Information reply method and device, computer storage medium and electronic equipment

Abstract

The invention relates to the technical field of internet medical treatment, and provides an information reply method, an information reply device, a computer storage medium and electronic equipment.The information reply method comprises the steps that inquiry information from an inquiry end is received, and a pre-configured inquiry mode is detected; the inquiry mode comprises an intelligent response mode and a manual response mode; responding to the inquiry mode as an intelligent response mode, and outputting reply information corresponding to the inquiry information through a trained inquiry reply model; and in response to the inquiry mode being a manual response mode, recommending target information to a reception end based on a pre-configured information recommendation strategy, so that the reception end determines reply information corresponding to the inquiry information based on the target information. In the scene that the same doctor faces a plurality of interrogation objects at the same moment, the doctor can conveniently and quickly provide the interrogation reply to each interrogation object, and the online interrogation efficiency is improved.

Description

Background Art

[0002] Online consultation service is one of the important ways for medical institutions to develop online services and information transformation, and supports a variety of different types of consultation forms. Whether for institutions, doctors, or patients, it is a way to meet the needs of personnel services in different scenarios at different times, which is very convenient and flexible.

[0003] However, in a scenario where the same doctor faces multiple patients in the same period of time, the timeliness of the response to each patient cannot be guaranteed.

[0004] In view of this, there is an urgent need in the art to develop an information reply method and device.

[0005] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure. Summary of the invention

[0006] The purpose of the present disclosure is to provide an information reply method, an information reply device, a computer storage medium and an electronic device, thereby overcoming, at least to a certain extent, the technical problem of being unable to guarantee the timeliness of medical consultation due to the limitations of related technologies.

[0007] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0008] According to the first aspect of the present disclosure, there is provided an information reply method, comprising: receiving medical inquiry information from a medical inquiry end, and detecting a preconfigured medical inquiry mode; the medical inquiry mode comprises an intelligent answering mode and a manual answering mode; in response to the medical inquiry mode being an intelligent answering mode, outputting reply information corresponding to the medical inquiry information through a trained medical inquiry reply model; in response to the medical inquiry mode being a manual answering mode, recommending target information to a medical reception end based on a preconfigured information recommendation strategy, so that the medical reception end determines the reply information corresponding to the medical inquiry information based on the target information.

[0009] In an exemplary embodiment of the present disclosure, the medical inquiry information includes the file information of the patient and the medical inquiry channel information of the patient; in response to the medical inquiry mode being the intelligent answering mode, the reply information corresponding to the medical inquiry information is output through a trained medical inquiry reply model, including: determining a reply style for the medical inquiry information according to the medical inquiry channel information; there is a preset correlation between the medical inquiry channel information and the reply style; the medical inquiry information, the file information and the reply style are input into the trained medical inquiry reply model, and the reply information corresponding to the medical inquiry information is determined according to the output information of the medical inquiry reply model.

[0010] In an exemplary embodiment of the present disclosure, after determining the response style for the inquiry information according to the inquiry channel information, the method further includes: obtaining a inquiry template associated with the inquiry channel information; the inquiry channel information and the inquiry template have a preset association relationship; filling the inquiry information, the file information and the response style into the inquiry template to obtain filled information; inputting the filled information into the trained inquiry response model, and determining the response information corresponding to the inquiry information according to the output information of the inquiry response model.

[0011] In an exemplary embodiment of the present disclosure, after obtaining the output information of the medical inquiry reply model, the method further includes: filtering the output information of the medical inquiry reply model for sensitive words to obtain reply information corresponding to the medical inquiry information.

[0012] In an exemplary embodiment of the present disclosure, in response to the consultation mode being the manual answering mode, the target information is recommended to the reception end based on a preconfigured information recommendation strategy, including: detecting whether an input keyword from the reception end is received; in response to receiving the input keyword from the reception end, recommending the target information to the reception end based on a first information recommendation strategy; in response to not receiving the input keyword from the reception end, recommending the target information to the reception end based on a second information recommendation strategy.

[0013] In an exemplary embodiment of the present disclosure, the reception end corresponds to a receiving doctor, and the recommending target information to the reception end based on the first information recommendation strategy includes: performing information processing of at least one dimension on the input keyword to obtain the target input keyword corresponding to each dimension of the at least one dimension; querying a preconfigured public information library and / or a private information set of the receiving doctor based on the target input keyword and the department information to which the receiving doctor belongs to obtain a query result set corresponding to each dimension; and filtering to obtain the target information from at least one query result set corresponding to the at least one dimension.

[0014] In an exemplary embodiment of the present disclosure, the information processing of at least one dimension includes at least one of the following: information completion processing, word segmentation processing, and error correction processing.

[0015] In an exemplary embodiment of the present disclosure, the public information database contains multiple sample reply information and a label corresponding to each of the sample reply information, and the label contains applicable department information of the sample reply information; the pre-configured public information database and / or the private information set of the attending doctor are queried according to the target input keyword and the department information to which the attending doctor belongs, to obtain the query result set corresponding to each dimension, including: searching the public information database according to the department information to which the attending doctor belongs and the label corresponding to each of the sample reply information to obtain a feature information set; querying the feature information set and / or the private information set according to the target input keyword to obtain the query result set corresponding to each dimension.

[0016] In an exemplary embodiment of the present disclosure, when the at least one dimension contains only one dimension, filtering to obtain the target information from at least one query result set corresponding to the at least one dimension includes: filtering out a preset number of target information from the query result set according to the reference frequency corresponding to each piece of information contained in the query result set.

[0017] In an exemplary embodiment of the present disclosure, when the query result set includes a first query result set determined according to a feature information set and a second query result set determined according to the private information set, filtering out a preset number of target information from the query result set according to the reference frequency corresponding to each piece of information contained in the query result set includes: splitting the preset number of pieces into a first part of pieces and a second part of pieces according to a preset weight coefficient; filtering out the first part of target information from the first query result set according to the reference frequency of each piece of information contained in the first query result set; and filtering out the second part of target information from the second query result set according to the reference frequency of each piece of information contained in the second query result set.

[0018] In an exemplary embodiment of the present disclosure, when the at least one dimension includes two or more dimensions, filtering the target information from at least one query result set corresponding to the at least one dimension includes: obtaining a priority order pre-set for each of the query result sets; and filtering a preset number of the target information from the at least one query result set according to the priority order.

[0019] In an exemplary embodiment of the present disclosure, the target information is recommended to the reception end based on the second information recommendation strategy, including: filtering the public information library according to the department information of the receiving doctor and the label corresponding to each of the sample reply information to obtain a feature information set; filtering out a preset number of target information from the feature information set and / or the private information set according to the reference frequency of each piece of information contained in the feature information set and / or the private information set.

[0020] According to a second aspect of the present disclosure, there is provided an information reply device, comprising: a mode detection module, for receiving medical consultation information from a medical consultation end, and detecting a preconfigured medical consultation mode; the medical consultation mode comprises an intelligent answering mode and a manual answering mode; an information output module, for, in response to the medical consultation mode being an intelligent answering mode, outputting reply information corresponding to the medical consultation information through a trained medical consultation reply model; an information recommendation module, for, in response to the medical consultation mode being a manual answering mode, recommending target information to a medical consultation end based on a preconfigured information recommendation strategy, so that the medical consultation end determines the reply information corresponding to the medical consultation information based on the target information.

[0021] According to a third aspect of the present disclosure, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the information reply method described in the first aspect is implemented.

[0022] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the information reply method described in the first aspect above by executing the executable instructions.

[0023] It can be seen from the above technical solutions that the information reply method, information reply device, computer storage medium and electronic device in the exemplary embodiments of the present disclosure have at least the following advantages and positive effects:

[0024] In the technical solutions provided in some embodiments of the present disclosure, by receiving the inquiry information from the inquiry end, detecting the pre-configured inquiry mode, and in response to the inquiry mode being the intelligent answering mode, outputting the reply information corresponding to the inquiry information through the trained inquiry reply model, it is possible to quickly reply to the inquiry information of the inquiry object without manual consultation or only requiring a small amount of manual work, and improve the doctor's consultation efficiency in the scenario where the same doctor faces multiple inquiry objects at the same time. In response to the inquiry mode being the manual answering mode, the target information is recommended to the consultation end based on the pre-configured information recommendation strategy, so that the consultation end determines the reply information corresponding to the inquiry information based on the target information, and can quickly provide the receiving doctor with a reply reference, solving the technical problems of the difficulty of searching and low consultation efficiency caused by manual search or input of reply information in the related technology, thereby further improving the timeliness and consultation efficiency of online consultation.

[0025] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0027] Figure 1 A schematic diagram showing a flow chart of an information reply method in an embodiment of the present disclosure;

[0028] Figure 2 A flow chart showing how to output reply information corresponding to the medical inquiry information through a trained medical inquiry reply model in an intelligent answering mode in an embodiment of the present disclosure;

[0029] Figure 3 A schematic diagram showing another process of outputting reply information corresponding to the medical inquiry information through a trained medical inquiry reply model in an intelligent answering mode in an embodiment of the present disclosure;

[0030] Figure 4 A schematic diagram showing a process of recommending target information to a receiving terminal based on a preconfigured information recommendation strategy in an embodiment of the present disclosure;

[0031] Figure 5 A schematic diagram showing a flow chart of how to recommend target information to the above-mentioned reception terminal based on the first information recommendation strategy when receiving an input keyword from the reception terminal in the embodiment of the present disclosure;

[0032] Figure 6 A schematic diagram showing a flow chart of how to obtain a query result set corresponding to each dimension in an embodiment of the present disclosure;

[0033] Figure 7 A flow chart showing how to recommend target information to the receiving terminal based on the second information recommendation strategy when no input keyword from the receiving terminal is received in the embodiment of the present disclosure;

[0034] Figure 8 A functional schematic diagram of the intelligent reply system in an embodiment of the present disclosure is shown;

[0035] Fig. 9 The overall architecture diagram of the intelligent medical consultation system involved in the embodiments of the present disclosure is shown;

[0036] Fig.10 A schematic diagram showing the structure of an information reply device in an exemplary embodiment of the present disclosure;

[0037] Fig.11 A schematic structural diagram of an electronic device in an exemplary embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0038] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0039] The terms "a", "an", "the" and "said" are used in this specification to indicate the presence of one or more elements / components / etc.; the terms "including" and "having" are used to express an open-ended inclusion and mean that additional elements / components / etc. may exist in addition to the listed elements / components / etc.; the terms "first" and "second" etc. are used only as labels and are not intended to limit the quantity of their objects.

[0040] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and their repeated descriptions will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities.

[0041] In recent years, with the rapid development of online medical consultations, more and more doctors have chosen to provide consultation services online. The provision of this service has enabled more and more patients to receive medical consultations online.

[0042] During online consultations, information exchange between doctors and patients is mainly in text form. Therefore, the speed and quality of doctors' responses to patients greatly affect the patients' consultation experience. At the same time, in a scenario where a doctor communicates with multiple patients at the same time, due to the doctor's personal energy, it is impossible to quickly provide effective responses to each patient, resulting in patients waiting too long, refunds, and bad reviews. This not only affects the patient's consultation experience, but is also not conducive to the doctor's consultation evaluation. Therefore, it is very necessary to improve the doctor's response speed and achieve quick responses.

[0043] Quick replies have the following functions in the doctor-patient communication process:

[0044] First, it saves time. Doctors can quickly answer common questions or provide general advice through quick replies. This saves doctors time and allows them to process large amounts of information more quickly.

[0045] Second, improve efficiency. Quick replies can help doctors answer patients' questions more quickly, thereby improving work efficiency;

[0046] Third, maintain consistency. Doctors can use pre-configured responses to ensure that consistent information and advice is provided to all patients, thus avoiding omissions or misunderstandings;

[0047] Fourth, reduce errors. When doctors need to answer the same questions repeatedly, using the copy and paste function can reduce manual errors and spelling errors;

[0048] Fifth, improve satisfaction. Quick replies can help doctors better handle patients’ needs, improve patient satisfaction, and thus enhance the doctor-patient relationship;

[0049] Sixth, enhance the unique advantages of online consultation. Quick replies are not simple replies for daily chats. To set up quick replies, it is necessary to ensure the characteristics of Internet medical care and have a certain degree of professionalism and accuracy.

[0050] In short, quick replies can improve efficiency, reduce errors, and increase satisfaction in the doctor-patient communication process, providing better services for doctors and patients.

[0051] In the related art, quick reply is generally achieved through the following three solutions:

[0052] The first is copying and pasting text. When doctors need to answer the same question repeatedly, they can use the copy and paste function to quickly send the written answer. They just need to make sure it matches the current patient's condition before sending the text.

[0053] However, this solution has the following problems:

[0054] In practice, doctors are usually required to summarize past consultations in advance, organize responses to different questions, and manually judge and search based on the questions to find the appropriate responses. Obviously, this solution will consume too much energy of doctors. In addition, due to the limitations of doctors' personal experience, the responses compiled by doctors can only cover a limited number of scenarios, and most cases still require manual responses.

[0055] The second is that the platform provides unified responses. The platform presets and stores responses to some common questions, and all doctors can share the platform's quick responses.

[0056] However, this solution has the following problems:

[0057] Since the quick replies on the platform are not broken down by department, most of the quick replies are invalid and unusable for individual doctors. Doctors need to spend a lot of energy searching and screening to find available replies. In extreme cases, the time and energy spent on searching and screening may be higher than the time and energy spent on directly entering the answers.

[0058] The third method is to pre-create responses to common questions. Doctors can prepare answers to some common questions in advance and store them as preset responses. In this way, when patients ask similar questions to doctors, doctors only need to select the corresponding preset responses.

[0059] However, this solution has the following problems:

[0060] When doctors have hundreds or even thousands of quick replies preset, it still takes a long time for doctors to select a specific quick reply. At the same time, since each doctor can only use the quick replies they have set, they cannot use other doctors' high-quality quick replies. In addition, since doctors cannot compare with the quick replies set by other doctors, they cannot find the shortcomings of their own reply words and cannot make improvements.

[0061] In addition, the above three solutions still have the following problems:

[0062] On the one hand, it is impossible to continuously enrich the shared quick replies, and the platform can only maintain the public quick replies and open them to all users, which makes the public quick replies too monotonous and slow to update. On the other hand, it is impossible to prompt doctors to reply based on patient information, problem description, etc., and it is impossible to quickly search and find the quick replies available to doctors. When there are a large number of available quick replies, it becomes more difficult to find quick replies.

[0063] In the embodiments of the present disclosure, firstly, an information reply method is provided, which at least to some extent overcomes the defect in the related art that the timeliness of the consultation cannot be guaranteed.

[0064] Figure 1 A flow chart of the information reply method in the embodiment of the present disclosure is shown. The execution subject of the information reply method may be a server that performs consultation processing.

[0065] refer to Figure 1 According to an embodiment of the present disclosure, the information reply method includes the following steps:

[0066] Step S110, receiving the inquiry information from the inquiry terminal, and detecting the pre-configured inquiry mode; the inquiry mode includes an intelligent answering mode and a manual answering mode;

[0067] Step S120, in response to the medical inquiry mode being the intelligent answering mode, outputting reply information corresponding to the medical inquiry information through the trained medical inquiry reply model;

[0068] Step S130, in response to the consultation mode being the manual answering mode, recommending target information to the reception end based on a preconfigured information recommendation strategy, so that the reception end determines reply information corresponding to the consultation information based on the target information.

[0069] exist Figure 1In the technical solution provided by the illustrated embodiment, by receiving the inquiry information from the inquiry end, detecting the pre-configured inquiry mode, and responding to the inquiry mode being the intelligent answering mode, the reply information corresponding to the inquiry information is output through the trained inquiry reply model, so that the inquiry information of the inquiry object can be quickly replied without manual consultation or only a small amount of manual work is required, and the doctor's consultation efficiency is improved in the scenario where the same doctor faces multiple inquiry objects at the same time. In response to the inquiry mode being the manual answering mode, the target information is recommended to the consultation end based on the pre-configured information recommendation strategy, so that the consultation end determines the reply information corresponding to the inquiry information based on the target information, and can quickly provide a reply reference for the receiving doctor, solving the technical problems of the difficulty of searching and low consultation efficiency caused by manual search or input of reply information in the related technology, thereby further improving the timeliness and consultation efficiency of online consultation.

[0070] The following Figure 1 The specific implementation process of each step is described in detail:

[0071] Before step S110, it should be noted that the present disclosure may pre-configure a public information database, which may contain multiple sample reply messages and labels corresponding to each sample reply message, and the label may contain the department information of the source doctor of the sample reply message (i.e., from which attending doctor the information was obtained). After the public information database is configured, all attending doctors connected to the intelligent reply system can view or use the relevant sample reply information from the public information database.

[0072] The specific contents contained in the above public information library are described below with reference to Table 1. As shown in Table 1, the public information library may contain the following fields (identification Id, content Content, contributor Contributor, citation frequency Num_of_quote, label Label and type Type):

[0073] Table 1

[0074]

[0075] Referring to Table 1, Id represents the identifier of each sample reply information, which is used to distinguish multiple sample reply information; Content represents the specific content of each sample reply information, Contributor represents the contributor of each sample reply information (i.e., the source doctor), Num_of_quote represents the citation frequency of the sample consultation reply information, Label represents the label of each sample reply information, and Type represents the type of each sample reply information. For example, Type 1 represents that the platform provides it, and Type 2 represents that the doctor publishes it independently (i.e., the present disclosure supports the attending doctor to publish the customized reply information in the private information set to the public information library).

[0076] Exemplarily, the present disclosure may perform real-time or timed expansion of the above public information database in the following manners:

[0077] Acquire the reply information sent by each attending doctor connected to the intelligent consultation system in real time or at a scheduled time, and then perform word segmentation on each reply information (word segmentation is to divide the continuous text into independent words) to obtain multiple segmented words. Furthermore, since diseases and drugs are the main information in the process of online consultation and communication between doctors and patients, in order to facilitate the subsequent rapid retrieval of relevant information, each segmented word can be matched with a preset tag library, etc., to obtain the label corresponding to each segmented word, so as to realize the rapid labeling of the segmented word. Furthermore, the applicable department information of the reply information can be obtained. Finally, the label corresponding to the above reply information can be determined based on the labels corresponding to the above multiple segmented words and the applicable department information of the above reply information.

[0078] Among them, the above-mentioned preset tag library can include a disease name library, a drug name library, etc., which can be set according to actual conditions, and the present disclosure does not make special restrictions on this. By marking the above-mentioned reply information, it can be convenient for subsequent doctors to better use the search function, improve the information recommendation rate, be more in line with the characteristics of Internet diagnosis and treatment, and meet the business needs of Internet diagnosis and treatment scenarios.

[0079] Furthermore, the above reply information can be used as sample reply information, and each sample reply information and its corresponding label can be stored in ES (Elasticsearch, a free and open distributed search and analysis engine suitable for all types of data including text, numbers, geospatial, structured and unstructured data, etc., which can provide word segmentation, suggest retrieval, associated query, and complete query functions) to achieve the expansion of the above public information library.

[0080] In addition, the attending doctor can also configure his / her own customized reply information and the label corresponding to each customized reply information according to his / her professional habits to form a private information library. The private information library can contain multiple customized reply information. Each attending doctor can only view or use the content configured by himself / herself in the private information library (i.e., his / her own private information set), but cannot view the content configured by other attending doctors.

[0081] The following is a description of the contents contained in the private information library in the present disclosure in conjunction with Table 2. As shown in Table 2, the private information library may include the following fields (ID, Content, Is_published, Label, Type, Usage_count):

[0082] Table 2

[0083]

[0084]

[0085] Id is the identifier of each custom reply message, which is used to distinguish multiple custom reply messages. Content represents the specific content of the consultation reply message. Ower represents the owner of each consultation reply message (the private information set of each doctor can be divided according to the Ower item). Is_published represents whether it has been published to the public information library. Label represents the label of each custom reply message. Type represents the type of each consultation reply message. For example, type 1 represents custom and type 2 represents favorite. Usage_count represents the frequency of use.

[0086] After configuring the above public information base and / or private information base, refer to Figure 1 In step S110, the consultation information from the consultation terminal is received and the pre-configured consultation mode is detected.

[0087] In this step, when the intelligent medical consultation system receives medical consultation information from the medical consultation terminal, the pre-configured medical consultation mode can be detected.

[0088] Exemplarily, the preconfigured consultation mode can be selected by the consultation end or the reception end, and can be set according to actual conditions. The present disclosure does not make any special limitation on this.

[0089] Exemplarily, the above-mentioned consultation mode may include an intelligent response mode and a manual response mode. The intelligent response mode is that the intelligent consultation system calls the trained consultation response model to intelligently respond to the consultation information sent by the above-mentioned consultation terminal, while the manual response mode is that the attending doctor himself responds to the consultation information sent by the consultation terminal. In step S120, in response to the consultation mode being the intelligent response mode, the trained consultation response model outputs the response information corresponding to the consultation information.

[0090] In this step, it should be noted that the above-mentioned consultation information may include the profile information of the consultation object and the consultation channel information of the above-mentioned consultation object. It should be noted that the consultation object may be the target user who is currently conducting the consultation operation on the consultation end, or it may be an associated user of the target user, such as: his family, friends, etc., which can be set according to the actual situation, and the present disclosure does not make any special restrictions on this. Among them, the profile information may include information such as age, gender, marital status, etc. (for example: 24 years old, female, unmarried), which can be set according to the actual situation, and the present disclosure does not make any special restrictions on this. The above-mentioned consultation channel information refers to the channel through which the above-mentioned consultation object accesses the above-mentioned intelligent consultation system. Exemplarily, the consultation channel may include mini-programs, web pages, APP (short for Application in English, generally refers to mobile phone software), etc., which can be set according to the actual situation, and the present disclosure does not make any special restrictions on this.

[0091] refer to Figure 2 , Figure 2 A flow chart showing how to output reply information corresponding to the medical inquiry information through a trained medical inquiry reply model in an intelligent answering mode in an embodiment of the present disclosure includes steps S201 and S202:

[0092] In step S201, a reply style for the consultation information is determined according to the consultation channel information.

[0093] In this step, the response style refers to the type or manner in which the response information for the above-mentioned medical consultation information is organized. Exemplarily, the response style may include: professional and rigorous, easy to understand, detailed and meticulous, etc. It can be set according to the actual situation, and the present disclosure does not make any special limitation on this.

[0094] There is a preset correlation between the inquiry channel information and the reply style. For example, the preset correlation between the inquiry channel information and the reply style can be obtained. For example, the preset correlation can be set as follows: inquiry channel 1-professional and rigorous; inquiry channel 2-easy to understand; inquiry channel 3-detailed and meticulous. The specific preset correlation can be set according to the actual situation, and the present disclosure does not make any special limitation on this.

[0095] After obtaining the preset association relationship between the above-mentioned consultation channel information and the response style, the response style for the consultation object's consultation information can be determined according to the above-mentioned preset association relationship.

[0096] After determining the reply style, the process proceeds to step S202 to input the medical inquiry information, file information and reply style into a trained medical inquiry reply model, and determine reply information corresponding to the medical inquiry information based on output information of the medical inquiry reply model.

[0097] In this step, the above-mentioned medical inquiry information, the above-mentioned file information and the above-mentioned reply style can be input into the trained medical inquiry reply model, and then, the output information of the above-mentioned medical inquiry reply model is obtained, and the output information of the medical inquiry reply model is determined as the reply information corresponding to the above-mentioned medical inquiry information.

[0098] Among them, the above-mentioned consultation response model can be a generative AI model (Artificial Intelligence). Generative AI is a specific type of AI that focuses on generating new content, such as text, images, and music. These systems are trained on large data sets and use machine learning algorithms to generate new content similar to the training data. Exemplary ones can be ChatGPT (Chat Generative Pre-trained Transformer, a natural language processing tool based on artificial intelligence), Wenxin Yiyan, etc., which can be set according to actual conditions, and this disclosure does not make special restrictions on this.

[0099] Optionally, after the above step S201, the present disclosure further provides another flow chart of determining the reply information corresponding to the medical inquiry information, referring to Figure 3 , Figure 3 A flowchart showing how to output reply information corresponding to the medical inquiry information through a trained medical inquiry reply model in another embodiment of the present disclosure in the intelligent answering mode includes steps S301 to S303:

[0100] In step S301, a consultation template associated with consultation channel information is obtained.

[0101] In this step, there is a preset association relationship between the consultation channel information and the consultation template. Thus, the association relationship between the preset consultation channel information and the consultation template can also be obtained. For example, the preset association relationship is set as follows: consultation channel 1-"Use #{} method to answer the following question for #{} patients: #{}"; consultation channel 2-"A #{} patient has a question: #{}, answer it in #{} method". The specific preset association relationship can be set according to the actual situation, and the present disclosure does not make any special restrictions on this.

[0102] After obtaining the consultation template associated with the consultation channel information, the process may proceed to step S302 to fill the consultation information, the archive information and the reply style into the consultation template to obtain filled information.

[0103] In this step, the above-mentioned medical inquiry information, the above-mentioned file information and the above-mentioned reply style may be filled into the above-mentioned medical inquiry template to obtain filled information.

[0104] For example, when the consultation information of the above-mentioned consultation subject is "What medicine should I take for stomachache", the consultation channel is channel 1, the profile information is 24 years old, female, unmarried, and the response style is "professional and rigorous", the above information is filled into the consultation template "Use #{} method to answer the following question for #{} patients: #{}", and the filled-in information obtained is: "Use #{professional and rigorous} method to answer the following question for #{24 years old, female, unmarried} patients: #{What medicine should I take for stomachache}".

[0105] After obtaining the above-mentioned filling information, the process may proceed to step S303 to input the filling information into the trained medical inquiry reply model, and determine the reply information corresponding to the medical inquiry information according to the output information of the medical inquiry reply model.

[0106] In this step, the above-mentioned filled-in information can be input into the trained medical consultation reply model so that the above-mentioned filled-in information can be identified and processed by the medical consultation reply model to obtain the output information of the medical consultation reply model, and the output information of the medical consultation reply model can be determined as the reply information corresponding to the above-mentioned medical consultation information.

[0107] Optionally, after obtaining the output information of the consultation reply model through the above step S202 or step S303, the above output information can also be filtered for sensitive words to obtain the reply information corresponding to the above consultation information. Exemplarily, sensitive words can be words with sensitive political tendencies, violent tendencies, unhealthy colors or uncivilized words, or special words pre-set according to the Internet medical consultation scenario, which can be set according to actual conditions, and the present disclosure does not make special restrictions on this.

[0108] After receiving the reply information, the reply information can be directly returned to the inquiry end, or the reply information can be returned to the reception end, so that the receiving doctor at the reception end can simply modify the reply information and then return it to the inquiry end. Thus, the present disclosure can realize the rapid reply of the inquiry information of the inquiry object without manual reception or with only a small amount of manual work.

[0109] In step S130, in response to the consultation mode being the manual answering mode, target information is recommended to the reception end based on a preconfigured information recommendation strategy, so that the reception end determines reply information corresponding to the consultation information based on the target information.

[0110] In this step, when the above-mentioned consultation mode is the manual response mode, the consulting doctor needs to input the response information by himself. In order to improve the response efficiency of the consulting doctor and avoid the consulting doctor inputting a complete response statement as much as possible, the present disclosure can recommend target information to the above-mentioned consulting end based on a pre-configured information recommendation strategy, so that the consulting end can quickly determine the response information corresponding to the above-mentioned consultation information based on the above-mentioned target information.

[0111] Exemplarily, refer to Figure 4 , Figure 4 FIG. shows a flowchart of how to recommend target information to the consulting end based on a pre-configured information recommendation strategy in an embodiment of the present disclosure, including step S401-step S403:

[0112] In step S401, it is detected whether an input keyword from the consulting end is received.

[0113] In this step, it can be detected whether an input keyword from the consulting end is received.

[0114] In step S402, in response to receiving the input keyword from the consulting end, target information is recommended to the consulting end based on the first information recommendation strategy.

[0115] In this step, if the above-mentioned input keyword is received, target information can be recommended to the above-mentioned consulting end based on the first information recommendation strategy.

[0116] Specifically, refer to Figure 5 , Figure 5 FIG. shows a flowchart of how to recommend target information to the above-mentioned consulting end based on the first information recommendation strategy when an input keyword from the consulting end is received in an embodiment of the present disclosure, including step S501-step S503:

[0117] In step S501, information processing is performed on the input keyword in at least one dimension to obtain target input keywords corresponding to each dimension in at least one dimension.

[0118] In this step, exemplarily, the following at least one dimension of information processing can be performed on the above-mentioned input keyword to obtain at least one target input keyword:

[0119] The first dimension (exemplarily, it can be named completion suggester): Information completion processing. Similar to information expansion, for example: when the consulting doctor inputs "amo", it can be automatically completed to "amoxicillin" to complete the information of the input keyword.

[0120] The second dimension (exemplarily, it can be named label query): Word segmentation processing. For reference, see the above relevant explanations and will not be elaborated here.

[0121] The third dimension (exemplarily, it can be named phase suggester): word segmentation and error correction processing, which can be word segmentation first and then error correction, or error correction first and then word segmentation. It can be set according to the actual situation, and the present disclosure does not make any special limitations on this.

[0122] The fourth dimension (exemplarily, it can be named as term suggester): word segmentation, error correction and information completion processing. Similarly, the sequence of these three processing steps can also be set according to actual conditions, and the present disclosure does not make any special limitations on this.

[0123] In step S502, based on the target input keyword and the department information to which the receiving doctor belongs, the pre-configured public information database and / or the private information set associated with the receiving end are queried to obtain the query result set corresponding to each dimension.

[0124] In this step, after obtaining the target input keywords corresponding to each dimension, the pre-configured public information database and / or the private information set associated with the receiving end can be queried based on the target input keywords and the department information to which the receiving doctor belongs to obtain the query result set corresponding to each dimension.

[0125] Specifically, refer to Figure 6 , Figure 6 A flowchart showing how to obtain a query result set corresponding to each dimension in an embodiment of the present disclosure includes steps S601 and S602:

[0126] In step S601, the public information database is searched according to the department information of the attending doctor and the label corresponding to each sample reply information to obtain a feature information set.

[0127] In this step, considering that each sample reply information in the above-mentioned public information library is provided with a corresponding label, and the label contains the applicable department information of the sample reply information, thus, the sample reply information whose applicable department information in the label is the same as the department information to which the above-mentioned attending doctor belongs can be retrieved from the above-mentioned public information library to form the above-mentioned feature information set.

[0128] In step S602, the characteristic information set and / or the private information set is queried according to the target input keyword to obtain a query result set corresponding to each dimension.

[0129] In this step, keywords can be input according to the above-mentioned target to query the above-mentioned feature information set and / or the private information set of the attending doctor to obtain the query result set corresponding to each dimension, until at least one query result set corresponding to at least one of the above-mentioned dimensions is obtained.

[0130] It should be noted that, if the above-mentioned characteristic information set is queried only according to the above-mentioned target keywords, the query result set obtained will only include the sample response information in the public information database; if the above-mentioned doctor's private information set is queried only according to the above-mentioned target keywords, the query result set obtained will only include the customized sample information in the private information set; and if the above-mentioned characteristic information set and the above-mentioned private information set are queried according to the above-mentioned target keywords, the query result set obtained will include both the sample response information in the public information database and the customized sample information in the private information set.

[0131] Next reference Figure 5 In step S503, target information is obtained by filtering at least one query result set corresponding to at least one dimension.

[0132] In this step, at least one query result set corresponding to the at least one dimension may be screened to obtain the target information.

[0133] In an optional implementation, when the at least one dimension contains only one dimension, a query result set can be obtained. At this time, a preset number of target information can be filtered out from the query result set according to the reference frequency corresponding to each piece of information contained in the query result set.

[0134] Referring to the relevant explanation of the above step S602, it can be known that, given that there are three optional query methods when querying (the first is to query only the feature information set in the public information database, the second is to query only the private information database of the attending doctor, and the third is to query both the public information database and the private information database of the attending doctor), therefore, for the first two query methods, illustratively, taking the preset number of 10 as an example, the information with the top 10 citation frequencies in the query result set can be determined as the above target information. For the third query method, the above query result set will include a first query result set and a second query result set. The first query result set may include sample reply information from a public information database, and the second query result set may include custom reply information from a private information database. Thus, the above preset number of items can be divided into a first part of items and a second part of items according to a preset weight coefficient. For example, assuming that the weight coefficient corresponding to the public information database is 0.2 and the weight coefficient corresponding to the private information database is 0.8, it can be determined that the first part of items is 2 and the second part of items is 8. Thus, the information with the top 2 citation frequencies in the first query result set and the information with the top 8 citation frequencies in the second query result set can be determined as the above target information.

[0135] In an optional implementation, when the at least one dimension includes two or more dimensions, at least two query result sets will be obtained. At this time, the priority order pre-set for each query result set can be obtained, and according to the priority order, a preset number of target information is filtered out from the at least one query result set. For example, assuming that information processing of four dimensions (i.e., the first dimension to the fourth dimension in step S501) is performed in parallel, four query result sets (e.g., the first dimension query result set, the second dimension query result set, the third dimension query result set, and the fourth dimension query result set) are obtained, and the priority order corresponding to the four query result sets is: the first dimension query result set> the second dimension query result set> the third dimension query result set> the fourth dimension query result set, so that it can be first determined whether a preset number of target information (e.g., 10) can be filtered out from the first dimension query result set. If so, the target information can be directly determined without filtering the second dimension query result, the third dimension query result, and the fourth dimension query result. Assuming that there is insufficient data in the first dimension query result set and only 6 pieces of target information can be filtered out, the remaining 4 pieces of target information can be filtered out from the second dimension query result set (or the second dimension query result set combined with the third dimension query result set, or the second dimension query result set combined with the third dimension query result set and the fourth dimension query result set) until the above-mentioned preset number of target information is obtained.

[0136] Next reference Figure 4 In step S403, in response to not receiving the input keyword from the receiving end, target information is recommended to the receiving end based on the second information recommendation strategy.

[0137] In this step, if the input keyword is not received from the receiving end, you can refer to Figure 7 , Figure 7 A flow chart showing how to recommend target information to the patient terminal based on the second information recommendation strategy when the input keyword of the patient terminal is not received in the embodiment of the present disclosure, including steps S701-S702:

[0138] In step S701, the public information database is screened according to the department information of the attending doctor and the label corresponding to each sample reply information to obtain a feature information set.

[0139] In this step, you can refer to the relevant explanation of step S601, which will not be repeated here.

[0140] In step S702, a preset number of target information pieces are screened out from the feature information set and / or the private information set according to the reference frequency of each piece of information contained in the feature information set and / or the private information set.

[0141] In this step, a preset number of target information can be screened out from the characteristic information set and / or the private information set according to the reference frequency of each information contained in the characteristic information set and / or the private information set. The specific information screening method can refer to the relevant explanation of step S503, which will not be repeated here.

[0142] After recommending the target information to the above-mentioned reception end based on the preconfigured information recommendation strategy, illustratively, the receiving doctor at the reception end can quickly filter out the reply information corresponding to the medical inquiry information from the above-mentioned preset number of target information, or, after selecting any one of the above-mentioned target information, simply modify it to generate the reply information corresponding to the medical inquiry information.

[0143] refer to Figure 8 , Figure 8 A functional schematic diagram of the intelligent reply system in an embodiment of the present disclosure is shown as follows: Figure 8 As shown:

[0144] First, the intelligent reply system disclosed in the present invention can realize intelligent response. Specifically, in the online consultation process, the consultation reply model is called mainly through the consultation information of the consultation object, combined with the consultation channel and archival information such as age, gender, marital status and other factors, and the reply information is automatically generated. The doctor can choose to modify or directly use the obtained reply information according to the actual situation, so as to achieve the goal of quickly responding to the consultation;

[0145] Second, the intelligent reply system disclosed in the present invention can realize information recommendation, which is mainly divided into two scenarios. The first category is the unconditional query scenario, that is, aggregating the custom reply information commonly used under the name of the doctor in the private information database and the sample reply information of the same department as the doctor in the public information database, and providing an unconditional query list based on the principle that the more frequently used, the higher the ranking. The second category is the parameter query scenario, that is, using the input keyword as the query condition, performing word segmentation processing on it, and selecting the custom reply information commonly used under the name of the doctor in the private information database and the sample reply information with the same department label as the doctor in the public information database through label precise search combined with correlation fuzzy search, based on the preset priority strategy and weighted algorithm, recommend multiple quick reply lists with high matching degree for the doctor to choose;

[0146] Third, the present disclosure provides collection and publishing functions. Collection means that each attending doctor can collect sample reply information in the public information database, which can be stored in the personal private information set after automatic word segmentation and labeling by the system, so as to add or modify the private information set. Publishing means that the attending doctor can publish the customized reply information contained in his own private information set to the public information database for public disclosure, so that all attending doctors can refer to the published information for information response.

[0147] refer to Fig. 9 , Fig. 9 The overall architecture diagram of the intelligent medical consultation system involved in the embodiments of the present disclosure is shown as follows: Fig. 9 As shown:

[0148] The above intelligent medical consultation system may include two main modules, namely, an intelligent question and answer module 901 and a quick reply module 902, wherein:

[0149] The intelligent question-answering module 901 may include an AI parameter processor 9011, a generative AI 9012, and an AI result processor 9013. Among them, the AI ​​parameter processor 9011 may determine the reply style for the inquiry information according to the inquiry channel information, and obtain the inquiry template associated with the inquiry channel information, and then fill the inquiry information, file information and reply style into the inquiry template to obtain the filled information, and then input the filled information into the above-mentioned generative AI 9012, and determine the reply information corresponding to the inquiry information according to the output information of the generative AI 9012. The AI ​​result processor 9013 may filter the output information of the inquiry reply model for sensitive words to obtain the reply information corresponding to the inquiry information;

[0150] The data layer of the quick reply module 902 may include a quick reply library 9021, which may include a public information library and a private information library. The public information library may exist in the form of multiple es, and the private information library may also exist in the form of multiple es. It can be set according to actual conditions, and the present disclosure does not make any special restrictions on this.

[0151] The collection / or publishing module 9022 of the quick reply module 902 is mainly used to provide two functions, the first function is the collection function, and the second function is the publishing function, wherein collection means that each attending doctor can collect the sample reply information in the public information library, and after automatic word segmentation by the word segmentation processor 9023 and automatic labeling by the label acquisition processor 9024, the above sample reply information and the label corresponding to the sample reply information can be stored in the personal private information set through the quick reply storage 9025. Publishing means that the attending doctor can publish the customized reply information contained in his own private information set to the public information library for public reference for all attending doctors connected to the system;

[0152] The retrieval module 9026 of the quick reply module 902 can perform query with or without parameters to recommend information to the attending doctor. Specifically, for the query with parameters scenario, after obtaining the input keywords of the attending doctor, after automatic word segmentation by the word segmentation processor 9023 and automatic labeling by the label acquisition processor 9024, the query strategy executor 9027 can execute the pre-configured information recommendation strategy to filter out the target information from the quick reply library 9021, and then, the filtered target information can be sent to the result processor 9028, so that the result processor 9028 returns it to the retrieval module 9026, so that the retrieval module 9026 can return the above target information to the receiving end, so that the attending doctor can quickly select the reply information for the consultation information from the above target information.

[0153] Based on the above technical solutions, the present disclosure can at least achieve the following technical effects:

[0154] First, by combining the consultation channel information, consultation information, and the patient's profile information such as age and gender, the trained consultation response model can be called to directly obtain specific response words. After that, the attending doctor can choose to modify or directly use the obtained response words according to the actual situation, thereby improving the doctor's response speed;

[0155] Second, based on the preset public information database and private information database, through the collection and publishing functions of the preset reply information, the preset reply information can be shared, learned and used among doctors to achieve continuous updating of the information database.

[0156] Third, when setting sample reply information, the sample reply information is automatically marked based on the applicable department information and other features, so as to facilitate the subsequent retrieval of the sample reply information based on the applicable department information, so as to quickly retrieve multiple reply information with higher adaptability to the scene;

[0157] Fourth, by completing and prompting the keywords input by the doctor in real time, it is possible to reduce the difficulty of use for users and quickly obtain reply information, save doctors' time, improve the efficiency of doctors' replies, and improve consultation satisfaction.

[0158] The present disclosure also provides an information reply device, Fig.10 FIG. 2 is a schematic diagram showing the structure of an information reply device in an exemplary embodiment of the present disclosure; Fig.10 As shown, the information reply device 1000 may include a mode detection module 1010, an information output module 1020 and an information recommendation module 1030. Among them:

[0159] Mode detection module 1010, used to receive the inquiry information from the inquiry terminal and detect the pre-configured inquiry mode; the inquiry mode includes intelligent answering mode and manual answering mode;

[0160] An information output module 1020 is used to output reply information corresponding to the medical inquiry information through a trained medical inquiry reply model in response to the medical inquiry mode being an intelligent answering mode;

[0161] The information recommendation module 1030 is used to recommend target information to the reception end based on a preconfigured information recommendation strategy in response to the consultation mode being a manual answering mode, so that the reception end determines reply information corresponding to the consultation information based on the target information.

[0162] In an exemplary embodiment of the present disclosure, the medical inquiry information includes the file information of the patient and the medical inquiry channel information of the patient; the information output module 1020 responds to the medical inquiry mode being the intelligent answering mode, and outputs the reply information corresponding to the medical inquiry information through a trained medical inquiry reply model, including: determining the reply style for the medical inquiry information according to the medical inquiry channel information; there is a preset correlation between the medical inquiry channel information and the reply style; the medical inquiry information, the file information and the reply style are input into the trained medical inquiry reply model, and the reply information corresponding to the medical inquiry information is determined according to the output information of the medical inquiry reply model.

[0163] In an exemplary embodiment of the present disclosure, after determining the reply style for the consultation information according to the consultation channel information, the information output module 1020 is configured to:

[0164] Acquire a consultation template associated with the consultation channel information; the consultation channel information and the consultation template have a preset association relationship; fill the consultation information, the file information and the reply style into the consultation template to obtain filled information; input the filled information into the trained consultation reply model, and determine the reply information corresponding to the consultation information according to the output information of the consultation reply model.

[0165] In an exemplary embodiment of the present disclosure, the information output module 1020 is configured to: after obtaining the output information of the medical inquiry reply model, filter the output information of the medical inquiry reply model for sensitive words to obtain reply information corresponding to the medical inquiry information.

[0166] In an exemplary embodiment of the present disclosure, in response to the consultation mode being a manual answering mode, the information recommendation module 1030 recommends target information to the reception end based on a preconfigured information recommendation strategy, including: detecting whether an input keyword from the reception end is received; in response to receiving an input keyword from the reception end, recommending target information to the reception end based on a first information recommendation strategy; in response to not receiving an input keyword from the reception end, recommending target information to the reception end based on a second information recommendation strategy.

[0167] In an exemplary embodiment of the present disclosure, the reception end corresponds to a receiving doctor, and the information recommendation module 1030 recommends target information to the reception end based on a first information recommendation strategy, including: performing information processing of at least one dimension on the input keyword to obtain the target input keyword corresponding to each dimension of the at least one dimension; querying a preconfigured public information library and / or a private information set of the receiving doctor based on the target input keyword and the department information to which the receiving doctor belongs to obtain a query result set corresponding to each dimension; and filtering to obtain the target information from at least one query result set corresponding to the at least one dimension.

[0168] In an exemplary embodiment of the present disclosure, the information processing of at least one dimension includes at least one of the following: information completion processing, word segmentation processing, and error correction processing.

[0169] In an exemplary embodiment of the present disclosure, the public information database contains multiple sample reply information and a label corresponding to each of the sample reply information, and the label contains applicable department information of the sample reply information; the information recommendation module 1030 queries the pre-configured public information database and / or the private information set of the attending doctor according to the target input keyword and the department information to which the attending doctor belongs, and obtains the query result set corresponding to each dimension, including: searching the public information database according to the department information to which the attending doctor belongs and the label corresponding to each of the sample reply information to obtain a feature information set; querying the feature information set and / or the private information set according to the target input keyword to obtain the query result set corresponding to each dimension.

[0170] In an exemplary embodiment of the present disclosure, when the at least one dimension contains only one dimension, the information recommendation module 1030 filters out the target information from at least one query result set corresponding to the at least one dimension, including: filtering out a preset number of target information from the query result set according to the reference frequency corresponding to each piece of information contained in the query result set.

[0171] In an exemplary embodiment of the present disclosure, when the query result set includes a first query result set determined according to a feature information set and a second query result set determined according to the private information set, the information recommendation module 1030 filters out a preset number of target information from the query result set according to a reference frequency corresponding to each piece of information contained in the query result set, including: splitting the preset number of pieces into a first part of pieces and a second part of pieces according to a preset weight coefficient; filtering out the first part of target information from the first query result set according to the reference frequency of each piece of information contained in the first query result set; filtering out the second part of target information from the second query result set according to the reference frequency of each piece of information contained in the second query result set.

[0172] In an exemplary embodiment of the present disclosure, when the at least one dimension includes two or more dimensions, the information recommendation module 1030 filters out the target information from at least one query result set corresponding to the at least one dimension, including: obtaining a priority order pre-set for each of the query result sets; and filtering out a preset number of the target information from the at least one query result set according to the priority order.

[0173] In an exemplary embodiment of the present disclosure, the information recommendation module 1030 recommends target information to the reception end based on the second information recommendation strategy, including: filtering the public information library according to the department information of the receiving doctor and the label corresponding to each of the sample reply information to obtain a characteristic information set; filtering out a preset number of target information from the characteristic information set and / or the private information set according to the frequency of citation of each piece of information contained in the characteristic information set and / or the private information set.

[0174] The specific details of each module in the above-mentioned information reply device have been described in detail in the corresponding information reply method, so they will not be repeated here.

[0175] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0176] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0177] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0178] The present application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device.

[0179] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wires, 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 thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0180] Computer-readable storage media can send, propagate or transmit programs for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable storage medium can be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0181] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiments.

[0182] In addition, an electronic device capable of implementing the above method is also provided in an embodiment of the present disclosure.

[0183] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0184] Refer to the following Fig.11 1100 according to this embodiment of the present disclosure is described. Fig.11 The electronic device 1100 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0185] like Fig.11 As shown, the electronic device 1100 is in the form of a general computing device. The components of the electronic device 1100 may include, but are not limited to: the at least one processing unit 1110, the at least one storage unit 1120, a bus 1130 connecting different system components (including the storage unit 1120 and the processing unit 1110), and a display unit 1140.

[0186] The storage unit stores a program code, which can be executed by the processing unit 1110, so that the processing unit 1110 performs the steps described in the "Exemplary Method" section of the present specification according to various exemplary embodiments of the present disclosure. For example, the processing unit 1110 can perform the following steps: Figure 1 As shown in: Step S110, receiving the inquiry information from the inquiry end, and detecting the pre-configured inquiry mode; the inquiry mode includes an intelligent answer mode and a manual answer mode; Step S120, in response to the inquiry mode being the intelligent answer mode, outputting the reply information corresponding to the inquiry information through the trained inquiry reply model; Step S130, in response to the inquiry mode being the manual answer mode, recommending target information to the reception end based on the pre-configured information recommendation strategy, so that the reception end determines the reply information corresponding to the inquiry information based on the target information.

[0187] The storage unit 1120 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 11201 and / or a cache storage unit 11202 , and may further include a read-only storage unit (ROM) 11203 .

[0188] The storage unit 1120 may also include a program / utility 11204 having a set (at least one) of program modules 11205, such program modules 11205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0189] Bus 1130 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0190] The electronic device 1100 may also communicate with one or more external devices 1200 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1100, and / or communicate with any device that enables the electronic device 1100 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 1150. Furthermore, the electronic device 1100 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 1160. As shown, the network adapter 1160 communicates with other modules of the electronic device 1100 via a bus 1130. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1100, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0191] It should be noted that in the technical solution of the embodiment of the present disclosure, the collection / collection, update, analysis, use, transmission, storage and other aspects of the user personal information involved are in compliance with the provisions of relevant laws and regulations, are used for legal and reasonable purposes, are not shared, leaked or sold outside of these legal uses, and are subject to supervision and management by national regulatory authorities. Necessary measures should be taken for user personal information to selectively block the use or access to personal information data to prevent illegal access to such personal information data, ensure that persons who have access to personal information data comply with the provisions of relevant laws and regulations, and ensure the security of user personal information. In addition, once these user personal information data are no longer needed, the risks should be minimized by limiting or even prohibiting data collection and / or deleting data.

[0192] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A method for replying information, characterized in that: include: Receive consultation information from the consultation terminal and detect the pre-configured consultation mode; The consultation mode includes an intelligent answering mode and a manual answering mode; In response to the medical inquiry mode being the intelligent answering mode, outputting reply information corresponding to the medical inquiry information through a trained medical inquiry reply model; In response to the consultation mode being the manual answering mode, target information is recommended to the reception end based on a preconfigured information recommendation strategy, so that the reception end determines reply information corresponding to the consultation information based on the target information.

2. The method according to claim 1, characterized in that The consultation information includes the patient's profile information and the patient's consultation channel information; In response to the medical consultation mode being the intelligent answering mode, outputting reply information corresponding to the medical consultation information through a trained medical consultation reply model includes: Determine a reply style for the consultation information according to the consultation channel information; there is a preset association relationship between the consultation channel information and the reply style; The medical inquiry information, the file information and the reply style are input into the trained medical inquiry reply model, and the reply information corresponding to the medical inquiry information is determined according to the output information of the medical inquiry reply model.

3. The method according to claim 2, characterized in that After determining a reply style for the consultation information according to the consultation channel information, the method further includes: Acquire a medical consultation template associated with the medical consultation channel information; the medical consultation channel information and the medical consultation template have a preset association relationship; Filling the medical consultation information, the file information and the reply style into the medical consultation template to obtain filled information; The filled information is input into the trained medical inquiry response model, and the response information corresponding to the medical inquiry information is determined according to the output information of the medical inquiry response model.

4. The method according to claim 2 or 3, characterized in that: After obtaining the output information of the consultation response model, the method further includes: The output information of the medical inquiry reply model is filtered for sensitive words to obtain reply information corresponding to the medical inquiry information.

5. The method according to claim 1, characterized in that In response to the consultation mode being the manual answering mode, recommending target information to the receiving terminal based on a preconfigured information recommendation strategy includes: Detecting whether an input keyword from the receiving terminal is received; In response to receiving an input keyword from the patient receiving terminal, recommending target information to the patient receiving terminal based on a first information recommendation strategy; In response to not receiving the input keyword from the receiving terminal, recommending target information to the receiving terminal based on a second information recommendation strategy.

6. The method according to claim 5, characterized in that The receiving end corresponds to a receiving doctor. The recommending target information to the receiving terminal based on the first information recommendation strategy includes: Performing information processing of at least one dimension on the input keyword to obtain a target input keyword corresponding to each dimension of the at least one dimension; According to the target input keyword and the department information to which the attending doctor belongs, a pre-configured public information database and / or the private information set of the attending doctor is queried to obtain a query result set corresponding to each dimension; The target information is obtained by filtering at least one query result set corresponding to the at least one dimension.

7. The method according to claim 6, characterized in that The information processing of at least one dimension includes at least one of the following: information completion processing, word segmentation processing, and error correction processing.

8. The method according to claim 6, characterized in that The public information database includes multiple pieces of sample response information and labels corresponding to each piece of the sample response information, wherein the labels include applicable department information of the sample response information; The step of querying a preconfigured public information database and / or a private information set of the attending doctor according to the target input keyword and the department information to which the attending doctor belongs, and obtaining a query result set corresponding to each dimension includes: According to the department information of the attending doctor and the label corresponding to each of the sample reply information, the public information database is searched to obtain a feature information set; According to the target input keyword, the feature information set and / or the private information set is queried to obtain a query result set corresponding to each dimension.

9. The method according to claim 6, characterized in that When the at least one dimension includes only one dimension, the step of filtering at least one query result set corresponding to the at least one dimension to obtain the target information includes: According to the reference frequency corresponding to each piece of information contained in the query result set, a preset number of target information pieces are filtered out from the query result set.

10. The method according to claim 9, characterized in that When the query result set includes a first query result set determined according to the feature information set and a second query result set determined according to the private information set, The step of filtering out a preset number of target information from the query result set according to the reference frequency corresponding to each piece of information included in the query result set includes: According to a preset weight coefficient, the preset number of items is divided into a first part of items and a second part of items; Filtering the first portion of target information from the first query result set according to the reference frequency of each piece of information included in the first query result set; According to the reference frequency of each piece of information included in the second query result set, a second portion of pieces of target information are filtered out from the second query result set.

11. The method according to claim 6, characterized in that When the at least one dimension includes two or more dimensions, The filtering and obtaining the target information from at least one query result set corresponding to the at least one dimension includes: Obtaining a priority order pre-set for each of the query result sets; According to the priority order, a preset number of pieces of target information are filtered out from the at least one query result set.

12. The method according to claim 5, characterized in that The recommending target information to the receiving terminal based on the second information recommendation strategy includes: According to the department information of the attending doctor and the label corresponding to each of the sample reply information, the public information database is screened to obtain a feature information set; According to the reference frequency of each piece of information contained in the characteristic information set and / or the private information set, a preset number of pieces of target information are screened out from the characteristic information set and / or the private information set.

13. An information reply device, characterized in that: include: A mode detection module, used for receiving the consultation information from the consultation terminal and detecting the pre-configured consultation mode; The consultation mode includes an intelligent answering mode and a manual answering mode; An information output module, configured to output reply information corresponding to the medical inquiry information through a trained medical inquiry reply model in response to the medical inquiry mode being an intelligent answering mode; The information recommendation module is used to recommend target information to the reception end based on a preconfigured information recommendation strategy in response to the consultation mode being a manual answering mode, so that the reception end determines reply information corresponding to the consultation information based on the target information.

14. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the information reply method according to any one of claims 1 to 12 is implemented.

15. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the information reply method according to any one of claims 1 to 12 by executing the executable instructions.

Images