Tuberculosis onset risk assessment method and model

Abstract

The invention discloses a tuberculosis onset risk assessment method and model, and relates to the technical field of medicines.The tuberculosis onset risk assessment method comprises the steps that tuberculosis epidemiological investigation is conducted, and the tuberculosis epidemiological investigation comprises collecting questionnaire data through screening infection detection, chest digital X-ray imaging and laboratory examination; based on the collected questionnaire data, in combination with the basic information, contact history, symptom expression and infection detection result of the investigated person, a primary database of tuberculosis risk assessment is constructed for the latent tuberculosis infected person. According to the tuberculosis onset risk assessment method and model provided by the invention, various epidemiological survey data, individual basic information, contact history, symptom expression and laboratory detection results are integrated, so that the tuberculosis onset risk of a latent infected person can be assessed more comprehensively and accurately, and misdiagnosis and missed diagnosis conditions are reduced; based on an accurate risk assessment result, high-risk crowds are identified more effectively, and corresponding prevention and control measures are taken.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a method and model for assessing the risk of tuberculosis. Background Art

[0002] Tuberculosis, a chronic infectious disease caused by Mycobacterium tuberculosis, remains a serious public health problem worldwide. Traditional tuberculosis diagnostic methods, such as sputum smear microscopy and recombinant tuberculosis fusion protein (EC) skin test, have certain diagnostic value, but have limitations such as low sensitivity and susceptibility to interference. Therefore, the development of a more accurate and comprehensive tuberculosis risk assessment method and model is of great significance for early detection, prevention and control of the spread of tuberculosis.

[0003] However, existing tuberculosis risk assessment methods often rely on single or multiple limited detection indicators, lack systematicity and comprehensiveness, and find it difficult to comprehensively assess an individual's risk of tuberculosis infection. In addition, there is a lack of scientific and systematic guidance for follow-up management and prevention measures for high-risk populations, resulting in limited tuberculosis prevention and control effects. In this regard, we proposed a tuberculosis incidence risk assessment method and model. Summary of the invention

[0004] In order to solve the above technical problems, a tuberculosis incidence risk assessment method and model are provided. This technical solution solves the above problems.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A tuberculosis risk assessment method and model, comprising:

[0007] Conduct epidemiological surveys on tuberculosis, including: infection testing and questionnaire data;

[0008] Based on the collected questionnaire data, combined with the respondents' basic information, contact history, symptoms and past medical history, a preliminary database for tuberculosis risk assessment was constructed;

[0009] Apply algorithms to process the preliminary database and establish a tuberculosis risk assessment model;

[0010] Using the established tuberculosis risk assessment model, the respondents were given tuberculosis risk scores, and were divided into different risk levels according to the risk scores;

[0011] Subsequent follow-up observation will be implemented for respondents with high risk levels to further confirm their tuberculosis infection status and take corresponding preventive measures.

[0012] Preferably, conducting an epidemiological survey on tuberculosis specifically includes:

[0013] The respondents were given a skin test of recombinant tuberculosis fusion protein (EC) to determine whether they were infected with Mycobacterium tuberculosis based on the test results; the respondents' basic information, contact history, past medical history, symptoms, etc. were collected. The collected data were checked and collated, and the accuracy and completeness of the data were checked.

[0014] Preferably, the method steps for constructing a preliminary database for tuberculosis risk assessment are:

[0015] The collected questionnaire data, basic information, contact history, symptoms and laboratory test results were classified and sorted, the answers to the questions in the questionnaire data were coded, the contact history was quantified, and the symptoms were coded according to the presence or absence of specific symptoms;

[0016] Standardize laboratory test results. For recombinant tuberculosis fusion protein (EC) skin test results, convert the nodule diameter into a specific risk level code. Set large and small thresholds for the nodule diameter based on experience. If it is below the small threshold, it is low risk. If it is between the small and large thresholds, it is medium risk. If it is greater than the large threshold, it is high risk. The digital chest X-ray imaging results are coded to distinguish between normal and abnormal after judgment by the doctor. The sputum smear examination and tuberculosis culture results are coded to distinguish between negative and positive.

[0017] Determine the fields of the database, the fields including: the basic information field of the investigated person, the contact history field, the symptom field and the laboratory test result field;

[0018] Use database management software to create tables and enter the sorted data into fields;

[0019] Enter the sorted data into the database and conduct a spot check on the entered data to check whether it is consistent with the original data.

[0020] Preferably, establishing a tuberculosis risk assessment model specifically includes:

[0021] The questionnaire data, basic information, contact history, symptom manifestations, and infection test results in the preliminary database were cleaned to remove outliers and missing values, and the data were standardized. The standardized calculation formula is:

[0022]

[0023] In the formula, represents the standardized value, X represents the original data, represents the mean of the data, Indicates the standard deviation of the data;

[0024] Analyze the correlation between various variables in the preliminary database and tuberculosis risk, select characteristic variables with high correlation with tuberculosis risk, and establish a risk assessment model;

[0025] Among them, the expression of the risk assessment model is:

[0026]

[0027] In the formula, Indicates the probability that the risk of tuberculosis is high, represents the characteristic variable, represents the model parameters;

[0028] The algorithm was trained with the selected feature variables as input and the tuberculosis risk level as output;

[0029] After training the established model with the training set, the model is evaluated with the test set. The evaluation indicators include accuracy, precision, recall, and F1 value.

[0030] Based on the model evaluation results, the model is optimized and the model building and evaluation steps are repeated until satisfactory model performance is obtained.

[0031] Preferably, the model is evaluated using a test set, and the evaluation indicators include accuracy, precision, recall, and F1 value, specifically including:

[0032] The calculation formula for accuracy is:

[0033]

[0034] In the formula, represents the accuracy, TP represents the true positive, that is, the number of samples that are actually high-risk and predicted by the model to be high-risk. represents true negative examples, that is, the number of samples that are actually low risk and predicted by the model to be low risk, represents false positives, that is, the number of samples that are actually low risk but predicted by the model to be high risk. represents false negative examples, that is, the number of samples that are actually high risk but predicted as low risk by the model;

[0035] The calculation formula of precision is:

[0036]

[0037] In the formula, Indicates accuracy;

[0038] The calculation formula of recall rate is:

[0039]

[0040] In the formula, represents the recall rate;

[0041] The calculation formula of F1 value is:

[0042] .

[0043] Preferably, the method for performing risk scoring and risk classification using the tuberculosis risk assessment model is:

[0044] Input the basic information, contact history, symptoms and laboratory test results of the subjects into the established tuberculosis risk assessment model;

[0045] Input the characteristics of the subjects into the established tuberculosis risk assessment model, including basic information, contact history, symptoms and laboratory test results;

[0046] The model calculates based on the input data and outputs a risk score value;

[0047] Based on expert experience and previous research results, the risk level division intervals are determined, and the risk scores of the respondents are classified into corresponding risk levels according to the determined division intervals.

[0048] Preferably, the risk level division interval is expressed as:

[0049] The distribution range of risk scores is determined by data analysis as follows: ;

[0050] The low risk range is set as:

[0051]

[0052] The medium risk interval is set as the risk score between the low risk interval and the high risk interval;

[0053] The high-risk range is set as:

[0054] .

[0055] Preferably, the method for implementing follow-up observation on high-risk level respondents is:

[0056] The high-risk respondents classified according to the tuberculosis risk assessment model were identified as the subjects of follow-up observation;

[0057] Follow-up visits were conducted at fixed intervals after the initial evaluation, and included symptom inquiry, physical examination, laboratory tests, and chest imaging examinations;

[0058] Based on the information obtained from the follow-up, the tuberculosis infection status of the subjects was comprehensively judged;

[0059] For patients diagnosed with tuberculosis, anti-tuberculosis treatment should be carried out in a timely manner and standardized management should be carried out according to the treatment plan. For respondents who have not been diagnosed but are still at high risk, preventive measures should be taken, including: health education, avoiding contact with sources of infection and enhancing immunity.

[0060] Preferably, when constructing a preliminary database, if there are ambiguous or uncertain answers in the questionnaire data, they should be verified and corrected through telephone follow-up or re-survey to ensure the reliability of the data.

[0061] Preferably, when selecting characteristic variables with a high correlation with tuberculosis risk, the stepwise regression method is used for variable screening, gradually introducing variables that contribute significantly to the model, while eliminating insignificant variables, so as to improve the simplicity and accuracy of the model.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] The tuberculosis risk assessment method and model proposed in the present invention can more comprehensively and accurately assess the risk of individual tuberculosis infected persons by integrating a variety of epidemiological survey data, individual basic information, contact history, symptom manifestations and laboratory test results, reduce misdiagnosis and missed diagnosis, and more effectively identify high-risk groups based on accurate risk assessment results, and take corresponding prevention and control measures, thereby reducing the incidence and transmission rate of tuberculosis and improving the level of public health. It can provide personalized risk assessment results and corresponding health management suggestions for the specific conditions of different individuals, which helps individuals to better understand their health conditions and take healthy lifestyle adjustments or medical intervention measures, provide new ideas and methods for research in tuberculosis-related fields, promote the in-depth development of scientific research, provide more scientific basis and technical support for the prevention, control and treatment of tuberculosis, and greatly improve the utilization efficiency of medical resources through an automated risk assessment model, reduce the time of manual judgment of doctors, and improve the accuracy and efficiency of diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0065] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0066] Reference Figure 1 As shown, a tuberculosis risk assessment method and model, including:

[0067] Conduct an epidemiological survey on tuberculosis. Covering multiple aspects, collect key information through screening and infection testing, make a preliminary judgment on the physical condition of the person being investigated to determine whether there is a possibility of tuberculosis infection. Collect more detailed and accurate questionnaire data through various means to provide a basis for subsequent analysis and evaluation.

[0068] Based on the questionnaire data, combined with the basic information, contact history, symptoms and infection test results of the respondents, a preliminary database for tuberculosis risk assessment was constructed. Basic information includes age, gender, occupation, etc., which provides background reference for the assessment. In terms of contact history, it is crucial to understand whether there has been any contact experience with tuberculosis patients and the extent and duration of contact, etc., in order to determine the risk of infection. Symptoms include coughing, sputum, fever, and fatigue, suggesting the possibility of tuberculosis. Infection test results provide objective data support. Integrate multi-faceted information to build a comprehensive preliminary database.

[0069] The algorithm is applied to process the preliminary database and establish a tuberculosis risk assessment model. Through the algorithm, a large amount of data in the database is analyzed and mined. Through in-depth processing, key features and patterns are extracted to establish a model for accurately assessing tuberculosis risk. The model comprehensively considers various factors according to the different situations of the respondents and gives accurate risk assessment results.

[0070] Using the established tuberculosis risk assessment model, the respondents were scored for tuberculosis risk and divided into different risk levels according to the risk scores. The risk scores were used to understand the risk level of latent tuberculosis infection. The respondents were divided into high, medium and low risk levels according to different scores. Targeted measures were taken for respondents of different risk levels.

[0071] Follow-up observation will be conducted for high-risk respondents to confirm the tuberculosis infection status and take preventive measures. Close follow-up observation will be conducted for high-risk individuals to confirm whether they have developed the disease through regular monitoring and examination.

[0072] High-risk individuals undergo regular digital chest X-rays. Chest X-rays are performed under medical equipment. After the scan is completed, the radiologist interprets the X-ray image. The doctor observes each area of ​​the lungs in the image to determine whether there are abnormal shadows in the lungs. Abnormal shadows include nodules, patchy shadows, cavities, etc., which are potential manifestations of tuberculosis.

[0073] Collect sputum samples from high-risk individuals. Cough up sputum under the guidance of professionals and collect it in a specific container. Subsequently, the sputum samples are subjected to sputum smear examination and tuberculosis culture. In the sputum smear examination, technicians use the Ziehl-Neelsen acid-fast staining method to observe whether there are acid-fast bacilli in the sputum under a microscope. Tuberculosis culture is to place the sputum sample in a specific culture medium, culture it for a period of time, and observe whether tuberculosis bacteria grow. These two examination methods can more accurately determine whether the person being investigated is diagnosed with pulmonary tuberculosis.

[0074] Patients diagnosed with tuberculosis should be treated, and those who have not developed the disease should take preventive measures. This can effectively control the spread of tuberculosis and improve public health.

[0075] This tuberculosis risk assessment method and model, by integrating multiple means and information, can more comprehensively and accurately assess individual tuberculosis infection risks, and provide support for tuberculosis prevention, control and treatment.

[0076] Conducting a tuberculosis epidemiological survey specifically includes:

[0077] The respondents were given a skin test of recombinant tuberculosis fusion protein (EC). Medical staff injected the recombinant tuberculosis fusion protein (EC) reagent into the skin on the palmar forearm of the respondents. The injection site reaction was checked 48 to 72 hours after the injection, and the transverse and longitudinal diameters of the red halo and nodules were measured and recorded in millimeters (mm). The larger the red halo or nodule, the more positive the reaction was. The average diameter of the reaction (the sum of the transverse and longitudinal diameters divided by 2) was not less than 5 mm, which indicated that the respondents were infected with Mycobacterium tuberculosis.

[0078] Verify and organize the collected data. Staff will check the data of the respondents and verify the accuracy and completeness of the data. Ensuring data accuracy can avoid misjudgment and analysis, and ensuring data integrity can make subsequent epidemiological investigations and analyses more comprehensive and in-depth. Verify and confirm questionable data to ensure the reliability and validity of the survey data.

[0079] The methodological steps for constructing a preliminary database for tuberculosis risk assessment are:

[0080] 1. Classification and Coding

[0081] The questionnaire data, basic information, contact history, symptoms and laboratory test results were sorted and organized. The answers to the questionnaire questions were coded, the contact history was quantified, and the symptoms were coded according to whether the respondents had specific symptoms.

[0082] 2. Standardization of infection test results

[0083] Standardize the laboratory test results. For the recombinant tuberculosis fusion protein (EC) skin test results, set a threshold for the size of the nodule diameter. If it is lower than the small threshold, it is coded as low risk, if it is between the two thresholds, it is coded as medium risk, and if it is greater than the large threshold, it is coded as high risk.

[0084] 3. Determine the database fields

[0085] Determine the database fields, including the basic information fields of the respondents (age, gender, occupation, etc.), contact history fields, symptom fields, and laboratory test results fields.

[0086] 4. Create tables and enter data

[0087] Use database management software to create tables and ensure that the structure is reasonable to accommodate the data in each field. Enter the sorted data by field and check the data to ensure accuracy and completeness.

[0088] V. Spot Check and Verification

[0089] After entering the data into the database, spot check is performed. Randomly select some data to check whether they are consistent with the original data. If they are inconsistent, verify and correct them to ensure data reliability. The constructed preliminary database for tuberculosis risk assessment provides the basis for subsequent risk assessment and analysis.

[0090] The establishment of a tuberculosis risk assessment model specifically includes:

[0091] The questionnaire data, basic information, contact history, symptom manifestations, and infection test results in the preliminary database were cleaned to remove outliers and missing values, and the data were standardized. The standardized calculation formula is:

[0092]

[0093] In the formula, represents the standardized value, X represents the original data, represents the mean of the data, Indicates the standard deviation of the data;

[0094] Analyze the correlation between various variables in the preliminary database and tuberculosis risk, select characteristic variables with high correlation with tuberculosis risk, and establish a risk assessment model;

[0095] Among them, the expression of the risk assessment model is:

[0096]

[0097] In the formula, Indicates the probability that the risk of tuberculosis is high, represents the characteristic variable, represents the model parameters;

[0098] The algorithm was trained with the selected feature variables as input and the tuberculosis risk level as output;

[0099] After training the established model with the training set, the model is evaluated with the test set. The evaluation indicators include accuracy, precision, recall, and F1 value.

[0100] Based on the model evaluation results, the model is optimized and the model building and evaluation steps are repeated until satisfactory model performance is obtained.

[0101] Use the test set to evaluate the model. The evaluation indicators include accuracy, precision, recall and F1 value. Specifically, they include:

[0102] The calculation formula for accuracy is:

[0103]

[0104] In the formula, represents the accuracy, TP represents the true positive, that is, the number of samples that are actually high-risk and predicted by the model to be high-risk. represents true negative examples, that is, the number of samples that are actually low risk and predicted by the model to be low risk, represents false positives, that is, the number of samples that are actually low risk but predicted by the model to be high risk. represents false negative examples, that is, the number of samples that are actually high risk but predicted as low risk by the model;

[0105] The calculation formula of precision is:

[0106]

[0107] In the formula, Indicates accuracy;

[0108] The calculation formula of recall rate is:

[0109]

[0110] In the formula, represents the recall rate;

[0111] The calculation formula of F1 value is:

[0112] .

[0113] The method for using the tuberculosis risk assessment model to score and classify risk levels is as follows:

[0114] The basic information, contact history, symptoms and infection test results of the subjects were input into the tuberculosis risk assessment model. Basic information includes age, gender, occupation, etc.; contact history includes whether there is any contact experience with tuberculosis patients and the frequency, duration and method of contact; symptoms include cough, sputum, fever, fatigue, night sweats and tuberculosis-related symptoms; infection test results are mainly recombinant tuberculosis fusion protein (EC) skin test results, etc.

[0115] The characteristic data of the subjects are input into the model. The characteristic data covers basic information, contact history, symptoms, and infection test results. By comprehensively considering multi-dimensional data, the model assesses the risk of tuberculosis infection of the subjects.

[0116] The model performs calculations based on the input data. During the calculation process, different feature data are assigned different weights based on internal algorithms and parameters, and a comprehensive analysis is performed. After the calculation, a risk score is output, reflecting the likelihood of the person being investigated being infected with tuberculosis.

[0117] Based on expert experience and research results, determine the risk level division interval. Experts determine scientific and accurate risk level division standards based on professional knowledge and clinical experience and summary and analysis of research results. According to the division interval, the risk score of the respondents is classified into the corresponding risk level, high risk, medium risk, and low risk. The risk of tuberculosis infection of the respondents is divided into different levels to provide guidance for follow-up observation and preventive measures.

[0118] The risk level division interval is expressed as:

[0119] The distribution range of risk scores is determined by data analysis as follows: ;

[0120] The low risk range is set as:

[0121]

[0122] The medium risk interval is set as the risk score between the low risk interval and the high risk interval;

[0123] The high-risk range is set as:

[0124] .

[0125] The method for implementing follow-up observation on high-risk respondents is as follows:

[0126] 1. Determine the subjects for follow-up observation

[0127] According to the tuberculosis risk assessment model, high-risk respondents are identified as follow-up observation subjects. High-risk respondents have a higher possibility of tuberculosis infection. Follow-up observation can confirm the infection status and take measures to control the spread and development of tuberculosis.

[0128] 2. Develop a follow-up plan

[0129] Follow-up visits at fixed intervals after the initial assessment can help monitor the health status of the person being investigated. Follow-up visits include symptom inquiries to understand whether there are new tuberculosis-related symptoms. Physical examinations can reveal abnormal signs. Laboratory tests cover inflammatory indicators and specific tests to provide data support for determining infection status. Chest imaging examinations can observe lung lesions and detect potential tuberculosis lesions.

[0130] 3. Confirmation of tuberculosis status

[0131] The tuberculosis infection status of the person under investigation is determined based on follow-up information. Patients with clear evidence of onset can be diagnosed with tuberculosis, and confirmed patients should receive anti-tuberculosis treatment and standardized management. Patients without clear evidence but with suspicious conditions can receive diagnostic treatment or close observation.

[0132] 4. Take preventive measures

[0133] Undiagnosed but high-risk persons under investigation need to take preventive measures. Health education can improve self-protection awareness and help people understand the transmission routes, symptoms and prevention methods. Avoid contact with infectious sources to reduce the risk of infection. Strengthening immunity, eating a proper diet, exercising moderately and getting enough rest can reduce the possibility of infection.

[0134] 5. Ensure data reliability

[0135] When constructing a preliminary database, any ambiguous or uncertain answers in the questionnaire should be verified and corrected through telephone interviews or re-investigations. Ensuring data reliability provides a basis for subsequent risk assessment and model building, and avoids erroneous assessment results and decisions.

[0136] 6. Variable Screening Method

[0137] The characteristic variables with high correlation with tuberculosis risk were selected and screened by stepwise regression method. Significant variables were gradually introduced and insignificant variables were eliminated to make the model accurately reflect the risk characteristics, improve simplicity, avoid irrelevant variables interfering with accuracy, and establish an efficient and accurate tuberculosis risk assessment model.

[0138] The use process of the present invention is: conduct a tuberculosis epidemiological survey, collect questionnaire data, construct a preliminary database for tuberculosis risk assessment, establish a tuberculosis risk assessment model, perform tuberculosis risk scoring and divide risk levels, follow up and observe high-risk individuals and take measures.

[0139] In summary, the advantages of the present invention are: accurately assessing the risk of latent tuberculosis infection in individuals by integrating multiple data and taking timely intervention; effectively identifying high-risk groups and taking measures to reduce the incidence and transmission rate; providing personalized assessment results and health management recommendations; providing new ideas and methods for tuberculosis research; and improving the efficiency of medical resource utilization.

[0140] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A tuberculosis risk assessment method and model, characterized in that: include: Conducting an epidemiological survey on tuberculosis, including: screening infection tests and collecting questionnaire data; Based on the collected questionnaire data, a preliminary database for tuberculosis risk assessment was constructed in combination with the basic information, contact history, and past history of the respondents; Apply algorithms to process the preliminary database and establish a tuberculosis risk assessment model; Using the established tuberculosis risk assessment model, the respondents were given tuberculosis risk scores, and were divided into different risk levels according to the risk scores; Subsequent follow-up observation will be implemented for respondents with high risk levels to further confirm their tuberculosis infection status and take corresponding preventive measures.

2. A tuberculosis risk assessment method and model according to claim 1, characterized in that: Conducting a tuberculosis epidemiological survey specifically includes: Conduct a skin test of recombinant tuberculosis fusion protein (EC) on the subjects, and determine whether the subjects are infected with Mycobacterium tuberculosis based on the test results; The collected data will be checked and collated as the respondents are investigated about their environment, number of sick people around them, contact history, etc., and the accuracy and completeness of the data will be checked.

3. A tuberculosis risk assessment method and model according to claim 1, characterized in that: The methodological steps for constructing a preliminary database for tuberculosis risk assessment are: The collected questionnaire data, basic information, contact history, symptoms and infection test results were classified and sorted, the answers to the questions in the questionnaire data were coded, the contact history was quantified, and the symptoms were coded according to the presence or absence of specific symptoms; The laboratory test results are standardized. For the recombinant tuberculosis fusion protein (EC) skin test results, the nodule diameter is converted into a specific risk level code. According to experience, large and small thresholds are set for the nodule diameter. If it is below the small threshold, it is low risk; if it is between the small and large thresholds, it is medium risk; if it is greater than the large threshold, it is high risk. Determine the fields of the database, the fields including: the basic information field of the investigated person, the contact history field, the symptom field and the laboratory test result field; Use database management software to create tables and enter the sorted data into fields; Enter the sorted data into the database and conduct a spot check on the entered data to check whether it is consistent with the original data.

4. A tuberculosis risk assessment method and model according to claim 1, characterized in that: The establishment of a tuberculosis risk assessment model specifically includes: The questionnaire data, basic information, contact history, symptom manifestations, and laboratory test results in the preliminary database were cleaned, outliers and missing values ​​were removed, and the data were standardized. The standardized calculation formula is: In the formula, represents the standardized value, X represents the original data, represents the mean of the data, Indicates the standard deviation of the data; Analyze the correlation between various variables in the preliminary database and tuberculosis risk, select characteristic variables with high correlation with tuberculosis risk, and establish a risk assessment model; Among them, the expression of the risk assessment model is: In the formula, Indicates the probability that the risk of tuberculosis is high, represents the characteristic variable, represents the model parameters; The algorithm was trained with the selected feature variables as input and the tuberculosis risk level as output; After training the established model with the training set, the model is evaluated with the test set. The evaluation indicators include accuracy, precision, recall, and F1 value. Based on the model evaluation results, the model is optimized and the model building and evaluation steps are repeated until satisfactory model performance is obtained.

5. A tuberculosis risk assessment method and model according to claim 4, characterized in that: Use the test set to evaluate the model. The evaluation indicators include accuracy, precision, recall and F1 value. include: The calculation formula for accuracy is: In the formula, represents the accuracy, TP represents the true positive, that is, the number of samples that are actually high-risk and predicted by the model to be high-risk. represents true negative examples, that is, the number of samples that are actually low risk and predicted by the model to be low risk, represents false positives, that is, the number of samples that are actually low risk but predicted by the model to be high risk. represents false negative examples, that is, the number of samples that are actually high risk but predicted as low risk by the model; The calculation formula of precision is: In the formula, Indicates accuracy; The calculation formula of recall rate is: In the formula, represents the recall rate; The calculation formula of F1 value is: 。 6. A tuberculosis risk assessment method and model according to claim 1, characterized in that: The method for using the tuberculosis risk assessment model to score and classify risk levels is as follows: Input the basic information, contact history, symptom manifestations, and infection test results of the subjects into the established tuberculosis risk assessment model; Input the characteristics of the subjects into the established tuberculosis risk assessment model, including basic information, contact history, symptoms and laboratory test results; The model calculates based on the input data and outputs a risk score value; Based on expert experience and previous research results, the risk level division intervals are determined, and the risk scores of the respondents are classified into corresponding risk levels according to the determined division intervals.

7. A tuberculosis risk assessment method and model according to claim 6, characterized in that: The risk level division interval is expressed as: The distribution range of risk scores is determined by data analysis as follows: ; The low risk range is set as: The medium risk interval is set as the risk score between the low risk interval and the high risk interval; The high-risk range is set as: 。 8. A tuberculosis risk assessment method and model according to claim 1, characterized in that: The method for implementing follow-up observation on high-risk respondents is as follows: The high-risk respondents classified according to the tuberculosis risk assessment model were identified as the subjects of follow-up observation; Follow-up visits were conducted at fixed intervals after the initial evaluation, and included symptom inquiry, physical examination, laboratory tests, and chest imaging examinations; Based on the information obtained from the follow-up, the tuberculosis infection status of the subjects was comprehensively judged; For patients diagnosed with tuberculosis, anti-tuberculosis treatment should be carried out in a timely manner and standardized management should be carried out according to the treatment plan. For those who are not diagnosed but still have a high risk, preventive measures should be taken, including: health education, avoiding contact with sources of infection and enhancing immunity.

9. A tuberculosis risk assessment method and model according to claim 1, characterized in that: When constructing a preliminary database, if there are vague or uncertain answers in the questionnaire data, they should be verified and corrected through telephone follow-up or re-survey to ensure the reliability of the data.

10. A tuberculosis risk assessment method and model according to claim 1, characterized in that: When selecting characteristic variables with a high correlation with tuberculosis risk, the stepwise regression method was used for variable screening, gradually introducing variables that contributed significantly to the model, while eliminating insignificant variables to improve the simplicity and accuracy of the model.

Images