Lung cancer early screening database management system

Abstract

The invention discloses a lung cancer early screening database management system, which relates to the technical field of database management and comprises a data acquisition module, a data arrangement module, a calculation module and an analysis module. The system integrates detection results of multiple biomarkers and individual clinical characteristics, provides a more accurate and more comprehensive lung cancer risk assessment database management system, can significantly improve the screening accuracy, reduces unnecessary excessive diagnosis and treatment, and improves the screening efficiency. The system considers clinical feature information such as age, gender, smoking history and family history, and constructs a comprehensive risk scoring system, so that the screening accuracy is further improved, the lung cancer risk of an individual can be comprehensively reflected, and the possibility of missed diagnosis and misdiagnosis is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of database management, and in particular to a database management system for early screening of lung cancer. Background Art

[0002] With the continuous advancement of medical technology, early screening and diagnosis of lung cancer have become the key to improving patient survival rates. However, traditional lung cancer screening methods, such as chest X-rays and CT scans, although able to detect lung abnormalities, have limitations in distinguishing benign from malignant nodules, leading to unnecessary overdiagnosis and treatment. In recent years, the application of biomarkers in lung cancer screening has received increasing attention. In addition, individual clinical characteristics are also important factors affecting the risk of lung cancer.

[0003] However, the current management of early lung cancer screening databases may only focus on a single biomarker and ignore the combined application of multiple biomarkers, which may lead to limited accuracy and sensitivity of screening results. In addition, the impact of individual clinical characteristics such as age, gender, smoking history and family history on lung cancer risk is ignored, resulting in limited predictive ability of screening models. Therefore, existing screening databases may often be simple and unable to fully capture the complex relationship between biomarkers and clinical characteristics and lung cancer risk. Summary of the invention

[0004] The purpose of the present invention is to provide a database management system for early screening of lung cancer, which solves the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a lung cancer early screening database management system, comprising: Data collection module: collects blood from patients and tests the biomarkers in the blood to output heat shock protein 90α, CEA, SHOX2 of ctDNA, and PTGER4 of ctDNA, and performs imaging examinations on the patient's lung area, as well as collects basic information about the patient; Data sorting module: input heat shock protein 90α, CEA, SHOX2 of ctDNA, PTGER4 of ctDNA, imaging examination data of the patient's lung area and basic information of the patient, output the original test results of the Hsp90α marker of the i-th patient, the original test results of the CEA marker of the i-th patient, the test evaluation value of the SHOX2 gene methylation of the i-th patient, the test evaluation value of the PTGER4 gene methylation of the i-th patient, establish a lung cancer early screening database, and verify the entered data; Calculation module: the original test results of the Hsp90α marker of the i-th patient, the original test results of the CEA marker of the i-th patient, the test evaluation value of the SHOX2 gene methylation of the i-th patient, and the test evaluation value of the PTGER4 gene methylation of the i-th patient are input, and the calculation module outputs the biomarker risk level of the i-th patient, the comprehensive risk score of the i-th patient, and the evaluation value of the i-th patient suffering from lung cancer; Analysis module: Analyze the biomarker risk level of the i-th patient, the comprehensive risk score of the i-th patient, and the assessment value of lung cancer for the i-th patient. Combined with the specific situation of the patient, further examination and diagnosis of high-risk patients are performed, and personalized screening and follow-up plans are formulated accordingly. The data in the database are then updated regularly.

[0006] Optionally, the calculation module includes: a biomarker risk submodule, a comprehensive risk submodule and a lung cancer management assessment submodule.

[0007] Optionally, the calculation formula of the biomarker risk submodule is as follows: ; in: RAE i refers to the biomarker risk level of the i-th patient, RAEh refers to the weight coefficient of Hsp90α, RAEc refers to the weight coefficient of CEA, RAEs refers to the weight coefficient of SHOX2 gene methylation, RAEp refers to the weight coefficient of PTGER4 gene methylation, BAEh i Refers to the detection evaluation value of Hsp90α and BAEha of the i-th patient i Refers to the original test results of Hsp90α marker of the i-th patient, BAEh min Refers to the minimum value of Hsp90α marker in all individuals, BAEh max Refers to the maximum value of Hsp90α marker in all individuals, BAEc min Refers to the minimum value of CEA markers among all individuals, BAEc max Refers to the maximum value of CEA markers among all individuals, BAEca i The original test results of CEA markers of the i-th patient, BAEc i Refers to the detection evaluation value of the tumor marker CEA of the i-th patient, BAEs i Refers to the detection evaluation value of SHOX2 gene methylation in the i-th patient, where positive is 1 and negative is 0, BAEp i Refers to the detection evaluation value of PTGER4 gene methylation of the i-th patient, where positive is 1 and negative is 0, and AA refers to a positive number to avoid the denominator being zero; The processing process of the biomarker risk submodule is as follows: the original test result of the Hsp90α marker of the i-th patient BAEha i , the original test result of CEA marker of the i-th patient BAEca i 、The detection evaluation value BAEs of SHOX2 gene methylation in the i-th patient i , the detection evaluation value of PTGER4 gene methylation of the i-th patient BAEp i Input to the biomarker risk submodule, output the biomarker risk level RAE of the i-th patient i .

[0008] Optionally, the calculation formula of the comprehensive risk submodule is as follows: ; in: CCD i refers to the comprehensive risk score of the i-th patient, CCDA refers to the weight coefficient of age factor in risk assessment, CCDB refers to the weight coefficient of gender factor in risk assessment, CCDC refers to the weight coefficient of smoking history in risk assessment, CCDD refers to the weight coefficient of family history in risk assessment, FCDA i Refers to the age of the i-th patient, FCDB i Refers to the gender of the i-th patient, 1 for male and 0 for female, FCDC i Refers to the number of years of smoking and FCDD of the i-th patient i Refers to the family history of lung cancer of the i-th patient, where a positive family history is 1 and a negative family history is 0; The processing process of the comprehensive risk submodule is as follows: the biomarker risk level RAE of the i-th patient i Input to the comprehensive risk submodule and based on the age of the i-th patient FCDA i and the gender FCDB of the i-th patient i Output the comprehensive risk score CCD of the i-th patient i .

[0009] Optionally, the calculation formula of the lung cancer management assessment submodule is as follows: ; in: PDG i Refers to the assessment value of lung cancer for the i-th patient, RAE i Refers to the biomarker risk level of the i-th patient, CCD i refers to the comprehensive risk score of the i-th patient, α refers to the impact parameter 1, and β refers to RAE i The influence coefficient of γ refers to CCD iThe influence coefficient of 1. θ refers to RAE i The influence coefficient of 2. ω refers to CCD i The influence coefficient of 2. φ refers to CCD i The influence coefficient of three; The processing process of the lung cancer management assessment submodule is as follows: the biomarker risk level RAE of the i-th patient i and the comprehensive risk score CCD of the i-th patient i Input into the lung cancer management evaluation submodule, and output the evaluation value PDG of lung cancer for the i-th patient i .

[0010] Optionally, the data acquisition module uses a blood testing device and an imaging examination device to extract and detect biomarkers in the blood. The imaging examination device is a low-dose spiral CT to detect the size and morphological characteristics of the lung lesions.

[0011] Optionally, the data sorting module is used to clean and sort the input data, eliminate invalid and erroneous data, and match the data with the patient.

[0012] Optionally, the establishment of the lung cancer early screening database in the data sorting module requires verification of input data after the database is established, and the database needs to be backed up regularly thereafter to prevent data loss.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention outputs the biomarker risk level of the ith patient through the biomarker risk submodule. This submodule provides a quantitative biomarker risk score for each patient by integrating the test results of multiple biomarkers. This score can comprehensively reflect the biological processes related to lung cancer in the patient, thereby providing more accurate and detailed individual risk information for the lung cancer early screening database, which helps database managers to better identify high-risk groups and conduct targeted follow-up and monitoring. The biomarker risk level can accurately reflect the level of biomarkers related to lung cancer in the patient, and has high specificity and sensitivity. In database management, this risk level can be used as a basis for preliminary screening, helping managers to quickly identify patients who may be at risk of lung cancer, and providing important clues for subsequent comprehensive evaluation.

[0014] 2. The present invention outputs the comprehensive risk score of the ith patient through the comprehensive risk submodule. This submodule further combines the patient's age, gender, smoking history, family history and other clinical characteristics on the basis of the biomarker risk score to provide a comprehensive risk score for each patient. This score not only considers biological factors but also incorporates the influence of environment and lifestyle, making the risk assessment more comprehensive and accurate. In database management, this score helps to achieve stratified risk management and optimize the allocation of medical resources. The comprehensive risk score of the ith patient, as a comprehensive risk score combined with multiple factors, can comprehensively assess the patient's lung cancer risk. In database management, the comprehensive risk score of the ith patient can be used as the basis for stratified management, and patients can be classified according to risk levels to achieve optimal resource allocation and precise management. In addition, the comprehensive risk score of the ith patient is used as an indicator for dynamic monitoring, which is updated as the patient's clinical characteristics change, providing support for continuous risk assessment.

[0015] 3. The present invention outputs the evaluation value of the i-th patient suffering from lung cancer through the lung cancer management evaluation submodule. This submodule uses the biomarker risk score and the comprehensive risk score to predict a probability of lung cancer for each patient. This probability value is intuitive and explainable, which helps doctors and patients to understand the level of risk and formulate personalized screening and treatment plans. In database management, this predicted probability is used as a decision support tool to help managers formulate more scientific screening strategies. In addition, the evaluation value of the i-th patient suffering from lung cancer is intuitive and explainable as the predicted probability of lung cancer, which can provide clear risk warnings for doctors and patients. In database management, this evaluation value is used as a decision support tool to help managers formulate more scientific screening strategies and treatment plans. At the same time, this evaluation value is used as an important tool for patient communication to improve patients' awareness and participation in early screening for lung cancer.

[0016] Fourth, the present invention is based on the calculation of the evaluation value of lung cancer of the ith patient, and can iterate the weight coefficient of family history in risk assessment in the comprehensive risk submodule, so that the comprehensive risk score of the ith patient and the evaluation value of lung cancer of the ith patient are continuously optimized and improved. The iteration can achieve the purpose of optimizing parameters of the feedback mechanism, thereby improving the prediction accuracy of the system. In the early screening of lung cancer, family history is an important risk factor. By using the evaluation value of lung cancer of the ith patient as feedback, the value of the family history of lung cancer of the ith patient is iteratively adjusted, so that the overall system can better adapt to the current data set. Improve the comprehensive risk score of the ith patient and the accuracy of the final predicted probability of the evaluation value of lung cancer of the ith patient. This iterative mechanism makes the system more robust and reliable in practical applications, because the evaluation value of lung cancer of the ith patient is comprehensively The information of biomarker risk score and comprehensive risk score is the system's estimate of the probability of the current individual suffering from lung cancer. The weight coefficient of family history in risk assessment, as an important component of the comprehensive risk score, directly affects the calculation result of the comprehensive risk score of the ith patient, and thus affects the accuracy of the assessment value of lung cancer for the ith patient. Therefore, the weight coefficient of family history in risk assessment is iteratively adjusted based on the assessment value of lung cancer for the ith patient. In fact, the rich information contained in the assessment value of lung cancer for the ith patient is used to optimize the system parameters, so that the system can accurately reflect the role of family history in lung cancer risk assessment. By iteratively adjusting the weight coefficient of family history in risk assessment, the system can more accurately reflect the impact of family history on lung cancer risk while keeping other parameters unchanged, thereby improving the prediction accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flowchart of the method steps of the early lung cancer screening database management system; Figure 2 This is the overall structure diagram of the lung cancer early screening database management system; Figure 3 This is a schematic diagram of the structure of the computing module in the early lung cancer screening database management system. 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 only 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] Regarding this lung cancer early screening database management system, it is different from the existing lung cancer early screening database management systems. The existing lung cancer early screening database management systems often only focus on a single biomarker and ignore the combined application of multiple biomarkers, which leads to limited accuracy and sensitivity of screening results, and the impact of individual clinical characteristics of patients such as age, gender, smoking history and family history on lung cancer risk is ignored, resulting in limited predictive ability of the screening model.

[0020] The module of the early screening database management system for lung cancer integrates the test results of multiple biomarkers such as heat shock protein 90α and carcinoembryonic antigen CEA. The present invention can more comprehensively evaluate the individual's lung cancer risk. These biomarkers play an important role in the occurrence and development of lung cancer. Their combined application can significantly improve the accuracy of screening. The present invention also takes into account clinical characteristic information such as age, gender, smoking history and family history, and further improves the accuracy of screening by constructing a comprehensive risk scoring system. This system can more comprehensively reflect the individual's lung cancer risk, thereby reducing the possibility of missed diagnosis and misdiagnosis.

[0021] Example 1: Please refer to Figures 1 to 3 This embodiment provides a lung cancer early screening database management system, including: Data collection module: collects blood from patients and tests the biomarkers in the blood to output heat shock protein 90α, CEA, SHOX2 of ctDNA, and PTGER4 of ctDNA, and performs imaging examinations on the patient's lung area, as well as collects basic information about the patient; Data sorting module: input heat shock protein 90α, CEA, SHOX2 of ctDNA, PTGER4 of ctDNA, imaging examination data of the patient's lung area and basic information of the patient, output the original test results of the Hsp90α marker of the i-th patient, the original test results of the CEA marker of the i-th patient, the test evaluation value of the SHOX2 gene methylation of the i-th patient, the test evaluation value of the PTGER4 gene methylation of the i-th patient, establish a lung cancer early screening database, and verify the entered data; Calculation module: the original test results of the Hsp90α marker of the i-th patient, the original test results of the CEA marker of the i-th patient, the test evaluation value of the SHOX2 gene methylation of the i-th patient, and the test evaluation value of the PTGER4 gene methylation of the i-th patient are input, and the calculation module outputs the biomarker risk level of the i-th patient, the comprehensive risk score of the i-th patient, and the evaluation value of the i-th patient suffering from lung cancer; Analysis module: Analyze the biomarker risk level of the i-th patient, the comprehensive risk score of the i-th patient, and the assessment value of lung cancer of the i-th patient. Combined with the specific conditions of the patients, further examine and diagnose high-risk patients, formulate personalized screening and follow-up plans accordingly, and then regularly update the data in the database; The calculation module includes: a biomarker risk submodule, a comprehensive risk submodule and a lung cancer management assessment submodule.

[0022] In this embodiment: the system comprehensively considers a variety of biomarkers and clinical characteristics, and provides a more comprehensive and accurate evaluation method for early screening of lung cancer. The system adopts advanced statistical methods and machine learning techniques, which can more accurately reflect the biological processes related to lung cancer in patients, improve the accuracy and reliability of risk assessment, and provide a quantitative risk score and prediction probability for each patient through the calculation of multiple groups of submodules, which is helpful to achieve stratified management and precision treatment of risks, which can not only improve the utilization efficiency of medical resources, but also reduce the medical burden and pain of patients. This submodule can also provide doctors and patients with clear risk warnings and decision support, which is helpful to improve the participation and satisfaction of early screening of lung cancer, and multiple groups of submodules are combined together to form a complete early screening evaluation system for lung cancer. This system can comprehensively consider a variety of biomarkers and clinical characteristics, and provide a comprehensive, accurate, and quantitative risk assessment and prediction probability for each patient. Through this evaluation system, stratified management and precision treatment of risks can be achieved, and the allocation and utilization efficiency of medical resources can be optimized. At the same time, it can also provide doctors and patients with clear risk warnings and decision-making support, and improve the participation and satisfaction of early screening for lung cancer. Compared with the existing technology, the multiple groups of sub-modules and their results of this system are combined with higher accuracy and reliability. By comprehensively considering multiple biomarkers and clinical characteristics, it can more accurately reflect the biological processes related to lung cancer in patients, and improve the accuracy and reliability of risk assessment. The multiple groups of sub-modules and their results of this system are also more operational and practical. Through quantitative risk scores and predicted probabilities, it can provide doctors and patients with clear risk warnings and decision-making support, which is conducive to the realization of stratified risk management and precise treatment. In addition, this evaluation system can also serve as an important tool for patient education and communication, and improve patients' awareness and participation in early screening for lung cancer.

[0023] See also Figures 1 to 3 , the processing of the biomarker risk submodule is as follows: ; in: RAE irefers to the biomarker risk level of the i-th patient, RAEh refers to the weight coefficient of Hsp90α, RAEc refers to the weight coefficient of CEA, RAEs refers to the weight coefficient of SHOX2 gene methylation, RAEp refers to the weight coefficient of PTGER4 gene methylation, BAEh i Refers to the detection evaluation value of Hsp90α and BAEha of the i-th patient i Refers to the original test results of Hsp90α marker of the i-th patient, BAEh min Refers to the minimum value of Hsp90α marker in all individuals, BAEh max Refers to the maximum value of Hsp90α marker in all individuals, BAEc min Refers to the minimum value of CEA markers among all individuals, BAEc max Refers to the maximum value of CEA markers among all individuals, BAEca i The original test results of CEA markers of the i-th patient, BAEc i Refers to the detection evaluation value of the tumor marker CEA of the i-th patient, BAEs i Refers to the detection evaluation value of SHOX2 gene methylation in the i-th patient, where positive is 1 and negative is 0, BAEp i Refers to the detection evaluation value of PTGER4 gene methylation of the i-th patient, where positive is 1 and negative is 0, and AA refers to a positive number to avoid the denominator being zero; The processing process of the biomarker risk submodule is as follows: the original test result of the Hsp90α marker of the i-th patient BAEha i , the original test result of CEA marker of the i-th patient BAEca i 、The detection evaluation value BAEs of SHOX2 gene methylation in the i-th patient i , the detection evaluation value of PTGER4 gene methylation of the i-th patient BAEp i Input to the biomarker risk submodule, output the biomarker risk level RAE of the i-th patient i .

[0024] In this embodiment: This submodule provides a quantitative biomarker risk score RAE for each patient by integrating the test results of multiple biomarkers and assigning different weights i This score can more comprehensively reflect the biological processes related to lung cancer in patients, thereby providing more accurate and detailed individual risk information for the lung cancer early screening database, which will help database managers better identify high-risk groups and conduct targeted follow-up and monitoring. iAs a biomarker risk score, it can accurately reflect the level of biomarkers related to lung cancer in patients, with high specificity and sensitivity. i It can be used as a basis for preliminary screening to help managers quickly identify patients who may be at risk of lung cancer and provide important clues for subsequent comprehensive evaluations.

[0025] See also Figures 1 to 3 , the processing process of the comprehensive risk submodule is as follows: ; in: CCD i refers to the comprehensive risk score of the i-th patient, CCDA refers to the weight coefficient of age factor in risk assessment, CCDB refers to the weight coefficient of gender factor in risk assessment, CCDC refers to the weight coefficient of smoking history in risk assessment, CCDD refers to the weight coefficient of family history in risk assessment, FCDA i Refers to the age of the i-th patient, FCDB i Refers to the gender of the i-th patient, 1 for male and 0 for female, FCDC i Refers to the number of years of smoking and FCDD of the i-th patient i Refers to the family history of lung cancer of the i-th patient, where a positive family history is 1 and a negative family history is 0; The processing process of the comprehensive risk submodule is as follows: the biomarker risk level RAE of the i-th patient i Input to the comprehensive risk submodule and based on the age of the i-th patient FCDA i and the gender FCDB of the i-th patient i Output the comprehensive risk score CCD of the i-th patient i .

[0026] In this embodiment: Based on the biomarker risk score, this submodule further combines the patient's age, gender, smoking history, family history and other clinical characteristics to provide a comprehensive risk score for each patient. This score not only takes into account biological factors, but also incorporates the influence of environment and lifestyle, making the risk assessment more comprehensive and accurate. In database management, this score helps to achieve stratified risk management and optimize the allocation of medical resources. CCD i As a comprehensive risk score, it combines multiple factors to comprehensively assess a patient's lung cancer risk. i It can be used as a basis for hierarchical management, and patients can be classified according to risk levels, so as to achieve optimal allocation and precise management of resources. In addition, CCD i It can also be used as an indicator for dynamic monitoring and updated as the patient's clinical characteristics change, providing support for ongoing risk assessment.

[0027] See also Figures 1 to 3 ,The processing process of the lung cancer management evaluation submodule is as follows: ; in: PDG i Refers to the assessment value of lung cancer for the i-th patient, RAE i Refers to the biomarker risk level of the i-th patient, CCD i refers to the comprehensive risk score of the i-th patient, α refers to the impact parameter 1, and β refers to RAE i The influence coefficient of γ refers to CCD i The influence coefficient of 1. θ refers to RAE i The influence coefficient of 2. ω refers to CCD i The influence coefficient of 2. φ refers to CCD i The influence coefficient of three; The processing process of the lung cancer management assessment submodule is as follows: the biomarker risk level RAE of the i-th patient i and the comprehensive risk score CCD of the i-th patient i Input into the lung cancer management evaluation submodule, and output the evaluation value PDG of lung cancer for the i-th patient i .

[0028] In this embodiment: This submodule uses biomarker risk scores and comprehensive risk scores, and uses statistical methods such as logistic regression to predict a probability of lung cancer PDG for each patient. i This probability value is intuitive and explainable, which helps doctors and patients understand the risk level and develop personalized screening and treatment plans. In database management, this predicted probability can be used as a decision support tool to help managers develop more scientific screening strategies, and PDG i As the predicted probability of lung cancer, it is intuitive and explainable, and can provide clear risk warnings for doctors and patients. In database management, PDG i It can be used as a decision support tool to help managers develop more scientific screening strategies and treatment plans. i It can also serve as an important tool for patient education and communication, increasing patients' awareness and participation in early screening for lung cancer.

[0029] It is worth noting that based on the evaluation value PDG of lung cancer in the i-th patient i The mutual operation with PSH can iterate the weight coefficient CCDD of family history in risk assessment in the comprehensive risk submodule, so that the comprehensive risk score CCD of the i-th patient is iand the estimated value of lung cancer for the i-th patient PDG i Continuously optimize and improve, the specific processing process is as follows: First up: CCDD new =CCDD old +CSC×(PDG i -PSH); Second: Set the iteration termination condition: Termination condition 1: The number of iterations is 100; Termination condition 2: |CCDD new -CCDD old |<0.001; in: CCDD new Refers to the weight coefficient of the iterated family history in risk assessment, CCDD old Refers to the weight coefficient of family history before iteration in risk assessment, CSC refers to the learning rate, which is used to control the step size of iteration, PDG i PSH refers to the evaluation value of lung cancer for the i-th patient and the average lung cancer evaluation value for all individuals in the current test batch.

[0030] In this embodiment: This submodule can optimize the parameters through the feedback mechanism through iteration, thereby improving the prediction accuracy of the model. In the early screening of lung cancer, family history is an important risk factor. By i As feedback, iteratively adjust FCDD i The value of can make the overall system better adapt to the current data set and improve the comprehensive risk score CCD i And the final predicted probability PDG i This iterative mechanism helps the model to be more robust and reliable in practical applications. i Iteration CCDD is due to PDG i Integrated biomarker risk score RAE i and the comprehensive risk score CCD i The information is the system's estimate of the probability of the current individual suffering from lung cancer, and CCDD is used as a comprehensive risk score CCD i An important component of CCD, the weight of which directly affects i The calculation results of PDG i The accuracy of PDG i To iteratively adjust CCDD, we are actually using PDG i The rich information contained in the model is used to optimize the model parameters, so that the system can more accurately reflect the role of family history in lung cancer risk assessment. The correlation and logic between the two lies in PDG iThe output of the system is an estimate of the current individual lung cancer risk, and CCDD is an input parameter of the system. Its weight determines the contribution of family history in risk assessment. By iteratively adjusting CCDD, the system can more accurately reflect the impact of family history on lung cancer risk while keeping other parameters unchanged, thereby improving the prediction accuracy of the model. This iteration has significant significance and contribution to the calculation results of multiple groups of submodules. First, by iteratively adjusting CCDD, the comprehensive risk score CCD can be optimized. i The calculation of PDG i Based on RAE i and CCD i Calculated, therefore the optimized CCD i Will directly affect PDG i The accuracy of the model can be improved, thereby further improving the prediction performance of the overall model. By iteratively adjusting CCDD, the system can be better adapted to the data sets of these studies, improving the applicability and accuracy of the system. Through the iterative mechanism, we can more clearly understand the role and contribution of risk factors such as family history in lung cancer risk assessment, which helps us better understand the research results in the two materials and provide a useful reference for future research. By continuously iterating and adjusting parameters, we can gradually improve the performance of the model and make it more accurately predict individual lung cancer risks. This is of great significance for improving the accuracy and efficiency of early screening for lung cancer.

[0031] In the specific implementation process, multiple submodules in this method are used to form a database management system for early screening of lung cancer. i , the original test result of CEA marker of the i-th patient BAEca i 、The detection evaluation value BAEs of SHOX2 gene methylation in the i-th patient i , the detection evaluation value of PTGER4 gene methylation of the i-th patient BAEp i Input to the biomarker risk submodule, output the biomarker risk level RAE of the i-th patient i This submodule provides a quantitative biomarker risk score (RAE) for each patient by integrating the test results of multiple biomarkers and assigning different weights. i , RAE i It can more comprehensively reflect the biological processes related to lung cancer in patients, thereby providing more accurate and detailed individual risk information for the lung cancer early screening database, helping database managers to better identify high-risk groups and conduct targeted follow-up and monitoring. iAs a biomarker risk score, it can accurately reflect the level of biomarkers related to lung cancer in patients, with high specificity and sensitivity. i It can be used as a basis for preliminary screening to help managers quickly identify patients who may be at risk of lung cancer and provide important clues for subsequent comprehensive evaluation; By inputting the biomarker risk level RAEi of the i-th patient into the comprehensive risk submodule, and based on the age of the i-th patient FCDA i and the gender FCDB of the i-th patient i Output the comprehensive risk score CCD of the i-th patient i Based on the biomarker risk score, this submodule further combines the patient's age, gender, smoking history, family history and other clinical characteristics to provide a comprehensive risk score for each patient. This score not only considers biological factors but also incorporates the impact of the environment and lifestyle, making the risk assessment more comprehensive and accurate. In database management, this score helps to achieve stratified risk management and optimize the allocation of medical resources. CCD i As a comprehensive risk score that combines multiple factors, it can comprehensively assess a patient's lung cancer risk in database management. i It can be used as a basis for hierarchical management, and patients can be classified according to risk levels to achieve optimal allocation and precise management of resources. In addition, CCD i It can also be used as an indicator for dynamic monitoring, updated as the patient's clinical characteristics change, and provide support for continuous risk assessment; By dividing the biomarker risk level RAE of the i-th patient i and the comprehensive risk score CCD of the i-th patient i Input into the lung cancer management evaluation submodule, and output the evaluation value PDG of lung cancer for the i-th patient i This submodule uses biomarker risk scores and comprehensive risk scores to predict a probability of lung cancer PDG for each patient. i This probability value is intuitive and explainable, which helps doctors and patients understand the risk level and develop personalized screening and treatment plans. In database management, this predicted probability can be used as a decision support tool to help managers develop more scientific screening strategies, and PDG i As the prediction probability of lung cancer, it is intuitive and explainable, and can provide clear risk warnings for doctors and patients. i It can be used as a decision support tool to help managers develop more scientific screening strategies and treatment plans. i It can also serve as an important tool for patient education and communication, increasing patient awareness and participation in early lung cancer screening; The estimated value of lung cancer in the i-th patient is PDG i The weight coefficient CCDD of family history in risk assessment in the comprehensive risk submodule can be iterated repeatedly, so that the comprehensive risk score CCD of the i-th patient i and the estimated value of lung cancer for the i-th patient PDG i Continuous optimization and improvement can be carried out. Through iteration, the feedback mechanism can optimize the parameters, thereby improving the prediction accuracy of the model. In the early screening of lung cancer, family history is an important risk factor. By integrating PDG i Iteratively adjust FCDD as feedback i The value of can make the overall system better adapt to the current data set and improve the comprehensive risk score CCD i And the final predicted probability PDG i This iterative mechanism helps the system to be more robust and reliable in practical applications, because PDG i Integrated biomarker risk score RAE i and the comprehensive risk score CCD i The information is the system's estimate of the probability of the current individual suffering from lung cancer, and CCDD is used as a comprehensive risk score CCD i An important component of the CCD. i The calculation results of PDG i The accuracy of PDG i To iteratively adjust CCDD, we are actually using PDG i The rich information contained in the system can be used to optimize system parameters, so that the system can more accurately reflect the role of family history in lung cancer risk assessment. By iteratively adjusting CCDD, the system can more accurately reflect the impact of family history on lung cancer risk while keeping other parameters unchanged, thereby improving the prediction accuracy of the model. This allows the various sub-modules to cooperate with each other in calculations, and to perform overall cycles and iterations, so that the overall system has the effect of automatic optimization and updating, and thus better adaptability.

[0032] Example 2: Please refer to Figure 1 , Figure 2 and Figure 3The data acquisition module uses a blood test device and an imaging examination device to extract and detect biomarkers in the blood. The imaging examination device is a low-dose spiral CT, which detects the size and morphological characteristics of lung lesions. The data sorting module is used to clean and sort the input data, eliminate invalid and erroneous data, and match the data with the patient. The establishment of the early screening database for lung cancer in the data sorting module requires verification of the input data after the database is established, and the database needs to be backed up regularly afterwards to prevent data loss.

[0033] In this embodiment: Currently, computed tomography screening, or CT examination, is mainly used for clinical screening of lung cancer. Although conventional CT examination has a high sensitivity to lung cancer, its high dose of radiation is difficult for patients to accept. Therefore, low-dose spiral CT screening is more applicable in the clinical diagnosis of lung cancer while ensuring sensitivity. Low-dose spiral CT is an effective, simple and economical method for lung cancer screening. It can detect lesions early and reduce the mortality rate of lung cancer. For example, children under 1 year old or elderly people over 60 years old, or patients with severe liver disease, kidney disease and heart disease are not suitable for CT examination. Therefore, surgical experts are still more reserved about carrying out large-scale low-dose spiral CT screening in a wide range of people in order to achieve early diagnosis and early treatment of lung cancer patients. In addition, combined with low Low-dose spiral CT and non-invasive body fluid tumor markers can also reduce the psychological burden of patients, assist patients and their families in making reasonable medical decisions, and improve patient prognosis. In general, low-dose spiral CT uses the smallest scanning range, lowest dose, and least amount of X-rays to diagnose lesions. Compared with traditional routine CT examinations, it has less radiation and can show tiny nodules more clearly. By combining low-dose spiral CT, tumor marker Hsp90α, and CEA detection, we dynamically track and follow up the changing characteristics and patterns of solid nodules with a diameter of 6 to 8 mm in the high-risk population for lung cancer, determine the value of Hsp90α in the diagnosis of benign and malignant solid nodules and in early warning of high-risk cancer populations, and observe whether Hsp90α detection has the clinical significance of reducing the number of CT follow-up visits and reducing CT radiation. It should be noted that heat shock protein 90α is an important molecular chaperone protein. It was first discovered by the Key Laboratory of Anti-tumor Protein Drugs of Tsinghua University. Through the combined use of low-dose spiral CT, tumor marker CEA and plasma Hsp90α detection, patients with 6-8 mm solid nodules screened from the clinical physical examination population were followed up and monitored to evaluate the diagnostic value of plasma Hsp90α detection for benign and malignant solid nodules and the early warning value for high-risk lung cancer. ctDNA is a new biological marker that is not only minimally invasive, rapid and accurate, but also easily accepted by patients. It can indirectly reflect the tumor burden in lung cancer patients and the overall picture of the tumor genome. It is an effective method for accurate diagnosis and prognosis. With people's understanding of epigenetic DNA methylation, it has been found that DNA methylation abnormalities exist in many human malignant tumors and appear in the early stages of lung cancer. Currently, SHOX2 and PTGER4 gene methylation can be detected in lung cancer tissues and the plasma of lung cancer patients. Lung cancer is one of the most malignant tumors. Non-small cell lung cancer (NSCLC) accounts for about 85% of lung cancer, and about 70% of NSCLC are found at the time of diagnosis. It is already in the late stage and its survival rate drops significantly. Therefore, early diagnosis is an important means to improve the treatment effect of lung cancer. The commonly used clinical lung cancer diagnosis methods have certain limitations in specificity, sensitivity and patient compliance. Research in recent years has continuously confirmed that tumor tissue continues to grow and release DNA into the periphery to become circulating tumor DNA, which has the same genetic characteristics as the primary tumor genome DNA and is closely related to the occurrence and metastasis of the tumor. ctDNA is a new biological molecular marker, which not only has the advantages of minimally invasive, rapid and accurate, but also is easy to be accepted by patients. It can indirectly reflect the tumor load in lung cancer patients and can also reflect the overall picture of the tumor genome. It is an effective method for accurate diagnosis and prognosis.

[0034] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A database management system for early screening of lung cancer, characterized in that: include: Data collection module: collects blood from patients and tests the biomarkers in the blood to output heat shock protein 90α, CEA, SHOX2 of ctDNA, and PTGER4 of ctDNA, and performs imaging examinations on the patient's lung area, as well as collects basic information about the patient; Data sorting module: input heat shock protein 90α, CEA, SHOX2 of ctDNA, PTGER4 of ctDNA, imaging examination data of the patient's lung area and basic information of the patient, output the original test results of the Hsp90α marker of the i-th patient, the original test results of the CEA marker of the i-th patient, the test evaluation value of the SHOX2 gene methylation of the i-th patient, the test evaluation value of the PTGER4 gene methylation of the i-th patient, establish a lung cancer early screening database, and verify the entered data; Calculation module: the original test results of the Hsp90α marker of the i-th patient, the original test results of the CEA marker of the i-th patient, the test evaluation value of the SHOX2 gene methylation of the i-th patient, and the test evaluation value of the PTGER4 gene methylation of the i-th patient are input, and the calculation module outputs the biomarker risk level of the i-th patient, the comprehensive risk score of the i-th patient, and the evaluation value of the i-th patient suffering from lung cancer; Analysis module: Analyze the biomarker risk level of the i-th patient, the comprehensive risk score of the i-th patient, and the assessment value of lung cancer for the i-th patient. Combined with the specific situation of the patient, further examination and diagnosis of high-risk patients are performed, and personalized screening and follow-up plans are formulated accordingly. The data in the database are then updated regularly.

2. The database management system for early screening of lung cancer according to claim 1, characterized in that: The calculation module includes: a biomarker risk submodule, a comprehensive risk submodule and a lung cancer management assessment submodule.

3. The database management system for early screening of lung cancer according to claim 2, characterized in that: The calculation formula of the biomarker risk submodule is as follows: ； in: RAE i refers to the biomarker risk level of the i-th patient, RAEh refers to the weight coefficient of Hsp90α, RAEc refers to the weight coefficient of CEA, RAEs refers to the weight coefficient of SHOX2 gene methylation, RAEp refers to the weight coefficient of PTGER4 gene methylation, BAEh i Refers to the detection evaluation value of Hsp90α and BAEha of the i-th patient i Refers to the original test results of Hsp90α marker of the i-th patient, BAEh min Refers to the minimum value of Hsp90α marker in all individuals, BAEh max Refers to the maximum value of Hsp90α marker in all individuals, BAEc min Refers to the minimum value of CEA markers among all individuals, BAEc max Refers to the maximum value of CEA markers among all individuals, BAEca i The original test results of CEA markers of the i-th patient, BAEc i Refers to the detection evaluation value of the tumor marker CEA of the i-th patient, BAEs i Refers to the detection evaluation value of SHOX2 gene methylation in the i-th patient, where positive is 1 and negative is 0, BAEp i Refers to the detection evaluation value of PTGER4 gene methylation of the i-th patient, where positive is 1 and negative is 0, and AA refers to a positive number to avoid the denominator being zero; The processing process of the biomarker risk submodule is as follows: the original test result of the Hsp90α marker of the i-th patient BAEha i , the original test result of CEA marker of the i-th patient BAEca i 、The detection evaluation value BAEs of SHOX2 gene methylation in the i-th patient i , the detection evaluation value of PTGER4 gene methylation of the i-th patient BAEp i Input to the biomarker risk submodule, output the biomarker risk level RAE of the i-th patient i .

4. The database management system for early screening of lung cancer according to claim 3, characterized in that: The calculation formula of the comprehensive risk submodule is as follows: ； in: CCD i refers to the comprehensive risk score of the i-th patient, CCDA refers to the weight coefficient of age factor in risk assessment, CCDB refers to the weight coefficient of gender factor in risk assessment, CCDC refers to the weight coefficient of smoking history in risk assessment, CCDD refers to the weight coefficient of family history in risk assessment, FCDA i Refers to the age of the i-th patient, FCDB i Refers to the gender of the i-th patient, 1 for male and 0 for female, FCDC i Refers to the number of years of smoking and FCDD of the i-th patient i Refers to the family history of lung cancer of the i-th patient, where a positive family history is 1 and a negative family history is 0; The processing process of the comprehensive risk submodule is as follows: the biomarker risk level RAE of the i-th patient i Input to the comprehensive risk submodule and based on the age of the i-th patient FCDA i and the gender FCDB of the i-th patient i Output the comprehensive risk score CCD of the i-th patient i .

5. The database management system for early screening of lung cancer according to claim 4, characterized in that: The calculation formula of the lung cancer management assessment submodule is as follows: ； in: PDG i Refers to the assessment value of lung cancer for the i-th patient, RAE i Refers to the biomarker risk level of the i-th patient, CCD i refers to the comprehensive risk score of the i-th patient, α refers to the impact parameter 1, and β refers to RAE i The influence coefficient of γ refers to CCD i The influence coefficient of θ refers to RAE i The influence coefficient of 2. ω refers to CCD i The influence coefficient of 2. φ refers to CCD i The influence coefficient of three; The processing process of the lung cancer management assessment submodule is as follows: the biomarker risk level RAE of the i-th patient i and the comprehensive risk score CCD of the i-th patient i Input into the lung cancer management evaluation submodule, and output the evaluation value PDG of lung cancer for the i-th patient i .

6. The database management system for early screening of lung cancer according to claim 1, characterized in that: The data acquisition module uses a blood testing device and an imaging examination device to extract and detect biomarkers in the blood. The imaging examination device is a low-dose spiral CT to detect the size and morphological characteristics of the lung diseased area.

7. The database management system for early screening of lung cancer according to claim 1, characterized in that: The data sorting module is used to clean and sort the input data, eliminate invalid and erroneous data, and match the data with the patient.

8. The database management system for early screening of lung cancer according to claim 1, characterized in that: The establishment of the lung cancer early screening database in the data sorting module requires verification of the input data after the database is established, and the database needs to be backed up regularly thereafter to prevent data loss.

Images