Method for constructing a prediction model of risk factors for pulmonary cavity occurrence in initial treatment sputum-positive pulmonary tuberculosis

Abstract

The application discloses a kind of risk factors prediction model of occurrence of initial treatment bacterium positive pulmonary tuberculosis lung cavity, belong to medical data processing technical field, the clinical data of initial treatment bacterium positive pulmonary tuberculosis patient is collected in the application, using single factor analysis, Lasso regression analysis and multivariate Logistic regression analysis, male, diabetes, smoking, white blood cell count and platelet / hemoglobin ratio are screened as the independent risk factors of initial treatment bacterium positive pulmonary tuberculosis causing lung cavity, and based on this, nomogram prediction model is constructed.H-L goodness-of-fit test, receiver operating characteristic curve and decision curve analysis are verified, the model has higher prediction efficiency and clinical practical value.The application can help clinical advance to predict and intervene the cavity caused by high-risk population of pulmonary tuberculosis, provide new ideas for the diagnosis and treatment of bacterium positive pulmonary tuberculosis causing cavity.

Description

[0001] Technical Field Methods / Systems This invention relates to the field of medical data processing technology, and in particular to a method for constructing a risk factor prediction model for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis. Background Technology

[0002] Pulmonary tuberculosis (TB) is a chronic infectious lung disease caused by Mycobacterium tuberculosis. Patients with sputum-positive TB have a high incidence of cavitation, and the occurrence of cavitation not only affects sputum conversion to negative but is also an independent risk factor for poor prognosis and drug resistance. Studies have confirmed a close correlation between the number of cavities and sputum conversion to negative, and existing literature also indicates that cavity formation can significantly affect disease prognosis. However, current research on risk factors for cavitation in newly diagnosed sputum-positive TB is still limited, and there is a lack of scientifically feasible risk factor prediction models, making it difficult to meet the needs of early clinical prevention and management. Traditional standard chemotherapy for cavitary TB has limited effectiveness, as drugs cannot effectively penetrate deep into the cavities. Although cavitary resection and bronchoscopic local drug administration are used clinically, many cavitary lesions are difficult to detect, and diagnosis requires highly sophisticated equipment. Many underdeveloped areas lack the necessary equipment and professional personnel. Early diagnosis remains crucial for improving the treatment outcomes of newly diagnosed sputum-positive TB.

[0003] Therefore, there is an urgent need in this field for a technical solution that can effectively predict the risk of pulmonary cavitation in newly diagnosed septic tuberculosis patients.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method for constructing a risk factor prediction model for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis, so as to help clinicians predict and intervene in advance the occurrence of cavities in high-risk groups of pulmonary tuberculosis.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for constructing a risk factor prediction model for the occurrence of pulmonary cavitation in newly diagnosed septic tuberculosis includes the following steps: Collect clinical data of newly diagnosed septic tuberculosis patients; Univariate analysis, Lasso regression analysis, and multivariate logistic regression analysis were performed on the clinical data to determine that male sex, diabetes, smoking, white blood cell count, and platelet / hemoglobin ratio were independent risk factors for pulmonary cavitation caused by newly diagnosed septic tuberculosis. A nomogram prediction model is constructed based on the aforementioned independent risk factors.

[0007] Optionally, the clinical data includes the patient's general information and hematological examination results; the general information includes gender, age, hypertension, diabetes, history of alcohol consumption, and history of smoking; the hematological examination results include white blood cell count, C-reactive protein, mean platelet volume, aspartate aminotransferase, gamma-glutamyl transferase, adenosine deaminase, uric acid, neutrophil / lymphocyte ratio, and platelet / hemoglobin ratio.

[0008] Optionally, the univariate analysis is used to screen variables with P < 0.05; the Lasso regression analysis uses 10-fold cross-validation for data processing and variable screening, and identifies predictors by identifying features with non-zero coefficients; the multivariate logistic regression analysis is used to identify independent risk factors.

[0009] Optionally, the nomogram prediction model is constructed using a regression equation: Logit[P / (1-P)]=1.052×male+3.075×diabetes+2.041×smoking+0.150×white blood cell count+0.612×platelet / hemoglobin value-5.642.

[0010] Optionally, the method may also include a step of evaluating the nomogram prediction model, wherein the evaluation includes assessing the calibration accuracy of the model using the Hosmer-Lemeshow goodness-of-fit test, assessing the discriminative power of the model using receiver operating characteristic (ROC) curves, and assessing the clinical utility of the model using decision curve analysis.

[0011] Compared with the prior art, the present invention has the following beneficial effects: This invention constructs a nomogram prediction model by systematically screening independent risk factors for pulmonary cavitation in newly diagnosed sputum-positive pulmonary tuberculosis patients. The factors included in this model are simple and readily available, possessing practical clinical applicability. Verification through HL goodness-of-fit test, receiver operating characteristic (ROC) curve, and decision curve analysis shows that the model has good predictive efficacy and clinical application value. It helps clinicians to detect pulmonary cavitation caused by sputum-positive pulmonary tuberculosis early and intervene in advance, guiding the treatment of cavitary pulmonary tuberculosis and improving patient prognosis. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1This diagram illustrates the predictor variables for the LASSO regression outcome prediction model provided in this embodiment of the invention. (A. Shows the selection results of the optimal parameter λ for LASSO regression based on 10-fold cross-validation, including the trend of misclassification error rate of each variable as a function of λ. The two vertical dashed lines in the figure correspond to the optimal λ values ​​determined under the minimum standard error criterion and the 1-standard error criterion, respectively, where λ is the parameter of the model.; B. Presents the distribution of LASSO regression coefficients corresponding to all baseline features.) Figure 2 This is a schematic diagram of a nomogram model of the risk of pulmonary cavitation caused by newly diagnosed seropositive pulmonary tuberculosis, provided in an embodiment of the present invention.

[0014] Figure 3 The diagram shows the ROC curve, calibration curve, and DCA analysis curve of the prediction model for pulmonary cavitation caused by newly diagnosed septic tuberculosis provided in this embodiment of the invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] The purpose of this invention is to provide a technical solution that can effectively predict the risk of pulmonary cavitation in newly diagnosed septic tuberculosis patients.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] Example 1: The purpose of this embodiment is to explore the risk factors for pulmonary cavitation in newly diagnosed sputum-positive pulmonary tuberculosis, establish a risk nomogram prediction model, help clinicians predict and intervene in the development of cavitation in high-risk groups of pulmonary tuberculosis in advance, improve the identification, intervention and prevention of early diseases, and thus provide new ideas for the diagnosis and treatment of cavitation caused by sputum-positive pulmonary tuberculosis.

[0019] This study collected data from newly diagnosed septic tuberculosis patients admitted to a hospital between January 2020 and February 2025. Based on the inclusion and exclusion criteria, 1240 patients meeting the requirements were included in the study; the patients' ages ranged from 18 to 92 years. The patients were randomly divided into a training set (869 cases) and a validation set (371 cases) at a ratio of 7:3.

[0020] Inclusion criteria: newly diagnosed pulmonary tuberculosis with cavitation (experimental group) Inclusion criteria: (1) meets the People's Republic of my country Health Industry Standard "Diagnosis of Pulmonary Tuberculosis" (WS 288-2017) issued in 2017

[10] ; (2) positive sputum smear and / or sputum culture; (3) imaging suggests pulmonary cavitation; (4) pulmonary tuberculosis is diagnosed for the first time and has not undergone anti-tuberculosis treatment before admission; (5) age > 18 years. Inclusion criteria for control group: meets (1), (2), (4), (5) and imaging suggests no pulmonary cavitation.

[0021] Exclusion criteria: (1) Comorbid serious diseases of other systems; (2) Malignant tumors; (3) Comorbid acute or chronic respiratory failure, heart failure, liver and kidney failure; (4) Incomplete clinical data; (5) Pregnant patients; (6) Pregnant patients; (7) Age <18 years.

[0022] Data collection: Collect general information about the patient, such as gender, age, hypertension, diabetes, alcohol consumption history, and smoking history. Collect the patient's blood cell analysis, liver and kidney function tests, coagulation routine tests, and other hematological test results upon admission.

[0023] Selection of predictor variables in a predictive model: Univariate analysis was performed on the training set to select variables with p < 0.05. Simultaneously, LASSO regression was performed on the training set, using 10-fold cross-validation for data processing and variable selection. Predictors were identified by recognizing features with non-zero coefficients. The selected common feature variables were then incorporated into multivariate logistic regression analysis to determine their independent risk factors and to construct the predictive model equation.

[0024] A nomogram prediction model was constructed using R software based on the regression equation. The Hosmer-Lemeshow (HL) goodness-of-fit test was used to evaluate the model's fit on the training and validation sets, and calibration curves were used to assess the model's fit. The model's discriminative power was evaluated by plotting receiver operating characteristic (ROC) curves; and the clinical application value of the model was evaluated using decision curve analysis (DCA).

[0025] Statistical methods: Data analysis was performed using R 4.4.3 software. Grouping variables were expressed as chi-square (x²). Quantitative data were represented using different methods depending on their distribution type. Normally distributed data were expressed as mean ± standard deviation (x ± s), and independent samples t-tests were used for comparisons between two groups. Non-normally distributed data were expressed as median (interquartiles) [M(Q1,Q3)], and independent samples rank-sum tests were used for comparisons between two groups. Lasso regression was implemented using the "glmnet" package in R software. For both univariate analysis and multivariate logistic regression models, a p-value < 0.05 was considered statistically significant.

[0026] result: Univariate analysis of pulmonary cavitation caused by newly diagnosed septic tuberculosis: Analysis of baseline data from the training set revealed significant differences between the experimental and control groups in several indicators, including male sex, diabetes, smoking, CRP, WBC, MPV, AST, GGT, ADA, UA, NLR, and PHR (P < 0.05). See Table 1.

[0027] Table 1 Univariate analysis of pulmonary cavitation caused by newly diagnosed septic tuberculosis.

[0028] Lasso regression analysis of pulmonary cavitation caused by newly diagnosed sputum-positive pulmonary tuberculosis: All variables were included in the LASSO regression model. Through 10-fold cross-validation, eight variables with non-zero coefficients were selected when Lambda 1Se = 0.02345: male sex, diabetes, hypertension, smoking, alcohol consumption, WBC, CRP, and PHR. See [link to model]. Figure 1 .

[0029] Multivariate logistic regression analysis of pulmonary cavitation caused by newly diagnosed sputum-positive pulmonary tuberculosis: Six common characteristic variables selected by univariate analysis and Lasso regression were included in a multivariate logistic regression analysis. The results showed that male sex, diabetes, smoking, WBC, and PHR were independent risk factors for pulmonary cavitation caused by newly diagnosed septic tuberculosis (P<0.05). See Table 2.

[0030] Table 2. Univariate binary logistic regression analysis of pulmonary cavitation caused by newly diagnosed septic tuberculosis.

[0031] Construction of a nomogram predictive model for pulmonary cavitation caused by newly diagnosed sputum-positive pulmonary tuberculosis: Based on the results of multivariate logistic regression analysis, a nomogram prediction model was constructed using R software. The regression equation for the prediction model is: Logit [P( / 1-P)] = 1.052 × "Male" + 3.075 × "Diabetes" + 2.041 × "Smoking" + 0.150 × "White blood cell count" + 0.612 × "Platelet / hemoglobin ratio" - 5.642. See [link / details]. Figure 2 .

[0032] Evaluate the predictive efficacy and clinical application value of the nomogram prediction model: The HL goodness-of-fit test results show that the training set χ² 2 =12.426 (P>0.05), validation set χ² 2=8.902 (P>0.05). ROC curve results show that the area under the curve (AUC) for the training set is 0.861 (95% CI 0.833, 0.887), and the AUC for the validation set is 0.844 (95% CI 0.792, 0.890). The calibration curves for both the training and validation sets fit the actual curves well, and the predicted probabilities are close to the actual probabilities. DCA analysis results show that the threshold probabilities in both the training and validation sets are within the 0-1 range, and the net return of the prediction model is >0. See... Figure 3 .

[0033] In summary, this embodiment included 1240 newly diagnosed septic tuberculosis patients admitted to a hospital from January 2020 to February 2025. Patient admission data, hematological examinations, and imaging data were collected and randomly assigned to a training set (869 cases) and a validation set (371 cases) at a ratio of 7:3. Based on inclusion and exclusion criteria, patients with newly diagnosed septic tuberculosis developing cavitation were assigned to the experimental group, and patients without cavitation were assigned to the control group. In the training set, univariate analysis and Lasso regression analysis were used to screen for risk factors influencing pulmonary cavitation in newly diagnosed septic tuberculosis. Common risk factors were included in multivariate logistic regression analysis to determine their independent risk factors. A nomogram prediction model for these independent risk factors was constructed using R language. The discriminative power of the model was evaluated by plotting receiver operating characteristic (ROC) curves, and the calibration accuracy of the model was verified using the Hosmer-Lemeshow (HL) goodness-of-fit test and calibration curve. The clinical applicability of the model was analyzed using decision curve analysis (DCA).

[0034] Results: Univariate analysis showed statistically significant differences between the experimental and control groups in terms of male sex, diabetes, smoking, C-reactive protein (CRP), white blood cell count (WBC), mean platelet volume (MPV), aspartate aminotransferase (AST), gamma-glutamyl transferase (GGT), adenosine deaminase (ADA), uric acid (UA), neutrophil / lymphocyte ratio (NLR), and platelet / hemoglobin ratio (PHR) (P<0.05). Characteristic variables screened by Lasso regression analysis included male sex, diabetes, hypertension, smoking, alcohol consumption, WBC, CRP, and PHR. Both univariate and Lasso regression analyses indicated that male sex, diabetes, smoking, WBC, CRP, and PHR were common and important variables in newly diagnosed septic tuberculosis leading to pulmonary cavitation. Multivariate logistic regression analysis showed that male sex (OR=2.864, 95%CI 1.612, 5.087), diabetes (OR=21.657, 95%CI 13.192, 35.555), smoking (OR=7.698, 95%CI 4.867, 12.177), WBC (OR=1.162, 95%CI 1.062, 1.272), and PHR (OR=1.844, 95%CI 1.464, 2.324) were independent risk factors for pulmonary cavitation caused by newly diagnosed septic tuberculosis (P<0.05). The regression equation for the predictive model is: Logit[P( / 1-P)]=1.052דMale”+3.075דDiabetes”+2.041דSmoking”+0.150דWhite Blood Cell Count”+0.612דPlatelet / Hemoglobin Ratio”-5.642. The H-L goodness-of-fit test results show that the training set χ² 2 =12.426 (P>0.05), validation set χ² 2 =8.902 (P>0.05). ROC curve results showed that the area under the curve (AUC) for the training set was 0.861 (95% CI 0.833, 0.887), and the AUC for the validation set was 0.844 (95% CI 0.792, 0.890). The calibration curves for both the training and validation sets fit the actual curves well, and the predicted probabilities were close to the actual probabilities. DCA analysis showed that the threshold probabilities in both the training and validation sets were within the 0-1 range, and the net return of the prediction model was >0. Conclusion: Male sex, diabetes, smoking, WBC, and PHR are important variables for pulmonary cavitation caused by newly diagnosed septic tuberculosis. The nomogram prediction model for pulmonary cavitation caused by newly diagnosed septic tuberculosis constructed using the above five predictive variables has good predictive efficacy and clinical application value.

[0035] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0036] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for constructing a predictive model of risk factors for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis, characterized in that, Includes the following steps: Collect clinical data of newly diagnosed septic tuberculosis patients; Univariate analysis, Lasso regression analysis, and multivariate logistic regression analysis were performed on the clinical data to determine that male sex, diabetes, smoking, white blood cell count, and platelet / hemoglobin ratio were independent risk factors for pulmonary cavitation caused by newly diagnosed septic tuberculosis. A nomogram prediction model is constructed based on the aforementioned independent risk factors.

2. The method for constructing a risk factor prediction model for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis according to claim 1, characterized in that, The clinical data includes the patient's general information and hematological examination results; the general information includes gender, age, hypertension, diabetes, history of alcohol consumption, and history of smoking; the hematological examination results include white blood cell count, C-reactive protein, mean platelet volume, aspartate aminotransferase, gamma-glutamyl transferase, adenosine deaminase, uric acid, neutrophil / lymphocyte ratio, and platelet / hemoglobin ratio.

3. The method for constructing a risk factor prediction model for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis patients according to claim 1, characterized in that, The univariate analysis was used to screen variables with p < 0.05; the Lasso regression analysis used 10-fold cross-validation for data processing and variable screening, and determined predictors by identifying features with non-zero coefficients; the multivariate logistic regression analysis was used to identify independent risk factors.

4. The method for constructing a risk factor prediction model for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis according to claim 1, characterized in that, The nomogram prediction model is constructed using the regression equation: Logit[P / (1-P)]=1.052×male+3.075×diabetes+2.041×smoking+0.150×white blood cell count+0.612×platelet / hemoglobin value-5.

642.

5. The method for constructing a risk factor prediction model for the occurrence of pulmonary cavities in newly diagnosed septic tuberculosis according to claim 1, characterized in that, It also includes a step of evaluating the nomogram prediction model, the evaluation of which includes assessing the calibration accuracy of the model using the Hosmer-Lemeshow goodness-of-fit test, assessing the discriminative power of the model using receiver operating characteristic (ROC) curves, and assessing the clinical utility of the model using decision curve analysis.

Images