Intelligent robot-based auxiliary treatment method, device and related equipment

Abstract

The present application belongs to the field of digital medical technology, and provides a method, apparatus and related equipment for assisting medical consultation based on an intelligent robot. In order to solve the problem that traditional technology cannot efficiently and automatically determine online which department of the hospital the patient should go to for medical consultation, a preset intelligent robot is used to respond to the patient's medical consultation assistance request corresponding to the registration, obtain the patient's symptom information data, and determine whether the symptom information data is multimodal data. When the symptom information data is multimodal data, based on a preset multimodal deep learning model and a preset medical consultation assistance knowledge graph, and according to the symptom information data, the registration suggestion corresponding to the medical consultation assistance request is determined, and then the registration suggestion is fed back to the patient. This can efficiently and accurately determine online which department of the hospital the patient should go to for medical consultation to provide medical consultation assistance suggestions, and can improve the efficiency and accuracy of registration processing in the case of a large number of patients visiting the hospital.

Description

Technical Field

[0001] The present application relates to the field of digital medical technology, and in particular to an intelligent robot-based auxiliary medical treatment method, device, computer equipment and computer-readable storage medium. Background Art

[0002] With the development of modern medical care, the divisions of hospital departments are becoming more and more detailed and professional, and new departments are constantly emerging. For ordinary patients with low professional skills, they can only describe their symptoms. In many cases, they are not clear about which department of the hospital they should go to for registration and treatment. Even if they consult the nurses or guides in the hospital, faced with the various symptoms provided by the patients and the division of professional departments, sometimes relatively professional nurses or guides find it difficult to give accurate advice.

[0003] In intelligent digital healthcare, especially with the rapid increase in the number of patients, the use of artificial intelligence to provide automated medical assistance based on patients' self-reported symptoms has significant practical value. When patients register at a hospital and then seek medical treatment, they generally have two requirements: first, the department they should visit must be accurately determined; second, this determination must be quick and efficient. This is especially true for in-person registration, where determining the appropriate department is time-consuming and inappropriate.

[0004] However, in traditional technology, with the development of artificial intelligence technology, various data of patients are generally processed based on artificial intelligence to assist doctors in diagnosis. The processing is relatively complex and time-consuming, and then relatively professional auxiliary diagnosis results are given. The auxiliary diagnosis results are used for doctors or patients to have a preliminary understanding and reference for the diagnosis of the disease, and are not suitable for telling patients which department of the hospital to go to for registration and treatment.

[0005] Therefore, how to automatically and efficiently and accurately tell patients which hospital department is suitable for registration and treatment online has become a technical problem that digital medicine urgently needs to solve. Summary of the Invention

[0006] The present application provides a method, apparatus, computer equipment and computer-readable storage medium for assisting medical consultation based on an intelligent robot, which can solve the technical problem in traditional technology that it is impossible to efficiently and automatically determine online which department of the hospital a patient should go to for medical consultation.

[0007] In the first aspect, the present application provides a method for assisting medical consultation based on an intelligent robot, comprising: based on a preset intelligent robot, responding to a patient's request for assistance in registering, and obtaining the patient's symptom information data; determining whether the symptom information data is multimodal data; when the symptom information data is multimodal data, based on a preset multimodal deep learning model and a preset medical knowledge graph for assistance in medical consultation, and according to the symptom information data, determining the registration suggestion corresponding to the request for assistance in medical consultation; and feeding back the registration suggestion to the patient.

[0008] In the second aspect, the present application provides a medical consultation assistance processing device based on an intelligent robot, including: a first acquisition unit, used to respond to the patient's medical consultation assistance request corresponding to the registration based on a preset intelligent robot, and obtain the patient's symptom information data; a first judgment unit, used to judge whether the symptom information data is multimodal data; a first determination unit, used to determine the registration suggestion corresponding to the medical consultation assistance request based on a preset multimodal deep learning model and a preset medical consultation assistance knowledge graph, and according to the symptom information data when the symptom information data is multimodal data; a feedback unit, used to feed back the registration suggestion to the patient.

[0009] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the medical consultation assistance method based on an intelligent robot are implemented.

[0010] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the intelligent robot-based medical assistance processing method.

[0011] The present application provides a method, apparatus, computer equipment and computer-readable storage medium for medical consultation assistance processing based on an intelligent robot. The method responds to a patient's medical consultation assistance request corresponding to registration based on a preset intelligent robot, obtains the patient's symptom information data, and determines whether the symptom information data is multimodal data. If the symptom information data is multimodal data, based on a preset multimodal deep learning model and a preset medical consultation assistance knowledge graph, and according to the symptom information data, determines the registration suggestion corresponding to the medical consultation assistance request, and then feeds back the registration suggestion to the patient. Based on the obtained multimodal information of the patient and the preset medical consultation assistance knowledge graph including the hospital department settings, and using multimodal data processing based on machine learning, it is possible to comprehensively determine which department of the hospital the patient should go to for medical consultation in combination with the hospital department settings. The method can efficiently and accurately automatically determine which department of the hospital the patient should go to for medical consultation online to provide medical consultation assistance suggestions, improve the efficiency and accuracy of registration processing in the case of a large number of patients, and improve the level and quality of automated medical services based on digital medicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0013] Figure 1 A flowchart of a method for assisting with medical consultation based on an intelligent robot provided in an embodiment of the present application;

[0014] Figure 2 This is a schematic diagram of the first sub-flow of the intelligent robot-based medical assistance processing method provided in an embodiment of the present application;

[0015] Figure 3 A schematic diagram of the second sub-flow of the intelligent robot-based medical consultation assistance method provided in an embodiment of the present application;

[0016] Figure 4 A schematic block diagram of an intelligent robot-based medical consultation assistance device provided in an embodiment of the present application;

[0017] Figure 5 A schematic block diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0020] The embodiment of the present application provides a method for assisting medical consultations based on an intelligent robot. The method can be applied to, but not limited to, computer devices such as smart phones, laptops, desktop computers, servers, intelligent robots, and intelligent medical guidance equipment, and can be used in the field of digital medicine, including but not limited to assisting medical consultations when providing self-service medical services.

[0021] Faced with the technical problem that traditional technology cannot efficiently and automatically determine online which department of the hospital the patient should go to for treatment, the inventor proposed the medical assistance processing method based on an intelligent robot in the embodiment of this application. The core idea of ​​the embodiment of this application is: based on the multimodal information provided and automatically collected by the patient, based on the medical assistance medical knowledge graph, using multimodal data processing based on machine learning, and combined with the department settings of the hospital, to comprehensively determine which department of the hospital the patient should go to for treatment. It can efficiently and accurately determine online which department of the hospital the patient should go to for treatment to provide medical assistance suggestions. Furthermore, the medical assistance medical knowledge graph is dynamically updated according to the feedback of the medical assistance, which can further improve the accuracy of the medical assistance suggestions and improve the registration processing efficiency in the case of a large number of patients.

[0022] The following embodiments of the present application are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0023] See also Figure 1 , Figure 1 This is a flow chart of the intelligent robot-based medical assistance processing method provided in the embodiment of the present application. Figure 1 As shown, the method includes but is not limited to the following steps S11-S15:

[0024] S11. Based on the preset intelligent robot, respond to the patient's request for medical assistance corresponding to the registration, and obtain the patient's disease information data.

[0025] Explanatoryly, when a patient registers for medical treatment at a medical institution including but not limited to a hospital, if there is a long queue of people waiting in line for registration at the medical institution, the patient can register based on a self-service device, or can make an automatic appointment for registration through an APP or other application. If the patient is not sure which department to register for treatment based on the representation corresponding to the symptoms he or she exhibits, the patient can initiate a request for medical assistance advice based on the self-service provided by a preset intelligent robot including but not limited to the self-service device, APP application, etc. The preset intelligent robot responds to the patient's medical assistance request corresponding to the registration and obtains the patient's symptom information data, which includes but is not limited to identity feature information data, representation information data, historical medical information data, and physiological signal information data. The identity feature information data represents the patient's identity feature information, which includes but is not limited to age, gender, and occupation. The representation information data represents the patient's symptoms. The corresponding characterization information data, the characterization information is not limited to headache, sweating, weakness, cough, historical medical information data represents the data of historical medical information when the patient has visited the medical unit or the data of valid past examination information that the patient can provide, the historical medical information data includes medical records in text form and medical imaging examination images in image form, the data of valid past examination information that the patient can provide includes but is not limited to medical imaging examinations between mutually recognized medical units within the validity period and examinations such as but not limited to blood routine and liver function, among which, mutually recognized medical units within the validity period refer to different medical units that recognize each other's examination results within the preset validity period of the examination results, medical imaging examination images include but are not limited to X-ray, CT, B-ultrasound, and magnetic resonance imaging examination images, physiological signal information data represents the data of patient physiological signal information that can be automatically collected on site, physiological signal information is but not limited to body temperature, blood pressure, respiration, pulse, and blood oxygen.

[0026] S12: Determine whether the disease information data is multimodal data.

[0027] Explanatoryally, multimodal data refers to data that contains at least two of the following data types: text, voice, image, video, and physiological signals. Similarly, unimodal data refers to data that contains one of the following data types: text, voice, image, video, and physiological signals.

[0028] After obtaining the patient's symptom information data, determine whether the symptom information data includes at least but not limited to data of the two data types described above. If the symptom information data includes at least but not limited to data of the two data types described above, determine that the symptom information data is multimodal data. If the symptom information data includes but not limited to data of one data type described above, determine that the symptom information data is not multimodal data.

[0029] S13. When the symptom information data is multimodal data, determining a registration suggestion corresponding to the medical assistance request based on a preset multimodal deep learning model and a preset medical knowledge graph for medical assistance, and according to the symptom information data;

[0030] S14. When the symptom information data is not multimodal data, based on the preset unimodal deep learning model and the preset medical knowledge graph for medical assistance, and according to the symptom information data, determine the registration suggestion corresponding to the medical assistance request.

[0031] Explanatory, a multimodal deep learning model is pre-set, that is, a preset multimodal deep learning model. The preset multimodal deep learning model represents a model that can process at least two data types of data among data types such as text, voice, image, video, physiological signals, etc. The preset multimodal deep learning model can process relatively complex data types and data content, and combine data of different data types to obtain the final result. It can improve the efficiency and effect of processing complex data, and improve the efficiency of processing data containing multiple data types. For example, for the patient's symptom information data containing three data types of voice, text, and video, a preset multimodal deep learning model is used to process the symptom information data. The content of the three data types of voice, text, and video can be combined to determine the patient's corresponding registration recommendation, thereby using the patient's relatively rich information to make a more accurate judgment on the patient, thereby improving the accuracy of the registration recommendation. Among them, the preset multimodal deep learning model can be implemented using a multimodal convolutional neural network, and the structure of the multimodal convolutional neural network can be expressed by the following formula (1):

[0032]

[0033] Among them, x i is the input information of the i-th mode, F i (x i ) is the feature extractor of the i-th modality, f i is the eigenvector of the ith mode, M is the number of input modes, H(f1,…,f M ) is a multimodal feature fusion device, h is a multimodal feature vector, G j (n) is the result generator of the jth mode, y jis the output information of the jth modality, and N is the number of output modalities. Furthermore, a multimodal convolutional neural network consists of three parts: a feature extractor, a feature fusion unit, and a result generator. The feature extractor is responsible for extracting feature vectors from the input information of each modality, the feature fusion unit is responsible for concatenating the feature vectors of different modalities into a multimodal feature vector, and the result generator is responsible for generating the output information of each modality based on the multimodal feature vector.

[0034] A single-modal deep learning model is pre-set, that is, a preset single-modal deep learning model. The preset single-modal deep learning model represents a data processing method that can only process one type of data, such as text, voice, image, video, physiological signal, etc. Since only a single data type needs to be processed, the preset single-modal deep learning model processes data relatively efficiently and accurately.

[0035] A medical knowledge graph for auxiliary medical consultation is pre-set, that is, a preset medical knowledge graph for auxiliary medical consultation. The preset medical knowledge graph for auxiliary medical consultation represents a graph based on medical knowledge used to assist in determining registration suggestions for registration and medical consultation. The preset medical knowledge graph for auxiliary medical consultation includes but is not limited to the following entities and the relationships between them: symptoms, diseases, symptom sites, medical unit department configuration, physicians, number of physician registrations, and consultation time. Among them, the disease represents the diagnosis result of the patient's condition, the symptom represents the symptoms corresponding to the disease, the symptom site represents the part of the body where the disease occurs, several symptoms correspond to one symptom site, and one symptom site corresponds to several diseases, and the correspondence between symptoms, diseases and symptom sites needs to be sorted out by medical personnel based on professional knowledge. The medical unit department configuration represents the department setting for consultation configured by the medical unit. The departments set up by each medical unit may be different. Each department will be equipped with several physicians, each physician will arrange consultation time, and each consultation time will be equipped with a corresponding number of registrations. The consultation time is the predicted medical consultation time. The preset auxiliary medical knowledge graph can be represented by a directed graph, in which the nodes represent the entities described above, the edges represent the relationships between the entities, and the attributes can represent the corresponding features involved.

[0036] Based on the above settings, when the symptom information data is multimodal data, based on the preset multimodal deep learning model and the preset medical knowledge graph for auxiliary consultation, and according to the symptom information data, the registration suggestion corresponding to the medical assistance request is determined. When the symptom information data is not multimodal data, based on the preset unimodal deep learning model and the preset medical knowledge graph for auxiliary consultation, and according to the symptom information data, the registration suggestion corresponding to the medical assistance request is determined, wherein the registration suggestion includes but is not limited to a summary description of the patient's main symptoms, the corresponding predicted symptom site, the predicted disease, the recommended registration department, the registration physician selection, and the registration time selection, so that the corresponding symptom information data processing method can be flexibly adopted according to the different situations of the patient's corresponding symptom information data, that is, When the symptom information data is multimodal data, processing based on a preset multimodal deep learning model can improve the efficiency of processing rich and complex data, and can combine different types of data to determine the content corresponding to the patient's registration recommendation, and obtain registration recommendations, thereby using the patient's relatively rich information to make a more accurate judgment on the patient's symptom location, to improve the accuracy of registration recommendations. When the symptom information data is not multimodal data, that is, when the symptom information data is unimodal data, processing based on a preset unimodal deep learning model and determining the registration recommendations corresponding to the medical assistance request based on the symptom information data can give full play to the relatively efficient and accurate data processing characteristics of the preset unimodal deep learning model, and improve the efficiency of determining registration recommendations.

[0037] It should be noted that the predicted disease involved in the embodiments of the present application means predicting the possible disease of the patient, which is used as a reference for registering the patient or as a basis for giving corresponding registration recommendations to the patient, and is not the result of disease diagnosis.

[0038] S15. Feedback the registration suggestion to the patient.

[0039] Explanatoryally, registration suggestions are fed back to patients based on a preset intelligent robot. The feedback methods include but are not limited to displaying through a preset display interface of the self-service device, broadcasting through audio, or displaying and broadcasting through video based on the preset intelligent robot, so as to feed back registration suggestions to patients, and then the patients register according to the registration suggestions and receive corresponding medical treatment.

[0040] Furthermore, feeding back the registration suggestion to the patient includes:

[0041] Based on a preset multimodal approach, the registration suggestion is fed back to the patient.

[0042] Explanatoryally, the preset multimodal method includes but is not limited to at least two of the following methods: text method, voice broadcast method, and video playback method, and then the preset multimodal method is used to feed back the registration suggestion to the patient.

[0043] For example, in one example, based on the preset display interface of the self-service device, the registration suggestion is displayed in text form and simultaneously announced by voice, thereby providing feedback to the patient through a combination of text display and voice announcement. In another example, based on the preset display interface of the self-service device, the registration suggestion is displayed in text form and simultaneously announced by video playback.

[0044] Based on the preset multimodal method, the registration suggestions are fed back to the patients, which can make the feedback of the registration suggestions rich, vivid, lively and intuitive, and have a better communication effect, thereby improving the self-service level and quality of digital medical care.

[0045] In an embodiment of the present application, a preset intelligent robot responds to a patient's request for medical assistance corresponding to registration, obtains the patient's symptom information data, and determines whether the symptom information data is multimodal data. If the symptom information data is multimodal data, based on a preset multimodal deep learning model and a preset medical assistance knowledge graph, and according to the symptom information data, a registration suggestion corresponding to the medical assistance request is determined, and then the registration suggestion is fed back to the patient. Based on the obtained multimodal information of the patient and the preset medical assistance knowledge graph including the hospital department settings, and using multimodal data processing based on machine learning, it is possible to comprehensively determine which department of the hospital the patient should go to for medical treatment in combination with the hospital department settings. This can efficiently and accurately determine which department of the hospital the patient should go to for medical treatment online to provide medical assistance suggestions, and can improve the efficiency and accuracy of registration processing in situations where there are a large number of patients and limited guides, thereby improving the level and quality of automated medical services based on digital medicine.

[0046] In one embodiment, obtaining the patient's symptom information data includes:

[0047] Prompting the patient to provide characterization information data corresponding to the condition;

[0048] Acquiring the characterization information data and collecting physiological signal data corresponding to the patient;

[0049] The characterization information data and the physiological signal data are used as the disease information data corresponding to the patient.

[0050] Explanatoryally, when a patient initiates a request for auxiliary medical advice, the patient is prompted to provide representational information data corresponding to the condition. The patient can describe his or her symptoms by text or voice, and upload the text, voice and other representational information data, and then the preset intelligent robot obtains the representational information data. At the same time, based on the preset intelligent robot, the patient's corresponding physiological signal data, including but not limited to body temperature, blood pressure, respiration, pulse, and blood oxygen, is automatically collected, and the obtained representational information data and the automatically collected physiological signal data are used as the patient's corresponding disease information data. The disease information data represents data reflecting the patient's symptom information.

[0051] For example, at a medical institution, when a request for medical assistance advice is initiated based on a self-service device, the patient can describe the symptoms he or she feels through text or voice, and upload the text, voice and other representative information data. At the same time, based on the automatic collection function integrated into the self-service device itself, the patient's corresponding physiological signal data, including but not limited to body temperature, blood pressure, etc., is automatically collected, so as to obtain as much multi-dimensional disease information data of the patient as possible.

[0052] The embodiment of the present application obtains the characterization information data provided by the patient and automatically collects the patient's corresponding physiological signal data, thereby obtaining as much multi-dimensional and rich disease information data as possible about the patient, and combining the data of different dimensions to determine the corresponding registration recommendation for the patient, thereby using the relatively rich information corresponding to the patient to make a more accurate judgment on the location of the patient's disease, and can efficiently and accurately determine online which department of the hospital the patient should go to for treatment to provide auxiliary treatment suggestions, thereby improving the accuracy of registration recommendations.

[0053] In one embodiment, determining whether the disease information data is multimodal data includes:

[0054] Identify the disease condition data type corresponding to the disease condition information data;

[0055] Determining whether the disease condition data type includes at least two data types;

[0056] When the disease condition data type includes at least two data types, determining that the disease condition information data is multimodal data;

[0057] If the disease condition data type does not include at least two data types, it is determined that the disease condition information data is not multimodal data.

[0058] Explanatoryly, the data types contained in the symptom information data are identified, the disease data types are obtained, and the number of disease data types contained in the symptom information data is counted. The disease data types include but are not limited to text, voice, image, video, and physiological signals. Then, it is determined whether the disease data type contains at least two data types. In the case that the disease data type contains at least two data types, the symptom information data is determined to be multimodal data. In the case that the disease data type does not contain at least two data types, that is, the symptom information data is a single data type, that is, the symptom information data is text, voice or video, it is determined that the symptom information data is not multimodal data, that is, the symptom information data is determined to be unimodal data.

[0059] The embodiment of the present application determines whether the symptom information data is multimodal data, and then flexibly adopts the corresponding symptom information data processing method according to the different situations of the symptom information data corresponding to the patient based on whether the symptom information data is multimodal data or unimodal data. It can efficiently and accurately determine online which department of the hospital the patient should go to for treatment to provide auxiliary treatment suggestions, thereby improving the accuracy of registration suggestions.

[0060] In one embodiment, see Figure 2 , Figure 2 This is a schematic diagram of the first sub-flow of the intelligent robot-based medical assistance processing method provided in the embodiment of the present application. Figure 2 As shown, the symptom information data includes the patient's identity feature information; based on a preset multimodal deep learning model and a preset consultation assistance medical knowledge graph, and according to the symptom information data, determining the registration suggestion corresponding to the consultation assistance request includes:

[0061] S21. Determine the patient type corresponding to the patient based on the identity feature information;

[0062] S22, determining the first type of data and the second type of data included in the disease information data;

[0063] S23. Determine, based on the patient type, a preset first multimodal deep learning model corresponding to the first type of data, and process the first type of data based on the preset first multimodal deep learning model to obtain first symptom information;

[0064] S24. Determine, based on the patient type, a preset second multimodal deep learning model corresponding to the second type of data, and process the second type of data based on the preset second multimodal deep learning model to obtain second symptom information;

[0065] S25. Determine a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information;

[0066] S26. Determine the registration suggestion corresponding to the medical assistance request based on the preset medical assistance knowledge graph.

[0067] Explanation-wise, the symptom information data includes the patient's identity characteristic information, which represents the patient's identity characteristics. The identity characteristic information includes but is not limited to gender, age, and occupation. Since some diseases have group characteristics, for example, geriatric diseases have the characteristics of the elderly group, occupational diseases have occupational commonalities, and gynecological diseases and men's diseases each have gender group characteristics, the characteristics and probability of the user suffering from a certain group disease can be inferred from the identity characteristic information.

[0068] According to the patient's identity characteristic information, the corresponding patient type of the patient is determined. The patient types include but are not limited to elderly men, elderly women, middle-aged men, middle-aged women, young men, young women, male teenagers, female teenagers, and occupational groups. By first classifying the patients into patient types according to their identity characteristic information, the corresponding symptom sites of the same symptoms but applicable to different types of diseases can be predicted, which can make the registration recommendations more accurate with a certain probability.

[0069] Then, the first and second types of data included in the symptom information data are determined. The symptom information data includes, but is not limited to, the first and second types of data. The symptom information data may also include third and fourth types of data, but the symptom information data at least includes the first and second types of data. The first and second types of data include, but are not limited to, text, voice, image, video, and physiological signals, respectively. It should be noted that the terms "first" and "second" in the first and second types of data are merely used to distinguish different types of data and are not used to limit the types of data. Other similar descriptions apply analogously.

[0070] For each type of patient and each type of data corresponding to each type of patient, a corresponding preset multimodal deep learning model is set. The preset multimodal deep learning model is used to process the corresponding type of data of the corresponding patient type. Each type of data includes the same type of data or similar type of data according to the setting. For example, since audio is eventually converted into text for processing, audio and text are similar types of data and the same preset multimodal deep learning model can be used. For another example, since video is eventually converted into image frames for processing, video and image are similar types of data and the same preset multimodal deep learning model can be used, thereby using the same preset multimodal deep learning model for the same or similar types of data. Using a multimodal deep learning model for processing can make the processing of each type of data corresponding to each type of patient more accurate and easier to train the model. Then, based on the patient type, the preset first multimodal deep learning model corresponding to the first type of data is determined, and the first type of data is processed based on the preset first multimodal deep learning model to obtain first symptom information. Symptom information includes but is not limited to symptom summary, symptom site prediction, and disease prediction. Symptom site prediction is the part of the patient's body that may have a symptom predicted based on the first type of data. Disease prediction means predicting the patient's possible disease for reference or as a basis for the patient's corresponding registration recommendation, and is not the final result of the disease diagnosis. Similarly, based on the patient type, the preset second multimodal deep learning model corresponding to the second type of data is determined and called, and the second type of data is processed based on the preset second multimodal deep learning model to obtain second symptom information.

[0071] Then, based on the first symptom information and the second symptom information, the first symptom information and the second symptom information are combined to determine and call several corresponding preset auxiliary medical knowledge graphs for medical consultation. The preset auxiliary medical knowledge graph for medical consultation represents the information required for registration of medical units such as the consultation department, corresponding symptoms, and visiting physicians set up by the hospital, which are described in a preset graph structure. Then, based on the preset auxiliary medical knowledge graph for medical consultation, the first symptom information and the second symptom information are matched with the preset auxiliary medical knowledge graph for medical consultation to obtain registration suggestions, thereby determining the registration suggestions corresponding to the auxiliary medical consultation request. The registration suggestions include but are not limited to registration information such as the consultation department, visiting physician, and consultation time.

[0072] In an embodiment of the present application, patients are first classified according to their identity feature information to obtain patient types, and then the corresponding multimodal deep learning model is determined according to the patient type and the data type corresponding to the symptom information data. Thus, for different data types of different patient types, corresponding multimodal deep learning models are used to process the symptom information data, and combined with a preset auxiliary medical knowledge graph for medical consultation, registration suggestions are determined. According to different situations of different types of symptom information data corresponding to different patient types, the corresponding multimodal deep learning model for processing symptom information data can be flexibly adopted. It can efficiently and accurately automatically determine online which department of the hospital the patient should go to for medical consultation to provide auxiliary medical consultation suggestions, thereby improving the accuracy of registration suggestions.

[0073] In one embodiment, determining a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information includes:

[0074] Determining whether there is a preset symptom association relationship between the first symptom information and the second symptom information;

[0075] If there is a preset symptom association relationship between the first symptom information and the second symptom information, obtaining a preset auxiliary medical knowledge graph containing the first symptom information and the second symptom information, and using the preset auxiliary medical knowledge graph corresponding to the auxiliary medical knowledge graph;

[0076] When there is no preset symptom association relationship between the first symptom information and the second symptom information, a preset first auxiliary medical knowledge graph containing the first symptom information is obtained, and a preset second auxiliary medical knowledge graph containing the second symptom information is obtained, and the preset first auxiliary medical knowledge graph and the preset second auxiliary medical knowledge graph are used as the preset auxiliary medical knowledge graphs corresponding to the auxiliary medical request.

[0077] Explanatoryly, a medical knowledge graph for auxiliary medical treatment is pre-set, that is, a preset medical knowledge graph for auxiliary medical treatment, which contains the relationship between symptoms, diseases and symptom sites, and different symptoms can individually or jointly determine a disease or symptom site and the corresponding medical department and medical doctor and other related registration requirements.

[0078] Pre-set symptom associations, also known as preset symptom associations, represent the associations that exist between different symptoms. This association allows different symptoms to collectively point to a symptom site and the disease corresponding to that site. These different symptoms are different manifestations of the same disease at a symptom site, so that these different symptoms can collectively identify a symptom site or the same disease. Pre-set symptom associations should generally be set by physicians in various departments based on their medical expertise. For example, in one example, the symptom "itch" and the symptom "rash" may both point to the skin and indicate the possibility of an allergic skin disease. In another example, the symptom "cough," the symptom "fever," and the symptom "headache" may collectively point to the disease corresponding to a cold.

[0079] Determine whether there is a preset symptom association relationship between the first symptom information and the second symptom information, wherein the "first" and "second" involved in the first symptom information and the second symptom information are only used to distinguish different symptom information and are not used to limit different symptom information, and other similar descriptions are similar.

[0080] When there is a preset symptom association relationship between the first symptom information and the second symptom information, a preset auxiliary medical knowledge graph containing the first symptom information and the second symptom information is obtained and used as the preset auxiliary medical knowledge graph corresponding to the auxiliary medical request, and then the registration suggestion is determined based on the preset auxiliary medical knowledge graph.

[0081] In the case where there is no preset symptom association relationship between the first symptom information and the second symptom information, that is, the first symptom information and the second symptom information each point to a different symptom site and the disease corresponding to the site, a preset first consultation auxiliary medical knowledge graph containing the first symptom information is obtained, and a preset second consultation auxiliary medical knowledge graph containing the second symptom information is obtained, and the preset first consultation auxiliary medical knowledge graph and the preset second consultation auxiliary medical knowledge graph are used as the preset consultation auxiliary medical knowledge graphs corresponding to the consultation auxiliary request, and then the registration suggestion is determined based on the preset consultation auxiliary medical knowledge graph, which is particularly suitable for situations where it is possible to register in this department or that department. For example, in one example, for skin problems, sometimes you can go to the dermatology department for treatment, and sometimes you can go to the plastic surgery department for treatment. In another example, for headache problems, sometimes you can go to the internal medicine department for treatment, and sometimes you can go to the neurology department for treatment. In such similar cases, the registration suggestions corresponding to the preset first consultation auxiliary medical knowledge graph and the preset second consultation auxiliary medical knowledge graph can be given, and the patient can choose by himself.

[0082] In the embodiment of the present application, the corresponding preset auxiliary medical knowledge graphs for medical consultation are determined respectively through the association relationship between the first symptom information and the second symptom information corresponding to different situations. The corresponding preset auxiliary medical knowledge graphs for medical consultation can be flexibly adopted according to the different symptom information of the patients. It can efficiently and accurately automatically determine online which department of the hospital the patient should go to for medical consultation to provide auxiliary medical consultation suggestions, thereby improving the accuracy of registration suggestions.

[0083] In one embodiment, see Figure 3 , Figure 3 This is a schematic diagram of the second sub-flow of the intelligent robot-based medical assistance processing method provided in the embodiment of the present application. Figure 3 As shown, in this embodiment, according to the preset consultation assistance medical knowledge graph, determining the registration suggestion corresponding to the consultation assistance request includes:

[0084] S31. Obtaining a first predicted disease and a second predicted disease corresponding to the preset auxiliary medical knowledge graph;

[0085] S32. Determine a first confirmed disease rate corresponding to the first predicted disease and a second confirmed disease rate corresponding to the second predicted disease based on the patient type and a preset case sample;

[0086] S33, using the predicted disease corresponding to the larger one of the first confirmed disease rate and the second confirmed disease rate as the target predicted disease corresponding to the patient;

[0087] S34. Determine a registration suggestion corresponding to the medical assistance request based on the target predicted disease.

[0088] Explanatoryally, several corresponding preset auxiliary medical knowledge graphs are obtained. When at least two preset auxiliary medical knowledge graphs are obtained, different preset auxiliary medical knowledge graphs generally have different registration suggestions. Registration suggestions include but are not limited to predicted diseases and symptom sites. In this case, the first predicted disease and the second predicted disease corresponding to the preset auxiliary medical knowledge graph are obtained, wherein the "first" and "second" involved in the first predicted disease and the second predicted disease are only used to distinguish different predicted diseases, and are not used to limit the predicted diseases.

[0089] Then, based on the patient type and the preset case sample, the first confirmed morbidity corresponding to the first predicted disease is determined, wherein the first confirmed morbidity represents the probability of the first predicted disease occurring in the preset case sample in patients of the patient type. Similarly, the second confirmed morbidity corresponding to the second predicted disease is determined, and the predicted disease corresponding to the larger of the first confirmed morbidity and the second confirmed morbidity is used as the target predicted disease corresponding to the patient. That is, when the first confirmed morbidity is greater than the second confirmed morbidity, the first predicted disease corresponding to the first confirmed morbidity is used as the target predicted disease; when the first confirmed morbidity is less than the second confirmed morbidity, the second predicted disease corresponding to the second confirmed morbidity is used as the target predicted disease; when the first confirmed morbidity is equal to the second confirmed morbidity, the first predicted disease and the second predicted disease can be used as target predicted diseases at the same time, indicating that the patient is suffering from multiple diseases. Then, the recommended content for registration corresponding to the target predicted disease is determined, and the recommended content includes but is not limited to the patient's symptom summary description, the corresponding predicted symptom site, the predicted disease, the recommended registration department, the registration physician selection, and the registration time selection, and the recommended content is used as the registration suggestion corresponding to the medical assistance request.

[0090] In an embodiment of the present application, in the presence of different predicted diseases, a target predicted disease is determined based on the confirmed disease rate corresponding to the predicted disease, and then a corresponding registration suggestion is determined based on the target predicted disease. From a probability perspective, based on the principle of distinguishing between primary and secondary, it is possible to automatically determine online which department of the hospital the patient should go to for treatment as accurately as possible to provide auxiliary treatment suggestions, thereby improving the accuracy of the registration suggestions.

[0091] In one embodiment, after feeding back the registration suggestion to the patient, the method further includes:

[0092] When monitoring that the patient has completed the medical consultation, obtaining the patient's medical consultation record;

[0093] According to the medical records, obtain the actual registration results and disease diagnosis corresponding to the patient;

[0094] The preset auxiliary medical knowledge graph is updated according to the actual registration result, the disease diagnosis and the registration suggestion.

[0095] Explanatory, the patient's medical treatment progress is monitored online in real time, and after the patient's medical treatment is completed, that is, when the patient's medical treatment is completed, the patient's medical treatment record is obtained, which includes but is not limited to the patient's actual registration result, medical department, medical doctor, and disease diagnosis. Based on the medical treatment record, the patient's corresponding actual registration result and disease diagnosis are obtained, and then based on the actual registration result, disease diagnosis and registration suggestion, the preset medical treatment auxiliary medical knowledge graph is updated to realize the iterative update of the preset medical treatment auxiliary medical knowledge graph.

[0096] The embodiment of the present application monitors the patient's medical treatment progress and realizes automatic iterative update of the preset medical auxiliary knowledge graph based on the patient's medical treatment record, which can realize automatic learning of the preset intelligent robot, thereby improving the accuracy and reliability of the preset intelligent robot, thereby improving the accuracy of the online automatic determination of which hospital department the patient should go to for treatment to provide medical auxiliary suggestions, thereby further improving the self-service level and quality of digital medicine.

[0097] It should be noted that the intelligent robot-based medical assistance treatment methods described in the above embodiments can recombine the technical features contained in different embodiments as needed to obtain a combined implementation plan, but they are all within the scope of protection required by this application.

[0098] See also Figure 4 , Figure 4 This is a schematic block diagram of an intelligent robot-based medical consultation assistance processing device provided in an embodiment of the present application. Corresponding to the above-mentioned intelligent robot-based medical consultation assistance processing method, an embodiment of the present application also provides an intelligent robot-based medical consultation assistance processing device. Figure 4 As shown, the intelligent robot-based medical consultation auxiliary processing device includes a unit for executing the above-mentioned intelligent robot-based medical consultation auxiliary processing method. The intelligent robot-based medical consultation auxiliary processing device can be configured in a computer device. Specifically, please refer to Figure 4 The intelligent robot-based medical consultation auxiliary processing device 40 includes a first acquisition unit 41, a first judgment unit 42, a first determination unit 43 and a feedback unit 44.

[0099] The first acquisition unit 41 is configured to respond to a patient's request for assistance in making an appointment based on a preset intelligent robot and acquire the patient's symptom information data.

[0100] A first determining unit 42 is configured to determine whether the disease information data is multimodal data;

[0101] A first determining unit 43 is configured to, when the symptom information data is multimodal data, determine a registration suggestion corresponding to the medical assistance request based on a preset multimodal deep learning model and a preset medical knowledge graph for medical assistance, and according to the symptom information data;

[0102] The feedback unit 44 is used to feed back the registration suggestion to the patient.

[0103] In one embodiment, the first acquiring unit 41 includes:

[0104] a prompting subunit, configured to prompt the patient to provide characterization information data corresponding to the condition;

[0105] a first acquisition subunit, configured to acquire the characterization information data and collect physiological signal data corresponding to the patient;

[0106] The first determining subunit is configured to use the characterization information data and the physiological signal data as the disease information data corresponding to the patient.

[0107] In one embodiment, the first determining unit 42 includes:

[0108] An identification subunit, configured to identify a type of disease condition data corresponding to the disease condition information data;

[0109] A first judging subunit is configured to judge whether the disease condition data type includes at least two data types;

[0110] The first determination subunit is configured to determine that the disease information data is multimodal data when the disease data type includes at least two data types.

[0111] In one embodiment, the symptom information data includes identity information of the patient; the first determining unit 43 includes:

[0112] a second determining subunit, configured to determine a patient type corresponding to the patient based on the identity feature information;

[0113] a third determining subunit, configured to determine the first type of data and the second type of data included in the symptom information data;

[0114] a first processing subunit, configured to determine, based on the patient type, a preset first multimodal deep learning model corresponding to the first type of data, and process the first type of data based on the preset first multimodal deep learning model to obtain first symptom information;

[0115] a second processing subunit, configured to determine, based on the patient type, a preset second multimodal deep learning model corresponding to the second type of data, and process the second type of data based on the preset second multimodal deep learning model to obtain second symptom information;

[0116] a fourth determining subunit, configured to determine a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information;

[0117] The fifth determining subunit is used to determine the registration suggestion corresponding to the medical assistance request based on the preset medical assistance knowledge graph.

[0118] In one embodiment, the fourth determining subunit includes:

[0119] A second judgment subunit is used to judge whether there is a preset disease association relationship between the first disease information and the second disease information;

[0120] The sixth determination subunit is used to obtain a preset medical consultation assistance knowledge graph containing the first symptom information and the second symptom information when there is a preset symptom association relationship between the first symptom information and the second symptom information, and use it as the preset medical consultation assistance medical knowledge graph corresponding to the medical consultation assistance request.

[0121] In one embodiment, the fifth determining subunit includes:

[0122] The second acquisition subunit is used to acquire the first predicted disease and the second predicted disease corresponding to the preset auxiliary medical knowledge graph;

[0123] a seventh determining subunit, configured to determine a first confirmed disease rate corresponding to the first predicted disease and a second confirmed disease rate corresponding to the second predicted disease based on the patient type and a preset case sample;

[0124] an eighth determining subunit, configured to use the predicted disease corresponding to the larger one of the first confirmed disease rate and the second confirmed disease rate as the target predicted disease corresponding to the patient;

[0125] The ninth determining subunit is configured to determine a registration suggestion corresponding to the medical assistance request based on the target predicted disease.

[0126] In one embodiment, the intelligent robot-based auxiliary medical treatment processing device 40 further includes:

[0127] A second acquiring unit is configured to acquire the patient's medical records when monitoring that the patient's medical consultation is completed;

[0128] A third obtaining unit is used to obtain the actual registration result and disease diagnosis corresponding to the patient according to the medical record;

[0129] An updating unit is used to update the preset auxiliary medical knowledge graph based on the actual registration result, the disease diagnosis and the registration suggestion.

[0130] It should be noted that technical personnel in the relevant field can clearly understand that the specific implementation process of the above-mentioned intelligent robot-based medical auxiliary processing device and each unit can refer to the corresponding description in the aforementioned method embodiment. For the convenience and conciseness of the description, it will not be repeated here.

[0131] At the same time, the division and connection methods of each unit in the above-mentioned intelligent robot-based medical assistance processing device are only used for illustration. In other embodiments, the intelligent robot-based medical assistance processing device can be divided into different units as needed, and the units in the intelligent robot-based medical assistance processing device can also adopt different connection sequences and methods to complete all or part of the functions of the above-mentioned intelligent robot-based medical assistance processing device.

[0132] The above-mentioned intelligent robot-based auxiliary treatment device can be implemented in the form of a computer program. The computer program can be used in Figure 5 Runs on the computer device shown.

[0133] See also Figure 5 , Figure 5 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a computer device such as a desktop computer or a server, or may be a component or part in other devices.

[0134] See Figure 5 The computer device 500 includes a processor 502, a memory and a network interface 505 connected through a system bus 501, wherein the memory may include a non-volatile storage medium 503 and an internal memory 504, and the memory may also be a volatile storage medium.

[0135] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute the above-mentioned intelligent robot-based medical consultation assistance method.

[0136] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500.

[0137] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the above-mentioned intelligent robot-based medical assistance processing method.

[0138] The network interface 505 is used to communicate with other devices through the network. Figure 5The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. For example, in some embodiments, the computer device may only include a memory and a processor. In such an embodiment, the structure and function of the memory and processor are the same as those in the embodiment of the present application. Figure 5 The embodiments shown are consistent and will not be described again here.

[0139] Among them, the processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps: based on the preset intelligent robot, respond to the patient's medical assistance request corresponding to the registration, and obtain the patient's symptom information data; determine whether the symptom information data is multimodal data; when the symptom information data is multimodal data, based on the preset multimodal deep learning model and the preset medical assistance knowledge graph, and according to the symptom information data, determine the registration suggestion corresponding to the medical assistance request; and feed back the registration suggestion to the patient.

[0140] In one embodiment, the processor 502 implements the following steps when acquiring the patient's symptom information data:

[0141] Prompting the patient to provide characterization information data corresponding to the condition;

[0142] Acquiring the characterization information data and collecting physiological signal data corresponding to the patient;

[0143] The characterization information data and the physiological signal data are used as the disease information data corresponding to the patient.

[0144] In one embodiment, when determining whether the disease information data is multimodal data, the processor 502 specifically implements the following steps:

[0145] Identify the disease condition data type corresponding to the disease condition information data;

[0146] Determining whether the disease condition data type includes at least two data types;

[0147] When the disease condition data type includes at least two data types, it is determined that the disease condition information data is multimodal data.

[0148] In one embodiment, the symptom information data includes identity feature information of the patient; when the processor 502 determines the registration suggestion corresponding to the medical assistance request based on the preset multimodal deep learning model and the preset medical knowledge graph for medical assistance, and according to the symptom information data, the processor 502 specifically implements the following steps:

[0149] Determining the patient type corresponding to the patient according to the identity characteristic information;

[0150] determining first type data and second type data included in the symptom information data;

[0151] Determining, according to the patient type, a preset first multimodal deep learning model corresponding to the first type of data, and processing the first type of data based on the preset first multimodal deep learning model to obtain first symptom information;

[0152] Determining, according to the patient type, a preset second multimodal deep learning model corresponding to the second type of data, and processing the second type of data based on the preset second multimodal deep learning model to obtain second symptom information;

[0153] Determining a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information;

[0154] According to the preset medical knowledge graph for auxiliary medical consultation, a registration suggestion corresponding to the auxiliary medical consultation request is determined.

[0155] In one embodiment, when determining the preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information, the processor 502 specifically implements the following steps:

[0156] Determining whether there is a preset symptom association relationship between the first symptom information and the second symptom information;

[0157] In the case that there is a preset symptom association relationship between the first symptom information and the second symptom information, a preset auxiliary medical knowledge graph containing the first symptom information and the second symptom information is obtained and used as the preset auxiliary medical knowledge graph corresponding to the auxiliary medical request.

[0158] In one embodiment, when determining the registration suggestion corresponding to the medical assistance request based on the preset medical assistance knowledge graph, the processor 502 specifically implements the following steps:

[0159] Obtaining a first predicted disease and a second predicted disease corresponding to the preset auxiliary medical knowledge graph;

[0160] Determining a first confirmed disease rate corresponding to the first predicted disease and a second confirmed disease rate corresponding to the second predicted disease based on the patient type and a preset case sample;

[0161] The predicted disease corresponding to the larger of the first confirmed disease rate and the second confirmed disease rate is used as the target predicted disease corresponding to the patient;

[0162] According to the target predicted disease, a registration suggestion corresponding to the medical assistance request is determined.

[0163] In one embodiment, after feeding back the registration suggestion to the patient, the processor 502 further implements the following steps:

[0164] When monitoring that the patient has completed the medical consultation, obtaining the patient's medical consultation record;

[0165] According to the medical records, obtain the actual registration results and disease diagnosis corresponding to the patient;

[0166] The preset auxiliary medical knowledge graph is updated according to the actual registration result, the disease diagnosis and the registration suggestion.

[0167] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0168] Those skilled in the art will appreciate that all or part of the steps in the method of the above-described embodiment can be implemented by a computer program, which can be stored in a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the steps in the method of the above-described embodiment.

[0169] Therefore, the present application also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, and the computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the following steps:

[0170] A computer program product, when running on a computer, enables the computer to execute the steps of the intelligent robot-based medical assistance processing method described in the above embodiments.

[0171] The computer-readable storage medium may be an internal storage unit of the aforementioned device, such as a hard disk or memory of the device. The computer-readable storage medium may also be an external storage device of the device, such as a plug-in hard disk equipped on the device, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Furthermore, the computer-readable storage medium may include both an internal storage unit of the device and an external storage device.

[0172] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0173] The storage medium is a physical, non-transient storage medium, for example, a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which can store computer programs.

[0174] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0175] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and other division methods may be used in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not implemented.

[0176] The steps in the method of the embodiment of the present application can be adjusted in order, combined, and deleted according to actual needs. The units in the device of the embodiment of the present application can be combined, divided, and deleted according to actual needs. In addition, the functional units in the various embodiments of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into a single unit.

[0177] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling an electronic device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0178] The software tools or components not produced by our company that appear in the embodiments of this application are merely examples and do not represent actual use.

[0179] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A medical consultation assistance method based on an intelligent robot, characterized in that: include: Based on the preset intelligent robot, respond to the patient's request for medical assistance corresponding to the registration, and obtain the patient's disease information data; Determining whether the disease information data is multimodal data; In the case where the symptom information data is multimodal data, based on a preset multimodal deep learning model and a preset consultation assistance medical knowledge graph, and according to the symptom information data, determining a registration suggestion corresponding to the consultation assistance request; Feedback the registration suggestion to the patient; Wherein, determining whether the disease information data is multimodal data includes: Identify the disease condition data type corresponding to the disease condition information data; Determining whether the disease condition data type includes at least two data types; When the disease condition data type includes at least two data types, determining that the disease condition information data is multimodal data; The symptom information data includes the patient's identity feature information; based on a preset multimodal deep learning model and a preset consultation assistance medical knowledge graph, and according to the symptom information data, determining a registration suggestion corresponding to the consultation assistance request includes: Determining the patient type corresponding to the patient according to the identity characteristic information; determining first type data and second type data included in the symptom information data; Determining, according to the patient type, a preset first multimodal deep learning model corresponding to the first type of data, and processing the first type of data based on the preset first multimodal deep learning model to obtain first symptom information; Determining, according to the patient type, a preset second multimodal deep learning model corresponding to the second type of data, and processing the second type of data based on the preset second multimodal deep learning model to obtain second symptom information; Determining a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information; Determining a registration suggestion corresponding to the medical assistance request based on the preset medical assistance knowledge graph; Wherein, determining the registration suggestion corresponding to the medical assistance request according to the preset medical assistance knowledge graph includes: Obtaining a first predicted disease and a second predicted disease corresponding to the preset auxiliary medical knowledge graph; Determining a first confirmed disease rate corresponding to the first predicted disease and a second confirmed disease rate corresponding to the second predicted disease based on the patient type and a preset case sample; The predicted disease corresponding to the larger of the first confirmed disease rate and the second confirmed disease rate is used as the target predicted disease corresponding to the patient; According to the target predicted disease, a registration suggestion corresponding to the medical assistance request is determined.

2. The intelligent robot-based medical assistance method according to claim 1, characterized in that: Acquiring the patient's symptom information data, including: Prompting the patient to provide characterization information data corresponding to the condition; Acquiring the characterization information data and collecting physiological signal data corresponding to the patient; The characterization information data and the physiological signal data are used as the disease information data corresponding to the patient.

3. The intelligent robot-based medical assistance method according to claim 1, characterized in that: Determining a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information includes: Determining whether there is a preset symptom association relationship between the first symptom information and the second symptom information; In the case that there is a preset symptom association relationship between the first symptom information and the second symptom information, a preset auxiliary medical knowledge graph containing the first symptom information and the second symptom information is obtained and used as the preset auxiliary medical knowledge graph corresponding to the auxiliary medical request.

4. The intelligent robot-based medical assistance method according to claim 1, characterized in that: After feeding back the registration suggestion to the patient, the method further includes: When monitoring that the patient has completed the medical consultation, obtaining the patient's medical consultation record; According to the medical records, obtain the actual registration results and disease diagnosis corresponding to the patient; The preset auxiliary medical knowledge graph is updated according to the actual registration result, the disease diagnosis and the registration suggestion.

5. A medical consultation auxiliary processing device based on an intelligent robot, characterized in that: include: The first acquisition unit is configured to respond to a patient's request for assistance in making an appointment based on a preset intelligent robot and acquire the patient's symptom information data; a first determining unit, configured to determine whether the disease information data is multimodal data; A first determining unit is configured to, when the symptom information data is multimodal data, determine a registration suggestion corresponding to the medical assistance request based on a preset multimodal deep learning model and a preset medical knowledge graph for medical assistance, and according to the symptom information data; A feedback unit, configured to feed back the registration suggestion to the patient; The first judgment unit includes: An identification subunit, configured to identify a type of disease condition data corresponding to the disease condition information data; A first judging subunit is configured to judge whether the disease condition data type includes at least two data types; a first determining subunit, configured to determine that the disease information data is multimodal data when the disease data type includes at least two data types; The symptom information data includes the patient's identity information; the first determining unit includes: a second determining subunit, configured to determine a patient type corresponding to the patient based on the identity feature information; a third determining subunit, configured to determine the first type of data and the second type of data included in the symptom information data; a first processing subunit, configured to determine, based on the patient type, a preset first multimodal deep learning model corresponding to the first type of data, and process the first type of data based on the preset first multimodal deep learning model to obtain first symptom information; a second processing subunit, configured to determine, based on the patient type, a preset second multimodal deep learning model corresponding to the second type of data, and process the second type of data based on the preset second multimodal deep learning model to obtain second symptom information; a fourth determining subunit, configured to determine a preset auxiliary medical knowledge graph based on the first symptom information and the second symptom information; a fifth determining subunit, configured to determine a registration suggestion corresponding to the medical assistance request based on the preset medical assistance knowledge graph; The fifth determining subunit includes: The second acquisition subunit is used to obtain the first predicted disease and the second predicted disease corresponding to the preset auxiliary medical knowledge graph; a seventh determining subunit, configured to determine a first confirmed disease rate corresponding to the first predicted disease and a second confirmed disease rate corresponding to the second predicted disease based on the patient type and a preset case sample; an eighth determining subunit, configured to use the predicted disease corresponding to the larger one of the first confirmed disease rate and the second confirmed disease rate as the target predicted disease corresponding to the patient; The ninth determining subunit is configured to determine a registration suggestion corresponding to the medical assistance request based on the target predicted disease.

6. A computer device, characterized in that: The computer device includes a memory and a processor connected to the memory; the memory is used to store a computer program; and the processor is used to run the computer program to perform the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 can be implemented.

Images