An intelligent treatment method, system, device and medium for AI virtual human accompanying diagnosis

Abstract

The application discloses an intelligent treatment method, system, equipment and medium for AI virtual person accompanying diagnosis, and the method comprises the following steps: receiving an intelligent guide diagnosis demand instruction, constructing a virtual clinic and generating an AI virtual person; after judging that a user enters the virtual clinic, real-time acquisition of user input information; determining the user object type and the user guide diagnosis application scene according to the user input information; identifying the abnormal emotion of the user and matching the preset emotion soothing strategy, so that the AI virtual person interacts with the user according to the matched emotion soothing strategy; according to the real-time interaction information of the user and the historical intelligent guide diagnosis information of the user, combining the user guide diagnosis application scene, matching the preset deep learning model to output the interaction reaction information and the preliminary diagnosis result of the AI virtual person; according to the preliminary diagnosis result, determining the registration department, generating the registration appointment recommendation and completing the registration appointment. The application can improve the user experience in the intelligent treatment process of AI virtual person accompanying diagnosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence medicine, in particular to an intelligent medical treatment method, system, device and medium for AI virtual person accompanying diagnosis. BACKGROUND

[0002] With the development of Internet technology, online medical consultation has gradually become popular, and more and more patients choose to conduct preliminary diagnosis through network platforms. This online consultation method not only improves the convenience and accessibility of medical services, but also provides preliminary medical advice for patients. However, most platforms only provide text or video communication, lacking intelligent diagnosis and treatment planning services, making it difficult for patients to obtain comprehensive and professional guidance before treatment.

[0003] In order to solve the above problems, in the prior art, the following methods are usually used: first, the patient judges the disease and selects the department and doctor by himself. Second, the customer service personnel on the platform assists the patient to judge the disease and select the appropriate department and doctor. Third, some platforms provide some simple treatment process planning services, but these services are only based on basic information and common symptoms, and lack personalized and intelligent recommendation, and cannot realize intelligent interaction and complete preliminary diagnosis and generate intelligent recommendation for different diagnosis objects and diagnosis needs.

[0004] There are also some intelligent accompanying diagnosis systems in the prior art to assist users in completing intelligent medical treatment, but these systems mainly focus on improving the medical service efficiency of virtual person accompanying diagnosis before, during and after treatment, ignoring the improvement of user treatment experience in terms of emotion recognition and pacification of diagnosis objects during treatment, different needs for treatment process and intelligent recommendation in different objects and different application scenarios. SUMMARY

[0005] In order to improve the user experience in the intelligent medical treatment process of AI virtual person accompanying diagnosis, the present application provides an intelligent medical treatment method, system, device and medium for AI virtual person accompanying diagnosis.

[0006] In a first aspect, the present application provides an intelligent medical treatment method for AI virtual person accompanying diagnosis, comprising:

[0007] receiving an intelligent guide diagnosis demand instruction sent by a user terminal, constructing a virtual clinic and generating an AI virtual person;

[0008] After determining that the user enters the virtual clinic based on the operation of the user terminal, real-time acquisition of user input information; the user input information includes pre-input user portrait information, user-selected intelligent guide diagnosis scene information, user real-time interaction information and user historical intelligent guide diagnosis information;

[0009] Determine a user object type according to the acquired user portrait information and / or user historical intelligent guide diagnosis information; the user object type includes: a child type, an adult type, an old person type, and an interactive obstacle population type;

[0010] Determine a user guide diagnosis application scene according to the acquired user historical intelligent guide diagnosis information and user-selected intelligent guide diagnosis scene information; the user guide diagnosis application scene includes: ordinary inquiry, emergency inquiry, ordinary re-inquiry, and emergency re-inquiry.

[0011] According to the user real-time interaction information, complete user abnormal emotion recognition by using a deep learning algorithm; the abnormal situation includes anxiety, tension, and sadness; based on the matching of the recognized abnormal emotion and the determined user object type, match a preset emotion pacification strategy, so that the AI virtual person interacts with the user according to the matched emotion pacification strategy; wherein each user object type and each abnormal emotion match a preset emotion pacification strategy;

[0012] According to the determined user guide diagnosis application scene, match a preset deep learning model; according to the user real-time interaction information and the user historical intelligent guide diagnosis information, combine the matched deep learning model to output AI virtual person interaction reaction information and a preliminary diagnosis result; wherein the AI virtual person interaction reaction information output times and the preliminary diagnosis result output time of different preset deep learning models are different.

[0013] According to the preliminary diagnosis result, determine a registration department, and generate a registration appointment recommendation according to the registration department; receive a registration appointment request information sent by a user terminal according to the registration appointment recommendation, connect a selected appointment registration department hospital information system, and complete the registration appointment.

[0014] By using the above scheme, the determination of the user object type and the guide diagnosis application scene based on the user portrait information and the historical intelligent guide diagnosis information is realized, which provides a basis for the subsequent AI virtual person accompanying diagnosis; the user abnormal emotion is recognized by using a deep learning algorithm, and the emotion pacification strategy that matches the user type is matched, which effectively pacifies the emotion of the user in the diagnosis process and improves the user experience; the deep learning model that matches the application scene is intelligently matched, the diagnosis process planning is adaptively simplified, and the preliminary diagnosis result is timely output, which improves the user experience.

[0015] Preferably, the determination of the registration department according to the preliminary diagnosis result, and the generation of the registration appointment recommendation according to the registration department include:

[0016] Before the output of the preliminary diagnosis result, receive the last interaction information of the user and the AI virtual person to acquire the geographical position information of the corresponding user terminal; according to the preliminary diagnosis result, acquire the current appointment registration quantity of the hospital with the determined registration department.

[0017] Based on the determined user object type and user triage application scenario, the obtained geographical location information of the user terminal, and the current number of appointments at the hospital with a determined registration department, an appointment recommendation order that meets the first rule is generated according to the user's historical intelligent triage information and the determined registration department.

[0018] The first rule includes: In the user-guided service application scenario, the hospital where the registration department for regular follow-up visits and emergency follow-up visits is located is the same as the hospital where the original registration department was located; the current number of appointments at the hospital with the definite registration department is less than the maximum appointment threshold matched by the user-guided service application scenario; the fewer the current number of appointments at the hospital with the definite registration department, the higher the priority recommendation; the closer the geographical location of the hospital where the registration department is located is to the geographical location of the user's terminal, the higher the priority recommendation; and the hospital where the registration department is located is a hospital exclusive to the user's object type, the higher the priority recommendation. Different user-guided service application scenarios have corresponding maximum appointment thresholds. The priority of the priority recommendations generated based on the user's terminal's geographical location, the priority of the priority recommendations generated based on the user's object type, and the priority of the priority recommendations generated based on the number of appointments are determined according to the user's pre-set priority level, with higher priority levels corresponding to earlier priority recommendations.

[0019] By adopting the above scheme, when generating appointment recommendations intelligently, the system comprehensively considers the user's terminal's geographical location information, the current number of appointments at the hospital, the user's type, and the user's triage application scenario. This results in a more accurate appointment recommendation order, ensuring that patients receive more reasonable medical arrangements, reducing waiting time, and improving medical efficiency and user experience.

[0020] Preferred options also include:

[0021] When connecting to the hospital information system of the selected appointment department, a preliminary diagnosis report is sent synchronously; when receiving a preliminary diagnosis report request instruction from the terminal of the attending physician of the selected appointment department in the hospital information system, the preliminary diagnosis report is sent to the terminal of the attending physician of the selected appointment department; the preliminary diagnosis report is generated by integrating key information extracted from real-time user interaction information and AI virtual human interaction response information converted into text using natural language processing technology, combined with the preliminary diagnosis results.

[0022] The real doctor examination item suggestion uploaded by the terminal of the doctor on duty of the registered department receiving the appointment is received, the preliminary diagnosis result and the examination item required for final diagnosis are determined according to the medical knowledge graph, the determined examination item is compared with the real doctor examination item suggestion in terms of similarity, if the similarity is lower than a first preset similarity, a real doctor examination item suggestion reconfirmation prompt information is generated and the examination item required for final diagnosis is displayed, and when the real doctor examination item suggestion reconfirmation is received, the real doctor examination item suggestion is taken as the final examination item suggestion;

[0023] When the user's examination item results are received from the hospital information system, the secondary diagnosis result is output according to the user's examination item results, real-time interaction information and historical intelligent guide information of the user, and in combination with a deep learning algorithm; the real doctor diagnosis result uploaded by the terminal of the doctor on duty of the registered department receiving the appointment is received, and the secondary diagnosis result is compared in terms of similarity, if the similarity is lower than a second preset similarity, a real doctor diagnosis result reconfirmation information is generated and the preliminary diagnosis result is displayed on the terminal of the doctor on duty of the registered department, and when the real doctor diagnosis result reconfirmation information is received, the real doctor diagnosis information is taken as the final diagnosis result.

[0024] By adopting the above scheme, the basic situation of the user's preliminary diagnosis report is informed to the doctor, so that the doctor can know the patient's condition in advance; at the same time, the preliminary diagnosis report is compared with the doctor's examination item suggestion in terms of similarity, and when the similarity is low, a confirmation information is generated to prompt the doctor, so as to ensure the accuracy of the examination item; after the user completes all the examination items, the secondary diagnosis result is generated again in combination with the examination results, and is compared with the doctor's diagnosis result in terms of similarity, and when the similarity is low, a confirmation information is generated to prompt the doctor, so as to ensure the accuracy of the final diagnosis result.

[0025] Preferably, it further comprises:

[0026] When the first route planning request instruction sent by the user terminal is received, the position information of each department of the hospital and the current flow density of each area of the hospital are obtained, the determined user object type and the user guide application scenario are combined, and a first planning route of the user reaching a single fixed target position in the hospital where the registered department receiving the appointment is generated by using a random algorithm, the first constraint condition includes: for the elderly type or the child type, the flow density in the area along the planning route is less than a first preset flow density threshold and the distance to the single fixed target position in the hospital where the registered department receiving the appointment is the shortest; for the adult type, the flow density in the area along the planning route is less than a second preset flow density threshold and the distance to the single fixed target position in the hospital where the registered department receiving the appointment is the shortest; the first preset flow density threshold is less than the second preset flow density threshold;

[0027] When the second route planning request instruction sent by the user terminal is received, the final recommended examination items are checked to obtain the location information of each examination item in each area of the hospital, the current crowd density of each area of the hospital, and the current waiting time of each examination item, the second planning route of the user to complete all the examination items is generated by using a random algorithm in combination with the determined user object type and the user guidance application scenario, and the second constraint condition is satisfied; the second constraint condition includes: for the elderly type or the child type, the crowd density in the area along the planning route is less than the first preset crowd density threshold, and the shortest waiting time required to complete all the examination items and the shortest distance along the route to complete all the examination items are satisfied; for the adult type, the crowd density in the area along the planning route is less than the second preset crowd density threshold, and the shortest waiting time required to complete all the examination items and the shortest distance along the route to complete all the examination items are satisfied.

[0028] By adopting the above scheme, the first planning route and the second planning route are generated according to the object type of the user and the guidance application scenario in combination with the real-time crowd density of the hospital, so that the first planning route ensures that the user can avoid the crowd dense area, ensuring the convenience and safety of the user; the second planning route ensures that all the examination items of the user are completed in the shortest time, and the distance along the route is the shortest and the high crowd dense area is avoided, improving the efficiency of the user and the user experience.

[0029] Preferably, it further comprises:

[0030] When the preliminary diagnosis report is sent to the terminal of the on-duty doctor of the booked registration department, the preliminary diagnosis result in the preliminary diagnosis report is in a to-be-unlocked state within a preset time period, and when the preliminary diagnosis result acquisition request sent by the terminal of the on-duty doctor of the booked registration department is received or after the preset time period is exceeded, the preliminary diagnosis result unlocking password in the preliminary diagnosis report is sent to the terminal of the on-duty doctor of the booked registration department.

[0031] By adopting the above scheme, the preliminary diagnosis result is set to be in a to-be-unlocked state within a preset time period, and is unlocked after a request is received or the reserved doctor's self-thinking preset time period is exceeded, which not only ensures that the preliminary diagnosis result will not be immediately exposed to unauthorized personnel, thereby enhancing information security and patient privacy protection, but also reduces the dependence of the doctor on the intelligent generated preliminary diagnosis result, thereby preventing the doctor from misdiagnosis.

[0032] Preferably, it further comprises:

[0033] The prescription information uploaded by the on-duty doctor terminal of the registered department is received, and user personalized health management information is generated by using a deep learning algorithm according to the prescription information and user portrait information; the prescription information includes diet taboos, medicine taking information and rehabilitation training information; and the user personalized health management information includes user personalized diet recommendations, user medicine taking prompt information and rehabilitation training prompt information.

[0034] By using the above scheme, the habits of the user are considered, and personalized health management based on the prescription information and user portrait information is realized, so that the scientificity and individuality of the diet, medicine taking and rehabilitation training of the user after the medical treatment are improved, and the user experience satisfaction is improved.

[0035] Preferably, it further comprises:

[0036] The medicine extraction completion information sent by the hospital information system of the registered department is received, and medicine inspection information is generated and displayed on the user terminal;

[0037] The medicine image and the medicine inspection request instruction uploaded by the user terminal are received, the medicine information is identified by using image recognition technology, the identified medicine information is compared with the medicine information in the medicine taking information, if there is a difference, the medicine error prompt information is generated and sent to the hospital information system of the registered department or the user terminal; if the identified medicine information includes prescription medicine information, in addition to comparing the prescription medicine information in the identified medicine information with the prescription medicine information in the medicine taking information, it is also necessary to find out whether there is doctor signature information in the received prescription information, if there is no doctor signature information, the medicine error prompt information is also generated and sent to the hospital information system of the registered department and the on-duty doctor terminal of the registered department.

[0038] By using the above scheme, the user is reminded to check the medicine extracted according to the prescription information, so as to avoid that the medicine taken by the user is accurate and correct, and at the same time, for the prescription medicine, it is determined again that it belongs to the medicine determined to be taken by the doctor, so as to avoid safety hazards and improve the user experience.

[0039] In a second aspect, the present application provides an intelligent medical treatment system of AI virtual person accompanying diagnosis, comprising:

[0040] The virtual clinic construction module is used for receiving the intelligent guidance demand instruction sent by the user terminal, constructing the virtual clinic and generating the AI virtual person;

[0041] The information required for guidance acquisition module is used for acquiring the user input information in real time after the user enters the virtual clinic based on the user terminal operation; the user input information includes pre-input user portrait information, user-selected intelligent guidance scene information, user real-time interaction information and user historical intelligent guidance information;

[0042] The guide diagnosis influencing factor judgment module is configured to determine a user object type according to the obtained user portrait information and / or user historical intelligent guide diagnosis information; the user object type includes a child type, an adult type, an old person type, and an interactive obstacle population type; and a user guide diagnosis application scene is determined according to the obtained user historical intelligent guide diagnosis information and user-selected intelligent guide diagnosis scene information; the user guide diagnosis application scene includes ordinary inquiry, emergency inquiry, ordinary reexamination, and emergency reexamination.

[0043] The guide diagnosis interactive emotion pacification module is configured to complete user abnormal emotion recognition according to user real-time interactive information by using a deep learning algorithm, the abnormal situation including anxiety, tension, and sadness; a preset emotion pacification strategy is matched based on matching of the recognized abnormal emotion and the determined user object type, so that the AI virtual person interacts with the user according to the matched emotion pacification strategy; wherein each user object type is matched with one preset emotion pacification strategy for each abnormal emotion.

[0044] The guide diagnosis diagnosis and recommendation module is configured to match a preset deep learning model according to the determined user guide diagnosis application scene; AI virtual person interactive response information and a preliminary diagnosis result are output according to user real-time interactive information and user historical intelligent guide diagnosis information, in combination with the matched deep learning model; wherein the AI virtual person interactive response information output times and the preliminary diagnosis result output time are different for different preset deep learning models.

[0045] The guide diagnosis registration reservation operation module is configured to determine a registration department according to the preliminary diagnosis result, and generate a registration reservation recommendation according to the registration department; registration reservation request information sent by a user terminal according to the registration reservation recommendation is received, a hospital information system in which the selected registration department is located is connected, and registration reservation is completed.

[0046] By adopting the above scheme, the construction and application of the intelligent diagnosis system are realized, personalized medical services can be provided according to specific needs of different users, and user experience in the intelligent diagnosis process with AI virtual person accompanying diagnosis is improved.

[0047] In a third aspect, a computer readable storage medium is provided, which includes a stored computer program, wherein the computer readable storage medium controls a device in which the computer readable storage medium is located to execute the method as described above when the computer program is run.

[0048] In a fourth aspect, a computer device is provided, which includes a memory, a processor, and a program stored in the memory and executable by the processor, and the program is executed by the processor to implement the steps of the method as described above.

[0049] In summary, the present application has the following beneficial effects:

[0050] 1、Obtain multiple information of the user, including user portrait information, intelligent guide diagnosis scene information, real-time interaction information and historical intelligent guide diagnosis information, determine the user object type and its application scene, and then match the corresponding emotional pacification strategy according to the abnormal emotion of the user object type, match the corresponding deep learning model according to the application scene, output the interactive response information and preliminary diagnosis result of the AI virtual person in time, and recommend the appropriate registration department and appointment registration service according to the preliminary diagnosis result, improve the experience of the user in the process of treatment;

[0051] 2、During the treatment process of the user according to the recommended registration department, adaptively recommend a planning route according to different user needs to meet the planning route needs of different users in different application scenarios, generate a planning route with small density and short time consumption, and improve the user experience; During the treatment process of the user according to the recommended registration department, compare the examination items corresponding to the preliminary diagnosis result with the examination items recommended by the real doctor, and compare the preliminary diagnosis result with the diagnosis result given by the real doctor, remind the doctor of possible errors, and ensure the accuracy of the diagnosis result;

[0052] 3、At the medicine taking stage after the treatment, instruct the user to check the medicine to ensure the accuracy of the medicine extraction; After the treatment, generate personalized health management information based on the medical order information and the user portrait, realize the scientific and personalized management of the diet, medication and rehabilitation training of the user after the treatment, and improve the user experience satisfaction. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 The flow chart of the intelligent treatment method of the AI virtual person accompanying diagnosis in the specific embodiment;

[0054] Figure 2 The structural schematic diagram of the intelligent treatment system of the AI virtual person accompanying diagnosis in the specific embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0056] As shown in Figure 1 The embodiment of the present application discloses an intelligent treatment method of AI virtual person accompanying diagnosis, and the specific steps include:

[0057] S1, receive an intelligent guide diagnosis demand instruction, build a virtual clinic and generate an AI virtual person.

[0058] Specifically, the user can log in to the AI virtual person accompanying diagnosis APP and pre-enter the user's portrait information, such as age, gender, ID card, personal basic information such as habits, historical illness, sensitive drug information and physical health information such as blood pressure and weight; generate an intelligent guide diagnosis demand instruction in the AI virtual person accompanying diagnosis APP, and the AI virtual person accompanying diagnosis APP corresponding server receives the intelligent guide diagnosis demand instruction sent by the user terminal. After receiving the intelligent guide diagnosis demand instruction sent by the user terminal, a virtual clinic is constructed and an AI virtual person is generated and displayed on the AI virtual person accompanying diagnosis APP interface. Among them, the AI virtual person accompanying diagnosis APP corresponding server generates an AI virtual person that can generate an AI virtual person image that meets the user's preferences according to the user's portrait information.

[0059] S2, after judging that the user enters the virtual clinic based on the user terminal operation, real-time user input information is obtained.

[0060] Specifically, the AI virtual person accompanying diagnosis APP corresponding server will judge whether the user enters the virtual clinic on the APP interface through the user terminal operation, and real-time user input information is obtained after judging that the user enters the virtual clinic based on the user terminal operation.

[0061] Among them, the AI virtual person accompanying diagnosis APP displays intelligent guide diagnosis scene mode information, and the user can select an intelligent guide diagnosis scene on the AI virtual person accompanying diagnosis APP interface according to his own needs. After judging that the user enters the virtual clinic based on the user terminal operation, the user's pre-entered portrait information, the user's selected intelligent guide diagnosis scene information and other user input information can be obtained in real time; the AI virtual person accompanying diagnosis APP corresponding server will store the user's historical intelligent guide diagnosis information corresponding to the user, and after judging that the user enters the virtual clinic based on the user terminal operation, the user's historical intelligent guide diagnosis information can be queried and called in real time; after the user enters the virtual clinic through the user terminal operation, real-time interaction information will be uploaded, including: input text information, video information, voice information and other types of interaction information, and the user's real-time interaction information is obtained.

[0062] S3, determine the user object type and user guide diagnosis application scene according to the obtained user input information.

[0063] Specifically, considering the difference in the demand of different user object types for interaction with the AI virtual person or the preliminary diagnosis given by the AI virtual person under different guide diagnosis application scenes, the current user object type and user guide diagnosis application scene are determined in priority.

[0064] Among them, considering that the needs of users of different ages for interaction with the AI virtual person are quite different, the user object type is mainly divided into the following types, including: child type, adult type, old person type, and interactive obstacle population type (such as: voice or hearing impaired population, etc.); the user object type for the current APP intelligent diagnosis demand operation is determined according to the age information and body health information in the obtained user portrait information and / or user historical intelligent diagnosis information;

[0065] Considering that the needs for the length of time for the AI virtual person to give a preliminary diagnosis or the degree of simplicity of interaction with the AI virtual person are quite different in the scenes of ordinary inquiry and emergency inquiry, and in the scenes of normal inquiry and follow-up inquiry, the user diagnosis application scene is mainly divided into the following types, including: ordinary inquiry, emergency inquiry, ordinary follow-up and emergency follow-up; the user diagnosis application scene is determined according to whether it is the first inquiry and the selected intelligent diagnosis scene mode (ordinary or emergency) in the selected intelligent diagnosis scene information and the user historical intelligent diagnosis information.

[0066] S4, according to the user input information, determine the user object type and the user diagnosis application scene matching the preset emotion soothing strategy and deep learning model, so that the AI virtual person interacts with the user according to the matched emotion soothing strategy and deep learning model, and generates a preliminary diagnosis result.

[0067] Specifically, after the user inputs the interaction information on the APP, the AI virtual person will interact according to the input interaction information, thereby completing the preliminary diagnosis of whether the user is ill. During the diagnosis process, the user may have some abnormal emotions such as anxiety, tension, sadness, etc. These abnormal emotions will affect the user's description of the feeling of being ill. In order to output as accurate a diagnosis result as possible, the user's abnormal emotions are recognized and the emotion soothing strategy is used to alleviate the user's abnormal emotions, so as to improve the emotional intelligence and individualization level of the AI virtual person accompanying diagnosis service, thereby realizing more considerate and efficient accompanying diagnosis experience in different scenes and populations.

[0068] Among them, the AI virtual person accompanying diagnosis APP corresponds to a server in which an emotion recognition model constructed and trained based on a deep learning algorithm is built in, the input of the emotion recognition model is real-time interaction information of the user, including: user text information, user face and body video information or user semantic information, etc. Multimodal real-time interaction information, the output of the model is whether the user has abnormal emotions and the type of abnormal emotions, the abnormal situation includes anxiety, tension, sadness; using the emotion recognition model, the user's abnormal emotion recognition is realized according to the user's real-time interaction information.

[0069] According to the matching of the identified abnormal emotion and the determined user object type, a preset emotion soothing strategy is matched, so that the AI virtual person interacts with the user according to the matched emotion soothing strategy; wherein each user object type is matched with a preset emotion soothing strategy for each abnormal emotion, and the preset emotion soothing strategies can be set by experts according to different user object types and different types of abnormal emotions. For example: for children, the first to third emotion soothing strategies are preset in the order of anxiety, tension and sadness, such as: for children with tension, the second emotion soothing strategy is preset, which includes playing animation, music and other elements that children like, distracting their attention, and relieving their tension, and the AI virtual person interacts with the user according to the matched second emotion soothing strategy; for example: for the elderly, the fourth to sixth emotion soothing strategies are preset in the order of anxiety, tension and sadness, such as: for the elderly with sadness, the sixth emotion soothing strategy is preset, which includes sharing successful treatment cases of diseases or outputting encouraging words to relieve the user's sadness, and the AI virtual person interacts with the user according to the matched sixth emotion soothing strategy; for example: for adults, the seventh to ninth emotion soothing strategies are preset in the order of anxiety, tension and sadness, such as: for adults with anxiety, the seventh emotion soothing strategy is preset, which includes guiding the user to explain the symptoms with calm language, and the AI virtual person interacts with the user according to the matched seventh emotion soothing strategy; for example: for the interactive barrier population, the tenth to twelfth emotion soothing strategies are preset in the order of anxiety, tension and sadness.

[0070] Considering that the user has different requirements for the output time of the preliminary diagnosis result and the interaction conciseness in different application scenarios, a preset deep learning model is matched according to the determined user diagnosis application scenario; wherein the number of AI virtual person interaction response information and the output time of the preliminary diagnosis result of different preset deep learning models are different, including: a first deep learning model matched for ordinary diagnosis, a second deep learning model for emergency diagnosis, a third deep learning model matched for ordinary re-diagnosis, and a fourth deep learning model matched for emergency re-diagnosis; the input of each deep learning model is the real-time interaction information of the user and the historical intelligent diagnosis information of the user, and the model output is the AI virtual person interaction response information or the preliminary diagnosis result, and the number of AI virtual person interaction response information (i.e. the number of AI interaction processes, reflecting the AI interaction conciseness) and the output time of the preliminary diagnosis result (reflecting the diagnosis efficiency) of each deep learning model gradually decrease from the first deep learning model, the second deep learning model, the third deep learning model and the fourth deep learning model.

[0071] According to the user real-time interaction information and the user history intelligent guide diagnosis information, the AI virtual person interaction response information and the preliminary diagnosis result are output in combination with the matched deep learning model.

[0072] S5, according to the preliminary diagnosis result, determine the registration department, and generate a registration appointment recommendation according to the registration department; receive the registration appointment request information sent by the user terminal according to the registration appointment recommendation, connect the selected appointment registration department in the hospital information system and complete the registration appointment.

[0073] Specifically, according to the registration department information corresponding to the diagnosis result of different diseases in the medical knowledge graph, the registration department corresponding to the preliminary diagnosis result is determined, such as: for the user output hand injury, the registration department is determined as the general surgery department. Subsequently, considering the position, registration appointment quantity and other factors, the registration appointment recommendation is generated according to the registration department, and the specific steps include:

[0074] Before outputting the preliminary diagnosis result, the last interaction information between the user and the AI virtual person is received to obtain the geographical position information of the corresponding user terminal, that is, the geographical position is determined according to the information transmission path and time;

[0075] According to the preliminary diagnosis result, the current appointment registration quantity of the hospital with the determined registration department is obtained, that is, the hospital information system of the hospital with the determined registration department is connected to query the current appointment registration quantity of the department, such as: the current appointment registration quantity of the general surgery department of the first people's hospital is 10, and the current appointment registration quantity of the general surgery department of the second people's hospital is 4.

[0076] In combination with the determined user object type and user guide diagnosis application scene, the obtained geographical position information of the user terminal and the current appointment registration quantity of the hospital with the determined registration department, the registration appointment recommendation order meeting the first rule is generated according to the user history intelligent guide diagnosis information and the determined registration department, and is displayed on the APP interface for the user to select.

[0077] Among them, the first rule includes: the hospital and the original registration department of the registration department corresponding to the general reexamination and emergency reexamination in the user guide diagnosis application scene are the same, that is, once it is determined that the current user guide diagnosis application scene is reexamination, the corresponding hospital and department corresponding to the registration appointment are determined according to the original registration department in the user history intelligent guide diagnosis information;

[0078] and the current number of pre-booking in the hospital with the determined booking department is less than the corresponding maximum pre-booking number threshold of the hospital matched by the user guide application scenario, that is, different user guide application scenarios are provided with corresponding matching maximum pre-booking number thresholds of the corresponding hospital, and the threshold is determined according to the maximum average value of the historical pre-booking number received under different user guide application scenarios of the hospital with the determined booking department. If the maximum pre-booking number threshold is exceeded, there is a risk of booking failure. Specifically, the first maximum pre-booking number threshold matched by the ordinary inquiry, the second maximum pre-booking number threshold of the emergency inquiry, the third maximum pre-booking number threshold matched by the ordinary re-inquiry, and the fourth maximum pre-booking number threshold matched by the emergency re-inquiry. Because the re-inquiry is relatively short in duration compared to the ordinary inquiry, and the emergency is relatively limited in the number of pre-booking numbers compared to the ordinary, the second maximum pre-booking number threshold, the first maximum pre-booking number threshold, the fourth maximum pre-booking number threshold, and the third maximum pre-booking number threshold of a specific hospital gradually increase under normal circumstances.

[0079] and the current number of pre-booking in the hospital with the determined booking department is less than the corresponding maximum pre-booking number threshold of the hospital matched by the user guide application scenario, that is, different user guide application scenarios are provided with corresponding matching maximum pre-booking number thresholds of the corresponding hospital, and the threshold is determined according to the maximum average value of the historical pre-booking number received under different user guide application scenarios of the hospital with the determined booking department. If the maximum pre-booking number threshold is exceeded, there is a risk of booking failure. Specifically, the first maximum pre-booking number threshold matched by the ordinary inquiry, the second maximum pre-booking number threshold of the emergency inquiry, the third maximum pre-booking number threshold matched by the ordinary re-inquiry, and the fourth maximum pre-booking number threshold matched by the emergency re-inquiry. Because the re-inquiry is relatively short in duration compared to the ordinary inquiry, and the emergency is relatively limited in the number of pre-booking numbers compared to the ordinary, the second maximum pre-booking number threshold, the first maximum pre-booking number threshold, the fourth maximum pre-booking number threshold, and the third maximum pre-booking number threshold of a specific hospital gradually increase under normal circumstances.

[0080] and the hospital where the booking department is located belongs to the user object type exclusive hospital, which is preferentially recommended, such as a child exclusive hospital, which is preferentially recommended for children of the same type;

[0081] In which, due to the possible contradiction of different factors preferentially recommended, the priority of the preferentially recommended generated based on the geographical location information of the user terminal, the priority of the preferentially recommended generated based on the user object type, and the priority of the preferentially recommended generated based on the pre-booking number are determined according to the user's pre-set priority level (i.e. the user's pre-entered login information). The larger the priority level, the higher the corresponding priority recommendation order, that is, the larger the corresponding priority recommendation is taken as the standard, such as: the user pre-sets the priority of the preferentially recommended generated based on the pre-booking number as the highest, that is, the user values the pre-booking number the most, and the second people's hospital general surgery department is preferentially recommended in the order of the first people's hospital general surgery department. In addition, the priority level of the preferentially recommended generated based on different factors can be generated according to the user's historical selection, and the factor with the highest selection frequency has the highest priority level of the preferentially recommended.

[0082] The receiving user terminal receives the registration appointment request information sent according to the registration appointment recommendation, connects the selected registration department of the hospital information system and completes the registration appointment, and facilitates the user to directly visit a doctor.

[0083] To sum up, the user experience in the intelligent diagnosis process accompanied by the AI virtual person can be improved by using the above-mentioned implementation methods, and the needs of different types of users in different application scenarios can be met.

[0084] In one specific embodiment, in the intelligent diagnosis process accompanied by the AI virtual person, in order to further improve the diagnosis efficiency and accuracy and improve the user experience in the diagnosis process, the method further comprises:

[0085] The generated preliminary diagnosis report is sent synchronously when connecting the selected registration department of the hospital information system; and when receiving the preliminary diagnosis report request instruction sent by the terminal of the doctor on duty of the registered registration department in the hospital information system of the selected registration department, the preliminary diagnosis report is sent to the terminal of the doctor on duty of the registered registration department, which is equivalent to informing the doctor of the basic condition of the patient in advance, so that the doctor can have a general understanding in advance, and the timely diagnosis output of the recommended examination items is realized.

[0086] The preliminary diagnosis report is generated by extracting key information such as disease description from the user real-time interaction information and the AI virtual person interaction response information converted into text content by using natural language technology or deep learning technology.

[0087] The real doctor recommended examination items uploaded by the terminal of the doctor on duty of the registered registration department are received, the required examination items for the preliminary diagnosis result are determined according to the medical knowledge graph, the determined examination items and the real doctor recommended examination items are compared in similarity, if the similarity is lower than the first preset similarity, it indicates that the real doctor recommended examination items may have missing items, the real doctor recommended examination items reconfirmation prompt information is generated and the required examination items for the preliminary diagnosis result are displayed, so that the real doctor can judge whether it is necessary to supplement, adjust or modify according to the required examination items for the preliminary diagnosis result; when the real doctor recommended examination items reconfirmation is received, it indicates that the real doctor determines that the examination items are correct, the real doctor recommended examination items are taken as the final recommended examination items, otherwise, the modified real doctor recommended examination items are taken as the final recommended examination items.

[0088] In view of the fact that combining the examination results with the user interaction information can analyze and obtain more accurate diagnosis results, upon receiving the user's examination item results uploaded by the hospital information system, a secondary diagnosis result is output according to the user's examination item results, the user's real-time interaction information and the user's historical intelligent guidance information, in combination with a deep learning algorithm; the real doctor's diagnosis result uploaded by the terminal of the on-duty doctor of the registered department is received, and a similarity comparison is made with the generated secondary diagnosis result; if the similarity is lower than a second preset similarity, it is indicated that the real doctor's diagnosis may have errors, the real doctor's diagnosis result confirmation information is generated and the preliminary diagnosis result is displayed on the terminal of the on-duty doctor of the registered department, so that the real doctor can reconsider his diagnosis result and judge whether the diagnosis result needs to be modified; upon receiving the real doctor's diagnosis result confirmation information, the real doctor's diagnosis information is taken as the final diagnosis result, and the modified real doctor's diagnosis information is taken as the final diagnosis result.

[0089] In one specific embodiment, in order to further consider the different needs of different users and application scenarios for route planning in the treatment process, the method further comprises:

[0090] After the user completes the registration appointment, he / she will go to the hospital where the registration appointment is located. In order to facilitate the user to quickly reach the registration department after arriving at the hospital, a route planning generation function is set in the APP, so that the user can move according to the planned route; wherein the route planning generation includes two levels of route planning, one is the planning of the route to reach a single fixed target position, such as reaching the registration department or the pharmacy, which corresponds to inputting a single fixed target position operation to generate a first route planning request instruction; the other is the planning of the route to continuously reach multiple detection positions according to the examination items, which corresponds to obtaining the final recommended examination item operation to generate a second route planning request instruction.

[0091] In this embodiment, taking the arrival of the registration department as an example, considering the dense flow of people and the distance the user needs to move, when receiving the first route planning request instruction sent by the user terminal, the single fixed target position (such as the location information of each department) in the hospital where the registered department is located and the current hospital area people flow density are obtained, combined with the determined user object type and the user guide application scenario, a first planning route is generated by using a random algorithm to meet the first constraint condition, which is that the user reaches the single fixed target position (such as the registered registration department) in the hospital where the registered registration department is located. Different types of users have different abilities to safely move in areas with different levels of people flow density. Compared with the elderly, children and people with interaction barriers, adults can safely move in areas with higher people flow density. The first constraint condition includes: for the elderly type or the child type or the interaction barrier type, the people flow density in the area along the planned route is less than the first preset people flow density threshold and the distance to the single fixed target position (such as the registered registration department) in the hospital where the registered registration department is located is the shortest; for the adult type, the people flow density in the area along the planned route is less than the second preset people flow density threshold and the distance to the single fixed target position (such as the registered registration department) in the hospital where the registered registration department is located is the shortest; The first preset people flow density threshold is less than the second preset people flow density threshold, and the specific value can be determined according to the expert setting.

[0092] Upon receiving the second route planning request instruction sent by the user terminal, the location information of each examination item in each area of the hospital, the current hospital area people flow density, and the current waiting time of each examination item are obtained according to the final recommended examination item, combined with the determined user object type and the user guide application scenario, a second planning route is generated by using a random algorithm to meet the second constraint condition, which is that the user completes all examination items; The second constraint condition includes: for the elderly type or the child type, the people flow density in the area along the planned route is less than the first preset people flow density threshold, and the shortest waiting time required to complete all examination items and the shortest distance to complete all examination items; For the adult type, the people flow density in the area along the planned route is less than the second preset people flow density threshold, and the shortest waiting time required to complete all examination items and the shortest distance to complete all examination items.

[0093] In one specific embodiment, in order to protect the user's personal safety information and reduce the dependence of the doctor on the AI intelligence, the method further comprises:

[0094] When the preliminary diagnosis report is sent to the terminal of the doctor on duty in the registered department, the preliminary diagnosis result in the preliminary diagnosis report is in a state of being unlocked within a preset time period, that is, the preliminary diagnosis result in the preliminary diagnosis report is encrypted and hidden, and needs to be inputted with an unlocking password to be displayed; the preset time period can be set artificially and adaptively according to the practice time of different doctors on duty, for example, if the practice time of the doctor on duty is more than a first preset practice time (10 years), a first preset time period (2 minutes) is set correspondingly, otherwise, a second preset time period (5 minutes) is set correspondingly.

[0095] Considering that the doctor on duty needs to refer to the preliminary diagnosis result output by the AI virtual person in the case of a difficult problem, the unlocking password of the preliminary diagnosis result in the preliminary diagnosis report is sent to the terminal of the doctor on duty in the registered department when the preliminary diagnosis result acquisition request sent by the terminal of the doctor on duty in the registered department is received or after the preset time period is exceeded.

[0096] In one specific embodiment, in the intelligent treatment process of the AI virtual person accompanying diagnosis, in addition to providing some auxiliary functions before and during treatment to improve user experience, the AI virtual person can also provide personalized management information to assist users in recovery during the recovery process after treatment and the period between treatment and follow-up visit. The method further comprises:

[0097] The medical order information uploaded by the terminal of the doctor on duty in the registered department is received, and user personalized health management information is generated by using a deep learning algorithm according to the medical order information and the user portrait information; the medical order information includes: diet information, drug taking information and rehabilitation training information, etc.; the user personalized health management information includes: user personalized diet recommendation, user drug taking prompt information, rehabilitation training prompt information, etc.

[0098] In one specific embodiment, considering that there is information that affects user safety such as drug errors during the drug taking process after the user finishes treatment, in order to further ensure the safety of the user during treatment, the method further comprises:

[0099] The staff uploads the drug extraction completion information in the terminal of the drug taking place in the hospital information system of the registered department, the APP receives the drug extraction completion information sent by the hospital information system of the registered department, generates drug checking information and displays it on the user terminal;

[0100] The drug image and the drug checking request instruction uploaded by the user terminal are received, the drug information is identified by using image recognition technology, and whether the identified drug information is the same as the drug information in the drug taking information is compared. If there is a difference, drug error prompt information is generated and sent to the terminal of the drug taking place in the hospital information system of the registered department or the user terminal, so as to remind the staff taking the drug or directly remind the staff taking the drug.

[0101] If the identified drug information includes prescription drug information, in addition to comparing whether the prescription drug information in the identified drug information and the prescription drug information in the drug taking information are the same, it is also necessary to find out whether there is a doctor's signature information in the received medical order information. If there is no doctor's signature information, a drug error prompt information is also generated and sent to the terminal of the dispensing department in the hospital information system of the registered department and the terminal of the on-site doctor of the registered department to remind the dispensing staff to confirm whether the drug taking voucher is correct or to remind the on-site doctor whether there is an error.

[0102] As shown in Figure 2 The embodiment of the application discloses an intelligent treatment system for AI virtual person accompanying diagnosis, which comprises:

[0103] The guidance virtual clinic construction module 101 is used for receiving the intelligent guidance demand instruction sent by the user terminal, constructing the virtual clinic and generating the AI virtual person.

[0104] The guidance required information acquisition module 102 is used for acquiring the user input information in real time after judging that the user enters the virtual clinic based on the operation of the user terminal; the user input information includes the pre-input user portrait information, the user selected intelligent guidance scene information, the user real-time interaction information and the user historical intelligent guidance information.

[0105] The guidance influencing factor judgment module 103 is used for determining the user object type according to the acquired user portrait information and / or user historical intelligent guidance information; the user object type includes: child type, adult type, old person type and interactive obstacle crowd type; determining the user guidance application scene according to the acquired user historical intelligent guidance information and the user selected intelligent guidance scene information; the user guidance application scene includes: ordinary inquiry, emergency inquiry, ordinary reexamination and emergency reexamination.

[0106] The guidance interaction emotion pacification module 104 is used for completing the user abnormal emotion recognition according to the user real-time interaction information by using the deep learning algorithm; the abnormal situation includes anxiety, tension and sadness; based on the matching of the identified abnormal emotion and the determined user object type with the preset emotion pacification strategy, the AI virtual person interacts with the user according to the matched emotion pacification strategy; wherein each user object type and each abnormal emotion matches a preset emotion pacification strategy.

[0107] The diagnosis and recommendation module 105 is configured to match a preset deep learning model according to the determined user diagnosis application scenario; output AI virtual human interactive response information and a preliminary diagnosis result in combination with the matched deep learning model according to real-time interactive information of the user and historical intelligent diagnosis information of the user; wherein the AI virtual human interactive response information output by different preset deep learning models is different in number and the preliminary diagnosis result output is different in time.

[0108] The diagnosis and recommendation module 105 is configured to match a preset deep learning model according to the determined user diagnosis application scenario; output AI virtual human interactive response information and a preliminary diagnosis result in combination with the matched deep learning model according to real-time interactive information of the user and historical intelligent diagnosis information of the user; wherein the AI virtual human interactive response information output by different preset deep learning models is different in number and the preliminary diagnosis result output is different in time.

[0109] The system further comprises:

[0110] The diagnosis and recommendation module 105 is configured to match a preset deep learning model according to the determined user diagnosis application scenario; output AI virtual human interactive response information and a preliminary diagnosis result in combination with the matched deep learning model according to real-time interactive information of the user and historical intelligent diagnosis information of the user; wherein the AI virtual human interactive response information output by different preset deep learning models is different in number and the preliminary diagnosis result output is different in time.

[0111] The real triage result acquisition module 108 is configured to: send the generated preliminary diagnosis report to the hospital information system of the selected appointment department; send the preliminary diagnosis report to the terminal of the doctor on duty of the department for which the appointment is made when receiving the preliminary diagnosis report request instruction sent by the terminal of the doctor on duty of the department for which the appointment is made; the preliminary diagnosis report is generated by extracting key information from the user real-time interaction information and the AI virtual person interaction response information converted into text content by using natural language technology, and combining the preliminary diagnosis result; receive the real doctor recommended examination item uploaded by the terminal of the doctor on duty of the department for which the appointment is made, determine the examination item required for the preliminary diagnosis result according to the medical knowledge graph, compare the determined examination item with the real doctor recommended examination item, if the similarity is lower than the first preset similarity, generate the real doctor recommended examination item reconfirmation prompt information and display the examination item required for the preliminary diagnosis result, and when receiving the real doctor recommended examination item reconfirmation, the real doctor recommended examination item is used as the final recommended examination item; receive the user examination item results uploaded by the hospital information system, output the secondary diagnosis result according to the user examination item results, user real-time interaction information and user historical intelligent triage information, and combine the deep learning algorithm; compare the real doctor diagnosis result uploaded by the terminal of the doctor on duty of the department for which the appointment is made with the generated secondary diagnosis result, if the similarity is lower than the second preset similarity, generate the real doctor diagnosis result reconfirmation information and display the preliminary diagnosis result on the terminal of the doctor on duty of the department for which the appointment is made, and when receiving the real doctor diagnosis result confirmation information, the real doctor diagnosis information is used as the final diagnosis result; and when sending the preliminary diagnosis report to the terminal of the doctor on duty of the department for which the appointment is made, the preliminary diagnosis result in the preliminary diagnosis report is in a to-be-unlocked state within a preset time period, and when receiving the preliminary diagnosis result acquisition request sent by the terminal of the doctor on duty of the department for which the appointment is made or after the preset time period is exceeded, the preliminary diagnosis result unlocking password in the preliminary diagnosis report is sent to the terminal of the doctor on duty of the department for which the appointment is made.

[0112] The guidance medicine inspection module 109 is configured to receive the medicine extraction completion information sent by the hospital information system of the registered department, generate medicine inspection information and display the medicine inspection information on the user terminal; receive the medicine image and the medicine inspection request instruction uploaded by the user terminal, identify the medicine information by using the image recognition technology, compare whether the identified medicine information is same as the medicine information in the medicine taking information, if there is a difference, generate a medicine error prompt information and send the medicine error prompt information to the hospital information system of the registered department or the user terminal; if the identified medicine information includes the prescription medicine information, in addition to comparing whether the prescription medicine information in the identified medicine information is same as the prescription medicine information in the medicine taking information, it is also necessary to find whether there is a doctor's signature information in the received medical order information, if there is no doctor's signature information, the same medicine error prompt information is generated and sent to the hospital information system of the registered department and the doctor terminal of the registered department.

[0113] The guidance health management module 110 is configured to receive the medical order information uploaded by the doctor terminal of the registered department, generate user personalized health management information by using a deep learning algorithm according to the medical order information and the user portrait information; the medical order information includes: dietary taboos information, medicine taking information and rehabilitation training information; the user personalized health management information includes: user personalized recipe recommendation, user medicine taking prompt information and rehabilitation training prompt information.

[0114] The embodiment of the application further discloses a computer readable storage medium.

[0115] Specifically, the computer readable storage medium stores a computer program capable of being loaded and executed by the processor to perform the intelligent treatment method of the AI virtual person accompanying diagnosis described above, and the computer readable storage medium includes, for example, a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk and various storage program codes.

[0116] The embodiment of the application further discloses a computer device.

[0117] Specifically, the computer device includes a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor to perform the intelligent treatment method of the AI virtual person accompanying diagnosis described above.

[0118] The above are preferred embodiments of the application, and do not limit the protection scope of the application, any feature disclosed in the specification (including the abstract and the drawings) can be replaced by other equivalent or similar features unless specifically described. That is, each feature is only an example of a series of equivalent or similar features unless specifically described.

Claims

1. An intelligent treatment method for AI virtual human accompanying diagnosis, characterized in that, Comprise: Receiving the intelligent guide diagnosis demand instruction sent by the user terminal, constructing a virtual clinic and generating an AI virtual person; After judging that the user enters the virtual clinic based on the user terminal operation, real-time user input information is obtained; the user input information includes pre-input user portrait information, user-selected intelligent guide diagnosis scene information, user real-time interaction information, and user historical intelligent guide diagnosis information; According to the obtained user portrait information and / or user historical intelligent guide diagnosis information, the user object type is determined; The user object type includes: child type, adult type, old person type and interactive obstacle population type; according to the obtained user historical intelligent guide diagnosis information and user-selected intelligent guide diagnosis scene information, the user guide diagnosis application scene is determined; the user guide diagnosis application scene includes: general inquiry, emergency inquiry, general reexamination and emergency reexamination; using a deep learning algorithm, user abnormal emotion recognition is completed according to user real-time interaction information, the abnormal emotion includes anxiety, tension, sadness; based on the matching of the identified abnormal emotion and the determined user object type, a preset emotion soothing strategy is matched, so that the AI virtual person interacts with the user according to the matched emotion soothing strategy; wherein each user object type and each abnormal emotion matches a preset emotion soothing strategy; according to the determined user guide diagnosis application scene, a preset deep learning model is matched; according to the user real-time interaction information and the user historical intelligent guide diagnosis information, the AI virtual person interaction response information and the preliminary diagnosis result are output by combining the matched deep learning model; wherein the AI virtual person interaction response information output by different preset deep learning models is different in frequency and the preliminary diagnosis result output time; according to the preliminary diagnosis result, the registration department is determined, and the registration appointment recommendation is generated according to the registration department; receiving the registration appointment request information sent by the user terminal according to the registration appointment recommendation, connecting the selected appointment registration department in the hospital information system and completing the registration appointment; the registration department is determined according to the preliminary diagnosis result, and the registration appointment recommendation is generated according to the registration department, which comprises: Before outputting the preliminary diagnosis result, the last interaction information between the user and the AI virtual person received is used to obtain the geographical location information of the corresponding user terminal; according to the preliminary diagnosis result, the current number of registered appointments of the hospital with the determined registration department is obtained; In combination with the determined user object type, the user guide application scenario, the obtained geographical location information of the user terminal, and the current number of appointments for the hospital with the determined registration department, a registration appointment recommendation order that satisfies a first rule is generated according to the user historical intelligent guide information and the determined registration department; the first rule includes: the hospital where the registration department corresponding to the ordinary reexamination and emergency reexamination in the user guide application scenario is the same as the hospital where the original registration department is located, and the current number of appointments for the hospital with the determined registration department is less than the maximum appointment number threshold of the corresponding hospital matched by the user guide application scenario, and the less the current number of appointments for the hospital with the determined registration department, the more preferentially recommended, and the smaller the distance between the geographical location information of the hospital where the registration department is located and the geographical location information of the user terminal, the more preferentially recommended, and the hospital where the registration department is located belongs to the user object type exclusive hospital and is preferentially recommended; wherein different user guide application scenarios are provided with corresponding matched maximum appointment number thresholds; the priority of the preferential recommendation generated based on the geographical location information of the user terminal, the priority of the preferential recommendation order generated based on the user object type, and the priority of the preferential recommendation generated based on the number of appointments are determined according to the user's pre-set priority level, and the larger the priority level, the more preferentially recommended the order.

2. The intelligent treatment method of AI virtual human accompanying diagnosis according to claim 1, characterized in that, Also includes: The generated preliminary diagnosis report is sent to the terminal of the doctor on duty of the registered department when a preliminary diagnosis report request instruction is received from the terminal of the doctor on duty of the registered department in the hospital information system of the registered department; the preliminary diagnosis report is key information extracted from user real-time interaction information and AI virtual human interaction response information converted into text content by using natural language technology, and is integrated and generated in combination with a preliminary diagnosis result; real doctor recommended examination items uploaded by the terminal of the doctor on duty of the registered department are received, examination items required for a final diagnosis are determined according to a medical knowledge graph, the determined examination items are compared with the real doctor recommended examination items in terms of similarity, if the similarity is lower than a first preset similarity, real doctor recommended examination item reconfirmation prompt information is generated and the examination items required for the final diagnosis are displayed, and when the real doctor recommended examination item reconfirmation is received, the real doctor recommended examination items are taken as the final recommended examination items; when the user's examination item results uploaded by the hospital information system are received, a secondary diagnosis result is output according to the user's examination item results, user real-time interaction information and user historical intelligent triage information in combination with a deep learning algorithm; real doctor diagnosis results uploaded by the terminal of the doctor on duty of the registered department are received, and are compared with the generated secondary diagnosis result in terms of similarity, if the similarity is lower than a second preset similarity, real doctor diagnosis result reconfirmation information is generated and the preliminary diagnosis result is displayed on the terminal of the doctor on duty of the registered department, and when the real doctor diagnosis result reconfirmation information is received, the real doctor diagnosis information is taken as the final diagnosis result.

3. The intelligent treatment method of AI virtual human accompanying diagnosis according to claim 2, characterized in that, Also includes: When a first route planning request instruction is received from the user terminal, a single fixed target position in the hospital where the registered department is located and a current hospital area crowd density are obtained, a first planning route of the user reaching the single fixed target position in the hospital where the registered department is located is generated by using a random algorithm in combination with the determined user object type and the user triage application scenario, which satisfies a first constraint condition; The first constraint condition includes: for the elderly type or the child type, the crowd density in the area along the planning route is less than a first preset crowd density threshold and the distance to the single fixed target position in the hospital where the registered department is located is the shortest; for the adult type, the crowd density in the area along the planning route is less than a second preset crowd density threshold and the distance to the single fixed target position in the hospital where the registered department is located is the shortest; the first preset crowd density threshold is less than the second preset crowd density threshold; When the second route planning request instruction sent by the user terminal is received, the final recommended examination items are checked to obtain the location information of each examination item in each area of the hospital, the current crowd density of each area of the hospital, and the current waiting time of each examination item, the determined user object type and the user guide application scenario are combined, and a random algorithm is used to generate a second planned route for the user to complete all the examination items, which satisfies a second constraint condition; the second constraint condition includes: for the elderly type or the child type, the crowd density in the area along the planned route is less than a first preset crowd density threshold, and the shortest waiting time and the shortest distance along the route for completing all the examination items are required; for the adult type, the crowd density in the area along the planned route is less than a second preset crowd density threshold, and the shortest waiting time and the shortest distance along the route for completing all the examination items are required.

4. The intelligent treatment method of AI virtual human accompanying diagnosis according to claim 1, characterized in that, Also includes: When the preliminary diagnosis report is sent to the terminal of the on-site doctor of the booked registration department, the preliminary diagnosis result in the preliminary diagnosis report is in a to-be-unlocked state within a preset time period, and when the preliminary diagnosis result acquisition request sent by the terminal of the on-site doctor of the booked registration department is received or after the preset time period is exceeded, the preliminary diagnosis result unlocking password in the preliminary diagnosis report is sent to the terminal of the on-site doctor of the booked registration department.

5. The intelligent treatment method of AI virtual human accompanying diagnosis according to claim 1, characterized in that, Also includes: Receiving the medical order information uploaded by the terminal of the on-site doctor of the booked registration department, generating user personalized health management information by using a deep learning algorithm according to the medical order information and the user portrait information; The medical order information includes: dietary information, drug taking information and rehabilitation training information; the user personalized health management information includes: user personalized diet recommendation, user drug taking prompt information and rehabilitation training prompt information.

6. The intelligent treatment method of AI virtual human accompanying diagnosis according to claim 5, characterized in that, Also includes: Receiving the drug extraction completion information sent by the hospital information system of the booked registration department, generating drug examination information and displaying it on the user terminal; Receiving the drug image and the drug examination request instruction uploaded by the user terminal, identifying the drug information by using image recognition technology, comparing whether the identified drug information and the drug information in the drug taking information are the same, if there is a difference, generating a drug error prompt information and sending it to the hospital information system of the booked registration department or the user terminal; if the identified drug information includes prescription drug information, in addition to comparing whether the prescription drug information in the identified drug information and the prescription drug information in the drug taking information are the same, it is also necessary to find whether there is a doctor's signature information in the received medical order information, if there is no doctor's signature information, a drug error prompt information is also generated and sent to the hospital information system of the booked registration department and the terminal of the on-site doctor of the booked registration department.

7. An intelligent treatment system for AI virtual human accompanying diagnosis, characterized in that, Includes: A guide virtual clinic construction module for receiving an intelligent guide demand instruction sent by a user terminal, constructing a virtual clinic and generating an AI virtual person; The information acquisition module is used for acquiring the user input information in real time after the user enters the virtual clinic based on the user terminal operation; the user input information includes the user portrait information input in advance, the intelligent guide diagnosis scene information selected by the user, the real-time interaction information of the user, and the historical intelligent guide diagnosis information of the user; The guide diagnosis influence factor judgment module is used for determining the user object type according to the acquired user portrait information and / or the historical intelligent guide diagnosis information of the user; The user object type includes the child type, the adult type, the old person type, and the interactive obstacle population type; the user guide diagnosis application scene is determined according to the acquired historical intelligent guide diagnosis information of the user and the intelligent guide diagnosis scene information selected by the user; the user guide diagnosis application scene includes the general inquiry, the emergency inquiry, the general reexamination, and the emergency reexamination; The guide diagnosis interaction emotion pacification module is used for completing the user abnormal emotion recognition according to the real-time interaction information of the user by using the deep learning algorithm; the abnormal emotion includes anxiety, tension, and sadness; the emotion pacification strategy is matched according to the recognized abnormal emotion and the determined user object type, so that the AI virtual person interacts with the user according to the matched emotion pacification strategy; wherein each user object type is matched with one preset emotion pacification strategy for each abnormal emotion; The guide diagnosis diagnosis and recommendation module is used for matching the preset deep learning model according to the determined user guide diagnosis application scene; the AI virtual person interaction response information and the preliminary diagnosis result are output according to the real-time interaction information of the user and the historical intelligent guide diagnosis information of the user, combined with the matched deep learning model; wherein the AI virtual person interaction response information times and the preliminary diagnosis result output time output by different preset deep learning models are different. The guidance registration appointment operation module is configured to determine a registration department according to the preliminary diagnosis result, and generate a registration appointment recommendation according to the registration department; receive a registration appointment request information sent by the user terminal according to the registration appointment recommendation, connect a hospital information system of the selected registration department, and complete the registration appointment; the determination of the registration department according to the preliminary diagnosis result and the generation of the registration appointment recommendation according to the registration department include: before the output of the preliminary diagnosis result, receiving the last interaction information between the user and the AI virtual person to obtain the geographical location information of the user terminal; obtaining the current registration appointment quantity of the hospital with the determined registration department according to the preliminary diagnosis result; combining the determined user object type and the user guidance application scenario, the obtained geographical location information of the user terminal, and the current registration appointment quantity of the hospital with the determined registration department, generating a registration appointment recommendation sequence that satisfies the first rule according to the user historical intelligent guidance information and the determined registration department; the first rule includes: the registration department corresponding to the ordinary re-examination and the emergency re-examination in the user guidance application scenario is in the same hospital as the original registration department, and the current registration appointment quantity of the hospital with the determined registration department is less than the maximum registration appointment quantity threshold of the corresponding hospital matched by the user guidance application scenario, and the less the current registration appointment quantity of the hospital with the determined registration department, the higher the priority of the recommendation, and the smaller the distance between the geographical location information of the hospital with the registration department and the geographical location information of the user terminal, the higher the priority of the recommendation, and the hospital with the registration department belongs to the user object type exclusive hospital, which is preferentially recommended; wherein, different user guidance application scenarios are provided with corresponding matched maximum registration appointment quantity thresholds; the priority of the priority recommendation generated based on the geographical location information of the user terminal, the priority of the priority recommendation sequence generated based on the satisfaction of the user object type, and the priority of the priority recommendation generated based on the registration appointment quantity are determined according to the user's pre-set priority level size, and the larger the priority level, the earlier the corresponding priority recommendation sequence.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium includes a stored computer program, wherein the computer readable storage medium controls the device where the computer readable storage medium is located to execute the method of any one of claims 1 to 6 when the computer program is running.

9. A computer device, comprising: The computer device includes a memory, a processor, and a program stored on the memory and executable by the processor, and the program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images