Skin repair efficacy evaluation method based on multiple linear regression model

Abstract

The invention discloses a skin repair efficacy evaluation method based on a multiple linear regression model. The method comprises the following steps: randomly grouping subjects into a control group and an experimental group; the method comprises the following steps: acquiring initial facial skin data and age data, and constructing facial skin data of a specified area position of a subject collected in a subject database at time t after a repair product is used, the facial skin data comprising epidermis data and skin color data; building a facial skin repair database of the subject, building a multi-element skin evaluation model based on the subject database and the facial skin repair database, and reading a skin repair scheme of the subject; and analyzing the skin repair scheme of the subject, notifying the subject to return visit for detection, and calculating through a multi-element skin evaluation model to obtain a repair capability evaluation index of a repair product. According to the evaluation method, the factors influencing the repairing ability of the cosmetics can be found out through the model, the relation between the repairing ability of the cosmetics and the skin can be researched, and meanwhile, the appropriate cosmetics with the repairing ability are recommended to the subject according to the influence factors.

Description

Technical Field

[0001] The present invention relates to the field of skin repair evaluation, and particularly to a method for evaluating skin repair efficacy based on a multiple linear regression model. Background Art

[0002] With the improvement of people's living standards, cosmetics have become a necessity in people's daily lives. Cosmetics can not only beautify the skin but also play a role in protecting the skin. However, the chemical components in cosmetics may cause harm to the skin, leading to problems such as skin sensitivity, redness, and itching. Therefore, the repair ability of cosmetics has become an important research direction.

[0003] Scientists will select experimental subjects to try and evaluate the repair ability of cosmetics. Through the feedback of the experimental subjects' experiences, it can help scientists understand the physiological and biochemical reactions during skin damage and repair. However, due to individual differences among subjects, simply relying on a general repair ability may not be suitable for that customer. Therefore, there is an urgent need in the market for an evaluation method that can recommend suitable repair ability cosmetics for customers. Summary of the Invention

[0004] In view of the above problems, the present invention aims to provide a method for evaluating skin repair efficacy based on a multiple linear regression model.

[0005] To achieve the technical purpose, the solution of the present invention is: A method for evaluating skin repair efficacy based on a multiple linear regression model, the method comprising:

[0006] Randomly dividing the subjects into a control group and an experimental group; obtaining initial facial skin data and age data to construct a subject database;

[0007] Collecting facial skin data at the specified area position of the subject at time t after using the repair product, where the facial skin data includes epidermal data and skin color data, and forming a facial skin repair database of the subject;

[0008] Constructing a multiple skin evaluation model based on the subject database and the facial skin repair database, and reading the skin repair plan of the subject;

[0009] Analyzing the skin repair plan of the subject, and notifying the subject to return for re - testing at a specified time. The repair ability evaluation index of the repair product is calculated through the multiple skin evaluation model.

[0010] Preferably, the epidermal data is specifically the transepidermal water loss rate. The transepidermal water loss rate of the subject is measured using a transepidermal water loss rate tester, and the transepidermal water loss rate at the specified area position of the control group is denoted as A 0, the trans-epidermal water loss rate at the designated area of the experimental group is denoted as A 0.1 .

[0011] Preferably, a skin erythema tester is used to conduct a skin erythema test on the subjects. The skin erythema at the designated area of the control group is denoted as B 0 , and the skin erythema at the designated area of the experimental group is denoted as B 0.1 ;

[0012] The subjects use it according to the skin repair plan. At the t1 follow-up visit, the trans-epidermal water loss rate at the designated area of the control group is denoted as A 1 , and the trans-epidermal water loss rate at the designated area of the experimental group is denoted as A 1.1 ; the skin erythema at the designated area of the control group is denoted as B 1 , and the skin erythema at the designated area of the experimental group is denoted as B 1.1 ;

[0013] At the t2 follow-up visit, the trans-epidermal water loss rate at the designated area of the control group is denoted as A 2 , and the trans-epidermal water loss rate at the designated area of the experimental group is denoted as A 2.1 ; the skin erythema at the designated area of the control group is denoted as B 2 , and the skin erythema at the designated area of the experimental group is denoted as B 2.1 .

[0014] Preferably, a trans-epidermal water loss rate function Na and a skin erythema function Nb are constructed respectively; the specific functions are as follows: Na = β 0 + β 1 F + β 2 A 0 + ε;

[0015] Nb = γ 0 + γ 1 F + γ 2 B 0 + μ;

[0016] where F is the age of the subject; β 0 , β 1 , β 2 , γ 0 , γ 1 , γ 2 are regression coefficients, and ε and μ are model random error terms.

[0017] Preferably, the regression coefficients β 0 , β 1 , β 2 , γ 0 , γ 1 , γ 2The values of, the random error terms ε and μ of the model;

[0018] Calculate the trans-epidermal water loss change rate ΔNa and the skin erythema change rate ΔNb through the following formula, which are used as evaluation indicators for the repair ability of repair products:

[0019] After the subject uses the product at t1, the evaluation indicators for the repair ability of the product:

[0020] ΔNa 1 = 100% × [(A 1.1 - A 0.1 ) / A 0.1 - (A 1 - A 0 ) / A 0 ;

[0021] ΔNb 1 = 100% × [(B 1.1 - B 0.1 ) / B 0.1 - (B 1 - B 0 ) / B 0 ;

[0022] After the subject uses the product at t2, the evaluation indicators for the repair ability of the product:

[0023] ΔNa 2 = 100% × [(A 2.1 - A 0.1 ) / A 0.1 - (A 2 - A 0 ) / A 0 ;

[0024] ΔNb 2 = 100% × [(B 2.1 - B 0.1 ) / B 0.1 - (B 2 - B 0 ) / B 0 .

[0025] Preferably, in the trans-epidermal water loss rate function Na and the skin erythema function Nb, the variables F, A 0 are deterministic variables and are not correlated with each other; the variables F, B 0 are deterministic variables and are not correlated with each other; the random error terms ε, μ have a mean of 0, the same variance, and are independent of each other at different sample points, and there is no serial correlation; and F, A 0 , B 0 are not correlated with ε, μ; the random error terms ε, μ follow a normal distribution;

[0026] According to the preconditions of the above multiple linear regression model, calculate the multiple linear regression model for the trans-epidermal water loss rate of the skin.

[0027] Preferably, when the subject needs to return to the laboratory for a follow-up visit at time t1 or t2, wait for 25 - 30 minutes in the test environment to balance and stay relaxed.

[0028] Preferably, the time t1 of the subject's first follow-up visit is greater than or equal to four days, and the time interval between the subject's second follow-up visit t2 and t1 is greater than or equal to seven days.

[0029] Preferably, before testing with the trans-epidermal water loss rate tester and the skin erythema tester, the subject waits for 25 - 30 minutes in a room at 21 ± 1°C and a humidity of 50 ± 10% to balance and stay relaxed.

[0030] The beneficial effects of the present invention are that the evaluation method of this application can find out the factors affecting the repair ability of cosmetics through the model, can study the relationship between the repair ability of cosmetics and the skin, and at the same time recommend suitable cosmetics with repair ability to the subject according to the influencing factors;

[0031] Since the repair ability of cosmetics varies greatly among different subjects; this application can correlate the repair ability of cosmetics with factors such as the age and skin type of the subject to construct a regression model, and through the regression model, the relationship between the age and skin type of the subject and the repair ability of cosmetics can be found. The subject only needs to return for a few follow-up visits to analyze the repair ability for the subject. This method can more specifically recommend suitable cosmetics for different subjects. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flowchart of the present invention;

[0033] Figure 2 is a specific data collection flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0034] The following further describes the present invention in detail in conjunction with the attached Figure 1-2 drawings and specific embodiments.

[0035] The specific embodiment of the present invention is a skin repair efficacy evaluation method based on a multiple linear regression model, and the steps are as follows:

[0036] S101. Randomly divide the subjects into a control group and an experimental group; obtain the initial facial skin data and age data, and construct a subject database;

[0037] S102. Collect facial skin data of the subject at the specified area at time t after using the repair product. The facial skin data includes epidermal data and skin color data, and establish a facial skin repair database for the subject.

[0038] S103. Build a multivariate skin assessment model based on the subject database and the facial skin repair database, and read the skin repair plan of the subject.

[0039] S104. Analyze the skin repair plan of the subject, and notify the subject to return for a re-examination at the specified time. Calculate the repair ability evaluation index of the repair product through the multivariate skin assessment model.

[0040] The specific data collection process includes the following steps: ① Select subjects with a relatively high transepidermal water loss rate of the skin (such as using a Tewameter to test the face), wait for 30 minutes in the test environment to balance, and keep relaxed; the room temperature is 21±1°C and the humidity is 50±10%.

[0041] ② Randomly divide the subject's face into a control group (left face / right face) and an experimental group (right face / left face), and use a transepidermal water loss rate tester (Tewameter, or equivalent) to measure the transepidermal water loss rate. The transepidermal water loss rate of the skin area in the control group is recorded as A 0 and the transepidermal water loss rate of the skin area in the experimental group is recorded as A 0.1 .

[0042] ③ Use a skin erythema tester (Mexameter, or equivalent) to measure skin erythema. The skin erythema of the skin area in the control group is recorded as B 0 and the skin erythema of the skin area in the experimental group is recorded as B 0.1 .

[0043] ④ Distribute the product and inform the usage method. The subject takes the product home for use; the control group area uses a basic ingredient product without repair efficacy ingredients, and the experimental group area uses a product with repair efficacy ingredients.

[0044] ⑤ The subject needs to return to the laboratory for a follow-up visit on the 14th day (t1 = 14). Wait for 30 minutes in the test environment to balance and keep relaxed. Use a transepidermal water loss rate tester (Tewameter, or equivalent) to measure the transepidermal water loss rate. The transepidermal water loss rate of the skin area in the control group is recorded as A 1 and the transepidermal water loss rate of the skin area in the experimental group is recorded as A 1.1 .

[0045] ⑥ The subjects need to return to the laboratory for a follow-up visit on the 14th day (t1 = 14). Wait for 30 minutes in the test environment to balance and stay relaxed. Use a skin erythema tester (Mexameter, or equivalent) to conduct a skin erythema test. Record the skin erythema of the control group's skin area as B 1 , and record the skin erythema of the experimental group's skin area as B 1.1 .

[0046] ⑦ The subjects need to return to the laboratory for a follow-up visit on the 28th day (t2 = 28). Wait for 30 minutes in the test environment to balance and stay relaxed. Use a trans-epidermal water loss rate tester (Tewameter) to conduct a trans-epidermal water loss rate test. Record the trans-epidermal water loss rate of the control group's skin area as A 2 , and record the trans-epidermal water loss rate of the experimental group's skin area as A 2.1 .

[0047] ⑧ The subjects need to return to the laboratory for a follow-up visit on the 28th day (t2 = 28). Wait for 30 minutes in the test environment to balance and stay relaxed. Use a skin erythema tester (Mexameter) to conduct a skin erythema test. Record the skin erythema measurement of the control group's skin area as B 2 , and record the skin erythema measurement of the experimental group's skin area as B 2.1 .

[0048] Table 1 Skin erythema data Serial number Age D0 D14 D28 1 58 11.478 11.311 11.185 2 57 7.714 7.658 5.972 3 54 11.988 12.968 13.949 4 53 11.354 10.497 10.576 5 52 31.208 30.592 27.124 6 51 14.866 13.430 12.928 7 50 10.374 9.921 15.824 8 50 16.879 16.755 14.502 9 47 8.167 9.187 9.152 10 47 8.329 8.187 9.341 11 47 10.010 8.652 9.076 12 45 10.007 11.168 11.570 13 43 9.736 9.222 8.794 14 41 21.997 21.226 21.214 15 39 12.469 12.201 12.026 16 39 14.536 11.923 12.027 17 39 7.731 7.338 7.232 18 39 9.867 8.728 10.356 19 38 8.334 7.961 7.654 20 38 9.921 9.100 9.333 21 37 14.581 14.763 13.695 22 37 13.458 13.178 12.145 23 36 9.551 7.035 8.051 24 33 14.428 10.560 10.244 25 33 10.810 10.068 10.169 26 31 13.604 11.544 9.004 27 31 12.889 12.884 9.702 28 30 9.007 7.981 8.249 29 26 12.280 9.567 8.075 30 26 19.263 12.215 8.935 31 26 11.360 11.002 9.148

[0049] Table 2 Trans-epidermal water loss rate data

[0050] Construct the trans-epidermal water loss rate function Na and the skin erythema function Nb respectively;

[0051] Na = β 0 + β 1 F + β 2 A 0 + ε;

[0052] Nb = γ 0 + γ 1 F + γ 2 B 0 + μ;

[0053] Among them, Na and Nb are the change rates of trans-epidermal water loss rate and skin erythema respectively after using the repair product; F, A 0 , B 0 are age, the trans-epidermal water loss rate of the subject's facial skin before using the repair product, and the skin erythema of the subject's facial skin before using the repair product respectively, and are also the independent variables of the model; β0 , β 1 , β 2 , γ 0 , γ 1 , γ 2 are regression coefficients, and ε and μ are the random error terms of the model.

[0054] The assumptions for the model setting are mainly as follows:

[0055] 1) The dependent variables F, A 0 are deterministic variables, not random variables, and are not correlated with each other; the dependent variables F, B 0 are deterministic variables, not random variables, and are not correlated with each other.

[0056] 2) The random error terms ε, μ have a mean of 0, the same variance, and are independent of each other at different sample points, and there is no serial correlation;

[0057] 3) The dependent variables are not correlated with the random error terms, that is, F, A 0 , B 0 are not correlated with ε, μ;

[0058] 4) The random error terms ε, μ follow a normal distribution;

[0059] 5) The regression model is correctly specified.

[0060] Among these 5 assumptions in the multiple linear regression model, the first 4 assumptions are classical assumptions. If the first two assumptions are satisfied, the third assumption is naturally satisfied, and from the second assumption, there is...

[0061] According to the above preconditions of the multiple linear regression model, calculate the regression model for the trans-epidermal water loss rate of the skin:

[0062] Table 3 One of the regression models

[0063] R-squared = 0.489 = 48.9%: It means that the change in the trans-epidermal water loss rate before using the repair product (A) and age (F) affects the trans-epidermal water loss rate after using the repair product (Na) by 48.9%. 0

[0064] Table 4 Coefficients of One of the Regression Models

[0066] In the regression model calculation, the significant difference in which age (F) affects Na (the trans-epidermal water loss rate after using the repair product) is 0.018 < 0.05, showing a significant difference; the influence coefficient B = 0.121 > 0, indicating that the larger F is, the larger Na is, showing a positive correlation.

[0067] In the regression model calculation, A 0 (the trans-epidermal water loss rate without using the repair product) affects the significant difference in Na (the trans-epidermal water loss rate after using the repair product) is 0.000 < 0.05, showing a significant difference; the influence coefficient B = 0.462 > 0, indicating that the larger A 0 is, the larger Na is, showing a positive correlation.

[0068] According to the above prerequisites of the multiple linear regression model, calculate the regression model of skin erythema:

[0069] Table 5 Regression Model II

[0071] R-squared = 0.922 = 92.2%: indicating that the change situation of the predicted variables B 0 (skin erythema before using the repair product) and age (F) affecting Nb (skin erythema after using the repair product) is 92.2%.

[0072] Table 6 Coefficients of Regression Model II

[0074] In the regression model calculation, the significant difference in which age (F) affects Nb (skin erythema after using the repair product) is 0.003 < 0.05, showing a significant difference; the influence coefficient B = 0.087 > 0, indicating that the larger F is, the larger Nb is, showing a positive correlation.

[0075] In the regression model calculation, B 0 (skin erythema before using the repair product) affects the significant difference in Nb (skin erythema after using the repair product) is 0.000 < 0.05, showing a significant difference; the influence coefficient B = 0.905 > 0, indicating that the larger B 0 is, the larger Nb is.

[0076] Calculate the change rate of trans-epidermal water loss ΔNa and the change rate of skin erythema measurement ΔNb using the following formula as an index to evaluate the repair ability of the repair product:

[0077] (i) After 14 days of using the product by the subject, the repair ability of the product:

[0078] ΔNa 1= 100%×[(A 1.1 - A 0.1 ) / A 0.1 -(A 1 - A 0 ) / A 0 ;

[0079] ΔNb 1 = 100%×[(B 1.1 - B 0.1 ) / B 0.1 -(B 1 - B 0 ) / B 0 ;

[0080] Evaluation index of the product repair ability after 28 days of use by the subject:

[0081] ΔNa 2 = 100%×[(A 2.1 - A 0.1 ) / A 0.1 -(A 2 - A 0 ) / A 0 ;

[0082] ΔNb 2 = 100%×[(B 2.1 - B 0.1 ) / B 0.1 -(B 2 - B 0 ) / B 0 ;

[0083] (iii) After t days of use by the subject, the product repair ability index is also evaluated using the evaluation formulas in (i) and (ii).

[0084] In the existing T / CAB 0152-2022 "Test Methods for Seven Efficacy of Cosmetics: Anti-Wrinkle, Firming, Moisturizing, Oil Control, Repair, Nourishing, and Soothing" of the China Association for the Promotion of Industry-University-Research Cooperation Group Standard, in the 10.4 test steps, "Measure the transepidermal water loss rate and heme content of each test area respectively, and record them as the baseline values. After the baseline values are measured, distribute the samples. Provide usage instructions to the subjects according to the sample usage instructions to ensure that the subjects can use the product correctly. The test areas are divided into the sample application area and the control area. At the set measurement time points, revisit and measure the transepidermal water loss rate and heme content of each test area, and record them as the measured values." Among them, heme and erythema have the same trend and are replaceable. The test cycle of the T / CAB 0152-2022 standard is relatively long, and it is difficult to meet the time arrangements of the general consumer subject population. And this method can find the relationship between the age and skin type of the subjects and the repair ability of cosmetics through a regression model. The subjects only need to be revisited a few times to analyze the repair ability for the subjects. This method can more specifically recommend suitable cosmetics for different subjects.

[0085] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any minor modifications, equivalent replacements, and improvements made to the above embodiments based on the technical essence of the present invention shall be included within the protection scope of the technical solution of the present invention.

Claims

1. A skin repair efficacy evaluation method based on a multiple linear regression model, characterized in that: The method comprises: The subjects were randomly divided into a control group and an experimental group; initial facial skin data and age data were obtained, and a subject database was constructed; Facial skin data of a designated area of ​​a subject collected at time t after using the repair product, wherein the facial skin data includes epidermal data and skin color data, to establish a facial skin repair database of the subject; Construct a multivariate skin assessment model based on the subject database and facial skin repair database, and read the subject's skin repair plan; The skin repair program of the subject is analyzed, and the subject is notified to return for retesting at a specified time, and the repair ability evaluation index of the repair product is calculated through a multivariate skin evaluation model.

2. The skin repair efficacy evaluation method based on the multivariate linear regression model according to claim 1, characterized in that: The epidermal data is specifically the transepidermal water loss rate. The transepidermal water loss rate of the subjects is tested using a transepidermal water loss rate tester. The transepidermal water loss rate of the designated area of ​​the control group is recorded as A0, and the transepidermal water loss rate of the designated area of ​​the experimental group is recorded as A 0.1 .

3. The skin repair efficacy evaluation method based on the multivariate linear regression model according to claim 2, characterized in that: The subjects were tested for skin erythema using a skin erythema tester. The skin erythema at the designated area of ​​the control group was marked as B0, and the skin erythema at the designated area of ​​the experimental group was marked as B0. 0.1 ; The subjects used the skin repair program. At the time of the return visit at time t1, the transepidermal water loss rate of the designated area of ​​the control group was recorded as A1, and the transepidermal water loss rate of the designated area of ​​the experimental group was recorded as A 1.1 The skin erythema at the designated area of ​​the control group was marked as B1, and the skin erythema at the designated area of ​​the experimental group was marked as B 1.1 ; At the time t2, the transepidermal water loss rate of the designated area of ​​the control group was recorded as A2, and the transepidermal water loss rate of the designated area of ​​the experimental group was recorded as A 2.1 The skin erythema at the designated area of ​​the control group was marked as B2, and the skin erythema at the designated area of ​​the experimental group was marked as B 2.1。 4. The skin repair efficacy evaluation method based on the multivariate linear regression model according to claim 3, characterized in that: The transepidermal water loss function Na and skin erythema function Nb are constructed respectively; the specific functions are as follows: Na = β0 + β1F + β2A0 + ε; Nb = γ0 + γ1F + γ2B0 + μ; Where F is the age of the subjects; β0, β1, β2, γ0, γ1, γ2 are regression coefficients, and ε and μ are random error terms of the model.

5. The skin repair efficacy evaluation method based on the multivariate linear regression model according to claim 4, characterized in that: The values ​​of regression coefficients β0, β1, β2, γ0, γ1, γ2, and the values ​​of model random error terms ε and μ are obtained through the multivariate linear regression model; The following formulas are used to calculate the transepidermal water loss change rate ΔNa and the skin erythema change rate ΔNb as evaluation indicators for the repair ability of repair products: After the subjects used the product for t1, the evaluation indicators of the product's repair ability were: ΔNa1=100%×[(A 1.1 -A 0.1 ) / A 0.1 -(A1-A0) / A0]; ΔNb1=100%×[(B 1.1 -B 0.1 ) / B 0.1 -(B1-B0) / B0]; After the subjects used the product t2, the evaluation indicators of the product's repair ability were: ΔNa2=100%×[(A 2.1 -A 0.1 ) / A 0.1 -(A2-A0) / A0]; ΔNb2=100%×[(B 2.1 -B 0.1 ) / B 0.1 -(B2-B0) / B0]。 6. The skin repair efficacy evaluation method based on the multivariate linear regression model according to claim 4, characterized in that: In the transepidermal water loss function Na and the skin erythema function Nb, the variables F and A0 are deterministic variables and are uncorrelated with each other; the variables F and B0 are deterministic variables and are uncorrelated with each other; the random error terms ε and μ have zero mean and homoscedasticity, and are independent of each other at different sample points, without serial correlation; and F, A0, B0 are uncorrelated with ε and μ; the random error terms ε and μ obey normal distribution; Based on the prerequisites of the above multiple linear regression model, the multiple linear regression model of transepidermal water loss rate was calculated.

7. The skin repair efficacy evaluation method based on the multivariate linear regression model according to any one of claims 3 to 5, characterized in that: When the subject needed to return to the laboratory at time t1 or t2, he / she waited in the test environment for 25-30 min and remained relaxed.

8. The method for evaluating skin repair efficacy based on a multiple linear regression model according to any one of claims 3 to 5, characterized in that: The time t1 of the subject's first return visit is greater than or equal to four days, and the time interval between the subject's second return visit t2 and t1 is greater than or equal to seven days.

9. The skin repair efficacy evaluation method based on the multivariate linear regression model according to claim 7, characterized in that: Before the transepidermal water loss tester and skin erythema tester tests, the subjects waited for 25-30 minutes in a room with a temperature of 21±1℃ and a humidity of 50±10% and remained relaxed.

Images