New infectious disease monitoring and early warning method and device based on multi-point triggering

Abstract

The embodiment of the invention discloses a new infectious disease monitoring and early warning method and device based on multi-point triggering. The method comprises the following steps: acquiring clinical data of a patient in HIS, LIS, PACS, operation and electronic medical records from a medical institution; setting a screening strategy of new infectious diseases; identifying suspected cases according to the clinical data of the patient; identifying abnormal diagnosis and treatment behaviors of the patient from clinical data of the patient based on big data analysis; determining a suspected case level according to the suspected case and the abnormal diagnosis and treatment behavior; obtaining relevant survey data of suspected cases; judging whether the suspected cases have aggregated cases or not according to the related survey data of the suspected cases; and if the suspected cases have aggregated cases, determining an infectious disease transmission route, and performing case management. By implementing the method provided by the embodiment of the invention, real-time monitoring and early warning of infectious disease epidemic situations and other public health events can be realized, and the processing efficiency and accuracy of new infectious diseases are improved.

Description

Technical Field

[0001] The present invention relates to a disease monitoring method, and more specifically to a method and device for monitoring and early warning of emerging infectious diseases based on multi-point triggering. Background Art

[0002] Infectious diseases have a profound impact on human health and socio-economics. Therefore, timely detection and early warning are crucial to take public health measures to reduce the harm caused by infectious diseases. However, the ability to monitor and warn of infectious diseases is still insufficient and needs to be improved. The current online direct reporting system relies on frontline medical staff to actively report cases, but there are often underreporting and late reporting, resulting in the number of reported cases being far lower than the actual number of cases, thus affecting the timing of epidemic assessment and control.

[0003] Existing methods for the prevention and control of emerging infectious diseases use big data models to screen suspected cases and assess the level of infection based on the number of patients and risk factors. However, these methods often ignore the transmission relationship between patients and lack detailed analysis of community clustering and hospital clustering investigations, resulting in the inability to accurately identify the transmission paths of emerging infectious diseases and find the source of infection. In addition, due to the wide variety of emerging infectious diseases, different transmission methods and clinical manifestations, and the lack of unified standards, artificial intelligence models for different transmission types are inefficient when processing large amounts of data in real time, and have high hardware requirements.

[0004] Therefore, it is necessary to design a new method to achieve real-time monitoring and early warning of infectious disease outbreaks and other public health events, and to improve the efficiency and accuracy of handling emerging infectious diseases. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the prior art and provide a method and device for monitoring and early warning of emerging infectious diseases based on multi-point triggering.

[0006] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a method for monitoring and early warning of emerging infectious diseases based on multi-point triggering, comprising: Obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; Setting screening strategies for emerging infectious diseases; Identify suspected cases based on the patient's clinical data; Identifying abnormal diagnosis and treatment behaviors of patients from the clinical data of the patients based on big data analysis; The level of suspected cases is determined according to the suspected cases and the abnormal diagnosis and treatment behaviors; wherein the suspected case levels include typical cases, highly suspected cases, moderately suspected cases and mildly suspected cases; typical cases are cases with particularly abnormal diagnosis and treatment behaviors and the patients are suspected cases; highly suspected cases are cases with abnormal diagnosis and treatment behaviors and the patients are suspected patients; moderately suspected cases meet one of the conditions that the patient's diagnosis and treatment behaviors are generally abnormal or the patient is a suspected patient; mildly suspected cases meet only the condition of unknown cause one and the patient is not a suspected patient; the condition of unknown cause one is that the medication does not match the laboratory test results; Obtain investigation data related to suspected cases; Determine whether there are clustered cases of the suspected cases based on the investigation data related to the suspected cases; If there are clusters of suspected cases, the transmission route of the infectious disease will be determined and case management will be carried out.

[0007] Its further technical solution is: after judging whether the suspected case is a cluster case according to the investigation data related to the suspected case, it also includes: If there is no clustered case for the suspected case, case management will be carried out for the suspected case.

[0008] Its further technical solution is: the screening strategy includes the characteristics of emerging infectious diseases, and the characteristics of emerging infectious diseases include symptoms of infection but unclear pathogens, unsatisfactory treatment effects, and epidemiological correlations between cases.

[0009] A further technical solution is: the identifying of suspected cases according to the clinical data of the patient comprises: Mapping the characteristics of the emerging infectious disease with the patient clinical data to obtain a mapping result; The patient information that meets the screening strategy is screened out from the mapping results to obtain suspected cases.

[0010] A further technical solution is: the method of identifying abnormal diagnosis and treatment behaviors of patients from the clinical data of the patients based on big data analysis includes: Based on big data analysis, the patient's diagnosis and treatment data that meets the condition of unknown etiology is screened out from the patient's clinical data to obtain abnormal diagnosis and treatment behavior; Among them, the conditions with unknown causes include: Except for preventive medication, medication is inconsistent with laboratory test results; Even when there were positive reports, clinicians still sent samples for respiratory-related pathogen tests, and the results were all negative.

[0011] A further technical solution is: the determination of the transmission route of infectious diseases and the case management include: Establish the transmission chain of suspected cases, obtain the transmission route of the unidentified infectious disease based on the level and clustering relationship of suspected cases, obtain the transmission route of the infectious disease, and conduct case management.

[0012] The present invention also provides an emerging infectious disease monitoring and early warning device based on multi-point triggering, comprising: Clinical data acquisition unit, used to obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; The setting unit is used to set the screening strategy for emerging infectious diseases; A case identification unit, used to identify suspected cases based on the patient's clinical data; A behavior recognition unit, used to identify abnormal diagnosis and treatment behaviors of the patient from the patient's clinical data based on big data analysis; A level determination unit is used to determine the level of the suspected case according to the suspected case and the abnormal diagnosis and treatment behavior; wherein the suspected case level includes typical cases, highly suspected cases, moderately suspected cases and mildly suspected cases; wherein a typical case is a case in which the test submission behavior is particularly abnormal and the patient's condition is abnormal; a highly suspected case is a case in which both the test submission behavior and the patient's condition are abnormal; a moderately suspected case is a case that meets the basic test submission requirements and either the test submission behavior or the patient's condition is abnormal; a mildly suspected case is a case that meets the basic test submission requirements but the test submission behavior and the patient's condition do not meet the abnormal standards; An investigation data acquisition unit, used to obtain investigation data related to suspected cases; A judgment unit, used to judge whether the suspected case is a cluster case according to the investigation data related to the suspected case; The route determination unit is used to determine the route of infectious disease transmission and perform case management if there are clustered cases among the suspected cases.

[0013] The present invention further provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above method when executing the computer program.

[0014] The present invention also provides a storage medium, wherein the storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0015] The beneficial effects of the present invention compared with the prior art are as follows: the present invention obtains clinical data from medical institutions, sets screening strategies for emerging infectious diseases, and identifies suspected cases; uses big data analysis to identify patients' abnormal diagnosis and treatment behaviors; determines case levels based on suspected cases and their abnormal behaviors, and obtains relevant investigation data; determines whether there are clustered cases, and if so, determines the transmission route and conducts case management; realizes real-time monitoring and early warning of infectious disease outbreaks and other public health events, and improves the efficiency and accuracy of handling emerging infectious diseases.

[0016] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of an application scenario of the emerging infectious disease monitoring and early warning method based on multi-point triggering provided by an embodiment of the present invention; Figure 2 A schematic diagram of a process for monitoring and early warning of emerging infectious diseases based on multi-point triggering provided by an embodiment of the present invention; Figure 3 A schematic diagram of a sub-process of a method for monitoring and early warning of emerging infectious diseases based on multi-point triggering provided by an embodiment of the present invention; Figure 4 A schematic diagram of screening suspected cases provided by an embodiment of the present invention; Figure 5 A schematic diagram of an infectious disease transmission route provided by an embodiment of the present invention; Figure 6 A schematic block diagram of an emerging infectious disease monitoring and early warning device based on multi-point triggering provided by an embodiment of the present invention; Figure 7 A schematic block diagram of a case identification unit of an emerging infectious disease monitoring and early warning device based on multi-point triggering provided by an embodiment of the present invention; Figure 8 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

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

[0019] See also Figure 1 and Figure 2 , Figure 1 Schematic diagram of application scenarios of the emerging infectious disease monitoring and early warning method based on multi-point triggering provided in an embodiment of the present invention. Figure 2 A schematic flow chart of a method for monitoring and early warning of emerging infectious diseases based on multi-point triggering provided by an embodiment of the present invention. The method for monitoring and early warning of emerging infectious diseases based on multi-point triggering is applied to a server. The server interacts with the hospital information system for data, screens out suspected cases by analyzing the hospital's patient diagnosis and treatment data and screening strategies, and then determines the suspected level of these cases based on the abnormal degree of the patient's diagnosis and treatment behavior and the change of the condition. At the same time, it is determined whether there are signs of "human-to-human transmission" in the suspected cases, such as whether there are clustered cases in the community or hospital. On this basis, actively identify those cases that may be infected with unknown causes, and promptly remind managers and clinicians to conduct in-depth investigations. For cases initially diagnosed as unexplained infections, they will be separated as required, epidemiological investigations will be carried out, an in-hospital expert group will be organized for consultation, and relevant information will be reported. Once the in-hospital expert group confirms the case of unexplained infection, it will start the corresponding prevention and control measures in combination with the infectiousness assessment results to ensure the safety of patients and the public.

[0020] Figure 2 FIG. 1 is a flow chart of a method for monitoring and early warning of emerging infectious diseases based on multi-point triggering provided by an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S110 to S190.

[0021] S110, obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions.

[0022] In this embodiment, data is acquired from multiple data sources of the hospital information system through interfaces in various forms, and then formed into a specific unified data format through a unified data conversion component to form a standard unified data structure.

[0023] The above-mentioned hospital information systems may include HIS (Hospital Information System), LIS (Laboratory Information Management System), RIS (Radiology Information System), EMR (Electronic Medical Record) system, PACS (Picture Archiving And Communication System) and other systems.

[0024] Patient clinical data refers to various information related to the patient's health status and medical process. These data usually include: Basic information: the patient's name, age, gender, contact information, etc. Medical history: including past medical history, family medical history and allergy history, etc. Physical signs: such as vital signs (heart rate, blood pressure, body temperature, etc.) and other physical examination results. Laboratory test results: test results of blood, urine and other biological samples. Imaging examination: X-ray, CT, MRI and other imaging data and their interpretation results. Diagnostic information: the doctor's diagnosis and classification of the patient's disease. Treatment records: including medication, surgical records, rehabilitation plans, etc. Follow-up records: subsequent health changes of patients and evaluation of treatment effects.

[0025] S120. Establish screening strategies for emerging infectious diseases.

[0026] In this embodiment, the screening strategy includes the characteristics of emerging infectious diseases, and the characteristics of emerging infectious diseases include the presence of infection symptoms but unclear pathogens, unsatisfactory treatment effects, and epidemiological correlations between cases.

[0027] Drawing on the definition of pneumonia of unknown cause, the main characteristics of emerging infectious diseases can be summarized as follows: Symptoms of infection but unclear pathogen: Patients usually show typical symptoms of infection, such as cough, fever, and difficulty breathing. However, despite these symptoms, the pathogen (such as bacteria, virus, or fungus) that causes the disease cannot be clearly identified through conventional testing. This may be because the pathogen has not yet been discovered or the current detection methods cannot effectively identify new pathogens.

[0028] Unsatisfactory treatment effects: Conventional treatment measures (such as antibiotics or antiviral drugs) are often ineffective for such diseases. Patients may experience significant symptom improvement, but may also experience progressive worsening of the disease, leading to worsening of the condition. This unsatisfactory treatment effect poses greater challenges to medical personnel and increases the urgent need for etiology research and the development of new treatment options.

[0029] Contagious: Such emerging infectious diseases are somewhat contagious and can be spread between cases. The transmission route may include direct contact between people, and may also involve transmission between people and animals, especially in some zoonotic diseases. There is an epidemiological correlation between cases, that is, by tracking the cases, the connection between them can be found, which is crucial for determining the source of the epidemic and formulating prevention and control measures.

[0030] These characteristics not only help identify and define emerging respiratory infectious diseases, but also provide important clues for public health response. Understanding these characteristics can promote early diagnosis, rational treatment and effective control of the spread of the epidemic. Researchers and public health experts need to pay close attention to the epidemic dynamics of these diseases and strengthen the monitoring and research of potential pathogens in order to respond to sudden public health events in a timely manner.

[0031] S130. Identify suspected cases based on the patient's clinical data.

[0032] In this embodiment, suspected cases refer to patients who exhibit symptoms associated with an emerging infectious disease but whose infection has not yet been confirmed by laboratory tests and who may be contagious.

[0033] This step identifies patients who may be infected with emerging infectious diseases by analyzing their clinical data.

[0034] In one embodiment, see Figure 3 , the above-mentioned step S130 may include steps S131~S132.

[0035] S131. Mapping the characteristics of the emerging infectious disease with the patient's clinical data to obtain a mapping result.

[0036] In this embodiment, data analysis techniques (such as machine learning and statistical analysis) are used to compare the characteristics of emerging infectious diseases with the clinical data of patients to generate mapping results. Through systematic analysis, it is helpful to more accurately identify possible suspected cases; it can capture some cases with mild or atypical symptoms and detect potential infected persons at an early stage. First, referring to the definition of pneumonia of unknown cause, the main characteristics of emerging respiratory infectious diseases are summarized as follows: 1) There are symptoms of infection, but the pathogen is unclear. 2) The treatment effect is not ideal, including no obvious improvement or progressive aggravation. 3) It is contagious. It is transmitted between people or between people and animals, that is, there is an epidemiological correlation between cases. According to the above summary, four screening nodes are analyzed: infection indication, unknown cause, poor treatment effect, and infectiousness. Through data analysis technology, the system identifies whether there are infection indications, unknown cause, poor treatment effect, and infectiousness based on the patient's clinical data.

[0037] First, a logistic regression model is used to analyze and identify the infection status of patients. The hospital's historical data information is obtained, and cases of patients with emerging infectious diseases are extracted as training data, and the training set is labeled.

[0038] First, data preparation: Before model training, it is necessary to extract relevant data from the patient's clinical and behavioral information to construct a data set. The input data includes the following types of information: Clinical Information: Physical sign data: such as fever, cough, difficulty breathing, etc.

[0039] Laboratory test results: including blood routine, C-reactive protein, pathogen examination, etc.

[0040] Imaging data: such as chest X-ray or CT results.

[0041] Behavioral Information: Diagnostic and treatment behavior: whether the physician prescribes etiological examination and whether the physician adjusts anti-infective drugs.

[0042] Records of antibiotic use: type of drug, time and frequency of dosage adjustment.

[0043] Hospitalization process: changes in the severity of the condition (such as transfer of patients from the general ward to the ICU).

[0044] Infection events: such as infection of medical staff or contacts during hospitalization.

[0045] Feature selection and label setting: In order for the logistic regression model to identify the most relevant influencing factors for each criterion, it is necessary to set corresponding labels for each criterion and select appropriate features for training.

[0046] Label settings: Infection signs: The label indicates whether there are infection signs (1 for yes, 0 for no).

[0047] Unknown cause: The label indicates whether the cause is unknown (1 for unknown, 0 for clear).

[0048] Poor treatment effect: The label is whether the treatment effect is poor (1 for poor and 0 for good).

[0049] Infectivity: Label is whether the virus is infectious (1 for yes, 0 for no).

[0050] Feature Selection: The prediction task for each label inputs the same features, but the model determines the most influential features for each judgment criterion based on the training results. The input features include: Anti-infection drug use records: whether antibiotics and antiviral drugs have been used.

[0051] Pathological examination: whether pathogenic examinations such as PCR and microbial culture have been prescribed.

[0052] Antimicrobial drug adjustment records: whether the type or dosage of antibiotics was adjusted during hospitalization.

[0053] Body temperature data: whether there is unexplained fever, persistent fever, or remittent fever.

[0054] Change in condition level: Whether the patient's condition level changes during hospitalization, such as transfer to the ICU.

[0055] Abnormal imaging examination: whether the chest X-ray or CT results show abnormalities.

[0056] Infection contact history: whether the patient has had contact with other infected persons (including medical staff or contacts).

[0057] Finally, model training and analysis are performed. After data preparation and feature selection are completed, the logistic regression model is used to train each judgment criterion independently. The model will try to find the most relevant features for each judgment criterion by minimizing the loss function. The training process can be divided into the following steps: Data splitting: Divide the dataset into training and test sets, 80% for training and 20% for testing to ensure the generalization ability of the model.

[0058] Model training: A separate logistic regression model is trained for each criterion (indications of infection, unknown etiology, poor treatment effect, and infectiousness). The logistic regression model maps the input features to the labels of each criterion, learns the coefficient weights of each feature, and determines which features are most relevant to the results.

[0059] Model optimization: The model is optimized through cross-validation, L1 / L2 regularization and other techniques to avoid overfitting and improve the stability of the model. Regularization can reduce the coefficients of unimportant features and make the model more concise.

[0060] Feature coefficient analysis: After training is completed, the model's regression coefficient (weight) is used to determine the contribution of each feature to the prediction result. A positive coefficient indicates that the feature is positively correlated, and a negative coefficient indicates that the feature is negatively correlated.

[0061] Through model training and feature weight analysis, the most relevant influencing factors for each criterion can be obtained. These results are based on the quantitative analysis of the correlation between each feature and the output result by the logistic regression model. The following are the results for each criterion: The most relevant influencing factor for those with infection indications: The most relevant influencing factor: having a record of using anti-infective drugs.

[0062] Analysis results: The model shows that the use of anti-infective drugs (such as antibiotics or antiviral drugs) is the most relevant influencing factor for "infection indications". The weight of this feature is high, indicating that the use of anti-infective drugs is significantly positively correlated with infection indications.

[0063] The most relevant influencing factor for unknown etiology: The most relevant influencing factor: prescribing etiological examination (such as PCR testing, microbial culture, etc.).

[0064] Analysis results: The logistic regression model found that the ordering of etiological tests was the most important influencing factor for "unclear etiology". Doctors usually ordered etiological tests because the cause of the disease was not yet clear, and the coefficient of etiological tests was significant in predicting the state of unknown etiology.

[0065] Factors most associated with poor treatment outcomes: Most relevant factors: The following four characteristics were most strongly associated with poor treatment outcomes: Antimicrobial drug adjustment: The physician adjusts the type or dosage of antimicrobial drugs during hospitalization.

[0066] Fever of unknown origin: During the screening period, the patient has fever of unknown origin, persistent fever or remittent fever.

[0067] Change in condition level: The patient's condition level changes during hospitalization, such as being transferred from a general ward to an ICU.

[0068] Abnormal imaging examination: chest X-ray or CT examination shows abnormalities.

[0069] Analysis results: The model found that these four features are the main influencing factors of "poor treatment effect". The coefficient weight of each feature indicates its importance in predicting poor treatment effect, especially the adjustment of antimicrobial drugs and the change of disease level, which have a high positive correlation.

[0070] The most relevant influencing factors for infectiousness: The most relevant influencing factors: medical staff are infected during hospitalization or other contacts with patients are confirmed to be infected.

[0071] Analysis results: Through the logistic regression model analysis, whether there are medical staff infected or other contacts confirmed to be infected during hospitalization is the most relevant influencing factor for predicting "infectious". The weight of this feature is very high, indicating that the patient's infectious contact with medical staff or other patients is a key factor in determining infectiousness.

[0072] By performing logistic regression analysis on the influencing factors of each criterion, the system can apply these most relevant features to the early screening and early warning mechanism of emerging infectious diseases: Indications of infection: If the model detects that the patient has used anti-infective drugs, the system will prompt that there may be indications of infection.

[0073] Unknown etiology: The system can automatically mark patients with unknown etiology by detecting the ordering of etiological tests.

[0074] Poor treatment effect: If the system monitors the adjustment of antimicrobial drugs, persistent fever or changes in the condition, it will indicate that the treatment effect may be poor.

[0075] Infectiousness: Once a contact or medical staff is found to be infected, the system will issue an infectious risk warning.

[0076] S132. Filter out patient information that meets the screening strategy from the mapping results to obtain suspected cases.

[0077] In this embodiment, the mapping results are screened to find patient information that meets the above strategy and mark them as suspected cases; the screening results are recorded and reported to relevant public health agencies in a timely manner. Suspected cases can be quickly identified and reported to promote epidemic control; resources are concentrated to further detect and manage suspected cases, improving the efficiency of public health response.

[0078] Through the above steps, suspected cases can be efficiently identified and managed, which helps to detect and control the spread of emerging infectious diseases at an early stage. This is not only conducive to early treatment of patients, but also crucial to overall public health security.

[0079] Specifically, see Figure 4 , map the main characteristics of emerging respiratory infectious diseases with relevant diagnosis and treatment data, and design an algorithm to screen suspected cases by computer. The specific steps are as follows: Infection indications: The patient has a record of using anti-infective drugs within the set screening time range.

[0080] Unknown cause: Clinicians prescribe etiological tests to patients to find potential causes.

[0081] Poor treatment effect: any of the following conditions are met: After the patient was hospitalized, the physician adjusted the type or dosage of antibiotics.

[0082] During the screening period, the patient develops unexplained fever (such as lasting for more than three weeks, body temperature above 38.5°C) or persistent fever (oral temperature above 37.3°C or rectal temperature above 37.6°C, with diurnal fluctuations exceeding 1.2°C), and meets different clinical classifications (low fever, moderate fever, high fever, very high fever).

[0083] The patient's condition changes during hospitalization, or imaging test results are abnormal.

[0084] Infectiousness: Medical staff come into contact with patients confirmed to be infected during hospitalization.

[0085] Only when all four of the above screening strategies (infection indications, unknown etiology, poor treatment response and infectiousness) are met can the patient be identified as a suspected case.

[0086] S140. Identify abnormal diagnosis and treatment behaviors of the patient from the patient's clinical data based on big data analysis.

[0087] In this embodiment, abnormal diagnosis and treatment behavior refers to the diagnosis and treatment data of patients that meet the condition of unknown cause.

[0088] Build a system based on big data analysis and automated rule detection to screen patients who meet the "unknown cause" condition from their clinical data, and automatically mark those situations that meet the criteria of abnormal diagnosis and treatment behaviors through big data processing and intelligent algorithms.

[0089] Preprocess anti-infective drug use records, laboratory test results, and medical records. Build a rule engine, where the rules automatically evaluate each patient's medical data based on the two main conditions of "unclear cause". The rule engine can be implemented based on a database query language.

[0090] Unknown cause Condition 1: The medication does not match the laboratory test results Identification logic: The system will first identify all patients who have used anti-infective drugs based on the patient's anti-infective drug use records. Then, the system will look for pathogen test results corresponding to the anti-infective drug from the patient's laboratory test results. If the patient's laboratory test results are negative (no pathogens are detected), but the clinician still prescribes the corresponding anti-infective drug for the patient (non-preventive medication), this behavior is considered to be inconsistent with the medication and test results. For example, the patient did not detect a positive result for bacterial infection, but the doctor prescribed antibiotic treatment for him.

[0091] If the medication behavior is inconsistent with the laboratory test results, the system will mark the behavior as abnormal diagnosis and treatment behavior.

[0092] Condition 2: Frequent respiratory pathogen testing with negative results despite positive reports Identification logic: The system will first confirm whether the patient has a positive report (i.e. a positive result for a certain pathogen).

[0093] If there is a positive report, the system will further check whether the patient continued to undergo respiratory-related pathogen tests after the positive result.

[0094] The system then checks to see if the results of these follow-up tests are all negative.

[0095] If a physician frequently orders respiratory pathogen testing despite a positive result, and all subsequent test results are negative, this behavior is considered abnormal, indicating that the physician may suspect other potential causes and does not believe the initial positive report.

[0096] This behavior meets the second condition of "unclear cause", and the system will automatically mark this behavior as abnormal diagnosis and treatment behavior.

[0097] After the rule engine is built, the system will use big data processing tools to screen all patient data in real time. The entire process is as follows: Real-time monitoring and updating: The system can set up regular automatic tasks (hourly / daily) to obtain the latest patient data from the database and perform automated rule detection. For inpatients, the system can monitor the updates of medication and laboratory results in real time.

[0098] Parallel computing: In the context of large-scale hospital data, the system uses Hadoop, a parallel processing technology, to speed up the data screening process, especially when processing large amounts of patient data. This approach helps to efficiently screen out eligible patients.

[0099] Abnormal behavior marking: When the system identifies qualified abnormal diagnosis and treatment behaviors, it will automatically record the behavior as an abnormal event and mark the relevant patients and doctors.

[0100] Feedback mechanism and abnormal reporting: After identifying abnormal diagnosis and treatment behavior, the system will automatically generate a report and provide it to medical management personnel or the expert team responsible for infection control. The report content may include: Patient information: patient ID, hospitalization department, main diagnosis, etc. Diagnosis and treatment behavior: abnormal medication records, laboratory results, historical records of etiology examinations, etc. Abnormal type: Indicate whether it is "drug use does not match laboratory test results" or "there are positive reports but frequent negative tests."

[0101] The system can also provide a feedback mechanism, allowing clinicians to review the system's recognition results after checking the report. If it is confirmed that a certain behavior meets medical standards, the system can mark it as a reasonable diagnosis and treatment behavior to reduce false positives in similar situations in the future.

[0102] As long as one of the above two conditions is met, it can be considered that abnormal diagnosis and treatment behavior exists. Detailed information on medication is shown in Table 1.

[0103] Table 1. Detailed information on medications Types of anti-infective drugs Normal application scenarios antibiotic Positive for any one of the antibiotic-treated pathogens such as bacteria, Mycoplasma pneumoniae, Chlamydia pneumoniae, and Rickettsia Antifungal drugs Positive results of fungal tests Antiviral drugs Virus-related test results are positive Anti-TB drugs Mycobacterium-related test results are positive Antimalarial drugs Plasmodium related test results are positive Antiparasitic drugs Positive for any one of the pathogens that can be treated with anthelmintics, such as echinococcosis, lung fluke, and schistosomiasis S150, determining the suspected case level according to the suspected case and the abnormal diagnosis and treatment behavior; wherein the suspected case level includes typical case, highly suspected case, moderately suspected case and slightly suspected case; Among them, typical cases are cases with particularly abnormal diagnosis and treatment behaviors (meeting conditions one and two of unknown cause, and the items submitted for condition two of unknown cause cover common pathogen examinations, and one or more uncommon pathogen examinations are submitted for examination) and the patients are suspected cases; highly suspected cases are cases with abnormal diagnosis and treatment behaviors (meeting conditions one and two of unknown cause) and the patients are suspected patients; moderately suspected cases meet the conditions that the patient's diagnosis and treatment behaviors are generally abnormal or the patient is one of the conditions that the patient is a suspected patient (meeting condition one of unknown cause, but only one of condition two of unknown cause and suspected patients is met); mildly suspected cases are cases that only meet condition one of unknown cause and the patients are not suspected patients.

[0104] Here is a detailed explanation: Unknown cause Condition 1: The medication does not match the laboratory test results; Condition 2 of the unknown etiology: Frequent respiratory pathogen testing with negative results despite positive reports; Suspected cases: Patients can only be identified as suspected cases when all the above four screening strategies (infection indications, unknown etiology (with pathogen examination), poor treatment effect and infectiousness) are met.

[0105] The above are the abnormal diagnosis and treatment behaviors and suspected cases. Among them. Abnormal diagnosis and treatment behavior one (condition one of unknown cause) is the basic condition for determining suspected patients. On this basis, the four screening strategies for suspected patients and abnormal diagnosis and treatment behavior two (condition two of unknown cause) are used as auxiliary conditions for determining the level of cases. If abnormal diagnosis and treatment behaviors one and two are met, and the items sent for examination under condition two of unknown cause cover common pathogen examinations, and one or more uncommon pathogen examinations are sent for examination, and the patient meets the four screening strategies for suspected cases, then the patient is judged to be a typical case; if abnormal diagnosis and treatment behaviors one and two are met, and the patient meets the four screening strategies for suspected cases, then the patient is judged to be a highly suspected case; if abnormal diagnosis and treatment behavior one is met, but abnormal diagnosis and treatment behavior two or the four screening conditions for suspected cases are met, then the patient is judged to be a moderate suspected case; a mild suspected case is a case that only meets condition one of unknown cause and the patient is not a suspected patient. Simply put: cause one + cause two (and covering at least one unconventional pathogenic examination) + suspected case = typical case; cause one + cause two + suspected case = highly suspected case; cause one + cause two / suspected case = moderately suspected case; only cause one = mildly suspected case.

[0106] Routine pathological examinations include virus-related examinations, bacteria-related examinations, fungi-related examinations, tuberculosis-related examinations, and atypical pathogen-related examinations.

[0107] Non-routine pathogen testing includes NGS external laboratory sequencing, etc.

[0108] Bacteria-related tests include blood (such as blood culture, for the diagnosis of sepsis), urine (for urinary tract infection), throat swabs (for pharyngitis, tonsillitis), sputum (for respiratory tract infections, such as pneumonia), wound secretions or pus (for skin or soft tissue infections), and cerebrospinal fluid (for the diagnosis of meningitis).

[0109] Virus-related testing uses PCR, antigen detection or culture methods to determine whether there is a viral infection. Common samples for testing include: blood (for testing HIV, hepatitis B, hepatitis C, etc.), throat swabs or nasal swabs (for testing influenza, COVID-19, RSV, etc.), urine (for testing certain viruses, such as adenovirus), and cerebrospinal fluid (for the diagnosis of viral meningitis).

[0110] Fungal related tests are mainly used to diagnose infections caused by fungi, such as Candida, Aspergillus, etc. Common samples for testing include: skin, nails, hair (such as mold infection), oral or pharyngeal secretions (such as Candida), and blood (such as systemic fungal infection).

[0111] Tuberculosis-related tests include tuberculosis-infected T cell testing, tuberculosis culture, etc.

[0112] Atypical pathogens sent for testing include Mycoplasma pneumoniae, Chlamydia pneumoniae, etc.

[0113] The automated process consists of three main steps: Real-time data collection: The system collects patients' examination records, physical sign data, laboratory test results and condition change records from the hospital database regularly or in real time.

[0114] Rule engine processing: The system automatically evaluates each patient's test submission behavior and condition changes through the constructed rule engine. Through SQL query or rule detection, the system can automatically determine the suspected level of each patient.

[0115] Dynamic update: As the patient's condition changes, the system will dynamically update their suspected level. For example, if the patient's test behavior or condition changes significantly, the system will automatically adjust their suspected level.

[0116] Data collection is to extract the inspection behavior data and disease change data from the data that has been collected.

[0117] Inspection behavior data: including the type of items submitted for inspection, type of etiological examination (common or uncommon pathogens), inspection time and frequency, etc.

[0118] Data on changes in condition: including the patient's vital signs (such as body temperature, blood pressure, etc.), changes in condition level (whether transferred to the ICU), laboratory test results, and imaging examination results.

[0119] The system needs to use a rule engine to evaluate the patient's test submission behavior and condition changes to determine their suspected level. Each level can be divided into clear rules or conditions, and the system automatically executes the following process.

[0120] Automatic identification of typical cases Conditions: The test submission behavior is particularly abnormal (the test items cover common pathogen tests, and one or more uncommon pathogen tests are submitted) and the patient's condition is abnormal.

[0121] Abnormal detection of sample submission behavior: The system automatically analyzes the sample submission records to determine whether common pathogenic examinations (such as bacterial culture, viral PCR testing, etc.) are covered, and further checks whether uncommon pathogens (such as rare viruses, parasites, fungi, etc.) are submitted for examination.

[0122] Rule 1: If the patient's test record contains uncommon pathogen tests and the types of common pathogen tests are wide enough, the test submission behavior is judged to be particularly abnormal.

[0123] Abnormal condition detection: The system analyzes the patient's physical signs and condition data to see if there are any abnormal changes in the condition, such as a sharp deterioration in the condition, abnormal body temperature, or imaging results showing serious lesions.

[0124] Rule 2: If there are major abnormalities in the condition, the system will mark it as an abnormal condition.

[0125] Automated judgment: When the above two conditions are met, the system automatically classifies the patient as a typical case.

[0126] Automatic identification of highly suspected cases Conditions: Cases where both the behavior of submitting for examination and the patient's condition are abnormal.

[0127] Abnormal test behavior detection: The system uses similar rules as typical cases, but does not require uncommon pathogen tests to be submitted. As long as the patient's test submission behavior is relatively frequent or covers multiple common pathogens, and the test submission behavior is abnormally frequent, it can be determined that the test submission behavior is abnormal.

[0128] Abnormal condition detection: Detect whether the change in condition is abnormal (similar to typical cases).

[0129] Automated judgment: If both the test submission behavior and the changes in the condition are abnormal, the system will mark the patient as a highly suspected case.

[0130] Automatic identification of moderately suspected cases Conditions: The case meets the basic requirements for submission for examination, and either the submission behavior or the patient's condition is abnormal.

[0131] Basic test submission requirements: The system checks whether the patient has completed basic etiological tests, such as bacterial culture, common virus testing, etc. If these basic requirements are met, the system will further check the test submission behavior and changes in the condition.

[0132] Abnormal inspection behavior detection: The system determines whether the inspection behavior is abnormal (same as above).

[0133] Abnormal condition detection: The system detects whether there are significant changes in the condition (same as above).

[0134] Automated judgment: If either the patient's test submission behavior or the change in condition is abnormal, the system will mark him or her as a moderate suspected case.

[0135] Automatic identification of mild suspected cases Conditions: Cases that meet basic requirements for testing, but the testing behavior and patient condition do not meet the abnormal standards.

[0136] Basic testing requirements: Same as for moderate suspected cases, check whether basic etiological testing has been completed.

[0137] Inspection behavior detection: Determine whether the inspection behavior has not reached the abnormal standard (the inspection items and frequency are within a reasonable range).

[0138] Detection of condition changes: Determine whether the condition changes have not reached the abnormal standard (the patient's physical signs and condition have no significant abnormalities).

[0139] Automated judgment: If the patient meets the basic requirements for examination, but the examination behavior and changes in the condition do not meet the abnormal standards, the system will mark the patient as a mild suspected case.

[0140] S160. Obtain investigation data related to suspected cases.

[0141] In this embodiment, the investigation data related to the suspected case includes the contact history information of the suspected case, etc.

[0142] S170. Determine whether the suspected case is a cluster case based on the investigation data related to the suspected case.

[0143] In this embodiment, the system first needs to collect relevant data of the patient, especially information that can reflect interpersonal contact and spatial association. The collected data includes the following categories: Epidemiological survey data: community-related information, patients’ family members, neighbors, contacts, etc.

[0144] Ward records: patient admission information, hospitalization time, bed usage records, ward number, etc.

[0145] Medical staff contact records: including the list of medical staff who had contact with the patient, the time of contact and the duration of contact.

[0146] Contact information of other hospitalized patients: Information on other hospitalized patients who are related to the suspected case in time and space (such as patients in the same ward, patients who have used the same bed, etc.).

[0147] The automated identification of clusters can be divided into community clusters, nosocomial cluster 1 (patients infecting healthcare workers), and nosocomial cluster 2 (patients infecting other hospitalized patients). These clusters are identified based on contact records by building a network analysis model.

[0148] Community cluster case screening: For community cluster infections, the system needs to establish a relationship network between patients and their contacts through data analysis, especially epidemiological survey data. The analysis content is as follows: Building a community relationship network: Based on the community epidemiological survey form filled out by the patient when admitted to the hospital, information on family members, relatives, friends, neighbors, roommates, etc. was collected.

[0149] Construct a community contact network, using nodes in graph theory to represent patients and their contacts, and edges to represent contacts or epidemiological correlations.

[0150] Prioritization: First priority: Family members (such as couples, parents, siblings, etc.), who have close contact and a higher probability of infection.

[0151] Second priority: neighbors, roommates, residents of the same building, community or street, who have a relatively low probability of infection.

[0152] Data analysis methods: Network analysis tools were used to generate the patient's contact network, and clustering algorithms were used to perform community infection cluster analysis on the patients.

[0153] Community infection risk assessment: By calculating the contact frequency and intensity in the network, the transmission path and risk of infection can be determined, especially those with the closest contact (highly central nodes in the network).

[0154] Automatically determine community cluster cases: The system analyzes the contact network of each suspected case to determine whether there are multiple cases sharing the same community contacts or close family members. If the number of cases in a community exceeds the preset threshold, the system will automatically mark the community as a community cluster of cases.

[0155] Screening for hospital cluster cases 1 (patients infecting medical staff). For hospital clusters 1, the system needs to identify whether there is medical staff infection by tracking the contact records of medical staff who have contact with patients. The system can use time and contact frequency data for analysis.

[0156] Building a relationship network: The contact network between patients and medical staff was established using a patient-medical staff contact matrix, in which rows represent patients, columns represent medical staff, and the values ​​in the matrix represent the number of contacts or the duration of contacts.

[0157] Data analysis methods: Use time series analysis to model the contact records of medical staff, track the time points of infection cases, and detect whether multiple medical staff are infected due to contact with the same patient within a certain period of time.

[0158] Frequent pattern mining: Use association rule learning algorithms (such as the Apriori algorithm) to analyze whether there are certain medical staff who have been diagnosed as infected after contacting a specific patient.

[0159] Automatic determination of hospital cluster cases 1: The system automatically determines whether any medical staff have been infected due to contact with a patient by analyzing the infection records of medical staff and combining them with their contact information with patients. If the number of infected medical staff exceeds the preset threshold, the system will automatically mark it as a hospital cluster case1.

[0160] Screening of nosocomial cluster cases 2 (patients infecting other inpatients). For nosocomial clusters 2, the system mainly determines whether there is transmission between patients by analyzing the spatial and temporal contact records between patients and other inpatients.

[0161] Spatial-temporal contact analysis: The time-space association matrix was constructed using the patient's hospitalization records (such as bed number, ward number, and hospitalization date) to record whether each patient was in the same space as other patients (such as the same ward, using the same bed, etc.) during a certain period of time.

[0162] Data analysis methods: Use spatial analysis algorithms (such as heat maps and kernel density estimation) to analyze the spatial contacts between different patients and identify patient groups with denser spatial contacts.

[0163] Through time window analysis, the risk of mutual infection between patients in the same ward or bed is evaluated, especially those who have stayed in the same bed before and after.

[0164] Automatic determination of hospital cluster cases 2: The system determines whether there is transmission between patients by analyzing their contact with each other in the same ward or bed. If the number of patients infected in the same area exceeds the preset threshold, the system will automatically mark it as a hospital cluster case2.

[0165] Automated screening process for cluster cases 1. Data collection and preprocessing: The system regularly collects patients’ community contact information, medical contact information, hospitalization time and ward records from the hospital information system and performs preprocessing.

[0166] 2. Relationship network construction and cluster analysis: The system constructs the patient's contact network (community, medical staff, and inpatients) through network analysis methods, and uses clustering algorithms to automatically analyze the data.

[0167] 3. Cluster case identification and threshold detection: The system determines whether there are cluster cases by detecting the number of contacts in the contact network, and classifies community cluster cases and hospital cluster cases.

[0168] 4. Early warning and report generation: When the system detects that the number of clustered cases in a certain area, ward or community exceeds the set threshold, it automatically generates an early warning report to prompt managers to take control measures in a timely manner.

[0169] S180. If there are clustered cases among the suspected cases, determine the transmission route of the infectious disease and conduct case management.

[0170] In this embodiment, a transmission chain of suspected cases is established, and the transmission route of the unexplained infectious disease is obtained based on the level and clustering relationship of the suspected cases, so as to obtain the transmission route of the infectious disease and perform case management. The details are as follows: Data collection and preprocessing: First, the system needs to automatically collect and integrate data related to case spread from the hospital's information system. The data types include: Suspected case information: the level of each suspected case (mild, moderate, severe, typical case), diagnosis data, and physical sign data.

[0171] Epidemiological survey data: including the patient’s community contact information, family members, address, contact list, etc.

[0172] Hospitalization data: wards, beds, length of hospital stay, transfer records, etc., reflecting the spatial and temporal contact between cases.

[0173] Medical staff contact records: List of medical staff who have come into contact with the patient and the time of contact.

[0174] Laboratory test and imaging examination results: including pathogen detection (positive or negative results).

[0175] Data preprocessing: Clean and standardize the data to remove redundant data and ensure the consistency of patient information.

[0176] Synchronize all time data related to case contact and transmission (such as contact time, hospitalization time, etc.) to a unified timeline to facilitate the analysis of the transmission chain.

[0177] Establishing transmission chains of suspected cases: The transmission chain of suspected cases is constructed by analyzing the time, space and contact relationships between cases. The system can use network analysis technology to construct the transmission path between cases, track the contact source of each suspected case, and finally confirm the transmission route.

[0178] Constructing a case transmission network: Nodes: Each case (confirmed or suspected) is regarded as a node, which contains the case's personal information, suspected level, diagnosis results, etc.

[0179] Edges: The contact relationship between cases is an edge, reflecting the contact or transmission possibility between two cases. The weight of the edge can represent the intensity or duration of the contact. The more frequent the contact, the greater the weight.

[0180] By constructing a transmission network, the system can clearly visualize the transmission paths between suspected cases.

[0181] Network analysis methods: Contact analysis: The system analyzes spatial contact, temporal overlap, and other potential transmission pathways between cases based on the patient's epidemiological survey data, hospitalization records, and medical contact records.

[0182] Community contact chain: By analyzing community contact relationships (family members, neighbors, friends, etc.), the system can establish the transmission chain of suspected cases in the community.

[0183] In-hospital contact chains: By analyzing the contacts between hospitalized patients and medical staff, the system can determine whether there are transmission chains between patients or between patients and medical staff within the hospital.

[0184] Spatiotemporal contact matrix: Using the patient's hospital stay and ward data, a spatiotemporal contact matrix is ​​constructed to analyze whether different patients are in the same ward or use the same bed during the same period of time. This helps identify potential transmission chains.

[0185] Propagation path algorithm: Use the shortest path algorithm or time series network analysis to calculate the shortest propagation path between suspected cases to determine the order of transmission and possible transmission chains.

[0186] Automated identification of transmission chains: Based on the above analysis results, the system automatically generates the transmission chain of each suspected case. By traversing the transmission network, the system can mark the contact source of each suspected case and determine the upper-level case infected by it (that is, the source of infection that directly transmitted to the case).

[0187] The transmission chain is classified according to the level of the case, starting from highly suspected cases or typical cases and extending down to moderate and mild suspected cases. As the system traverses the transmission chain, it records the transmission path of each node.

[0188] Identify case zero and transmission routes: Patient zero refers to the first person infected with an infectious disease and the starting point of the entire transmission chain. By analyzing the transmission chain, the system can trace back to the first case of an infectious disease and confirm its transmission route.

[0189] Confirmed case zero: Transmission path backtracking: The system finds the starting node in the transmission network by backtracking the transmission path, that is, the case that has contact with other cases but has not been infected by other cases. This node is marked as case zero.

[0190] For example, in community transmission, case zero may be the first confirmed patient who has multiple contacts with other cases in the community.

[0191] The basis for confirming case zero: The patient's onset time is the earliest.

[0192] Other cases who had contact with the patient were confirmed or suspected to be infected in subsequent periods of time.

[0193] Patient zero may be a case imported from the community or the first case admitted to the hospital and confirmed.

[0194] Determination of transmission routes: Transmission path assessment: After confirming case zero, the system automatically generates a transmission route map by analyzing the contact chain between cases, indicating the source and destination of each case.

[0195] The system can display the chronological order of transmission through a time-series diagram, and mark the status of each case (suspected, confirmed) and the direction of transmission through color or graphics.

[0196] Update of transmission route map: The system dynamically updates the transmission route map based on the information of new cases. If new transmission paths or contacts are found, the system will automatically adjust the transmission chain.

[0197] Case Management System Integration: After confirming the transmission route and patient zero, all suspected and confirmed cases will be included in the case management system for subsequent tracking, treatment and implementation of prevention and control measures.

[0198] Case management features: Case file creation: The system creates an electronic file for each case, including its disease progression, epidemiological investigation information, hospitalization records and transmission chain information.

[0199] Case tracking: The system continuously tracks the case's condition changes, treatment effects and contact information through the case management module.

[0200] Separate management: For suspected cases that have been in contact with confirmed cases, the system can automatically generate separation recommendations and notify management personnel to take separation measures for relevant patients and contacts.

[0201] Automated alerts and reporting: The system will automatically generate a transmission report based on the dynamic changes of the transmission chain, which will include information such as the transmission path, case zero, and high-risk contacts in the transmission chain.

[0202] Early warning mechanism: The system can set thresholds. For example, when the number of confirmed cases or high-risk contacts in the transmission chain reaches a certain scale, it will automatically issue an early warning to managers and recommend more stringent prevention and control measures.

[0203] When the number of community cluster cases and hospital cluster cases exceeds the preset threshold, a suspected case transmission chain is established, and the transmission route of the unexplained infectious disease is obtained based on the suspected case level and cluster relationship. Figure 5 .

[0204] Additionally, case management includes: Implement separation: Enable clinical reminders to ensure timely separation.

[0205] Conduct epidemiological surveys: The system automatically generates epidemiological survey forms.

[0206] Organize in-hospital expert group consultation: the system automatically generates a consultation application form.

[0207] Separation of contacts: If the consultation determines that the infection is of unknown cause and may be contagious, contacts need to be separated. The system can provide partial contact information based on hospitalization data.

[0208] Reporting and filing: The system provides reporting materials for confirmed cases.

[0209] Material sending: After confirmation by experts, the above materials will be sent to the responsible physician of the suspected case and the relevant ward terminal.

[0210] Specifically, it includes: Push warning: The system automatically sends warning information to the responsible physicians of suspected cases and the ward area terminals to ensure that relevant personnel are aware of the epidemic situation in a timely manner, strictly implement separation measures, and make prevention and control preparations in advance.

[0211] Epidemiological survey form generation: Based on the patient's clinical data, the system automatically generates an epidemiological survey form, which will cover the hospital and community transmission routes of the new infectious disease, helping to track the transmission chain and risk assessment.

[0212] At the same time, the system provides a time-space route map of each suspected case in the hospital, making it easier for medical staff to identify potential contacts.

[0213] Expert consultation notice: Within the campus, the system sends meeting notifications to hospital experts and organizes expert consultations to assess case situations and develop corresponding treatment plans.

[0214] The system automatically generates consultation application forms for suspected cases to ensure the accuracy and timeliness of information transmission.

[0215] Contact list generation: Based on the transmission route of suspected cases, the system generates a list of patients who have been in contact with the suspected cases to provide support for early prevention and control. This process helps to quickly identify and separate potentially infected patients and reduce the risk of in-hospital transmission.

[0216] Medical port establishment and information sending: Establish an information receiving mechanism at the work ports of each medical staff in the hospital to ensure that the above materials and warning information can be quickly sent to the terminal of each medical staff to improve response efficiency.

[0217] Case report card generation and reporting: Generate case report cards for suspected patients and summarize relevant information, including patient basic information, clinical manifestations, epidemiological investigation results, etc.

[0218] Finally, the case report card will be reported to the regional epidemic prevention and control platform to facilitate higher-level epidemic monitoring and management.

[0219] Through this series of measures, medical institutions are able to identify and control epidemics at an early stage, reduce the risk of transmission, and improve overall prevention and control capabilities.

[0220] By establishing the transmission chain of suspected cases, potential sources of infection can be quickly identified, and separation measures can be taken to reduce the risk of transmission; the system automatically generates epidemiological survey forms, which simplifies the process and improves the efficiency and accuracy of the investigation; organizing in-hospital expert consultations helps to formulate more reasonable treatment plans and improve patient treatment outcomes; the system can quickly generate a list of contacts, which helps to carry out prevention and control in advance and reduce the possibility of hospital infection; through the information receiving mechanism of the medical port, it ensures that early warnings and related materials are quickly conveyed and the response speed to the epidemic is improved; case report cards are generated and information summary is standardized, which helps with regional epidemic monitoring and management and improves overall prevention and control capabilities.

[0221] In short, these measures will help improve the response speed and efficiency of medical institutions in dealing with epidemics, thereby effectively controlling the spread of infectious diseases.

[0222] S190. If there is no clustered case among the suspected cases, case management will be performed on the suspected cases.

[0223] The details of case management in step S190 are consistent with those in step S180 above, and will not be repeated here.

[0224] For example: obtain hospitalization process information A, antibiotic use information B, etiology examination information C, imaging examination information D, epidemiological investigation information F, transfer information G, and laboratory test information J; obtain the patient list from the hospitalization process information A, and retrieve the patient's clinical data in sequence; retrieve the antibiotic use information B to determine whether antibiotics are used. If so, proceed to the next step, otherwise return to the previous step; retrieve the patient's etiology examination information C to determine whether etiology examinations are prescribed. If so, proceed to the next step, otherwise return to "retrieve patient clinical data in sequence"; from the data in the hospitalization process information A, antibiotic use information B, and imaging examination information D, determine whether the patient has one of the following conditions: antibiotic adjustment, abnormal fever, change in disease level, abnormal imaging examination. If yes, go to the next step, otherwise return to "retrieve patient clinical data in order"; retrieve antimicrobial drug use information B, laboratory test information J, and etiology examination information C to determine whether the patient has one of the following conditions: that is, there is no corresponding positive result, but the clinician is using a certain anti-infective drug; in the case of a positive report, respiratory tract-related etiology examinations are still frequently sent for inspection and the test results are all negative; the inspection items cover common etiology examinations, and one or more uncommon etiology examinations are sent for inspection. Go to the next step; comprehensively consider the patient's condition and the abnormal inspection behavior, and judge the level of suspected cases, which are divided into typical cases> severe cases> moderate cases> mild cases. Go to the next step; retrieve the suspected case hospitalization process data information A, epidemiological survey data F, and transfer department information G, obtain the list of patients with intention to contact, count the number of suspected case patients in their relationship network, and determine whether it exceeds the preset threshold; if it exceeds, go to the next step, otherwise return to "retrieve patient clinical data in order"; according to the suspected case level and epidemiological survey data F, establish a spatiotemporal relationship network, obtain the patient's transmission route and trace the source, that is, case zero, and confirm case zero. Go to the next step; send warnings, transmission route maps, symptom information and other data to the responsible physician terminal and regional terminal to implement intervention separation measures. Go to the next step; provide suspected case information, pre-fill the reporting card, and report to the designated platform with one click after review.

[0225] The above-mentioned emerging infectious disease monitoring and early warning method based on multi-point triggering obtains clinical data from medical institutions, sets screening strategies for emerging infectious diseases, and identifies suspected cases; uses big data analysis to identify patients' abnormal diagnosis and treatment behaviors; determines the case level based on suspected cases and their abnormal behaviors, and obtains relevant investigation data; determines whether there are clustered cases, and if so, determines the transmission route and conducts case management; realizes real-time monitoring and early warning of infectious disease outbreaks and other public health events, and improves the efficiency and accuracy of handling emerging infectious diseases.

[0226] Figure 6 FIG. 3 is a schematic block diagram of an emerging infectious disease monitoring and early warning device 300 based on multi-point triggering provided by an embodiment of the present invention. Figure 6 As shown, corresponding to the above-mentioned emerging infectious disease monitoring and early warning method based on multi-point triggering, the present invention also provides an emerging infectious disease monitoring and early warning device 300 based on multi-point triggering. The emerging infectious disease monitoring and early warning device 300 based on multi-point triggering includes a unit for executing the above-mentioned emerging infectious disease monitoring and early warning method based on multi-point triggering, and the device can be configured in a server. Specifically, please refer to Figure 6 The emerging infectious disease monitoring and early warning device 300 based on multi-point triggering includes a clinical data acquisition unit 301, a setting unit 302, a case identification unit 303, a behavior identification unit 304, a level determination unit 305, a survey data acquisition unit 306, a judgment unit 307, a route determination unit 308 and a management unit 309.

[0227] The clinical data acquisition unit 301 is used to acquire the clinical data of patients from the HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; the setting unit 302 is used to set the screening strategy for emerging infectious diseases; the case identification unit 303 is used to identify suspected cases based on the patient clinical data; the behavior identification unit 304 is used to identify the abnormal diagnosis and treatment behavior of the patient from the patient clinical data based on big data analysis; the level determination unit 305 is used to determine the level of the suspected case based on the suspected case and the abnormal diagnosis and treatment behavior; the investigation data acquisition unit 306 is used to acquire the investigation data related to the suspected case; the judgment unit 307 is used to judge whether the suspected case has a cluster case based on the investigation data related to the suspected case; the route determination unit 308 is used to determine the infectious disease transmission route and perform case management if the suspected case has a cluster case. The management unit 309 is used to perform case management on the suspected case if the suspected case does not have a cluster case.

[0228] In one embodiment, if Figure 7 As shown, the case identification unit 303 includes a mapping subunit 3031 and a screening subunit 3032 .

[0229] The mapping subunit 3031 is used to map the characteristics of the emerging infectious disease with the patient clinical data to obtain a mapping result; the screening subunit 3032 is used to screen out patient information that meets the screening strategy from the mapping result to obtain suspected cases.

[0230] In one embodiment, the behavior recognition unit 304 is used to filter out the patient diagnosis and treatment data that meets the condition of unknown cause of disease from the patient clinical data based on big data analysis to obtain abnormal diagnosis and treatment behavior.

[0231] In one embodiment, the route determination unit 308 is used to establish a transmission chain of suspected cases, obtain the transmission route of the unidentified infectious disease based on the level and clustering relationship of the suspected cases, so as to obtain the transmission route of the infectious disease and perform case management.

[0232] It should be noted that technical personnel in the relevant field can clearly understand that the specific implementation process of the above-mentioned emerging infectious disease monitoring and early warning device 300 based on multi-point triggering and each unit can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and conciseness of the description, it will not be repeated here.

[0233] The above-mentioned emerging infectious disease monitoring and early warning device 300 based on multi-point triggering can be implemented in the form of a computer program. The computer program can be used in Figure 8 Runs on the computer device shown.

[0234] See also Figure 8 , Figure 8 5 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a server, wherein the server may be an independent server or a server cluster composed of multiple servers.

[0235] See also Figure 8 The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0236] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, and when the program instructions are executed, the processor 502 can execute a method for monitoring and early warning of emerging infectious diseases based on multi-point triggering.

[0237] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0238] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a method for monitoring and early warning of emerging infectious diseases based on multi-point triggering.

[0239] The network interface 505 is used to communicate with other devices over the network. Figure 8The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0240] The processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps: Obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; set screening strategies for emerging infectious diseases; identify suspected cases based on the patient clinical data; identify patients' abnormal diagnosis and treatment behaviors from the patient clinical data based on big data analysis; determine the level of suspected cases based on the suspected cases and the abnormal diagnosis and treatment behaviors; obtain relevant investigation data on suspected cases; determine whether the suspected cases have clustered cases based on the relevant investigation data on suspected cases; if the suspected cases have clustered cases, determine the infectious disease transmission route and conduct case management.

[0241] Among them, the screening strategy includes the characteristics of emerging infectious diseases, and the characteristics of emerging infectious diseases include symptoms of infection but unclear pathogens, unsatisfactory treatment effects, and epidemiological correlations between cases.

[0242] In one embodiment, after implementing the step of determining whether the suspected case is a cluster case according to the investigation data related to the suspected case, the processor 502 further implements the following steps: If there is no clustered case for the suspected case, case management will be carried out for the suspected case.

[0243] In one embodiment, when the processor 502 implements the step of identifying a suspected case according to the patient's clinical data, the processor 502 specifically implements the following steps: The characteristics of the emerging infectious disease are mapped with the clinical data of the patient to obtain a mapping result; and patient information that meets the screening strategy is screened out from the mapping result to obtain suspected cases.

[0244] In one embodiment, when the processor 502 implements the step of identifying the abnormal diagnosis and treatment behavior of the patient from the patient's clinical data based on big data analysis, the processor 502 specifically implements the following steps: Based on big data analysis, the patient's diagnosis and treatment data that meets the condition of unknown etiology is screened out from the patient's clinical data to obtain abnormal diagnosis and treatment behavior; Among them, the conditions with unknown causes include: Except for preventive medication, medication is inconsistent with laboratory test results; Even when there were positive reports, clinicians still sent samples for respiratory-related pathogen tests, and the results were all negative.

[0245] The suspected case levels include typical cases, highly suspected cases, moderately suspected cases and mildly suspected cases; typical cases are cases in which the diagnosis and treatment behaviors are particularly abnormal and the patients are suspected cases; highly suspected cases are cases in which the diagnosis and treatment behaviors are abnormal and the patients are suspected patients; moderately suspected cases meet one of the conditions that the patient's diagnosis and treatment behaviors are generally abnormal or the patient is a suspected patient; mildly suspected cases only meet the condition of unknown cause and the patient is not a suspected patient.

[0246] In one embodiment, the processor 502 implements the following steps when implementing the steps of determining the infectious disease transmission route and performing case management: Establish the transmission chain of suspected cases, obtain the transmission route of the unidentified infectious disease based on the level and clustering relationship of suspected cases, obtain the transmission route of the infectious disease, and conduct case management.

[0247] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0248] It can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiment of the above method.

[0249] Therefore, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps: Obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; set screening strategies for emerging infectious diseases; identify suspected cases based on the patient clinical data; identify patients' abnormal diagnosis and treatment behaviors from the patient clinical data based on big data analysis; determine the level of suspected cases based on the suspected cases and the abnormal diagnosis and treatment behaviors; obtain relevant investigation data on suspected cases; determine whether the suspected cases have clustered cases based on the relevant investigation data on suspected cases; if the suspected cases have clustered cases, determine the infectious disease transmission route and conduct case management.

[0250] Among them, the screening strategy includes the characteristics of emerging infectious diseases, and the characteristics of emerging infectious diseases include symptoms of infection but unclear pathogens, unsatisfactory treatment effects, and epidemiological correlations between cases.

[0251] In one embodiment, after executing the computer program to implement the step of determining whether the suspected case exists as a cluster case based on the investigation data related to the suspected case, the processor further implements the following step: if the suspected case does not exist as a cluster case, case management is performed on the suspected case.

[0252] In one embodiment, when the processor executes the computer program to implement the step of identifying suspected cases based on the patient's clinical data, it specifically implements the following steps: mapping the characteristics of the emerging infectious disease with the patient's clinical data to obtain a mapping result; and screening out patient information that meets the screening strategy from the mapping result to obtain suspected cases.

[0253] In one embodiment, when the processor executes the computer program to implement the step of identifying the abnormal diagnosis and treatment behavior of the patient from the patient's clinical data based on big data analysis, the processor specifically implements the following steps: based on big data analysis, the diagnosis and treatment data of the patient that meets the condition of unknown cause is screened from the patient's clinical data to obtain the abnormal diagnosis and treatment behavior; Among them, the condition of unknown cause includes: after preventive medication, the medication is inconsistent with the laboratory test results; in the case of a positive report, the clinician still sends for respiratory tract related pathogen tests and the results are all negative. The levels of suspected cases include typical cases, highly suspected cases, moderately suspected cases and mildly suspected cases; typical cases are cases with particularly abnormal diagnosis and treatment behaviors and the patients are suspected cases; highly suspected cases are cases with abnormal diagnosis and treatment behaviors and the patients are suspected patients; moderately suspected cases meet one of the conditions that the patient's diagnosis and treatment behaviors are generally abnormal or the patient is a suspected patient; mildly suspected cases only meet the condition of unknown cause one and the patient is not a suspected patient.

[0254] In one embodiment, when the processor executes the computer program to implement the determination of the infectious disease transmission route and perform case management steps, it specifically implements the following steps: establishing a suspected case transmission chain, obtaining the unknown infectious disease transmission route based on the suspected case level and clustering relationship, so as to obtain the infectious disease transmission route, and perform case management.

[0255] The storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

[0256] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for monitoring and early warning of emerging infectious diseases based on multi-point triggering, characterized in that: include: Obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; Setting screening strategies for emerging infectious diseases; Identify suspected cases based on the patient's clinical data; Identifying abnormal diagnosis and treatment behaviors of patients from the clinical data of the patients based on big data analysis; The level of suspected cases is determined according to the suspected cases and the abnormal diagnosis and treatment behaviors; wherein the suspected case levels include typical cases, highly suspected cases, moderately suspected cases and mildly suspected cases; typical cases are cases with particularly abnormal diagnosis and treatment behaviors and the patients are suspected cases; highly suspected cases are cases with abnormal diagnosis and treatment behaviors and the patients are suspected patients; moderately suspected cases meet one of the conditions that the patient's diagnosis and treatment behaviors are generally abnormal or the patient is a suspected patient; mildly suspected cases meet only the condition of unknown cause one and the patient is not a suspected patient; the condition of unknown cause one is that the medication does not match the laboratory test results; Obtain investigation data related to suspected cases; Determine whether there are clustered cases of the suspected cases based on the investigation data related to the suspected cases; If there are clusters of suspected cases, the transmission route of the infectious disease will be determined and case management will be carried out.

2. The emerging infectious disease monitoring and early warning method based on multi-point triggering according to claim 1 is characterized in that: After determining whether the suspected case is a cluster case according to the investigation data related to the suspected case, the method further includes: If there is no clustered case for the suspected case, case management will be carried out for the suspected case.

3. The emerging infectious disease monitoring and early warning method based on multi-point triggering according to claim 1 is characterized in that: The screening strategy includes the characteristics of emerging infectious diseases, including symptoms of infection but unclear causative pathogens, unsatisfactory treatment effects, and epidemiological correlations between cases.

4. The emerging infectious disease monitoring and early warning method based on multi-point triggering according to claim 3 is characterized in that: The identifying of suspected cases according to the patient's clinical data comprises: Mapping the characteristics of the emerging infectious disease with the patient clinical data to obtain a mapping result; The patient information that meets the screening strategy is screened out from the mapping results to obtain suspected cases.

5. The emerging infectious disease monitoring and early warning method based on multi-point triggering according to claim 4 is characterized in that: The identifying the abnormal diagnosis and treatment behavior of the patient from the patient's clinical data based on big data analysis includes: Based on big data analysis, the patient's diagnosis and treatment data that meets the condition of unknown etiology is screened out from the patient's clinical data to obtain abnormal diagnosis and treatment behavior; Among them, the conditions with unknown causes include: Except for preventive medication, medication is inconsistent with laboratory test results; Even when there were positive reports, clinicians still sent samples for respiratory-related pathogen tests, and the results were all negative.

6. The emerging infectious disease monitoring and early warning method based on multi-point triggering according to claim 1 is characterized in that: Determine the transmission route of infectious diseases and conduct case management, including: Establish the transmission chain of suspected cases, obtain the transmission route of the unidentified infectious disease based on the level and clustering relationship of suspected cases, obtain the transmission route of the infectious disease, and conduct case management.

7. The emerging infectious disease monitoring and early warning device based on multi-point triggering is characterized by: include: Clinical data acquisition unit, used to obtain patient clinical data from HIS, LIS, PACS, surgery, and electronic medical records of medical institutions; The setting unit is used to set the screening strategy for emerging infectious diseases; A case identification unit, used to identify suspected cases based on the patient's clinical data; A behavior recognition unit, used to identify abnormal diagnosis and treatment behaviors of the patient from the patient's clinical data based on big data analysis; A level determination unit is used to determine the level of the suspected case according to the suspected case and the abnormal diagnosis and treatment behavior; wherein the suspected case level includes typical cases, highly suspected cases, moderately suspected cases and mildly suspected cases; wherein a typical case is a case in which the test submission behavior is particularly abnormal and the patient's condition is abnormal; a highly suspected case is a case in which both the test submission behavior and the patient's condition are abnormal; a moderately suspected case is a case that meets the basic test submission requirements and either the test submission behavior or the patient's condition is abnormal; a mildly suspected case is a case that meets the basic test submission requirements but the test submission behavior and the patient's condition do not meet the abnormal standards; An investigation data acquisition unit, used to obtain investigation data related to suspected cases; A judgment unit, used to judge whether the suspected case is a cluster case according to the investigation data related to the suspected case; The route determination unit is used to determine the route of infectious disease transmission and perform case management if there are clustered cases among the suspected cases.

8. A computer device, characterized in that: The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 6 when executing the computer program.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images