A method for studying the similarities and differences between left and right heart failure in cardiopulmonary exercise testing.

Abstract

This invention discloses a method for studying the similarities and differences between left and right heart failure in cardiopulmonary exercise testing (CPET), relating to the field of data analysis technology. The method includes: S1: Selection and grouping of study subjects: Selecting heart failure patients with stable clinical symptoms and enrolling them according to New York Heart Classification II-IV; dividing patients into left and right heart failure groups based on their clinical diagnosis; S2: Standardization of CPET operation: Performing quality control on the equipment, including environmental control and regular equipment calibration; subjects first undergo static pulmonary function testing, followed by testing using the testing device. This invention significantly improves the accuracy, reliability, and clinical applicability of CPET data analysis by introducing and improving multivariate statistical methods and machine learning algorithms. Compared with traditional methods, it not only allows for a deeper understanding of the complex physiological characteristics exhibited by heart failure patients in CPET but also provides more accurate predictive models.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a method for studying the similarities and differences between left and right heart failure in cardiopulmonary exercise tests. Background Technology

[0002] In studies exploring the application of cardiopulmonary exercise testing (CPET) in patients with heart failure, traditional methods often focus on single physiological parameters or simple statistical analyses, which limits a comprehensive understanding of the complexity of the disease.

[0003] A search revealed Chinese patent application CN202410268819.X, which discloses a multi-parameter cardiopulmonary function testing device and its data processing method, belonging to the field of electronic digital data processing technology. The method includes: Step 1: acquiring the testing mode of the cardiopulmonary function testing device; Step 2: acquiring first acquisition parameters based on the cardiopulmonary exercise test protocol and the testing mode; Step 3: acquiring a regression data model; Step 4: acquiring second acquisition parameters based on the regression data model and according to analysis requirements; Step 5: performing data processing based on the parameter type of the second acquisition parameters. The data processing method in the aforementioned patent has the following shortcomings: the accuracy of data analysis is insufficient, and the universality of the data processing strategy is also lacking. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for studying the similarities and differences between left and right heart failure in cardiopulmonary exercise tests.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for studying the differences between left and right heart failure in cardiopulmonary exercise testing includes:

[0007] S1: Selection and grouping of study subjects: Heart failure patients with stable clinical symptoms were selected and enrolled according to New York Heart Classification II to IV; based on the patients' clinical diagnosis results, the patients were divided into left heart failure group and right heart failure group;

[0008] S2: CPET Operation Standardization: Quality control of equipment, including environmental control and regular equipment calibration; subjects first undergo static lung function tests, followed by testing through the testing device, with respiratory, electrocardiogram, and blood pressure vital signs monitored throughout the process;

[0009] S3: Data Collection and Analysis: Based on the CPET index, conduct continuous dynamic monitoring and detailed recording of blood oxygen saturation, electrocardiogram changes, and gas exchange rate; use statistical software to process the data; introduce partial correlation analysis to evaluate the correlation between different variables;

[0010] S4: Results Analysis: Comparative analysis of the differences in performance between patients with left heart failure and those with right heart failure in CPET.

[0011] Preferably, in step S1, the selection and grouping of research subjects includes:

[0012] S11: Screening criteria: Inclusion and exclusion criteria are established based on medical history, clinical symptoms and preliminary examination results;

[0013] S12: Clinical assessment: Clinical assessment of patients who initially meet the screening criteria;

[0014] S13: Grouping: Patients are divided into corresponding study groups based on clinical assessment results;

[0015] S14: Baseline Characteristics Recording: Record the patient's baseline characteristics in detail;

[0016] In S2, the standardization of CPET operation includes:

[0017] S21: Static pulmonary function test: Subjects undergo static pulmonary function tests in a seated position to measure vital capacity, slow vital capacity, and maximum minute ventilation.

[0018] S22: Dynamic stress test: Set the parameters of the test device according to the patient's actual condition;

[0019] S23: Full-process monitoring and recording: Starting from the resting state, continuously monitor and record the patient's heart rate, blood pressure, respiratory rate, and blood oxygen saturation until the end of the recovery period.

[0020] Preferably, in step S3, data collection and analysis includes:

[0021] S31: Data Acquisition: Acquire CPET indexes and perform continuous dynamic monitoring of blood oxygen saturation, electrocardiogram changes, and gas exchange rate indicators;

[0022] S32: Statistical Analysis: Use statistical software to process data and employ independent samples t-test, analysis of variance, and partial correlation analysis to analyze the relationships between data.

[0023] S33: Subgroup analysis: Subgroup analysis was conducted based on the baseline characteristics of patients to explore the differences in CPET indices among different subgroups and their potential influencing factors.

[0024] S34: Sensitivity Analysis: Monte Carlo simulation was used to assess the confidence intervals and sensitivity of the statistical data.

[0025] Preferably, in step S32, the statistical analysis includes:

[0026] Multivariate analysis of variance was used to explore the relationship between multiple response variables and grouping variables;

[0027] Based on the random forest algorithm, feature selection and importance scoring are performed to identify the most important parameters for classifying left and right heart failure in CPET.

[0028] Regression analysis was used to evaluate the predictive ability of CPET parameters for heart failure type.

[0029] Preferably, in S33, the subgroup analysis includes:

[0030] Based on factors such as age, gender, and severity of illness, patients were subgrouped according to CPET parameters using a clustering algorithm.

[0031] Comparative analysis of CPET parameters was performed within each subgroup.

[0032] Preferably, in step S32, multivariate analysis of variance is used to explore the relationship between multiple response variables and grouping variables, as follows:

[0033] Preparation phase: Collect CPET data, including continuous variables such as heart rate and blood pressure, and record the group information for each patient;

[0034] Data preprocessing: handling missing values ​​and performing variable transformations;

[0035] Perform MANOVA: In the statistical software, set heart rate and blood pressure as dependent variables and grouping variables as independent variables, and perform multivariate analysis of variance.

[0036] Results evaluation: Wilks' lambda value, F-value, and significance level were analyzed to assess differences between groups and correlations between response variables.

[0037] Preferably, in step S32, regression analysis is used to evaluate the predictive ability of CPET parameters for heart failure type, as follows:

[0038] Select variables: Select CPET parameters that affect the type of heart failure;

[0039] Model building: Using heart failure type as the dependent variable and CPET parameter as the independent variable, a logistic regression model is constructed;

[0040] Model evaluation: The explanatory power and predictive ability of the model are evaluated through goodness-of-fit index, prediction accuracy, and likelihood ratio test.

[0041] Model optimization: Select the optimal feature subset through stepwise regression or information-based criteria.

[0042] Preferably, in step S32, feature selection and importance scoring are performed based on the random forest algorithm, as follows:

[0043] Data preparation: All parameters of CPET were used as features, and the classification label of heart failure was used as the target variable;

[0044] Data set partitioning: The data is divided into a training set and a test set;

[0045] Model training: The model is trained using the random forest algorithm;

[0046] Feature importance assessment: Ranking the importance of each feature using feature importance scores provided by the algorithm;

[0047] Feature selection: Features with low importance scores are removed, and features that contribute more to the model's predictions are selected.

[0048] Preferably, in step S33, a decision tree is applied to discover key distinguishing features within the subgroups, as follows:

[0049] Subgroup identification: Patients are subgrouped based on clinical characteristics or CPET results;

[0050] Data preparation: For each subgroup, prepare the CPET parameters as input for the decision tree or GBM model;

[0051] Model building: Build a decision tree model or GBM model, and adjust the tree depth and learning rate to optimize performance;

[0052] Key feature identification: Analyze the model results to identify features that have key discriminative power for classifying heart failure within subgroups.

[0053] Preferably, in step S34, Monte Carlo simulation is used to evaluate the confidence interval and sensitivity of statistical data, as follows:

[0054] Define the analytical objective: Identify the statistical measures of uncertainty that need to be assessed;

[0055] Monte Carlo simulation: Set a distribution under preset conditions, perform random sampling, and calculate the statistics for each simulation;

[0056] Bootstrap resampling: It calculates the statistics obtained from each sampling by repeatedly sampling the original dataset.

[0057] Construct confidence intervals: Based on the simulation or resampling results, calculate the confidence intervals for the statistic, and take the 2.5% and 97.5% quantiles as the 95% confidence intervals;

[0058] Sensitivity analysis: By changing the preconditions of simulation or resampling, the sensitivity of the statistic to changes in the hypothesis is assessed.

[0059] The beneficial effects of this invention are as follows:

[0060] 1. This invention significantly improves the accuracy, reliability, and clinical applicability of CPET data analysis by introducing and improving multivariate statistical methods and machine learning algorithms.

[0061] 2. Compared with traditional methods, this invention can not only more deeply explore and understand the complex physiological characteristics of heart failure patients in CPET, but also provide a more accurate prediction model.

[0062] 3. This invention further enhances the stability and credibility of the research results through sensitivity analysis and uncertainty assessment. Attached Figure Description

[0063] Figure 1 This is a flowchart of a method proposed in this invention for studying the similarities and differences between left and right heart failure in cardiopulmonary exercise tests. Detailed Implementation

[0064] The technical solution of the present invention will be further described in detail below with reference to specific embodiments.

[0065] Example 1:

[0066] A method for studying the differences between left and right heart failure in cardiopulmonary exercise testing includes:

[0067] S1: Selection and grouping of study subjects: Heart failure patients with stable clinical symptoms were selected and enrolled according to New York Heart Classification II to IV; based on the patients' clinical diagnosis results, the patients were divided into left heart failure group and right heart failure group;

[0068] S2: CPET Operation Standardization: Quality control of equipment, including environmental control and regular equipment calibration; Subjects first undergo static lung function tests, and then undergo tests through testing devices, such as incremental load tests on a power bicycle, with vital signs such as respiration, electrocardiogram, and blood pressure monitored throughout the process;

[0069] S3: Data Collection and Analysis: Based on the CPET index, detailed records of continuous dynamic monitoring of blood oxygen saturation, electrocardiogram changes, and gas exchange rate were recorded; statistical software, such as SPSS 26.0, was used to process the data; partial correlation analysis was introduced to evaluate the correlation between different variables.

[0070] S4: Results Analysis: Comparative analysis of the differences in performance between patients with left heart failure and those with right heart failure in CPET.

[0071] In S1, the selection and grouping of research subjects includes:

[0072] S11: Screening criteria: Inclusion and exclusion criteria are established based on medical history, clinical symptoms and preliminary examination results;

[0073] S12: Clinical assessment: Clinical assessment of patients who initially meet the screening criteria, including but not limited to examinations such as electrocardiogram and echocardiography;

[0074] S13: Grouping: Patients are divided into corresponding study groups based on clinical assessment results;

[0075] S14: Baseline Characteristics Recording: Record the patient's baseline characteristics in detail, including age, gender, weight, height, disease course, etc., to provide basic data for subsequent analysis.

[0076] In S2, the standardization of CPET operation includes:

[0077] S21: Static pulmonary function test: Subjects undergo static pulmonary function tests such as vital capacity, slow vital capacity, and maximum minute ventilation while seated.

[0078] S22: Dynamic load test: Set the parameters of the test device according to the patient's actual situation, such as the starting load and rate of increase of the power bicycle;

[0079] S23: Full-process monitoring and recording: Starting from the resting state, continuously monitor and record the patient's heart rate, blood pressure, respiratory rate, blood oxygen saturation and other indicators until the end of the recovery period.

[0080] Specifically, in S3, data collection and analysis includes:

[0081] S31: Data Acquisition: Collect CPET index data and perform continuous dynamic monitoring of indicators such as blood oxygen saturation, electrocardiogram changes, and gas exchange rate;

[0082] S32: Statistical Analysis: Data processing was performed using statistical software such as SPSS 26.0, and methods such as independent samples t-test, analysis of variance, and partial correlation analysis were used to analyze the relationships between data.

[0083] S33: Subgroup analysis: Subgroup analysis was conducted based on the baseline characteristics of patients to explore the differences in CPET indices among different subgroups and their potential influencing factors.

[0084] S34: Sensitivity analysis: Use Monte Carlo simulation or bootstrap resampling to assess the confidence intervals and sensitivity of the statistical data.

[0085] In S32, the statistical analysis includes:

[0086] Multivariate analysis of variance (MANOVA) was used to explore the relationship between multiple response variables (such as heart rate and blood pressure) and grouping variables (left heart failure and right heart failure);

[0087] Based on the random forest algorithm, feature selection and importance scoring are performed to identify the most important parameters for classifying left and right heart failure in CPET.

[0088] Regression analysis was used to evaluate the predictive ability of CPET parameters for heart failure type (left heart failure and right heart failure).

[0089] In S33, the subgroup analysis includes:

[0090] Based on factors such as age, gender, and severity of illness, clustering algorithms, such as K-means or hierarchical clustering, are used to subgroup patients according to CPET parameters.

[0091] Comparative analysis of CPET parameters was performed within each subgroup.

[0092] In step S32, multivariate analysis of variance is used to explore the relationship between multiple response variables and grouping variables, as detailed below:

[0093] Preparation phase: Collect CPET data, including continuous variables such as heart rate and blood pressure, and record the grouping information (left heart failure and right heart failure) for each patient;

[0094] Data preprocessing: handling missing values ​​and performing variable transformations (e.g., logarithmic transformation of heart rate to meet analytical hypotheses);

[0095] Perform MANOVA: In the statistical software, set heart rate, blood pressure, etc. as dependent variables, and grouping variables (left heart failure and right heart failure) as independent variables, and perform multivariate analysis of variance.

[0096] Results evaluation: Wilks' lambda value, F-value, and significance level were analyzed to assess differences between groups and correlations between response variables.

[0097] In step S32, regression analysis is used to evaluate the predictive ability of the CPET parameter for heart failure type (left heart failure and right heart failure), as follows:

[0098] Select variables: Select CPET parameters that affect the type of heart failure (such as peak oxygen uptake, heart rate recovery time, etc.);

[0099] Model building: Using heart failure type as the dependent variable and CPET parameter as the independent variable, a logistic regression model is constructed;

[0100] Model evaluation: The explanatory power and predictive ability of the model are evaluated through goodness-of-fit indices (such as R-squared), prediction accuracy, and likelihood ratio tests.

[0101] Model optimization: Select the optimal feature subset through stepwise regression methods or based on information criteria (such as AIC, BIC).

[0102] In step S32, feature selection and importance scoring are performed based on the random forest algorithm, as detailed below:

[0103] Data preparation: All parameters of CPET were used as features, and the classification labels of heart failure (left heart failure and right heart failure) were used as target variables;

[0104] Data set partitioning: Divide the data into a training set and a test set, with a ratio of 70% training set and 30% test set;

[0105] Model training: The model is trained using the random forest algorithm;

[0106] Feature importance assessment: Ranking the importance of each feature using feature importance scores provided by the algorithm;

[0107] Feature selection: Features with low importance scores are removed, and features that contribute more to the model's predictions are selected.

[0108] In step S33, decision trees are applied to discover key distinguishing features within subgroups, as detailed below:

[0109] Subgroup identification: Patients are subgrouped based on clinical characteristics or CPET results;

[0110] Data preparation: For each subgroup, prepare the CPET parameters as input for the decision tree or GBM model;

[0111] Model building: Build a decision tree model or GBM model, and adjust hyperparameters such as tree depth and learning rate to optimize performance;

[0112] Key feature identification: Analyze the model results to identify features that have key discriminative power for classifying heart failure within subgroups.

[0113] Specifically, in S34, Monte Carlo simulation is used to evaluate the confidence interval and sensitivity of statistical data, as follows:

[0114] Define the analysis objective: Identify the statistics that need to assess uncertainty, such as the mean difference of a certain CPET parameter;

[0115] Monte Carlo simulation: Set a distribution under preset conditions (such as normal distribution), perform random sampling, and calculate the statistics for each simulation;

[0116] Bootstrap resampling: It calculates the statistics obtained from each sampling by repeatedly sampling the original dataset (with replacement);

[0117] Constructing confidence intervals: Based on the simulation or resampling results, calculate the confidence intervals for the statistic, typically using the 2.5% and 97.5% quantiles as the 95% confidence intervals;

[0118] Sensitivity analysis: By changing the preconditions of simulation or resampling (such as changing the distribution assumptions), the sensitivity of the statistic to changes in the assumptions is assessed.

[0119] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for studying the similarities and differences between left and right heart failure in cardiopulmonary exercise testing, characterized in that, include: S1: Selection and grouping of study subjects: Heart failure patients with stable clinical symptoms were selected and enrolled according to New York Heart Classification II to IV; Based on the patients' clinical diagnosis results, the patients were divided into a left heart failure group and a right heart failure group; S2: CPET Operation Standardization: Quality control of equipment, including environmental control and regular equipment calibration; subjects first undergo static lung function tests, followed by testing through the testing device, with respiratory, electrocardiogram, and blood pressure vital signs monitored throughout the process; S3: Data collection and analysis, including: S31: Data acquisition, based on the CPET index, involves continuous dynamic monitoring and detailed recording of blood oxygen saturation, electrocardiogram changes, and gas exchange rate. S32: Statistical Analysis: Data processing was performed using statistical software, and multivariate analysis of variance was used to explore the relationship between multiple response variables and grouping variables, as detailed below: Preparation phase: Record the grouping information for each patient; Data preprocessing: handling missing values ​​and performing variable transformations; Perform MANOVA: In the statistical software, set heart rate and blood pressure as dependent variables and grouping variables as independent variables, and perform multivariate analysis of variance. Results evaluation: Wilks' lambda value, F value, and significance level were analyzed to assess differences between groups and correlations between response variables; S33: Subgroup Analysis: Subgroup analysis was performed based on patients' baseline characteristics to explore the differences in CPET indices among different subgroups and their potential influencing factors. Decision trees were used to identify key distinguishing features within subgroups, as detailed below: Subgroup identification: Patients are subgrouped based on clinical characteristics or CPET results; Data preparation: For each subgroup, prepare the CPET parameters as input for the decision tree or GBM model; Model building: Build a decision tree model or GBM model, and adjust the tree depth and learning rate to optimize performance; Key Feature Identification: Analyze the model results to identify features that have key discriminative power for classifying heart failure within subgroups; S34: Sensitivity Analysis: Monte Carlo simulation was used to evaluate the confidence intervals and sensitivity of the statistical data, as detailed below: Define the analytical objective: Identify the statistical measures of uncertainty that need to be assessed; Monte Carlo simulation: Set a distribution under preset conditions, perform random sampling, and calculate the statistics for each simulation; Bootstrap resampling: It calculates the statistics obtained from each sampling by repeatedly sampling the original dataset. Construct confidence intervals: Based on the simulation or resampling results, calculate the confidence intervals for the statistic, and take the 2.5% and 97.5% quantiles as the 95% confidence intervals; Sensitivity analysis: assessing the sensitivity of a statistic to changes in the assumptions by altering the preconditions of the simulation or resampling. S4: Results Analysis: Comparative analysis of the differences in performance between patients with left heart failure and those with right heart failure in CPET.

2. The method for studying the differences between left and right heart failure in cardiopulmonary exercise testing according to claim 1, characterized in that, In S1, the selection and grouping of research subjects includes: S11: Screening criteria: Inclusion and exclusion criteria are established based on medical history, clinical symptoms and preliminary examination results; S12: Clinical assessment: Clinical assessment of patients who initially meet the screening criteria; S13: Grouping: Patients are divided into corresponding study groups based on clinical assessment results; S14: Baseline Characteristics Recording: Record the patient's baseline characteristics in detail; In S2, the standardization of CPET operation includes: S21: Static pulmonary function test: Subjects undergo static pulmonary function tests in a seated position to measure vital capacity, slow vital capacity, and maximum minute ventilation. S22: Dynamic stress test: Set the parameters of the test device according to the patient's actual condition; S23: Full-process monitoring and recording: Starting from the resting state, continuously monitor and record the patient's heart rate, blood pressure, respiratory rate, and blood oxygen saturation until the end of the recovery period.

3. The method for studying the differences between left and right heart failure in cardiopulmonary exercise testing according to claim 1, characterized in that, In S32, the statistical analysis includes: Multivariate analysis of variance was used to explore the relationship between multiple response variables and grouping variables; Based on the random forest algorithm, feature selection and importance scoring are performed to identify the most important parameters for classifying left and right heart failure in CPET. Regression analysis was used to evaluate the predictive ability of CPET parameters for heart failure type.

4. A method for studying the differences between left and right heart failure in cardiopulmonary exercise testing according to claim 1, characterized in that, In S33, the subgroup analysis includes: Based on factors such as age, gender, and severity of illness, patients were subgrouped according to CPET parameters using a clustering algorithm. Comparative analysis of CPET parameters was performed within each subgroup.

5. A method for studying the differences between left and right heart failure in cardiopulmonary exercise testing according to claim 1, characterized in that, In step S32, regression analysis is used to evaluate the predictive ability of CPET parameters for heart failure type, as follows: Select variables: Select CPET parameters that affect the type of heart failure; Model building: Using heart failure type as the dependent variable and CPET parameter as the independent variable, a logistic regression model is constructed; Model evaluation: The explanatory power and predictive ability of the model are evaluated through goodness-of-fit index, prediction accuracy, and likelihood ratio test. Model optimization: Select the optimal feature subset through stepwise regression or information-based criteria.

6. A method for studying the differences between left and right heart failure in cardiopulmonary exercise testing according to claim 1, characterized in that, In step S32, feature selection and importance scoring are performed based on the random forest algorithm, as detailed below: Data preparation: All parameters of CPET were used as features, and the classification label of heart failure was used as the target variable; Data set partitioning: The data is divided into a training set and a test set; Model training: The model is trained using the random forest algorithm; Feature importance assessment: Ranking the importance of each feature using feature importance scores provided by the algorithm; Feature selection: Features with low importance scores are removed, and features that contribute more to the model's predictions are selected.

Images