Method for predicting risk of chronic obstructive pulmonary disease in frigid region, electronic equipment and storage medium

Abstract

The invention discloses a cold region chronic obstructive pulmonary disease risk prediction method, electronic equipment and a storage medium. The cold region chronic obstructive pulmonary disease risk prediction method comprises the steps of obtaining conventional indexes of a person to be predicted; and inputting the conventional indexes into a preset COPD risk prediction model so as to predict and display the disease risk of the to-be-predicted person. Wherein the conventional indexes comprise age, smoking condition, drinking condition, chronic bronchitis and emphysema history, airflow blocking condition, average diastolic pressure, heart rate and BMI. The conventional indexes can be obtained without depending on expensive special equipment (such as imaging equipment) or complex biomarker detection, so that the data acquisition cost and the operation difficulty are greatly reduced. The preset COPD risk prediction model is constructed based on an XGBoost model and a Logistic Regression model, so that efficient and accurate COPD disease risk prediction can be realized, and the applicability of the model can be improved.

Description

Technical Field

[0001] This invention relates to the field of chronic obstructive pulmonary disease (COPD) prevention and treatment technology, and in particular to a method, electronic device and storage medium for predicting the risk of COPD in cold regions. Background Technology

[0002] Chronic obstructive pulmonary disease (COPD) is a major chronic disease that seriously endangers public health, especially in cold-climate regions where its prevalence and disease burden are significantly higher than in other regions. Existing research indicates that complex factors such as environmental and occupational exposures are closely related to the high incidence of COPD in cold-climate regions. However, in large-scale application scenarios such as primary healthcare institutions, community screening, and health checkups in cold-climate regions, systematically collecting individualized environmental exposure data or conducting complex physiological tests (such as imaging examinations) faces practical bottlenecks such as high costs, high operational expertise requirements, and poor reproducibility. This contradiction makes the development of a widely applicable COPD risk screening tool an urgent need.

[0003] Currently, existing COPD risk screening tools mainly rely on two types of methods: one is based on complex models using specialized examinations for prediction. These models typically use complex clinical data obtained through pulmonary function tests or specific biomarker tests as core variables, resulting in high predictive accuracy. The other is based on simplified screening questionnaires or scoring cards, relying on traditional risk factor combinations for rapid assessment. However, the former requires specialized equipment and operators and is highly dependent on complex clinical data, leading to high costs and making it difficult to popularize in resource-constrained primary healthcare systems in cold regions. The latter, on the other hand, often neglects basic vital sign information (such as heart rate) readily available in routine medical care, resulting in insufficient utilization of variables and limited predictive efficacy and applicability.

[0004] Therefore, there is still an urgent need to develop a new COPD risk screening tool based on conventional and easily accessible indicators, which can both ensure predictive accuracy and have good scalability and applicability, so as to effectively meet the actual needs of primary healthcare and public health screening in cold regions. Summary of the Invention

[0005] Based on this, the present invention proposes a method, electronic device and storage medium for predicting the risk of chronic obstructive pulmonary disease in cold regions, aiming to solve the problems of existing COPD risk screening tools being highly dependent on complex clinical data and having limited applicability.

[0006] To solve the above problems, the present invention adopts the following technical solution: In a first aspect, this invention proposes a method for predicting the risk of chronic obstructive pulmonary disease in cold regions, including: Obtain the routine indicators of the person to be predicted; wherein, the routine indicators are eight influencing factors, namely, age, smoking status, drinking status, history of chronic bronchitis and emphysema, airflow obstruction status, mean diastolic blood pressure, heart rate and BMI; The conventional indicators are input into a preset COPD risk prediction model to predict the risk of disease in the individuals to be predicted and to display the predicted risk; wherein, the preset COPD risk prediction model is constructed based on the XGBoost model and the Logistic Regression model.

[0007] Furthermore, the preset COPD risk prediction model is as follows: logit(p) = intercept term + 0.089[0.063, 0.116] × age +0(smoking_category="Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482](smoking_category="Previous smoking") +0(Drinking="No")-0.501(Drinking="Yes") +0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicbronchitisandemphysemahistory="Yes") +0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes") +0.018[0.001,0.035]×mean_dbp+0.013[-0.002,0.027]×heart_rate +0(BMI="Normal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity"); Wherein, logit(p) represents the linear prediction value of the preset COPD risk prediction model, and the intercept term includes the point estimate of the regression coefficient and the corresponding 95% confidence interval; 0.089[0.063,0.116]×age represents the point estimate of the regression coefficient for age, which is 0.089 and the corresponding 95% confidence interval is [0.063,0.116]. 0(smoking_category="Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482](smoking_category="Previous smoking") indicates that smoking is categorized into three classes: Never, Smoking, and Previous smoking. The point estimate of the regression coefficient for the Never class is 0; the point estimate of the regression coefficient for the Smoking class is 0.385 with a corresponding 95% confidence interval of [-0.019,0.791]; and the point estimate of the regression coefficient for the Previous smoking class is 0.713 with a corresponding 95% confidence interval of [-0.102,1.482]. Only the point estimates of the regression coefficients for the selected classes are included in the calculation; the point estimates of the regression coefficients for other classes are considered 0. 0(Drinking="No")-0.501(Drinking="Yes") indicates that drinking behavior is divided into two categories: No and Yes; the point estimate of the regression coefficient for the No category is 0, and the point estimate of the regression coefficient for the Yes category is -0.501. 0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicbronchitisandemphysemahistory="Yes") indicates that the history of chronic bronchitis and emphysema is divided into two categories: No and Yes. The point estimate of the regression coefficient for the No category is 0, and the point estimate of the regression coefficient for the Yes category is 1.389 with a corresponding 95% confidence interval of [0.801,1.987]. 0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes") indicates that the airflow obstruction situation is divided into two categories: No and Yes; where the point estimate of the regression coefficient for the No category is 0; the point estimate of the regression coefficient for the Yes category is 0.619 and the corresponding 95% confidence interval is [0.158,1.069]. 0.018[0.001,0.035]×mean_dbp represents the point estimate of the regression coefficient of mean diastolic blood pressure as 0.018 and the corresponding 95% confidence interval as [0.001,0.035]. 0.013[-0.002,0.027]×heart_rate represents the point estimate of the regression coefficient of heart rate, which is 0.013 and the corresponding 95% confidence interval is [-0.002,0.027]. 0(BMI="Nomal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity") indicates that BMI is divided into four categories: Normal, Underweight, Overweight, and Obesity; where the point estimate of the regression coefficient for the Normal category is 0; Underweight, Overweight, and Obesity are all considered to be 0. The point estimate of the regression coefficient for the erweight category is 0.505, and the corresponding 95% confidence interval is [-0.210, 1.211]; the point estimate of the regression coefficient for the overweight category is -0.708, and the corresponding 95% confidence interval is [-1.164, -0.265]; the point estimate of the regression coefficient for the obesity category is -0.772, and the corresponding 95% confidence interval is [-1.387, -0.195]. During the calculation, only the point estimates of the regression coefficients of the selected categories are included in the calculation, and the point estimates of the regression coefficients of other categories are considered to be 0.

[0008] Furthermore, the point estimate of the regression coefficient of the intercept term is -9.511 and the corresponding 95% confidence interval is [-11.920, -7.227].

[0009] Furthermore, the step of inputting the conventional indicators into a preset COPD risk prediction model to predict the disease risk of the person to be predicted includes: The conventional indicators are input into a preset COPD risk prediction model to calculate logit(p); Substitute logit(p) into the formula p=1 / {1+exp[-logit(p)]} to calculate the disease risk of the person to be predicted.

[0010] Furthermore, the conventional indicators are input through a nomogram tool, and the input process is simultaneously visualized; wherein, the nomogram tool is developed based on a preset COPD risk prediction model.

[0011] Furthermore, the nomogram tool includes input boxes for the eight influencing factors displayed in order of importance, with the age input box at the top.

[0012] Furthermore, the displayed predicted risk of disease includes: displaying the calculated probability of disease and the corresponding 95% confidence interval, so as to clarify that the actual probability of disease of the person to be predicted falls between the lower and upper limits of the corresponding 95% confidence interval with a 95% confidence level.

[0013] Furthermore, the 95% confidence interval corresponding to the probability of disease is formed by estimating the standard error of the linear prediction value of the preset COPD risk prediction model based on the covariance matrix of the preset COPD risk prediction model, and converting the standard error into an interval range of the probability of disease.

[0014] Secondly, the present invention also proposes an electronic device, which is an embedded device, comprising: Memory, used to store program instructions; and A processor for executing the program instructions to implement the steps of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions as described above.

[0015] Thirdly, the present invention also proposes a storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions as described above.

[0016] Beneficial Effects: This method, electronic device, and storage medium for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions selects the most basic general demographic characteristics and vital signs as routine indicators, including age, smoking status, alcohol consumption, history of chronic bronchitis and emphysema, airflow obstruction, mean diastolic blood pressure, heart rate, and BMI. These routine indicators can be obtained without relying on expensive specialized equipment (such as imaging equipment) or complex biomarker detection, overcoming the over-reliance on complex clinical indicators and greatly reducing the cost and difficulty of data collection. In the process of predicting COPD risk, simply inputting the routine indicators of the person to be predicted into the preset COPD risk prediction model can predict and display the risk of the person's disease. This preset COPD risk prediction model is built based on the XGBoost model and the Logistic Regression model. The XGBoost model, with its strong nonlinear fitting ability and high prediction accuracy, serves as the main prediction model to achieve efficient and accurate COPD risk prediction; while the Logistic Regression model serves as an auxiliary explanatory model. Furthermore, logistic regression, as a mature and computationally efficient statistical method, is easy to understand and maintain, and can be quickly deployed and applied in primary healthcare institutions or community health service centers, significantly improving the applicability and promotional value of the model. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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 the structures shown in these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to the present invention. Figure 2 ROC curves for four machine learning models on the training set; Figure 3 ROC curves of four machine learning models on the validation set; Figure 4 This is a flowchart illustrating the second embodiment of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to the present invention. Figure 5 This is a nomogram illustrating the risk prediction method for chronic obstructive pulmonary disease in cold regions according to the present invention. Figure 6 A visual interface for applying a pre-defined COPD risk prediction model; Figure 7 A visualization of the importance of each influencing factor in a pre-defined COPD risk prediction model; Figure 8 SHAP analysis diagram of each influencing factor in the preset COPD risk prediction model; Figure 9 This is a schematic diagram of the structure of an electronic device in the hardware operating environment involved in the embodiments of the present invention.

[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0022] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the term "and / or" throughout the text includes three solutions; taking A and / or B as an example, it includes technical solution A, technical solution B, and a technical solution that simultaneously satisfies A and B. Furthermore, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of a person skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0023] Please see Figure 1 The diagram shown is a flowchart of the first embodiment of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to the present invention.

[0024] In this embodiment, the method for predicting the risk of chronic obstructive pulmonary disease in cold regions includes: Step S100: Obtain the routine indicators of the person to be predicted.

[0025] The conventional indicators are eight influencing factors, namely age, smoking status, alcohol consumption, history of chronic bronchitis and emphysema, airflow obstruction, mean diastolic blood pressure, heart rate, and BMI.

[0026] Specifically, this invention is based on the natural population cohort of chronic diseases in Heilongjiang Province. It collected 224 newly diagnosed patients with chronic obstructive pulmonary disease from two hospitals in Mingshui County, Suihua City, Heilongjiang Province between 2019 and 2025, and selected 896 control individuals from the natural population cohort of chronic diseases in Heilongjiang Province by random sampling at a ratio of 1:4.

[0027] Based on relevant literature and professional knowledge, we have preliminarily selected risk factors associated with COPD, including: gender, age, education level, marital status, occupation type, annual family income level, smoking status (current smoker, quit smoker, never smoker), smoking level (smoking index = number of cigarettes smoked per day × number of years of smoking; smoking index ≤ 200 is light smoking, 200 < smoking index < 400 is moderate smoking, and smoking index ≥ 400 is heavy smoking), passive smoking, alcohol consumption, physical exercise, history of hypertension, history of diabetes, history of chronic bronchitis and emphysema, history of asthma, airflow obstruction (%FEV1 < 70%), systolic blood pressure, mean diastolic blood pressure, heart rate, BMI (Body Mass Index), frequency of cooking at home, type of fuel used for cooking, and use of range hood.

[0028] First, the rank-sum test and chi-square test (Fisher's exact test was used when the theoretical frequency was less than 5) were employed to analyze the differences in risk factors between the case group (224 COPD patients) and the control group (896 control individuals). The results showed statistically significant differences between the case group and the control group in terms of age, education level, annual household income, smoking status, smoking level, alcohol consumption, history of chronic bronchitis and emphysema, history of asthma, airflow obstruction, systolic blood pressure, mean diastolic blood pressure, heart rate, BMI, and use of a range hood. Then, optimal regression models were constructed using forward elimination, backward elimination, and forward-backward elimination (also known as stepwise elimination) methods to further screen influencing factors. The screening results showed that the regression models constructed using backward elimination and forward-backward elimination included the same influencing factors, and the AIC (Akaike Information Criterion) values ​​(AIC=719.17) were all lower than those constructed using forward elimination (AIC=727.42). Furthermore, a multicollinearity test was performed on the regression model. The results showed that the VIF (variance inflation factor) values ​​for all variables were less than 2, indicating that the multicollinearity problem was not serious and the constructed regression model was in good condition. Finally, the optimal regression model was confirmed to include eight influencing factors: age, smoking status, alcohol consumption, history of chronic bronchitis and emphysema, airflow obstruction, mean diastolic blood pressure, heart rate, and BMI.

[0029] Step S200: Input the conventional indicators into the preset COPD risk prediction model to predict the disease risk of the person to be predicted and display the predicted disease risk.

[0030] The preset COPD risk prediction model is constructed based on the XGBoost model and the Logistic Regression model.

[0031] Specifically, the data input module acquires the routine indicators of the individuals to be predicted and inputs these indicators into a preset COPD risk prediction model. The calculation module automatically calculates and predicts the disease risk of the individuals based on the routine indicators and the preset COPD risk prediction model. The display module presents the predicted disease risk in real time through a graphical interface. This method for predicting the risk of chronic obstructive pulmonary disease in cold regions combines specific computer hardware and software to achieve accurate prediction and intuitive display of COPD risk.

[0032] Specifically, the sample information of 1,120 subjects (224 COPD patients and 896 control individuals) was divided into training set and validation set in an 8:2 ratio. Based on the eight confirmed influencing factors, four machine learning models (Logistic Regression, RandomForest, XGBoost and Decision Tree) were selected to construct a preset COPD risk prediction model.

[0033] The predictive performance of the four machine learning models on the training and validation sets was evaluated using ROC (Receiver Operating Characteristic) curves. The evaluation results show that on the training set, the XGBoost model had the highest AUC (Area Under the Curve) value (0.8673), sensitivity (0.8095), specificity (0.7720), and accuracy (0.7792); followed by the RandomForest model (AUC = 0.8559). On the validation set, the Logistic Regression model had the highest AUC value (0.7468), sensitivity (0.6190), specificity (0.7784), and accuracy (0.7477); followed by the XGBoost model (AUC = 0.7083). A comprehensive comparison of the performance of these four machine learning models confirms that the XGBoost and Logistic Regression models exhibit superior predictive performance and possess high application value.

[0034] Specifically, the prediction performance of the four machine learning models on the training and validation sets is as follows: Table 1 Comparison of Predictive Performance of Four Machine Learning Models on the Training Set Table 1 shows the predictive performance metrics of the four machine learning models on the training set. The training set was used to train the models, and these predictive performance metrics reflect the fit of the four machine learning models on the available data.

[0035] Table 2 Comparison of Predictive Performance of Four Machine Learning Models on the Validation Set Table 2 shows the predictive performance metrics of the four machine learning models on the validation set. The validation set is used to evaluate the model's generalization ability and typically contains data that has not been seen during model training. These predictive performance metrics reflect the performance of the four machine learning models on new data.

[0036] By comparing the specific predictive performance metrics of the four machine learning models during the training and validation phases, a more comprehensive evaluation of the effectiveness and stability of different models can be achieved.

[0037] Please see Figure 2 and Figure 3 The figures show the ROC curves of four machine learning models on the training and validation sets, respectively. The ROC curve (Receiver Operating Characteristic Curve) is used to evaluate the classification performance of a model and reflects its ability to distinguish data on a dataset. Comparing the different ROC performances of the four machine learning models during the training and validation phases allows for a more comprehensive evaluation of the overall performance of each model.

[0038] Compared with existing technologies, the method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions according to this invention collected sample information from 1120 subjects in two hospitals in Heilongjiang Province, including 224 COPD patients and 896 healthy controls. The collected routine indicators covered eight influencing factors: age, smoking status, alcohol consumption, history of chronic bronchitis and emphysema, airflow obstruction, mean diastolic blood pressure, heart rate, and BMI. In the construction phase of the pre-defined COPD risk prediction model, the study first used the eight influencing factors identified through precise screening. Then, the sample information was proportionally divided into training and validation sets, and four machine learning models—Logistic Regression, Random Forest, XGBoost, and Decision Tree—were used for modeling. After comprehensively comparing the overall performance of the four models, the XGBoost model (highest prediction accuracy) and the Logistic Regression model (strong generalization ability and easy interpretation) were ultimately selected as the best prediction models. The XGBoost model was selected as the primary prediction model to achieve the highest prediction accuracy, while the Logistic Regression model was retained for clinical interpretation and assessment of generalization ability.

[0039] This method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions can also be based on a Logistic Regression model, integrating eight influencing factors and transforming them into an intuitive graphical scoring tool, namely a nomogram. Doctors or users simply draw lines on the graph according to the individual's routine indicators and score them, converting the total score into a COPD risk (this COPD risk is based on the risk predicted by the pre-set COPD risk prediction model). Furthermore, using SHAP analysis to clarify the importance ranking of each influencing factor (with age being the most important) and the direction of its influence on the model's prediction can significantly improve the transparency and credibility of the pre-set COPD risk prediction model.

[0040] In some embodiments of this application, the preset COPD risk prediction model is: logit(p) = intercept term + 0.089[0.063, 0.116] × age +0(smoking_category="Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482](smoking_category="Previous smoking") +0(Drinking="No")-0.501(Drinking="Yes") +0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicbronchitisandemphysemahistory="Yes") +0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes") +0.018[0.001,0.035]×mean_dbp+0.013[-0.002,0.027]×heart_rate +0(BMI="Normal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity"); Wherein, logit(p) represents the linear prediction value of the preset COPD risk prediction model, and the intercept term includes the point estimate of the regression coefficient and the corresponding 95% confidence interval; 0.089[0.063,0.116]×age represents the point estimate of the regression coefficient for age, which is 0.089 and the corresponding 95% confidence interval is [0.063,0.116]. 0(smoking_category="Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482](smoking_category="Previous smoking") indicates that smoking is categorized into three classes: Never, Smoking, and Previous smoking. The point estimate of the regression coefficient for the Never class is 0; the point estimate of the regression coefficient for the Smoking class is 0.385 with a corresponding 95% confidence interval of [-0.019,0.791]; and the point estimate of the regression coefficient for the Previous smoking class is 0.713 with a corresponding 95% confidence interval of [-0.102,1.482]. Only the point estimates of the regression coefficients for the selected classes are included in the calculation; the point estimates of the regression coefficients for other classes are considered 0. 0(Drinking="No")-0.501(Drinking="Yes") indicates that drinking behavior is divided into two categories: No and Yes; the point estimate of the regression coefficient for the No category is 0, and the point estimate of the regression coefficient for the Yes category is -0.501. 0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicbronchitisandemphysemahistory="Yes") indicates that the history of chronic bronchitis and emphysema is divided into two categories: No and Yes. The point estimate of the regression coefficient for the No category is 0, and the point estimate of the regression coefficient for the Yes category is 1.389 with a corresponding 95% confidence interval of [0.801,1.987]. 0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes") indicates that the airflow obstruction situation is divided into two categories: No and Yes; where the point estimate of the regression coefficient for the No category is 0; the point estimate of the regression coefficient for the Yes category is 0.619 and the corresponding 95% confidence interval is [0.158,1.069]. 0.018[0.001,0.035]×mean_dbp represents the point estimate of the regression coefficient of mean diastolic blood pressure as 0.018 and the corresponding 95% confidence interval as [0.001,0.035]. 0.013[-0.002,0.027]×heart_rate represents the point estimate of the regression coefficient of heart rate, which is 0.013 and the corresponding 95% confidence interval is [-0.002,0.027]. 0(BMI="Nomal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity") indicates that BMI is divided into four categories: Normal, Underweight, Overweight, and Obesity; where the point estimate of the regression coefficient for the Normal category is 0; Underweight, Overweight, and Obesity are all considered to be 0. The point estimate of the regression coefficient for the erweight category is 0.505, and the corresponding 95% confidence interval is [-0.210, 1.211]; the point estimate of the regression coefficient for the overweight category is -0.708, and the corresponding 95% confidence interval is [-1.164, -0.265]; the point estimate of the regression coefficient for the obesity category is -0.772, and the corresponding 95% confidence interval is [-1.387, -0.195]. During the calculation, only the point estimates of the regression coefficients of the selected categories are included in the calculation, and the point estimates of the regression coefficients of other categories are considered to be 0.

[0041] Specifically, the preset COPD risk prediction model was developed specifically for the characteristics of residents in cold regions of China, making it more suitable for the local environment and lifestyle; and it integrates multiple machine learning models, selecting the best among the best, to ensure the accuracy and stability of the prediction.

[0042] In some embodiments of this application, the point estimate of the regression coefficient of the intercept term is -9.511 and the corresponding 95% confidence interval is [-11.920, -7.227].

[0043] Specifically, the intercept is an important parameter in the Logistic Regression model, typically representing the predicted value of the dependent variable (target variable) when all independent variables (features) are zero. The default COPD risk prediction model is as follows: logit(p)=-9.511[-11.920,-7.227]+0.089[0.063,0.116]×age+0(smoking_category= "Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482 ](smoking_category="Previoussmoking")+0(Drinking="No")-0.501(Drinking="Yes")+0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicb ronchitisandemphysemahistory="Yes")+0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes")+0.018[0.001,0.035]×mean_dbp+0.013[-0.00 2,0.027]×heart_rate+0(BMI="Normal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity").

[0044] Please see Figure 4 The diagram shown is a flowchart illustrating a second embodiment of the method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions according to the present invention. This embodiment provides a detailed description of the steps for calculating and displaying the risk of COPD in individuals to be predicted, compared to other embodiments of the method for predicting the risk of COPD in cold regions.

[0045] In this embodiment, the method for predicting the risk of chronic obstructive pulmonary disease in cold regions includes: Step S100: Obtain the routine indicators of the person to be predicted.

[0046] The conventional indicators are eight influencing factors, namely age, smoking status, alcohol consumption, history of chronic bronchitis and emphysema, airflow obstruction, mean diastolic blood pressure, heart rate, and BMI.

[0047] Specifically, the method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions belongs to the field of risk prediction models and is suitable for preliminary screening of COPD risk in populations in cold regions. This method breaks through the dependence on complex data and innovatively utilizes the most basic and readily available general demographic characteristics and vital signs to construct a tool for preliminary COPD screening and risk prediction applicable to large-scale populations in cold regions.

[0048] Step S201: Input the conventional indicators into the preset COPD risk prediction model to calculate logit(p); substitute logit(p) into the formula p=1 / {1+exp[-logit(p)]} to calculate the disease risk of the person to be predicted.

[0049] Specifically, the calculated risk of disease includes the probability of disease and the corresponding 95% confidence interval, the latter reflecting the statistical uncertainty of the prediction result.

[0050] Step S202: Display the calculated probability of disease and the corresponding 95% confidence interval to confirm that the true probability of disease of the person to be predicted falls between the lower and upper limits of the corresponding 95% confidence interval with a 95% confidence level.

[0051] The method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions according to the present invention overcomes the shortcomings of existing COPD risk prediction models, which are highly dependent on complex clinical indicators or environmental data and have limited applicability. It provides a COPD early screening and risk prediction model specifically for large-scale populations in cold regions.

[0052] In some embodiments of this application, the 95% confidence interval corresponding to the probability of disease is formed by estimating the standard error of the linear prediction value of the preset COPD risk prediction model based on the covariance matrix of the preset COPD risk prediction model, and converting the standard error into an interval range of the probability of disease.

[0053] Specifically, the linear predictor is logit(p). Displaying the 95% confidence interval corresponding to the probability of developing the disease is to facilitate a comprehensive assessment of the COPD risk in the individuals being predicted.

[0054] Please see Figure 5 The diagram shown is a nomogram illustrating the risk prediction method for chronic obstructive pulmonary disease (COPD) in cold regions according to the present invention. Please refer to [link / reference]. Figure 6 As shown, it is a visualization interface for applying a preset COPD risk prediction model.

[0055] In some embodiments of this application, the conventional indicators are input through a nomogram tool, and the input process is simultaneously visualized; wherein, the nomogram tool is developed based on a preset COPD risk prediction model.

[0056] Specifically, a nomogram is an intuitive and simple tool that integrates multiple risk factors to graphically predict the probability of an event. It is developed based on statistical or machine learning models, facilitating clinical application and personalized risk assessment. Figure 5 As shown, in this embodiment, the nomogram tool is developed based on a preset COPD risk prediction model. It is used to calculate the probability of COPD in an individual based on the different positions of eight influencing factors on the graph, facilitating personalized risk assessment and decision-making by clinicians. The nomogram tool can also be combined with a dynamic interactive interface or web application to display the predicted COPD risk in real time, allowing clinicians and patients to intuitively view and understand the risk level. In other words, the preset COPD risk prediction model can be used to build an interactive COPD prediction tool using RStudio software. Figure 6 As shown, this interface is a visualization interface that combines the nomogram tool with a dynamic interactive interface. It adopts a graphical user interface (GUI) design and uses intuitive elements such as graphics, charts, and buttons to facilitate users to input influencing factors and view the predicted risk of disease, thereby improving the ease of use and practical value of the preset COPD risk prediction model.

[0057] The nomogram tool is specifically built based on the Logistic Regression model. By scoring eight influencing factors on the nomogram and calculating the total score, the corresponding probability of COPD can be determined on the probability line. This nomogram tool is a visual risk calculator that is easy to use, requires no complex calculations, and is convenient for promotion and application in primary healthcare institutions. It provides practical technical support for conducting large-scale preliminary screening and early intervention for COPD in resource-limited environments.

[0058] In some embodiments of this application, the nomogram tool includes input boxes for the eight influencing factors displayed in order of importance, with the age input box at the top.

[0059] Please see Figure 7 The image shown is a visualization of the importance of various influencing factors in the pre-defined COPD risk prediction model. Please refer to [link / reference]. Figure 8 As shown, it is a SHAP analysis diagram of each influencing factor in the preset COPD risk prediction model.

[0060] Specifically, the Shap value was used to interpret the prediction results of the pre-defined COPD risk prediction model and to rank the importance of influencing factors. The results are as follows: Figure 7 As shown, the influencing factors are ranked from most important to least important: age, heart rate, history of chronic bronchitis and emphysema, BMI, mean diastolic blood pressure, airflow obstruction, smoking, and alcohol consumption. Among these, [the following is a list of factors]. Figure 8 As shown, the Shp values ​​for age, heart rate, history of chronic bronchitis and emphysema, mean diastolic blood pressure, and airway obstruction are widely distributed and mostly concentrated in the positive influence area, having a significant positive impact on the prediction results. That is, the older the age, the higher the heart rate, the history of chronic bronchitis and emphysema, the higher the mean diastolic blood pressure, and the more likely the person to be predicted has chronic obstructive pulmonary disease.

[0061] The COPD risk prediction method provided in this embodiment combines innovative variable screening, multi-model comparison, SHAP interpretation, and visualization tools to construct a high-precision, interpretable, and clinically practical COPD risk prediction system, which significantly improves the accuracy and efficiency of early COPD screening and risk assessment.

[0062] The method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions provided in this embodiment is unique in that it uses only general population characteristics and basic vital signs such as heart rate as routine indicators to construct an efficient and convenient early COPD screening tool. It aims to achieve rapid and low-cost preliminary identification and stratification of individual COPD risk in cold regions, and provide practical technical support for carrying out large-scale population COPD early screening and early intervention in resource-limited scenarios.

[0063] The following are embodiments of the electronic device provided by the present invention. These embodiments of the electronic device belong to the same concept as the embodiments of the method for predicting the risk of chronic obstructive pulmonary disease (COPD) in cold regions described above. Details not fully described in the embodiments of the electronic device can be found in the embodiments of the method for predicting the risk of COPD in cold regions described above.

[0064] Please see Figure 9 As shown, it is a schematic diagram of the structure of an electronic device in the hardware operating environment involved in the embodiment of the present invention.

[0065] In this embodiment, an electronic device includes: Memory 1005 is used to store program instructions; and Processor 1001 is configured to execute the program instructions to implement the steps of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions as described above.

[0066] The electronic device in this embodiment of the invention can be a desktop computer, laptop, handheld computer, server, or other computing device. Figure 9 As shown, the electronic device may include: a processor 1001 (e.g., a CPU), a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen and an input unit, such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0067] Those skilled in the art will understand that Figure 9 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0068] like Figure 9 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and computer programs.

[0069] exist Figure 9 In the electronic device shown, the network interface 1004 is mainly used to connect to the backend server and communicate with the backend server; the user interface 1003 is mainly used to connect to the client (user end) and communicate with the client; and the processor 1001 can be used to call the computer program stored in the memory 1005. When the computer program is called and executed by the processor 1001, it implements the steps of the above-mentioned method for predicting the risk of chronic obstructive pulmonary disease in cold regions.

[0070] The following are embodiments of the storage medium provided by the present invention. The embodiments of the storage medium belong to the same concept as the embodiments of the method and electronic device for predicting the risk of chronic obstructive pulmonary disease in cold regions described above. Details not fully described in the embodiments of the storage medium can be found in the embodiments of the method and electronic device for predicting the risk of chronic obstructive pulmonary disease in cold regions described above.

[0071] In this embodiment, a storage medium stores computer-executable instructions, which, when executed by a processor, implement the steps of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions as described above.

[0072] Since the electronic device and storage medium of the present invention can implement the steps of the above-described method for predicting the risk of chronic obstructive pulmonary disease in cold regions, they at least have all the beneficial effects brought about by the technical solutions of the above-described step embodiments of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions, which will not be repeated here.

[0073] The above description is only a part or preferred embodiment of the present invention. Neither the text nor the drawings should limit the scope of protection of the present invention. All equivalent structural transformations made using the content of the present invention specification and drawings under the overall concept of the present invention, or direct / indirect applications in other related technical fields, are included within the scope of protection of the present invention.

[0074] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0075] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0076] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0077] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting the risk of chronic obstructive pulmonary disease in cold regions, characterized in that, include: Obtain the routine indicators of the person to be predicted; wherein, the routine indicators are eight influencing factors, namely, age, smoking status, drinking status, history of chronic bronchitis and emphysema, airflow obstruction status, mean diastolic blood pressure, heart rate and BMI; The conventional indicators are input into a preset COPD risk prediction model to predict the risk of disease in the individuals to be predicted and to display the predicted risk; wherein, the preset COPD risk prediction model is constructed based on the XGBoost model and the LogisticRegression model.

2. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 1, characterized in that, The preset COPD risk prediction model is as follows: logit(p) = intercept term + 0.089[0.063, 0.116] × age +0(smoking_category="Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482](smoking_category="Previoussmoking") +0(Drinking="No")-0.501(Drinking="Yes") +0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicbronchitisandemphysemahistory="Yes") +0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes") +0.018[0.001,0.035]×mean_dbp+0.013[-0.002,0.027]×heart_rate +0(BMI="Normal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity"); Wherein, logit(p) represents the linear prediction value of the preset COPD risk prediction model, and the intercept term includes the point estimate of the regression coefficient and the corresponding 95% confidence interval; 0.089[0.063,0.116]×age represents the point estimate of the regression coefficient for age, which is 0.089 and the corresponding 95% confidence interval is [0.063,0.116]. 0(smoking_category="Never")+0.385[-0.019,0.791](smoking_category="smoking")+0.713[-0.102,1.482](smoking_category="Previoussmoking") indicates that smoking is categorized into three classes: Never, Smoking, and Previoussmoking. The point estimate of the regression coefficient for the Never class is 0; the point estimate of the regression coefficient for the Smoking class is 0.385 with a corresponding 95% confidence interval of [-0.019,0.791]; and the point estimate of the regression coefficient for the Previoussmoking class is 0.713 with a corresponding 95% confidence interval of [-0.102,1.482]. During calculation, only the point estimates of the regression coefficients for the selected classes are included in the calculation; the point estimates of the regression coefficients for other classes are considered 0. 0(Drinking="No")-0.501(Drinking="Yes") indicates that drinking behavior is divided into two categories: No and Yes; the point estimate of the regression coefficient for the No category is 0, and the point estimate of the regression coefficient for the Yes category is -0.

501. 0(Chronicbronchitisandemphysemahistory="No")+1.389[0.801,1.987](Chronicbronchitisandemphysemahistory="Yes") indicates that the history of chronic bronchitis and emphysema is divided into two categories: No and Yes. The point estimate of the regression coefficient for the No category is 0, and the point estimate of the regression coefficient for the Yes category is 1.389 with a corresponding 95% confidence interval of [0.801,1.987]. 0(Airflow_obstruction="No")+0.619[0.158,1.069](Airflow_obstruction="Yes") indicates that the airflow obstruction situation is divided into two categories: No and Yes; where the point estimate of the regression coefficient for the No category is 0; the point estimate of the regression coefficient for the Yes category is 0.619 and the corresponding 95% confidence interval is [0.158,1.069]. 0.018[0.001,0.035]×mean_dbp represents the point estimate of the regression coefficient of mean diastolic blood pressure as 0.018 and the corresponding 95% confidence interval as [0.001,0.035]. 0.013[-0.002,0.027]×heart_rate represents the point estimate of the regression coefficient of heart rate, which is 0.013 and the corresponding 95% confidence interval is [-0.002,0.027]. 0(BMI="Nomal")+0.505[-0.210,1.211](BMI="Underweight")-0.708[-1.164,-0.265](BMI="Overweight")-0.772[-1.387,-0.195](BMI="Obesity") indicates that BMI is divided into four categories: Normal, Underweight, Overweight, and Obesity; where the point estimate of the regression coefficient for the Normal category is 0; Underweight, Overweight, and Obesity are all considered to be 0. The point estimate of the regression coefficient for the erweight category is 0.505, and the corresponding 95% confidence interval is [-0.210, 1.211]; the point estimate of the regression coefficient for the overweight category is -0.708, and the corresponding 95% confidence interval is [-1.164, -0.265]; the point estimate of the regression coefficient for the obesity category is -0.772, and the corresponding 95% confidence interval is [-1.387, -0.195]. During the calculation, only the point estimates of the regression coefficients of the selected categories are included in the calculation, and the point estimates of the regression coefficients of other categories are considered to be 0.

3. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 2, characterized in that, The point estimate of the regression coefficient of the intercept term is -9.511 and the corresponding 95% confidence interval is [-11.920, -7.227].

4. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 3, characterized in that, The step of inputting the conventional indicators into a preset COPD risk prediction model to predict the disease risk of the individuals to be predicted includes: The conventional indicators are input into a preset COPD risk prediction model to calculate logit(p); Substitute logit(p) into the formula p=1 / {1+exp[-logit(p)]} to calculate the disease risk of the person to be predicted.

5. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 1, characterized in that, The conventional indicators are input through a nomogram tool, and the input process is simultaneously visualized; the nomogram tool is developed based on a preset COPD risk prediction model.

6. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 5, characterized in that, The nomogram tool includes input boxes for the eight influencing factors, which are displayed in order of importance, with the age input box at the top.

7. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 1, characterized in that, The displayed predicted risk of disease includes: displaying the calculated probability of disease and the corresponding 95% confidence interval, so as to clarify that the actual probability of disease of the person to be predicted falls between the lower and upper limits of the corresponding 95% confidence interval with a 95% confidence level.

8. The method for predicting the risk of chronic obstructive pulmonary disease in cold regions according to claim 7, characterized in that, The 95% confidence interval corresponding to the probability of disease is formed by estimating the standard error of the linear prediction value of the preset COPD risk prediction model and converting the standard error into an interval range of the probability of disease based on the covariance matrix of the preset COPD risk prediction model.

9. An electronic device, wherein the electronic device is an embedded device, characterized in that, include: Memory, used to store program instructions; as well as A processor for executing the program instructions to implement the steps of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium stores computer-executable instructions, which, when executed by a processor, implement the steps of the method for predicting the risk of chronic obstructive pulmonary disease in cold regions as described in any one of claims 1 to 8.

Images