Medical service method based on large medical model and related device

Abstract

The invention discloses a doctor service method based on a medical large model and a related device. The method comprises the following steps: acquiring service request data of a target doctor in a target service scene; wherein the target service scene is any one of the following medical service scenes: diagnosis decision suggestion, treatment scheme recommendation, medical record generation, medical record quality control and medical record information management; and analyzing the service request data based on the large medical model to obtain a medical service suggestion output by the large medical model to the target doctor in the target service scene. In this way, medical service suggestions can be automatically provided for doctors.

Description

Technical Field

[0001] The present application relates to the field of audio processing technology, and in particular to a doctor service method and related devices based on a medical big model. Background Art

[0002] At present, in the medical field, medical staff generally make diagnoses through interviews and examination results to obtain medical service recommendations, such as recommended diseases and causes. However, manual diagnosis is inefficient. In view of this, how to enable medical staff to obtain medical service recommendations in a timely manner has become an urgent problem to be solved. Summary of the invention

[0003] The main technical problem solved by this application is to provide a doctor service method and related devices based on a medical big model, which can automatically provide medical service suggestions to doctors.

[0004] In order to solve the above-mentioned technical problems, the first aspect of the present application provides a doctor service method based on a medical big model, including: obtaining service request data of a target doctor in a target service scenario; wherein the target service scenario is any one of the following medical service scenarios: diagnostic decision recommendations, treatment plan recommendations, medical record generation, medical record quality control, and medical record information management; analyzing the service request data based on the medical big model to obtain the medical service recommendations output by the medical big model for the target doctor in the target service scenario.

[0005] In order to solve the above-mentioned technical problems, the second aspect of the present application provides a doctor service device based on a medical big model, which includes an acquisition module and an analysis module; the acquisition module is used to obtain the service request data of the target doctor in the target service scenario; wherein the target service scenario is any one of the following medical service scenarios: diagnostic decision recommendations, treatment plan recommendations, medical record generation, medical record quality control, and medical record information management; the analysis module is used to analyze the service request data based on the medical big model, and obtain the medical service recommendations output by the medical big model for the target doctor in the target service scenario.

[0006] In order to solve the above-mentioned technical problems, the third aspect of the present application provides an electronic device, which includes a memory and a processor coupled to each other, wherein the memory stores program instructions, and the processor is used to execute the program instructions to implement the medical big model-based doctor service method of the above-mentioned first aspect.

[0007] In order to solve the above technical problems, the fourth aspect of the present application provides a computer-readable storage medium, which stores program instructions that can be executed by a processor, and the program instructions are used to implement the medical service method based on the medical big model of the above first aspect.

[0008] The above technical solution obtains the service request data of the target doctor in the target service scenario, and analyzes the service request data based on the medical big model to obtain the medical service suggestions output by the medical big model for the target doctor in the target service scenario. Therefore, the medical big model can automatically provide medical service suggestions for the target doctor and improve the efficiency and quality of medical and health services. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a flowchart of an embodiment of a doctor service method based on a medical big model provided by the present application;

[0010] Figure 2 It is a schematic diagram of the framework of an embodiment of a doctor service device based on a medical big model provided by the present application;

[0011] Figure 3 It is a schematic diagram of a framework of an embodiment of an electronic device provided by the present application;

[0012] Figure 4 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium provided by the present application. DETAILED DESCRIPTION

[0013] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0014] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0015] The terms "system" and "network" are often used interchangeably in this article. The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the fragment " / " in this article generally indicates that the associated objects before and after are in an "or" relationship. In addition, "many" in this article means two or more than two.

[0016] See also Figure 1 , Figure 1 It is a flowchart of an embodiment of a doctor service method based on a medical big model provided by the present application. It should be noted that if there are substantially the same results, the embodiments of the present application are not based on Figure 1 The process sequence shown is limited. Figure 1 As shown, specifically including:

[0017] Step S11: Obtain the service request data of the target doctor in the target service scenario.

[0018] The method of this embodiment is used to use the medical big model to analyze the service request data of the target doctor in the target service scenario, and obtain the medical service suggestions output by the medical big model for the target doctor, which can intelligently provide medical service suggestions for the target doctor and improve the efficiency and quality of medical and health services. The medical big model is a multimodal big model of sustainable learning based on deep learning technology, integrating medical and health knowledge and reasoning patterns, and understanding and executing medical and health-related tasks based on natural dialogue. It has the capabilities of complex language understanding, medical professional text generation, multimodal recognition and retrieval, evidence-based logical reasoning, multi-round interaction, and ethical safety assurance in the medical and health industry.

[0019] In this implementation, the service request data of the target doctor in the target service scenario is obtained; wherein the target service scenario is any one of the following medical service scenarios: diagnosis decision suggestion, treatment plan recommendation, medical record generation, medical record quality control, and medical record information management. In other words, the medical service based on the medical big model can perform at least one of the following tasks: diagnosis decision suggestion, treatment plan recommendation, medical record generation, medical record quality control, and medical record information management, and improve the efficiency and quality of at least one of the following tasks: diagnosis decision suggestion, treatment plan recommendation, medical record generation, medical record quality control, and medical record information management.

[0020] Diagnostic decision suggestions can be understood as follows: the medical big model can make diagnostic decision suggestions based on clinical manifestations and test results, give complete decision path suggestions and basis, or conduct personalized and disease-compliant diagnostic questions and answers. Treatment plan recommendation can be understood as follows: the medical big model can comprehensively consider treatment plans, assess risks, and give treatment plans based on factors such as patient medical history, physical condition, and treatment progress. Medical record generation can be understood as follows: the medical big model can generate outpatient (emergency) medical records, inpatient medical records, ward nursing records, and examination result documents based on existing medical record content, input content, and examination results. Medical record quality control can be understood as follows: the medical big model can perform quality control on medical record content in medical services, provide the quality of electronic medical records, and assist medical staff in effectively improving the quality and efficiency of medical record writing activities. Medical record information management can be understood as follows: the medical big model can manage medical record information in medical services.

[0021] In one embodiment, the service request data includes at least one modality of data including text, images, voice, and video.

[0022] Step S12: Analyze the service request data based on the medical big model to obtain the medical service recommendations output by the medical big model for the target doctor in the target service scenario.

[0023] In this embodiment, the service request data is analyzed based on the medical big model to obtain the medical service recommendations output by the medical big model for the target doctor in the target service scenario. That is to say, the service request data is analyzed using the medical big model, and the medical big model will analyze the service request data based on its knowledge understanding ability and the medical knowledge learned in the training stage to obtain medical service recommendations, so as to provide medical service recommendations to the target doctor, that is, it can intelligently, accurately and efficiently provide medical service recommendations to the target doctor, and improve the efficiency and quality of medical and health services.

[0024] In one embodiment, when the target service scenario is a diagnosis decision suggestion, in response to obtaining at least one of the physical examination information and the consultation information of the target doctor through the analysis of the service request data based on the medical big model, a medical service suggestion including a recommended diagnosis and at least one of its cause and the patient's constitution is output. That is, the medical big model can extract the physical examination information and the consultation information of the target doctor from the service request data, and can perform disease diagnosis based on the extracted physical examination information and the consultation information to obtain a disease diagnosis result, and can also identify the cause and the patient's constitution, etc.

[0025] There is no limitation on physical examination information and medical interview information; for example, physical examination information may be patient symptoms, patient signs, patient medical history, or patient western medicine examination results, and medical interview information may be patient symptoms, visual examination, pulse diagnosis, and other Chinese medicine examination results. For example, the medical big model quickly diagnoses the disease and identifies the cause of the disease based on physical examination information such as patient symptoms, patient signs, patient medical history, and patient western medicine examination results; the medical big model identifies the patient's constitution based on the patient's symptoms, visual examination, pulse diagnosis, and other Chinese medicine examination results.

[0026] In one embodiment, when the target service scenario is a diagnosis decision suggestion, in response to obtaining at least one of clinical manifestation information, examination and testing information, and medical imaging information through analysis of the service request data based on the medical big model, a medical service suggestion including at least one of a recommended diagnosis and a diagnosis basis is output. That is, the medical big model can extract clinical manifestation information, examination and testing information, medical imaging information, etc. from the service request data, and can perform disease diagnosis based on the extracted clinical manifestation information, examination and testing information, medical imaging information, etc. to obtain a disease diagnosis result, and can also generate a diagnosis basis, etc., to assist the target doctor in making a correct diagnosis.

[0027] In a specific embodiment, the medical big model analyzes the service request data to obtain at least one of clinical manifestation information, examination and test information, and medical imaging information, and combines it with evidence-based medicine evidence such as clinical guidelines to output medical service recommendations including at least one of a recommended diagnosis and a diagnostic basis.

[0028] In one embodiment, when the target service scenario is a diagnosis decision recommendation, in response to obtaining clinical manifestation information by analyzing the service request data based on the medical big model, a medical service recommendation including a recommended disease and its diagnosis probability is output. That is, the medical big model can extract clinical manifestation information from the service request data, and can perform disease diagnosis based on the extracted clinical manifestation information, and can also generate the probability of diagnosing the disease, so as to assist the target doctor in making a correct diagnosis.

[0029] In one embodiment, when the target service scenario is a diagnosis decision suggestion, in response to obtaining at least one of a disease risk factor and a lifestyle information by analyzing the service request data based on the medical big model, a medical service suggestion including a disease risk is output. That is, the medical big model can extract disease risk factors, lifestyle information, etc. from the service request data, and can assist the target doctor in making an early diagnosis of the disease based on the extracted disease risk factors, lifestyle information, etc.

[0030] In one embodiment, when the target service scenario is a diagnosis decision suggestion, in response to obtaining at least one of individual differences, physical condition information, and disease influencing factors through analysis of the service request data based on the medical big model, a medical service suggestion including at least one of personalized examination contraindications and recommended diagnosis is output. That is, the medical big model can extract individual differences, physical condition information, disease influencing factors, etc. from the service request data, and can perform disease diagnosis based on the extracted individual differences, physical condition information, disease influencing factors, etc. to obtain personalized examination contraindications, disease diagnosis results, etc., so as to assist the target doctor in making a correct diagnosis.

[0031] In one embodiment, when the target service scenario is a diagnosis decision suggestion, in response to obtaining medication history and disease information by analyzing the service request data based on the medical big model, a medical service suggestion including drug interactions is output. That is, the medical big model can extract medication history and disease information from the service request data, and can perform a diagnosis based on the extracted medication history and disease information to obtain drug interactions, so as to assist the target doctor in evaluating drug interactions.

[0032] In one embodiment, when the target service scenario is a diagnosis decision suggestion, in response to obtaining treatment feedback information and condition change information by analyzing the service request data based on the medical big model, a medical service suggestion including a diagnosis update result is output. That is, the medical big model can extract treatment feedback information and condition change information from the service request data, and can perform a diagnosis based on the extracted treatment feedback information and condition change information to obtain a diagnosis update result, so as to assist the target doctor to timely update the patient's diagnosis result.

[0033] It should be noted that the above-mentioned recommended diagnosis includes one candidate disease, or multiple concurrent candidate diseases, that is, the above-mentioned disease diagnosis results include one candidate disease, or multiple concurrent candidate diseases.

[0034] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining patient medical history information by analyzing the service request data based on the medical big model, a medical service suggestion including a patient portrait is output. That is, the medical big model can extract the patient medical history information from the service request data, and can construct a patient screen based on the extracted patient medical history information.

[0035] There is no limitation on the patient's medical record information; for example, the patient's medical record information may be admission records, outpatient medical records, surgical records, medication records, discharge summaries, etc.

[0036] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to a rehabilitation formulation request obtained by analyzing the service request data based on the medical big model, and the medical big model also inputs a patient portrait, outputs a medical service suggestion including at least one of a rehabilitation plan and a node prompt; wherein the node prompt is used to prompt the execution of relevant rehabilitation actions at each node in the rehabilitation plan. In other words, the medical big model can extract the rehabilitation formulation request from the service request data, and can formulate a rehabilitation plan, prompt plan nodes, etc. according to the extracted rehabilitation formulation request.

[0037] Specifically, the medical big model can generate a patient-specific rehabilitation plan based on the patient's picture, and according to the plan nodes, remind and guide the patient to perform relevant rehabilitation actions (such as follow-up visits, exercise, diet, etc.).

[0038] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining information about the condition changes during the rehabilitation process through analysis of the service request data based on the medical big model, a medical service suggestion including an updated patient portrait is output. In other words, the medical big model can extract information about the condition changes during the rehabilitation process from the service request data, and can update the patient portrait based on the extracted information about the condition changes during the rehabilitation process.

[0039] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining patient information by analyzing the service request data based on the medical big model, a medical service suggestion including at least one of emergency treatment, auxiliary guidance, treatment plan formulation, treatment plan evidence-based, treatment plan explanation, and drug selection is output; wherein the treatment plan explanation is provided based on at least one of three-dimensional rendering and generative speech. In other words, the medical big model can extract patient information from the service request data, and can generate suggestions such as emergency treatment, auxiliary guidance, treatment plan formulation, treatment plan evidence-based, treatment plan explanation, and drug selection based on the extracted patient information.

[0040] Among them, there is no limitation on patient information; for example, patient information may include patient basic information, medical condition, medication history, drug interactions, etc.

[0041] In one embodiment, the medical big model can combine the treatment plan and medical guidelines to generate treatment evidence. In one embodiment, the medical big model can combine the relevant medical guideline recommendations or authoritative expert recommendations to generate opinions or deficiencies corresponding to the treatment plan.

[0042] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining at least one of the patient's condition and the treatment progress through analysis of the service request data based on the medical big model, a medical service suggestion including an updated treatment plan is output. That is, the medical big model can extract information such as the patient's condition and the treatment progress from the service request data, and can generate a new treatment plan based on the extracted information such as the patient's condition and the treatment progress.

[0043] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining at least one of the patient's physique, physical condition, individual differences, and treatment progress through analysis of the service request data based on the medical big model, a medical service suggestion including at least one of a surgical plan and surgical contraindications is output. That is, the medical big model can extract information such as the patient's physique, physical condition, individual differences, and treatment progress from the service request data, and can generate a surgical plan, surgical contraindications, etc. based on the extracted information such as the patient's physique, physical condition, individual differences, and treatment progress. That is, the medical big model can comprehensively specify surgical planning, prompt surgical contraindications, etc.

[0044] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining at least one of the patient's clinical characteristics and test data from the service request data based on the medical big model, and the medical big model also inputs historical treatment plans, a medical service suggestion including a treatment plan risk assessment is output. That is, the medical big model can extract information such as the patient's clinical characteristics and test data from the service request data, and can generate a treatment plan risk assessment based on the extracted patient clinical characteristics, test data, and other information combined with the input historical treatment plans. That is, the medical big model can generate treatment plan risk management suggestions, compare the patient's clinical characteristics, test data, and other data based on the historical treatment plans, and provide a treatment plan risk assessment.

[0045] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to a digital treatment request obtained by analyzing the service request data based on the medical big model, a medical service recommendation including an evidence-based treatment intervention is output; wherein the evidence-based treatment intervention is driven by the medical big model. That is, the medical big model can extract the digital treatment request from the service request data, and can generate an evidence-based treatment intervention plan based on the extracted digital treatment request. It should be noted that the evidence-based treatment intervention driven by the medical big model can prevent, manage, and treat medical disorders or diseases, and can be used independently or in conjunction with drugs, equipment, and other therapies.

[0046] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to the historical medical records obtained by analyzing the service request data based on the medical big model, and the medical big model also inputs the medical guidelines, a medical service suggestion including a disease recovery path framework is output. In other words, the medical big model can extract the historical medical records from the service request data, and can generate a recovery path framework for the corresponding disease based on the extracted historical medical records and the input medical guidelines.

[0047] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining an updated patient portrait based on the analysis of the service request data by the medical big model, a medical service suggestion including an updated rehabilitation plan is output. That is, the medical big model can extract the updated patient portrait from the service request data, and can synchronously update the patient's disease-specific rehabilitation plan based on the extracted updated patient portrait, that is, the medical big model can dynamically update the patient's disease-specific rehabilitation plan based on the update of the patient portrait.

[0048] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining an updated rehabilitation plan based on the medical big model analyzing the service request data, a medical service suggestion including updated node prompts is output. That is, the medical big model can extract the updated rehabilitation plan from the service request data, and can generate updated node prompts based on the extracted updated rehabilitation plan to remind and guide the patient to perform rehabilitation actions.

[0049] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to the patient's condition and the customized education request obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing education knowledge is output; wherein the education knowledge is consistent with the patient's condition. In other words, the medical big model can extract the patient's condition and the customized education request from the service request data, and can generate education knowledge that is consistent with the patient's condition based on the extracted patient's condition and the customized education request, so that the education knowledge provided to the patient is more targeted.

[0050] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining at least one of clinical characteristics, medical history information, diagnosis information, and treatment information through analysis of the service request data based on the medical big model, a medical service suggestion including at least one of the course of the disease and the predicted treatment result is output. That is, the medical big model can extract information such as clinical characteristics, medical history information, diagnosis information, and treatment information from the service request data, and can generate the course of the disease and the predicted treatment result based on the extracted clinical characteristics, medical history information, diagnosis information, and treatment information. That is, the medical big model can evaluate the course of the disease and predict the treatment result based on the patient's clinical characteristics, medical history information, diagnosis information, and treatment information.

[0051] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to obtaining at least one of the patient's condition information and medication history information by analyzing the service request data based on the medical big model, a medical service suggestion including the predicted risk of adverse events is output. That is, the medical big model can extract information such as the patient's condition information and medication history information from the service request data, and can generate the predicted risk of adverse events based on the extracted information such as the patient's condition information and medication history information. That is, the medical big model can predict the risk of adverse events and provide risk management suggestions based on the patient's condition information, medication history information and other information for early warning and risk management of adverse events.

[0052] In one embodiment, when the target service scenario is a treatment plan recommendation, in response to a medical search request obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing medical knowledge related to the medical search request is output. That is, the medical big model can extract the medical search request from the service request data, and can generate medical knowledge related to the medical search request based on the extracted medical search request. That is, the medical big model can retrieve and learn medical knowledge, and retrieve medical literature by calling relevant plug-ins to obtain the latest medical knowledge.

[0053] In one embodiment, when the target service scenario is to recommend a treatment plan, the medical big model can also output guidance information such as patient medical treatment and patient medication in the form of text, pictures, videos, etc. based on information such as the patient's condition, treatment status, and personal characteristics.

[0054] In one embodiment, when the target service scenario is medical record generation, in response to obtaining at least one of existing medical record content, doctor input content, and examination and test results through analysis of the service request data based on the medical big model, a medical service suggestion containing medical record content data under the first subdivided scenario of medical record generation is output; wherein the first subdivided scenario includes any one of outpatient / emergency medical record generation, inpatient medical record generation, ward nursing record generation, and examination result document generation. That is, the medical big model can extract information such as existing medical record content, doctor input content, and examination and test results from the service request data, and can generate outpatient / emergency medical records, inpatient medical records, ward nursing records, examination result documents, etc. based on the extracted existing medical record content, doctor input content, and examination and test results. That is, the medical big model can automatically generate medical record content data under the first subdivided scenario.

[0055] In a specific implementation, when the first subdivided scenario is outpatient / emergency medical record generation, the medical record content data includes at least one of the following: medical record homepage, medical record record, emergency department visit time, emergency observation record, rescue record, prescription record, external prescription, test application record, test report record, inspection application record, and inspection execution record. For example, the medical big model can automatically generate an outpatient medical record containing a medical record homepage and medical record record; the medical big model can automatically generate an emergency medical record containing an emergency department visit time, emergency observation record, and rescue record; the medical big model can automatically generate prescription records and external prescriptions; the medical big model can automatically generate a test application record and a test report record; the medical big model can automatically generate an inspection application record and an inspection execution record.

[0056] In a specific implementation, when the first subdivided scenario is the generation of inpatient medical records, the medical record content data includes at least one of the following: automatic coding of at least one of medical records, medical history, disease diagnosis, and surgery, doctor's order record, clinical pathway record, test application record, examination application record, examination report record, and examination critical value record. For example, the medical big model can automatically encode patient medical records and medical records; the medical big model encodes various inpatient medical records such as disease diagnosis and surgery according to ICD-10; the medical big model can automatically generate doctor's order records; the medical big model can generate clinical pathway records, and the clinical pathway can be the enrollment status, the cause of variation, etc.; the medical big model can generate test application records; the medical big model can generate examination application records, examination report records, and examination critical value records.

[0057] In a specific implementation, when the first subdivided scenario is the generation of ward nursing records, the medical record content data includes at least one of the following: ward patient information, nursing assessment records, nursing-related clinical pathway records, doctor's order execution records, descriptive nursing items, nursing electronic signature records, and adverse event records. For example, the medical big model can generate nursing records, such as descriptive nursing items with content of no less than 10 words; the medical big model can generate ward patient information, nursing assessment records, nursing-related clinical pathway records; the medical big model can generate doctor's order execution records, nursing electronic signature records, and adverse event records.

[0058] In a specific embodiment, when the first subdivision scenario is the generation of an examination result document, the medical record content data includes at least one of the following: specimen records, specimen transmission records, test report records, test result records, test critical value records, and quality control records, and the medical record content belongs to an internal report of the institution, a report on the return of outsourced specimens, or a report requested by an external institution.

[0059] Among them, the report returned by the sent specimens can include patient identification, test report items, hospital records, etc. In addition, the report requested by the external institution can include test result records, external hospital application records, etc.

[0060] In one embodiment, when the target service scenario is medical record generation, in response to analyzing the service request data based on the medical big model and obtaining at least one of the diagnosis and treatment conditions and medical advice information inside and outside the medical institution in the past, a medical service suggestion including at least medical advice verification is output. In other words, the medical big model can verify medical advice and provide prompts based on the diagnosis and treatment conditions and medical advice inside and outside the medical institution in the past.

[0061] In one embodiment, when the target service scenario is medical record generation, the patient's medical and health information is aggregated and shared; wherein the medical and health information includes at least one of the following: medical information inside and outside the institution, health records, physical sign detection, and follow-up information. In other words, the medical big model can aggregate and share the patient's medical and health information and display it.

[0062] Among them, medical and health information can also include health records collected by patients themselves, such as health records, wearable device data, etc.

[0063] In one embodiment, when the target service scenario is medical record quality control, in response to obtaining a medical record quality control request by analyzing the service request data based on the medical big model, a medical service suggestion including the medical record quality control result in the second subdivided scenario of medical record quality control is output; wherein the second subdivided scenario includes any one of outpatient / emergency medical record quality control, inpatient medical record quality control, ward nursing record quality control, and examination result document quality control. In other words, the medical big model can extract the medical record quality control request from the service request data, and can generate the medical record quality control result in the second subdivided scenario including the medical record quality control according to the extracted medical record quality control request. That is, the medical big model can automatically generate the medical record quality control result in the second subdivided scenario.

[0064] In the case where the second sub-scenario is the outpatient / emergency medical record quality control, in response to the query request for historical test results obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing a summary of test reports made by the same patient in different medical institutions is output. In other words, the medical big model can extract the query request for historical test results from the service request data, and can generate a summary of test reports made by the same patient in different medical institutions based on the extracted query request for historical test results. In other words, the medical big model can query historical test results, and can also query or automatically summarize the test results and reports of the same patient in other medical institutions.

[0065] In the case where the second sub-scenario is the outpatient / emergency medical record quality control, in response to obtaining at least one of the diagnosis and treatment conditions and medication conditions in the past inside and outside the medical institution through the analysis of the service request data based on the medical big model, a medical service suggestion including at least medical order verification is output. In other words, the medical big model can extract information such as the diagnosis and treatment conditions and medication conditions in the past inside and outside the medical institution from the service request data, and can verify the medical order based on the extracted information such as the diagnosis and treatment conditions and medication conditions in the past inside and outside the medical institution. For example, abnormality prompts can also be verified based on information such as the diagnosis and treatment conditions and medication conditions in the past inside and outside the medical institution.

[0066] In the case where the second sub-scenario is the outpatient / emergency medical record quality control, in response to the complete quality control request for the outpatient / emergency medical record obtained by analyzing the service request data based on the medical big model, a medical service suggestion including the complete quality control result is output. In other words, the medical big model can extract the complete quality control request for the outpatient / emergency medical record from the service request data, and can generate the complete quality control result based on the extracted complete quality control request for the outpatient / emergency medical record.

[0067] Among them, outpatient / emergency medical records can be outpatient test applications, outpatient test reports, outpatient examination applications, outpatient examination reports, outpatient medical record records, etc.

[0068] In the case where the second sub-scenario is the quality control of inpatient medical records, in response to obtaining at least one of the terminology verification request and the logic verification request through the analysis of the service request data based on the medical big model, a medical service suggestion including the verification result is output. The medical big model can extract the terminology verification request, the logic verification request and other requests from the service request data, and can generate the verification result according to the extracted terminology verification request, the logic verification request and other requests. That is, the medical big model can verify the logical conformity of the terminology and description of the relevant medical record content.

[0069] In the case of the second sub-scenario of inpatient medical record quality control, in response to the prescription review request obtained by analyzing the service request data based on the medical big model, and the medical big model also inputs medical-related laws, it outputs a medical service recommendation including a drug dispensing decision. In other words, the medical big model can extract the prescription review request from the service request data, and can generate a drug dispensing decision based on the extracted prescription review request and medical-related laws. That is, the medical big model can conduct prescription review through pharmaceutical expertise, and according to relevant laws, regulations, rules and technical specifications, the prescriptions issued by the physician to the patient during the diagnosis and treatment activities are reviewed for legality, standardization and suitability, and the pharmaceutical technical service that decides whether to agree to dispense the drug is made.

[0070] In the case where the second sub-scenario is the quality control of inpatient medical records, in response to the request for complete quality control of inpatient medical records obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing the integrity quality control result is output. In other words, the medical big model can extract the request for complete quality control of inpatient medical records from the service request data, and can generate integrity quality control results based on the extracted request for complete quality control of inpatient medical records. For example, the medical big model can process documents such as ward medical orders, test applications, test reports, inspection applications, inspection reports, medical records, etc., and generate corresponding integrity quality control results.

[0071] In the case where the second sub-scenario is the quality control of inpatient medical records, in response to the consistency quality control request for the inpatient medical record content obtained by analyzing the service request data based on the medical big model, a medical service recommendation containing the consistency quality control result is output. In other words, the medical big model can extract the consistency quality control request for the inpatient medical record content from the service request data, and can generate a consistency quality control result based on the extracted consistency quality control request for the inpatient medical record content. For example, the medical big model can process documents such as ward medical orders, test applications, test reports, inspection applications, inspection reports, medical records, etc., and generate corresponding consistency quality control results.

[0072] In the case of the second sub-scenario of inpatient medical record quality control, in response to at least one of the test results and medication status obtained by analyzing the service request data based on the medical big model, a medical service suggestion including an infection warning prompt is output. That is, the medical big model can extract information such as test results and medication status from the service request data, and can provide infection warning prompts based on the extracted test results, medication status, etc. That is, the medical big model can automatically warn and give prompts for infectious diseases, hospital infection outbreaks, etc. based on test results, medication, etc., and can also supplement information on confirmed infectious diseases, hospital infection outbreaks, etc. and report them to the medical administration department.

[0073] In the case where the second subdivided scenario is the quality control of the ward nursing records, in response to the patient information request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including at least one record of admission, discharge, and transfer is output. In other words, the medical big model can extract the patient information request from the service request data, and can generate at least one record including admission, discharge, and transfer according to the extracted patient information request. In other words, the medical big model can manage patient information such as admission, discharge, and transfer records.

[0074] In the case where the second subdivided scenario is the quality control of the ward nursing records, in response to the image recognition request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including the image recognition result is output; wherein the image recognition result includes the result of identifying at least one of the patient, the medicine, and the test specimen through the image. In other words, the medical big model can extract the image recognition request from the service request data, and can generate the image recognition result according to the extracted image recognition request. In other words, the medical big model can automatically identify and check the patient, medicine, test specimen, etc. through image recognition technology.

[0075] In the case where the second sub-scenario is the quality control of the nursing records in the ward, in response to the information import request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including the information import result is output; wherein the information import result represents that at least one of vital signs and nursing treatment is automatically imported into the corresponding record sheet through a mobile device. In other words, the medical big model can extract the information import request from the service request data, and can generate the information import result according to the extracted information import request. That is, the medical big model can automatically import information such as vital signs and nursing treatment into the corresponding record sheet through a mobile device (mobile nursing).

[0076] In the case of the second sub-scenario of ward nursing record quality control, in response to the patient's activity records in other departments of the hospital obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing nursing measures is output. In other words, the medical big model can extract the patient's activity records in other departments of the hospital from the service request data, and can generate nursing measures based on the extracted patient's activity records in other departments of the hospital. In other words, the medical big model can automatically generate nursing measures suggestions based on the patient's diagnosis and treatment activity records in other departments of the hospital.

[0077] In the case where the second subdivided scenario is the quality control of the ward nursing records, in response to the patient condition obtained by analyzing the service request data based on the medical big model, a medical service suggestion including a nursing quality assessment is output. That is, the medical big model can extract the patient condition from the service request data, and can generate a nursing quality assessment based on the extracted patient condition. That is, the medical big model can automatically generate a nursing quality assessment based on the patient condition.

[0078] In the case where the second sub-scenario is the quality control of ward nursing records, in response to the consistency quality control request for ward nursing records obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing the consistency quality control result is output. In other words, the medical big model can extract the consistency quality control request for ward nursing records from the service request data, and can generate a consistency quality control result based on the extracted consistency quality control request for ward nursing records. In other words, the medical big model can process documents such as patient care and evaluation records, medical order execution records, and nursing records, and generate corresponding consistency quality control results.

[0079] In the case where the second sub-scenario is the quality control of ward nursing records, in response to the integrity quality control request for ward nursing records obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing the integrity quality control result is output. In other words, the medical big model can extract the integrity quality control request for ward nursing records from the service request data, and can generate integrity quality control results based on the extracted integrity quality control request for ward nursing records. In other words, the medical big model can process documents such as patient care and evaluation records, medical order execution records, and nursing records, and generate corresponding integrity quality control results.

[0080] In the case of the second sub-scenario of quality control of the inspection result documents, in response to the automatic corresponding request obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing the corresponding result is output; wherein the corresponding result includes the corresponding relationship between the test report and the clinical test application. In other words, the medical big model can extract the automatic corresponding request from the service request data, and can generate the corresponding relationship between the test report and the clinical test application based on the extracted automatic corresponding request. That is, the medical big model can automatically correspond the test report to the clinical test application.

[0081] In the case where the second subdivided scenario is the quality control of the examination result document, in response to the automatic judgment prompt obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing automatic prompt information related to the examination measurement value is output; wherein the automatic prompt information includes at least one of a reasonable range of the examination measurement value and a reasonable vocabulary range of the annotation description. In other words, the medical big model can extract the automatic judgment prompt from the service request data, and can generate the automatic prompt information related to the examination measurement value based on the extracted automatic judgment prompt.

[0082] In the case of the second sub-scenario of document quality control of the examination results, in response to the content verification request obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing report verification information is output. In other words, the medical big model can extract the content verification request from the service request data, and can generate report verification information based on the extracted content verification request.

[0083] Among them, the report verification information may include the test department report data, historical test results, result reference range, abnormal result mark, and project description.

[0084] In the case where the second sub-scenario is the quality control of the examination result documents, in response to the consistency quality control request for the examination result documents obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing the consistency quality control result is output. In other words, the medical big model can extract the consistency quality control request for the examination result documents from the service request data, and can generate the consistency quality control result based on the extracted consistency quality control request for the examination result documents. In other words, the medical big model can process documents such as specimen processing records and test result records to generate corresponding consistency quality control results.

[0085] In the case where the second sub-scenario is the quality control of the examination result documents, in response to the request for complete quality control of the examination result documents obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing the integrity quality control result is output. In other words, the medical big model can extract the request for complete quality control of the examination result documents from the service request data, and can generate the integrity quality control result based on the extracted request for complete quality control of the examination result documents. In other words, the medical big model can process documents such as specimen processing records and test result records to generate corresponding integrity quality control results.

[0086] In one embodiment, when the target service scenario is medical record information management, in response to the quality verification request for the first page of the medical record obtained by analyzing the service request data based on the medical big model, a medical service suggestion including at least one of the automatic verification result and the verification abnormality prompt is output. That is to say, the medical big model can extract the quality verification request for the first page of the medical record from the service request data, and can verify the various contents of the medical record based on the quality management specification according to the extracted quality verification request for the first page of the medical record to generate automatic verification results, verification abnormality prompts, etc. The provision of automatic verification results can enable the target doctor to know whether the various contents of the medical record are standardized and accurate in a timely manner; the verification abnormality prompt can enable the target doctor to know at a glance the various non-standard and inaccurate contents in the medical record. Thereby, the target doctor can adjust and modify the non-standard and inaccurate contents in a timely manner.

[0087] In one embodiment, when the target service scenario is medical record information management, in response to a defect verification request obtained by analyzing the service request data based on the medical big model, a medical service suggestion containing medical record content defects is output. That is, the medical big model can extract the defect verification request from the service request data, and can perform defect verification on various contents of the medical record according to the extracted defect verification request to generate medical record content defects. The provision of medical record content defects can enable the target doctor to promptly know the various contents with defects in the medical record, and specifically what kind of defects caused the abnormality, so that the target doctor can adjust and modify the content defects in a timely manner.

[0088] In one embodiment, when the target service scenario is medical record information management, in response to an automatic judgment request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including a medical record quality control record is output. That is, the medical big model can extract the automatic judgment request from the service request data, and can verify the content that must be written in the medical record according to the extracted automatic judgment request to generate a medical record quality control record. The provision of the medical record quality control record can enable the target doctor to promptly know that the content that must be written in the medical record has been verified and whether there is any missing content that must be written, so that the target doctor can promptly adjust and modify the medical record content with abnormal verification.

[0089] For example, assuming that the time needs to be written in Item A of the medical record, but the actual Item A of the medical record does not have the time written in it, it is determined that the content of Item A in the medical record is abnormal, and the abnormality is caused by the defect of missing time.

[0090] In one embodiment, when the target service scenario is medical record information management, in response to obtaining a writing verification request based on the medical big model analysis of the service request data, a medical service suggestion including a writing verification result is output. That is, the medical big model can extract the writing verification request from the service request data, and can generate a writing verification result based on the extracted writing verification request, realize intelligent inspection and prompting of the writing content, and can process the in-hospital consultation record, compare it with the consultation application, and incorporate the consultation record into the electronic medical record system.

[0091] In one embodiment, when the target service scenario is medical record information management, in response to the quality verification request for the complete medical record obtained by analyzing the service request data based on the medical big model, a medical service suggestion including the quality verification results of both the running medical record and the terminal medical record is output. That is, the medical big model can extract the quality verification request for the complete medical record from the service request data, and can generate the quality verification results of the running medical record and the terminal medical record according to the extracted quality verification request for the complete medical record, thereby realizing the automatic verification of the running medical record and the terminal medical record.

[0092] In one embodiment, when the target service scenario is medical record information management, in response to obtaining a quality control verification template by analyzing the service request data based on the medical big model, a medical service suggestion including a medical record quality control report is output; wherein the medical record quality control report conforms to the quality control verification template. In other words, the medical big model can extract the quality control verification template from the service request data, and can generate a medical record quality control report based on the extracted quality control verification template. The provision of the medical record quality control report can enable the target doctor to intuitively understand the quality control status of the medical record information.

[0093] In one embodiment, when the target service scenario is medical record information management, in response to a data comparison request for medical records and quality control records obtained by analyzing the service request data based on the medical big model, a medical service suggestion including at least a customized structured medical record is output. That is, the medical big model can extract the data comparison request for medical records and quality control records from the service request data, and can generate at least a customized structured medical record according to the extracted data comparison request for medical records and quality control records.

[0094] In a specific implementation, a customized structured medical record including examination and test results, a customized structured medical record corresponding to the target doctor's search content, etc. may also be output, which is not limited here.

[0095] In one embodiment, when the target service scenario is medical record information management, in response to abnormal data operation behavior obtained by analyzing the service request data based on the medical big model, a medical service suggestion including an automatic alarm prompt is output. In other words, the medical big model can extract abnormal data operation behavior from the service request data, and can monitor and alarm according to the extracted abnormal data operation behavior. The monitoring and alarm of abnormal data operation behavior can avoid the leakage of relevant information such as medical records and ensure the security of data.

[0096] There is no limitation on abnormal data operation behavior; for example, abnormal data operation behavior is data copying, etc.

[0097] In one embodiment, when the target service scenario is medical record information management, in response to the quality control analysis request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including at least one of a quality control data summary and a multi-dimensional analysis report is output. In other words, the medical big model can extract the quality control analysis request from the service request data, and can summarize and perform multi-dimensional analysis on the quality control data of the medical records according to the extracted quality control analysis request to generate summary data and a multi-dimensional analysis report, and intelligently manage the quality control data of the medical records. The provision of summary data can enable the target doctor to intuitively understand the overall situation of the quality control data of the medical records. The provision of a multi-dimensional analysis report can facilitate the target doctor to understand the situation of the quality control data of the medical records from multiple angles.

[0098] It should be noted that the relevant information obtained by the medical big model based on the analysis of the service request information may be information contained in the service request information itself. After analyzing the service request information, the medical big model determines that the service request information itself contains relevant information, and directly extracts the relevant information from the service request information. It may also be information not included in the service request information. After analyzing the service request information, the medical big model determines that the service request information itself does not directly contain relevant information, and obtains the relevant information based on reasoning about the service request information.

[0099] In one embodiment, the medical big model provided by the present application can be deployed in the cloud and can be quickly integrated with the business system or hardware device through independent API call model. In order to ensure the data privacy of the medical big model and the adaptability of the medical big model in the intranet / no network environment, the deployment on the private server can be realized through two integration methods: API (application programming interface) and SDK (software development kit); further, the model can be packaged into an SDK adapted to smart hardware, adapted to the operating system and deployed on the local device side, so that the medical big model can adapt to mainstream operating systems such as iOS, Android, Linux, and Windows.

[0100] In one embodiment, the medical big model can use API, SDK, SAAS (software as a service), PAAS (platform as a service), MAAS (model as a service), etc. in the application process. And / or, the medical big model can also configure the permissions of the operator, for example, it can review the data permission management strategy and verify the configuration personnel's operation permission classification, specifically, it can add permissions, delete permissions, edit permissions, view permissions, etc. Further, it can also configure the operator access control strategy to verify the permission operation level of different roles in the system, specifically, the system administrator has all operation permissions, the business administrator has all operation permissions under the corresponding section corresponding to the business needs, and the business members need to apply to the administrator for the allocation of permissions.

[0101] In one embodiment, the medical big model can also provide different types of big models, application cases, technical documents, model call introductions, and model compression and deployment process guidance; among them, different types of big models can include big models for computer vision tasks, natural speech processing big models, multimodal big models, scientific computing big models, etc.; the medical big model can also include model fine-tuning tools, model compression tools, model conversion tools, model call tools and interface instructions. Furthermore, the medical big model can also provide a visual interaction component, which can include a first viewing component, a second viewing component and a debugging component. The first viewing component can be used to view the model fine-tuning iteration performance trajectory, specifically, the iteration gap, communication time, calculation time, communication operator training time statistics, etc.; the second viewing component is used to view the data flow direction, model structure, and calculation graph node properties of the operator in the model network; the debugging component can be used to set monitoring points, monitor training anomalies, and view parameter changes.

[0102] In one embodiment, the medical big model has multiple performances after training, and the multiple performances may include at least one of stability, robustness, fairness, explainability, auditability, and maintainability. Specifically, stability means that when the medical big model is running under preset pressure conditions, resources such as TPS, CPU, memory, and disk can be used normally; in the same business scenario, within a period of time determined according to business characteristics, the deviation ratio between the actual prediction accuracy of the medical big model and the expected prediction accuracy is no more than 1%; unified planning and analysis of various links such as network design, selection, installation, and debugging can isolate key business partitions, strengthen business links and security group policies, etc. Robustness means that after training the medical big model, data disturbance, noise, and outliers will not have a significant impact on the performance of the medical big model. Explainability means that the medical big model can explain and understand the model decisions and prediction results, help users understand how the model makes decisions or predicts results, and increase the reliability of the model. Auditability means that the medical big model stores operation files such as operation manuals and process logs, and the operation and maintenance management of databases, platforms, etc. are recorded in detail. Maintainability means that the medical big model can diagnose and repair faults during use.

[0103] The fault diagnosis of the medical big model specifically includes: fault detection of task-related states and / or fault detection of system-related states. Fault detection of task-related states includes: information related to computational graph variables, collective communication-related indicators, data enhancement-related indicators, computing device-related indicators, task scheduling trajectories, etc.; fault detection of system-related states: inter-chip communication bandwidth, CPU three-level cache-related indicators, SDRAM (synchronous dynamic random access memory) read-write bandwidth-related indicators, system and process memory usage, high-bandwidth memory-related indicators, network card rate, error rate, packet loss rate, PCIe (a high-speed serial computer expansion bus standard) read-write bandwidth-related indicators. It can be understood that when a fault occurs, the medical big model can save necessary platform system information, such as software and hardware parameters; and can support fault analysis and problem location when a fault occurs. Furthermore, the medical big model can also perform a variety of health monitoring methods, such as automatic state collection, log analysis, heartbeat monitoring, etc.

[0104] In one embodiment, the fault repair of the medical big model specifically includes: detecting and repairing application faults, for example, determining whether the medical big model is invalid through concept drift detection, determining reasoning faults through operator overflow, and also through distributed computing fault detection and disaster recovery. When a node managed by the platform fails, the faulty resources are isolated and the medical big model tasks that were running when the failure occurred are automatically rescheduled. Furthermore, the present application also provides for resetting, repairing and automatic isolation of some specific faults, see Table 1.

[0105] Table 1 Fault list for resetting, repairing, and automatically isolating specific faults

[0106]

[0107] In one embodiment, the medical big model can also provide feedback channels and optimization mechanisms. The feedback channels can support telephone, SMS, APP, public account, web page, email, etc. The feedback content can include typical error examples of the medical big model, incompatibility issues during resource customization, and suggestions for the use of the medical big model. The optimization mechanism can optimize service capabilities based on the feedback content and improve the model effect.

[0108] In one embodiment, the medical big model also includes a communication community module, which is mainly used for target patients to conduct knowledge questions and answers, topic discussions, initiate activities, etc. Furthermore, the medical big model can also provide a technical support mechanism, which can optimize the software installation and configuration backup, build the operating environment, introduce the working principle of the medical big model, train the operation process of the medical big model, upgrade the version, etc.

[0109] In one embodiment, technicians associated with the large medical model can provide remote support services, that is, they can communicate with users in real time through telephone, email or other effective means to assist users in solving related problems and system failures; when remote services cannot solve the problem, technicians need to go to the user's site to solve the problem to ensure the normal operation of the large medical model; and provide regular on-site maintenance services to respond to questions raised by customers in a timely manner.

[0110] In one embodiment, the degree of change in the business effect of the medical model is also evaluated. Specifically, under the same business evaluation index system, the following formula is used to calculate the ratio of the improved business effect value to the original value to determine the degree of change in the business effect of the medical model.

[0111]

[0112]

[0113] Among them, S1 and S2 represent the improvement rate (degree of change) of business effect; M after It indicates the business effect value after applying the medical big model under the evaluation of indicator A; M before It indicates the business effect value before applying the medical big model under the evaluation of indicator A. It should be noted that in the formula corresponding to S1, the business evaluation indicator value is directly proportional to the business effect, and in the formula corresponding to S2, the business evaluation indicator value is inversely proportional to the business effect.

[0114] In one embodiment, the business optimization degree of the medical big model is also evaluated. Specifically, the following formula is used to calculate the labor time reduction ratio brought by the medical big model, and the labor replacement rate of the medical big model is determined to determine the business optimization degree of the medical big model.

[0115]

[0116] Where R represents the labor replacement rate; T after It represents the man-hours required for the business after applying the medical big model; T before Indicates the man-hours required for the business before the application of the medical big model.

[0117] In one embodiment, the resource optimization scheduling of the large medical model will also be evaluated; wherein, the resource optimization scheduling of the large medical model includes the richness of the resource optimization strategies of the large medical model during the training phase, and the optimization strategies include but are not limited to machine combing, cluster merging, machine memory expansion, GPU / CPU switching, task classification and resource grouping.

[0118] Specifically, the following formula is used to calculate the actual resource utilization rate within a specified time based on the training log of the medical big model to determine the resource optimization scheduling capability.

[0119]

[0120] Among them, s 1 Indicates that resources have been allocated; 2 Indicates the actual usage of resources; L 1 Indicates resource utilization.

[0121] In one embodiment, the real-time response of the medical big model is also evaluated. Specifically, the following formula is used to determine the real-time response of the medical big model based on the time from the user initiating the request to the medical big model returning the result.

[0122] ES_T=R_T finish -R_T start

[0123] Among them, ES_T represents the response time; R_T finish Indicates the time when the medical big model returns the result (non-streaming) or the time when the medical big model returns the first byte result (streaming); R_T start Indicates the start time of the user's request.

[0124] In one embodiment, the mean time between failures of the medical model is also evaluated. Specifically, the following formula is used to calculate the reliability of the medical model in the application process by the time the medical model can operate normally and the longer the mean time between failures.

[0125]

[0126] Among them, MTTF stands for mean time to failure; S all represents the uptime in the medical large model; S fail Indicates the number of failures of the medical large model.

[0127] In one embodiment, the medical big model can process modality data such as text, image, voice, video, music, code, 3D model, etc.

[0128] In one embodiment, the medical big model can perform image-text retrieval, image-based text question and answer, image-based text description, text-to-image generation, speech synthesis, speech recognition, music video retrieval, and video description generation.

[0129] In one embodiment, the medical big model can perform sequence labeling. The type of sequence labeling is not limited; for example, sequence labeling is named entity recognition, semantic labeling, part-of-speech labeling, word segmentation, etc. In one embodiment, the medical big model can perform text classification, sentiment analysis, etc. In one embodiment, the medical big model can perform sentence pair relationship processing such as natural language processing, question answering, and text semantic similarity. In one embodiment, the medical big model can perform text generation processing such as machine translation, text summarization, and dialogue system. In one embodiment, the medical big model can perform knowledge extraction processing such as relationship extraction, event extraction, and table extraction.

[0130] In one embodiment, the medical model can perform spam judgment and image classification processing. In one embodiment, the medical model can perform image recognition processing such as face recognition, character recognition, and vehicle recognition. In one embodiment, the medical model can perform target detection processing such as face detection, pedestrian detection, text detection, and key point detection. In one embodiment, the medical model can perform image recognition processing such as face recognition, character recognition, and vehicle recognition. In one embodiment, the medical model can perform image segmentation processing such as semantic segmentation, instance segmentation, and panorama segmentation. In one embodiment, the medical model can perform image generation processing such as picture generation, video generation, and image editing.

[0131] In one embodiment, the medical large model can perform speech recognition processing such as Chinese speech recognition, multi-language recognition, multi-dialect recognition, multi-language mixed reading recognition, automatic recognition of meeting records, speaker information recognition, and voice wake-up.

[0132] In one embodiment, the medical large model is capable of performing speech generation processing such as Chinese speech generation, streaming speech generation, multi-language generation, multi-timbre generation, music generation, and intelligent dubbing.

[0133] In one embodiment, the language comprehension ability of the medical big model is also evaluated. Specifically, the medical big model identifies the intent of each data in the test data set and the key information contained therein through multiple rounds of task-based dialogues, and calculates the precision, recall and F1 value respectively using the following formulas.

[0134]

[0135] Among them, P A Indicates the prediction accuracy; A 1 It indicates the number of positive examples predicted correctly; A indicates the number of data predicted as positive examples.

[0136]

[0137] Among them, R A A represents the prediction recall rate; 1 Indicates the number of positive examples predicted correctly; A 答 Indicates the actual number of positive examples.

[0138]

[0139] Among them, F A Indicates the predicted F1 value; P A Represents the prediction accuracy; R A Represents the prediction recall.

[0140] In other implementations, the accuracy, recall and F1 value of the question-answering dialogue need to be calculated separately based on single-round question-answering and multi-round question-answering, where the single-round question-answering capability mainly determines the direct recall and accuracy as well as the recall and accuracy of the Top3, and the multi-round question-answering capability determines the context association capability based on the single-round capability.

[0141] In one embodiment, the text generation capability of the medical model is also evaluated.

[0142] In a specific implementation, for a reading comprehension task, the EM value of a text generation result is calculated using the following formula.

[0143]

[0144] Among them, C represents the candidate answer generated by the medical big model; R represents the correct answer marked by humans; Count c Indicates the number of answers generated by the medical big model; Count exactmatch(c,r) Indicates the number of sentences where the answers generated by the medical big model completely match the correct answers.

[0145] In a specific implementation, for the summary task, the objective indicator ROUGE-N is calculated using the following formula.

[0146]

[0147] Where N is n-gram, which means the number of bytes in the text content sliding window, and the reference value is 2; Count match (gram n ) represents the number of n-grams in common between the reference summary and the machine-generated summary; Count(gram n ) represents the number of n-grams in the reference summary.

[0148] In a specific implementation, the text generation results of medical professional documents such as doctor's orders and electronic medical records can also be scored according to fluency, accuracy, and interpretability, and the average score of the generated text scores is used as the acceptability score of the generated text.

[0149] In one embodiment, the multi-round active interaction capability of the medical large model is also evaluated.

[0150] In a specific implementation, the following formula is used to calculate the multi-round dialogue task completion rate of the dialogue system.

[0151]

[0152] Among them, P G represents the completion rate of multi-round dialogue tasks; G 1 It represents the number of multi-round dialogues that actually complete the task; G represents the total number of multi-round dialogue data.

[0153] In a specific implementation, the following formula is used to calculate the accuracy of the following text recognition of the dialogue system, that is, if a certain round of multiple rounds of dialogue is correct, the following text recognition is considered to be correct.

[0154]

[0155] Among them, P D represents the completion rate of multi-round dialogue tasks; G 1 It represents the number of multi-round dialogues that actually complete the task; D represents the total number of multi-round dialogue data.

[0156] In a specific implementation, the multi-round active interaction capability of the medical large model can also be evaluated based on the average scores of fluency scores, coherence scores, and interestingness scores of the multi-round dialogue tasks.

[0157] In a specific implementation, the multi-round active interaction capability of the medical large model can also be evaluated from three dimensions: the fun, sensitivity, and clarity of the multi-round dialogue tasks.

[0158] See also Figure 2 , Figure 2 It is a schematic diagram of the framework of an embodiment of a medical service device based on a medical big model provided by the present application. The medical service device 20 based on the medical big model includes an acquisition module 21 and an analysis module 22; the acquisition module 21 is used to acquire the service request data of the target doctor in the target service scenario; wherein the target service scenario is any one of the following medical service scenarios: diagnosis decision suggestion, treatment plan recommendation, medical record generation, medical record quality control, and medical record information management; the analysis module 22 is used to analyze the service request data based on the medical big model, and obtain the medical service suggestion output by the medical big model for the target doctor in the target service scenario.

[0159] Among them, in the case where the target service scenario is a diagnostic decision suggestion, the analysis module 22 is also used to respond to at least one of the physical examination information and consultation information of the target doctor obtained by analyzing the service request data based on the medical big model, and output a medical service suggestion including a recommended diagnosis and its cause, and at least one of the patient's physique; respond to at least one of the clinical manifestation information, examination and inspection information, and medical imaging information obtained by analyzing the service request data based on the medical big model, and output a medical service suggestion including at least one of the recommended diagnosis and the basis for diagnosis; respond to the clinical manifestation information obtained by analyzing the service request data based on the medical big model, and output a medical service suggestion including a recommended disease and its diagnosis probability; respond to the analysis of the service request data based on the medical big model, and output a medical service suggestion including a recommended disease and its diagnosis probability. The medical service system outputs medical service suggestions including disease risks based on the analysis of at least one of disease risk factors and life habit information; in response to the analysis of service request data based on the medical big model to obtain at least one of individual differences, physical condition information, and disease influencing factors, the medical service suggestions including at least one of personalized examination contraindications and recommended diagnoses are output; in response to the analysis of service request data based on the medical big model to obtain medication history and disease information, the medical service suggestions including drug interactions are output; in response to the analysis of service request data based on the medical big model to obtain treatment feedback information and disease change information, the medical service suggestions including diagnosis update results are output; wherein the recommended diagnosis includes one candidate disease, or multiple concurrent candidate diseases.

[0160] Wherein, in the case where the target service scenario is the treatment plan recommendation, the analysis module 22 is also used for: in response to obtaining patient medical history information through analysis of service request data based on the medical big model, outputting medical service suggestions including patient portraits; in response to obtaining rehabilitation formulation requests through analysis of service request data based on the medical big model, and the medical big model also inputs patient portraits, outputting medical service suggestions including at least one of rehabilitation plans and node prompts; wherein the node prompts are used to prompt execution of relevant rehabilitation actions at each node in the rehabilitation plan; in response to obtaining condition change information during the rehabilitation process through analysis of service request data based on the medical big model, outputting medical service suggestions including updated patient portraits; in response to obtaining rehabilitation formulation requests through analysis of service request data based on the medical big model, and the medical big model also inputs patient portraits, outputting medical service suggestions including at least one of rehabilitation plans and node prompts; wherein the node prompts are used to prompt execution of relevant rehabilitation actions at each node in the rehabilitation plan; The model analyzes the service request data to obtain patient information, and outputs medical service suggestions including at least one of emergency treatment, auxiliary guidance, treatment plan formulation, evidence-based treatment plan, treatment plan explanation, and drug selection; wherein the treatment plan explanation is provided based on at least one of three-dimensional rendering and generative speech; in response to obtaining at least one of the patient's condition and treatment progress through the analysis of the service request data based on the medical big model, outputs medical service suggestions including an updated treatment plan; in response to obtaining at least one of the patient's physique, physical condition, individual difference, and treatment progress through the analysis of the service request data based on the medical big model, outputs medical service suggestions including at least one of surgical plan and surgical contraindications; in response to obtaining at least one of the patient's physique, physical condition, individual difference, and treatment progress through the analysis of the service request data based on the medical big model, outputs medical service suggestions including at least one of surgical plan and surgical contraindications; Based on the analysis of at least one of the patient's clinical characteristics and test and examination data, and the medical big model also inputs historical diagnosis and treatment plans, it outputs medical service recommendations including risk assessments of treatment plans; in response to a digital treatment request obtained by analysis of the service request data based on the medical big model, it outputs medical service recommendations including evidence-based treatment interventions; wherein the evidence-based treatment interventions are driven by the medical big model; in response to historical medical records obtained by analysis of the service request data based on the medical big model, and the medical big model also inputs medical guidelines, it outputs medical service recommendations including a disease rehabilitation pathway framework; in response to an updated patient portrait obtained by analysis of the service request data based on the medical big model, it outputs medical service recommendations including an updated rehabilitation plan; The medical big model is used to analyze the service request data to obtain an updated rehabilitation plan, and a medical service suggestion containing updated node prompts is output; in response to the patient's condition and the customized education request obtained by the medical big model through the analysis of the service request data, a medical service suggestion containing education knowledge is output; wherein the education knowledge is consistent with the patient's condition; in response to at least one of clinical characteristics, medical history information, diagnosis information, and treatment information obtained by the medical big model through the analysis of the service request data, a medical service suggestion containing at least one of the course of the disease and the predicted treatment result is output; in response to at least one of the patient's condition information and medication history information obtained by the medical big model through the analysis of the service request data, a medical service suggestion containing the predicted risk of adverse events is output;In response to obtaining a medical search request by analyzing the service request data based on the medical big model, a medical service suggestion containing medical knowledge related to the medical search request is output. ;

[0161] Among them, when the target service scenario is medical record generation, the analysis module 22 is also used for: in response to analyzing the service request data based on the medical big model to obtain at least one of the existing medical record content, doctor input content, and examination and inspection results, outputting medical service recommendations containing medical record content data in the first subdivided scenario of medical record generation; wherein the first subdivided scenario includes any one of outpatient / emergency medical record generation, inpatient medical record generation, ward nursing record generation, and examination result document generation.

[0162] Wherein, in the case where the first subdivision scenario is the generation of the outpatient / emergency medical record, the above-mentioned medical record content data includes at least one of the following: medical record homepage, medical record record, emergency consultation time, emergency observation record, rescue record, prescription record, external prescription, test application record, test report record, inspection application record, and inspection execution record; and / or, in the case where the first subdivision scenario is the generation of the inpatient medical record, the above-mentioned medical record content data includes at least one of the following: automatic coding of at least one of the medical record, medical history, disease diagnosis, and surgery, medical order record, clinical pathway record, test application record, inspection application record, inspection report record, and inspection critical value record. Record; and / or; in the case where the first subdivision scenario is the generation of ward nursing records, the above-mentioned medical record content data includes at least one of the following: ward patient information, nursing assessment records, nursing-related clinical pathway records, doctor's order execution records, descriptive nursing items, nursing electronic signature records, adverse event records; and / or, in the case where the first subdivision scenario is the generation of examination result documents, the above-mentioned medical record content data includes at least one of the following: specimen records, specimen transmission records, test report records, test result records, test critical value records, quality control records, and the medical record content belongs to internal institutional reports, reports on the return of sent specimens, and reports requested by external institutions.

[0163] Among them, the analysis module 22 is also used for: in response to the service request data based on the medical big model, it also analyzes at least one of the previous diagnosis and treatment conditions and medical order information inside and outside the medical institution, and outputs medical service recommendations that at least include medical order verification; summarizes and shares the patient's medical health information; wherein the medical health information includes at least one of the following: medical information inside and outside the institution, health records, physical sign detection, and follow-up information.

[0164] Among them, when the target service scenario is medical record quality control, the analysis module 22 is also used to: respond to the medical record quality control request obtained by analyzing the service request data based on the medical big model, and output a medical service recommendation including the medical record quality control results in the second subdivided scenario of medical record quality control; wherein the second subdivided scenario includes any one of outpatient / emergency medical record quality control, inpatient medical record quality control, ward nursing record quality control, and examination result document quality control.

[0165] Among them, in the case where the second sub-scenario is outpatient / emergency medical record quality control, the analysis module is also used to: respond to a query request for historical test results obtained by analyzing the service request data based on the medical big model, and output a medical service recommendation that includes a summary of test reports of the same patient in different medical institutions; respond to at least one of the previous diagnosis and treatment conditions and medication conditions inside and outside the medical institution obtained by analyzing the service request data based on the medical big model, and output a medical service recommendation that includes at least medical order verification; respond to a request for integrity quality control of outpatient / emergency medical records obtained by analyzing the service request data based on the medical big model, and output a medical service recommendation that includes integrity quality control results.

[0166] Among them, in the case where the second sub-scenario is quality control of inpatient medical records, the analysis module 22 is also used for: in response to at least one of a terminology verification request and a logic verification request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including the verification result; in response to a prescription review request obtained by analyzing the service request data based on the medical big model, and the medical big model also inputs medical-related laws, outputting a medical service suggestion including a drug dispensing decision; in response to a request for integrity quality control of inpatient medical records obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an integrity quality control result; in response to a request for consistency quality control of the content of inpatient medical records obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a consistency quality control result; in response to at least one of a test result and a medication situation obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an infection warning prompt.

[0167] Among them, in the case where the second subdivided scenario is the quality control of ward nursing records, the analysis module 22 is also used to: in response to a patient information request obtained by analyzing the service request data based on the medical big model, output a medical service suggestion including at least one record of admission, discharge, and transfer; in response to an image recognition request obtained by analyzing the service request data based on the medical big model, output a medical service suggestion including an image recognition result; wherein the image recognition result includes the result of identifying at least one of the patient, the drug, and the test specimen through the image; in response to an information import request obtained by analyzing the service request data based on the medical big model, output a medical service suggestion including an information import result; wherein the information import result represents vital signs, At least one kind of information in the nursing treatment is automatically imported into the corresponding record sheet through the mobile device; in response to the patient's activity records in other departments in the hospital obtained by analyzing the service request data based on the medical big model, medical service suggestions including nursing measures are output; in response to the patient's condition obtained by analyzing the service request data based on the medical big model, medical service suggestions including nursing quality assessment are output; in response to the consistency quality control request for the ward nursing record obtained by analyzing the service request data based on the medical big model, medical service suggestions including consistency quality control results are output; in response to the integrity quality control request for the ward nursing record obtained by analyzing the service request data based on the medical big model, medical service suggestions including integrity quality control results are output.

[0168] Among them, in the case where the second subdivided scenario is the quality control of the examination result document, the analysis module 22 is also used for: in response to an automatic corresponding request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion containing the corresponding result; wherein the corresponding result includes the correspondence between the test report and the clinical test application; in response to an automatic judgment prompt obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion containing automatic prompt information related to the examination measurement value; wherein the automatic prompt information includes at least one of a reasonable range of the examination measurement value and a reasonable vocabulary range of the annotation description; in response to a content verification request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion containing report verification information; in response to a consistency quality control request for the examination result document obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion containing a consistency quality control result; in response to an integrity quality control request for the examination result document obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion containing an integrity quality control result.

[0169] Among them, when the target service scenario is medical record information management, the analysis module 22 is also used for: in response to a quality verification request for the medical record homepage obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one of an automatic verification result and a verification abnormality prompt during the medical record generation process; in response to a defect verification request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including defects in the medical record content; in response to an automatic judgment request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a medical record quality control record; in response to a writing verification request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a writing verification result; in response to a complete medical record obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a defect in the medical record content; In response to a quality verification request, a medical service suggestion including the quality verification results of both the running medical record and the terminal medical record is output; in response to a quality control verification template obtained by analyzing the service request data based on the medical big model, a medical service suggestion including a medical record quality control report is output; wherein the medical record quality control report complies with the quality control verification template; in response to a data comparison request for medical records and quality control records obtained by analyzing the service request data based on the medical big model, a medical service suggestion including at least a customized structured medical record is output; in response to abnormal data operation behavior obtained by analyzing the service request data based on the medical big model, a medical service suggestion including an automatic alarm prompt is output; in response to a quality control analysis request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including at least one of a quality control data summary and a multi-dimensional analysis report is output.

[0170] Among them, the above-mentioned service request data includes at least one modal data among text, picture, voice and video.

[0171] See also Figure 3 , Figure 3 : is a schematic diagram of a framework of an embodiment of an electronic device provided by the present application. The electronic device 30 includes a memory 31 and a processor 32. The memory 31 stores program instructions, and the processor 32 is used to execute the program instructions to implement the steps in any of the above-mentioned embodiments of the medical service method based on the medical big model. Specifically, the electronic device 30 may include but is not limited to: a mobile phone, a tablet computer, a laptop computer, a server, etc., which are not limited here.

[0172] Specifically, the processor 32 is used to control itself and the memory 31 to implement the steps in any of the above-mentioned embodiments of the medical service method based on the medical big model. The processor 32 can also be called a CPU (Central Processing Unit). The processor 32 may be an integrated circuit chip with signal processing capabilities. The processor 32 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 32 can be implemented by an integrated circuit chip.

[0173] In one embodiment, the electronic device 30 is deployed based on at least one of the physical security requirements, functional security requirements, and management security requirements; the physical security requirements include fire protection, lightning protection, waterproofing, disaster recovery, authorization, etc., the functional security requirements include device label security, hardware interface security, firmware security, driver security, etc., and the management security requirements include management mechanisms, management personnel, etc. That is, the electronic device has functions such as fire protection, lightning protection, waterproofing, disaster recovery, authorization, etc., and the labels, hardware interfaces, firmware, drivers, etc. of the electronic device are all safe. Furthermore, the electronic device 30 may also include a management mechanism and be provided with corresponding management personnel.

[0174] In one embodiment, the electronic device 30 is used in at least one stage of training and reasoning of a large medical model. In at least one stage of training and reasoning, the electronic device 30 is deployed based on at least one security requirement of network bandwidth, network latency, network packet loss rate, and network jitter.

[0175] In one embodiment, the electronic device 30 may also be provided with at least one of information theft protection and security vulnerability protection; and / or, the electronic device 30 operates in a chip heterogeneous scenario, that is, the chip in the electronic device 30 can operate stably in a heterogeneous scenario, and the heterogeneous scenario may include a scenario in which a central processing unit and a graphics processing unit are combined.

[0176] In the above implementation, the hardware devices in the electronic device 30 are mainly limited. In other implementations, the software devices in the electronic device 30 may also be limited.

[0177] In one embodiment, the electronic device 30 runs multiple software facilities, and the multiple software facilities include any two or more of the following: dependency library, artificial intelligence framework, vector database, large model middleware, interface; wherein the electronic device 30 is provided with protection means for the multiple software facilities, and the protection means include at least one of vulnerability management and security update. Specifically, the electronic device 30 can scan and repair vulnerabilities at intervals of preset time, and update security patches in a timely manner, and the preset time is set by the user and is not limited here.

[0178] Furthermore, the electronic device 30 may also be provided with protection measures related to at least one of data collection, data preprocessing, and data usage, and the protection measures include at least one of storage security, privacy protection, process safety, and destruction security; and / or, the electronic device 30 is provided with protection measures related to at least one of model training security, model reasoning security, and model file security.

[0179] In one embodiment, the electronic device 30 may also be provided with protection means related to at least one of service security and content security; wherein service security includes: at least one of mean time between failures, service compliance, and feedback handling mechanism, and content security includes: at least one of content reliability, content compliance, and value alignment. Service security and content security may indicate that both the service and the content meet the requirements. Taking the example that service security includes mean time between failures, service security may indicate that the mean time between failures is less than the preset time between failures.

[0180] See also Figure 4 , Figure 4 The computer-readable storage medium 40 stores program instructions 41 that can be executed by a processor, and the program instructions 41 are used to implement the steps in any of the above-mentioned doctor service method embodiments based on the medical big model.

[0181] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0182] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0183] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0184] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0185] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0186] 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 computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0187] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A doctor service method based on a medical big model, characterized in that: include: Obtaining service request data of a target doctor in a target service scenario; wherein the target service scenario is any one of the following medical service scenarios: diagnosis decision suggestion, treatment plan recommendation, medical record generation, medical record quality control, and medical record information management; The service request data is analyzed based on the medical big model to obtain the medical service recommendations output by the medical big model for the target doctor in the target service scenario.

2. The method according to claim 1, characterized in that In the case where the target service scenario is the diagnostic decision suggestion, the method further includes at least one of the following: In response to obtaining at least one of the physical examination information and the consultation information of the target doctor through analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one of a recommended diagnosis and its cause, and a patient's physical condition; In response to obtaining at least one of clinical manifestation information, examination and test information, and medical imaging information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one of a recommended diagnosis and a diagnosis basis; In response to obtaining clinical performance information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a recommended disease and its diagnosis probability; In response to obtaining at least one of a disease risk factor and a life habit information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a disease risk; In response to analyzing the service request data based on the medical big model to obtain at least one of individual difference conditions, physical condition information, and disease influencing factors, outputting a medical service suggestion including at least one of personalized examination contraindications and recommended diagnosis; In response to obtaining medication history and disease information by analyzing the service request data based on the medical big model, outputting medical service recommendations including interactions between drugs; In response to obtaining treatment feedback information and condition change information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a diagnosis update result; The recommended diagnosis includes one candidate disease, or multiple concurrent candidate diseases.

3. The method according to claim 1, characterized in that: When the target service scenario is the treatment plan recommendation, the method further includes at least one of the following: In response to obtaining patient medical record information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a patient portrait; In response to obtaining a rehabilitation formulation request by analyzing the service request data based on the medical big model, and the medical big model also inputting a patient portrait, outputting a medical service suggestion including at least one of a rehabilitation plan and a node prompt; wherein the node prompt is used to prompt execution of relevant rehabilitation actions at each node in the rehabilitation plan; In response to analyzing the service request data based on the medical big model to obtain information on changes in the condition during the rehabilitation process, output a medical service suggestion including an updated patient portrait; In response to obtaining patient information by analyzing the service request data based on the medical big model, outputting medical service suggestions including at least one of emergency treatment, auxiliary guidance, treatment plan formulation, treatment plan evidence-based, treatment plan explanation, and drug selection; wherein the treatment plan explanation is provided based on at least one of three-dimensional rendering and generative speech; In response to obtaining at least one of a patient condition and a treatment process by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an updated treatment plan; In response to obtaining at least one of the patient's constitution, physical condition, individual difference, and treatment process by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one of a surgical plan and surgical contraindications; In response to obtaining at least one of the patient's clinical characteristics and test data by analyzing the service request data based on the medical big model, and the medical big model also inputs historical diagnosis and treatment plans, outputting a medical service recommendation including a risk assessment of the treatment plan; In response to a digital treatment request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an evidence-based treatment intervention; wherein the evidence-based treatment intervention is driven by the medical big model; In response to obtaining historical medical records by analyzing the service request data based on the medical big model, and the medical big model also inputs medical guidelines, outputting medical service recommendations including a disease recovery path framework; In response to obtaining an updated patient portrait by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an updated rehabilitation plan; In response to obtaining an updated rehabilitation plan by analyzing the service request data based on the medical big model, outputting a medical service suggestion including updated node prompts; In response to obtaining the patient's condition and the customized education request by analyzing the service request data based on the medical big model, outputting a medical service suggestion containing education knowledge; wherein the education knowledge is consistent with the patient's condition; In response to obtaining at least one of clinical characteristics, medical history information, diagnosis information, and treatment information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one of a course of disease and a predicted treatment result; In response to obtaining at least one of the patient's condition information and medication history information by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a predicted risk of an adverse event; In response to obtaining a medical search request by analyzing the service request data based on the medical big model, a medical service suggestion including medical knowledge related to the medical search request is output.

4. The method according to claim 1, characterized in that In the case where the target service scenario is the medical record generation, the method further includes: In response to obtaining at least one of existing medical record content, doctor input content, and examination and test results by analyzing the service request data based on the medical big model, outputting a medical service suggestion including medical record content data in a first segmented scenario generated by the medical record; Among them, the first subdivided scenario includes any one of outpatient / emergency medical record generation, inpatient medical record generation, ward nursing record generation, and examination result document generation.

5. The method according to claim 4, characterized in that In the case where the first subdivided scenario is the generation of the outpatient / emergency medical record, the medical record content data includes at least one of the following: medical record homepage, medical record record, emergency consultation time, emergency observation record, rescue record, prescription record, external prescription, test application record, test report record, inspection application record, inspection execution record; and / or, In the case where the first subdivided scenario is the generation of the inpatient medical record, the medical record content data includes at least one of the following: automatic coding of at least one of medical records, medical histories, disease diagnoses, and surgeries, medical order records, clinical pathway records, test application records, examination application records, examination report records, and examination critical value records; and / or; In the case where the first subdivided scenario is the generation of the ward nursing record, the medical record content data includes at least one of the following: ward patient information, nursing assessment records, nursing-related clinical pathway records, doctor's order execution records, descriptive nursing items, nursing electronic signature records, and adverse event records; and / or, When the first subdivided scenario is the generation of the examination result document, the medical record content data includes at least one of the following: specimen records, specimen transmission records, test report records, test result records, test critical value records, quality control records, and the medical record content belongs to internal institution reports, reports on returned specimens sent out, and reports requested by external institutions.

6. The method according to claim 4, characterized in that The method further comprises at least one of the following: In response to analyzing the service request data based on the medical big model and obtaining at least one of the diagnosis and treatment conditions and medical advice information in the past inside and outside the medical institution, outputting a medical service suggestion including at least medical advice verification; Aggregate and share patients' medical and health information; wherein the medical and health information includes at least one of the following: medical information inside and outside the institution, health records, physical sign detection, and follow-up information.

7. The method according to claim 1, characterized in that When the target service scenario is the medical record quality control, the method further includes: In response to obtaining a medical record quality control request by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a medical record quality control result in a second subdivided scenario of the medical record quality control; Among them, the second subdivided scenario includes any one of outpatient / emergency medical record quality control, inpatient medical record quality control, ward nursing record quality control, and examination result document quality control.

8. The method according to claim 7, characterized in that When the second subdivided scenario is the outpatient / emergency medical record quality control, the method further includes at least one of the following: In response to a query request for historical test results obtained by analyzing the service request data based on the medical big model, output a medical service suggestion including a summary of test reports of the same patient made in different medical institutions; In response to analyzing the service request data based on the medical big model to obtain at least one of the diagnosis and treatment conditions and medication conditions in and outside the medical institution in the past, outputting a medical service suggestion including at least a doctor's order verification; In response to an analysis of the service request data based on the medical big model to obtain a request for complete quality control of outpatient / emergency medical records, a medical service recommendation including an complete quality control result is output.

9. The method according to claim 7, characterized in that: When the second subdivision scenario is the inpatient medical record quality control, the method further includes at least one of the following: In response to obtaining at least one of a terminology verification request and a logic verification request by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a verification result; In response to obtaining a prescription review request by analyzing the service request data based on the medical big model, and the medical big model also inputs medical-related laws, outputting a medical service suggestion including a drug dispensing decision; In response to an analysis of the service request data based on the medical big model to obtain a request for complete quality control of the inpatient medical record, outputting a medical service suggestion including a complete quality control result; In response to a consistency quality control request for inpatient medical records obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a consistency quality control result; In response to obtaining at least one of a test result and a medication situation by analyzing the service request data based on the medical big model, a medical service suggestion including an infection early warning prompt is output.

10. The method according to claim 7, characterized in that In the case where the second subdivision scenario is the quality control of the ward nursing records, the method further includes at least one of the following: In response to a patient information request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one record of admission, discharge, and transfer; In response to an image recognition request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an image recognition result; wherein the image recognition result includes a result of recognizing at least one of a patient, a drug, and a test specimen through an image; In response to an information import request obtained by analyzing the service request data based on the medical big model, a medical service suggestion including an information import result is output; wherein the information import result represents at least one of vital signs and nursing treatment information automatically imported into a corresponding record sheet through a mobile device; In response to obtaining the patient's activity records in other departments of the hospital by analyzing the service request data based on the medical big model, outputting medical service suggestions including nursing measures; In response to obtaining a patient condition by analyzing the service request data based on the medical big model, outputting a medical service recommendation including a nursing quality assessment; In response to a request for consistency quality control of ward nursing records obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a consistency quality control result; In response to an analysis of the service request data based on the medical big model to obtain an integrity quality control request for ward nursing records, a medical service recommendation including an integrity quality control result is output.

11. The method according to claim 7, characterized in that In the case where the second subdivided scenario is the inspection result document quality control, the method further includes at least one of the following: In response to automatically obtaining a corresponding request by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a corresponding result; wherein the corresponding result includes a corresponding relationship between a test report and a clinical test application; In response to an automatic judgment prompt obtained by analyzing the service request data based on the medical big model, a medical service suggestion including automatic prompt information related to the examination measurement value is output; wherein the automatic prompt information includes at least one of a reasonable range of the examination measurement value and a reasonable vocabulary range of the annotation description; In response to analyzing the service request data based on the medical big model to obtain a content verification request, outputting a medical service suggestion including report verification information; In response to a consistency quality control request for the examination result document obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a consistency quality control result; In response to an analysis of the service request data based on the medical big model to obtain an integrity quality control request for the inspection result document, a medical service suggestion including the integrity quality control result is output.

12. The method according to claim 1, characterized in that When the target service scenario is the medical record information management, the method further includes at least one of the following: In response to a quality check request for the medical record front page obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least one of an automatic check result and a check abnormality prompt during the medical record generation process; In response to a defect verification request obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including defects in medical record content; In response to automatically determining a request based on the medical big model analyzing the service request data, outputting a medical service suggestion including a medical record quality control record; In response to obtaining a writing verification request by analyzing the service request data based on the medical macro model, outputting a medical service suggestion including a writing verification result; In response to a quality check request for a complete medical record obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including quality check results of both the running medical record and the terminal medical record; In response to obtaining a quality control verification template by analyzing the service request data based on the medical big model, outputting a medical service suggestion including a medical record quality control report; wherein the medical record quality control report complies with the quality control verification template; In response to a data comparison request for medical records and quality control records obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including at least a customized structured medical record; In response to abnormal data operation behavior obtained by analyzing the service request data based on the medical big model, outputting a medical service suggestion including an automatic alarm prompt; In response to a quality control analysis request obtained by analyzing the service request data based on the medical big model, a medical service recommendation including at least one of a quality control data summary and a multi-dimensional analysis report is output.

13. The method according to any one of claims 1 to 12, characterized in that: The service request data includes at least one modality of data including text, picture, voice, and video.

14. A doctor service device based on a medical big model, characterized in that: include: An acquisition module is used to acquire service request data of a target doctor in a target service scenario; wherein the target service scenario is any one of the following medical service scenarios: diagnosis decision suggestion, treatment plan recommendation, medical record generation, medical record quality control, and medical record information management; The analysis module is used to analyze the service request data based on the medical big model to obtain the medical service recommendations output by the medical big model for the target doctor in the target service scenario.

15. An electronic device, characterized in that: The method comprises a memory and a processor coupled to each other, wherein the memory stores program instructions, and the processor is used to execute the program instructions to implement the method according to any one of claims 1 to 13.

16. The device according to claim 15, characterized in that The electronic equipment is deployed based on at least one security requirement among physical security requirements, functional security requirements, and management security requirements; And / or, the electronic device is used in at least one stage of training and reasoning of a large medical model, and in at least one stage of the training and reasoning, the electronic device is deployed based on at least one guarantee requirement of network bandwidth, network delay, network packet loss rate, and network jitter; And / or, the electronic device is provided with at least one of information theft protection and security vulnerability protection; And / or, the electronic device operates in a chip heterogeneous scenario.

17. The device according to claim 15, characterized in that The electronic device runs a variety of software facilities, and the various software facilities include any two or more of the following: dependency library, artificial intelligence framework, vector database, large model middleware, interface; The electronic device is provided with protection means for the multiple software facilities, and the protection means include at least one of vulnerability management and security update.

18. The device according to claim 15, characterized in that The electronic device is provided with protection means related to at least one of data collection, data preprocessing, and data use, and the protection means includes at least one of storage security, privacy protection, process security, and destruction security; and / or, The electronic device is provided with protection measures related to at least one of model training security, model reasoning security, and model file security.

19. The device according to claim 15, characterized in that The electronic device is provided with protection means related to at least one of service security and content security; Among them, the service security includes: at least one of mean time between failures, service compliance, and feedback handling mechanism; the content security includes: at least one of content reliability, content compliance, and value alignment.

20. A computer-readable storage medium, characterized in that: Program instructions that can be executed by a processor are stored, and the program instructions are used to implement the method according to any one of claims 1 to 13.

Images