospital from different hospitals based on user information as the medical resource for treatment.

[0049] For example, diagnostic reference information can indicate the type of disease a user has and obtain the types of diseases that different hospitals can treat and the treatment effects. Based on the type of disease and the treatment effects, candidate hospitals can be identified, and combined with user information, such as the user's location, the nearest candidate hospital can be selected as the target hospital, i.e., the medical resources available for treatment.

[0050] In other words, user information can include the user's location information, thereby enabling the selection of medical resources based on the user's location.

[0051] In some embodiments, diagnostic reference information may also include treatment recommendations. If the treatment recommendations carry information related to medical resources, medical resources can be determined based on the treatment recommendations and user information.

[0052] For example, if the treatment recommendation is to seek treatment at a specialized hospital, then the hospital closest to the user can be selected from among the specialized hospitals as the medical resource for treatment.

[0053] In some embodiments, after obtaining medical resources, these resources can be fed back to the user, and at the same time, the user's medical treatment path can be planned so that the user can receive medical treatment conveniently and quickly.

[0054] According to the agent-based consultation method provided in this disclosure, multiple rounds of consultation guidance are conducted based on the consultation agent to obtain target symptom profile data. During the consultation process of the consultation agent, a retrieval engine is used to search the target knowledge base based on the content obtained from the consultation guidance to obtain target knowledge information. This allows the diagnostic agent to perform a diagnosis based on the target symptom profile data and target knowledge information, obtaining diagnostic reference information. In response to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, a decision-making agent is invoked to determine the appropriate medical resources based on the diagnostic reference information and user information. Therefore, this solution improves the professionalism and accuracy of consultation by using an agent and combining it with a target knowledge base for diagnosis. Multiple rounds of consultation guidance during the consultation phase reduce communication barriers and improve the completeness of symptom collection, further enhancing diagnostic accuracy. Online consultations can reduce medical costs and alleviate pressure on offline medical resources, and by guiding users to seek medical treatment through relevant medical resources, it improves medical efficiency and user experience.

[0055] Based on the above embodiments, before conducting multiple rounds of consultation guidance based on the consultation intelligence agent, the consultation guidance script can be determined in advance to reduce semantic ambiguity, improve the accuracy of information collection, so that users can answer the questions with the set script and improve the comprehensiveness of the user's response information.

[0056] In some embodiments, based on the interaction between the consultation agent and the user, it can receive the initial input of natural language symptom description information, and perform intent recognition on the initial input symptom description information to determine the consultation department and the type of disease.

[0057] In some embodiments, when interacting with the intelligent diagnostic agent, users can directly input symptom descriptions into the agent, or they can input symptom descriptions after multiple rounds of interaction. That is to say, the initial input of natural language symptom descriptions is not necessarily the same as the input information during the user's first interaction with the intelligent diagnostic agent.

[0058] In some embodiments, keywords related to the consultation can be determined from the initially entered symptom description information, and the consultation department and disease type can be determined based on the keywords. For example, if the keywords are headache, abdominal pain, etc., it can be determined that the intention is to consult a doctor, and the consultation department and disease type can be further determined based on the keywords in the symptom description information.

[0059] For example, the initial symptom description information can be input into a natural language processing model for keyword recognition.

[0060] In some embodiments, when it is determined that the symptom description information has the intention to seek medical advice, the department and disease type can be matched based on the keywords in the symptom description information to determine the department and disease type to be consulted.

[0061] For example, you can first determine the department to consult, and then determine the type of disease to consult within the scope of consultation in that department.

[0062] In some embodiments, different departments may use different scripts to guide the consultation. For example, considering the specificity of symptom descriptions, consultation methods, and privacy protection requirements of different diseases, a specialized terminology database for different departments can be pre-determined. This database can understand professional terms and non-standard expressions in user descriptions and take into account the sensitivity and privacy of diseases, adopting a gentle and professional consultation approach.

[0063] For example, you can pre-set the target knowledge base to be called and the corresponding dialogue for consultation guidance for different departments. Once the consultation department is determined, you can determine the target knowledge base to be called and the corresponding dialogue for consultation guidance based on the consultation department.

[0064] Based on the above embodiments, this disclosure can further explain the process of multi-round consultation guidance. For the i-th round of the multi-round consultation guidance process, symptom information can be extracted from the natural language symptom description information input in the i-th round, based on the consultation agent, to update the symptom profile data.

[0065] In other words, each round of consultation guidance can extract symptom information from the natural language symptom description information entered by the user, obtain symptom profile data, and use the symptom information of the next round to update the symptom profile data of the previous round.

[0066] In some embodiments, a first guiding question for the (i+1)th round can be generated based on updated symptom profile data and candidate knowledge information retrieved in the i-th round, so that the guiding question can more accurately ask questions about the user's symptoms and improve the efficiency of the consultation.

[0067] In some embodiments, in order to make the guiding questions more in line with the consultation needs of the consultation department, the first guiding question can be optimized based on the corresponding dialogue of the consultation guidance to generate the second guiding question for the (i+1)th round, and displayed on the front end to obtain the natural language symptom description information for the (i+1)th round. The above process can be repeated until the preset consultation guidance end condition is met to obtain the target symptom profile data.

[0068] One of the conditions for ending the consultation guidance is that the number of consultation rounds has reached the consultation limit. For example, if the consultation limit is 4 rounds of consultation guidance, when the natural language symptom description information of the 4th round is obtained, it is determined that the consultation number has reached the consultation limit, and the consultation guidance ends.

[0069] Figure 2 This is a schematic diagram of the structure of the intelligent diagnostic agent provided in an embodiment of this disclosure. Figure 2The intelligent agent for diagnosis in the system adopts a layered architecture design, including an intent recognition layer, a symptom extraction layer, a diagnosis control layer, and a diagnosis decision layer.

[0070] The intent recognition layer is used to distinguish a user's intent to seek medical advice from other intents to consult.

[0071] The symptom extraction layer is used to extract structured symptom information from users' responses.

[0072] The consultation control layer is used to guide the dialogue process according to the preset consultation strategy.

[0073] The diagnostic decision-making layer is used to make preliminary diagnoses based on medical knowledge.

[0074] In some embodiments, the intent recognition layer may include an intent classifier and intent label output. The intent classifier interacts with the user and performs intent recognition on the initial input symptom description information, thereby outputting the consultation department and the type of disease to be consulted by the intent label output.

[0075] The symptom extraction layer can include a symptom recognizer and a symptom analyzer. The symptom recognizer can identify symptom entities during the multi-round consultation process, and the symptom analyzer can analyze the symptom entities to determine the target symptom profile data.

[0076] The consultation control layer can include a state manager and a question generator. The state manager can manage the consultation status and the question generator can guide the next round of consultation with guiding questions.

[0077] The diagnostic decision layer may include a retrieval support and a diagnostic model, such as a diagnostic agent. The retrieval support obtains the content of the consultation guidance output by the consultation control layer and interacts with the knowledge retrieval interface to perform retrieval in the target knowledge base based on the content obtained from the consultation guidance. The retrieved target knowledge information is then input into the diagnostic model to obtain diagnostic reference information.

[0078] Figure 3 This is a flowchart illustrating a multi-round consultation guidance process provided in an embodiment of this disclosure. By receiving the user's initial natural language symptom description and performing intent recognition on the initial symptom description, it is determined whether to initiate a consultation. If not, the process can proceed to a consultation agent to interact with the user.

[0079] If a consultation is initiated, symptom information is extracted from the symptom description to obtain symptom profile data. It is then determined whether the diagnostic criteria are met. If not, a knowledge retrieval is performed to obtain candidate knowledge information. Based on the symptom profile data and the candidate knowledge information, the first guiding question for the next round is generated. This first guiding question can be updated based on the corresponding consultation guidance script to obtain a second guiding question. This second guiding question is then used to receive the user's symptom description information and update the symptom profile data.

[0080] The system continues to determine if the number of consultation rounds exceeds 10. If it does, it outputs a "diagnosis not possible" message and suggests seeking medical attention. If it does not exceed the limit, it checks if the diagnostic criteria are met. If so, it can perform a diagnosis based on the target symptom profile data and target knowledge information to obtain diagnostic reference information.

[0081] One diagnostic criterion is that the obtained symptom information is sufficient for diagnosis.

[0082] Figure 4 This is a schematic diagram illustrating the process of retrieving the target knowledge base according to an embodiment of this disclosure. The content obtained based on the consultation guidance is vectorized to obtain a query vector. Based on vector similarity, a search is performed in the target knowledge base to obtain the top-k vectors as search results. The target knowledge base includes a vector database storing disease knowledge bases, medical knowledge bases, and professional medical data databases.

[0083] By determining the contextual information of the search results and inputting this contextual information into the large language model, the large language model generates target knowledge information based on the contextual information.

[0084] Figure 5 This is a flowchart illustrating another agent-based diagnostic method provided in an embodiment of this disclosure.

[0085] like Figure 5 As shown, this agent-based diagnostic method may include: S501 uses a diagnostic intelligence agent to conduct multiple rounds of consultation guidance in order to obtain target symptom profile data.

[0086] S502, during the consultation process of the consultation agent, the search engine searches the target knowledge base based on the content obtained from the consultation guidance to obtain target knowledge information.

[0087] S503, based on the diagnostic agent, performs diagnosis according to the target symptom profile data and target knowledge information to obtain diagnostic reference information.

[0088] The relevant content of steps S501-S503 can be found in the above embodiments, and will not be repeated here.

[0089] S504, in response to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, the decision-making agent constructs search keywords based on the diagnostic reference information.

[0090] In some embodiments, keyword extraction can be performed on diagnostic reference information based on the decision-making agent to extract keywords related to the disease type, and keywords related to the department to which the disease type belongs can also be extracted.

[0091] In some embodiments, the diagnostic reference information may include possible disease types, diagnostic criteria, and treatment recommendations. That is, keywords related to the disease type can be extracted from the possible disease types, and keywords related to the department to which the disease type belongs can be extracted from the treatment recommendations.

[0092] In some embodiments, by vectorizing the extracted keywords, search keywords can be obtained, wherein the search keywords include at least the target disease type and / or the target department to which the target disease type belongs.

[0093] S505: Search for medical resources based on search keywords and user information, and determine the medical resources for treatment from the first candidate medical resources found.

[0094] In some embodiments, by acquiring medical resources associated with the target disease, and by first searching for medical resources using keywords to obtain first candidate medical resources, and then filtering the first candidate medical resources based on user information, the medical resources for treatment can be determined.

[0095] In some embodiments, if the user information includes at least user location information, then the medical resources available for treatment can be those related to the user's location. For example, the medical resources closest to the user can be determined based on the user's location information.

[0096] In this disclosure, medical resources for treatment are determined based on the user's location information, which ensures that the determined medical resources are those that the user can reach, thereby achieving a reasonable allocation of medical resources, shortening the time for seeking medical treatment, and improving the convenience of seeking medical treatment.

[0097] In some embodiments, in order to improve retrieval efficiency, the resource retrieval scope can be determined first based on the user's location information, and medical resources can be retrieved within the resource retrieval scope based on the search keywords to obtain the first candidate medical resources.

[0098] For example, each medical resource has its own location information. Based on the user's location information, a range within which the user sets a distance can be determined as the resource retrieval range. Then, based on the location information of each medical resource, it can be determined whether it is within the resource retrieval range. Medical resources within the resource retrieval range can be searched according to search keywords to obtain the first candidate medical resource.

[0099] In some embodiments, medical resource retrieval based on search keywords within the resource retrieval scope may involve matching search keywords with medical resource keywords and using the successfully matched medical resources as the first candidate medical resources.

[0100] For example, if the search keyword is "department A", then medical resources within the search scope that contain "department A" will be considered the first candidate medical resources. Medical resources containing "department A" are considered successfully matched medical resources.

[0101] In some embodiments, the diagnostic reference information also includes symptom levels, which can be quantified based on the severity, urgency, stage of development, or degree of threat to life and health of the disease type.

[0102] In other words, after obtaining possible disease types as diagnostic reference information, the severity, urgency, development stage, or threat to life and health of the disease type can be combined to quantify the disease type and obtain the symptom level of the disease type.

[0103] In some embodiments, the first candidate medical resource identified by the medical resource retrieval can treat the target disease type; that is, the resource level of the first candidate medical resource should be greater than the symptom level.

[0104] In some embodiments, medical resources associated with the target disease type can be obtained, and the resource level of the associated medical resources can be determined. This allows for the screening of associated medical resources based on symptom level and resource level to obtain first-candidate medical resources. The quantification standards for resource level and symptom level are the same.

[0105] For example, medical resources with a resource level greater than or equal to the symptom level are selected as the first candidate medical resources.

[0106] In some embodiments, to improve the convenience of seeking medical treatment, a route between a first candidate medical resource and the user's location information can be determined, and the first candidate medical resource can be screened according to the route to obtain the medical resource for treatment.

[0107] In some embodiments, the user's travel time and distance along the route can be determined, thereby identifying the medical resources available for treatment from a first pool of candidate medical resources based on the travel time and distance along the route.

[0108] For example, candidate medical resources can be ranked based on travel time to obtain a first ranking result, and then ranked based on travel distance to obtain a second ranking result. The top N candidate medical resources in each of the first and second ranking results can be identified, and the intersection of these N candidate medical resources from both rankings can be performed to obtain the same candidate medical resource as the designated medical resource. Here, N is a natural number greater than 1.

[0109] For example, when N is 3, the top 3 candidate medical resources in the first ranking result are Resource 1, Resource 2 and Resource 3, and the top 3 candidate medical resources in the second ranking result are Resource 1, Resource 3 and Resource 4.

[0110] By performing intersection processing on resources 1, 2, and 3, as well as resources 1, 3, and 4, resources 1 and 3 are obtained as medical resources for outpatient visits.

[0111] In some embodiments, the process of determining medical resources for treatment can also be combined with the comprehensive evaluation data of the first candidate medical resources to determine the medical resources for treatment, thereby achieving multi-dimensional balancing, avoiding the limitations of screening medical resources from a single dimension, and making the determined medical resources for treatment have high medical quality.

[0112] In some embodiments, comprehensive evaluation data of a first candidate medical resource is obtained, and the first candidate medical resource is scored based on travel time, travel distance, and comprehensive evaluation data. For example, the first candidate medical resource can be scored based on travel time, travel distance, and comprehensive evaluation data respectively, resulting in a first score corresponding to travel time, a second score corresponding to travel distance, and a third score corresponding to comprehensive evaluation data. The first, second, and third scores are then weighted and fused to obtain the final score of the first candidate medical resource, thereby determining the appropriate medical resource from the first candidate medical resource based on the score.

[0113] In other words, you can choose the highest-rated first-choice medical resource as your medical resource for treatment.

[0114] In some embodiments, the first candidate medical resources are scored based on travel time and travel distance, and the first candidate medical resources with shorter travel time and shorter travel distance may be scored higher; the first candidate medical resources are scored based on comprehensive evaluation data, and the first candidate medical resources with higher medical quality indicated by the comprehensive evaluation data may be scored higher.

[0115] In some embodiments, after identifying the medical resources available for treatment, a navigation path between the user and the medical resources can be planned and provided to the user for selection.

[0116] According to the agent-based consultation method provided in this disclosure, the decision-making agent constructs search keywords based on diagnostic reference information, and performs medical resource retrieval based on the search keywords and user information. It then determines the medical resources for treatment from the first candidate medical resources retrieved, thereby realizing a service loop from consultation to medical treatment and achieving seamless connection from online consultation to offline medical treatment, thus improving the convenience of medical treatment.

[0117] Figure 6 This is a schematic diagram illustrating the process of determining medical resources for treatment, provided in an embodiment of this disclosure. Figure 6 The user information in the data is the user's location information.

[0118] By identifying diagnostic reference information and constructing search keywords based on it, the resource search scope can be determined according to user location information. Medical resources can then be searched within this scope using the search keywords to obtain the first-choice medical resources.

[0119] By determining the route between the first candidate medical resource and the user's location information, the travel time and distance of the route are determined. Comprehensive evaluation data of the first candidate medical resource is obtained, and a comprehensive score is calculated based on the route's travel time, distance, and overall evaluation data. Based on this score, the appropriate medical resource is selected from the first candidate pool.

[0120] In the above embodiments, user information also includes user travel information. The process of determining medical resources based on user travel information can be combined with... Figure 7 To understand further. Figure 7 This is a flowchart illustrating another agent-based diagnostic method provided in an embodiment of this disclosure.

[0121] like Figure 7 As shown, this agent-based diagnostic method may include: S701, retrieve the outpatient information of each candidate doctor in the first candidate medical resource.

[0122] S702 determines the medical resources for treatment from the first candidate medical resources based on outpatient information and user travel information.

[0123] It's important to note that user travel information refers to data related to a user's travel plans or intended destination. In other words, user travel information can indicate what a user needs to do at a specific time or period. Determining medical resources based on user travel information can avoid conflicts between user schedules and medical appointments, enabling personalized medical services.

[0124] In some embodiments, the outpatient information of candidate doctors is determined from a first pool of candidate medical resources, ultimately identifying the candidate doctors as those in the target department. For example, the outpatient information of candidate doctors can be determined from official consultations of the first pool of candidate medical resources.

[0125] In some embodiments, medical resources for treatment are determined based on outpatient information and user travel information in order to avoid time conflicts. In this case, a time conflict check is performed on the outpatient information and user travel information to identify a second candidate medical resource with a time conflict.

[0126] In other words, it can be determined whether the time information indicated by the outpatient information is the same as the time information indicated by the user's travel information, and if there is the same time information, it is determined as a time conflict, thereby identifying the second candidate medical resource with the time conflict.

[0127] For example, a user's itinerary information includes time period A and time period B, and outpatient information includes time period 1 for medical resource 1, time period 2 for medical resource 2, and time period 3 for medical resource 3. If there is a time conflict between time period A and time period 2, then medical resource 2 is determined as the second candidate medical resource.

[0128] In some embodiments, a second candidate medical resource can be removed from a first candidate medical resource to obtain a third candidate medical resource, and the medical resource to be treated can be determined from the third candidate medical resource.

[0129] It should be noted that the third candidate medical resource refers to a medical resource that does not conflict with the user's travel information in terms of time. In order to improve the efficiency of the user's travel, a fourth candidate medical resource can be selected from the third candidate medical resource whose consultation information is on the same day as the user's travel information.

[0130] In some embodiments, to improve the convenience of users seeking medical treatment, a fifth candidate medical resource with the shortest travel time can be selected from the fourth candidate medical resources based on the location information of the fourth candidate medical resource and the destination of the user's travel information, and used as the medical resource for treatment.

[0131] In some embodiments, if there is a conflict between the outpatient time of the first candidate medical resource and the user's travel information, the user's travel information can be adjusted, and the medical resource to be treated can be determined based on the adjusted user travel information.

[0132] In some embodiments, user travel information can be adjusted based on the priority of travel information, thereby ensuring that the user can seek medical treatment without delaying the user's important travel plans.

[0133] In other words, in response to the conflict between the outpatient time of the first candidate medical resource and the user's travel information, the user's travel information is prioritized, and based on the priority, the first user travel information that can be adjusted is determined.

[0134] For example, the lowest priority travel information can be selected as the first adjustable user travel information. This allows for time conflict checks between outpatient information and the first user travel information, identifying the first candidate medical resource that does not conflict with the first user travel information, and using it as the medical resource for outpatient treatment.

[0135] In some embodiments, after identifying medical resources, appointments can be made for those resources, and reference information and target symptom profile data can be fed back to the medical resources to ensure accessibility to medical care, reduce the time spent on repeated consultations and information collection, and improve diagnostic efficiency.

[0136] According to the intelligent agent-based consultation method provided in this disclosure, by obtaining the outpatient information of each candidate doctor in the first candidate medical resources, and determining the medical resources for treatment from the first candidate medical resources based on the outpatient information and user travel information, it is possible to avoid conflicts between user travel and medical treatment travel and realize personalized medical services.

[0137] Figure 8 This is a schematic diagram of the online consultation process provided in an embodiment of this disclosure.

[0138] By acquiring the symptom description information input by the user and performing intent recognition on the symptom description information, it can be determined whether to conduct a medical consultation. If a medical consultation is not conducted, the interaction can be redirected to a consultation agent to interact with the user.

[0139] If a consultation is initiated, symptom information is extracted from the symptom description to obtain symptom profile data, and the first guiding question for the next round is generated. This first guiding question can be updated based on the corresponding consultation guidance script to obtain a second guiding question. This second guiding question is then used to receive the user's symptom description information and update the symptom profile data.

[0140] The search engine retrieves the target knowledge base based on the content obtained from the consultation guidance to obtain target knowledge information and determines whether the consultation rounds have reached the consultation limit. If the limit has been reached, a diagnosis is made based on the target symptom profile data and target knowledge information to obtain diagnostic reference information; if the limit has not been reached, the consultation guidance continues.

[0141] After obtaining diagnostic reference information, search keywords can be constructed, and medical resources can be searched based on the search keywords and user information to determine the medical resources available for treatment.

[0142] Based on the above embodiments, a consultation system can also be proposed based on the agent-based consultation method of the present disclosure. Figure 9 This is a schematic diagram of the structure of the consultation system provided in an embodiment of this disclosure. Figure 9 The online consultation system comprises a user interaction layer, a consultation processing layer, and a knowledge support layer. These layers communicate through standardized interfaces to ensure the system's scalability and maintainability.

[0143] The user interaction layer provides a natural language dialogue interface to interact with users and obtain symptom descriptions input by them.

[0144] The consultation processing layer includes a medical resource recommendation module and a consultation module. The consultation module includes an intent recognition submodule, a symptom extraction submodule, a consultation control submodule, and a diagnostic decision submodule.

[0145] The knowledge support layer includes a knowledge retrieval module and a knowledge base module.

[0146] By inputting the symptom description information provided by the user into the intent recognition submodule, the user's consultation intent is identified. The symptom extraction submodule extracts symptom information from the symptom description information to update the symptom profile data. The consultation control submodule determines the questions for the next round of consultation guidance based on the candidate knowledge retrieved from the knowledge base module by the knowledge retrieval module, thereby realizing multi-round consultation guidance.

[0147] The knowledge retrieval module can also retrieve target knowledge information based on the content obtained during the consultation process, and the diagnostic decision-making submodule can then make a diagnosis based on the target symptom profile data and the target knowledge information to obtain diagnostic reference information.

[0148] Furthermore, the medical resource recommendation module determines the user's medical resources based on diagnostic reference information, and recommends medical resources to the user.

[0149] Corresponding to the agent-based consultation methods provided in the above embodiments, an embodiment of this disclosure also provides an agent-based consultation device. Since the agent-based consultation device provided in this disclosure corresponds to the agent-based consultation methods provided in the above embodiments, the implementation methods of the above agent-based consultation methods are also applicable to the agent-based consultation device provided in this disclosure, and will not be described in detail in the following embodiments.

[0150] Figure 10This is a schematic diagram of the structure of a medical consultation device based on an intelligent agent, provided in an embodiment of this disclosure.

[0151] like Figure 10 As shown, the intelligent agent-based consultation device 1000 of this disclosure includes a consultation guidance module 1001, a retrieval module 1002, a diagnosis module 1003, and a determination module 1004.

[0152] The consultation guidance module 1001 is used to conduct multiple rounds of consultation guidance based on the consultation intelligence agent in order to obtain target symptom profile data. The retrieval module 1002 is used to retrieve target knowledge information from the target knowledge base based on the content obtained according to the consultation guidance during the consultation process of the consultation intelligent agent. Diagnostic module 1003 is used to perform diagnosis based on the target symptom profile data and the target knowledge information by the diagnostic agent to obtain diagnostic reference information; The determination module 1004 is used to respond to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, and to invoke the decision-making intelligent agent to determine the medical resources for seeking medical treatment based on the diagnostic reference information and user information.

[0153] In one embodiment of this disclosure, the determining module 1004 is further configured to: construct search keywords based on the diagnostic reference information by the decision-making agent, wherein the search keywords include at least the target disease type and / or the target department to which the target disease type belongs; perform medical resource retrieval based on the search keywords and the user information, and determine the medical resource for treatment from the first candidate medical resources retrieved.

[0154] In one embodiment of this disclosure, the user information includes at least user location information. The determining module 1004 is further configured to: determine a resource retrieval range based on the user location information, and perform a medical resource retrieval within the resource retrieval range based on the retrieval keywords to obtain the first candidate medical resource; determine a route between the first candidate medical resource and the user location information; and determine the medical resource for treatment from the first candidate medical resource based on the travel time and distance of the route.

[0155] In one embodiment of this disclosure, the determining module 1004 is further configured to: acquire comprehensive evaluation data of the first candidate medical resources; score the first candidate medical resources based on the travel time of the route, the travel distance, and the comprehensive evaluation data; and determine the medical resource for treatment from the first candidate medical resources based on the score.

[0156] In one embodiment of this disclosure, the diagnostic reference information further includes a symptom level, and the determining module 1004 is further configured to: acquire medical resources associated with the target disease type; determine the resource level of the associated medical resources; and filter the associated medical resources according to the symptom level and the resource level to obtain the first candidate medical resource.

[0157] In one embodiment of this disclosure, the user information includes at least user travel information, and the determining module 1004 is further configured to: obtain the outpatient information of each candidate doctor in the first candidate medical resources; and determine the medical resource for treatment from the first candidate medical resources based on the outpatient information and the user travel information.

[0158] In one embodiment of this disclosure, the determining module 1004 is further configured to: perform a time conflict check on the outpatient information and the user travel information to determine a second candidate medical resource with a time conflict; remove the second candidate medical resource from the first candidate medical resource to obtain a third candidate medical resource; and determine the medical resource for outpatient treatment from the third candidate medical resource.

[0159] In one embodiment of this disclosure, the determining module 1004 is further configured to: filter out a fourth candidate medical resource from the third candidate medical resource whose consultation information is on the same day as the user's travel information; and, based on the location information of the fourth candidate medical resource and the destination of the user's travel information, filter out a fifth candidate medical resource with the shortest travel time from the fourth candidate medical resource as the medical resource for consultation.

[0160] In one embodiment of this disclosure, the determining module 1004 is further configured to: prioritize the user's travel information in response to a conflict between the outpatient time of the first candidate medical resource and the user's travel information; determine the first user's travel information that can be adjusted based on the priority; and determine the first candidate medical resource that does not conflict with the first user's travel information as the medical resource for treatment.

[0161] In one embodiment of this disclosure, the determining module 1004 is further configured to: register and make appointments for the medical resources for treatment, and provide feedback of the medical reference information and the target symptom profile data to the medical resources for treatment.

[0162] In one embodiment of this disclosure, the consultation guidance module 1001 is further configured to: receive the first input natural language symptom description information, and perform intent recognition on the first input symptom description information to determine the consultation department and the type of disease to be consulted; and determine the target knowledge base to be invoked and the corresponding consultation guidance script based on the consultation department.

[0163] In one embodiment of this disclosure, the consultation guidance module 1001 is further configured to: extract symptom information from the natural language symptom description information input in the i-th round based on the consultation agent, so as to update the symptom profile data; generate a first guidance question for the (i+1)-th round based on the updated symptom profile data and the candidate knowledge information retrieved in the i-th round; optimize the first guidance question based on the corresponding dialogue for consultation guidance, generate a second guidance question for the (i+1)-th round, and display it on the front end to obtain the natural language symptom description information for the (i+1)-th round; repeat the above process until the preset consultation guidance end condition is met to obtain the target symptom profile data.

[0164] According to the agent-based consultation device provided in this disclosure, multiple rounds of consultation guidance are conducted based on the consultation agent to obtain target symptom profile data. During the consultation process of the consultation agent, a retrieval engine is used to search the target knowledge base based on the content obtained from the consultation guidance to obtain target knowledge information. This allows the diagnostic agent to perform a diagnosis based on the target symptom profile data and target knowledge information, obtaining diagnostic reference information. In response to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, a decision-making agent is invoked to determine the appropriate medical resources based on the diagnostic reference information and user information. Therefore, this solution improves the professionalism and accuracy of consultation by using an agent for consultation and combining it with a target knowledge base for diagnosis. Multiple rounds of consultation guidance during the consultation phase reduce communication barriers and improve the completeness of symptom collection, further enhancing diagnostic accuracy. Online consultations can reduce medical costs and alleviate pressure on offline medical resources, and by guiding users to seek medical treatment through relevant medical resources, it improves medical efficiency and user experience.

[0165] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0166] The examples in this disclosure may involve user data, data acquisition, and / or use. All of these aspects comply with relevant laws, regulations, and provisions. In the examples of this disclosure, all data collection, acquisition, processing, manipulation, forwarding, and use are conducted with the user's knowledge and confirmation. Accordingly, when implementing each example, the type, scope of use, and usage scenarios of any data or information that may be involved should be communicated to the user and their authorization obtained through appropriate means in accordance with relevant laws and regulations. The specific methods of notification and / or authorization can vary depending on the actual situation and application scenario, and the scope of the solution is not limited in this regard.

[0167] In this manual and the sample solutions, any processing of personal information will be conducted only under legal grounds (such as obtaining the consent of the data subject or being necessary for the performance of a contract) and will only be carried out within the scope stipulated or agreed upon. A user's refusal to process personal information beyond what is necessary for basic functions will not affect the user's use of basic functions.

[0168] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0169] Figure 11 A schematic block diagram of an example electronic device 1100 that can be used to implement embodiments 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 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0170] like Figure 11 As shown, device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes based on computer programs / instructions stored in read-only memory (ROM) 1102 or loaded from storage unit 1106 into random access memory (RAM) 1103. The RAM 1103 may also store various programs and data required for the operation of device 1100. The computing unit 1101, ROM 1102, and RAM 1103 are interconnected via bus 1104. Input / output (I / O) interface 1105 is also connected to bus 1104.

[0171] Multiple components in device 1100 are connected to I / O interface 1105, including: input units 1106 such as keyboard, mouse, etc.; output units 1107 such as various types of displays, speakers, etc.; storage units 1108 such as disks, optical disks, etc.; and communication units 1109 such as network cards, modems, wireless transceivers, etc. Communication unit 1109 allows device 1100 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0172] The computing unit 1101 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the various methods and processes described above, such as agent-based diagnostic methods. For example, in some embodiments, the agent-based diagnostic method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1106. In some embodiments, part or all of the computer program / instructions can be loaded and / or installed on device 1100 via ROM 1102 and / or communication unit 1109. When the computer program / instructions are loaded into RAM 1103 and executed by the computing unit 1101, one or more steps of the agent-based diagnostic method described above can be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform an agent-based diagnostic method by any other suitable means (e.g., by means of firmware).

[0173] Various implementations 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-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs / instructions that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0174] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0175] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0176] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, 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 sound input, voice input, or tactile input).

[0177] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0178] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. The client-server relationship is created by computer programs / instructions running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0179] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in the disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this document does not impose any restrictions.

[0180] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A diagnostic method based on intelligent agents, wherein, The method includes: Multiple rounds of consultation guidance are conducted based on the intelligent consultation agent to obtain target symptom profile data; During the consultation process of the consultation intelligent agent, the target knowledge information is obtained by searching the target knowledge base based on the content obtained by the consultation guidance using a retrieval engine. Based on the diagnostic agent's diagnosis using the target symptom profile data and the target knowledge information, diagnostic reference information is obtained. In response to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, the decision-making agent is invoked to determine the medical resources for seeking medical treatment based on the diagnostic reference information and user information.

2. The method according to claim 1, wherein, The decision-making agent determines the medical resources for treatment based on the diagnostic reference information and user information, including: Based on the diagnostic reference information, the decision-making agent constructs search keywords, which include at least the target disease type and / or the target department to which the target disease type belongs; Medical resources are retrieved based on the search keywords and the user information, and the medical resources for treatment are determined from the first candidate medical resources retrieved.

3. The method according to claim 2, wherein, The user information includes at least user location information. The step of retrieving medical resources based on the search keywords and the user information, and determining the medical resource for treatment from the first candidate medical resources retrieved, includes: Based on the user location information, the resource retrieval range is determined, and medical resources are retrieved within the resource retrieval range based on the search keywords to obtain the first candidate medical resource; Determine the route between the first candidate medical resource and the user's location information; Based on the travel time and distance of the route, the medical resource for treatment is determined from the first candidate medical resources.

4. The method according to claim 3, wherein, The step of determining the medical resource for treatment from the first candidate medical resources based on the travel time and distance of the route includes: Obtain comprehensive evaluation data of the first candidate medical resource; The first candidate medical resource is scored based on the travel time of the route, the travel distance, and the comprehensive evaluation data. The medical resource for treatment is determined from the first candidate medical resources based on the score.

5. The method according to claim 3, wherein, The diagnostic reference information also includes symptom levels, and the medical resource retrieval based on the search keywords and user information includes: Obtain medical resources associated with the target disease type; Determine the resource level of the associated medical resources; Based on the symptom level and the resource level, the associated medical resources are screened to obtain the first candidate medical resource.

6. The method according to claim 2 or 3, wherein, The user information includes at least user travel information, and determining the medical resource from the retrieved first candidate medical resources includes: Obtain the outpatient information of each candidate doctor in the first candidate medical resources; Based on the outpatient information and the user's travel information, the medical resource for treatment is determined from the first candidate medical resources.

7. The method according to claim 6, wherein, The step of determining the medical resource for treatment from the first candidate medical resources based on the outpatient information and the user's travel information includes: Perform time conflict checks on the outpatient information and the user's travel information to identify a second candidate medical resource with a time conflict; Remove the second candidate medical resource from the first candidate medical resource to obtain the third candidate medical resource; The medical resource for treatment is determined from the third candidate medical resources.

8. The method according to claim 7, wherein, The step of determining the medical resource for treatment from the third candidate medical resources includes: From the third candidate medical resources, a fourth candidate medical resource is selected whose consultation information is on the same day as the user's travel information; Based on the location information of the fourth candidate medical resource and the destination of the user's travel information, the fifth candidate medical resource with the shortest travel time is selected from the fourth candidate medical resources and used as the medical resource for treatment.

9. The method according to claim 7, wherein, The method further includes: In response to the conflict between the outpatient time of the first candidate medical resource and the user's travel information, the user's travel information is prioritized. Based on priority, determine the first user's trip information that can be adjusted; A first candidate medical resource that does not conflict with the first user's travel information is identified as the medical resource for the patient's visit.

10. The method according to claim 1 or 2, wherein, After determining the available medical resources, the process also includes: The system allows users to register and make appointments with the medical resources, and then feeds back the consultation reference information and the target symptom profile data to the medical resources.

11. The method according to claim 1 or 2, wherein, The method further includes: Receive the initial input of natural language symptom description information, and perform intent recognition on the initial input symptom description information to determine the consultation department and the type of disease to be consulted; Based on the department being consulted, determine the target knowledge base to be accessed and the corresponding consultation guidance script.

12. The method according to claim 1 or 2, wherein, The method of guiding multiple rounds of consultation based on a consultation intelligence agent to obtain target symptom profile data includes: Based on the aforementioned intelligent consultation agent, symptom information is extracted from the natural language symptom description information input in the i-th round to update the symptom profile data; Based on the updated symptom profile data and the candidate knowledge information retrieved in the i-th round, the first guiding question for the (i+1)-th round is generated; Based on the corresponding dialogue for the consultation guidance, the first guidance question is optimized to generate the second guidance question for the (i+1)th round, and displayed on the front end to obtain the natural language symptom description information for the (i+1)th round. Repeat the above process until the preset end conditions for the consultation guidance are met, and obtain the target symptom profile data.

13. A diagnostic device based on intelligent agents, wherein, The device includes: The consultation guidance module is used to guide multiple rounds of consultations based on the consultation intelligence agent in order to obtain target symptom profile data; The retrieval module is used to retrieve target knowledge information from the target knowledge base based on the content obtained according to the consultation guidance during the consultation process of the consultation intelligent agent. The diagnostic module is used to perform a diagnosis based on the target symptom profile data and the target knowledge information by the diagnostic intelligent agent, and obtain diagnostic reference information. The determination module is used to respond to the diagnostic reference information indicating that the conditions for seeking medical treatment are met, and to invoke the decision-making agent to determine the medical resources for seeking medical treatment based on the diagnostic reference information and user information.

14. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-12.

15. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-12.

16. A computer program product comprising a computer program / instructions, wherein, When the computer program / instructions are executed by the processor, they implement the method of any one of claims 1-12.

Images