CN115443095A

CN115443095A - Method for evaluating skin wrinkles

Info

Publication number: CN115443095A
Application number: CN202180030981.2A
Authority: CN
Inventors: 中村理惠
Original assignee: Kose Corp
Current assignee: Kose Corp
Priority date: 2020-06-15
Filing date: 2021-06-09
Publication date: 2022-12-06
Also published as: KR20230023611A; WO2021256339A1; JPWO2021256339A1; TW202201425A

Abstract

The present invention provides a technique capable of evaluating skin wrinkles more accurately for each person. A method for evaluating skin wrinkles, which estimates an index value of skin wrinkles of a subject person based on at least 2 values of an age value and a measured value of skin brightness of the subject person. A method for evaluating or exploring a wrinkle-improving composition, comprising: estimating a wrinkle index value of a subject to which a test substance has been applied, using the wrinkle evaluation method; and determining the test substance as a wrinkle-improving substance when the estimated wrinkle index value is lower than the wrinkle index value of the subject before application.

Description

Method for evaluating skin wrinkles

Technical Field

The present invention relates to a method for evaluating skin wrinkles and the like.

Background

Generally, skin wrinkles are likely to occur in many parts such as the forehead, the glabella, the mouth corner, and the canthus as they age, but the state of skin wrinkles is very different. Therefore, in the fields of cosmetics, beauty, medicines, and the like, techniques for preventing, improving, or treating skin wrinkles have been proposed for each individual for the purpose of anti-aging, and the like. Therefore, a method capable of accurately evaluating the state of skin wrinkles is required.

Conventionally, in order to evaluate the state of skin wrinkles, visual evaluation methods for evaluating the state of skin, device measurement methods for measuring the state of skin, and the like have been used.

For example, in order to evaluate the state of wrinkles, a method is known in which a visual evaluation is performed by a professional evaluator according to a standard table (0 to 7) of wrinkle level in a guideline (non-patent document 1) published by the japan cosmetic industry association.

For example, as a device measurement method for evaluating a wrinkle state, a two-dimensional image analysis method using oblique light illumination using a replica of a wrinkle part, a three-dimensional measurement method using a replica, and the like are known.

Various studies have also been made on methods for precisely evaluating the state of skin wrinkles.

For example, patent document 1 proposes a method for identifying skin texture and/or wrinkles, which includes: a step of performing image processing including cross-shaped binarization processing and/or short line matching processing on the skin image to obtain a physical quantity of the skin; and a step of substituting the physical quantity of the skin obtained in the above step into a prepared prediction formula, and identifying the obtained evaluation value as an evaluation value of skin texture and/or wrinkles.

Documents of the prior art

Patent document

Patent document 1: international publication No. 2009/142069

Non-patent document

Non-patent document 1: journal of the japan cosmetics society, "research council of cosmetics function evaluation law report: guide to the evaluation of cosmetic function method ", vol.30, no.4, pp.316-332 (2006).

Disclosure of Invention

Problems to be solved by the invention

Further, the present inventors have considered that, by being able to accurately and more easily evaluate skin wrinkles for each person, a better skin care technique suitable for each person can be easily proposed to prevent, improve or treat skin wrinkles.

Therefore, a main object of the present invention is to provide a technique capable of accurately and more easily evaluating skin wrinkles for each person.

Means for solving the problems

The present inventors have conducted extensive studies and, based on the accumulation of the long-term evaluation results of the equipment for each person over time and the results of multivariate analysis, have found for the first time an optimum combination of age and skin condition that allows estimation of an individual skin wrinkle index value, and have also found for the first time a wrinkle prediction model that is well-adapted. The present inventors have thus completed the present invention as follows.

The invention provides a method for evaluating skin wrinkles, which estimates an index value of skin wrinkles of a subject person based on at least 2 values of measured values of the age and skin brightness of the subject person.

In the method for evaluating skin wrinkles, the index value of skin wrinkles of the subject may be estimated based on 1 or 2 of the 2 values and the measured value of the redness of the skin and/or the measured value of the amount of skin fat.

In the method for evaluating skin wrinkles, the index value of skin wrinkles of the subject may be estimated by applying 2 values of the measurement values of the age and skin brightness of the subject to the wrinkle prediction model.

In the method for evaluating skin wrinkles, the index value of skin wrinkles of the subject may be estimated by applying 2 values of the measured value of the age and skin brightness of the subject and 1 or 2 values of the measured value of the skin redness and/or the measured value of the skin fat mass to the wrinkle prediction model.

The wrinkle prediction model may be a model obtained by deriving a linear mixed effect model in which an index value of a wrinkle of a person is used as a target variable and an explanatory variable includes at least a measurement value of age and skin brightness of the person.

In the wrinkle prediction model, the parameters of the linear mixed effect model may be parameters obtained by a maximum likelihood method or a constrained maximum likelihood method.

The present invention can provide a method for evaluating or searching for a wrinkle-improving composition, comprising:

estimating an index value of skin wrinkles of a subject to which a test substance has been applied, by using the wrinkle evaluation method; and

and determining the test substance as a wrinkle-improving substance when the estimated skin wrinkle index value is lower than the skin wrinkle index value of the subject before application and the wrinkles are improved.

Effects of the invention

The present invention can provide a technique capable of accurately and more easily evaluating skin wrinkles for each person. The effects described herein are not limited, and may be any of the effects described in the present invention.

Drawings

Fig. 1 is a flowchart showing an example of the skin wrinkle evaluation method of the present invention.

Fig. 2 is a flowchart showing an example of the skin wrinkle evaluation method of the present invention.

Fig. 3 is a flowchart showing an example of the skin wrinkle evaluation method of the present invention.

Fig. 4 is a hardware configuration diagram showing an embodiment of the present invention.

Fig. 5 is a schematic diagram showing an example of the skin wrinkle evaluation device or system according to the present invention.

Fig. 6 is a diagram showing an example of the skin wrinkle evaluation method and the information providing method for wrinkle improvement according to the present invention.

Fig. 7 is a block diagram schematically showing the processing procedure of the normal specialized AI.

Detailed Description

Preferred embodiments for carrying out the present invention will be described below. The embodiments described below are merely examples of typical embodiments of the present invention, and the scope of the present invention should not be construed as being limited thereto. In the present specification, the percentages refer to mass percentages unless otherwise specified. The upper limit and the lower limit of each numerical range (-), may be arbitrarily combined as desired.

1. Method for evaluating skin wrinkles according to the present invention

The invention provides a method for evaluating skin wrinkles, which estimates an index value of skin wrinkles of a subject person based on at least 2 values out of a measurement value of skin brightness and an age value of the subject person. Thereby enabling accurate and easier evaluation of skin wrinkles for each person.

In the present invention, it is preferable that 1 or 2 values of the measured value of the redness of the skin and/or the measured value of the amount of skin fat are combined with the 2 values (the measured value of the age value and the measured value of the skin brightness), and the index value of the skin wrinkles of the subject person is estimated based on these values. Thereby enabling accurate and easier evaluation of skin wrinkles for each person.

In addition, the present invention preferably uses a wrinkle prediction model, and estimates the skin wrinkle index value of the subject person using the 2 values (age value, measurement value of skin brightness) of the subject person or using 2 or more values selected from the 4 values of the subject person. The wrinkle prediction model has good model performance, can be evaluated by equipment, and can provide a wrinkle prediction model suitable for each person. Thereby enabling accurate and easier evaluation of skin wrinkles for each person.

In addition, the present invention can evaluate skin wrinkles of a subject by comparing the index value of the skin wrinkles with a reference value of a skin state. Thereby, skin care advice (particularly, wrinkle skin care) suitable for the subject person can be more easily and accurately presented.

The reference value of the skin state may be a value obtained using the wrinkle prediction model, and specifically may be obtained by applying (e.g., substituting or the like) preset values (an age value, a skin brightness value, a skin redness value, a sebum quantity value, and the like) in the wrinkle prediction model. As a value for obtaining the reference value, it is preferable that an age value is associated with a skin state (skin brightness, skin redness, sebum amount, etc.) value.

The reference value of the skin condition may be a reference value of a normal skin condition (for example, an average value of skin conditions of ages or ages corresponding to ages or ages of the subject).

The present invention will be described in more detail below.

1-1. The subject

The subject person is not particularly limited, but preferably a person, a man, a woman, a young and old, and more preferably a woman in view of the good wrinkle prediction accuracy of the subject person.

The human species of the subject is not particularly limited, but is preferably caucasian species (white family) or Mongolian species (yellow family), and more preferably Mongolian species (preferably Mongolian species A, B, C). Preferably, the subject's national system is selected from 1 or more than 2 of China, korea and Japan, more preferably Japan.

In the present specification, the term "subject" refers to a person who receives the wrinkle index value estimation by the evaluation method of the present invention.

The age group of the subject is not particularly limited, and the lower limit value is preferably 18 years or more, more preferably 20 years or more, and still more preferably 22 years or more, and the upper limit value is not particularly limited, and may be 80 years or less, 70 years or less, 60 years or less, or the like, for example. The appropriate numerical range is more preferably 18 to 70 years old, and still more preferably 22 to 60 years old, from the viewpoint of satisfactory wrinkle prediction accuracy.

1-2 evaluation of age and skin Condition of subject

In the evaluation method of the present invention, a value of a certain age of the subject and a measured value of the skin condition at that age (a measured value of skin brightness, a measured value of skin redness, a measured value of Pi Fupi fat mass, etc.) are used. The age value of the subject may be correlated or correlated with the measured value of each skin condition at that age.

1-2-1. Age value

The age value is not particularly limited, and is preferably an actual age value of the subject when evaluating the current skin wrinkle.

The "age" in the present specification can be represented by "chronological age +" the number of months elapsed after birth month "or" chronological age + "the number of months/12 months elapsed after birth date month". For example, the age of 30 years for 6 months may be 30.5. The "age" may be converted to "month age" of "number of months elapsed after birth date".

1-2-2 measurement of skin Brightness

The skin brightness is "brightness L" when skin color is represented by L a b color system.

In the system of "L × a × b" ", the brightness is denoted by" L ", and the chromaticities indicating the hue and chroma are denoted by" a "," b "" (a × and b × denote the directions of colors, a × denotes the red direction, -a × denotes the green direction, b × denotes the yellow direction, and-b × denotes the blue direction).

The measured value of skin color such as skin brightness and redness used in the present invention is not limited to the value measured by a measuring device capable of measuring the L a b color system, and the measured value (for example, the value of XYZ color system, RGB color system, or the like) measured by a measuring device capable of measuring other color systems may be converted into the value of L a b color system as the measured value of skin brightness, redness, or the like (for example, "measurement of color" of tea tree clean ", color material 57[10], p-568, 1984; temple macro light," development of color space ", japan society of image, no. 43 No. 2, p73-81, 2004).

The measured value of the skin brightness may be obtained using a measuring device or an imaging device that can measure L α b color system or the like. As described above, "lightness L" which is a value obtained by converting the value of another color system such as an XYZ color system or an RGB color system into the value of an L a b color system can be used as the measured value of the skin lightness.

The skin site to be measured is not particularly limited, but is preferably facial skin, and more preferably cheek (more specifically, upper cheek).

In the present invention, the operator or the control unit may select a palette panel (image or the like) having the same or similar skin brightness as the target person, and the control unit may select the measured skin brightness value data associated with the data of the selected palette panel as the measured skin brightness value of the target person. It is preferable that measurement value data of the color tone (value of the color system) of each color panel and the skin brightness corresponding to the color tone (value of the color system) of each color panel be stored in advance, and the measurement value data be set and stored as measurement value data of the skin brightness associated with the data of the color panel, whereby both can be easily converted to each other.

The "association storage" may be, for example, storage of specific data and other data in association with each other, but is not limited to this.

The skin brightness is preferably measured using a measuring device that can usually measure the L a b color system (for example, a hand-held color difference meter, a spectrocolorimeter, etc. conforming to JIS standards), but is not limited thereto, and an imaging device (for example, a camera, a video camera, a mobile device, etc.) may be used. In addition, a color matching panel may also be used.

The color difference meter is not particularly limited, and a color difference meter conforming to JIS standard is preferable. As the above-mentioned spectrocolorimeter, for example, an integrating sphere type spectrocolorimeter CM-700d (manufactured by Konika Meinenda) can be used. Color matching panels may also be used.

1-2-3 measurement of skin redness

The skin redness is "chromaticity a" when skin color is expressed by a color system of L a b.

Note that "L × a × b color system", "measurement device capable of measuring L × a × b color system", and "imaging device" are the same as those described in the above "measurement value of skin brightness", and therefore, description thereof is omitted.

The measured value of the skin redness can be obtained using a measuring device or an imaging device that can measure L α a b color system or the like. As described above in the "measured value of skin brightness", the "chromaticity a" obtained by converting the value of another color system such as an XYZ color system or an RGB color system into a value of L a b color system can be used as the measured value of skin red chromaticity.

In the present invention, the operator or the control unit may select, as the measured value of the skin redness of the subject person, the measured value data of the skin redness associated with the data of the selected palette panel by selecting the palette panel (image or the like) having the same or similar skin redness of the subject person. It is preferable that measurement value data of the skin redness corresponding to the hue (value of the color system) of each color panel and the hue (value of the color system) of each color panel be stored in advance, and the measurement value data be set and stored as measurement value data of the skin redness associated with the color panel data, whereby both can be easily converted to each other.

1-2-4 measurement of skin fat mass

The sebum amount of the skin refers to the sebum amount secreted from the skin surface.

The Pi Fupi measurement of the fat mass can be obtained using a measurement device or an imaging device capable of measuring the amount of sebum. In this case, the skin area to be measured is not particularly limited, but is preferably facial skin, and more preferably forehead (more specifically, center portion of forehead) from the viewpoint of enabling a stable amount of sebum to be collected in a short measurement time.

The Pi Fupi lipid level is preferably measured using a measuring instrument (e.g., sebumeter) that can measure the lipid level on the skin surface, but is not limited thereto, and an imaging device (e.g., camera, video camera, mobile terminal, etc.) may be used.

The sebum amount measuring device may be, for example, a spectrophotometer Sebumeter SM815 (manufactured by Courage + Khazaka). When a sebum amount measuring device is used, a measurement value of the sebum amount is obtained by attaching sebum secreted from the skin surface to an adhesive tape and measuring the light transmittance.

In the present invention, as described above, the measurement value of the skin brightness, the measurement value of the skin redness, and the measurement value of the sebum amount can be measured using a skin condition measurement device such as a measurement device or an imaging device, and sensory evaluation by a person can be omitted. When an imaging device is used, captured data or image data obtained by imaging can be converted into a measurement value of each skin condition (for example, skin brightness, skin redness, skin fat amount, etc.) by a program capable of converting data into a measurement value of each skin condition (for example, WO2013/094442, japanese patent application laid-open No. 8-308634, etc.). By using the measurement value obtained by such a skin condition measurement device, it is possible to more easily and accurately obtain a measurement value of a skin condition with less variation in numerical value than in sensory evaluation by a human.

When the measurement is performed, the cosmetic condition (presence or absence, etc.) of the subject is not particularly limited, and may be a cosmetic-on condition. From the viewpoint of enabling more accurate wrinkle evaluation, the subject is preferably a bare muscle (more preferably a plain skin), and the measurement is more preferably performed after the subject washes the skin (face) with a skin-cleansing agent such as a facial cleanser and the like to bring the skin (face) into a bare muscle (plain skin) state. In addition, the part where the measurement is performed may be bare muscle or skin after washing. Note that the "bare muscle (plain skin)" in the present specification means "original skin (face) without makeup.

More specifically, from the viewpoint of improving the wrinkle evaluation accuracy, it is preferable to wash the face with a facial cleanser to obtain a plain state, and then measure the skin after conditioning the skin for 20 minutes to 60 minutes in an indoor environment. More specifically, it is preferable to perform the measurement after the face washing and the measurement after the conditioning for 30 minutes in an environment controlled to 20 to 22 ℃ and 50. + -. 5%.

1-3 estimation of index value of skin wrinkle of subject

In the present invention, as described above, the index value of skin wrinkles of the subject person can be estimated based on 2 or more values selected from the group consisting of the age value, the skin brightness measurement value, the skin redness measurement value, and Pi Zhiliang measurement value of the subject person. Preferably, at least 2 values of age value and skin brightness measurement value are included.

As shown in the following [ examples ], the "index value of skin wrinkles" shows a high positive correlation with the "wrinkle level standard table (0,1,2,3,4,5,6,7)" in the guidelines published by the Japan cosmetic industry Association (journal of the Japan society of cosmetics, "guide on the study of the evaluation of cosmetic function, report by the Committee for the research of cosmetic function evaluation method", vol.30, no.4, pp.316 to 332 (2006)). Thus, the "index value of skin wrinkles" can be applied to a "wrinkle level standard table" which has been conventionally used as a standard, without conversion. Therefore, the index value of skin wrinkles according to the present invention can be said to be a value having high accuracy and high reliability in predicting wrinkles. In addition, the index value of skin wrinkles of the present invention can also be used as the wrinkle level of the "wrinkle level standard table". In addition, in the skin wrinkle evaluation method of the present invention, information such as a document using the "wrinkle level standard table" can be easily utilized.

Conventionally, in the "wrinkle level standard table" (wrinkle level) based on the guidelines published by the japan cosmetic industry association, since sensory evaluation is performed by visual observation, it is necessary to have a high degree of familiarity at the level of a professional evaluator in order to ensure the accuracy of wrinkle evaluation. In addition, when performing sensory evaluation, the greater the number of subjects, the more likely the determination is to be biased, and there is a risk of deterioration in wrinkle evaluation accuracy.

However, in the present invention, since the measurement value of the skin condition is evaluated by a device such as a measurement device or an imaging device, the measurement person who measures the skin condition does not need to have a high degree of familiarity required for sensory evaluation. Further, since the measurement value used in the present invention is a value obtained by performing device evaluation using a measurement device, it is possible to reduce numerical value fluctuations as compared with variations in sensory evaluation. In the present invention, the ability of the measuring person and the measuring method can be made uniform by using the measuring apparatus. Therefore, even if a plurality of measurements or a long-term measurement is performed, high reliability of the obtained measurement value data can be ensured.

The present invention thus enables easier and more accurate evaluation of skin wrinkles of a subject person, since it can be evaluated using the apparatus.

1-4 wrinkle prediction model

In the skin wrinkle evaluation method of the present invention, a wrinkle prediction model is preferably used.

The wrinkle prediction model is preferably a model obtained by deriving a linear mixed effect model in which an index value of a wrinkle of a person is used as a target variable and explanatory variables at least include an age (age value) and skin brightness (measured value of skin brightness) of the person.

Furthermore, the parameters (variables) of the linear mixed-effect model are preferably obtained using a maximum likelihood method or a constrained maximum likelihood method. The maximum likelihood method is a method of statistically estimating the point of the overall parameter of the probability distribution to be followed by given data. The constrained maximum likelihood method is an estimation method of parameters related to dispersion of a linear mixture model, is a maximum likelihood method with respect to a likelihood function, and is constrained to "error contrast" rather than data itself.

The explanatory variable more preferably contains 1 or 2 values of the skin redness (measured value of redness) and/or the skin sebum amount (measured value of sebum amount) in addition to the measured values of the age value and skin lightness, and the explanatory variable is more preferably 4 values thereof.

In the skin wrinkle evaluation method of the present invention, it is further preferable to apply the above-mentioned at least 2 values (age value and measurement value of skin brightness) to the wrinkle prediction model. It is further preferable to apply 2 values of the age value and the measured value of the skin lightness and 1 or 2 values of the measured value of the skin redness and/or the measured value of the skin fat mass to the wrinkle prediction model. Thereby enabling easier and more accurate evaluation of skin wrinkles.

In the wrinkle evaluation method of the present invention, as described above, by applying 2 or more values selected from the group consisting of "an age value", "a measured value of skin brightness", "a measured value of skin redness", and "a measured value of skin sebum amount" of the subject to the wrinkle prediction model, an index value (wrinkle level) of a current wrinkle can be estimated. In addition, by appropriately selecting values for predicting the future wrinkle state of the subject (the "age", "skin brightness", "skin redness", "Pi Fupi lipid level"), a future wrinkle reference value can be obtained using the values for predicting the future wrinkles. The wrinkle prediction model for obtaining the future wrinkle reference value may be the current wrinkle prediction model.

Further, by performing a comparative study of the current wrinkle criterion value and a future wrinkle criterion value (prediction), it is possible to suggest skin care to the subject.

The past wrinkle index value can be obtained as the wrinkle reference value using the past age of the subject and the measured value of the past skin condition at that time, in the same manner as described above. Thus, by comparing and studying these values based on the current wrinkle index value, the past wrinkle index value, the future wrinkle reference value (prediction), and the like, skin care advice of the subject can be proposed. In obtaining the past wrinkle index value, the past wrinkle prediction model may be used, or the current wrinkle prediction model may be used.

1-4-1. Predictor for wrinkle prediction model

The wrinkle prediction model is preferably a wrinkle prediction model composed of a specific wrinkle prediction factor.

The number of the specific wrinkle predicting factors is not particularly limited, but is preferably 2 to 5 (specifically, 2,3,4, 5), more preferably 3 to 5, even more preferably 4 to 5, and even more preferably 5.

Examples of the specific wrinkle predicting factor include "age", "skin brightness", "redness of skin", "sebum amount of skin", "interaction term between sebum amount and redness of skin", and the like. As the interaction, a synergistic interaction is preferable.

Among them, 2 or more of the 5 wrinkle predicting factors "age", "skin brightness", "skin redness", "skin sebum amount", "interaction of skin sebum amount and skin redness" are preferable.

At least 2 wrinkle prediction factors (specifically, "age" and "skin brightness") among the 5 wrinkle prediction factors are important for the wrinkle prediction model of the present invention, from the viewpoint of good prediction accuracy of wrinkles. Therefore, in the wrinkle evaluation method of the present invention, a wrinkle prediction model including a wrinkle prediction factor containing at least these 2 factors is preferable. By applying 2 values of "age value" and "measured value of skin brightness" of the subject person to this wrinkle prediction model, the wrinkle index value can be obtained more easily and more accurately.

Further, from the viewpoint of enabling wrinkle evaluation to be performed more accurately, it is more preferable that the 2 wrinkle prediction factors (specifically, "age" and "skin brightness") include 1 or 2 wrinkle prediction factors selected from the group consisting of "degree of redness of skin", "amount of sebum of skin", "interaction term of amount of sebum and degree of redness of skin", and a wrinkle prediction model composed of these wrinkle prediction factors.

More specifically, a wrinkle prediction model composed of 2 wrinkle prediction factors of "age" and "skin brightness" and 1 or 2 wrinkle prediction factors of "skin redness" and/or "skin fat mass" is more preferable. By applying 3 or 4 values (specifically, "age value", "measured value of skin brightness", "measured value of skin redness", "measured value of skin lipid amount") of the subject person to the wrinkle prediction model composed of these 3 or 4 factors, the wrinkle index value can be obtained easily and more accurately.

In the wrinkle prediction model of the present invention, from the viewpoint of particularly good accuracy of wrinkle prediction, a wrinkle prediction model composed of the above-described 5 wrinkle prediction factors (specifically, "age", "skin brightness", "skin redness", "skin sebum amount", "interaction term between skin sebum amount and skin redness") is more preferable. By applying 4 values of the subject (the "age value", "the measurement value of skin brightness", "the measurement value of skin redness", and the measurement value of skin fat mass) to the wrinkle prediction model, the wrinkle index value can be obtained easily and more accurately.

In the wrinkle prediction model of the present invention, wrinkle prediction factors other than the above-mentioned 5 wrinkle prediction factors (for example, skin water content, amount of transepidermal water transpiration, etc.) may be included in the configuration of the wrinkle prediction model of the present invention, and as described above, the wrinkle prediction model using the above-mentioned 5 wrinkle prediction factors is an optimal model from the viewpoint of the accuracy of wrinkle prediction.

The following wrinkle prediction equations 1 to 7 can be given as the equations of the wrinkle prediction model of the present invention (hereinafter also referred to as "wrinkle prediction equation"), but the present invention is not limited thereto.

Among the wrinkle prediction formulas 1 to 7, the wrinkle prediction formula 1 and the wrinkle prediction formula 2 are preferable because their wrinkle prediction accuracy is better and can be used as a standard for the subject, and further, the wrinkle prediction accuracy of the wrinkle prediction formula 1 is better, and thus more preferable.

< wrinkle prediction equation 1 (equation 1) >

Wrinkle rating _i =0.1469 × age

+0.7540 XLn (sebum)

+0.3270 × skin color a

+0.1654 × mean value [ skin color L ]

-0.1044 × [ Ln (sebum) × [ skin color a ] ]

－15.90

+b _0，i

(formula 1)

b _0，i ～N(0，0.4847)

Final equation for predicting Wrinkle grade (Wrinkle grade) of Japanese female aged 22-60 years

< wrinkle prediction equation 2 (equation 2) >

Wrinkle rating _i = 0.1200-0.2053 × age

+ 0.2200-1.280 XLn (sebum)

+ 0.050-0.5710 x skin color a

+ 0.050-0.4000 Xmean value [ skin color L ]

-0.030 to 0.1800 × [ Ln (sebum). Times.skin color a ]

－25.00～7.600

(formula 2)

In the method of the present invention, by executing the steps of the present invention, it is possible to estimate an index value of skin wrinkles of a subject, which has a high correlation with sensory evaluation based on a "wrinkle level standard table" viewed by a professional evaluator, and thus can be used as an evaluation value of skin wrinkles of the same degree as the "wrinkle level standard table".

Therefore, a target person can estimate an index value of skin wrinkles of the target person by selecting 2 or more values from among the measurement values of age, skin color (brightness, redness) and sebum amount and applying the selected values to the wrinkle prediction model or the wrinkle prediction formula of the present invention, and can evaluate skin wrinkles based on the index value.

Therefore, with the wrinkle evaluation method of the present invention, it is possible to easily and more accurately provide an index value of wrinkles that is highly positively correlated with the wrinkle evaluation derived by a professional evaluator from the "wrinkle level standard table".

With the method of the present invention, since the device evaluation can be performed by a skin condition measuring device such as a measuring device or an imaging device without depending on the sensory evaluation, the skin wrinkles can be evaluated more easily, accurately, and objectively. Therefore, it is possible to reduce the deterioration of wrinkle evaluation accuracy due to the proficiency of the measurer or individual differences. Furthermore, by using the wrinkle prediction model of the present invention, it is also possible to easily and accurately predict each person's future wrinkle index value. In addition, skin care advice corresponding to the future wrinkles can also be suggested.

The method for evaluating skin wrinkles of the present invention can be used for non-treatment purposes, and the evaluation result thereof can be finally used for treatment purposes. The present invention can be applied to, for example, a method for assisting in diagnosing skin wrinkles, etc., rather than the direct medical action of a doctor. Here, the "non-therapeutic purpose" is a concept excluding a medical action, that is, a therapeutic action of a human body, and examples of the non-therapeutic purpose include a cosmetic purpose, a provision of a skin care product, and the like. The present invention is advantageous in that the measurement of the age and skin condition is not performed by a doctor, and can be evaluated or measured by a measuring instrument, an imaging device, or the like.

In the present invention, "prevention" means prevention or delay of onset of symptoms or diseases in a subject to whom the composition is applied, or reduction of risk of onset of symptoms or diseases in a subject to whom the composition is applied. In the present technology, "improvement" refers to improvement or maintenance of a disease, symptom, or state of an applicable subject; prevention or delay of progression; reversal, prevention or delay of progression.

1-4-2. Wrinkle prediction model making method

The wrinkle prediction model of the present invention can be created according to the procedure of [ example ] below, and an example will be described below.

The number of participants in creating the wrinkle prediction model is, for example, 40 to 60, but is not limited thereto. The participants were subjected to at least 1 evaluation of skin properties for 1 year as the number of trials. The test period is not particularly limited, but is preferably 4 to 6 years. As a test method, the skin condition was measured after the participants had been washed with facial cleanser to develop bare skin and acclimatized in an environment controlled at 20 to 22 ℃ and 50 ± 5% for 30 minutes. More specific test methods can be carried out as described in [ examples ] below.

For statistical analysis for making a wrinkle prediction model, a general linear mixed effect model can be used. The wrinkle prediction factor is selected by adopting a stepwise regression method based on a backward elimination method, and a model with the minimum AIC (Akaike Information Criterion) is determined as an optimal model. Sensitivity analysis was performed using the obtained wrinkle predicting factor. It is preferable to use rver.3.5.2statistical software for analysis, but the analysis is not limited thereto as long as statistical analysis for creating a wrinkle prediction model can be performed.

2. Application example of skin wrinkle evaluation method according to the present invention

The method for evaluating skin wrinkles, the apparatus for evaluating skin wrinkles, the system for evaluating skin wrinkles, and the like according to the present invention will be described with reference to fig. 1 to 7, but the present invention is not limited to these drawings.

2-1. Skin wrinkle evaluation device and evaluation system

The method for evaluating skin wrinkles according to the present invention can also be implemented by a control Unit including a CPU (Central Processing Unit) or the like in a skin wrinkle evaluation device (e.g., a computer or the like) (see, for example, fig. 4 to 6). For example, the present invention may be a method for evaluating skin wrinkles (wrinkle prediction model, wrinkle evaluation procedure, program, and the like) according to the present invention, a method for evaluating skin wrinkles by a computer, or a method for providing skin wrinkle evaluation (for example, see fig. 1 to 3, fig. 7, and the like).

The method of the present invention may be implemented by a control unit that stores a program in a hardware resource such as a recording medium (e.g., a nonvolatile memory (e.g., a USB memory), an HDD, a CD, a DVD, a Blu-ray (registered trademark) Disc, or a web server) and performs determination of skin wrinkle evaluation. Alternatively, a skin wrinkle evaluation system, a wrinkle improvement composition evaluation or search system, or any of these devices may be provided by providing or using the control unit. The recording medium is preferably a computer-readable recording medium.

The device or system for wrinkle evaluation may include a keyboard input unit, a communication unit such as a network, an output unit such as a display, a storage unit such as an HDD, a measurement unit for performing skin condition measurement such as the measurement device or the imaging device, and the like. The device or system preferably includes an input unit, an output unit, a storage unit, and preferably further includes a communication unit and/or a measurement unit.

The input section may receive a user operation by an operator using the wrinkle evaluation method. The input section may include, for example, a mouse and/or a keyboard. The display surface of the display device may be an input unit that receives a touch operation.

The output unit may output the obtained skin wrinkle evaluation and information (e.g., tables, figures, explanatory text, etc.) related thereto. Examples of the output unit include a display device for displaying an image, a speaker for outputting sound, and a printing device for printing on a printing medium such as paper, but the output unit is not limited thereto.

The storage unit may store data input by an operator and data for wrinkle evaluation (for example, a wrinkle prediction model) set in advance. The storage unit may include a recording medium, for example.

Specific examples of the skin wrinkle evaluation device are not particularly limited as long as the device includes a CPU, and examples thereof include, but are not limited to, a mobile terminal (e.g., a notebook computer, a smartphone, a tablet terminal, and the like), a desktop personal computer, a server, and cloud computing. The skin wrinkle evaluation device preferably further includes a measurement unit, and for example, preferably a mobile terminal with a camera (e.g., a Web camera).

Data related to wrinkle evaluation, such as a program according to the method for evaluating skin wrinkles of the present invention, a wrinkle prediction model, wrinkle evaluation results obtained by the wrinkle evaluation method, and data for executing the steps of the present invention, may be stored in a storage unit or a server inside or outside the apparatus related to wrinkle evaluation, or on a cloud end.

< example of hardware configuration of skin wrinkle evaluation device and evaluation System according to the present invention >

The skin wrinkle evaluation device and evaluation system of the present invention can be executed by a program and hardware. An embodiment of the computer 1 according to an embodiment of the present invention will be described below with reference to fig. 4 to 6, but the present invention is not limited to this. The computer 1 includes at least a CPU as its components, and may further include 1 or 2 kinds selected from a RAM, a storage unit, an output unit, an input unit, a communication unit, a ROM, a measurement unit, and the like, preferably includes a RAM, a storage unit, an output unit, and an input unit, and preferably further includes at least 1 kind selected from a communication unit, a measurement unit, a ROM, and the like. The components are connected by a bus as a data transfer path, for example (see fig. 4).

The CPU is realized by, for example, a microcomputer, and controls each component of the computer 1. The CPU may use a control unit capable of performing determination such as data acquisition, data processing, evaluation, and the like, for example. These data acquisition and the like can be realized or executed by a program, for example, and the program can be read by the CPU to function. The ROM may store control data such as programs and operation parameters used by the CPU. The RAM may temporarily store programs executed by the CPU, for example.

The storage unit may store various data, and may store part or all of the data participating in the execution of the present invention. The data is not particularly limited, and examples thereof include calculation parameters such as personal data, wrinkle evaluation data, and product proposal data for each subject person, and programs. The storage unit may function as a database that exists inside or outside the apparatus. The storage unit may be implemented by a storage device or the like, for example.

The output unit can output personal data for each subject person, such as items (for example, name, arrangement number, etc.) for specifying the subject person, an age value of the subject person, and a measurement value of skin condition of the subject person, to the operator or the subject person; wrinkle evaluation data such as an index value of wrinkles, an evaluation result of wrinkles, an optimum wrinkle risk, future wrinkle prediction, a graph such as a graph, a table, and a figure thereof, and an image of a wrinkle part; information such as product proposal data including a display field of a single or a plurality of recommended products (product names, product images, etc.) such as "current recommended items, recommended items from the age of …", and "more extensive recommended items" than these items. The output unit may output, for example, each measurement value input field (numerical value input, color panel selection, and the like), each index value, a wrinkle evaluation result, a recommended product, and the like to the operator or the target person. The output unit can be realized by a Display unit such as an LCD (Liquid Crystal Display) or an OLED (Organic Light-Emitting Diode).

The input unit may acquire information such as an item for specifying the subject, an age value of the subject, and a measurement value of the skin condition of the subject, which are input by the operator or the subject. The input unit may be implemented by, for example, a microphone, a touch sensor, a keyboard, a mouse, a camera, or the like. The input unit may acquire data for evaluating skin wrinkles, for example, based on input of measurement values, selection of color panel images, selection of levels, selection of index values, and the like.

The communication unit may have a function of performing communication via an information communication network by using a communication technology such as Wi-Fi, bluetooth (registered trademark), or LTE (Long Term Evolution). The computer example 1 may be provided with a communication I/F (interface).

The Computer 1 may be, for example, a PC (Personal Computer) 1a, a server 1b, a smartphone terminal 1c, a tablet terminal, a mobile phone terminal, a PDA (Personal Digital Assistant), a wearable terminal (HMD: head Mounted Display, glasses type HMD, tape terminal, or the like) (see fig. 5). They may be independent (standalone) or they may be connected via a network.

The program for executing the method of the present invention may be stored in a computer device or a computer system other than the computer 1. In this case, the computer can use a cloud service that provides the functions of the program (see fig. 5). Examples of the cloud Service include SaaS (Software as a Service), iaaS (Infrastructure as a Service), paaS (Platform as a Service), and the like.

Further, the program may be stored using various types of computer-readable recording media, and supplied to the computer. Examples of the computer-readable recording medium include magnetic recording media (e.g., floppy disks, magnetic tapes, and hard disk drives), optical and magnetic recording media (e.g., magneto-optical disks), compact Read Only Memory (CD-ROM), CD-R, CD-R/W, and semiconductor memories (e.g., mask ROM, programmable ROM (PROM), erasable PROM (EPROM), flash ROM, random Access Memory (RAM)). In addition, the above-described program may be supplied to the computer through various types of computer recording media.

The computer 1 may have the above configuration selected as an alternative to the above configuration, or may be appropriately modified to another configuration.

The operator or the subject inputs, to the input unit of the computer 1, data of at least 2 values of the age value and the measured value of skin lightness of the subject, and 1 or 2 values of the measured value of skin redness and/or the measured value of skin fat mass. The control unit can select such data for input, and at this time, the control unit can access the storage unit and transmit the data related to them from the storage unit to the control unit. When data is input, in addition to numerical value input, the control unit may select a color panel image from among a plurality of color panel images, select a level from among a horizontal bar, or the like, and select these images or levels, thereby obtaining data such as age and measurement values based on these images or levels.

For example, in the case of skin brightness, an image having the color closest to or the same as the skin brightness of the subject person is selected from the brightness palette panel image, and the measured value of brightness associated with the selected image data is used as the measured value data of brightness of the subject person. In the case of the skin redness, an image having the color closest to or the same as the skin redness of the subject person is similarly selected from the red chroma color panel image, and the measurement value of the redness associated with the selected image is used as data of the measurement value of the redness of the subject person. In the case of the sebum amount of the skin, a level is selected from a horizontal bar or the like, and the numerical value of the selected level is used as data of the sebum amount measurement value of the subject.

When the computer 1 includes 1 or 2 or more types of measurement units, the data of each measurement value can be transmitted from the measurement unit to the control unit and input. Further, measurement value data obtained by measuring the subject person using the measurement device 2 existing outside the computer 1 may be transmitted to the computer 1 or the control unit via a network and input (see fig. 5).

Thus, the computer 1 estimates the index value of the skin wrinkle of the subject person by using the skin wrinkle evaluation method of the present invention based on these data, obtains the wrinkle evaluation result of the subject person, and provides the evaluation result as wrinkle information to the operator or the subject person.

The computer 1 may output wrinkle evaluation data obtained based on the skin wrinkle evaluation result, personal data, or the like to the output unit. The computer 1 may output a recommended product based on the skin wrinkle evaluation result to the output unit.

The computer 1 can output, for example, a wrinkle risk of the subject person as wrinkle evaluation data of the subject person, and provide the information to the operator or the subject person. More specifically, for example, a pattern (for example, a graph of a linear function) of the present wrinkle risk and the optimum wrinkle risk of the subject person, a degree of reduction of the target risk of the subject person (for example, past, present, future face images of the subject person), and the like may be output. Further, the computer 1 may output a single or a plurality of recommended articles (for example, cosmetics, external skin preparations, or the like) for the subject person as recommended items. Examples of the "recommended item" include a "current recommended item" for presenting a product suitable for the current situation, "a" replacement item from the age of … "for presenting a product suitable for the future, and a" broader recommended item "for expanding the range of product selection for the target person, but the present invention is not limited thereto, and the operator or the target person can appropriately switch or select the item of the" recommended item "to be output.

The computer example 1 may store the data of these results in the storage unit as personal data of the subject person or data attached to an item for specifying the subject person. Thus, the control unit can appropriately access the storage unit and use the data of the results in the storage unit for various purposes. The proposal or provision of recommended products will be described below in "3. Method for evaluating or searching for a wrinkle-improving composition according to the present invention".

2-2 examples of skin wrinkle evaluation according to the present invention

The procedure of skin wrinkle evaluation according to the present invention will be described more specifically, but the skin wrinkle evaluation according to the present invention is not limited thereto. The explanation of the skin wrinkle evaluation procedure may be an explanation of the operation of the skin wrinkle evaluation method, the skin wrinkle evaluation device, and the evaluation system according to the present invention.

The operator who performs the skin wrinkle evaluation can input various data such as an age value of the subject person, a measurement value of skin brightness, and items (for example, name, identification number, and the like) for specifying the subject person through the input unit. This allows the input unit to transmit the data to the control unit.

In addition, the data specifying the items (for example, name, house number, etc.) of the subject person may be stored in the storage unit in advance in association with the measurement value of the skin condition and/or the age data of the subject person by the input of the operator. The control unit may be configured to control transmission of data associated with the input specific item of the target person from the storage unit to the control unit.

In addition, when the measurement unit measures the skin condition of the subject, data of the measurement value may be transmitted from the measurement unit to the control unit. In this case, the control unit may be configured to store the data in association with the specific item of the target person by the storage unit.

2-2-1. Embodiments 1 and 2 relating to the evaluation of skin wrinkles of the present invention

In

embodiments

1 and 2, the index value of skin wrinkles of the subject can be estimated based on 4 values of the age value and the measured value of skin brightness, the measured value of skin redness, and the measured value of Pi Fupi fat mass of the subject (see, for example, fig. 1).

In step 101, 4 values, that is, the age value of the subject, the measurement value of skin brightness, the measurement value of skin redness, and the measurement value of Pi Fupi fat mass, are transmitted to the control unit.

In step 102, the control unit estimates an index value of skin wrinkles of the subject based on 4 values of the age value of the subject, the measured value of skin brightness, the measured value of skin redness, and the measured value of Pi Fupi fat mass. The control unit outputs an index value of wrinkles of the subject person.

In embodiment 1a, the index value of skin wrinkles of the subject person can be estimated based on 4 values of the age value, the measured value of skin brightness, the measured value of skin redness, the measured value of Pi Fupi fat mass, and a wrinkle prediction model (wrinkle prediction formula 1 below) (see, for example, fig. 2).

Note that, in embodiment 2a, the index value of skin wrinkles of the subject person can be estimated in the same manner as in embodiment 1a except that the wrinkle prediction equation 1 is replaced with the wrinkle prediction equation 2 (for example, see fig. 2).

The wrinkle prediction formula 1 shown in formula 1 below is a wrinkle prediction formula composed of 5 wrinkle prediction factors, i.e., "age", "skin brightness", "skin redness", "skin sebum amount", "interaction term between sebum amount and skin redness".

In step 101a, the control unit may substitute 4 values of "age value", "measurement value of skin brightness", "measurement value of skin redness", and "measurement value of skin sebum amount" of the subject into a wrinkle prediction model (wrinkle prediction formula 1 below) to estimate an index value of wrinkles (wrinkle level) (see, for example, fig. 2).

In step 102a, a wrinkle reference value may be obtained based on a value preset for predicting the future wrinkle state of the subject and the wrinkle prediction formula 1 (for example, see fig. 2). The preset value is 1 or 2 or 3 or more selected from an age value, a skin brightness value, a skin redness value, and a skin sebum value, and the control part may obtain the wrinkle reference value by substituting the selected value into the wrinkle prediction formula 1. The operator can also appropriately substitute a value for predicting the future wrinkle state of the subject person.

In step 103a, a skin care suggestion may be made for the subject by comparing the index value of the wrinkles with the wrinkle reference value.

Specifically, when the wrinkle index value is lower than the wrinkle reference value, the time at which the wrinkle reference value is reached is outputted to the subject as the time at which the use of the wrinkle-improving composition is started. When the index value of wrinkles is equal to or greater than the wrinkle reference value, information indicating the start of use of the wrinkle-improving composition is output to the subject.

As described above, the operator may suggest skin care based on the wrinkle index value and the wrinkle reference value output from the control unit.

< wrinkle prediction equation 1 (equation 1) >

Wrinkle rating _i =0.1469 × age

+0.7540 XLn (sebum)

+0.3270 × skin color a

+0.1654 × mean value [ skin color L ]

-0.1044 × [ Ln (sebum) x skin color a ]

－15.90

+b _0，i

(formula 1)

b _0，i ～N(0，0.4847)

Final equation for predicting wrinkle grade in Japanese women aged 22-60 years

< wrinkle prediction equation 2 (equation 2) >

The wrinkle prediction formula 2 shown in the following formula 2 is a wrinkle prediction formula composed of 5 wrinkle prediction factors of "age", "skin brightness", "skin redness", "skin sebum amount", "interaction term of sebum amount and skin redness". Note that, portions overlapping with those described in

embodiments

1 and 1a above are appropriately omitted from description.

By substituting 4 values of "age", "skin brightness", "skin redness", "Pi Fupi fat mass" of the subject into the wrinkle prediction model (wrinkle prediction formula 2), an index value (wrinkle level) of wrinkles can be estimated (for example, see fig. 2). Further, by substituting 4 values of preset values ("age value", "skin brightness value", "skin redness value", and skin sebum value ") into the wrinkle prediction formula 2, a wrinkle reference value can also be obtained (for example, see fig. 2).

In embodiment 2a, as in embodiment 1a using the wrinkle prediction equation 1, the future wrinkle state of the subject can be predicted, and skin care advice of the subject can be provided.

Wrinkle rating _i = 0.1200-0.2053 × age

+0.2200 ~ 1.280 XLn (sebum)

+ 0.050-0.5710 x skin color a

+ 0.050-0.4000 × mean value [ skin color L ]

-0.030 to 0.1800 × [ Ln (sebum). Times.skin color a ]

－25.00～7.600

(formula 2)

2-2-2. Embodiment 3 relating to skin wrinkle evaluation of the present invention

As embodiment 3, an index value of skin wrinkles of the subject person can be estimated based on at least 2 values of the measured values of the age and skin brightness of the subject person (see, for example, fig. 1). Note that, although a diagram of the flow of steps of embodiment b itself is omitted here, the flow can be understood by referring to fig. 1 and 2.

In step 301, at least 2 values of the age value and the skin brightness measurement value of the subject person are transmitted to the control unit.

In step 302, the control unit estimates an index value of skin wrinkles of the subject person based on 2 values of the age value and the measured value of skin brightness of the subject person. The control unit outputs an index value of wrinkles of the subject person.

In embodiment 3a, the index value of skin wrinkles of the subject person can be estimated based on 2 values of the age value and the measured value of skin brightness of the subject person and based on a wrinkle prediction model (wrinkle prediction formula 3 below) (for example, see fig. 2).

The wrinkle prediction formula 3 shown in the following formula 3 is a wrinkle prediction formula composed of 2 wrinkle prediction factors of "age" and "skin brightness". Note that portions overlapping with those in embodiments 1 to 2 described above are appropriately omitted from description.

In step 301a, the control unit can estimate an index value (wrinkle level) of wrinkles by substituting 2 values of "age value" and "measured value of skin brightness" of the subject into a wrinkle prediction model (wrinkle prediction formula 3, below) (see, for example, fig. 2). Further, by substituting preset values ("2 values of age value and skin brightness value") into the wrinkle prediction equation 3, a wrinkle reference value can also be obtained (for example, see fig. 2).

As in embodiments 1 to 2, the wrinkle state of the subject in the future can be predicted, and skin care advice can be provided to the subject.

< wrinkle prediction equation 3 (equation 3) >

Wrinkle rating _i =0.1484 × age

+0.1844 × mean value [ skin color L ]

－14.80

(formula 3)

2-2-3. Embodiment 4 relating to skin wrinkle evaluation of the present invention

In embodiment 4, the index value of skin wrinkles of the subject can be estimated based on 3 values, which are the age value and the measured value of skin brightness of the subject and the measured value of Pi Fupi fat mass (see, for example, fig. 1). Note that, the flowchart of the steps of the embodiment itself is omitted, but the flowchart can be understood by referring to fig. 1 and 2.

In step 401, at least 3 values of the age value, the measured value of skin brightness, and the measured value of Pi Fupi fat mass of the subject are transmitted to the control unit.

In step 402, the control unit estimates an index value of skin wrinkles of the subject based on 3 values of the age value, the measured value of skin brightness, and the measured value of Pi Fupi fat mass of the subject. The control unit outputs an index value of wrinkles of the subject person.

In embodiment 4a, the index value of skin wrinkles of the subject person can be estimated based on 3 values of the age value, the measured value of skin brightness, and the measured value of Pi Fupi fat mass of the subject person and based on a wrinkle prediction model (wrinkle prediction formula 4 described below) (see, for example, fig. 2).

The wrinkle prediction formula 4 shown in the following formula 4 is a wrinkle prediction formula composed of 3 wrinkle prediction factors of "age", "skin brightness", "Pi Fupi lipid level". Note that description of portions overlapping with those in embodiments 1 to 3 is appropriately omitted.

In step 401a, the control unit can estimate an index value (wrinkle level) of a wrinkle by substituting 3 values of "age value", "skin brightness value", and "skin sebum value" of the subject into a wrinkle prediction model (wrinkle prediction formula 3). Further, by substituting 3 values of preset values ("age value", "skin brightness value", and skin sebum value ") into the wrinkle prediction formula 4, a wrinkle reference value can also be obtained (for example, see fig. 2).

As in embodiments 1 to 3, the wrinkle state of the subject in the future can be predicted, and skin care advice can be provided to the subject.

< wrinkle prediction equation 4 (equation 4) >

Wrinkle rating _i =0.1504 × age

+0.1880 × mean value [ skin color L ]

+0.044 XLn (sebum)

－15.28

(formula 4)

2-2-4. Embodiment 5 relating to skin wrinkle evaluation of the present invention

In embodiment 5, an index value of skin wrinkles of a subject can be estimated based on 4 values of the age value, the measured value of skin brightness, the measured value of Pi Fupi fat mass, and the measured value of skin redness of the subject (see, for example, fig. 1). Note that, although a flowchart of the steps of this embodiment itself is omitted, this flowchart can be understood by referring to fig. 1 and 2.

In step 501, 4 values of the age value, the measured value of skin brightness, the measured value of Pi Fupi fat mass, and the measured value of skin redness of the subject are transmitted to the control unit.

In step 502, the control unit estimates an index value of skin wrinkles of the subject based on 4 values of the age value, the measured value of skin brightness, the measured value of Pi Fupi fat mass, and the measured value of skin redness of the subject. The control unit outputs an index value of wrinkles of the subject person.

In embodiment 5a, the index value of skin wrinkles of the subject can be estimated based on 4 values of the age value, the measured value of skin brightness, the measured value of Pi Fupi fat mass, the measured value of skin redness, and the wrinkle prediction model (the wrinkle prediction formula 5 described below) (for example, see fig. 2).

The wrinkle prediction formula 5 shown in the following formula 5 is a wrinkle prediction formula composed of 4 wrinkle prediction factors of "age", "Pi Fupi fat mass", "skin redness", "skin brightness". Note that portions overlapping with those in embodiments 1 to 4 described above are appropriately omitted from description.

In step 501a, the control unit can estimate an index value (wrinkle level) of wrinkles by substituting 4 values of "age value", "measurement value of skin sebum amount", "measurement value of skin redness", "measurement value of skin brightness" of the subject person into a wrinkle prediction model (wrinkle prediction formula 5, described below) (see, for example, fig. 2).

Further, a wrinkle reference value can also be obtained by substituting 4 values of preset values ("age value", "skin sebum value", "skin redness value", and skin brightness value ") into the wrinkle prediction equation 5 (for example, see fig. 2).

As in embodiments 1 to 4, the wrinkle state of the subject in the future can be predicted, and skin care advice can be provided to the subject.

< wrinkle prediction equation 5 (equation 5) >

Wrinkle rating _i =0.1460 × age

+0.035 XLn (sebum)

+0.013 × skin color a

+0.1812 × mean value [ skin color L ]

(formula 5)

2-2-5. Embodiment 6 relating to skin wrinkle evaluation of the present invention

In embodiment 6, an index value of skin wrinkles of the subject can be estimated based on 4 values of the age value, the measured value of skin brightness, the measured value of Pi Fupi fat mass, and the measured value of skin redness of the subject (see, for example, fig. 1 to 3). By applying 2 values of the age value and the measured value of the skin brightness of the subject person to the wrinkle prediction model of embodiment 6, the index value of the skin wrinkle of the subject person can be estimated. More specifically, embodiment 6 can more accurately suggest skin care advice suitable for the subject by determining whether the skin is bright or dark based on the measured value of the skin brightness of the subject, distinguishing the skin brightness state (bright or dark) of the subject, and evaluating skin wrinkles for each case.

Note that description of portions overlapping with those in embodiments 1 to 5 is omitted as appropriate. The configurations of embodiments 1 to 5 described above can be used as appropriate.

< determination of skin Brightness >

In step 601, the control unit determines the brightness of the skin of the subject person based on the measured value of the skin brightness. Specifically, the control unit compares the measured value of the skin brightness of the subject person with a reference value of the skin brightness. The reference value of the skin brightness may be set as appropriate, and may be, for example, an average value of skin brightness of ordinary people (preferably women, particularly yellow race (Mongolian spook)).

When the skin brightness of the subject person is equal to or higher than the reference value of the skin brightness (Yes), the skin of the subject person is judged to be bright, and when the skin brightness of the subject person is not equal to or higher than the reference value (No), the skin of the subject person can be judged to be dark (for example, see fig. 3).

< case where skin is bright (S602 to S607) >

In step 602, the control unit selects wrinkle prediction equation 6 when determining that the skin of the subject person is bright (Yes).

When the subject is a bright skin, the control unit can estimate the wrinkle index value (wrinkle level) by substituting 3 values of "age value", "measured value of redness of skin", and "measured value of skin fat mass" of the subject into wrinkle prediction equation 6.

In step 603, the control part may obtain a wrinkle reference value of the bright skin based on the value for predicting the future wrinkle state and wrinkle prediction equation 6, unlike in step 602. Specifically, values of the skin states for predicting the future wrinkle state of the subject person can be selected, and the selected values can be substituted into wrinkle prediction equation 6. The operator may also input values.

In step 604, the control part may obtain a standard skin wrinkle reference value based on the value for predicting the future wrinkle state and the wrinkle prediction equation 1 or the wrinkle prediction equation 2, unlike in step 603. Specifically, using the wrinkle prediction formula 1 representing a standard person, a standard skin wrinkle reference value can be obtained based on the value for predicting the future wrinkle state and the

wrinkle prediction formula

1 or 2, as in the

above embodiments

1 and 2. The operator may also input values.

If step 604 is present, step 603 may be skipped, and the control unit may compare the wrinkle index value with the standard skin wrinkle reference value. In the case where step 603 is present, step 604 may be skipped, and the control unit may compare the wrinkle index value with the wrinkle reference value for bright skin.

In step 605, the control unit compares the wrinkle reference value of the bright color and/or the standard wrinkle reference value with the wrinkle index value. This makes it possible to suggest skin care to a subject with bright skin.

Specifically, in step 606, when the index value of wrinkles is lower than the wrinkle reference value, the time at which the wrinkle reference value is reached is output to the subject as the time at which the wrinkle-improving composition starts to be used.

In step 607, when the index value of wrinkles is equal to or more than the wrinkle reference value, information for starting the use of the wrinkle-improving composition is output to the subject.

From the prediction formulas 6 and 7, it is considered that the skin tends to be wrinkled when it is bright. Therefore, by using the bright wrinkle reference value, it is easy to prevent or improve wrinkles and the like in a subject whose skin is bright.

As described above, the operator may propose skin care advice based on the wrinkle index value and the wrinkle reference value output from the control unit.

< skin color (S608 to S613) >

In step 608, the control unit selects wrinkle prediction equation 7 when determining that the target person is not bright skin (No) (i.e., the skin of the target person is dark).

When the subject is a dark skin, the control unit can estimate the wrinkle index value (wrinkle level) by substituting 3 values of "age value", "measured value of skin redness", and "measured value of skin fat mass" of the subject into wrinkle prediction equation 7.

In step 609, the control section may obtain a wrinkle reference value of the dark skin based on the value for predicting the future wrinkle state of the subject person and wrinkle prediction equation 7, differently from step 608. Specifically, values of the skin states for predicting the future wrinkle state of the subject person may be selected, and the selected values may be substituted into wrinkle prediction equation 7. The operator may also input values.

In step 610, the control unit performs the same operation as in step 603, and can obtain a standard skin wrinkle reference value, unlike in step 609.

If step 609 is present, step 610 may be skipped, and the control unit may compare the wrinkle index value with the standard skin wrinkle reference value. In the case where step 610 is present, step 609 may be skipped, and the control unit may compare the wrinkle index value with the bright skin wrinkle reference value.

In step 611, the control unit compares the wrinkle reference value for dark wrinkles and/or the standard wrinkle reference value with the wrinkle index value. Thereby skin care advice can be presented to the dark-skinned subject.

Specifically, in step 612, when the index value of wrinkles is lower than the wrinkle reference value, the time at which the wrinkle reference value is reached is output to the subject as the time at which the wrinkle-improving composition starts to be used.

In step 613, when the index value of wrinkles is equal to or greater than the wrinkle reference value, information is output to the subject that use of the wrinkle-improving composition is started.

From the prediction formulas 6 and 7, it is considered that the skin tends to be wrinkled when it is bright. Therefore, by using the standard wrinkle reference value, it becomes easy to prevent or improve wrinkles and the like in a subject whose skin is dark.

The wrinkle prediction equation 6 that the subject can apply to the case of bright skin is preferably composed of 4 wrinkle prediction factors of "age", "skin redness", "skin sebum amount", "interaction term of sebum amount and skin redness". On the other hand, when the subject is a dark skin, it is preferable to use wrinkle prediction formula 7 including 3 wrinkle prediction factors of "age", "skin redness", "Pi Fupi fat mass". When these wrinkle prediction formulas are used, the subject preferably has a "age" value and measured values of "skin brightness", "skin redness", and "Pi Fupi fat mass" at that age.

Each wrinkle prediction formula can be obtained by the <1-4-2. Method for preparing wrinkle prediction model >, and can be obtained using the data obtained in test example 1 below [ example ].

3. Method for evaluating or searching for wrinkle-improving composition according to the present invention

The following describes embodiments of the method for evaluating or searching for a wrinkle-improving composition of the present invention. In the description of the present embodiment, the description of each configuration, evaluation method, and the like overlapping with the above-described "1." and "2." is appropriately omitted, and the description of "1." and "2." is also applicable to the present embodiment and can be appropriately adopted. In the present embodiment, the skin wrinkle evaluation and wrinkle prediction models (for example, wrinkle prediction equations 1 to 7) described above can be used as appropriate.

estimating an index value of skin wrinkles of the subject to which the test substance has been applied, by using the wrinkle evaluation method described in the above "1." and "2."; and

and determining the test substance as a wrinkle-improving substance when the estimated skin wrinkle index value of the subject is lower than the skin wrinkle index value of the subject before application.

This makes it possible to easily and accurately estimate the wrinkle-improving index value of a subject and to more easily and accurately evaluate or search for a wrinkle-improving composition suitable for each subject. In the method for evaluating or searching for a wrinkle-improving composition of the present invention, it is preferable to use the mobile terminal with a camera of the present invention because recording can be performed continuously and finely and the operation is simple.

The test substance is not particularly limited, and may be a commercially available product that is marketed as a composition for preventing, ameliorating or treating wrinkles. The degree of efficacy of a commercial product varies depending on individual differences, and therefore, a commercial product suitable for an individual can be selected in the present invention. The test substance may be a new substance or an existing substance expected to prevent, improve or treat wrinkles, and thus a more suitable substance for preventing, improving or treating wrinkles can be selected and searched for.

The administration (administration) may be either oral administration or non-oral administration (e.g., injection, coating, etc.), and is preferably coating.

The administration period is appropriately selected, and may be 1 time for 1 day, may be divided into a plurality of times for 1 day, and may be 1 time for several days or several weeks.

In addition, the present invention can provide a method for providing a wrinkle-improving composition, comprising:

a step of estimating a skin wrinkle index value of the subject person by using the wrinkle evaluation method described in the above "1." and "2."; and

and determining a single or a plurality of wrinkle-improving compositions based on the estimated skin wrinkle index value of the subject. This makes it possible to determine a wrinkle-improving composition corresponding to the wrinkle index value, and to propose or provide a wrinkle-improving composition suitable for the subject. Preferably, the method further comprises a step of outputting the determined wrinkle-improving composition through an output unit. The wrinkle-improving composition that presents or provides an output to the subject may be performed by the operator or the control unit.

The wrinkle-improving composition may be a wrinkle-preventing or wrinkle-treating composition, and is not particularly limited and may be a commercial product.

For each wrinkle-improving composition, data of skin wrinkle index values to which each wrinkle-improving composition can be associated with or attached to specific data. The skin wrinkle index value that can be corresponded to can be obtained based on the known efficacy of the wrinkle-improving composition or the results of the above-described method for evaluating or searching for a wrinkle-improving composition of the present invention. By associating data of skin wrinkle improvement indices with data of wrinkle improvement compositions or attaching data to wrinkle improvement compositions, it is possible to more easily and accurately search for wrinkle improvement compositions corresponding to the skin wrinkle state (for example, wrinkle index values, wrinkle levels, etc.) of a subject person, and it is possible to more easily and accurately provide information of the searched wrinkle improvement compositions to the subject person. This can provide information on a wrinkle-improving composition more suitable for the state of skin wrinkles of a subject.

In addition, the data of each wrinkle-improving composition may be grouped in units of a predetermined range based on skin wrinkle index value data to which each wrinkle-improving composition may correspond. The control unit may perform grouping based on data of wrinkle index values that can be associated with data given to each wrinkle-improving composition. The grouping in units of the predetermined range can be performed by sorting the standard table of wrinkle levels, and examples thereof include a wrinkle level of 3 to 4: a wrinkle-improving composition A, B, C; wrinkle grade 4-5: wrinkle compositions D, F, and the like; wrinkle grade 5-6: wrinkle-improving compositions G, H, i, J, etc.; wrinkle grade 6-7: a wrinkle-improving composition O, P. Data for each composition may be assigned a corresponding wrinkle index value (e.g., 4.5 or 6.7, etc.). The interval in the predetermined range is not particularly limited, and may be 1 or 0.5 intervals, and may be classified into 8 classes of 0,1,2,3,4,5,6, and 7 at intervals of 1, for example. By grouping, it is possible to present or provide the operator or the subject person with single or plural pieces of information of the group containing the wrinkle-improving composition corresponding to the skin wrinkle state of the subject person more easily and accurately.

Various data relating to the wrinkle-improving composition, such as data of the wrinkle-improving composition, and applicable wrinkle index value data and group data, can be stored in the storage unit.

The wrinkle-improving composition proposed or provided to the operator or the subject may be a wrinkle-improving composition corresponding to the present or future level of wrinkles of the subject. For example, the control unit may determine a single or a plurality of wrinkle-improving compositions or groups equal to or higher than (more preferably equal to or higher than 1 rank) the skin wrinkle index value of the subject, and output, present, or provide information on the determined single or plurality of wrinkle-improving compositions or groups to the operator or subject. The control unit may present or provide the information of the determined wrinkle-improving composition or group as the "currently recommended item" of the subject person.

The control unit may obtain the future skin wrinkle index value of the subject based on the skin wrinkle index value of the subject and the wrinkle prediction model, and output, present, or provide to the operator or the subject, as the future wrinkle improvement composition, information on the wrinkle improvement composition or the wrinkle improvement group that is equal to or higher than (more preferably equal to or higher than 1 st) the future wrinkle index value based on the future skin wrinkle index value of the subject, as in the case of the past. The control unit may set the determined wrinkle-improving composition or group as a "future recommended item" such as "replacement item from the age of …" of the subject person. In addition, the control unit may further expand the range of the level to the upper 1 st to the lower 1 st, for example, as to "the skin wrinkle index value of the subject is equal to or greater than (more preferably equal to or higher than 1 st)", thereby expanding the range of information of the wrinkle-improving composition proposed or provided, and the number of products may be increased, and information in such an expanded range may be presented or provided to the operator or the subject as "a broader recommended item".

The following shows methods for providing a wrinkle preventing, improving or treating composition using the wrinkle prediction model of the present invention as examples 1 and 2. The wrinkle prediction model may use the wrinkle prediction models or wrinkle prediction formulas of "1." and "2." described above as appropriate.

< example 1>

In step 701, the control unit inputs 4 values of a measured value of the skin brightness, a measured value of the redness, and a measured value of the sebum amount of the subject to the wrinkle prediction model. This enables the wrinkle prediction formula a to be set for 1 subject.

In step 702, the control unit inputs the age value of the subject person into the wrinkle prediction formula a, and outputs the index value of the current wrinkle of the subject person. This enables the operator or the subject to understand the present wrinkle risk (wrinkle level) of the subject.

In step 703, the control unit can output a wrinkle reference value and the age at that time by inputting the wrinkle levels 3 to 5 as the optimal wrinkle risk value to the wrinkle prediction formula a. The lower the wrinkle level is set, the more optimal the risk of wrinkles becomes.

The output is not particularly limited to screen display, voice display, and the like. The operator of the wrinkle prediction model is not particularly limited to the sales staff, the advisor, the operator, the subject person, and the like.

When a numerical value input field exists on the input screen for data of measured values (a measured value of brightness, a measured value of redness, and a measured value of sebum) of each skin condition of the subject person, the operator can input the measured values in the numerical value input field. When an input field of an image of a palette or a selection level exists on the input screen, the operator or the subject selects a palette or a level that is the same as or similar to the skin condition of the subject from among images or levels in which a plurality of palette panels exist, and the control unit selects measurement value data associated with or attached to the data from the palette panels, thereby obtaining data of the measurement values. When selecting the palette panel, the operator may select a palette panel that is the same as or similar to the skin condition of the subject person by comparing the skin condition visually observed or the captured image data, or the control unit may select a palette panel that is the same as or closest to the color tone of the captured skin image data of the subject person.

In step 704, when the index value of the current wrinkle of the subject is equal to or higher than the wrinkle reference value (Yes), the control unit compares the efficacy of the wrinkle-improving composition currently used by the subject and

outputs

1 or 2 or more types of wrinkle-improving compositions recommended to have higher efficacy.

In step 705, when the index value of the current wrinkle of the subject is not equal to or higher than the wrinkle reference value (No) (that is, when the index value is lower than the wrinkle reference value), the control unit outputs the estimated age at which the wrinkle level (3 to 5) set as the optimal wrinkle risk value is reached, based on the wrinkle prediction model a. Outputting 1 or 2 or more wrinkle-improving compositions recommended to be used after the expected age.

< example 2>

In step 801, the control unit inputs 3 values of the measurement value of skin brightness, the measurement value of skin redness, and the measurement value of Pi Fupi fat mass of the subject person to the wrinkle prediction model. This enables the wrinkle prediction formula b for 1 subject to be set.

At step 802, the control unit outputs the current wrinkle improvement index value, and also outputs 1 or 2 or more kinds of wrinkle improvement compositions that can correspond to the current wrinkle level, in addition to the currently used wrinkle improvement agent.

< example 3>

The operator selects a palette panel or a level similar or similar to the skin condition of the subject person from among palette panels or images (images, etc.) of selection levels of the skin condition (measured values of brightness, measured values of redness, measured values of sebum amount) of the subject person on the input screen (see fig. 6). The control unit may obtain measurement value data of each skin condition of the subject person based on personal data such as image data of the subject person. The age value of the subject person may be directly input by the operator or may be read from personal data by the control unit. The control unit can obtain data of each measurement value corresponding to the selected color panel or level (for example, measurement value data associated with data of the color panel).

In step 901, the control unit can input 4 values of a measured value of the brightness, a measured value of the redness, and a measured value of the sebum amount of the skin of the subject to the wrinkle prediction model. This enables the wrinkle prediction formula a to be set for 1 subject.

In step 902, the control unit inputs the age value of the subject person into the wrinkle prediction formula a, and outputs the wrinkle index value of the subject person. Thus, the control unit can output or provide the operator or the subject with the current wrinkle risk (wrinkle level) of the subject.

In step 903, the control unit can output the wrinkle reference value and the age at that time, as in step 703 of < example 1 >. Further, as shown in fig. 6, the wrinkle risk of the subject person may be output as wrinkle evaluation data of the subject person. Examples of the pattern output include the current wrinkle risk and the optimum wrinkle risk of the subject person, and the face image of the subject person in the future and the present face image showing the degree of reduction in the target risk of the subject person.

In step 904, the control unit may determine a single or a plurality of wrinkle-improving compositions associated with or attached to wrinkle index values equal to or more than the index values based on the index values of the wrinkles of the subject, output the determined wrinkle-improving compositions or groups to the output unit as recommended items, and propose or provide the recommended items to the operator or the subject. The control unit may select one or more wrinkle-improving compositions for the operator or the subject from the recommended items output to the input unit, and thereby the control unit may provide the selected wrinkle-improving composition to the subject as a product by sale, distribution, or the like.

Example 3 of the present invention can provide the following configuration.

The present invention can provide a method for providing information on a wrinkle-improving composition suitable for a subject. The method comprises the following steps:

a target person skin wrinkle index value estimation step of estimating an index value of skin wrinkles of the target person based on the age value, the measured value of skin brightness, the measured value of skin redness, and the measured value of sebum amount of the target person; and

and determining a wrinkle-improving composition suitable for the subject based on the estimated skin wrinkle index value of the subject.

Preferably, the data of the wrinkle-improving composition is associated with or associated with data of a skin wrinkle index value that can be correlated with each other, and the control unit can determine the wrinkle-improving composition or the group thereof suitable for the subject by selecting the data of the skin wrinkle index value that can be correlated with each other.

Preferably, the method further comprises a step of outputting, proposing or providing information on the determined single or plurality of wrinkle-improving compositions or groups thereof via an output unit.

More preferably, the estimation step includes the following steps: the method includes selecting a palette panel having the same or similar skin brightness as a subject, selecting measurement value data of skin brightness associated with data of the palette panel, and selecting measurement value data of skin redness associated with the data of the palette panel as the measurement value of skin brightness and/or selecting a palette panel having the same or similar skin redness as the subject, thereby selecting measurement value data of skin redness associated with the data of the palette panel and using the selected measurement value data as the measurement value of skin redness.

4. Evaluation system relating to wrinkle prediction

An embodiment of an evaluation system relating to wrinkle prediction according to the present invention will be described below. In the description of the present embodiment, the description of each configuration, evaluation method, and the like overlapping with the above-described "1." - "3." is appropriately omitted, and the description of "1." - "3." is also applicable to the present embodiment and can be appropriately employed.

In the skin wrinkle evaluation system according to the present invention, the wrinkle prediction model of the subject person can be derived by applying the measured values of the age and skin condition of the subject person to the wrinkle prediction model.

Further, the control section may estimate the skin wrinkle index value of the subject person based on the actual age of the subject person.

On the other hand, by setting the future age of the subject, the control unit estimates the future wrinkle index value. The future age may be an age desired by the subject, or may be a future age preset in the storage unit or the like, such as an actual age of the subject + α years. This enables prediction of the wrinkle state of the target person in the future.

The control unit can determine the start time of use of the wrinkle improvement composition by comparing the wrinkle index value with a set wrinkle reference value in the same wrinkle prediction model.

Preferably, the control unit compares the wrinkle index value with the wrinkle levels "3 to 5" of the set wrinkle reference value, and thereby the control unit can determine the use start timing of the wrinkle-improving composition.

The set wrinkle reference value may be stored in advance, set in a storage unit, or the like.

The control unit can estimate a wrinkle index value from the actual age of the subject using the wrinkle prediction model. On the other hand, the control unit may estimate the age of the subject when the target person becomes a set wrinkle reference value, for example, "3 to 5", using the same wrinkle prediction model, and determine the age estimated from the reference value as the use start timing of the wrinkle-improving composition.

When the wrinkle index value estimated by the wrinkle prediction model reaches a preset wrinkle reference value, the control unit may determine the age at which the value reaches the preset wrinkle reference value as the use start time of the wrinkle-improving composition.

The measurement value is preferably a measurement value by a skin condition measurement device such as a measurement device or an imaging device (e.g., a digital camera, a video image, a still image). This can reduce variations in measurement values by the measurer.

The wrinkle evaluation system of the present invention preferably includes:

determining that the wrinkle index value estimated by the prediction model reaches a preset wrinkle reference value; and

the age at which the composition reaches is determined as the start of use of the wrinkle-improving composition.

More preferably, the preset wrinkle reference value is a standard wrinkle reference value of the subject at the same age +5 years old.

5. Evaluation program relating to wrinkle prediction

An embodiment of an evaluation system relating to wrinkle prediction according to the present invention will be described below. In the description of the present embodiment, the description of each configuration, evaluation method, and the like overlapping with the above-described "1." - "4." is appropriately omitted, and the description of "1." - "4." is also applicable to the present embodiment and can be appropriately employed.

The skin wrinkle evaluation program was implemented by a computer with the following functions:

a function of inputting an age value and a measured value of a skin condition of a subject; and

and a function of applying the age value and the measured value of the skin state to a wrinkle prediction model to estimate an index value of skin wrinkles of the subject person.

a function of inputting an age value and a measured value of a skin condition of a subject;

a function of applying the age value and the measured value of the skin condition to a wrinkle prediction model to estimate an index value of skin wrinkles of the subject;

by comparing the estimated wrinkle index value with a preset wrinkle reference value,

when the estimated evaluation value of wrinkles reaches a preset wrinkle reference value, the age at the time of arrival is used as a function of the start time of use of the wrinkle-improving composition.

6. Learning model (Learned model) related to the invention

The following description explains an embodiment of a learning model according to the present invention. In the description of the present embodiment, the description of each configuration, evaluation method, and the like overlapping with the above-described "1." - "5." is appropriately omitted, but the description of "1." - "5." is also applicable to the present embodiment and can be appropriately adopted. In the present embodiment, the skin wrinkle evaluation and wrinkle prediction models (for example, wrinkle prediction equations 1 to 7) described above can be used as appropriate.

The wrinkle prediction model used in the present invention may be a wrinkle prediction model generated by a method of generating a learning model, and the explanation of "1-4-2. Method of generating wrinkle prediction model" described above is preferably used.

An example of the processing procedure of specialized AI used as the learning model in the present invention can be briefly expressed as "(1) learning data → (2) algorithm → (3) learning model" → (4) input data → (3) learning model → (5) fruit formation ", but is not limited thereto (for example, see fig. 7). The specialized AI is configured as a framework for acquiring a result by applying arbitrary input data to a learning model constructed by integrating learning data (teacher data) into an algorithm functioning as a learning program.

In the present invention, the subject in the learning data (teacher data) may include the participant in the above-mentioned "1-4-2. Wrinkle prediction model creation method".

The control unit acquires at least 2 values of the age value and the skin brightness measurement value of the subject person as learning data (teacher data), and acquires data of more than one person. Preferably, at least 2 values of the age value and the skin brightness measurement value of the same subject are set as 1 data set. As the teacher data, for example, the above-mentioned "prediction factor of the 1-4-1 wrinkle prediction model" can be used, and 2 or 3 or more kinds selected from age, skin brightness, skin redness, skin fat amount, interaction items of skin fat amount and skin redness, and the like are preferable.

More preferably, the control unit further acquires 1 or 2 of the 2 values and the measured value of the degree of redness of the skin and/or the measured value of the amount of skin fat as data. The control unit may read out from the storage unit, teacher data selected from 2 or more of these data. The storage unit stores in advance age values and measurement values acquired from a plurality of subjects as data.

Next, the control unit may construct a wrinkle prediction model (learning model) by incorporating the learning data read from the storage unit into a preset algorithm. Thus, the control unit has a configuration including a wrinkle prediction model. As illustrated in "1-4. Wrinkle prediction models", learning models (wrinkle prediction models) can be derived using "statistical analysis".

The algorithm may function as a machine learning algorithm, for example. The type of the mechanical learning algorithm is not particularly limited, and may be an algorithm using a Neural Network such as RNN (recurrent Neural Network), CNN (Convolutional Neural Network), or MLP (multi layer Perceptron), or may be any algorithm.

Next, the control unit can generate data ((5) result (output layer)) relating to wrinkle evaluation for display output by inputting input data ((4) input data (input layer)) from the operator into the constructed wrinkle prediction model (learning model).

The learning model may be, for example, a learning model generated by deep learning (deep learning). For example, the learning model may be a multi-layer Neural Network, such as a Deep Neural Network (DNN), and more specifically, a Convolutional Neural Network (CNN). As the learning model, a multilayer neural network may be used, which may have an input layer for inputting an age value and a measurement value of a subject person or the like, an output layer for outputting wrinkle evaluation results of the subject person, and at least 1 intermediate layer provided between the input layer and the output layer.

The control unit can implement a method of generating a learning model, a method of evaluating skin wrinkles of a subject using the learning model, or a method of providing skin wrinkles for evaluation, as described below.

The learning model generation method comprises the following steps: (a) Acquiring teacher data including at least 2 values of age values and skin brightness measurement values of a plurality of subjects as data; (b) Inputting at least 2 values of an age value and a skin brightness measurement value of each subject person using the teacher data; (c) A learning model for outputting a wrinkle prediction model for estimating an index value of skin wrinkles of a subject person is generated based on the input data.

A method in which (d) at least 2 values of an age value and a measured value of skin brightness of a subject are acquired as data; (e) Applying at least 2 values of the age value and the measured value of skin brightness of the subject person to a learning model generated by using at least 2 values of the age value and the measured value of skin brightness of the subject person by the learning model generation method to generate an index value of an estimated skin wrinkle of the subject person, and (f) providing the generated index value as an evaluation of the skin wrinkle or as an evaluation of the skin wrinkle.

A program for causing a computer to execute the method for evaluating skin wrinkles or the method provided for evaluating skin wrinkles, or a recording medium storing the program. A skin wrinkle evaluation device or a skin wrinkle evaluation system containing the program or the recording medium.

Note that the present technology may also adopt the following configuration.

[1 ] A method for evaluating skin wrinkles or a method for providing an evaluation of the skin wrinkles, wherein an index value of skin wrinkles of a subject is estimated based on at least 2 values of an age value and a measured value of skin brightness of the subject.

The method described in the above [1 ] is preferably used by 1 or 2 or more selected from an operator, a provider providing an evaluation result to a third party, a target person, an end user, a commodity purchaser, an operator (a counselor, an operator, a commodity seller, etc.), and the like. The method described in the above [1 ] may be a method for assisting skin wrinkle evaluation including presenting, displaying, or providing a result of estimating an index value of the wrinkle. The method described in said [1 ] can be executed by a computer.

[ 2 ] the evaluation method or the method of providing an evaluation according to [1 ], wherein the index value of skin wrinkles of the subject is estimated based on the 2 values and 1 or 2 values of the measured value of the degree of redness of the skin and/or the measured value of the amount of skin fat.

[ 3 ] the evaluation method or the method of providing an evaluation according to [1 ], wherein the index value of skin wrinkles of the subject is estimated by applying 2 values of the age value and the measured value of skin brightness of the subject to the wrinkle prediction model.

[ 4 ] the evaluation method or the method of providing an evaluation according to [ 2 ] above, wherein the index value of skin wrinkles of the subject is estimated by applying 2 values, which are the measured values of the age and skin brightness of the subject, and 1 or 2 values, which are the measured values of the skin redness and/or the measured values of the skin fat mass, to the wrinkle prediction model.

[ 5 ] the evaluation method or the method for providing an evaluation as described in [ 3 ] or [ 4 ], wherein the wrinkle prediction model is derived by using a linear mixture effect model in which an index value of wrinkles of a person is used as a target variable and an explanatory variable includes at least a measured value of age and skin brightness of the person.

[ 6 ] the evaluation method or the provision method of evaluation as described in [ 5 ], wherein the parameters of the linear mixture effect model are obtained using a maximum likelihood method or a constrained maximum likelihood method.

[ 7 ] A method for evaluating or searching for a wrinkle-improving composition, or a method for providing an evaluation result or a search result, which comprises:

a step of estimating an index value of skin wrinkles of a subject to which a test substance has been applied, using the wrinkle evaluation method according to any one of [1 ] to [ 6 ]; and

and determining the test substance as a wrinkle-improving substance when the estimated skin wrinkle index value is lower than the skin wrinkle index value of the subject before application.

A method for providing information on a wrinkle-improving composition suitable for a subject, which comprises:

estimating an index value of skin wrinkles of the subject based on at least 2 values of the age value and the measured value of skin brightness of the subject; and

and determining a wrinkle-improving composition based on the estimated index value of the skin wrinkles of the subject. The estimation step may estimate the wrinkle by using the wrinkle evaluation method according to any one of [1 ] to [ 6 ].

[ 9 ] A skin wrinkle evaluation device that causes a computer to execute the skin wrinkle evaluation method or the method for providing skin wrinkle evaluation according to any one of [1 ] to [ 6 ]. The wrinkle evaluation device preferably includes (a) a skin wrinkle evaluation program and/or (b) a recording medium on which the program is recorded or stored, configured to execute the method according to any one of [1 ] to [ 6 ]. Preferably, the wrinkle evaluation device accesses the program or the medium existing outside the device to execute the method according to any one of [1 ] to [ 6 ]. The recording medium may be a computer recording medium having commands executable by a computer.

[10] A method for evaluating skin wrinkles or a method for providing skin wrinkle evaluation, wherein (a) a computer is caused to make a wrinkle prediction model using a learning model; (b) The index value of skin wrinkles of the subject person is estimated by applying at least 2 values of the age value and the measured value of skin brightness of the subject person to the obtained wrinkle prediction model. The wrinkle prediction model is preferably any one of the above-mentioned models [ 3 ] to [ 6 ].

11 a learning model generation method, wherein,

acquiring teacher data including at least 2 values of age values and skin brightness measurement values of a plurality of subjects as data;

a learning model is generated using the teacher data, which outputs a wrinkle prediction model for estimating an index value of skin wrinkles of a subject by inputting at least 2 values of an age value and a measured value of skin brightness of each subject. The data preferably further contains 1 or 2 values of the measured value of the redness of the skin and/or the measured value of the amount of skin fat.

[ 12 ] A method for evaluating skin wrinkles or a method for providing an evaluation of the skin wrinkles, wherein,

at least 2 values of the age value and the skin brightness measurement value of the subject person are acquired as data,

in a learning model which uses teacher data to learn, and outputs a wrinkle prediction model for inputting at least 2 values of an age value and a skin brightness measurement value of each subject person and estimating an index value of skin wrinkles of the subject person,

an index value of skin wrinkles of the subject person is estimated based on at least 2 values of the age value and the measured value of skin brightness of the subject person. The data preferably further contains 1 or 2 values of the measured value of the redness of the skin and/or the measured value of the amount of skin fat.

[ 13 ] A program or a recording medium storing the program, which executes the method for evaluating skin wrinkles or the method for providing an evaluation of skin wrinkles described in [ 11 ] by a computer. The program, and a skin wrinkle evaluation device or a skin wrinkle evaluation system including the recording medium.

Examples

The present invention will be described in further detail below based on test examples and the like. The test examples and the like described below are representative examples of the present invention, and the scope of the present invention should not be construed narrowly.

< test example 1: wrinkle prediction model >

Test example 1-1 test participants

In the high-silk institute of japan, a total of 48 japanese women who participated in the wrinkle and skin condition evaluation study performed every year from 2011 to 2013 were enrolled in the study. All subjects signed a consent specification made by the high silk study based on helsinki treaty.

The study was a time-repeated study, and evaluation of wrinkles and skin characteristics of participants was continuously performed every year from 2011 to 2017. The age of the participants (women) was 22 to 60 years throughout the survey. The age at the end of 2017 is 20 to 29 years: 6 persons; 30 to 39 years old: 15 persons; age 40 to 49 years: 13 persons; age 50 to 59: 13 persons; 60 to 69 years old: 1 person. The number of participants participating in the test was 6 times or more and 3 times or more at the maximum, and the number of subjects participating in the survey was 5 times or more at the maximum.

Test examples 1-2 evaluation of wrinkles and evaluation of skin Condition by Each apparatus

After washing the face with the same facial cleanser, the panelists were acclimatized in an environment controlled to 20 to 22 ℃ and 50 ± 5% for 30 minutes, and then subjected to all the following evaluations (wrinkle, water content of stratum corneum, amount of transepidermal water transpiration, amount of sebum, skin color, and the like). For each evaluation other than wrinkle evaluation, measurement was performed using a general measurement apparatus and a measurement site as described below.

The wrinkle state of the participants was evaluated visually by a professional evaluator (Trained expert) according to the guidelines published by the Japan Society for cosmetics industry (Journal of Japan Cosmetic Science Society, "the Committee for the research and study of Cosmetic function evaluation methods report the guidelines for the evaluation of Cosmetic function", vol.30, no.4, pp.316-332 (2006)). The wrinkle state of the right canthus was evaluated as a wrinkle grade in units of 0.25 dots up to "0,1,2,3,4,5,6,7" according to the wrinkle grade standard table. In the wrinkle level standard tables, for example, "no wrinkles" is observed at a level 0, and "clearly deep wrinkles are observed" at a level 7.

The water content of the horny layer of the participants was measured by measuring the capacitance of the horny layer in the upper part of the right cheek of the participants using SKICON-200EX (manufactured by IBS Co., ltd.).

The right cheek area was measured by a Vapometer (Delfin technologies, inc.) for transepidermal water evaporation.

Measurement of sebum amount of participants the center of forehead was measured using a Sebumeter (manufactured by Courage + Khazaka electronic GmbH).

The skin color of the panelists was measured for L (lightness index), a (redness index) and b (yellowness index) on the upper portion of the right cheek of the panelists using a Spectrophotometer CM-700d (manufactured by KONICA MINOLTA).

Test examples 1 to 3 statistical analysis for developing wrinkle prediction model

Using the data obtained in "test example 1-1." test example 1-2. "the following wrinkle prediction models (formulas 1 to 7) were obtained.

In order to verify a model for predicting the state of wrinkles due to aging, which is a main object of the present study, the distribution of the obtained wrinkle grade and the relationship between age and wrinkle grade were confirmed.

Next, as a candidate for the wrinkle prediction factor, the average value of all measurement points of the device for each evaluation is added to the wrinkle prediction factor as an important characteristic value representing the skin condition of the individual, except for the age and the skin condition measurement values at all times acquired by the device evaluation.

In the prediction, since the number of data is small, the data is not divided into verification data and test data, and all data is used. Furthermore, the predictive model uses a general linear mixed-effect model that can represent unpredictable individual differences in repeated data over time as variable effects.

[ number 1 ]

Multivariate linear mixed effect model

Reaction Y of the ith participant at time j _ij And definition of

Y _i ＝X _i β+Z _i b _i +ε _i ，b _i ～MVN(0，G)，ε _i ～MyN(0，R _i )

The wrinkle prediction factor is selected by adopting a stepwise regression method based on a backward elimination method (algorithm), and a model with the minimum AIC (Akaike Information Criterion) is determined as an optimal model. (Algorithm: mallows (1973) techniques 15, 661-675.; steyerberg (2009) Statistics In Medicine 19 (8), 1059-1079.; akaike (1973) In 2nd International Symposium on Information Theory and an Extension of the Maximum Likeliod principle, B.N.Petrov, and F.Csaki (eds), 267-281.). The selection of the wrinkle predictor was sequentially examined from the use of a complete model of all wrinkle predictors.

Finally, sensitivity analysis is carried out by using the wrinkle predictor. All analyses were performed using R ver.3.5.2Statistical software (R Foundation for Statistical Computing, vienna, austria).

Test examples 1 to 4 results and examination

[ test examples 1-4.1 participants ]

The mean age of the participants in the test period was shifted from 38.64 years (95% ci.

Next, with respect to the temporal change in wrinkle level of each person, it was observed that such a strong linear relationship that the wrinkle level increases with age and that there is a large individual difference in the average value of wrinkles of each person. Therefore, as the wrinkle prediction model in the present study, a linear mixed effect model that introduces a variable effect in the z slice was selected.

[ test examples 1-4.2 development of wrinkle prediction model ]

As a final wrinkle prediction model, an average L value (skin brightness) of all the measurement years of each person including 5 wrinkle prediction factors (1) "age", 2) "sebum amount (skin sebum amount)", 3) "a value (skin redness)", 4) "and an interaction term of 5)" sebum amount and a value ") were estimated (table 1).

To verify the accuracy of the wrinkle prediction model of the present invention, R, which represents the accuracy of the prediction of the wrinkle level described by the wrinkle prediction factor, is calculated ² (%) and the result (R) obtained by RMSE representing the difference between the predicted value and the observed value ² =84.75,rmse = 0.685), from which it was found that the wrinkle prediction model showed good prediction accuracy.

[ TABLE 1 ]

Table 1: final generalized linear mixture model in dataset

The result of the model is a wrinkle rating

Natural logarithmic transformation

Average skin color over all measurement periods

Test example 1-4.3 wrinkle prediction model (formula 1)

Further, the following wrinkle prediction model (equation 1) using the 5 wrinkle prediction factors derived as described above and the linear mixed effect model is shown.

Wrinkle rating _i =0.1469 × age

+0.7540 XLn (sebum)

+0.3270 × skin color a

+0.1654 × mean value [ skin color L ]

0.1044X (Ln (sebum) x skin color a)

-15.90

+b _0，i

(formula 1)

b _0，i ～N(0，0.4847)

Final equation for predicting wrinkle grade in Japanese women aged 22-60 years

Considering the above test example 1 and the following test examples 2 to 5, it is important that at least 2 values of the measured values of age and skin brightness among the 5 wrinkle prediction factors are measured in the wrinkle prediction model of the present invention. Therefore, it is considered that the target person who wants to evaluate skin wrinkles can estimate the index value of wrinkles of the target person with good prediction accuracy based on these 2 values.

Generally, the wrinkle grade becomes high together with age, and in order to prevent wrinkles, a moisturizing measure may be taken. Therefore, it has been considered that age and measurement values related to moisture retention (for example, skin moisture content and transepidermal water transpiration) are the most important values. However, in the wrinkle prediction model of the present study, there may be no measured values of the moisture content of the skin and the transepidermal water transpiration amount, and it is totally unexpected for the present inventors that age and skin brightness are the most important values. By accumulating data of age and values indicating skin conditions (evaluation and measurement values of skin, etc.) for each individual over a long period of time and using multivariate analysis (particularly, a linear mixed effect model), unprecedented results and ideas have been successfully derived.

It was also confirmed that the prediction accuracy of the wrinkle prediction model composed of 5 wrinkle prediction factors, i.e., age, skin brightness, skin redness, skin sebum amount, and interaction between the skin sebum amount and the skin redness, was particularly good.

For example, when the "age value" of the subject (female) was 35 years, "the measured value of skin lightness" was 68, the measured value of skin redness was 8.3, and the measured value of Pi Fupi fat mass was 6.0, the "measured value of skin lightness" was substituted for the "age" of the above (formula 1), "the" measured value of skin redness "was substituted for the" skin color L "," the "measured value of skin redness" was substituted for the "skin color a", and "the measured value of skin fat mass" was substituted for the "sebum", the skin wrinkle grade was 3.0.

In addition, the present study also yielded formula 1 of the wrinkle prediction model and formulae 2 to 7 below. Conventionally, wrinkle evaluation (wrinkle grade) has been based on visual observation by a human, and therefore has required an experienced specialist, which has not been easy. However, in the present technology, the wrinkle level can be more easily and objectively quantified. Furthermore, the measured value of the skin condition used in the present study can be more easily obtained by a skin condition measurement device such as a measurement instrument. Therefore, the present invention does not require an experienced specialist and can more easily and accurately obtain the numerical value of the wrinkle level. In addition, the invention utilizes a smart phone. Since the wrinkle level can be digitized even in the camera function or the camera function of a personal computer with a Web camera, the present invention can objectively and accurately evaluate wrinkles at any time and place with a simpler operation.

Furthermore, by executing the wrinkle prediction model of the present invention with a computer including a CPU, each person can more easily and accurately evaluate wrinkles using a mobile terminal or the like.

Therefore, the present invention can provide a technique that can accurately and more easily evaluate skin wrinkles for each person.

< test example 2: wrinkle prediction model (equation 2) >

Using the data obtained in test example 1, when each slice in the wrinkle prediction model (formula 1) is in the following range, the accuracy of wrinkle prediction on an individual level is good when it is 60% or more. Therefore, the following equation 2 can be used as the wrinkle prediction model (equation 2).

Therefore, the present invention can accurately and more easily estimate the index value of skin wrinkles of the subject person based on 4 values of the age value, the measurement value of skin brightness, the measurement value of skin redness, and the measurement value of Pi Fupi fat mass by using the wrinkle prediction formula 1 or the wrinkle prediction formula 2.

Wrinkle rating _i = 0.1200-0.2053 × age

+ 0.2200-1.280 XLn (sebum)

+0.050 to 0.5710 × skin color a

+ 0.050-0.4000 × mean value [ skin color L ]

-0.030 to 0.1800 × [ Ln (sebum). Times.skin color a ]

－25.00～7.600

(formula 2)

< test example 3: number of wrinkle-predicting factors >

Using the data obtained in test example 1, equipment evaluation of skin conditions was performed, except for the values used for the wrinkle predicting factors (age, sebum amount, skin redness, skin brightness). For example, the moisture content of the skin, the amount of transepidermal water transpiration, the degree of skin yellowness, and the like were evaluated by a facility. However, the wrinkle level predicted by using all the measured values as the prediction factors is larger than the best model because the prediction accuracy is about 60% or the RMSE value is about 1.5, and thus no good prediction accuracy is exhibited. Therefore, it is considered that when other wrinkle prediction factors (for example, the moisture content of the skin, the amount of transepidermal water transpiration, etc.) are further added to the above 5 wrinkle prediction factors (the interactive terms of age, sebum amount, redness of the skin, skin brightness, sebum amount, and a value), the prediction accuracy is lowered. Therefore, the wrinkle prediction model using the above 5 wrinkle prediction factors is considered to be the optimal model.

< test example 4: the above-described 2 wrinkle prediction factors (age and skin lightness) >

The "age" and "skin brightness" of the 5 wrinkle predictors were selected, and a wrinkle prediction model (formula 3) using the 2 wrinkle predictors is shown below. The accuracy of wrinkle prediction at the individual level was 80% or more, which was good. The wrinkle prediction model and the wrinkle prediction accuracy used R described in "test examples 1 to 3, statistical analysis for developing wrinkle prediction model" and "test examples 1 to 4.2 development of wrinkle prediction model" above ² 。

Therefore, the present invention can accurately and more easily estimate the index value of skin wrinkles of the subject person based on 2 values of the age value and the measured value of skin brightness by using the wrinkle prediction formula 3.

Wrinkle rating _i =0.1484 × age

+0.1844 × mean value [ skin color L ]

－14.80

(formula 3)

< test example 5: the case of 2 wrinkle prediction factors + wrinkle prediction factor (sebum amount) described above >

Using the data obtained in test example 1, "age" and "skin brightness" among the 5 wrinkle predictors described above, and "sebum amount" were selected, and a wrinkle prediction model (formula 4) using the 3 wrinkle predictors is shown below. The accuracy of wrinkle prediction at the individual level was 80% or more, which was good. The wrinkle prediction model and the wrinkle prediction accuracy used R described in "test examples 1 to 3, statistical analysis for developing wrinkle prediction model" and "test examples 1 to 4.2 development of wrinkle prediction model" above ² 。

Therefore, the present invention can accurately and more easily estimate the index value of skin wrinkles of the subject person based on 3 values of the age value, the measured value of skin brightness, and the measured value of skin fat mass by using the wrinkle prediction formula 4.

Wrinkle rating _i =0.1504 × age

+0.1880 × mean value [ skin color L ]

+0.044 XLn (sebum)

－15.28

(formula 4)

< test example 6: the 2 wrinkle prediction factors + wrinkle prediction factors (redness, sebum amount) described above >

Using the data obtained in test example 1, "age" and "skin brightness" among the 5 wrinkle prediction factors described above, and "redness" and "sebum amount" were selected, and a wrinkle prediction model (formula 5) using these 4 wrinkle prediction factors is shown below. The accuracy of wrinkle prediction at the individual level was 80% or more, which was good. The wrinkle prediction model and the wrinkle prediction accuracy used R described in "test examples 1 to 3, statistical analysis for developing wrinkle prediction model" and "test examples 1 to 4.2 development of wrinkle prediction model" above ² 。

Therefore, the present invention can accurately and more easily estimate the index value of skin wrinkles of the subject person based on 4 values of the age value, the measured value of skin brightness, and the measured values of redness and sebum amount by using wrinkle prediction formula 5.

Wrinkle rating _i =0.1460 × age

+0.035 XLn (sebum)

+0.013 × skin color a

+0.1812 × mean value [ skin color L ]

(formula 5)

Claims

1. A method for evaluating skin wrinkles, which estimates an index value of skin wrinkles of a subject person based on at least 2 values of an age value and a measured value of skin brightness of the subject person.

2. The method for evaluating skin wrinkles according to claim 1, wherein the index value of skin wrinkles of the subject is estimated based on the 2 values and 1 or 2 values of the measured value of the degree of redness of the skin and/or the measured value of the amount of skin fat.

3. The method for evaluating skin wrinkles according to claim 1, wherein the index value of skin wrinkles of the subject is estimated by applying 2 values, which are an age value and a skin brightness measurement value of the subject, to the wrinkle prediction model.

4. The method for evaluating skin wrinkles according to claim 2, wherein the index value of skin wrinkles of the subject is estimated by applying 2 values of the measured value of the age and skin brightness of the subject and 1 or 2 values of the measured value of the skin redness and/or the measured value of the skin sebum amount to the wrinkle prediction model.

5. The method for evaluating skin wrinkles according to claim 3 or 4, wherein the wrinkle prediction model is derived by using a linear mixed effect model in which an index value of a wrinkle of a person is used as a target variable and an explanatory variable includes at least a measurement value of age and skin brightness of the person.

6. The method for evaluating skin wrinkles according to claim 5, wherein the parameters of the linear mixed effect model are obtained using a maximum likelihood method or a limited maximum likelihood method.

7. A method for evaluating or searching for a wrinkle-improving composition, comprising:

a step of estimating an index value of skin wrinkles of a subject to which a test substance has been applied, using the method for evaluating skin wrinkles according to any one of claims 1 to 6; and

and a step of judging the test substance as a wrinkle-improving substance when the estimated skin wrinkle index value is lower than the skin wrinkle index value of the subject before application.