An infectious disease early warning method and device based on electronic medical records

Abstract

The present application relates to disease early warning technology, the present application provides an infectious disease early warning method, device and storage medium based on electronic medical record, wherein the method comprises: obtaining the electronic medical record text of patients in each medical institution; extracting the disease symptoms of each patient from each electronic medical record text in turn; determine whether each disease symptom meets the preset infectious disease symptom condition; if yes, the patient corresponding to the disease symptom is recorded as a case of infectious disease, and the early warning level of the case of infectious disease is determined, different infectious disease types correspond to different early warning levels; early warning is carried out according to the early warning level corresponding to the case. In this way, the cases meeting the infectious disease symptoms can be monitored in time through the electronic medical record text, and then early warning is carried out, which is fast and efficient.

Description

Technical Field

[0001] This invention relates to the field of disease early warning technology, and in particular to a method, device and readable storage medium for early warning of infectious diseases based on electronic medical records. Background Technology

[0002] Currently, disease control centers across the country primarily monitor infectious diseases by having staff at medical institutions and monitoring points (schools, communities, etc.) identify cases and then report them regularly. However, this current model is generally plagued by significant reporting delays, underreporting, and misreporting. For example, doctors may diagnose a patient with a non-infectious disease or fever, but staff may report all such cases. In densely populated areas, it is also necessary to manually analyze and determine whether there has been contact between cases reported by multiple medical institutions. This results in staff spending a significant amount of time screening and analyzing high-risk cases, which is not only inefficient but also wasteful of resources. Summary of the Invention

[0003] The main objective of this invention is to provide a method, device, and storage medium for early warning of infectious diseases based on electronic medical records, aiming to solve the technical problem of time-consuming and labor-intensive screening of infectious disease cases in the prior art.

[0004] Based on the above-mentioned objectives, this invention proposes a method for early warning of infectious diseases based on electronic medical records, comprising:

[0005] Obtain electronic medical record texts of patients from various medical institutions;

[0006] Extract the symptoms of each patient from each of the aforementioned electronic medical record texts in sequence;

[0007] Determine whether the symptoms of each disease meet the preset criteria for infectious disease symptoms;

[0008] If so, the patient corresponding to the symptoms is recorded as a case of infectious disease, and the warning level of infectious disease for the case is determined. Different types of infectious diseases correspond to different warning levels.

[0009] Warnings will be issued based on the warning level corresponding to the case.

[0010] Furthermore, the step of sequentially extracting the patient's symptoms from each of the electronic medical record texts includes:

[0011] The electronic medical record text is preprocessed, and the preprocessing steps include word segmentation of the electronic medical record text and filtering of interference words according to a preset stop word list.

[0012] The preprocessed electronic medical record text is used to extract keywords through a trained extraction model. The keywords are the symptoms of the disease.

[0013] The further step of determining whether each of the disease symptoms meets the preset infectious disease symptom conditions includes:

[0014] The symptoms of each disease are compared with the symptoms of a variety of pre-set infectious diseases to determine whether the symptoms are positive symptoms of the corresponding infectious diseases.

[0015] If the symptoms are positive, then the patient is deemed to meet the criteria for symptoms of the infectious disease.

[0016] If it is not a symptom of the infectious disease, then determine whether the symptom of the disease includes the necessary symptoms of the infectious disease. The necessary symptoms are those that are extracted from the symptoms of various infectious diseases and have common characteristics.

[0017] If the necessary symptoms are included, the condition is determined to meet the criteria for infectious disease symptoms; otherwise, the symptoms are recorded as target symptoms, and it is determined whether the target symptoms include secondary symptoms.

[0018] If the secondary symptoms are present, then a target patient matching the target symptoms is identified from the patients, and the patient information of the target patient is obtained, including spatial and temporal activity information.

[0019] Based on the patient information, determine whether the spatial and temporal activity trajectories of each target patient overlap;

[0020] If the symptoms overlap, the condition is determined to meet the criteria for the infectious disease symptoms.

[0021] The further step of classifying patients corresponding to the symptoms of the disease as cases of infectious diseases includes:

[0022] The patient's information is sent to the terminal of the designated verification personnel;

[0023] Obtain the feedback information input by the verification personnel, wherein the feedback information is the feedback information after manual verification;

[0024] When the opinion information indicates that the patient's symptoms are incorrect, the patient's electronic medical record text is corrected based on the input information from the verification personnel.

[0025] When the information provided is accurate regarding the patient's symptoms, the patient is recorded as a case of infectious disease.

[0026] The steps for further determining the warning level of the infectious disease in the case include:

[0027] Obtain the disease and symptom information of the case to determine the type of disease;

[0028] Determine whether the disease type of the case is a special disease for single-case early warning;

[0029] If so, obtain the warning level information corresponding to the specific disease;

[0030] If not, obtain information on patients with the same symptoms as the case to determine the number of cases of the same infectious disease;

[0031] Based on the patient information, determine the number of new cases within a preset time period in a designated area, and calculate a warning value based on the number of new cases.

[0032] And determine the corresponding warning level based on the warning value;

[0033] The warning value can be calculated using the following formula:

[0034]

[0035] Where, x j The number of cases on day j within the period is represented by μ, the average number of cases in the same period in the past is represented by S, and the deviation of the number of cases in the budget period from the number of cases in the same period in the past is represented by S. When the deviation exceeds a preset threshold, it is recorded as the warning value.

[0036] Following the step of issuing a warning based on the warning level corresponding to the case, the following steps are further included:

[0037] Obtain case information of cases with the same symptoms, including information on residential address and work address;

[0038] Based on case information, GIS technology was used to mark geospatial heatmaps of cases;

[0039] A monitoring chart is generated based on the labeled information. The monitoring chart includes the number of cases, the geographical distribution of cases, and the trend changes.

[0040] The further steps of issuing an early warning based on the warning level include:

[0041] Warnings will be issued via system alerts, SMS alerts, and automated phone alerts.

[0042] The present invention also provides an infectious disease early warning device based on electronic medical records, included in the first terminal, the device comprising:

[0043] The text acquisition unit is used to acquire the electronic medical record texts of patients seeking medical treatment at various medical institutions.

[0044] The symptom extraction unit is used to sequentially extract the symptoms of each patient from each of the electronic medical record texts;

[0045] The judgment condition unit is used to determine whether the symptoms of each disease meet the preset infectious disease symptom conditions;

[0046] The judgment level unit is used to determine that when the preset infectious disease symptom conditions are met, the patient corresponding to the symptom is recorded as an infectious disease case, and the infectious disease warning level of the case is determined. Different infectious disease types correspond to different warning levels.

[0047] The level-based early warning unit is used to issue an early warning based on the warning level corresponding to the case.

[0048] Furthermore, the symptom extraction unit includes:

[0049] The text processing subunit is used to preprocess the electronic medical record text. The preprocessing steps include word segmentation of the electronic medical record text and filtering out interference words according to a preset stop word list.

[0050] The symptom extraction subunit is used to extract keywords from the preprocessed electronic medical record text using a trained extraction model. The keywords are the symptoms of the disease.

[0051] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method for early warning of infectious diseases based on electronic medical records.

[0052] The beneficial effects of this invention are as follows: cases that meet the symptoms of infectious diseases can be monitored in a timely manner through electronic medical record texts, and then early warnings can be issued quickly and efficiently. Furthermore, the potential transmission relationships and geographical distribution between cases can be analyzed, thereby improving the response speed of infectious disease prevention and control and avoiding the occurrence of mass infection events. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the steps of an infectious disease early warning method based on electronic medical records in one embodiment of the present invention;

[0054] Figure 2 This is a schematic block diagram of an infectious disease early warning device based on electronic medical records in one embodiment of the present invention;

[0055] Figure 3 This is a schematic block diagram of the structure of a storage medium in one embodiment of the present invention.

[0056] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0057] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0058] Reference Figure 1 The infectious disease early warning method based on electronic medical records in this embodiment includes:

[0059] Step S1: Obtain the electronic medical record texts of patients seeking medical treatment at various medical institutions;

[0060] Step S2: Extract the symptoms of each patient from each of the electronic medical record texts in sequence;

[0061] Step S3: Determine whether the symptoms of each disease meet the preset criteria for infectious disease symptoms;

[0062] Step S4: If so, the patient corresponding to the symptoms is recorded as a case of infectious disease, and the warning level of the infectious disease of the case is determined. Different types of infectious diseases correspond to different warning levels.

[0063] Step S5: Issue an early warning based on the warning level corresponding to the case.

[0064] As described in step S1 above, the above method can be applied to various disease control centers. The above electronic medical record texts are medical records recorded after a patient visits a medical institution and is examined by a doctor. Each electronic medical record text includes patient information, such as patient name, gender, place of illness, time of consultation, name of disease, symptoms of disease, and institution of consultation. By accessing the system interface of each medical institution, the electronic medical record texts of each patient in each medical institution are obtained. Then, the electronic medical record texts are processed to extract the patient's symptoms in order to monitor whether the patient has an infectious disease, thereby providing early warning and taking preventive measures in advance.

[0065] As described in step S2 above, the symptoms of each patient are extracted sequentially from the electronic medical record text of each patient in order to determine whether they are positive symptoms of an infectious disease. Specifically, this extraction can be performed using NLP technology. In one embodiment, step S2 includes:

[0066] Step S21: Preprocess the electronic medical record text. The preprocessing steps include segmenting the electronic medical record text into words and filtering out interference words according to a preset stop word list.

[0067] Step S22: Extract keywords from the preprocessed electronic medical record text using a trained extraction model. The keywords are the symptoms of the disease.

[0068] In this embodiment, the electronic medical record text is first preprocessed. The preprocessing includes word segmentation, part-of-speech tagging, and named entity recognition. Then, interference words in the electronic medical record text are filtered according to a preset stop word list. Furthermore, vocabulary normalization can be performed. Then, the preprocessed electronic medical record text is processed by a trained extraction model to extract the corresponding keywords. The keywords here are the symptoms of the disease. The extraction model is a model trained on a training set based on a preset algorithm. The preset algorithm may include the TF-IDF algorithm and the LDA algorithm, which will not be described in detail here.

[0069] As described in step S3 above, it is determined whether the symptoms of each disease meet the preset conditions for infectious disease symptoms. If they do not meet the conditions, no warning is required. In this embodiment, the determination of whether the symptoms meet the conditions for infectious disease symptoms can be made through the specific manifestations of the symptoms. Specifically, in one embodiment, step S3 above includes:

[0070] Step S31: Compare each of the described symptoms with a set of preset symptoms of multiple infectious diseases to determine whether the symptoms are positive symptoms of the corresponding infectious disease.

[0071] Step S32: If the symptoms are positive, then the condition for infectious disease symptoms is determined.

[0072] Step S33: If it is not a symptom of the infectious disease, then determine whether the symptom of the disease includes the necessary symptoms of the infectious disease. The necessary symptoms are those that are extracted from the symptoms of multiple infectious diseases and have common characteristics.

[0073] Step S34: If the necessary symptoms are included, it is determined that the symptoms meet the criteria for infectious disease symptoms; otherwise, the symptoms are recorded as target symptoms, and it is determined whether the target symptoms include secondary symptoms.

[0074] Step S35: If the secondary symptoms are present, find the target patient who matches the target symptoms from the patients, and obtain the patient information of the target patient, which includes spatial and temporal activity information;

[0075] Step S36: Determine whether the spatial and temporal activity trajectories of each target patient overlap based on the patient information;

[0076] Step S37: If there is an overlap, it is determined that the symptoms meet the criteria for the infectious disease.

[0077] In this embodiment, the symptoms of each disease are first compared with the symptoms of various infectious diseases to determine whether the symptoms are positive symptoms of the corresponding infectious disease. These infectious diseases are pre-defined known diseases, such as influenza, scarlet fever, typhoid fever, hepatitis E, dysentery, rubella, plague, cholera, severe acute respiratory syndrome (SARS), pneumonia of unknown cause, human infection with highly pathogenic avian influenza, diphtheria, filariasis, measles, etc. When the symptoms match the pre-defined infectious disease symptoms, the disease is deemed to meet the criteria for infectious disease symptoms. Each symptom is compared to identify the corresponding infectious disease case. When the symptoms do not match the pre-defined infectious disease symptoms, it indicates that the patient does not have a known infectious disease. In this case, it is first determined whether the symptoms contain the necessary symptoms of an infectious disease. Necessary symptoms are those extracted from the symptoms of various infectious diseases and possessing common characteristics; these are the symptoms that confirm the disease as an infectious disease. It can also be determined that the symptoms meet the criteria for infectious diseases. If the symptoms do not include the essential symptoms, it is then determined whether they include secondary symptoms. For ease of distinction, the above symptoms are referred to as target symptoms, and the secondary symptoms are suspected symptoms of infectious diseases. When secondary symptoms are included, it can be further determined whether the disease is infectious by looking at multiple patients with the same symptoms. At this time, target patients with the same target symptoms can be found among all patients. Then, the patient information of these target patients is obtained. The patient information includes spatial and temporal activity information, such as records of residential address, work address, and places visited, as well as the corresponding activity time. Then, based on this patient information, it is determined whether the spatial and temporal activity trajectories of these target patients overlap. If they overlap, it means that the two have been in contact. If both have the same suspected infectious disease symptoms and have been in contact, then it can be determined that the symptoms meet the criteria for infectious diseases.

[0078] As described in steps S4-S5 above, when the symptoms meet the criteria for infectious disease symptoms, patients with these symptoms can be recorded as cases of the corresponding infectious disease. For example, patients with influenza symptoms are recorded as influenza cases, patients with scarlet fever symptoms are recorded as scarlet fever cases, and patients with pneumonia of unknown cause are recorded as cases of unknown cause. Different levels of warnings can be issued for cases with different symptoms. Because different infectious diseases have different severity, different types of infectious diseases correspond to different warning levels. Warnings can be issued based on the warning level corresponding to the case, such as Level 1, Level 2, Level 3, etc. Specifically, warnings can be issued through system message reminders, SMS reminders, or telephone reminders, or through voice broadcasts.

[0079] In one embodiment, step S4 above includes:

[0080] Step S41: Send the patient's information to the terminal of the designated verification personnel;

[0081] Step S42: Obtain the opinion information input by the verification personnel, wherein the opinion information is the opinion information after manual verification;

[0082] Step S43: When the opinion information indicates that the patient's symptoms are incorrect, the patient's electronic medical record text is corrected based on the input information from the verification personnel;

[0083] Step S44: When the opinion information is correct regarding the patient's symptoms, the patient is recorded as a case of infectious disease.

[0084] To avoid errors in early warnings, a manual correction method is also provided. In this embodiment, once the symptoms are determined to match those of an infectious disease, verification can be performed. First, the patient's information is sent to the terminal of a designated verifier. After receiving the information, the verifier performs manual verification, such as reviewing medical records or consulting the corresponding doctor. After verification, the verifier can input the verification result and provide feedback through the terminal. This feedback information includes whether the patient's symptoms are incorrect or correct. If the feedback indicates that the patient's symptoms are incorrect, the patient's electronic medical record is corrected based on the verifier's input. If the feedback indicates that the patient's symptoms are correct, the patient is recorded as a case of an infectious disease. Furthermore, manual verification or no manual verification can be selected for different types of infectious diseases, depending on the actual situation.

[0085] In one embodiment, step S4 above includes:

[0086] Step S401: Obtain the disease symptom information of the case to determine the type of disease in the case;

[0087] Step S402: Determine whether the disease type of the case is a special disease for single-case early warning;

[0088] Step S403: If yes, obtain the warning level information corresponding to the specific disease;

[0089] Step S404: If not, obtain patient information with the same symptoms as the case to determine the number of cases of the same infectious disease;

[0090] Step S405: Determine the number of patients who have increased within a preset time period in a designated area based on the patient information, and calculate the warning value based on the number of patients.

[0091] Step S406: Determine the corresponding warning level based on the warning value.

[0092] In this embodiment, the symptoms of the cases are obtained to determine the type of disease. Since different diseases have different degrees of severity, for example, for highly severe infectious diseases, a warning is needed when a single case is found. However, for less severe infectious diseases, in order to avoid wasting resources, a warning can be issued only when the number of cases or the warning value reaches a certain level. Therefore, it is first determined whether the disease of the case is a special disease for single-case warning, such as plague, cholera, severe acute respiratory syndrome (SARS), pneumonia of unknown cause, human infection with highly pathogenic avian influenza, diphtheria, filariasis, measles, etc. If so, the warning level table is retrieved. This warning level table is used to represent the relationship between different infectious diseases and their corresponding warning levels. The corresponding warning level information is obtained through the warning level table, and then a warning is issued based on the warning level.

[0093] When the disease is not a specific disease for which a single case warning is issued, information on patients with the same symptoms can be obtained to determine the number of cases of the same infectious disease. Simultaneously, the number of cases increasing within a specified area over a predetermined time period can be obtained, such as the number of cases in the area for a week and the number of cases each day. Then, a warning value is calculated based on these numbers. Specifically, this can be achieved by obtaining the number of cases and the number of cases each day within a specified time period in the specified area, such as within 3 years or 2 years, and comparing the number of cases within that week with the same period in previous years to determine if the warning value has been exceeded. This warning value can be the average of the same period in previous years, or it can be calculated using a preset formula.

[0094] Where, x j This represents the number of cases on day j within the period, μ represents the average number of cases in the same period in the past, and S represents the deviation of the number of cases in the budget period from the number of cases in the same period in the past. When the deviation exceeds a preset threshold, it is recorded as a warning value. Different warning values ​​correspond to different levels.

[0095] In one embodiment, after step S5, the following is included:

[0096] Step S6: Obtain case information of cases with the same symptoms, including residential address and work address information;

[0097] Step S7: Based on the case information, use GIS technology to mark the geospatial heatmap of the case;

[0098] Step S8: Generate a monitoring chart based on the annotation information. The monitoring chart includes the number of cases, the geographical distribution of cases, and the trend changes.

[0099] In this embodiment, GIS (Geographic Information Systems) technology can also be used to analyze the distribution and development trend of cases. First, case information of cases with the same symptoms is obtained, namely the aforementioned patient information, specifically including residential address, work address, short-term activity time and trajectory. Then, GIS technology is used to geospatially label the cases, marking the geographical location of the onset of the cases on the GIS map. GIS technology is an existing technology, and the specific labeling process will not be described in detail here. Then, monitoring charts are generated based on the labeling information and the number of cases counted. These charts specifically include the number of cases, the geographical distribution map of cases, and the trend change map, etc., and these monitoring charts can be displayed on a display screen.

[0100] This allows for timely monitoring of cases exhibiting symptoms of infectious diseases via electronic medical records, enabling rapid and efficient early warning. Furthermore, it allows for analysis of potential transmission relationships and geographic spatial distribution among cases, thereby improving the response speed for infectious disease prevention and control and preventing mass infection events.

[0101] Reference Figure 2 This embodiment provides an infectious disease early warning device based on electronic medical records. This device corresponds to the aforementioned infectious disease early warning method based on electronic medical records, and includes:

[0102] Text Acquisition Unit 1 is used to acquire the electronic medical record texts of patients seeking medical treatment at various medical institutions.

[0103] Symptom extraction unit 2 is used to sequentially extract the symptoms of each patient from each of the electronic medical record texts;

[0104] Judgment condition unit 3 is used to determine whether the symptoms of each disease meet the preset infectious disease symptom conditions;

[0105] The judgment level unit 4 is used to determine that when the preset infectious disease symptom conditions are met, the patient corresponding to the symptom is recorded as an infectious disease case, and the infectious disease warning level of the case is determined. Different infectious disease types correspond to different warning levels.

[0106] Level warning unit 5 is used to issue a warning based on the warning level corresponding to the case.

[0107] As described in the above-mentioned text acquisition unit 1, the above method can be applied to various disease control centers. The above-mentioned electronic medical record text is the medical record recorded by the doctor after the patient seeks medical treatment at the medical institution. Each electronic medical record text includes patient information, such as patient name, gender, place of disease occurrence, time of consultation, disease name, disease symptoms, and institution of treatment. By accessing the system interface of each medical institution, the electronic medical record text of each patient in each medical institution is obtained. Then, the electronic medical record text is processed to extract the patient's symptoms in order to monitor whether the patient has an infectious disease, thereby providing early warning and taking preventive measures in advance.

[0108] As described in the above-mentioned symptom extraction unit 2, the symptoms of each patient are extracted sequentially from the electronic medical record text of each patient in order to determine whether they are positive symptoms of an infectious disease. Specifically, the extraction can be performed using NLP technology. In one embodiment, the above-mentioned symptom extraction unit 2 includes:

[0109] The text processing subunit is used to preprocess the electronic medical record text. The preprocessing steps include word segmentation of the electronic medical record text and filtering out interference words according to a preset stop word list.

[0110] The symptom extraction subunit is used to extract keywords from the preprocessed electronic medical record text using a trained extraction model. The keywords are the symptoms of the disease.

[0111] In this embodiment, the electronic medical record text is first preprocessed. The preprocessing includes word segmentation, part-of-speech tagging, and named entity recognition. Then, interference words in the electronic medical record text are filtered according to a preset stop word list. Furthermore, vocabulary normalization can be performed. Then, the preprocessed electronic medical record text is processed by a trained extraction model to extract the corresponding keywords. The keywords here are the symptoms of the disease. The extraction model is a model trained on a training set based on a preset algorithm. The preset algorithm may include the TF-IDF algorithm and the LDA algorithm.

[0112] As described in the judgment condition unit 3 above, it determines whether the symptoms of each disease meet the preset infectious disease symptom conditions. If they do not meet the conditions, no warning is required. In this embodiment, the determination of whether the symptoms meet the infectious disease symptom conditions can be made through the specific symptom manifestations. Specifically, in one embodiment, the judgment condition unit 3 includes:

[0113] The comparison symptom subunit is used to compare each of the disease symptoms with a variety of preset infectious disease symptoms in order to determine whether the disease symptoms are positive symptoms of the corresponding infectious disease.

[0114] The determination condition subunit is used to determine that if the positive symptom is present, the symptom meets the criteria for the infectious disease symptom.

[0115] The necessary subunit is used to determine whether the symptoms of the disease contain necessary symptoms of the infectious disease if the symptoms are not symptoms of the infectious disease. The necessary symptoms are symptoms that are extracted from the symptoms of various infectious diseases and have common characteristics.

[0116] The secondary subunit is used to determine whether the infectious disease symptom condition is met if the necessary symptom is included; otherwise, the symptom is recorded as the target symptom, and it is determined whether the target symptom includes secondary symptoms.

[0117] An information acquisition subunit is configured to, if the secondary symptoms are present, identify a target patient from the patients whose symptoms match the target symptoms, and acquire patient information of the target patient, the patient information including spatial and temporal activity information;

[0118] The overlapping sub-unit is determined to determine whether the spatial and temporal activity trajectories of each target patient overlap based on the patient information.

[0119] The determination of the conformity sub-unit is used to determine whether the symptoms of the infectious disease are met if an overlap is determined.

[0120] In this embodiment, the symptoms of each disease are first compared with the symptoms of various infectious diseases to determine whether the symptoms are positive symptoms of the corresponding infectious disease. These infectious diseases are pre-defined known diseases, such as influenza, scarlet fever, typhoid fever, hepatitis E, dysentery, rubella, plague, cholera, severe acute respiratory syndrome (SARS), pneumonia of unknown cause, human infection with highly pathogenic avian influenza, diphtheria, filariasis, measles, etc. When the symptoms match the pre-defined infectious disease symptoms, the disease is deemed to meet the criteria for infectious disease symptoms. Each symptom is compared to identify the corresponding infectious disease case. When the symptoms do not match the pre-defined infectious disease symptoms, it indicates that the patient does not have a known infectious disease. In this case, it is first determined whether the symptoms contain the necessary symptoms of an infectious disease. Necessary symptoms are those extracted from the symptoms of various infectious diseases and possessing common characteristics; these are the symptoms that confirm the disease as an infectious disease. It can also be determined that the symptoms meet the criteria for infectious diseases. If the symptoms do not include the essential symptoms, it is then determined whether they include secondary symptoms. For ease of distinction, the above symptoms are referred to as target symptoms, and the secondary symptoms are suspected symptoms of infectious diseases. When secondary symptoms are included, it can be further determined whether the disease is infectious by looking at multiple patients with the same symptoms. At this time, target patients with the same target symptoms can be found among all patients. Then, the patient information of these target patients is obtained. The patient information includes spatial and temporal activity information, such as records of residential address, work address, and places visited, as well as the corresponding activity time. Then, based on this patient information, it is determined whether the spatial and temporal activity trajectories of these target patients overlap. If they overlap, it means that the two have been in contact. If both have the same suspected infectious disease symptoms and have been in contact, then it can be determined that the symptoms meet the criteria for infectious diseases.

[0121] As described in the aforementioned judgment level unit 4 and level warning unit 5, when it is determined that the symptoms meet the criteria for infectious disease symptoms, patients with these symptoms can be recorded as cases of the corresponding infectious disease. For example, patients with influenza symptoms are recorded as influenza cases, patients with scarlet fever symptoms are recorded as scarlet fever cases, and patients with pneumonia of unknown cause are recorded as cases of unknown cause. Different levels of warnings can be issued for cases with different symptoms. Because different infectious diseases have different severity, different types of infectious diseases correspond to different warning levels. Warnings can be issued based on the warning level corresponding to the case, such as Level 1 warning, Level 2 warning, Level 3 warning, etc. Specifically, warnings can be issued through system-sent information reminders, SMS reminders, or telephone reminders, or through voice broadcasts.

[0122] In one embodiment, the aforementioned level determination unit 4 includes:

[0123] A transmitting terminal subunit is used to send the patient's information to the terminal of a designated verification personnel;

[0124] The opinion acquisition subunit is used to acquire the opinion information input by the verification personnel, wherein the opinion information is the opinion information after manual verification.

[0125] The medical record correction subunit is used to correct the patient's electronic medical record text based on the input information of the verification personnel when the opinion information indicates that the patient's symptoms are incorrect.

[0126] This is recorded as a case subunit, used to record the patient as a case of infectious disease when the opinion information is correct regarding the patient's symptoms.

[0127] To avoid errors in early warnings, a manual correction method is also provided. In this embodiment, once the symptoms are determined to match those of an infectious disease, verification can be performed. First, the patient's information is sent to the terminal of a designated verifier. After receiving the information, the verifier performs manual verification, such as reviewing medical records or consulting the corresponding doctor. After verification, the verifier can input the verification result and provide feedback through the terminal. This feedback information includes whether the patient's symptoms are incorrect or correct. If the feedback indicates that the patient's symptoms are incorrect, the patient's electronic medical record is corrected based on the verifier's input. If the feedback indicates that the patient's symptoms are correct, the patient is recorded as a case of an infectious disease. Furthermore, manual verification or no manual verification can be selected for different types of infectious diseases, depending on the actual situation.

[0128] In one embodiment, the aforementioned level determination unit 4 includes:

[0129] A disease seed unit is used to obtain the disease symptom information of the case in order to determine the disease type of the case;

[0130] The disease seed determination unit is used to determine whether the disease of the case is a special disease for single-case early warning;

[0131] The acquisition level sub-unit is used to determine that when the disease type of the case is a special disease for single-case warning, the warning level information corresponding to the special disease is acquired.

[0132] The patient sub-unit is used to determine the number of cases of the same infectious disease when the disease type of the case is not a special disease for single-case warning.

[0133] The calculation and early warning subunit is used to determine the number of increases within a preset time period in a designated area based on the patient information, and to calculate the early warning value based on the number.

[0134] A warning subunit is defined, which is used to determine the corresponding warning level based on the warning value.

[0135] In this embodiment, the symptoms of the cases are obtained to determine the type of disease. Since different diseases have different degrees of severity, for example, for highly severe infectious diseases, a warning is needed when a single case is found. However, for less severe infectious diseases, in order to avoid wasting resources, a warning can be issued only when the number of cases or the warning value reaches a certain level. Therefore, it is first determined whether the disease of the case is a special disease for single-case warning, such as plague, cholera, severe acute respiratory syndrome (SARS), pneumonia of unknown cause, human infection with highly pathogenic avian influenza, diphtheria, filariasis, measles, etc. If so, the warning level table is retrieved. This warning level table is used to represent the relationship between different infectious diseases and their corresponding warning levels. The corresponding warning level information is obtained through the warning level table, and then a warning is issued based on the warning level.

[0136] When the disease is not a specific disease for which a single case warning is issued, information on patients with the same symptoms can be obtained to determine the number of cases of the same infectious disease. Simultaneously, the number of cases increasing within a specified area over a predetermined time period can be obtained, such as the number of cases in the area for a week and the number of cases each day. Then, a warning value is calculated based on these numbers. Specifically, this can be achieved by obtaining the number of cases and the number of cases each day within a specified time period in the specified area, such as within 3 years or 2 years, and comparing the number of cases within that week with the same period in previous years to determine if the warning value has been exceeded. This warning value can be the average of the same period in previous years, or it can be calculated using a preset formula.

[0137] Where, x j This represents the number of cases on day j within the period, μ represents the average number of cases in the same period in the past, and S represents the deviation of the number of cases in the budget period from the number of cases in the same period in the past. When the deviation exceeds a preset threshold, it is recorded as a warning value. Different warning values ​​correspond to different levels.

[0138] In one embodiment, the above-mentioned apparatus includes:

[0139] The case acquisition unit is used to acquire case information of cases with the same symptoms, including information on residential address and work address;

[0140] Case unit annotation is used to annotate the geospatial heatmap of cases using GIS technology based on case information:

[0141] A monitoring unit is generated to generate monitoring charts based on the labeled information. The monitoring charts include the number of cases, the geographical distribution of cases, and the trend changes.

[0142] In this embodiment, GIS (Geographic Information Systems) technology can also be used to analyze the distribution and development trend of cases. First, case information of cases with the same symptoms is obtained, namely the aforementioned patient information, specifically including residential address, work address, short-term activity time and trajectory. Then, GIS technology is used to geospatially label the cases, marking the geographical location of the onset of the cases on the GIS map. GIS technology is an existing technology, and the specific labeling process will not be described in detail here. Then, monitoring charts are generated based on the labeling information and the number of cases counted. These charts specifically include the number of cases, the geographical distribution map of cases, and the trend change map, etc., and these monitoring charts can be displayed on a display screen.

[0143] This allows for timely monitoring of cases exhibiting symptoms of infectious diseases via electronic medical records, enabling rapid and efficient early warning. Furthermore, it allows for analysis of potential transmission relationships and geographic spatial distribution among cases, thereby improving the response speed for infectious disease prevention and control and preventing mass infection events.

[0144] refer to Figure 3 The present invention also provides a computer-readable storage medium 10, which stores a computer program 20, which, when run on a computer, causes the computer to execute the infectious disease early warning method based on electronic medical records described in the above embodiments.

[0145] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media.

[0146] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for early warning of infectious diseases based on electronic medical records, characterized in that, include: Obtain electronic medical record texts of patients from various medical institutions; Extract the symptoms of each patient from each of the aforementioned electronic medical record texts in sequence; Determine whether the symptoms of each disease meet the preset criteria for infectious disease symptoms; If so, the patient corresponding to the symptoms is recorded as a case of infectious disease, and the warning level of infectious disease for the case is determined. Different types of infectious diseases correspond to different warning levels. An early warning will be issued based on the warning level corresponding to the case. The step of sequentially extracting the patient's symptoms from each of the electronic medical record texts includes: The electronic medical record text is preprocessed, and the preprocessing steps include word segmentation of the electronic medical record text and filtering of interference words according to a preset stop word list. The preprocessed electronic medical record text is used to extract keywords through a trained extraction model, where the keywords are the symptoms of the disease. The step of determining whether each of the disease symptoms meets the preset infectious disease symptom conditions includes: The symptoms of each disease are compared with the symptoms of a variety of pre-set infectious diseases to determine whether the symptoms are positive symptoms of the corresponding infectious diseases. If the symptoms are positive, then the condition is determined to meet the criteria for symptoms of the infectious disease. If it is not a symptom of the infectious disease, then determine whether the symptom of the disease includes the necessary symptoms of the infectious disease. The necessary symptoms are those that are extracted from the symptoms of various infectious diseases and have common characteristics. If the necessary symptoms are included, the condition is determined to meet the criteria for infectious disease symptoms; otherwise, the symptoms are recorded as target symptoms, and it is determined whether the target symptoms include secondary symptoms. If the secondary symptoms are present, then a target patient matching the target symptoms is identified from the patients, and the patient information of the target patient is obtained, including spatial and temporal activity information. Based on the patient information, determine whether the spatial and temporal activity trajectories of each target patient overlap; If the symptoms overlap, the condition is determined to meet the criteria for the infectious disease symptoms.

2. The infectious disease early warning method based on electronic medical records according to claim 1, characterized in that, The step of classifying patients whose symptoms correspond to infectious diseases as cases includes: The patient's information is sent to the terminal of the designated verification personnel; Obtain the feedback information input by the verification personnel, wherein the feedback information is the feedback information after manual verification; When the opinion information indicates that the patient's symptoms are incorrect, the patient's electronic medical record text is corrected based on the input information from the verification personnel. When the information provided is accurate regarding the patient's symptoms, the patient is recorded as a case of infectious disease.

3. The infectious disease early warning method based on electronic medical records according to claim 1, characterized in that, The steps for determining the warning level of the infectious disease in the case include: Obtain the disease and symptom information of the case to determine the type of disease; Determine whether the disease type of the case is a special disease for single-case early warning; If so, obtain the warning level information corresponding to the specific disease; If not, obtain information on patients with the same symptoms as the case to determine the number of cases of the same infectious disease; Based on the patient information, determine the number of new cases within a preset time period in a designated area, and calculate a warning value based on the number of new cases. The corresponding warning level is determined based on the warning value. The warning value can be calculated using the following formula: ； in, This represents the number of cases on day j within the period. S represents the average number of cases in the same period in the past, and S represents the deviation of the number of cases in the budget period from the number of cases in the same period in the past. When the deviation exceeds a preset threshold, it is recorded as the warning value.

4. The infectious disease early warning method based on electronic medical records according to claim 1, characterized in that, Following the step of issuing a warning based on the warning level corresponding to the case, the following steps are included: Obtain case information of cases with the same symptoms, including residential address and work address; Based on case information, GIS technology was used to mark geospatial heatmaps of cases; A monitoring chart is generated based on the annotation information. The monitoring chart includes the number of cases, the geographical distribution of cases, and the trend changes.

5. The infectious disease early warning method based on electronic medical records according to claim 1, characterized in that, The steps for issuing an early warning based on the warning level include: Warnings will be issued via system alerts, SMS alerts, and automated phone alerts.

Images