Auxiliary diagnosis system, auxiliary diagnosis deployment system, related method, device, equipment and medium

Abstract

The invention discloses an auxiliary diagnosis system, an auxiliary diagnosis deployment system and a related method, device, equipment and medium, and the auxiliary diagnosis system comprises an interface layer, a data layer, a capability layer, a service layer and an application layer. The interface layer comprises a plurality of interfaces which are used for being in butt joint with different medical systems of a medical institution respectively; the data layer is used for acquiring the data called by the interface layer so as to carry out data connection with a medical system; the capability layer comprises a plurality of artificial intelligence algorithms and is used for processing data of the medical system; the service layer comprises a plurality of basic business modules and is used for providing standard services for the application layer; and the application layer comprises a plurality of medical service applications and is used for respectively realizing different medical auxiliary functions, and the medical service applications are realized by at least one basic service module. According to the scheme, the operation and maintenance convenience can be improved on the premise that auxiliary diagnosis and treatment are met.

Description

Technical Field

[0001] This application relates to the field of intelligent medical technology, and particularly to a diagnosis assistance system, a diagnosis assistance deployment system, and related methods, devices, equipment, and media. Background Art

[0002] To facilitate medical staff and reduce their work burden, the development of various auxiliary means based on existing medical systems has attracted increasing attention.

[0003] In the prior art, it is mainly achieved by integrating intelligent modules for auxiliary means on the existing medical system. However, on the one hand, this method makes the existing medical system increasingly bloated, and on the other hand, when either the intelligent module or the existing medical system is upgraded, compatibility needs to be considered again. Thus, there are problems such as inconvenient operation and maintenance in the prior art. In view of this, how to improve the operation and maintenance convenience on the premise of meeting the requirements of auxiliary diagnosis and treatment has become an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem to be solved by this application is to provide a diagnosis assistance system, a diagnosis assistance deployment system, and related methods, devices, equipment, and media, which can improve the operation and maintenance convenience on the premise of meeting the requirements of auxiliary diagnosis and treatment.

[0005] To solve the above technical problem, in the first aspect of this application, a diagnosis assistance system is provided, including: an interface layer, a data layer, a capability layer, a service layer, and an application layer. The interface layer includes several interfaces for respectively docking different medical systems of medical institutions; the data layer is used to obtain the data retrieved by the interface layer to perform data docking with the medical system; the capability layer includes several artificial intelligence algorithms for processing the data of the medical system; the service layer includes several basic business modules for providing standard services for the application layer; the application layer includes several medical business applications for respectively implementing different medical assistance functions, and the medical business applications are implemented by at least one basic business module.

[0006] To solve the above technical problems, the second aspect of the present application provides a diagnosis and treatment control method, which is applied to the auxiliary diagnosis system in the first aspect. The diagnosis and treatment control method includes: calling the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system; improving the original medical record based on the quality inspection result of the original medical record to obtain the target medical record, and calling the interface layer to write back the target medical record to the hospital information system, and performing diagnosis and treatment recommendations based on the target medical record to obtain the recommendation result, so as to call the interface layer to feedback the recommendation result to the hospital information system for doctors' reference; calling the interface layer to obtain the doctor's diagnosis from the hospital information system; performing a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable; wherein, when the first detection result indicates that the doctor's diagnosis is unreasonable, calling the interface layer to feedback the first detection result to the hospital information system; calling the interface layer to obtain the doctor's prescription from the hospital information system; performing a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable; wherein, when the second detection result indicates that the doctor's prescription is unreasonable, calling the interface layer to feedback the second detection result to the hospital information system.

[0007] To solve the above technical problems, the third aspect of the present application provides a diagnosis and treatment control device, which is applied to the auxiliary diagnosis system in the first aspect. The diagnosis and treatment control device includes: a first calling module, an information interaction module, a second calling module, a first detection module, a third calling module, and a second detection module. The first calling module is used to call the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system; the information interaction module is used to improve the original medical record based on the quality inspection result of the original medical record to obtain the target medical record, and call the interface layer to write back the target medical record to the hospital information system, and perform diagnosis and treatment recommendations based on the target medical record to obtain the recommendation result, so as to call the interface layer to feedback the recommendation result to the hospital information system for doctors' reference; the second calling module is used to call the interface layer to obtain the doctor's diagnosis from the hospital information system; the first detection module is used to perform a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable; wherein, when the first detection result indicates that the doctor's diagnosis is unreasonable, calling the interface layer to feedback the first detection result to the hospital information system; the third calling module is used to call the interface layer to obtain the doctor's prescription from the hospital information system; the second detection module is used to perform a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable; wherein, when the second detection result indicates that the doctor's prescription is unreasonable, calling the interface layer to feedback the second detection result to the hospital information system.

[0008] To solve the above technical problems, the fourth aspect of the present application provides an electronic device, including a communication circuit, a memory, and a processor. The communication circuit and the memory are respectively coupled to the processor. The memory stores program instructions, and the processor is used to execute the program instructions to implement the diagnosis and treatment control method in the second aspect.

[0009] To solve the above technical problems, a fifth aspect of the present application provides a computer-readable storage medium storing program instructions executable by a processor, and the program instructions are used to implement the diagnosis and treatment control method in the second aspect above.

[0010] To solve the above technical problems, a sixth aspect of the present application provides an auxiliary diagnosis deployment system, which includes: a customer subsystem, a service subsystem, and an interface subsystem communicatively connected between the customer subsystem and the service subsystem. The customer subsystem is used to interface with the medical system of a medical institution, and the customer subsystem is the auxiliary diagnosis system in the first aspect above.

[0011] In the above solution, the auxiliary diagnosis system includes an interface layer, a data layer, a capability layer, a service layer, an application layer, and an application layer. The interface layer includes several interfaces for respectively interfacing with different medical systems of medical institutions. The data layer is used to obtain the data retrieved by the interface layer to perform data docking with the medical system. The capability layer includes several artificial intelligence algorithms for processing the data of the medical system. The service layer includes several basic business modules for providing standard services for the application layer. The application layer includes several medical business applications for respectively implementing different medical assistance functions, and the dependent business applications are implemented by at least one basic business module. Therefore, it directly interfaces with the medical system through the interface layer, and the data layer in the auxiliary diagnosis system obtains the data retrieved by the interface layer to achieve data docking with the medical system, and the data is processed by each artificial intelligence algorithm in the capability layer and encapsulated into several basic business modules in the service layer, so that at least one basic business module forms a medical business application in the application layer, thus eliminating the need for further integration on the existing medical system. On the one hand, it can minimize the impact on the existing medical system as much as possible, neither making the existing medical system more bloated nor damaging the normal use of the existing medical system. On the other hand, since the auxiliary diagnosis system and the medical system are separated from each other, even if either party is updated, the impact on the other party is very small. Therefore, it can improve the operation and maintenance convenience on the premise of meeting the auxiliary diagnosis. Description of the Drawings

[0012] Figure 1 is a schematic framework diagram of an embodiment of the auxiliary diagnosis system of the present application;

[0013] Figure 2 is a schematic framework diagram of an embodiment of the auxiliary diagnosis deployment system of the present application;

[0014] Figure 3 is a schematic framework diagram of another embodiment of the auxiliary diagnosis system of the present application;

[0015] Figure 4a is a schematic process diagram of an embodiment of the diagnosis and treatment control method of the present application;

[0016] Figure 4b It is a schematic flowchart of an embodiment of the diagnosis and treatment control method of the present application;

[0017] Figure 5 It is a schematic framework diagram of an embodiment of the diagnosis and treatment control device of the present application;

[0018] Figure 6 It is a schematic framework diagram of an embodiment of the electronic device of the present application;

[0019] Figure 7 It is a schematic framework diagram of an embodiment of the computer-readable storage medium of the present application. Detailed implementation manners

[0020] The following will combine the accompanying drawings of the specification to elaborate on the solutions of the embodiments of the present application in detail.

[0021] In the following description, specific details such as specific system structures, interfaces, and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the present application.

[0022] The terms "system" and "network" are often used interchangeably in this article. The term " / and" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the fragment " / " in this article generally represents an "or" relationship between the front and back associated objects. In addition, "multiple" in this article means two or more than two.

[0023] Please refer to Figure 1 , Figure 1 It is a schematic framework diagram of an embodiment of the auxiliary diagnosis system 10 of the present application. Specifically, the auxiliary diagnosis system 10 may include an interface layer 11, a data layer 12, a capability layer 13, a service layer 14, and an application layer 15. Among them, the interface layer 11 includes several interfaces for respectively docking different medical systems of medical institutions; the data layer 12 is used to obtain the data retrieved by the interface layer to perform data docking with the medical system; the capability layer 13 includes several artificial intelligence algorithms for processing the data of the medical system; the service layer 14 includes several basic business modules for providing standard services for the application layer; the application layer 15 includes several medical business applications for respectively implementing different medical assistance functions, and the medical business applications are implemented by at least one basic business module.

[0024] In an implementation scenario, the auxiliary diagnosis system 10 can be docked with different medical systems of a medical institution through the unified interfaces provided by the interface layer 11. Exemplarily, the interface layer 11 can include a first interface (not shown) for docking with a Hospital Information System (HIS), the interface layer 11 can include a second interface (not shown) for docking with a Laboratory Information System (LIS), the interface layer 11 can further include a third interface (not shown) for docking with a Picture Archiving and Communication System (PACS), and the interface layer 11 can further include a fourth interface (not shown) for docking with a basic public health management system. Of course, the above examples are only a possible setting method in the actual application process, and do not thus limit the interfaces defined in the interface layer 11 and the various medical systems docked by the interface layer 11. For example, the interface layer 11 can further include, but is not limited to, different interfaces for docking with the following medical systems: an electronic medical record system, a regional national health information platform, etc., which are not limited herein. It should be noted that the interface layer 11 mainly realizes data intercommunication and sharing between systems and integrates the service capabilities of all parties to provide support for the overall auxiliary diagnosis system 10.

[0025] In an implementation scenario, please refer to Figure 2 , Figure 2 which is a schematic framework diagram of an embodiment of the auxiliary diagnosis deployment system of the present application. As Figure 2 shown, the auxiliary diagnosis deployment system can include: a client subsystem, a service subsystem, and an interface subsystem communicatively connected between the client subsystem and the service subsystem. The client subsystem is used to dock with the medical system of a medical institution, and the client subsystem can be the above-mentioned auxiliary diagnosis system 10. As Figure 2 shown, the medical systems of each medical institution such as medical institution A, medical institution B, and medical institution C can all be docked with the auxiliary diagnosis system 10 through a communication interface to provide auxiliary diagnosis services for each medical institution such as medical institution A, medical institution B, and medical institution C without affecting the existing medical systems of the medical institutions. Of course, Figure 2 the shown is only a possible example in the actual application process, and does not thus limit the docking method between the auxiliary diagnosis system 10 and the medical system. For example, a medical institution can have multiple medical systems, that is, the relationship between the medical institution and the medical system can be one-to-many, or a medical institution can have one medical system, that is, the relationship between the medical institution and the medical system can be one-to-one; or, multiple medical institutions share one medical system, that is, the relationship between the medical institution and the medical system can be many-to-one, which is not limited herein. Please continue to refer to Figure 2, the interface subsystem may include, but is not limited to, enterprise business bus connections, firewalls, private networks, or e-government networks, etc., to achieve communication connections between the customer subsystem and the service subsystem. Of course, Figure 2 The interface subsystem shown is only a possible implementation method in the actual application process, and does not thereby limit the interface network for realizing the communication connection between the customer subsystem and the service subsystem.

[0026] In a specific implementation scenario, the service subsystem in the auxiliary diagnosis deployment system can be deployed in the cloud. In addition, the service subsystem can also be differentiated according to standards such as administration, etc., and can include, for example, prefecture-level service subsystems, provincial-level service subsystems, etc., which are not limited here.

[0027] In a specific implementation scenario, the service subsystem can also include an artificial intelligence-assisted diagnosis platform, and the artificial intelligence-assisted diagnosis platform provides artificial intelligence algorithms.

[0028] In a specific implementation scenario, the service subsystem can also include a business application service platform, and the business application service platform provides medical business applications.

[0029] In a specific implementation scenario, the service subsystem can also include a basic capability service platform, and the basic capability service platform provides basic service algorithms such as speech recognition.

[0030] Therefore, through the above-mentioned auxiliary diagnosis deployment system, the construction of the auxiliary diagnosis deployment can be realized through the local + cloud deployment method. The above examples are only several possible implementation methods in the actual application process.

[0031] In an implementation scenario, the data layer 12 in the auxiliary diagnosis system 10 can specifically be used to provide the auxiliary diagnosis system 10 with capabilities such as data docking, storage, and calculation, and complete the data docking with external systems (such as medical systems) by means of real-time calling the interface layer 11 to synchronize data in full volume / increment.

[0032] In an implementation scenario, the capability layer 13 in the auxiliary diagnosis system 10 can provide unified core medical artificial intelligence capability services for the auxiliary diagnosis system 10. Specifically, it can include, but is not limited to, the following artificial intelligence algorithms: knowledge graph capability, natural language processing capability, speech recognition capability, big data processing capability, auxiliary diagnosis reasoning capability, etc. The artificial intelligence algorithms in the capability layer 13 are not specifically limited here.

[0033] In a specific implementation scenario, the artificial intelligence algorithm may include text structuring. Medical record texts are usually entered in free text form. Text structuring is used to perform structured parsing on medical record texts in order to extract useful information such as diseases, examinations, treatments, diagnoses, etc. Its specific steps may include entity extraction, attribute extraction, and relationship extraction. For example, for data that is relatively common and standardized in expression (such as time, examinations, etc.) and not easily enumerable (such as incentives, etc.), structured extraction can be performed using methods such as sequence labeling; or, for information with relatively divergent expressions and not easily enumerable possible expressions (such as symptoms, etc.), key information after regularization can be directly obtained using techniques similar to translation.

[0034] In a specific implementation scenario, the artificial intelligence algorithm may include knowledge graph construction. For example, for multi-source heterogeneous, unstructured or semi-structured data with complex representations, medical knowledge bases can be constructed through data acquisition, information mining, knowledge fusion, knowledge processing, etc. to obtain a knowledge graph. Among them, information mining can extract the constituent elements of the knowledge graph such as entities, relationships, and attributes.

[0035] In a specific implementation scenario, the artificial intelligence algorithm may also include speech recognition technology. For example, for medical record entry, medical speech recognition technology based on doctor-patient multi-accent adaptation and medical large language fusion models can be used to adapt to different accent differences, realize the collection of voice information such as oral chief complaints and medical histories, and generate result texts in combination with medical language models to improve the efficiency of medical record entry.

[0036] In a specific implementation scenario, the artificial intelligence algorithm may also include diagnostic reasoning. Specifically, diagnostic reasoning can be based on the structured medical record content, integrate deep learning and expert knowledge, understand the input electronic medical record, and output suspected diagnoses and probability results.

[0037] In a specific implementation scenario, the artificial intelligence algorithm may also include knowledge graph recommendation. Specifically, deep learning algorithms can be used to learn from a vast amount of medical professional knowledge and medical record information, draw on the advantages of traditional rule-based schemes, and combine with the medical knowledge graph to provide doctors with reasonable treatment plan suggestions after obtaining the patient's condition information.

[0038] In an implementation scenario, each basic business module (not shown) in the service layer 14 of the auxiliary diagnosis system 10 is obtained by modular encapsulation of basic services through object-oriented technology, so as to provide unified standard services for the upper-layer application layer 15. It should be noted that the basic business module can be implemented through at least one artificial intelligence algorithm, or can also be implemented only through business logic processes, which is not limited here. Exemplarily, for example, the speech recognition business module can be implemented through the artificial intelligence algorithm of the aforementioned speech recognition ability; or, for example, the outpatient call number business module can be implemented by combining business logic processes such as outpatient registration sorting and person-number matching verification. The above examples are only several possible examples in the actual application process, and do not limit the specific content of the basic business module accordingly.

[0039] In an implementation scenario, the application layer 15 in the auxiliary diagnosis system 10 is used to provide targeted medical service applications for medical workers in medical institutions. Exemplarily, the medical service applications can provide application functions related to medical record assistance, examination and inspection assistance, diagnosis assistance, treatment assistance, and medical knowledge assistance around the outpatient clinical SOAP (Subjective-Objective-Assessment-Plan, that is, subjective data, objective data, assessment, plan) diagnosis and treatment process. For specific details, please refer to the subsequent relevant descriptions, which will not be elaborated here for the time being.

[0040] In an implementation scenario, the auxiliary diagnosis system 10 can also be integrated with a standard and specification system (not shown). Exemplarily, the standard and specification system can include but is not limited to unified standards and the health industry specification system, so as to follow a mature, reliable, and stable technical route during the auxiliary diagnosis process and actively avoid risks during the construction process.

[0041] In an implementation scenario, the auxiliary diagnosis system 10 can also be integrated with security and service management (not shown). Exemplarily, the auxiliary diagnosis system 10 can specifically implement the requirements related to network security and information security, design a complete security and service management plan, and conduct information security risk management to ensure the safe and stable operation of the system.

[0042] In an implementation scenario, please refer to Figure 3 , Figure 3 which is a schematic framework diagram of another embodiment of the auxiliary diagnosis system 10 of the present application. As Figure 3As shown in the figure, the functional architecture of the auxiliary diagnosis system 10 may include, but is not limited to: medical record assistance, examination and test assistance, diagnosis assistance, treatment assistance, medical knowledge assistance, etc. Among them, medical record assistance may include, but is not limited to: assisted consultation, voice-assisted input, intelligent association-assisted input, medical record format quality control, medical record content quality control, etc.; examination and test assistance may include, but is not limited to: examination and test item recommendations, examination and test item quality control, examination and test item interpretation, etc.; diagnosis assistance may include, but is not limited to: auxiliary diagnosis of common and frequently-occurring diseases, recommended disease ranking, recommended disease grouping, diversified disease recommendations, assessment scale recommendations, early warnings for high-risk diseases, auxiliary diagnosis by combining multi-source data, auxiliary diagnosis of special diseases, dynamic disease knowledge graphs, visualization of the reasoning process, hierarchical diagnosis and treatment referral interconnection, etc.; treatment assistance may include, but is not limited to: treatment plan recommendations, treatment path recommendations, common medication recommendations, similar medical record recommendations, health education, prescription rationality review, drug knowledge recommendations, pre-prescription review intervention, etc.; medical knowledge assistance may include, but is not limited to: medical knowledge acquisition, medical knowledge construction, medical knowledge display, medical knowledge update, etc. Of course, Figure 3 The figure shown is merely a possible example of the functional architecture of the auxiliary diagnosis system 10, and does not thereby limit the specific functional architecture of the auxiliary diagnosis system 10. For ease of understanding, please refer to Table 1 in conjunction. Table 1 is a schematic table of an embodiment of the docking interface between the auxiliary diagnosis system 10 and the medical system. As shown in Table 1, based on the business application scenarios of the auxiliary diagnosis system 10, the interfaces can generally be divided into two categories, namely, basic interfaces and business data interfaces. For the former (i.e., basic interfaces), they are used to obtain the basic data of the hospital information system (such as, institutional information, personnel information, drug information, diagnosis information, etc.) at regular intervals for the auxiliary diagnosis system 10 to call; while for the latter (i.e., business data interfaces), they are used to push the business data generated in relevant systems (such as, LIS, PACS, etc.) to the auxiliary diagnosis system 10 in real time (or near real time) for the auxiliary diagnosis system 10 to call. The following combines Table 1 with Figure 3 Specifically illustrate the functional framework of the auxiliary diagnosis system 10.

[0043] Table 1 Schematic table of an embodiment of the docking interface between the auxiliary diagnosis system and the medical system

[0044]

[0045]

[0046]

[0047] In an implementation scenario, please refer to Figure 3, several medical business applications can at least include assisted medical history taking, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and obtain the basic patient information and current symptom manifestations obtained by the first interface docking with the medical information system, and make recommendations based on the basic patient information and current symptom manifestations to obtain recommended potential diseases and corresponding disease attributes, guiding doctors into the accurate medical history taking process. In addition, it can also be used to record the process and results of the medical history taking to help doctors comprehensively control the patient's condition and generate standardized electronic medical records. On this basis, it can also be used to feedback the aforementioned recommended potential diseases, corresponding disease attributes, electronic medical records, etc. to the hospital information system through the first interface, so that doctors can focus on communicating with patients for medical history taking by combining the auxiliary information provided by the auxiliary diagnosis system 10 without being distracted by recording, which helps to improve the efficiency of medical history taking with the assistance of the auxiliary diagnosis system 10.

[0048] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include voice-assisted medical record generation, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and process the doctor-patient voice conversation retrieved by the first interface based on the speech recognition algorithm and natural language understanding in the capability layer 13 to obtain the target medical record text. In addition, as a possible implementation method, the target medical record text can also be written back to the hospital information system through the first interface. Specifically, based on speech recognition and natural language understanding technologies, medical voices can be quickly and accurately recognized as text. For example, single-person mode speech transcription can be used, and then the doctor-patient conversation can be transcribed in real time, and keyword information can be quickly extracted to form a standardized medical record. Furthermore, through the data docking of the first interface, a standardized medical record can be directly formed without manual entry by medical staff, which helps to further improve the efficiency of medical history taking through the auxiliary diagnosis system 10.

[0049] In an implementation scenario, please refer to Figure 3, several medical business applications can at least include intelligent contact assistance input, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and obtain the historical behavior data and the current input content retrieved by the first interface, and recommend the next input content for the current input content based on the input recommendation algorithm in the capability layer 13 with reference to the historical behavior data. In addition, as a possible implementation, the next input content can also be transmitted back to the hospital information system through the first interface, which helps to further improve the input efficiency through the auxiliary diagnosis system 10. Exemplarily, the historical behavior data can include but is not limited to the historical record content of medical workers, such as historical medical records, etc. Taking the current input content as "The patient is obese and has nutrition", combined with the historical medical record, the next input content can be recommended for the current input content, such as "excess", "good", etc., and the above next input content is transmitted back to the hospital information system through the first interface, so that the medical worker can select the desired content from the above two next input contents. For example, if "good" is selected, the current medical record is automatically updated to "The patient is obese and has good nutrition", and other situations can be deduced by analogy, and no further examples will be given here.

[0050] In an implementation scenario, please refer to Figure 3 and Figure 4a , Figure 4a is a process schematic diagram of an embodiment of the diagnosis and treatment control method of this application. Several medical business applications can at least include medical record form quality control, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and perform form quality inspection on the medical record text to be inspected retrieved by the first interface based on natural language understanding in the capability layer 13, obtain the form quality inspection result of the medical record text to be inspected, and feedback the form quality control result to the hospital information system through the first interface to perform backwriting on the medical record text to be inspected based on the connotation quality control result, so as to perform in-process monitoring and real-time intervention on writing errors such as omissions and typos in the medical record text to be inspected. Specifically, as mentioned above, the auxiliary diagnosis system 10 can also include a standard and specification system, such as "Basic Norms for Medical Record Writing", "Regulations on the Application and Management of Electronic Medical Records (Trial)", and the medical record quality scoring standards of each province, etc., and perform form quality control on the medical record text to be inspected based on this. Exemplarily, for the situation where the patient's gender is not detected on the first page of the medical record text to be inspected, the form quality control result can include "The medical record first page lacks the patient's gender", and it is written back to the hospital information system through the first interface, such as a prompt can be given in the hospital information system, so that the medical worker can timely correct the form problem in the medical record text to be inspected. Other situations can be deduced by analogy, and no further examples will be given here.

[0051] In an implementation scenario, please refer to Figure 3 and Figure 4a, several medical business applications can at least include medical record content quality control, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and perform content quality inspection on the medical record text to be inspected retrieved by the first interface based on natural language understanding in the capability layer 13, so as to obtain the content quality inspection result of the medical record text to be inspected. In addition, as a possible implementation method, the content quality control result can also be fed back to the hospital information system through the first interface, so as to perform backwriting on the medical record text to be inspected based on the content quality control result, so as to perform in-process monitoring and real-time intervention on content problems such as logical defects and contradictions in the medical record text to be inspected. Specifically, as mentioned above, the auxiliary diagnosis system 10 can also include a standard and specification system, such as the "Basic Norms for Medical Record Writing", the "Regulations on the Application and Management of Electronic Medical Records (Trial)", and the medical record quality scoring standards of each province, etc., and perform content quality control on the medical record text to be inspected based on this. Exemplarily, for the content problem of self-contradiction that "the patient's gender is male" on the first page of the medical record text to be inspected and "the patient is pregnant" appears in the main body of the medical record text, or for the content problem of logical defect that "the patient's body temperature is 38 degrees and the diagnosis is pulmonary infection" appears in the main body of the medical record text, corresponding content quality control results can be obtained respectively, and written back to the hospital information system through the first interface, such as prompts can be made in the hospital information system, so that medical staff can timely correct the content problems in the medical record text to be inspected. Other situations can be deduced by analogy, and no more examples will be given here.

[0052] It should be noted that the above form quality control and content quality control are both parts of medical record quality control. Generally speaking, medical record quality control can combine logical reasoning with resource libraries such as medical record standard libraries, disease knowledge libraries, and massive medical record libraries, and use the idea of combining deep learning models and medical knowledge graphs centered on diagnosis to judge the rationality and standardization of the input electronic medical record content, and output the final quality control result. In terms of process steps, as Figure 4a shown, the medical record information of the patient's current visit (such as chief complaint, current medical history, past medical history, family history, physical examination, etc.) can be obtained through the interface. The auxiliary diagnosis system 10 can call the medical record quality control capability engine, comprehensively judge the rationality of the medical record content in combination with the patient's basic information, and remind of unreasonable content to assist doctors in standardizing the writing of electronic medical records.

[0053] In an implementation scenario, please refer to Figure 3, several medical business applications may at least further include test / examination item recommendations, which are used to connect to the hospital information system in different medical systems through the first interface in the interface layer 11, and make recommendations based on the basic information and condition information of the target patient to obtain several test / examination items, so as to improve the efficiency of medical consultations. Exemplarily, the hospital information system can be connected through the first interface in the interface layer 11, and the basic information and condition information of the target patient, such as "persistent high fever", "cough", "pulmonary rales", etc., can be obtained, and then the following test / examination items can be recommended: chest CT, etc. Other situations can be inferred by analogy and will not be exemplified one by one here.

[0054] In an implementation scenario, please refer to Figure 3 , several medical business applications may at least include quality control of test / examination items, which are used to connect to the hospital information system in different medical systems through the first interface in the interface layer 11, and perform quality control on the test / examination items retrieved by the first interface based on the basic information and condition information of the target patient to obtain a quality control result that at least characterizes whether the test / examination items are reasonable, so as to further improve the rationality of recommending test / examination items to patients. Exemplarily, the basic information and condition information of the target patient, such as "persistent high fever", "cough", "pulmonary rales", etc., can be obtained through the interface layer 11, and quality control can be performed on the test / examination items retrieved by the first interface: chest CT, blood routine, urine routine, cranial MR, etc., to obtain a quality control result, such as the test / examination item "cranial MR" is unreasonable. Other situations can be inferred by analogy and will not be exemplified one by one here. Of course, in addition to whether the test / examination items are reasonable, the quality control result can also include an interpretability description. Still taking the previous example, it can include the interpretability description "According to the basic information and condition information of the patient, the patient's illness is probably not related to the cranium, so cranial MR is not recommended for the time being". Other situations can be inferred by analogy and will not be exemplified one by one here.

[0055] In an implementation scenario, please refer to Figure 3 , several medical business applications may at least further include test report interpretation, which is used to connect to the laboratory information system in different medical systems through the second interface in the interface layer 11, and interpret each index in the test report retrieved by the second interface to obtain a first interpretation result of the test report, so as to improve the convenience of test report interpretation. Exemplarily, the test report "blood routine report" can be obtained from the laboratory information system through the second interface in the interface layer 11 or the test report "blood routine report" can be obtained through optical character recognition (i.e., OCR), and each index in it (such as red blood cell count, hemoglobin, white blood cells, white blood cell differential count, platelets, etc.) can be interpreted to obtain the first interpretation result of the "blood routine report". Other situations can be inferred by analogy and will not be exemplified one by one here.

[0056] In one implementation scenario, please refer to Figure 3 , several medical service applications can at least include inspection report interpretation, which is used to dock with the picture archiving and communication system in different medical systems through the third interface in the interface layer 11, and interpret each index in the inspection report retrieved by the third interface to obtain the second interpretation result of the inspection report, so as to improve the convenience of inspection report interpretation. Exemplarily, the inspection report "lung CT report" can be obtained from the picture archiving and communication system through the third interface in the interface layer 11, or the inspection report "lung CT report" can be obtained through optical character recognition (i.e., OCR), and each index therein (such as lung markings, fibrous streaks, nodules, calcified foci, effusion, etc.) is interpreted to obtain the second interpretation result of the "lung CT report". Other situations can be deduced by analogy and will not be elaborated one by one here.

[0057] It should be noted that in addition to obtaining medical report data through the above-mentioned second interface, third interface and other interfaces, for the above-mentioned report interpretation applications such as test report interpretation and inspection report interpretation, medical report images can also be recognized through optical character recognition to obtain medical report data.

[0058] In one implementation scenario, please refer to Figure 3 , several medical service applications can at least include common disease assistant diagnosis, which is used to dock with the hospital information system through the first interface in the interface layer 11, so as to obtain the patient's basic information and condition information retrieved by the first interface from the hospital information system, and call the first engine in the capability layer to predict the basic information and condition information to obtain several recommended diseases belonging to common diseases. It should be noted that the first engine is a calculation engine dedicated to common disease recommendation, which mainly targets common diseases and frequently-occurring diseases at the grass-roots level, so as to intelligently prompt suspected diseases and detailed symptoms, signs and other information according to the patient's basic information and condition information, and give the probability of suspected diseases and the basis for judgment. Exemplarily, the first engine can be used to predict basic information and condition information, such as "continuous high fever" and "cough", to obtain several recommended diseases belonging to common diseases: common cold, influenza, etc. Of course, the above examples are only one possible example in the actual application process and do not limit other possible situations accordingly.

[0059] In one implementation scenario, please refer to Figure 3, several medical business applications can at least include recommended disease ranking, which is used to rank the recommended results of suspected diseases according to the likelihood of corresponding diseases. For example, factors such as the current season and the patient's symptoms can be combined to rank the suspected diseases. Taking the aforementioned basic information and condition information of "continuous high temperature" and "cough" as an example, several recommended diseases can include: common cold, influenza, etc. Then, considering that the current season is winter or spring, the recommended disease "influenza" can be placed before the recommended disease "common cold". Of course, the above example is only one possible example in the actual application process and does not limit other possible situations.

[0060] In one implementation scenario, please refer to Figure 3 , several medical business applications can at least include recommended disease grouping, which is used to group the recommended diseases according to departments or systems. Taking the aforementioned basic information and condition information of "continuous high temperature" and "cough" as an example, several recommended diseases can include: common cold, influenza, etc. Then, the above recommended diseases can be grouped according to departments or systems: Department of Respiratory Medicine. Of course, the above example is only one possible example in the actual application process and does not limit other possible situations.

[0061] In one implementation scenario, please refer to Figure 3 , several medical business applications can at least include diverse disease recommendations, which are used to make multi-dimensional diagnostic recommendations for basic information and condition information, especially when there is less medical record information for symptom-based diseases (such as abdominal pain, diarrhea), such as common diseases, critical diseases, infectious diseases, etc. Taking the aforementioned basic information and condition information of "continuous high temperature" and "cough" as an example, several recommended diseases belonging to common diseases can be predicted: common cold, influenza, etc., several recommended diseases belonging to critical diseases can be predicted: Mycoplasma pneumoniae infection of the lungs, etc., and several recommended diseases belonging to infectious diseases can be predicted: Mycoplasma pneumoniae infection of the lungs, influenza, etc. Of course, the above example is only one possible example in the actual application process and does not limit other possible situations.

[0062] In one implementation scenario, please refer to Figure 3 , several medical business applications can at least include assessment scale recommendations, which are used to recommend assessment scales for suspected diseases. Exemplarily, taking the aforementioned basic information and condition information of "continuous high temperature" and "cough" as an example, several recommended diseases can be predicted: common cold, influenza, and assessment scales can be provided for the above two recommended diseases respectively, and the assessment scales can be fed back to the hospital information system through the first interface so that medical staff can further evaluate the patient according to the assessment scales.

[0063] In one implementation scenario, please refer to Figure 3, several medical service applications can at least include disease risk warning, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, predict the basic information and disease condition information obtained by the first interface, obtain several recommended diseases, identify the risk levels of each recommended disease, and give real-time warnings for the recommended diseases whose risk levels meet the preset conditions. Exemplarily, still taking the aforementioned basic information and disease condition information of "continuous high temperature" and "cough" as an example, several recommended diseases can be predicted: common cold, influenza, and mycoplasma pneumoniae infection of the lungs. On this basis, the risk levels of the above-mentioned recommended diseases can be identified. For example, the risk level can be marked with a value from 0 to 1. The higher the risk level, the larger the value, and vice versa, the lower the risk level, the smaller the value. And corresponding thresholds can be set, such as 0.6, 0.7, etc. The preset conditions can specifically include that the risk level is greater than the above-mentioned preset threshold. For example, if the risk level of "mycoplasma pneumoniae infection of the lungs" among the above-mentioned recommended diseases is greater than the preset threshold, then a real-time warning can be given for the recommended disease "mycoplasma pneumoniae infection of the lungs". In addition, the auxiliary diagnosis system 10 can also feedback the above-mentioned real-time warning to the hospital information system through the first interface so that medical workers can intervene in time.

[0064] In a specific implementation scenario, as described above, several artificial intelligence algorithms include a first engine for predicting common diseases, and several medical service applications can at least include auxiliary diagnosis of common diseases, which is used to call the first engine to predict the basic information and disease condition information, and obtain several recommended diseases belonging to common diseases.

[0065] In another specific implementation scenario, please refer to Figure 3 , in order to expand the applicable scope of the auxiliary diagnosis system 10, several medical service applications can also include auxiliary diagnosis of special diseases, which is used to predict the basic information and disease condition information of the target patient, and obtain several recommended diseases belonging to special diseases. In this case, as a possible implementation manner, several artificial intelligence algorithms can include, in addition to the first engine for predicting common diseases, a second engine for predicting special diseases, and the first engine and the second engine work independently. Several medical service applications can at least include auxiliary diagnosis of special diseases, which is used to call the second engine to predict the basic information and disease condition information, and obtain several recommended diseases belonging to special diseases. For the convenience of distinction, the aforementioned auxiliary diagnosis of common diseases can be called the first disease recommendation, and the auxiliary diagnosis of special diseases can be called the second disease recommendation. It should be noted that special diseases can include but are not limited to: infectious diseases, endemic diseases, occupational diseases, rare diseases, etc., and are not limited here. In addition, similar to the aforementioned medical service application "disease risk warning", after obtaining several recommended diseases belonging to special diseases, real-time warnings can also be given for the recommended diseases belonging to special diseases and fed back to the hospital information system through the first interface.

[0066] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include multi-source data combined auxiliary diagnosis, which is used to dock the hospital information system in different medical systems through the first interface in the interface layer 11 to obtain historical medical records from the first interface, and dock the laboratory information system in different medical systems through the second interface in the interface layer 11 to obtain inspection reports from the second interface, dock the picture archiving and communication system in different medical systems through the third interface in the interface layer 11 to obtain examination reports from the third interface, and dock the basic public health management system in different medical systems through the fourth interface in the interface layer 11 to obtain health records. Then, multi-source data including historical medical records, inspection reports, examination reports, and health records are predicted to obtain recommended diseases, so as to recommend diseases comprehensively in multiple dimensions, which helps to improve the accuracy of disease recommendation. Exemplarily, still taking the aforementioned basic information and condition information of "persistent high temperature" and "cough" as an example, docking the hospital information system through the first interface can obtain historical medical records. For example, the historical medical records record the patient's recent visit records in the Department of Respiratory Medicine. Docking the laboratory information system through the second interface can obtain inspection reports. For example, the inspection report shows that the white blood cell index is on the high side. Docking the picture archiving and communication system through the third interface can obtain CT images. For example, shadows can be observed from the CT images. And docking the basic public health management system through the fourth interface can obtain health records. Combining the multi-source data of the above historical medical records, inspection reports, examination reports, and health records can be predicted to obtain recommended diseases. Of course, the above examples are only one possible example in the actual application process and do not limit other possible situations.

[0067] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include a dynamic disease knowledge graph, which is used to build the dynamic relationship between the recommended disease and clinical manifestations, pathogenic factors, inspection and examination, treatment drugs, etc. by calling the knowledge graph construction ability with the recommended disease as the center, so as to obtain a knowledge graph centered on the recommended disease, and feedback the knowledge graph to the hospital information system through the first interface for medical staff to view in time, so as to obtain efficient, intuitive and accurate diagnosis and treatment knowledge.

[0068] In an implementation scenario, please refer to Figure 3, several medical business applications can at least include inference visualization, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and infer the reference data retrieved by the first interface to obtain the inference paths of several recommended diseases, and the medical basis can be attached to the inference paths. It should be noted that the reference data can be data related to the auxiliary diagnosis and treatment inference process, such as but not limited to: medical record data, test data, examination data, etc., which are not limited here. In addition, as a possible implementation method, the reference data can be inferred step by step. In this case, the medical basis can be attached to each step of the inference in the inference path. Of course, after obtaining the inference path, the inference path can also be fed back to the hospital information system through the first interface, so that medical workers can learn the machine inference process and improve the interpretability of disease recommendation. Exemplarily, the starting point of the inference path can include many suspected diseases, and each step of the inference is attached with a medical basis (such as patient symptoms, past history, etc., and medical guidelines) to exclude some suspected diseases. Thus, with each step of the inference, the number of suspected diseases can be reduced until the last step of recommendation, when the suspected diseases will be limited to a few. Of course, the above example is only a possible manifestation of the inference path and does not limit other expressions of the inference path accordingly.

[0069] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include hierarchical referral interconnection, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and identify the diagnosed diseases retrieved by the first interface to obtain the identification results of the diagnosed diseases, and the identification results include whether the diagnosed diseases exceed the service scope of the medical institution, and in response to the identification result including that the diagnosed diseases exceed the service scope of the medical institution, initiate a referral recommendation and / or dock with the referral system in real time through the fifth interface in the interface layer 11. Exemplarily, if the medical institution is a primary clinic and a level 4 operation is required to treat the recommended disease, the identification result can include exceeding the service scope of the medical institution, and then it can dock with the referral system in real time through the fifth interface to transfer the patient for treatment in time and minimize the risk of delayed treatment.

[0070] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include differential diagnosis, which is used to intelligently recommend relevant easily confused differential diagnoses for the recommended diseases, and at the same time provide the key points of diagnostic differentiation, including but not limited to information such as symptoms, signs, tests, and examinations. Exemplarily, if the recommended disease is disease A and disease B is easily confused with disease A, then when recommending disease A, the key points of diagnostic differentiation for recommending disease A instead of disease B can also be provided to further improve the credibility of disease 6 recommendation.

[0071] In an implementation scenario, please refer toFigure 3 Among them, several medical business applications can at least include diagnostic quality inspection, which is used to mine the correlation between patient medical record information and doctor diagnoses, and conduct a reasonable quality inspection on doctor diagnoses to provide decision-making support for doctors. Specifically, the assistant diagnosis system 10 can call the diagnostic quality inspection capability engine in the capability layer 13, comprehensively judge whether the current diagnosis and treatment are reasonable in combination with patient diagnosis and treatment information, remind of unreasonable diagnosis and treatment information, and assist medical staff in making clinical decisions.

[0072] In an implementation scenario, please refer to Figure 3 Among them, several medical business applications can at least include clinical treatment plan recommendation, which is used to dock with the hospital information system through the first interface in the interface layer 11 to obtain patient basic information, condition information, and doctor diagnoses, and make recommendations based on the patient basic information, condition information, and patient diagnoses to obtain treatment suggestions and treatment plan suggestions, and can feedback the treatment suggestions and treatment plan suggestions to the hospital information system through the first interface for medical staff to refer to. Thus, medical staff can provide medical services to patients in combination with the treatment suggestions and treatment plan suggestions provided by the assistant diagnosis system 10.

[0073] In an implementation scenario, please refer to Figure 3 Among them, several medical business applications can at least include common medication recommendation, which is used to dock with the hospital information system through the first interface in the interface layer 11 to obtain patient basic information, condition information, and doctor diagnoses, and make recommendations based on the patient basic information, condition information, and patient diagnoses to obtain medication suggestions, and can feedback the medication suggestions to the hospital information system through the first interface for medical staff to refer to. Thus, medical staff can provide medical services to patients in combination with the medication suggestions provided by the assistant diagnosis system 10.

[0074] In an implementation scenario, please refer to Figure 3 Among them, several medical business applications can at least include health education knowledge recommendation, which is used to dock with the hospital information system through the first interface in the interface layer 11 to obtain patient basic information, condition information, and doctor diagnoses, and make recommendations based on the patient basic information, condition information, and patient diagnoses to obtain health education knowledge, and can feedback the health education knowledge to the hospital information system through the first interface for medical staff to refer to. Thus, medical staff can provide medical services to patients in combination with the health education knowledge provided by the assistant diagnosis system 10.

[0075] In an implementation scenario, please refer to Figure 3, several medical service applications may at least include drug knowledge recommendation, which is used to connect to the hospital information system through the first interface in the interface layer 11 to obtain a drug inquiry request, parse the drug inquiry request to obtain the target drug to be inquired and the target entry of the target drug, and feedback the specific content of the target entry to the hospital information system through the first interface. Exemplarily, by parsing the drug inquiry request, it can be obtained that the target drug to be inquired is drug A, and the target entry of the target drug is the indications and usage dosage of drug A. Then, the indications and usage dosage of drug A can be fed back to the hospital information system through the first interface.

[0076] In an implementation scenario, please refer to Figure 3 , several medical service applications may at least include similar medical record recommendation, which is used to connect to the hospital information system through the first interface in the interface layer 11 to obtain the current medical record, search for historical medical records similar to the current medical record in the medical record library based on the current medical record, and feedback the historical medical records similar to the current medical record to the hospital information system through the first interface for medical staff to learn and draw on the diagnosis and treatment ideas from the historical medical records similar to the current medical record.

[0077] In an implementation scenario, please refer to Figure 3 , several medical service applications may at least include diagnosis and treatment path recommendation, which is used to connect to the hospital information system in different medical systems through the first interface in the interface layer 11, and recommend a diagnosis and treatment path for the diagnosed disease obtained by the first interface to obtain the diagnosis and treatment path of the diagnosed disease. In addition, as a possible implementation method, the diagnosis and treatment path may include medical treatments at each stage during the diagnosis and treatment process of the diagnosed disease. Of course, after obtaining the diagnosis and treatment path of the diagnosed disease, the diagnosis and treatment path of the diagnosed disease can also be fed back to the hospital information system through the first interface for medical staff to refer to and standardize the diagnosis and treatment behavior.

[0078] In an implementation scenario, please refer to Figure 3 , several medical service applications may at least include prescription rationality review, which is used to connect to the hospital information system through the first interface in the interface layer 11 and conduct a rationality review on the medical prescription obtained by the first interface to obtain a review result. It should be noted that the review result may include risks such as drug interactions, population taboos, and incompatibilities in the medical prescription. In addition, after obtaining the review result, the assistant diagnosis system 10 can also feed back the review result to the hospital information system through the first interface for medical staff to determine whether to adjust the medical prescription according to the review result. Specifically, the rational drug use review engine in the capability layer 13 can be called to comprehensively judge whether the current prescription is reasonable in combination with data such as the patient's basic information, medical record information, and historical diagnosis and treatment information, so as to remind of unreasonable information and assist doctors in clinical decision-making.

[0079] In one implementation scenario, please refer to Figure 3 , several medical business applications may at least include prescription review intervention, which is used to dock with the hospital information system in different medical systems through the first interface in the interface layer 11, and pre-place the medical prescriptions obtained by the first interface in advance for pharmacists to review. Before outputting the medical prescriptions to the pharmacists, the review results of the medical prescriptions are sent back to the hospital information system through the first interface for doctors to intervene in unreasonable prescriptions. That is to say, the auxiliary diagnosis system 10 can call the first interface to obtain medical prescriptions, and before submitting them to the pharmacists, first review the medical prescriptions to obtain the review results of the medical prescriptions, and then feedback the review results to the hospital information system through the first interface for medical staff to determine whether to output or modify the medical prescriptions according to the review results. Exemplarily, when the review results include that the medical prescriptions are reviewed correctly, the medical prescriptions can be output to prescribe medicine for the patients accordingly, or when the review results include that the medical prescriptions are reviewed incorrectly, the medical prescriptions can be adjusted, so that the prescription review link can be pre-placed to reduce the occurrence of unreasonable prescriptions from the source.

[0080] In one implementation scenario, please refer to Figure 3 , several medical business applications may at least include medical knowledge acquisition, which is used to dock with the medical guidance department through other interfaces in the interface layer 11 except the above interfaces to obtain authoritative medical knowledge, including but not limited to: laws and regulations, departmental rules, regulatory documents, authoritative drug instructions, medical device registration certificates, clinical pathways, clinical practice guidelines, technical operation specifications, standards, medical textbooks, expert consensus, monographs, literature, etc.

[0081] In one implementation scenario, please refer to Figure 3, several medical business applications can at least include knowledge base construction, which is used to call capabilities such as knowledge graph construction in the capability layer 13 to construct a knowledge base for medical knowledge, so as to construct, including but not limited to: disease knowledge base, drug knowledge base, inspection and examination knowledge base, clinical pathway knowledge base, typical disease knowledge base, literature knowledge base, etc. The knowledge base is not limited here. It should be noted that the disease knowledge base can include but not limited to: knowledge such as disease overview, epidemiology, etiology and classification, pathogenesis, pathology, clinical manifestations, laboratory tests, diagnosis and differential diagnosis, treatment, prevention, etc.; the drug knowledge base can include but not limited to: drug instruction information, drug ingredients, properties, chemical structural formulas, indications, specifications, usage and dosage, adverse reactions, taboos, precautions, drug interactions, etc.; the inspection and examination knowledge base can include but not limited to: knowledge such as inspection and examination item overview, precautions, clinical significance, etc.; the typical disease knowledge base can include but not limited to: knowledge of typical cases of diseases; the clinical pathway knowledge base can include but not limited to: clinical pathway knowledge of relevant disease types issued by the medical guidance department; the literature knowledge base can include but not limited to: knowledge such as medical textbooks, expert consensus, monographs, literature, etc.

[0082] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include medical knowledge retrieval, which is used to connect to the hospital information system through the interface layer 11, and obtain a knowledge retrieval request through the first interface, so as to parse the knowledge retrieval request, determine the target knowledge expected to be obtained, and based on this, retrieve in the aforementioned knowledge base, such as medical textbooks, expert consensus, monographs, literature, etc., to obtain a retrieval result. Of course, on this basis, the retrieval result can be fed back to the hospital information system through the first interface to meet the needs of users for learning, research, and retrieval.

[0083] In an implementation scenario, please refer to Figure 3 , several medical business applications can at least include dynamic knowledge recommendation, which is used to connect to the hospital information system in different medical systems through the first interface in the interface layer 11, and predict the user portrait and user behavior retrieved by the first interface to obtain recommended knowledge. Exemplarily, when the user portrait includes but not limited to the following portrait tags: Department of Respiratory Medicine, bronchus, recommended knowledge such as the latest released guidelines and research reports in the field of bronchus can be recommended to the user. Other situations can be inferred by analogy and will not be exemplified one by one here.

[0084] In an implementation scenario, please refer to Figure 3, several medical business applications can at least include medical knowledge update, which is used to, in response to the knowledge base not being updated for a preset duration, call capabilities such as knowledge graph construction in the capability layer 13 to reconstruct the knowledge base with the latest medical knowledge, so as to timely delete or update inapplicable medical knowledge, that is, endow the knowledge base with an exit mechanism. In addition, the preset duration can be set according to the actual application situation, such as it can be set to: one quarter, half a year, etc., which is not limited here.

[0085] In an implementation scenario, to make the auxiliary diagnosis system 10 comply with information security, the auxiliary diagnosis system 10 can be specifically built in accordance with relevant regulations, such as the relevant requirements of the "Cybersecurity Law", the "Data Security Law", and the "Personal Information Protection Law", etc., so that the auxiliary diagnosis system 10 can implement the network security level protection system and related standards to ensure network security. In addition, as mentioned above, the auxiliary diagnosis system 10 can be deployed in the controllable internal network of the medical institution. In this case, the auxiliary diagnosis system 10 can be isolated from the Internet to make its security control level reach the same security level as the internal information system of the medical institution; or, as mentioned above, the auxiliary diagnosis system 10 can also be deployed outside the internal network of the medical institution. In this case, through management and technical control measures such as approval, authorization, and personal information hiding, the safety of patients and medical information can be ensured.

[0086] In an implementation scenario, the auxiliary diagnosis system 10 can also support the compatibility of various related hardware, operating systems, databases, middleware, related support software, and other related application systems.

[0087] In an implementation scenario, the auxiliary diagnosis system 10 can support the adaptation of hardware servers, operating systems, database management systems, middleware, client operating systems, browsers, virtualization / hyperscale platforms of specific manufacturers (such as domestic manufacturers).

[0088] In an implementation scenario, the auxiliary diagnosis system 10 can provide relevant interfaces so that users such as managers can check the interfaces (or through the check menu) to implement obvious functions. In addition, it also helps that various problems, messages, and results executed by the software product should be easy to understand, and it also helps that the screen input format, reports, and other outputs of the software product should be clear and easy for users to understand.

[0089] In an implementation scenario, in addition to providing relevant interfaces, the auxiliary diagnosis system 10 can also provide relevant means such as help functions and user documents for users such as administrators to learn how to use relevant functions.

[0090] In an implementation scenario, the interface composition style of the auxiliary diagnosis system 10 should be consistent; users should easily understand the messages of the software system; users should be able to easily fix their errors or re-execute tasks during use; the execution of functions with serious consequences should be reversible, or an obvious warning of the consequences should be given, and confirmation should be required before the execution of such commands.

[0091] In an implementation scenario, to make the auxiliary diagnosis system 10 meet the online requirements and improve the user experience as much as possible, the performance requirements of the auxiliary diagnosis system 10 can also be pre-set, and operations such as testing and optimization of the auxiliary diagnosis system 10 can be carried out according to the performance requirements before going online until the test passes. It should be noted that the performance requirements can include, but are not limited to: business requirements, system resource requirements, database requirements, and stability requirements.

[0092] In a specific implementation scenario, the business requirements can include transaction response requirements. For example, the average response time of a single business interface can be set to less than 1 second, and the maximum response time for batch transactions and complex reports can be set to less than 3 minutes; or the business requirements can also include the core processing ability of the system. For example, in the core artificial intelligence ability scenario (such as auxiliary consultation, auxiliary diagnosis, medical record quality control, rational drug use, etc.), the number of transactions processed per second (TPS) of the system should be no less than 200 transactions per second; or, the business requirements can also include the number of online users. For example, the concurrency of the system running on a single server should be no less than 1000 users; or, the business requirements can also include the transaction success rate. For example, under the load condition of the system, the transaction success rate should be no less than 99.4%.

[0093] In a specific implementation scenario, the system resource requirements can include the average utilization rate. For example, the average utilization rate should be lower than 75%, and corresponding expansion should be carried out when it exceeds this range; or, the system resource requirements can include the average occupancy rate. For example, the average occupancy rate should be lower than 70%; or, the system resource requirements can include the occupancy time. For example, the server disk occupancy time (IOWAIT) should be lower than 10%.

[0094] In a specific implementation scenario, the database requirements may include high availability. For example, the database should have the ability of automatic failover and can quickly resume the database service within 2 minutes, reducing the downtime of business services. In addition, after any database slave node fails, it should have no impact on the business, and the database cluster should take the failed node offline within 10 seconds. Or, the database requirements may include backup availability. For example, the database should perform automatic backup and automatic verification of the backup. In addition, the database should be able to restore at any time point within the backup period to prevent data loss. Or, the database requirements may also include monitoring integrity. For example, database anomalies should be obtained in a timely manner and alarms should be triggered in a timely manner to reduce the downtime of system failures. In addition, complete database operation metric data should be provided, and through the accumulation of performance metrics over a certain period of time, the overall operation of the database can be comprehensively understood, providing reference data basis for database / business system optimization and sufficient data metric basis for system resource management.

[0095] In a specific implementation scenario, the stability requirements may specifically include ensuring the stable operation of the business system 7*24 hours (except for system updates and upgrades).

[0096] In an implementation scenario, the auxiliary diagnosis system 10 can collect and analyze a large amount of clinical data in real time, and then seamlessly connect with the business scenarios of the HIS system. The auxiliary diagnosis system 10 can be integrated with the HIS system to uniformly access, analyze, process, and text structure the multi-source data stored by patients in different systems. Then, combined with key technologies such as medical knowledge graphs, medical record quality control, diagnostic reasoning, and rational drug use, finally, through the human-computer interaction interface, decision-making support is provided for doctors. According to the depth of the integration between the auxiliary diagnosis system and the HIS business, the interface docking mode between systems can be divided into two modes: basic data docking and deep interactive integration.

[0097] In a specific implementation scenario, the auxiliary diagnosis system 10 can use the plug-in mode to perform basic data docking with the HIS system without any impact on the HIS business process. After the HIS system completes a complete outpatient visit activity (writing medical records, making diagnoses, prescribing prescriptions, etc.), it calls an interface once through an independent interface call button to transmit the patient's basic information, medical record information, prescription information, etc. to the auxiliary diagnosis system 10 uniformly, so as to perform business processes such as medical record quality control, auxiliary diagnosis, and rational drug use review. After the auxiliary diagnosis system 10 finishes processing, it transmits the results to the HIS to assist doctors in making decisions.

[0098] In a specific implementation scenario, the auxiliary diagnosis system 10 can be deeply integrated with the HIS system. Based on the docking of basic data, deep interactive docking of relevant services can be carried out to achieve real-time interaction between the auxiliary diagnosis system 10 and the HIS system. For example, during the process of writing a medical record, real-time quality control is performed on the content that has been written, and the rationality of drug selection is reviewed when prescribing drugs. Specific interactive scenarios include but are not limited to real-time quality control of medical records, writing back of medical record data, citation of recommended diagnoses, citation of recommended drugs and examination and inspection items, and pre-prescription review intervention, etc.

[0099] In the above solution, the auxiliary diagnosis system includes an interface layer, a data layer, a capability layer, a service layer, an application layer, and an application layer. The interface layer includes several interfaces for respectively docking different medical systems in a medical institution. The data layer is used to obtain the data retrieved by the interface layer to perform data docking with the medical system. The capability layer includes several artificial intelligence algorithms for processing the data of the medical system. The service layer includes several basic business modules for providing standard services for the application layer. The application layer includes several medical business applications for respectively implementing different medical assistance functions, and the dependent business applications are implemented by at least one basic business module. Therefore, it is directly docked to the medical system through the interface layer, and the data retrieved by the interface layer is obtained by the data layer in the auxiliary diagnosis system to achieve data docking with the medical system, and the data is processed by each artificial intelligence algorithm in the capability layer and encapsulated into several basic business modules in the service layer. Thus, in the application layer, at least one basic business module forms a medical business application, so there is no need to further integrate on the existing medical system. On the one hand, it can minimize the impact on the existing medical system as much as possible, neither making the existing medical system more bloated nor damaging the normal use of the existing medical system. On the other hand, since the auxiliary diagnosis system and the medical system are separated from each other, even if either party is updated, the impact on the other party is very small. Therefore, it is possible to improve the convenience of operation and maintenance on the premise of meeting auxiliary diagnosis.

[0100] Please refer to Figure 4a and Figure 4b , Figure 4b is a schematic flowchart of an embodiment of the diagnosis and treatment control method of the present application. It should be noted that the diagnosis and treatment control system is applied to the auxiliary diagnosis system in the above-mentioned disclosed embodiment. For specific details, reference can be made to the relevant descriptions in the foregoing disclosed embodiment, which will not be elaborated here. The specific embodiments of the present disclosure may include the following steps:

[0101] Step S41: Invoke the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system.

[0102] Specifically, the first interface in the interface layer can be called to obtain the original medical record from the hospital information system. For the specific meanings of the interface layer and the relevant interfaces in the interface layer, reference can be made to the relevant descriptions in the foregoing disclosed embodiments, which will not be elaborated herein. In addition, the original case can be manually input by a doctor, or can be obtained by calling the speech recognition ability in the ability layer of the auxiliary diagnosis system to recognize the doctor-patient conversation, and then organizing and analyzing it. For the specific details, reference can be made to the relevant descriptions in the foregoing disclosed embodiments, which will not be elaborated herein either.

[0103] Step S42: Improve the original medical record based on the quality inspection result of the original medical record to obtain the target medical record, call the interface layer to write back the target medical record to the hospital information system, and perform a diagnosis recommendation based on the target medical record to obtain a recommendation result, so as to call the interface layer to feedback the recommendation result to the hospital information system for the doctor's reference.

[0104] Specifically, the quality inspection of the original medical record can include format quality inspection, content quality inspection, etc., which are not limited herein. For the specific processes of format quality inspection and content quality inspection, reference can be made to the relevant descriptions in the foregoing disclosed embodiments, which will not be elaborated herein. In addition, the recommendation result can specifically include but is not limited to: suspected diagnosis, recommended medication, etc., which are not limited herein. For the specific process of diagnosis recommendation, reference can be made to the foregoing disclosed embodiments, which will not be elaborated herein either.

[0105] Step S43: Call the interface layer to obtain the doctor's diagnosis from the hospital information system.

[0106] Specifically, the first interface in the interface layer can be called to obtain the doctor's diagnosis from the hospital information system. As mentioned above, for the specific meanings of the interface layer and the relevant interfaces in the interface layer, reference can be made to the relevant descriptions in the foregoing disclosed embodiments, which will not be elaborated herein.

[0107] Step S44: Perform a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable.

[0108] In the embodiments of the present disclosure, when the first detection result indicates that the doctor's diagnosis is unreasonable, the interface layer is called to feedback the first detection result to the hospital information system to remind the doctor. It should be noted that for the specific process of diagnosing reasonableness detection, reference can be made to the relevant descriptions in the foregoing disclosed embodiments, which will not be elaborated herein.

[0109] Step S45: Call the interface layer to obtain the doctor's prescription from the hospital information system.

[0110] Specifically, the first interface in the interface layer can be called to obtain the doctor's prescription from the hospital information system. As mentioned above, for the specific meanings of the interface layer and the relevant interfaces in the interface layer, reference can be made to the relevant descriptions in the foregoing disclosed embodiments, which will not be elaborated herein.

[0111] Step S46: Perform a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable.

[0112] In the embodiments of the present disclosure, when the second detection result indicates that the doctor's prescription is unreasonable, the interface layer is called to feedback the second detection result to the hospital information system to remind the doctor. It should be noted that the specific process of prescription rationality detection can refer to the relevant descriptions in the foregoing disclosed embodiments and will not be elaborated herein.

[0113] In the above solution, the interface layer in the auxiliary diagnosis system is called to obtain the original medical record from the hospital information system, and then the original medical record is improved based on the quality inspection result of the original medical record to obtain the target medical record. The interface layer is called to write back the target medical record to the hospital information system, and a diagnosis recommendation is made based on the target medical record to obtain a recommendation result. The interface layer is called to feedback the recommendation result to the hospital information system for the doctor's reference. On this basis, the interface layer is called to obtain the doctor's diagnosis from the hospital information system, and a first detection is performed based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable. When the first detection result indicates that the doctor's diagnosis is unreasonable, the interface layer is called to feedback the first detection result to the hospital information system. Then the interface layer is called to obtain the doctor's prescription from the hospital information system, and a second detection is performed based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable. When the second detection result indicates that the doctor's prescription is unreasonable, the interface layer is called to feedback the second detection result to the hospital information system. Therefore, by docking with the auxiliary diagnosis system, it can assist doctors in various links such as medical record entry, diagnosis issuance, and prescription writing. On the one hand, it can minimize the impact on the existing medical system as much as possible, neither making the existing medical system more bloated nor damaging the normal use of the existing medical system. On the other hand, since the auxiliary diagnosis system and the medical system are separated from each other, even if either party is updated, the impact on the other party is very small. Therefore, it can improve the operation and maintenance convenience on the premise of meeting auxiliary diagnosis.

[0114] Please refer to Figure 5 , Figure 5It is a schematic framework diagram of an embodiment of the diagnosis and treatment control device 50 of the present application. The diagnosis and treatment control device 50 is applied to the auxiliary diagnosis system in the above-mentioned disclosed embodiment. For specific details, reference can be made to the relevant descriptions in the foregoing disclosed embodiment, which will not be elaborated here. The diagnosis and treatment control device 50 may include: a first invocation module 51, an information interaction module 52, a second invocation module 53, a first detection module 54, a third invocation module 55, and a second detection module 56. The first invocation module 51 is configured to invoke the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system. The information interaction module 52 is configured to improve the original medical record based on the quality inspection result of the original medical record to obtain a target medical record, and invoke the interface layer to write back the target medical record to the hospital information system, and perform diagnosis and treatment recommendations based on the target medical record to obtain a recommendation result, so as to invoke the interface layer to feedback the recommendation result to the hospital information system for the doctor's reference. The second invocation module 53 is configured to invoke the interface layer to obtain the doctor's diagnosis from the hospital information system. The first detection module 54 is configured to perform a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable. Wherein, when the first detection result indicates that the doctor's diagnosis is unreasonable, the interface layer is invoked to feedback the first detection result to the hospital information system. The third invocation module 55 is configured to invoke the interface layer to obtain the doctor's prescription from the hospital information system. The second detection module 56 is configured to perform a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable. Wherein, when the second detection result indicates that the doctor's prescription is unreasonable, the interface layer is invoked to feedback the second detection result to the hospital information system.

[0115] In the above solution, the diagnosis and treatment control device 50 calls the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system, then improves the original medical record based on the quality inspection result of the original medical record to obtain the target medical record, and calls the interface layer to write back the target medical record to the hospital information system, and performs diagnosis and treatment recommendations based on the target medical record to obtain the recommendation result, so as to call the interface layer to feedback the recommendation result to the hospital information system for doctors' reference. On this basis, it further calls the interface layer to obtain the doctor's diagnosis from the hospital information system, and performs a first detection based on the doctor's diagnosis to obtain the first detection result indicating whether the doctor's diagnosis is reasonable. When the first detection result indicates that the doctor's diagnosis is unreasonable, it calls the interface layer to feedback the first detection result to the hospital information system, then calls the interface layer to obtain the doctor's prescription from the hospital information system, and performs a second detection based on the doctor's prescription to obtain the second detection result indicating whether the doctor's prescription is reasonable. When the second detection result indicates that the doctor's prescription is unreasonable, it calls the interface layer to feedback the second detection result to the hospital information system. Therefore, by docking with the auxiliary diagnosis system, it can assist doctors in various links such as medical record entry, diagnosis issuance, and prescription writing. On the one hand, it can minimize the impact on the existing medical system as much as possible, neither making the existing medical system more bloated nor damaging the normal use of the existing medical system. On the other hand, since the auxiliary diagnosis system and the medical system are separated from each other, even if either party is updated, the impact on the other party is very small. Therefore, it can improve the operation and maintenance convenience on the premise of meeting the auxiliary diagnosis and treatment requirements.

[0116] Please refer to Figure 6 , Figure 6 which is a framework schematic diagram of an embodiment of the electronic device 60 of the present application. The electronic device 60 includes a communication circuit 61, a memory 63, and a processor 62. The communication circuit 61 and the memory 63 are respectively coupled to the processor 62. Program instructions are stored in the memory 63, and the processor 62 is configured to execute the program instructions to implement the steps in any of the above embodiments of the diagnosis and treatment control method. Specifically, reference can be made to the foregoing disclosed embodiments, which will not be elaborated herein. The electronic device 60 may specifically include, but is not limited to, a server, a microcomputer, etc., which are not limited herein.

[0117] Specifically, the processor 62 is used to control itself, the communication circuit 61, and the memory 63 to implement the steps in any of the above-described diagnosis and treatment control method embodiments. The processor 62 may also be referred to as a CPU (Central Processing Unit). The processor 62 may be an integrated circuit chip with signal processing capabilities. The processor 62 may also be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. Additionally, the processor 62 may be implemented jointly by integrated circuit chips.

[0118] In the above solution, the electronic device 60 calls the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system, then improves the original medical record based on the quality inspection result of the original medical record to obtain the target medical record, and calls the interface layer to write the target medical record back to the hospital information system, and performs diagnosis and treatment recommendations based on the target medical record to obtain a recommendation result, so as to call the interface layer to feedback the recommendation result to the hospital information system for doctors to refer to. On this basis, it further calls the interface layer to obtain the doctor's diagnosis from the hospital information system, and performs a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable. When the first detection result indicates that the doctor's diagnosis is unreasonable, it calls the interface layer to feedback the first detection result to the hospital information system, then calls the interface layer to obtain the doctor's prescription from the hospital information system, and performs a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable. When the second detection result indicates that the doctor's prescription is unreasonable, it calls the interface layer to feedback the second detection result to the hospital information system. Therefore, by docking with the auxiliary diagnosis system, it can assist doctors in various links such as entering medical records, issuing diagnoses, and prescribing medications. On the one hand, it can minimize the impact on the existing medical system as much as possible, neither making the existing medical system more bloated nor damaging the normal use of the existing medical system. On the other hand, since the auxiliary diagnosis system and the medical system are separated from each other, even if either party is updated, the impact on the other party is very small. Therefore, it can improve the operation and maintenance convenience on the premise of meeting the auxiliary diagnosis and treatment requirements.

[0119] Please refer to Figure 7 , Figure 7It is a schematic framework diagram of an embodiment of the computer-readable storage medium 70 of the present application. The computer-readable storage medium 70 stores program instructions 71 that can be run by a processor, and the program instructions 71 are used to implement the steps in any of the above-described embodiments of the diagnosis and treatment control method.

[0120] In the above solution, the computer-readable storage medium 70 calls the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system, then improves the original medical record based on the quality inspection result of the original medical record to obtain the target medical record, and calls the interface layer to write the target medical record back to the hospital information system, and performs diagnosis and treatment recommendations based on the target medical record to obtain a recommendation result, so as to call the interface layer to feedback the recommendation result to the hospital information system for doctors to refer to. On this basis, it then calls the interface layer to obtain the doctor's diagnosis from the hospital information system, and performs a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable. When the first detection result indicates that the doctor's diagnosis is unreasonable, it calls the interface layer to feedback the first detection result to the hospital information system, then calls the interface layer to obtain the doctor's prescription from the hospital information system, and performs a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable. When the second detection result indicates that the doctor's prescription is unreasonable, it calls the interface layer to feedback the second detection result to the hospital information system. Therefore, by docking with the auxiliary diagnosis system, it can assist doctors in various links such as entering medical records, issuing diagnoses, and prescribing prescriptions. On the one hand, it can minimize the impact on the existing medical system as much as possible, neither making the existing medical system more bloated nor damaging the normal use of the existing medical system. On the other hand, since the auxiliary diagnosis system and the medical system are separated from each other, even if either party is updated, the impact on the other party is very small. Therefore, it can improve the operation and maintenance convenience on the premise of meeting the auxiliary diagnosis and treatment.

[0121] 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 methods described in the above method embodiments, and their specific implementations can refer to the descriptions of the above method embodiments. For the sake of brevity, they will not be repeated here.

[0122] The descriptions of the above embodiments tend to emphasize the differences between the embodiments, and their similarities can be referred to each other. For the sake of brevity, they will not be repeated in this article.

[0123] In several embodiments provided by the present application, it should be understood that the disclosed methods and apparatuses can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, 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 displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of apparatuses or units can be in electrical, mechanical or other forms.

[0124] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0126] 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, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in each embodiment of the present application. And the aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks or optical discs that can store program codes.

[0127] 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 diagnosis assistance system, characterized in that, Including: An interface layer, including a number of interfaces, which are used to respectively connect to different medical systems of medical institutions; A data layer, which is used to obtain the data retrieved by the interface layer to perform data docking with the medical systems; A capability layer, including a number of artificial intelligence algorithms, which are used to process the data of the medical systems; A service layer, including a number of basic business modules, which are used to provide standard services for the application layer; An application layer, including a number of medical business applications, which are used to respectively implement different medical assistance functions, and the medical business applications are implemented by at least one of the basic business modules.

2. The diagnosis assistance system according to claim 1, characterized in that, The number of medical business applications at least includes voice-assisted medical record generation, which is used to connect to the hospital information system in the different medical systems through the first interface in the interface layer, and based on the speech recognition algorithm and natural language understanding in the capability layer, process the doctor-patient voice conversation retrieved by the first interface to obtain a target medical record text.

3. The diagnosis assistance system according to claim 1, characterized in that, The number of medical business applications at least includes intelligent contact-assisted input, which is used to connect to the hospital information system in the different medical systems through the first interface in the interface layer, obtain the historical behavior data and the current input content retrieved by the first interface, and based on the input recommendation algorithm in the capability layer, recommend the next input content for the current input content with reference to the historical behavior data.

4. The diagnosis assistance system according to claim 1, characterized in that, The number of medical business applications at least includes medical record content quality control, which is used to connect to the hospital information system in the different medical systems through the first interface in the interface layer, and based on the natural language understanding in the capability layer, perform content quality inspection on the medical record text to be inspected retrieved by the first interface to obtain the content quality inspection result of the medical record text to be inspected.

5. The diagnosis assistance system according to claim 1, characterized in that, The number of medical business applications at least includes inspection / examination item quality control, which is used to connect to the hospital information system in the different medical systems through the first interface in the interface layer, and perform item quality control on the inspection / examination items retrieved by the first interface based on the basic information and condition information of the target patient to obtain an item quality control result that at least characterizes whether the inspection / examination items are reasonable.

6. The diagnosis assistance system according to claim 1, characterized in that, The number of medical business applications at least includes inspection report interpretation, which is used to connect to the laboratory information system in the different medical systems through the second interface in the interface layer, and interpret each index in the inspection report retrieved by the second interface to obtain a first interpretation result of the inspection report; and / or The number of medical business applications at least includes examination report interpretation, which is used to connect to the picture archiving and communication system in the different medical systems through the third interface in the interface layer, and interpret each index in the examination report retrieved by the third interface to obtain a second interpretation result of the examination report; and / or, the number of medical business applications at least includes a report interpretation application, which is used to identify the medical report image through optical character recognition to obtain medical report data.

7. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include disease risk warning, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer, predict the basic information and disease conditions obtained by the first interface, obtain several recommended diseases, identify the risk levels of each of the recommended diseases, and give real-time warnings for the recommended diseases whose risk levels meet the preset conditions.

8. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications further include special disease assistant diagnosis, which is used to predict the basic information and disease conditions of a target patient and obtain several recommended diseases belonging to special diseases.

9. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include multi-source data combined assistant diagnosis, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer to obtain historical medical records from the first interface, dock with the laboratory information system in the different medical systems through the second interface in the interface layer to obtain inspection reports from the second interface, dock with the picture archiving and communication system in the different medical systems through the third interface in the interface layer to obtain examination reports from the third interface, and dock with the basic public health management system in the different medical systems through the fourth interface in the interface layer to obtain health records, and then predict the multi-source data including the historical medical records, the inspection reports, the examination reports, and the health records to obtain recommended diseases.

10. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include inference visualization, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer, infer the reference data obtained by the first interface, obtain the inference paths of several recommended diseases, and medical evidences are attached to the inference paths, and the reference data are data related to the auxiliary diagnosis and treatment inference process.

11. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include hierarchical referral interconnection, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer, identify the diagnosed diseases obtained by the first interface, obtain the identification results of the diagnosed diseases, and the identification results include whether the diagnosed diseases exceed the service scope of the medical institution, and in response to the identification results including that the diagnosed diseases exceed the service scope of the medical institution, initiate referral suggestions and / or dock with the referral system in real time through the fifth interface in the interface layer.

12. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include diagnosis and treatment path recommendation, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer, and recommend the diagnosis and treatment paths for the diagnosed diseases obtained by the first interface to obtain the diagnosis and treatment paths of the diagnosed diseases.

13. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include prescription review intervention, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer, and pre-place the medical prescriptions obtained by the first interface in advance for review by pharmacists, and before outputting the medical prescriptions to the pharmacists, transmit the review results of the medical prescriptions back to the hospital information system through the first interface for the doctor side to intervene in unreasonable prescriptions.

14. The diagnosis assistance system according to claim 1, characterized in that, The several medical service applications at least include dynamic knowledge recommendation, which is used to dock with the hospital information system in the different medical systems through the first interface in the interface layer, and predict the user portraits and user behaviors obtained by the first interface to obtain recommended knowledge.

15. A diagnosis and treatment control method, characterized in that, Applied to the auxiliary diagnosis system according to any one of claims 1 to 14, the diagnosis and treatment control method includes: Invoking the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system; Based on the quality inspection result of the original medical record, improving the original medical record to obtain a target medical record, invoking the interface layer to write back the target medical record to the hospital information system, and making a diagnosis and treatment recommendation based on the target medical record to obtain a recommendation result, so as to invoke the interface layer to feedback the recommendation result to the hospital information system for the doctor's reference; Invoking the interface layer to obtain the doctor's diagnosis from the hospital information system; Performing a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable; wherein, when the first detection result indicates that the doctor's diagnosis is unreasonable, invoking the interface layer to feedback the first detection result to the hospital information system; Invoking the interface layer to obtain the doctor's prescription from the hospital information system; Performing a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable; wherein, when the second detection result indicates that the doctor's prescription is unreasonable, invoking the interface layer to feedback the second detection result to the hospital information system.

16. A diagnosis and treatment control device, characterized in that, Applied to the auxiliary diagnosis system according to any one of claims 1 to 14, the diagnosis and treatment control device includes: A first invocation module, which is used to invoke the interface layer in the auxiliary diagnosis system to obtain the original medical record from the hospital information system; An information interaction module, which is used to improve the original medical record based on the quality inspection result of the original medical record to obtain a target medical record; and invoke the interface layer to write back the target medical record to the hospital information system; and make a diagnosis and treatment recommendation based on the target medical record to obtain a recommendation result, so as to invoke the interface layer to feedback the recommendation result to the hospital information system for the doctor's reference; A second invocation module, which is used to invoke the interface layer to obtain the doctor's diagnosis from the hospital information system; A first detection module, which is used to perform a first detection based on the doctor's diagnosis to obtain a first detection result indicating whether the doctor's diagnosis is reasonable; wherein, when the first detection result indicates that the doctor's diagnosis is unreasonable, invoking the interface layer to feedback the first detection result to the hospital information system; A third calling module, configured to call the interface layer to obtain a doctor's prescription from the hospital information system; A second detection module, configured to perform a second detection based on the doctor's prescription to obtain a second detection result indicating whether the doctor's prescription is reasonable; wherein, when the second detection result indicates that the doctor's prescription is unreasonable, the interface layer is called to feedback the second detection result to the hospital information system.

17. An electronic device, characterized in that, It includes a communication circuit, a memory, and a processor. The communication circuit and the memory are respectively coupled to the processor. Program instructions are stored in the memory, and the processor is configured to execute the program instructions to implement the diagnosis and treatment control method according to claim 15.

18. A computer-readable storage medium, characterized in that, Program instructions that can be run by a processor are stored, and the program instructions are configured to implement the diagnosis and treatment control method according to claim 15.

19. An auxiliary diagnosis deployment system, characterized in that, The auxiliary diagnosis deployment system includes: a customer subsystem, a service subsystem, and an interface subsystem communicatively connected between the customer subsystem and the service subsystem. The customer subsystem is configured to interface with the medical system of a medical institution, and the customer subsystem is the auxiliary diagnosis system according to any one of claims 1 to 14.

20. The auxiliary diagnosis deployment system according to claim 19, characterized in that, The service subsystem is deployed in the cloud; and / or, the service subsystem at least includes an artificial intelligence auxiliary diagnosis platform, and the artificial intelligence auxiliary diagnosis platform provides artificial intelligence algorithms; and / or, the service subsystem at least includes a business application service platform, and the business application service platform provides medical business applications; and / or, the service subsystem at least includes a basic capability service platform, and the basic capability service platform provides basic service algorithms.

Images