CN114649098A - Modeling method of coronary heart disease PCI operation based on SYNTAX-II integral - Google Patents

Abstract

The invention discloses a modeling method of PCI (peripheral component interconnect) surgery for coronary heart disease based on SYNTAX-II integration in the field of model construction, which comprises the steps of acquiring the factors of age, sex, inosine clearance, left ventricular ejection fraction, chronic obstructive pulmonary artery disease and peripheral vascular disease of a patient, endowing the factors with different weights within 4 years after PIC (peripheral artery disease) surgery treatment, then adding all the weights to obtain the death rate after surgery, providing guidance for the implementation of the PIC surgery, carrying out statistical analysis on the obtained data, carrying out correlation analysis and partial correlation analysis on data for judging whether postoperative complications of the patient occur or not by judging whether the postoperative complications of the patient occur by systolic pressure and diastolic pressure, determining that the complications of the patient after surgery and the value of the systolic pressure have a certain positive correlation relationship, and because the systolic pressure has stronger linear influence on the diastolic pressure, a linear equation can be used for really fitting the relationship between the value of the systolic pressure and the postoperative complications generating probability of the patient, and the goodness of fit is higher and higher, so that the modeling accuracy is tested.

Description

The modeling method of coronary heart disease PCI based on SYNTAX-Ⅱ score

technical field

The invention belongs to the field of model construction, in particular to a modeling method for coronary heart disease PCI operation based on SYNTAX-II integral.

Background technique

Coronary heart disease is an important disease threatening human life and health, and its morbidity and mortality are increasing year by year. Its pathological basis is atherosclerosis, and low-density lipoprotein cholesterol plays a central role in AS. Dyslipidemia is an independent risk factor leading to the further deterioration of CHD. Guidelines for the prevention and treatment of dyslipidemia in adults emphasize that reducing the level of LDL-C can significantly reduce the risk of CHD morbidity and mortality. Therefore, LDL-C is recommended as the primary intervention target in the prevention and treatment of CHD. A large number of clinical studies and epidemiological evidence have shown that lipid-lowering therapy is beneficial to cardiovascular disease. Studies have shown that the degree of coronary event reduction is positively correlated with the level of LDL-C reduction.

Due to the differences in the standards of blood lipid control at home and abroad, it is necessary to explore the impact of further reducing the level of LDL-C on the progression of coronary artery disease. SYNTAX and SYNTAXⅡ scores can effectively evaluate the complexity and severity of coronary artery lesions in CHD patients and judge the prognosis, and provide guidance for the implementation of PIC surgery.

SUMMARY OF THE INVENTION

In order to solve the above problems, the purpose of the present invention is to provide guidance for the implementation of PIC based on the modeling method of coronary heart disease PCI based on SYNTAX-II score.

In order to achieve the above object, the technical solution of the present invention is as follows: a modeling method for coronary heart disease PCI based on SYNTAX-II score, including collecting the patient's age, gender, inosine clearance rate, left ventricular ejection fraction, chronic obstructive pulmonary artery Disease and peripheral vascular disease factors gave different weights to the above factors within 4 years after PIC treatment, and then the postoperative mortality rate was calculated according to the addition of all weights, which provided guidance for the implementation of PIC surgery.

Further, the following steps are also included:

S1, screen the patient's condition, classify the patient's disease, the classification is mainly left main disease, and the specific patient types are left main alone, left main combined with single-vessel and multi-vessel disease;

S2, the detailed information of the patient is divided, and the infection index, liver function, renal function and coagulation function are used as postoperative evaluation indicators;

S3: Sampling the patient's cells to analyze the index difference between the second-generation sirolimus drug-eluting stent and paclitaxel drug-eluting stent;

S4, assisting standard calibration, recording the systolic and diastolic blood pressure of the patient to generate a three-dimensional scattergram;

S5, establish a univariate linear regression model, take the midpoint of the systolic blood pressure grouping as the independent variable x, and the prevalence rate as the dependent variable y to obtain the expectation and variance, and compare the expectation and variance to the evaluation indicators to eliminate wrong parameters.

Further, the S2 also includes a data correction step, which includes the exclusion of outliers and excessively incomplete samples: due to possible recording errors in the investigation, for numerical characteristics, it is necessary to base on experience based on the distribution range of numerical values. Identify outliers and leave them empty; in addition, in order to avoid excessively mutilated samples from adversely affecting the results of disease risk assessment, while maximizing data retention.

Further, the correction step also includes missing value filling: when there are missing values in the data set, the risk assessment model cannot be learned correctly, and the missing values in the data set are filled step by step. The filling step is to measure the difference between different eigenvalues. If the k nearest samples of a sample in the feature space belong to a certain category, the sample should also belong to this category.

Further, in order to avoid the deviation caused by using fewer features to predict a large number of features when the missing data is scattered, a step-by-step filling method is used, that is, according to the completeness of the feature column, it is divided into several different levels, located in The first level is 14 features with 100% completeness. The number of samples in each subsequent level can be customized, and the missing values in the nth level feature are used for all the 1st to n-1th levels. feature to learn.

Further, the clinical data of the collected subjects in S1 include gender, age, infection indicators, leukocytes, neutrophils, lymphocytes, monocytes, total protein, albumin, alanine aminotransferase, aspartate aminotransferase, blood urea nitrogen, Serum creatinine, coagulation function, prothrombin time, activated partial thromboplastin time, fibrinogen, thrombin time, D-dimer and glycosylated hemoglobin.

The following beneficial effects were achieved after adopting the above scheme: All subjects in this study SYNTAX and SYNTAX II scores were calculated according to the calculator on the SYNTAX official website (http://www.syntaxscore.com), and the characteristics and baseline data of coronary artery lesions were input by inputting Automatically calculated, the higher the score, the more complex the lesion and the worse the patient's prognosis, especially the higher the revascularization rate.

Statistical analysis was performed on the obtained data, and correlation analysis and partial correlation analysis were performed on the data of systolic blood pressure and diastolic blood pressure to determine whether postoperative complications occurred. , because systolic blood pressure has a strong linear effect on diastolic blood pressure, we can use a linear equation to fit the relationship between the value of systolic blood pressure and the probability of postoperative complications of patients very realistically, and the goodness of fit is getting better and better. The higher it is, to test the accuracy of the modeling.

This study innovatively used the changes of SYNTAX and SYNTAXⅡ scores to quantitatively evaluate the progression of coronary artery disease in CHD patients with different LDL-C levels. The results showed that the SYNTAX and SYNTAXⅡ scores in the low LDL-C level group were lower than those in the non-low LDL-C level group and LDL-C unreached group, suggesting that low LDL-C level can further reduce the risk of coronary artery disease progression in CHD patients. This technical solution clarifies that LDL-C plays a very important role in the progression of coronary atherosclerotic lesions, suggesting that LDL-C levels may be a new potential therapeutic target for delaying or even preventing the progression of coronary artery lesions.

Description of drawings

Fig. 1 is the patient information statistics chart in the present invention;

Fig. 2 is the process steps of numerical correction;

Fig. 3 is the discrete model diagram of diastolic blood pressure in the auxiliary standard proofreading step;

Figure 4 shows the relationship between systolic blood pressure and the probability of postoperative complications in patients.

Detailed ways

The following is further described in detail by specific embodiments:

The embodiment is basically shown in Figure 1: the modeling method of coronary heart disease PCI based on SYNTAX-II score: including collecting the patient's age, gender, inosine clearance, left ventricular ejection fraction, chronic obstructive pulmonary artery disease and Peripheral vascular disease factors give different weights to the above factors within 4 years after PIC surgery, and then the postoperative mortality rate is obtained according to the addition of all weights, which provides guidance for the implementation of PIC surgery.

Also includes the following steps:

S1. Screen the patient's condition, classify the patient's symptoms and collect the clinical data of the research object. The classification is mainly left main disease, and the specific patient types are left main alone, left main combined with single-vessel and multi-vessel disease, and the research objects are collected. The clinical data include gender, age, infection indicators, white blood cells, neutrophils, lymphocytes, monocytes, total protein, albumin, alanine aminotransferase, aspartate aminotransferase, blood urea nitrogen, blood creatinine, coagulation function, thrombin Protime, activated partial thromboplastin time, fibrinogen, thrombin time, D-dimer and glycosylated hemoglobin;

Please refer to Figure 2, S2, to partition the patient's detailed information, and use infection indicators, liver function, renal function and coagulation function as postoperative evaluation indicators, and also include data correction steps, which include the exclusion of abnormal values and excessively incomplete samples: due to Recording errors may occur in the survey. For numerical characteristics, it is necessary to identify outliers based on the distribution range of the numerical values on the basis of experience, and set them as empty; produce adverse effects while maximizing the retention of data information.

The correction step also includes missing value filling: when the data set has missing values, the risk assessment model cannot be learned correctly, and the missing values in the data set are filled step by step. The filling step is to measure the difference between different eigenvalues. If the k nearest samples of a sample in the feature space belong to a certain category, the sample should also belong to this category.

In order to avoid the deviation caused by using fewer features to predict a large number of features when the missing data is scattered, a step-by-step filling method is used, that is, according to the completeness of the feature column, it is divided into several different levels, which are located in the first The level is 14 features with 100% completeness. The number of samples at each level can be customized, and the missing value in the nth level feature is used for all the features from the 1st to the n-1th level. study.

S3: Sampling the patient's cells to analyze the index difference between the second-generation sirolimus drug-eluting stent and paclitaxel drug-eluting stent;

Please refer to Figure 3 and Figure 4, S4, auxiliary standard calibration, record the patient's systolic blood pressure and diastolic blood pressure to generate a three-dimensional scattergram;

S5, establish a univariate linear regression model, take the midpoint of the systolic blood pressure grouping as the independent variable x, and the prevalence rate as the dependent variable y to obtain the expectation and variance, and compare the expectation and variance to the evaluation indicators to eliminate wrong parameters.

The SYNTAX and SYNTAX II scores of all subjects in this study were calculated based on the calculator on the SYNTAX official website (http://www.syntaxscore.com), and were automatically calculated by inputting their coronary artery lesion characteristics and baseline data. The more complex the lesion, the poorer the patient's prognosis, especially the higher the revascularization rate.

Statistical analysis was performed on the obtained data, and correlation analysis and partial correlation analysis were performed on the data of systolic blood pressure and diastolic blood pressure to determine whether postoperative complications occurred. , because systolic blood pressure has a strong linear effect on diastolic blood pressure, we can use a linear equation to fit the relationship between the value of systolic blood pressure and the probability of postoperative complications of patients very realistically, and the goodness of fit is getting better and better. The higher it is, to test the accuracy of the modeling.

This study innovatively used the changes of SYNTAX and SYNTAXⅡ scores to quantitatively evaluate the progression of coronary artery disease in CHD patients with different LDL-C levels. The results showed that the SYNTAX and SYNTAXⅡ scores in the low LDL-C level group were lower than those in the non-low LDL-C level group and LDL-C unreached group, suggesting that low LDL-C level can further reduce the risk of coronary artery disease progression in CHD patients. This technical solution clarifies that LDL-C plays a very important role in the progression of coronary atherosclerosis, suggesting that LDL-C levels may be a new potential therapeutic target for delaying or even preventing the progression of coronary artery disease.

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus.

The above are only the embodiments of the present invention, and the common knowledge such as the well-known specific structures and characteristics in the scheme has not been described too much here. Those of ordinary skill in the art know that the invention belongs to the technical field before the filing date or the priority date. Technical knowledge, can know all the prior art in this field, and have the ability to apply conventional experimental means before the date, those of ordinary skill in the art can improve and implement this scheme in combination with their own ability under the enlightenment given in this application, Some typical well-known structures or well-known methods should not be an obstacle to those skilled in the art from practicing the present application. It should be pointed out that for those skilled in the art, some modifications and improvements can be made without departing from the structure of the present invention. These should also be regarded as the protection scope of the present invention, and these will not affect the implementation of the present invention. Effectiveness and utility of patents. The scope of protection claimed in this application shall be based on the content of the claims, and the descriptions of the specific implementation manners in the description can be used to interpret the content of the claims.

Claims (6)

1. The modeling method of the coronary heart disease PCI operation based on SYNTAX-II integration is characterized in that: the method comprises the steps of collecting factors of age, sex, inosine clearance rate, left ventricular ejection fraction, chronic obstructive pulmonary artery disease and peripheral vascular disease of a patient, endowing the factors with different weights within 4 years after PIC operation treatment, and then adding all the weights to obtain the postoperative fatality rate, thereby providing guidance for the implementation of the PIC operation.

2. The modeling method for coronary heart disease PCI surgery based on SYNTAX-ii integration according to claim 1, wherein: further comprising the steps of:

s1, screening the conditions of patients, and classifying the diseases of the patients, wherein the classified conditions are mainly left main stem lesions, and the specific types of the patients are single left main stem, left main stem combined single branch lesions and multiple branch lesions;

s2, dividing the detailed information of the patient into regions, and taking the infection index, the liver function, the kidney function and the blood coagulation function as postoperative judgment indexes;

s3, sampling cells of the patient, and analyzing the index difference between the second generation sirolimus drug eluting stent and the paclitaxel drug eluting stent;

s4, correcting the auxiliary standard, recording the systolic pressure and the diastolic pressure of the patient to generate a three-dimensional scatter diagram;

s5, establishing a unitary linear regression model, taking the midpoint of the contraction compression grouping as an independent variable x and the disease proportion as a dependent variable y to obtain an expectation and a variance, and comparing the expectation and the variance with a judgment index to eliminate error parameters.

3. The modeling method for coronary heart disease PCI surgery based on SYNTAX-ii integration according to claim 2, wherein: the step of S2 further includes a data correction step, where the data correction step includes outlier and excessive incomplete sample exclusion: because recording errors may occur in the investigation, for the numerical characteristics, abnormal values need to be identified according to the distribution range of the numerical values on the basis of experience support, and the abnormal values are set to be null; in addition, in order to avoid the adverse effect of excessive incomplete samples on the disease onset risk assessment result, the data information is kept to the maximum extent.

4. The modeling method for coronary heart disease PCI surgery based on SYNTAX-ii integration according to claim 3, wherein: the correction step also comprises missing value padding: when the data set has a missing value, the risk assessment model cannot be correctly learned, the missing value of the data set is filled in a stepping mode, the filling step is to classify by measuring the distance between different characteristic values, and if k nearest samples of one sample in the characteristic space belong to a certain class, the sample also belongs to the class.

5. The modeling method for coronary heart disease PCI surgery based on SYNTAX-ii integration according to claim 4, wherein: in order to avoid the deviation caused by predicting a large number of features by using fewer features when missing data is more dispersed, a step-by-step filling method is used, namely the feature columns are divided into a plurality of different grades according to the integrity degree of the feature columns, 14 features with the integrity degree of 100% are positioned in the first grade, the number of samples of each grade can be customized, and the missing values in the features of the nth grade are learned by using all the features of the 1 st to the n-1 st grades.

6. The modeling method for coronary heart disease PCI surgery based on SYNTAX-ii integration according to claim 1, wherein: clinical data from subjects collected in S1 include sex, age, infection index, leukocytes, neutrophils, lymphocytes, monocytes, total protein, albumin, glutamic-pyruvic transaminase, glutamic-oxalacetic transaminase, blood urea nitrogen, blood creatinine, blood coagulation function, prothrombin time, activated partial thromboplastin time, fibrinogen, thrombin time, D-dimer, and glycated hemoglobin.

Images