KR20200021844A - Method for stratified grouping of continuous variables and Method and Apparatus for analyzing correlation using as the same - Google Patents

Abstract

A method for stratified grouping of continuous variables and a method and an apparatus for analyzing correlation using the same are provided. A method for stratified grouping of continuous variables according to an embodiment comprises the steps of: classifying analysis data into a plurality of subgroups having the same combination of disturbance variables; classifying each subgroup into high continuous variables and lower continuous variables based on the representative value of the continuous variable distribution; and generating a high-level group by using the high continuous variables of each subgroup and generating a low-level group by using the continuous variables of each subgroup.

Description

Method for stratified grouping of continuous variables and Method and Apparatus for analyzing correlation using as the same}

The present invention relates to a hierarchical grouping method of a continuous variable, a correlation analysis method using the same, and a device thereof. A method and apparatus for correlation analysis with variables.

Modern medicine frequently uses software that is equipped with various statistical analysis algorithms to find the cause of the disease, the factors related to the occurrence of the disease, and to analyze the effects of newly developed drugs or treatments.

Liver levels, cholesterol, blood pressure, body mass index (BMI), and smoking status are typical clinical and epidemiological variables that can be obtained in hospitals, and are measured, observed, or tested depending on the purpose of treatment or research. Can be extended to more than a dozen cases.

On the other hand, continuous variables (independent variables) may be correlated with certain dependent variables. Continuous variables are numbers that are numerically significant and in which the size itself is meaningful, such as the amount and / or number of drugs administered to a patient. The dependent variable is a variable (eg, the patient's blood pressure) that is assumed to depend on the value of the continuous variable (independent variable). Various statistical algorithms can be used to determine whether there is actually a relationship between these continuous and dependent variables (eg, whether the amount of medicine the patient receives is related to the patient's blood pressure).

When grouping continuous variables to analyze specific dependent variables and correlations, we usually use the "simple grouping" method. As a simple grouping method, as shown in FIG. 1, a method of dividing into two groups based on a specific value (average value or median value) is widely used.

However, in the case of the simple grouping method, the correlation may not be revealed due to the influence of a confounder. Disturbance variables (those that systematically vary with the level of the continuous variable) can affect the dependent variable, identifying which factor (variable) caused any observed change in the dependent variable (s). Makes things impossible. Thus, the presence of disturbing variables makes it impossible to make statistical inferences about the causal relationship between continuous and dependent variables.

Therefore, through the grouping of continuous variables, the development of technology that can analyze the correlation between the continuous variable and the dependent variable even when the continuous variable has a correlation with a specific dependent variable but the correlation variable cannot be revealed due to the disturbing variable There is a need for.

Related prior arts include Korea Patent Publication No. 10-2015-0116121 (name of the invention: continuous dependent variable prediction system and method, air freight freight prediction system and method using the same, published date: October 15, 2015).

The problem to be solved by the present invention is the correlation between the continuous variable and the dependent variable even when the continuous variable has a correlation with a specific dependent variable through grouping of the continuous variable, but the correlation is not revealed due to the disturbing variable. The present invention provides a method of hierarchical grouping of continuous continuous variables, a correlation analysis method using the same, and an apparatus thereof.

The problem to be solved by the present invention is not limited to the above-mentioned task (s), another task (s) not mentioned will be clearly understood by those skilled in the art from the following description.

In the method of hierarchical grouping of continuous variables according to an embodiment of the present invention, in the method of hierarchical grouping of continuous variables, the apparatus may include classifying analysis data into a plurality of subgroups having the same combination of disturbance variables, each subgroup. Dividing the group into upper continuous variable and lower continuous variable based on the representative value of the corresponding continuous variable distribution, generating the upper group by the upper continuous variable of each subgroup, and lowering each subgroup Generating a subgroup by the continuous variables.

Preferably, the representative value is a median value, and the dividing may include: arranging the continuous variable distribution for each subgroup in an ascending order, selecting a median value in the continuous variable distribution for each subgroup, and The method may include dividing the continuous variable above the median into the upper continuous variable and the continuous variable below the median into the lower continuous variable for each subgroup.

In a method for analyzing a correlation between a continuous variable and a specific dependent variable with respect to data to be analyzed by an apparatus according to another embodiment of the present invention, information about the continuous variable, the dependent variable, and the disturbance variable may be obtained from the data. And extracting the hierarchical grouping using the disturbance variable with respect to the extracted continuous variable, and analyzing the correlation between the hierarchical grouped continuous variable and the dependent variable.

Preferably, the hierarchical grouping comprises: classifying the data into a plurality of subgroups having the same combination of disturbance variables, and classifying each subgroup based on a representative value of a corresponding continuous variable distribution. Distinguishing each of the continuous variables may include generating an upper group by upper continuous variables of each subgroup and generating a lower group by lower continuous variables of each subgroup.

Correlation analysis apparatus using the hierarchical grouping of the continuous variable according to another embodiment of the present invention, the hierarchical grouping unit for the hierarchical grouping of the data to be analyzed to have the same disturbance variable for the continuous variable, the layered grouping continuous It includes a correlation analysis unit that analyzes the correlation between type variables and dependent variables.

Preferably, the hierarchical grouping unit classifies the data into a plurality of subgroups having the same combination of disturbance variables, and classifies each subgroup based on a representative value of the distribution of the continuous variable. Each of the subgroups may be classified into upper groups by upper continuous variables of each subgroup, and lower groups may be generated by lower continuous variables of each subgroup.

According to the present invention, by grouping the continuous variable using the disturbance variable, even if the continuous variable has a correlation with the specific dependent variable but the correlation is not revealed due to the disturbing variable, Correlation can be analyzed.

Effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a view for explaining a conventional simple grouping method.
2 is a view for explaining the process of finding the correlation according to an embodiment of the present invention.
3 is a flowchart illustrating a correlation analysis method of continuous variables using hierarchical grouping according to an embodiment of the present invention.
4 is a flowchart illustrating a hierarchical grouping of continuous variables according to an embodiment of the present invention.
5 and 6 are exemplary diagrams for explaining the hierarchical grouping of continuous variables according to an embodiment of the present invention.
7 is a diagram illustrating an apparatus for correlation analysis of continuous variables using hierarchical grouping according to an embodiment of the present invention.
FIG. 8 is an exemplary diagram for explaining simple grouping and stratified grouping in which patients are classified into two groups according to lung dose to investigate a correlation between lung dose and survival rate.
FIG. 9 is a diagram for explaining overall survival rate, survival without local recurrence, and survival without distant metastasis between high and low lung dose groups by simple grouping and stratified grouping.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. In describing the drawings, similar reference numerals are used for similar elements.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term and / or includes a combination of a plurality of related items or any item of a plurality of related items.

When a component is said to be "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that another component may be present in the middle. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and shall not be construed in ideal or excessively formal meanings unless expressly defined in this application. Do not.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

2 is a view for explaining the process of finding the correlation according to an embodiment of the present invention.

Referring to FIG. 2, stratified grouping using a disturbance variable is performed on a continuous variable, and the correlation between the continuous variable and the dependent variable is analyzed using the result. Then, even if there is a correlation with a specific dependent variable of the continuous variable through grouping of the continuous variable, even if the correlation cannot be revealed due to the disturbing variable, the correlation between the continuous variable and the dependent variable can be analyzed. .

3 is a flowchart illustrating a correlation analysis method of continuous variables using hierarchical grouping according to an embodiment of the present invention.

Referring to FIG. 3, when data to be analyzed is input (S310), the device hierarchically groups the data using a disturbance variable for a continuous variable (S320). In this case, the data may include continuous variables (independent variables), dependent variables, disturbing variables, and the like. Therefore, when data is input, the device analyzes the data to obtain information about continuous variables, disturbing variables, dependent variables, and the like. In this case, the continuous variable, the dependent variable, and the disturbance variable may be set in advance. The device then groups the hierarchies to have the same disturbances for the continuous variables. Detailed description of the hierarchical grouping of continuous variables will be described with reference to FIG. 4.

When the step S320 is performed, the device analyzes the correlation between the hierarchical grouped continuous variable and the dependent variable (S330). In this case, the device uses various statistical algorithms such as Student's T test, Welch's T test, Kruskal-Wallis test, etc. Correlation can be analyzed.

4 is a flowchart illustrating a hierarchical grouping of continuous variables according to an embodiment of the present invention, and FIGS. 5 and 6 are exemplary diagrams for describing the hierarchical grouping of continuous variables according to an embodiment of the present invention. .

Referring to FIG. 4, the device classifies data to be analyzed into a plurality of subgroups having the same combination of disturbance variables (S410).

For example, a case in which the data to be analyzed is (a) of FIG. 5 and the continuous variable (exposure) is grouped using the first disturbance variable var_01 and the second disturbance variable var_02 will be described. Shall be. In this case, it is assumed that the value of the first disturbance variable is composed of 1 and 0, and the value of the second disturbance variable is composed of 1 and 0. In this case, since the combination of the first disturbance variable and the second disturbance variable is (0,0), (0,1), (1,0), (1,1), the apparatus is as shown in FIG. The combination of the first disturbance variable and the second disturbance variable creates four subgroups. That is, the apparatus collects data of the combination (0,0) in which the value of the first disturbance variable and the value of the second disturbance variable are both '0' in the first subgroup A and the value of the first disturbance variable is '0'. , Data of the combination (0, 1) having the value of the second disturbance variable is' 1 'in the second subgroup (B), the value of the first disturbance variable is' 1', and the value of the second disturbance variable is' The data of the combination (1, 0) of 0 'is combined to the third subgroup (C), the data of the combination (1, 1) of which the value of the first disturbance variable and the value of the second disturbance variable are both' 4 '. Create as a subgroup (D).

When the step S410 is performed, the apparatus checks the distribution of the corresponding continuous variable of each subgroup (S420), and based on the representative value of the corresponding continuous variable distribution for each subgroup, the apparatus determines the upper continuous variable and the lower continuous variable. (S430). Here, the representative value may include a median value, an average value, and the like, but the following description will be limited to the median value. In this case, the device sorts the continuous variable distribution by each subgroup in ascending order, selects the median value in the continuous variable distribution by each subgroup, and selects the continuous variable above the median value by each subgroup as the upper continuous variable and the median value. Separate less than continuous variables into lower continuous variables.

When step S430 is performed, the device generates an upper group by upper continuous variables of each subgroup, and generates a lower group by lower continuous variables of each subgroup (S440).

For example, referring to FIG. 5B, since the median value of the first subgroup is identification number 20, continuous variables corresponding to 1 and 18 correspond to lower continuous variables (light purple) and 21 and 23. FIG. Continuous variables can be classified into upper continuous variables (beige). Since the center of the second subgroup is between 9 and 10, continuous variables corresponding to 3, 4, 5, 6, 7, 8, and 9 are lower continuous variables, 10, 12, 13, 15, 16, 17 For example, continuous variables corresponding to 25 may be divided into upper continuous variables. Since the center of the third subgroup is between 26 and 27, continuous variables corresponding to 14, 19, 22, 24, and 26 are lower continuous variables, and 27, 28, 29, and 30 are upper 12, 13, 15, Continuous variables corresponding to 16, 17, and 25 may be divided into upper continuous variables. Continuous variables corresponding to 2 of the fourth subgroup may be divided into lower continuous variables, and continuous variables corresponding to 11 may be divided into upper continuous variables.

Then, the device divides the continuous variable into a high exposure group and a low exposure group as shown in (c). In other words, continuous variables smaller than the median in the first subgroup are subgroups, continuous variables larger than the median are upper groups, continuous variables smaller than the median in the second subgroup are subgroups, and continuous variables larger than the median in the second subgroup. In the upper group, the third subgroup, the continuous variable smaller than the median is the lower group, the continuous variable larger than the median is the upper group, and in the fourth subgroup, the continuous variable smaller than the median is the subgroup, and the continuous variable larger than the median is Can be a parent group.

Here, the case where two cross variables are illustrated as an example, but the number of cross variables used for the hierarchical grouping of continuous variables is not limited.

For example, as shown in FIG. 6, 10 disturbance variables, SEX, COPD, CCI, SURG, pT, pN, RM, CTx, PET, and RT Dose, are hierarchically grouped for continuous variable MLD to group Low Lung Dose Group and High Lung. You can create a Dose Group.

Specifically, referring to FIG. 6 (a), 33 subgroups separated by red borders are generated by a combination of 10 disturbance variables. Each of the 33 subgroups is divided into upper continuous variables (beige) and lower continuous variables (mauve) by the median. Low Lung Dose Group is created by the low continuous variable of each subgroup, and High Lung Dose Group is created by the high continuous variable of each subgroup.

A low Lung Dose Group and a High Lung Dose Group generated by the hierarchical grouping may be represented as a continuous variable Mean Lung Dose (MLD) as shown in FIG. 6B.

7 is a diagram illustrating an apparatus for correlation analysis of continuous variables using hierarchical grouping according to an embodiment of the present invention.

Referring to FIG. 7, an apparatus for correlation analysis of continuous variables using hierarchical grouping according to an embodiment of the present invention may include a storage unit 710, an output unit 720, a controller 730, and a layered grouping unit. 740, a correlation analyzer 750.

The storage unit 710 is configured to store data related to the operation of the apparatus 700 for correlation analysis of continuous variables using hierarchical grouping. The storage unit 710 may use a known storage medium. For example, the storage unit 710 may use any one or more of known storage media, such as a ROM, a PROM, an EPROM, an EEPROM, and a RAM.

In particular, the storage unit 710 may store a program or an application for hierarchical grouping and correlation analysis of continuous variables. In addition, the storage unit 710 may store various related algorithms (or equations) for hierarchical grouping and correlation analysis of continuous variables. In this case, the controller 730 may call the storage unit 710 to obtain a necessary algorithm.

The output unit 720 is a component for displaying various information related to the operation of the apparatus 700 for correlation analysis of continuous variables using hierarchical grouping. In particular, the output unit 720 may display various information such as a group generated by the layered grouping in the layered grouping unit 740 and a correlation between the continuous variable and the dependent variable analyzed in the correlation analyzer 750. have. The output unit 720 may be implemented through various display devices including LCDs, LEDs, and the like.

The controller 730 is configured to control operations of various components of the apparatus 700 for correlation analysis of continuous variables using hierarchical grouping. The controller 730 may include at least one computing device. It may be a general purpose central processing unit (CPU), programmable device elements (CPLD, FPGA) implemented for a particular purpose, an application specific semiconductor computing unit (ASIC), or a microcontroller chip.

The hierarchical grouping unit 740 hierarchically groups the data to be analyzed to have the same disturbance variable with respect to the continuous variable. That is, the hierarchical grouping unit 740 classifies the data to be analyzed into a plurality of subgroups having the same combination of disturbance variables and checks the distribution of the corresponding continuous variable of each subgroup. Then, the hierarchical grouping unit 740 divides the upper continuous variable and the lower continuous variable based on the representative value of the corresponding continuous variable distribution for each subgroup, and ranks the upper continuous variable by the upper continuous variable of each subgroup. Create a group and create a subgroup by subcontinuous variables in each subgroup.

The correlation analyzer 750 analyzes the correlation between the hierarchical grouped continuous variable and the dependent variable.

According to an embodiment of the present invention, the layered grouping unit 740 and the correlation analyzer 750 may be program modules communicating with an external terminal device or an external server. Such program modules may be included in device 700 as an operating system, an application module, and other program modules, and may be physically stored on a variety of known storage devices. In addition, these program modules may be stored in a remote storage device capable of communicating with the device 700. Such program modules include, but are not limited to, routines, subroutines, programs, objects, components, data structures, etc. that perform particular operations described above or execute particular abstract data types in accordance with the present invention.

Meanwhile, the device 700 may be a communication terminal such as a computer, a notebook, a netbook, a PDA, or a smart device such as a smart phone, a smart note, a tablet PC, a smart TV, or the like. In addition, the apparatus 700 may be implemented as a single computing device or in the form of a collection device in which two or more computing devices are connected to each other. For example, the device 700 may be implemented as a single server or in a form in which two or more servers are connected.

Hereinafter, the lung dose and the risk of death of post-operative radiation therapy for non-small cell lung cancer will be described with reference to the present invention.

For this purpose, 178 patients with non-small cell lung cancer who received postoperative radiotherapy were analyzed. Mean lung dose was calculated from dose-volume data and classified into high and low dose groups using simple grouping and stratified grouping. In other words, the simple grouping method using the median lung dose of all patients and the stratified grouping method using the median lung dose of each subgroup having the same confounders. They were divided into high lung dose group and low lung dose group. We compared the clinical variables and survival rates between the high-dose group and the low-dose group.

Specifically, staging evaluations were routinely performed by bronchoscopy, computed tomography (CT), brain CT or magnetic resonance imaging, bone scan, or positron emission tomography (PET) -CT in the chest and upper abdomen. All patients who underwent mediastinal lymph node dissection underwent lobectomy or pneumonectomy. Postoperative radiotherapy was performed in patients with pathologically benign (mainly N2), benign or near resection. Radiation therapy started within 4-5 weeks after surgery, and three-dimensional conformal radiotherapy was performed using a mega-voltage photon beam. Initial clinical target volumes included bronchial stumps, mediastinal nodal stations, and then draining nodal stations. The boost clinical target volume included bronchial stumps and only mediastinal lymph node positive stations. The planning target volume was set to extend in all directions by a margin of 1-1.5 cm from the clinical target volume. According to the risk of recurrence, the boost planning target volume was investigated up to 50.4-60 Gy and the initial planning target volume was examined for a total of 44-45 Gy. In the case of positive resection margin, the risk area increased to a dose of 66-70 Gy. Adjuvant chemotherapy was performed with 4-6 cycles of platinum-based regimens. In the first year after completion of the adjuvant therapy, follow-up was followed every three months, followed up every six months for the next two years, and yearly examinations. During the follow-up, a brief chest radiograph or chest CT scan was performed, and a PET-CT scan was done at the doctor's discretion. The age-adjusted Charlson comorbidity score was adopted for the assessment of underlying comorbidities, and was determined using the ICD-10 diagnostic code previously identified in the inpatient and outpatient records from the first visit of each patient to the date of radical surgery. It was. To calculate the dose of lung tissue, Varian Eclipse External Beam Planning System version 7.1 (Varian Medical System, Palo, Alto, CA) was used, and a pencil-beam algorithm with heterogeneous correction was applied. Mean lung dose for each patient was obtained from dose-volume data. At this time, the entire remaining lung was considered as a single organ.

In addition, to investigate the association between lung dose and survival rate, patients were divided into two groups according to lung dose. Simple grouping method using median lung dose of all patients as shown in FIG. 8 (a) to classify into high and low dose group, same disturbance as in FIG. 8 (b) A stratified grouping method using the median lung dose of each subgroup of patients with confounders was used. At this time, the disturbance variables were male (male vs female), chronic obstructive pulmonary disorder (yes vs no), complication index (less than 3 vs 3), type of operation (lobectomy vs pneumonectomy) pneumonectomy), pathologic T-stage (112 vs 3-4), pathological N-stage (0 vs 1 vs 2), surgical margin (+ vs-), adjuvant chemotherapy Use (yes vs no), use of PET-CT staging (yes vs no), and radiation therapy doses (each individual dose level). For comparison between high and low dose groups, Fisher's exact test or chi-squared test was used for categorical variables and Student t-test or Kruskal-Wallis rank sum test was used for continuous variables. Also, locoregional recurrence-free survival, no distant metastasis Clinical variables that influenced survival and overall survival were analyzed: survival time was defined by the interval between the date of surgical treatment and the last follow-up visit, death or relapse date. Was calculated using the Kaplan-Meier method, and Cox proportional hazards regression models were implemented for single-variable and multivariate analysis, and variables with P values less than 0.10 were included in the multivariate analysis. Interactions between potential disturbances and lung doses were tested by adding an interaction term, and double-sided P values below 0.05 were considered statistically significant. All statistical analyzes were performed using R software.

The analysis results using the method described above are as follows.

Of the 178 patients, 145 (81.5%) were male and ages ranged from 30-78 years (median, 62 years). The type of operation was lobectomy / bilobectomy in 143 patients (80.3%) and pneumonectomy in 35 patients (19.7%). Squamous cell carcinoma was observed in 79 patients (44.4%) and adenocarcinoma was found in 80 patients (44.9%). Fourteen patients (7.9%) had a positive microscopic resection margin. The median dose of radiation therapy was 54 Gy (range, 45.0-70.0 Gy) and the mean lung dose was 0.0 to 25.3 Gy (median, 12.5). Total radiotherapy time ranged from 32 to 90 days (median, 39). 25 patients (14.0%) received steroids for symptomatic radiation pneumonitis, were treated with conservative therapy, and mild to moderate in 36 patients (20.2%). Esophagitis developed. Adjuvant chemotherapy with cisplatin was performed in 60 patients (33.7%) and the frequency of chemotherapy was 3 to 6 (median, 4).

Patient characteristics between high and low lung dose groups according to simple grouping and stratified grouping are shown in Table 1 below.

TABLE 1

In Table 1, standard deviation (SD) is the standard deviation, COPD (chronic obstructive pulmonary disorder) is chronic obstructive pulmonary disease, ECOG PS (Eastern Cooperative Oncology Group performance status), positron emission tomography-computed tomography (PET-CT), and FEV1. (forced expiratory volume in 1 s), postoperative chemotherapy (POCT) means postoperative chemotherapy, and postoperative radiation therapy (PORT) means postoperative radiation therapy.

Referring to Table 1, in the simple grouping, the high lung dose group had a high complication index, and there were much more patients who had lobectomy than the low lung dose group. The incidence of radiation pneumonitis and esophagitis was higher in the high-dose group than in the low-dose group (pneumonitis 20.9% vs. 6.9%, p = 0.014; esophagitis 28.6% vs. 11.5%, p = 0.008).

There was also no significant difference between the two groups except lung dose in stratified grouping (mean, 15.3 Gy vs. 11.5 Gy, p <0.001). The incidence of radiation pneumonia and esophagitis was not significantly different between the two stratified groups.

The median follow-up period ranged from 3 months to 123 months (median 30) in an average of 178 patients with a 5 year survival rate of 43.7%. Five-year locoregional recurrence-free survival and distant metastasis-free survival were 53.4% and 42.5%, respectively.

FIG. 9 is a diagram for explaining overall survival rate, survival without local recurrence, and survival without distant metastasis between high and low lung dose groups by simple grouping and stratified grouping. Referring to (a) of FIG. 9, in the simple grouping, there is no significant difference in the overall survival rate, the survival rate without local recurrence, and the survival without distant metastasis between the high lung dose group and the low lung dose group. Referring to FIG. 9 (b), in the grouping stratification, the overall survival rate of the low lung dose group is higher than the overall survival rate of the high lung dose group (5 year survival rate: 60.1% vs. 35.3%, p = 0.039), and there is no local recurrence. There is no significant difference in survival without distant metastasis.

Single and multivariate analyzes of clinical factors affecting survival of patients grouped by simple grouping are shown in Table 2 below.

TABLE 2

In Table 2, OS (overall survival) is overall survival, LRRFS (locoregional recurrence-free survival) is local recurrence-free survival, DMFS (distant metastasis-free survival) is distant metastasis-free survival, hazard ratio (HR) is risk ratio, CI (confidence interval) is the confidence interval, COPD (chronic obstructive pulmonary disorder) is chronic obstructive pulmonary disease, Eastern Cooperative Oncology Group performance status (ECOG PS), positron emission tomography-computed tomography (PET-CT), and force expiratory volume in 1s), postoperative chemotherapy (POCT), and postoperative radiotherapy (PORT).

Referring to Table 2, Pneumonectomy is a negative determinant of overall survival (risk ratio, HR = 1.98, p = 0.006) and survival without distant metastasis (HR = 1.71, p = 0.040). Positive surgical margin can also be seen as a poor prognostic factor for survival without local recurrence and survival without distant metastasis. In addition, it can be seen that the use of PET-CT staging (no vs. yes) is an important factor (HR = 0.38, p = 0.036) in overall survival rate.

The results of univariate and multivariate analysis among patients in stratified grouping are shown in Table 3 below.

TABLE 3

Referring to Table 3, it can be seen that lung dose is an important prognostic factor (HR = 2.08, p = 0.019) for overall survival.

In addition, the effects of other dose-volume parameters (V-dose: statistics from V50 to V50 at 5 Gy intervals) on survival were analyzed using simple grouping and stratified grouping. The results are shown in Table 4.

TABLE 4

Referring to Table 4, it can be seen that all V-dose parameters are not significant predictors that affect survival in both the simple grouping method and the stratified grouping method.

On the other hand, the present invention assumes that lung dose levels are associated with the risk of death in patients treated with postoperative radiation therapy for non-small cell lung cancer, and the initial analysis using simple grouping did not demonstrate this association, but stratified grouping After implementation, there was a significant relationship between lung dose and mortality risk. Among all patients, simple grouping to classify patients into groups according to specific exposure values was adopted as a general method of investigating the effects of exposure dose in the regression setting as shown in FIG. Although simple grouping is valid in some cases, it may not work when the risk group is severely disturbed by key variables affecting the outcome.

Simple grouping created a biased distribution of the variables between the high and low dose groups, as shown in Table 1, and this imbalance appeared to be at risk of death at the expense of the effect of lung dose. In particular, the rate of pneumonectomy with poor prognostic factors (HR = 1.98, p = 0.006) was significantly higher in the low-dose group than in the high-dose group (36.8% vs. 3.3%, p <0.001). As a result, the survival rate between the two groups by simple grouping was almost the same in the survival rate without local recurrence, survival without distant metastasis, and overall survival as shown in FIG. 9 (a). With this insignificant result of simple grouping, it is not meaningful to adjust the disturbing variables using multivariate analysis.

Since the simple grouping method does not show a correlation between lung exposure dose and survival rate, the risk group is classified using the stratified grouping method to control disturbance variables. In the stratified grouping, patients were divided into subgroups having the same disturbance parameters as shown in FIG. 8 (b), and then divided into high lung dose groups and low lung dose groups. In Table 1, the stratification went well and the disturbance variables were balanced between the two risk groups, including the proportion of the types of surgery.

Although lung dose values overlap between groups, significant lung dose differences of simple grouping were maintained even after stratified grouping. The survival rate of the low lung dose group was significantly higher than that of the high lung dose group as shown in FIG. 9 (b), and the lung dose was a significant prognostic factor affecting the overall survival rate in the multivariate analysis. . However, there was no difference in survival without local recurrence or survival without distant metastasis.

The pattern of survival outcomes in stratified groupings may confirm that higher lung doses are associated with an increased risk of potential death in patients with non-small cell lung cancer undergoing postoperative radiation therapy. Existing studies compared patients who did not receive radiotherapy with those who did not, and reported that harm was rather harmful if the radiation therapy itself was low. However, these studies have been limited in assessing the potential harm, or the risk of death, among patients receiving radiation. However, in the present invention, the risk of death from lung exposure among patients receiving radiation therapy was analyzed. The risk of radiation-induced potential death may differ between radiation exposures to normal tissue, even among patients receiving radiation at the same prescribed dose. Accordingly, the present invention was intended to demonstrate the risk of potential death according to lung dose of patients undergoing postoperative radiation therapy. However, it was difficult to assess this association because lung dose was closely related to the prescribed dose. The higher the dose prescribed, the more likely the tumor is to be controlled, offsetting the risk of potential death from radiation therapy. Pulmonary dose may also be influenced by target volume, which depends primarily on the pathological N-stage, and may depend on the extent of surgery that affects the volume of the lung contained in the radiation fields. Therefore, stratified grouping was applied to control various potential disturbance variables that influence the risk and benefit of radiation therapy according to the lung dose. As a result, significant differences in the risk of death from lung dose were found. These results show that stratified groupings can be an effective way of adjusting disturbance variables and are useful for exploring the effects of exposure in future studies.

Cardiopulmonary-associated deaths have been proposed as the main cause of radiation therapy-induced potential deaths, and the results of the present invention support that lung dose is primarily associated with an increased risk of potential deaths due to increased lung deaths. . Postoperative pulmonary function has been reported as a significant prognostic factor in patients treated with postoperative radiation therapy. Since a decrease in lung volume, such as radiation-induced lung fibrosis, is associated with lung dose, additional loss of lung function can be assumed to be proportional to lung dose. In stratified groupings, the difference in lung dose between the high and low dose groups was relatively small (high vs. low dose group, average 15.3 Gy vs. 11.5 Gy). Small differences in lung dose were too small to cause a large difference in the incidence of radiation pneumonitis, but had a significant impact on the risk of death. These results remind us that relatively small differences in radiation exposure to the lungs may affect the risk of death. In this context, lung dose should be kept as low as possible to reduce the risk of death in patients treated with postoperative radiation therapy. In addition, individual patients will have different dose limits that are determined by their health status, including lung function. In addition, it should be recognized that chest radiotherapy increases the risk of death as well as radiation pneumonia. Further research is needed to determine individual risks of death according to pulmonary exposure dose levels to provide tailored radiotherapy to balance the risk of death and potential benefits.

Thus, as shown in Table 4, the effect of the other volume-based parameters in addition to the average lung dose was analyzed. Unlike the average lung dose, other dose-volume parameters did not significantly affect the risk of death after stratified grouping. This suggests that mean lung dose is the most useful dose-volume parameter indicative of lung dose level. In conclusion, lung dose is associated with the risk of death in patients with non-small cell lung cancer with the same confounders, suggesting a link between lung dose and an increased risk of potential death, which is more likely to assess mortality risk according to lung dose. The research will help to implement more accurate and individualized radiation therapy and to balance the risks and benefits of radiation therapy.

On the other hand, embodiments of the present invention may be implemented in the form of program instructions that can be executed by various computer means may be recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, and the like, alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Examples of program instructions such as magneto-optical, ROM, RAM, flash memory, etc. may be executed by a computer using an interpreter as well as machine code such as produced by a compiler. Contains high-level language codes. The hardware device described above may be configured to operate as one or more software modules to perform the operations of one embodiment of the present invention, and vice versa.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the appended claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

700: device
710: storage unit
720: output unit
730 control unit
740: layered grouping unit
750: correlation analysis unit

Claims (6)

In the method in which the device hierarchically grouping continuous variables,
Classifying the analysis data into a plurality of subgroups having the same combination of disturbance variables;
Dividing each subgroup into an upper continuous variable and a lower continuous variable based on a representative value of a corresponding continuous variable distribution; And
Generating an upper group by upper continuous variables of each subgroup, and generating a lower group by lower continuous variables of each subgroup
Hierarchical grouping method of a continuous variable comprising a. The method of claim 1,
The step of separating,
Sorting the continuous variable distribution in ascending order for each subgroup;
Selecting a representative value from the continuous variable distribution for each subgroup; And
And classifying the continuous variable above the median into the upper continuous variable and the continuous variable below the median into the lower continuous variable for each subgroup. In a method for analyzing the correlation between a continuous variable and a specific dependent variable for the data to be analyzed by the device,
Extracting information on the continuous variable, the dependent variable, and the disturbance variable from the data;
Hierarchically grouping the extracted continuous variable using a disturbance variable; And
Analyzing a correlation between the hierarchical grouped continuous variable and the dependent variable
Correlation analysis method using the hierarchical grouping of continuous variables comprising a. The method of claim 3,
The layering grouping step,
Classifying the data into a plurality of subgroups having the same combination of disturbance variables;
Dividing each subgroup into an upper continuous variable and a lower continuous variable based on a representative value of a corresponding continuous variable distribution; And
Generating an upper group by the upper continuous variables of each subgroup, and generating a lower group by the lower continuous variables of each subgroup, using the hierarchical grouping of the continuous variables. Relationship Analysis Method. A hierarchical grouping unit which hierarchically groups the data to be analyzed to have the same disturbance variable with respect to the continuous variable; And
Correlation analysis unit for analyzing the correlation between the hierarchical grouped continuous variable and the dependent variable
Correlation analysis apparatus using the hierarchical grouping of continuous variables comprising a. The method of claim 5,
The layered grouping unit,
The data is classified into a plurality of subgroups having the same combination of disturbance variables, and each subgroup is divided into an upper continuous variable and a lower continuous variable based on a representative value of the distribution of the continuous variable. An apparatus for correlation analysis using hierarchical grouping of continuous variables, characterized in that an upper group is generated by upper continuous variables and a lower group is generated by lower continuous variables of each subgroup.

Images