KR102367491B1 - An hair loss management device and a method of providing hair loss management guidelines thereof - Google Patents

Abstract

The present disclosure is to provide a hair loss management device that provides a management index for slowing hair loss according to a hair loss condition, and a method for providing a hair loss management guideline therefor, a measuring unit for diagnosing a current user condition, a predetermined value according to the current user condition It may include a prediction unit for calculating the predicted hair loss state, which is the hair loss state after the elapse of the period, and a guide unit for outputting a management index for delaying the progress to the predicted hair loss state.

Description

A hair loss management device and a method for providing a hair loss management guideline thereof

The present disclosure relates to a hair loss management device and a method for providing hair loss management guidelines for the hair loss management device.

Hair loss refers to a state in which there is no hair in a part where hair should normally exist.

Hair loss generally progresses with aging, and causes include heredity and stress. Specifically, it is known that the cause of hair loss is heredity as the biggest reason, but recently environmental reasons such as frequent hair treatment, stress, environmental pollution, scalp health, etc. continue to be the main cause of hair loss. Interest is growing.

Methods for diagnosing hair loss include the Savin scale and the Norwood Hamilton scale. The Sabine scale is a method of diagnosing the progress of hair loss more precisely based on the image by dividing the parietal region into eight stages. When diagnosing hair loss according to the Sabine scale, it may be difficult to accurately diagnose hair loss because the subjective opinion of the person diagnosing hair loss is involved. In particular, when self-diagnosing a hair loss condition, it is difficult to make an objective judgment on your own, and most users do not know what kind of action is required depending on the hair loss condition and how to respond.

In addition, although users think that their scalp condition changes according to the season, and care should be taken according to the changed scalp condition, it is difficult to understand the exact scalp/hair condition. In addition, although studies have been conducted on changes in scalp/hair conditions according to aging, gender, and race, research data on the degree of hair loss according to scalp/hair conditions are insufficient.

The present disclosure aims to provide a hair loss management device that diagnoses a hair loss condition according to a scalp/hair condition, and provides a management index for slowing hair loss according to the hair loss condition, and a method for providing a hair loss management guideline thereof.

An object of the present disclosure is to provide a hair loss management device for constructing a decision tree capable of diagnosing a hair loss condition by acquiring scalp/hair related variables affecting the hair loss condition, and a method for providing a hair loss management guideline therefor.

An object of the present disclosure is to provide a hair loss management device for re-learning a decision tree through the scalp/hair condition at the time when a predetermined time has elapsed, and a method for providing a hair loss management guideline thereof.

An object of the present disclosure is to provide a hair loss management device that recommends a product for delaying the progression to a predicted hair loss condition according to the hair loss condition, and a method for providing a hair loss management guideline therefor.

A hair loss management apparatus according to an embodiment of the present disclosure includes a measuring unit for diagnosing a current user state, a predicting unit for calculating a predicted hair loss state that is a hair loss state after a predetermined period has elapsed according to the current user state, and a predictive hair loss state It may include a guide unit for outputting a management index for slowing the progress.

The prediction unit may calculate the predicted hair loss state through a decision tree in which the explanatory variable is at least one of age, hair thickness, scalp moisture, and scalp pH, and the dependent variable is at least one of part area and hair loss grade.

A hair loss management device according to an embodiment of the present disclosure generates a decision tree for generating a decision tree using at least one of the subject's age, hair thickness, scalp moisture, and scalp pH, and at least one of the subject's parting area and hair loss grade It may include more wealth.

The decision tree generating unit may generate the decision tree by comparing the prediction rates of the parting area according to each of a plurality of combinations including at least one of age, hair thickness, scalp moisture, and scalp pH.

The hair loss management apparatus according to an embodiment of the present disclosure may further include a decision tree re-learning unit for re-learning the decision tree generated by the decision tree generating unit.

The decision tree re-learning unit at the first time point, when the correlation between at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area follows the decision tree, a predetermined time from the first time point Decision making is made by comparing the predicted parting area output as the subject's age, hair thickness, scalp moisture, or scalp pH measured at the second time point is input to the decision tree with the subject's parting area measured at the second time point Trees can be relearned.

The hair loss management apparatus according to an embodiment of the present disclosure may further include a display unit for displaying a parting area according to the predicted hair loss state.

The management index may include at least one of hair thickness, scalp moisture, and scalp pH required to slow progression to a predicted hair loss condition.

The hair loss management apparatus according to an embodiment of the present disclosure may further include a product recommendation unit for recommending a hair-related product suitable for reaching at least one of a required hair thickness, scalp moisture, and scalp pH.

The hair loss condition may include a hair loss grade determined by the area of the parting.

The hair loss management apparatus according to an embodiment of the present disclosure further includes a counseling unit that calculates a predicted hair loss state when the current user state is managed with an index including at least one of hair thickness, scalp moisture, and scalp pH input by the user. may include

The guide unit may further output a management index for managing the predicted hair loss state calculated by the counseling unit.

The method of providing a hair loss management guideline for a hair loss management device according to an embodiment of the present disclosure includes the steps of diagnosing a current user condition, calculating a predicted hair loss condition that is a hair loss condition after a predetermined period has elapsed according to the current user condition, and prediction It may include outputting a management index for slowing the progression to the hair loss state.

The step of calculating the predicted hair loss state is a step of calculating the predicted hair loss status through a decision tree in which the explanatory variable is at least one of age, hair thickness, scalp moisture, and scalp pH, and the dependent variable is at least one of part area and hair loss grade may include.

The method for providing guidelines for hair loss management of the hair loss management device according to an embodiment of the present disclosure further includes generating a decision tree using at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area. can do.

The step of generating a decision tree may include generating a decision tree by comparing the prediction rates of a plurality of combinations consisting of at least one of age, hair thickness, scalp moisture, and scalp pH, and parting areas according to each of the plurality of combinations. can

The method of providing a hair loss management guideline for a hair loss management device according to an embodiment of the present disclosure may further include re-learning a decision tree.

The step of re-learning the decision tree is performed from the first time point when the correlation between at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area follows the decision tree at the first time point. At a second time point when a predetermined time has elapsed, measuring at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and a parting area, the subject's age, hair thickness, scalp moisture, and scalp pH measured at the second time point Comprising the step of comparing the predicted parting width output as at least one of the input to the decision tree with the parting width of the subject measured at the second time point, and adjusting the correlation according to the decision tree according to the comparison result can do.

The method of providing a hair loss management guideline for a hair loss management apparatus according to an embodiment of the present disclosure may further include displaying a parting area according to the predicted hair loss state.

Outputting the management index may include outputting at least one of a hair thickness, scalp moisture, and scalp pH required to delay the progression to the predicted hair loss state.

The method of providing a hair loss management guideline for a hair loss management device according to an embodiment of the present disclosure may further include recommending a hair-related product suitable for reaching at least one of a required hair thickness, scalp moisture, and scalp pH.

Diagnosing the current user condition may include measuring at least one of the user's age, hair thickness, scalp moisture, and scalp pH.

The method of providing a hair loss management guideline for a hair loss management device according to an embodiment of the present disclosure includes receiving an input of at least one index of hair thickness, scalp moisture, and scalp pH, and a case in which the current user state is managed by the input index. The method may further include calculating a predicted hair loss state.

The method of providing a hair loss management guideline for a hair loss management device according to an embodiment of the present disclosure may further include outputting a management index for managing the predicted hair loss state when managed with the input index.

According to the present disclosure, since the hair loss management device outputs a management index for delaying the hair loss condition predicted according to the current user condition, the user can more easily understand and manage the method for delaying the hair loss condition.

In particular, when the hair loss management device outputs the recommended product together with the management index, there is an advantage in that it is possible to provide the user with the ease of management according to the management index.

In addition, since the hair loss management device sets a significant variable related to hair loss and predicts the hair loss state through a decision tree generated according to the set variable, there is an advantage in that the accuracy of the hair loss condition prediction is improved.

In addition, the hair loss management apparatus has the advantage of increasing the accuracy of the decision tree by re-learning the decision tree when the data for the same subject is acquired after a predetermined time has elapsed.

In addition, the hair loss management device has the advantage of being able to build a hair loss severity prediction model (HLSPM) in order to suggest a scalp/hair condition improvement direction suitable for the current user condition.

1 is a control block diagram of a hair loss management apparatus according to an embodiment of the present disclosure.
2 is a flowchart illustrating a method of operating a hair loss management apparatus according to an embodiment of the present disclosure.
3 is a graph analyzing the correlation between the image-based parting area and the sabine scale.
4 is a diagram illustrating a hair loss grade constructed according to an image-based parting area.
5 is a flowchart illustrating step S30 of FIG. 2 .
6 is a graph showing the correlation between the scalp/hair condition and age.
7 is a graph showing the correlation between the scalp/hair condition and the area of a part.
8 is a flowchart illustrating a method of a decision tree generating unit generating a decision tree according to an embodiment of the present disclosure.
9 is a graph showing the prediction rate of a decision tree generated for each combination of variables related to scalp/hair.
10 is an exemplary diagram of a decision tree according to an embodiment of the present disclosure.
11 is an exemplary diagram of a method for displaying a management index by a hair loss management apparatus according to an embodiment of the present disclosure.
12 is an exemplary diagram of a method of displaying, by the hair loss management apparatus, a management index and a recommended product according to an embodiment of the present disclosure.
13 is a control block diagram of a hair loss management apparatus according to another embodiment of the present disclosure.
14 is a flowchart illustrating a first example of a method of operating an apparatus for managing hair loss according to another embodiment of the present disclosure.
15 is an exemplary diagram of a method for displaying a predicted hair loss state by a hair loss management apparatus operating according to the flowchart shown in FIG. 14 .
16 is a flowchart illustrating a second example of a method of operating a hair loss management apparatus according to another embodiment of the present disclosure.
17 is an exemplary diagram of a method for displaying a predicted hair loss state by the hair loss management apparatus operating according to the flowchart shown in FIG. 16 .

Hereinafter, the embodiments disclosed in the present specification will be described in detail with reference to the accompanying drawings, but the same or similar components are assigned the same reference numbers regardless of reference numerals, and redundant description thereof will be omitted. The suffix “module” or “part” for components used in the following description is given or mixed in consideration of only the ease of writing the specification, and does not have a meaning or role distinct from each other by itself.

In addition, in describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed herein, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in the present specification, and the technical spirit disclosed in the present specification is not limited by the accompanying drawings, and all changes included in the spirit and scope of the present disclosure , should be understood to include equivalents or substitutes.

Terms including an ordinal number such as 1st, 2nd, etc. may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

The singular expression includes the plural expression unless the context clearly dictates otherwise.

In the present application, the term "comprising" is intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, and includes one or more other features, number, step, or the like. , it should be understood that it does not preclude in advance the possibility of the presence or addition of an operation, component, part, or combination thereof.

1 is a control block diagram of a hair loss management apparatus according to an embodiment of the present disclosure.

The hair loss management device 1 includes a controller 11, a measuring unit 12, a predicting unit 13, a guide unit 14, a decision tree generating unit 16, a decision tree re-learning unit 17, and a display. It may include at least some or all of the unit 18 and the product recommendation unit 19 .

The controller 11 may control the operation of the hair loss management device 1 . The controller 11 includes a measurement unit 12 , a prediction unit 13 , a guide unit 14 , a decision tree generation unit 16 , a decision tree re-learning unit 17 , a display unit 18 , and a product recommendation Each of the units 19 can be controlled.

The measurement unit 12 may diagnose the current user state. The user state may include at least one of the user's age, hair thickness, scalp moisture, scalp pH, and parting width.

The prediction unit 13 may calculate the predicted hair loss state, which is the hair loss state after the lapse of a predetermined period according to the current user state. The prediction unit 13 may calculate the predicted hair loss state of the user expected when a predetermined time elapses by using the user state of the user measured by the measurement unit 12 .

The prediction unit 13 may calculate the predicted hair loss state through a decision tree in which the explanatory variable is at least one of age, hair thickness, scalp moisture, and scalp pH, and the dependent variable is at least one of parting width and hair loss grade. In this case, the decision tree may be at least one of a decision tree generated by the decision tree generating unit 16 to be described later and a decision tree enrolled by the decision tree re-learning unit 17 .

The guide unit 14 may output a management index for delaying the progress to the predicted hair loss state.

The management index may be information provided so that a user can more easily manage a scalp condition or a hair condition. The management index is required to slow the progression to the predicted hair loss state, and may include, for example, at least one of hair thickness, scalp moisture, and scalp pH.

The decision tree generating unit 16 may generate the decision tree by comparing the prediction rates of the parting width according to each of a plurality of combinations including at least one of age, hair thickness, scalp moisture, and scalp pH.

The decision tree generating unit 16 may generate a decision tree using at least one of the subject's age, hair thickness, scalp moisture, and scalp pH, and at least one of the subject's parting area and hair loss grade.

The decision tree generating unit 16 is composed of a root node, at least one intermediate node, and at least one terminal node, and the root node and the intermediate node are related to hair loss. It is a variable set as a significant variable among the variables, and the end node can create a decision tree with a parting area or a hair loss grade. Here, examples of variables set as significant variables among variables related to hair loss may include age, hair thickness, scalp moisture, and scalp pH.

The decision tree re-learning unit 17 may relearn the decision tree generated by the decision tree generating unit 16 .

The decision tree re-learning unit 17 generates the correlation between at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area at the first time point and the decision tree generating unit 16 In the case of following the decision tree, the age, hair thickness, scalp moisture, or scalp pH of the subject measured at the second time point when a predetermined time has elapsed from the first time point is the decision generated by the decision tree generating unit 16 . The decision tree can be retrained by comparing the predicted parting width output as it is input to the tree with the subject's parting width measured at the second time point.

The display unit 18 may display at least one of a parting area and a hair loss grade according to the predicted hair loss state.

The display unit 18 includes a user state diagnosed by the measurement unit 12, a predicted hair loss state calculated by the prediction unit 13, a management index output by the guide unit 14, and a product recommendation unit 19 to be described later recommended At least one of the hair related products may be further displayed.

The product recommendation unit 19 may output a recommended product so that the scalp condition or hair condition can be managed with the management index output by the guide unit 14 . The product recommendation unit 19 may recommend a hair-related product suitable for reaching the management index output by the guide unit 14 .

2 is a flowchart illustrating a method of operating a hair loss management apparatus according to an embodiment of the present disclosure.

The controller 11 may control the measurement unit 12 to diagnose the current user state (S10).

The measurement unit 12 may acquire at least one of the user's age, hair thickness, scalp moisture, scalp pH, and parting width as the current user state.

The measurement unit 12 may measure a user state under a predetermined condition. For example, the predetermined conditions may include prohibition of use of hair products such as shampoo and conditioner 12 hours before measurement, and a constant temperature and humidity evaluation room (eg, temperature 22±2° C., humidity 50±5%). That is, the measurement unit 12 may measure the state of the user who has not used hair products such as shampoo and conditioner for at least 12 hours in the constant temperature and humidity evaluation room.

The measurement unit 12 may receive the user's age through an input module (not shown). The input module (not shown) may be a physical key button, a touch pad, or the like, but is not limited thereto as this is merely exemplary.

The measuring unit 12 may measure the hair thickness by magnifying the hair 200 times using ASW (Aram HUVIS Co., Seoul, Korea). The measuring unit 12 may recognize the average thickness of the six strands at five locations on the scalp as the hair thickness.

The measurement unit 12 may measure the scalp moisture content (Scalp Hydration) with a DermaLab USB Hydration probe (Cortex Technology, Denmark) at two locations 8 times along the parting, and recognize the average value as the scalp moisture.

The measuring unit 12 may recognize the scalp pH as an average value of two or more points measured along the scalp pH with a Skin-pH-Meter (MPA580 (CK electronic, Germany)).

The measurement unit 12 may acquire the width of the parting through the parting image of the user. The measuring unit 12 may further include a camera (not shown), and after converting the garma image taken with the camera (not shown) into grayscale, it is analyzed with Image Pro Premier 9.2 (Media Cybemetics, Maryland, USA). You can measure the width of the part. The measurement unit 12 may measure the width of a part of a user at six places and measure the width of the part through an image-based hair midline area (IHMA) assessment.

As a specific example, the measurement unit 12 performs a preset condition (eg, aperture F/5, shutter speed 1/40 seconds, ISO 200, automatic white balance, 72 dot-per-inch, Resolution 5472 x 3648), the user's parting image is taken, and at this time, the parting image can be taken so that the tape measure is located next to the user's part for accurate calibration. After processing the photographed image in black and white and fixing the ROI setting to a height of 5 cm, the measurement unit 12 can automatically measure the width of the white part (garma region) by setting the ROI at 6 random places along the part. The measuring unit 12 may measure the width of the parting through the above-described method.

Next, a method of verifying the image-based parting area measurement method will be described with reference to FIGS. 3 and 4 .

3 is a graph analyzing the correlation between the image-based parting area and the Sabine scale, and FIG. 4 is a diagram showing the hair loss grades constructed according to the image-based parting area.

3 and 4 , the average IHMA for Sabine scale I-1 is 2.55 cm ² , the average IHMA for Sabine scale I-2 is 3.28 cm ² , and the average IHMA for Sabine scale I-3 is 3.75 cm ² , the mean IHMA for Sabine scale I-4 may be 4.9 cm ² , the mean IHMA for Sabine scale II-1 may be 6.47 cm ² , and the mean IHMA for Sabine scale II-2 may be 8.76 cm ² . .

The measurement unit 12 may obtain a parting width and a sabine scale according to the user's parting image by using the graph and construction information as shown in FIGS. 3 and 4 .

On the other hand, since the above-described measuring unit 12 measures each of the hair thickness, scalp moisture content, and scalp pH, it is just an example, and therefore it is reasonable that the type of device is not limited. That is, the measuring unit 12 may measure the thickness of the hair, the amount of moisture in the scalp, the pH of the scalp, and the like, through other devices in addition to the above-described devices.

The controller 11 may control the prediction unit 13 to calculate the predicted hair loss state according to the current user state (S30).

The prediction unit 13 may calculate a predicted hair loss state that is a hair loss state after a predetermined period has elapsed according to the current user state, and a decision tree may be used at this time.

The hair loss state may be expressed as a hair loss grade, and the hair loss grade may be determined by a parting area. That is, the hair loss grade may be a hair loss grade according to the Sabine scale constructed through FIGS. 3 and 4, but this is not limited thereto because it is only an example.

Next, with reference to FIG. 5, a method for generating, using, and learning a decision tree of the hair loss management apparatus according to an embodiment of the present disclosure will be described. 5 is a flowchart illustrating step S30 of FIG. 2 .

The decision tree generating unit 16 may obtain a parting width and a significant variable among various variables related to scalp/hair (S31).

6 is a graph showing the correlation between the scalp/hair condition and age, and FIG. 7 is a graph showing the correlation between the scalp/hair condition and the area of the part.

Fig. 6 (a) shows the correlation between the parting width and age, Fig. 6 (b) shows the correlation between hair thickness and age, and Fig. 6 (c) shows the scalp pH and A correlation between age is shown, and in FIG. 6(d), a correlation between scalp moisture and age is shown.

Fig. 7(a) shows the correlation between the parting area and the hair thickness, Fig. 7(b) shows the correlation between the parting area and the scalp pH, and Fig. 7(c) shows the correlation between the parting area and the hair thickness. The correlation between area and scalp moisture is shown.

The decision tree generating unit 16 may set age, hair thickness, scalp pH, and scalp moisture as statistically significant variables related to hair loss through correlation analysis as shown in FIGS. 6 and 7 .

That is, the decision tree generating unit 16 can obtain a statistically significant variable among various variables related to scalp/hair, and calculate a predicted hair loss state such as the future parting area through the variables obtained as significant. You can create a decision tree for

The decision tree generating unit 16 may generate a decision tree with respect to the variable obtained through step S31 and the area of the division (S33).

The decision tree generating unit 16 may generate a decision tree by using at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area.

8 is a flowchart illustrating a method of a decision tree generating unit generating a decision tree according to an embodiment of the present disclosure. FIG. 8 may be a flow chart embodying step S33 of FIG. 5 .

The decision tree generating unit 16 may set the explanatory variable to at least one of age, hair thickness, scalp moisture, and scalp pH, and set the dependent variable to the parting width (S41).

The decision tree generating unit 16 may determine a root node, an intermediate node, and an end node in the direction of increasing purity (S43).

Specifically, the decision tree generating unit 16 may determine a root node, an intermediate node, and an end node in a direction in which purity (homogeneity) increases according to a recursive partitioning method.

For example, it is assumed that the decision tree generating unit 16 determines a root node, a middle node, and an end node based on age, hair thickness, scalp moisture, and scalp pH. The decision tree generating unit 16 sorts the parting width data values in ascending order according to at least one of age, hair thickness, scalp moisture, and scalp pH, and assumes a point between the sorted data values as a branch before branching. Information acquisition can be investigated by calculating the Gini index and the Gini coefficient after branching, and calculating the difference between the Gini coefficient before branching and the Gini coefficient after branching. According to an embodiment, the decision tree generating unit 16 calculates entropy before branching and entropy after branching instead of the Gini coefficient before branching and the Gini coefficient after branching, and by calculating the difference between entropy before branching and entropy after branching Information acquisition can be investigated. That is, the decision tree generating unit 16 may investigate information acquisition by calculating various reference values indicating uncertainty such as the p-value of a chi-square statistic in addition to the Gini coefficient and entropy.

The decision tree generating unit 16 may determine a root node, a middle node, or an end node based on a branch point at which information acquisition is greatest. The control unit 15 determines the root node in the case of the first branching point, determines the end node in the case of a branching point having a purity of 100%, and determines the intermediate node in the case of other branching points that do not correspond to the root node and the end node. can

The decision tree generating unit 16 calculates the uncertainty before branching (eg, Gini coefficient or entropy) and the uncertainty after branching for the number of N1×N2 cases when the number of data is N1 and the number of variables is N2 Thus, it is possible to determine the root node, the middle node, and the terminal node by examining the information acquisition.

After determining the root node, the middle node, and the end node, the decision tree generating unit 16 may combine the end nodes by pruning (S45).

The decision tree generating unit 16 may determine a root node, a middle node, and an end node so that the purity of all end nodes becomes 100%, and then combine the end nodes so that overfitting is minimized. As branch points increase, it is common for the misclassification rate to decrease when new data is added. It can be said that Accordingly, when the number of branching points exceeds a predetermined number, the decision tree generating unit 16 may perform pruning by combining at least two or more end nodes. The controller 15 may perform pruning according to Equation 1 below.

[Equation 1]

CC(T)=Err(T)+α×L(T)

In this case, CC(T) is the cost and complexity of the decision tree (smaller values are the simpler models with fewer errors and fewer end nodes), Err(T) is the misclassification rate for the validation data, and L(T) is the number of end nodes, and α may mean a weight combining Err(T) and L(T), and may be a value in the range of 0.01 to 0.01.

The decision tree generating unit 16 may perform a validity evaluation after combining the end nodes (S47).

The decision tree generating unit 16 may perform feasibility evaluation through cross validation by at least one of a gain chart, a risk chart, and a cost chart.

Specifically, the decision tree generating unit 16 divides the data into k pieces of data based on at least one of a gain chart, a risk chart, and a cost chart, and divides the divided k Among the data, k-1 pieces of data can be used as training data, and one piece of data can be used as test data. The controller 15 may evaluate the validity of the decision tree generated through steps S41, S43, and S45 while changing the training data and the test data.

The decision tree generating unit 16 may construct a decision tree by interpreting the analysis result of the decision tree after feasibility evaluation (S49).

The decision tree generating unit 16 may perform verification on the decision tree generated by the method as shown in FIG. 8 .

9 is a graph showing the prediction rate of a decision tree generated for each combination of variables related to scalp/hair.

Fig. 9 (a) is a graph showing the prediction rate of the parting width of a decision tree after 6 months according to two variables (age, hair thickness), and Fig. 9 (b) is a graph showing three variables (age, hair thickness, It is a graph showing the prediction rate of the parting area after 6 months of the decision tree according to the scalp moisture), and FIG. It is a graph showing the prediction rate of the area, and FIG. 9 (d) is a graph showing the prediction rate of the area of the parting after 6 months of the decision tree according to four variables (age, hair thickness, scalp moisture, scalp pH).

The closer the slope of the trend line shown in the graph to 1, the higher the prediction rate, and the closer the coefficient of determination (R ²⁾ is to 1, the higher the accuracy.

That is, as shown in FIG. 9 , the decision tree generating unit 16 generates a decision tree by variously combining variables related to scalp/hair, and compares the prediction rates of the generated decision trees, so that the prediction rate is the highest. A high decision tree can be created as the final decision tree.

For example, the decision tree generating unit 16 generates a decision tree by comparing the prediction rates of a plurality of combinations consisting of at least one of age, hair thickness, scalp moisture, and scalp pH with the prediction rates of parting widths according to each of the plurality of combinations. can do. In the example of FIG. 9 , a decision tree according to four variables (age, hair thickness, scalp moisture, scalp pH) as shown in (d) may be generated as a final decision tree.

Again, FIG. 5 will be described.

The prediction unit 13 may acquire the predicted hair loss state output when the current user state is input to the decision tree (S35).

In this case, the decision tree may be a decision tree finally generated by the decision tree generating unit 16 or a decision tree relearned by the decision tree re-learning unit 17 .

10 is an exemplary diagram of a decision tree according to an embodiment of the present disclosure.

In the decision tree illustrated in the example of FIG. 10 , a root node may include a plurality of intermediate nodes and a plurality of end nodes. Each of the root node and the plurality of intermediate nodes may include reference information for each of age, hair thickness, scalp moisture, or scalp pH, and the end node may include a range of parting widths.

When each of the age, hair thickness, scalp moisture, and scalp pH diagnosed by the measuring unit 12 is input to the decision tree as shown in FIG. By acquiring one end node corresponding to each of age, hair thickness, scalp moisture, and scalp pH, the range of parting width can be obtained. The prediction unit 13 may acquire parting width information or a hair loss grade corresponding to parting width information through a decision tree.

The decision tree generating unit 16 may learn the decision tree when age, hair thickness, scalp moisture, scalp pH, and parting area are input.

On the other hand, the decision tree re-learning unit 17 may relearn the decision tree by using the variables and the division width measured for the same subject after a predetermined time has elapsed (S37).

For example, the predetermined time may be 6 months, but this is not limited thereto as it is merely an example.

The decision tree re-learning unit 17 at the first time point, when the correlation between at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area follows the decision tree, the first time point At a second time point when a predetermined time has elapsed, at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the area of the parting may be measured. The decision tree re-learning unit 17 measures the predicted segment width output as at least one of the subject's age, hair thickness, scalp moisture, and scalp pH measured at the second time point is input to the decision tree at the second time point It is compared with the divided part width of the tested subject, and the correlation according to the decision tree can be adjusted according to the comparison result.

Specifically, the decision tree re-learning unit 17 learns the decision tree using measurement information of the subject's first part width, first age, first hair thickness, first scalp moisture, and first scalp pH. Then, when a predetermined time has elapsed, the second parting area, the second age, the second hair thickness, the second scalp moisture, and the second scalp pH of the same subject may be measured. The decision tree re-learning unit 17 may acquire the predicted parting area output as the second age, the second hair thickness, the second scalp moisture, and the second scalp pH are input into the decision tree, and the predicted parting width can be compared with the width of the second part to analyze the correlation. The decision tree re-learning unit 17 may relearn the decision tree by adjusting the decision tree according to the difference when there is a difference between the width of the predicted part and the width of the second part.

Below, Table 1 is an example showing measurement information of the first parting area, age, first hair thickness, first scalp moisture, and first scalp pH of the test subject, and Table 2 is the second parting width and second hair of the same test subject It may be an example showing measurement information of the thickness, the second scalp moisture, and the second scalp pH. In the examples of Tables 1 and 2, it is assumed that the age of the subjects remains the same.

Participant Age Initial Measurements Scalp
Capacitance
(microsiemens) Scalp pH Hair Thickness (mm) IHMA (cm ² ) ID1 23 52.0 6.0 0.066 2.74 ID2 38 128.5 6.3 0.060 3.89 ID3 37 59.5 6.3 0.054 3.03 ID4 38 40.5 6.1 0.063 2.99 ID5 47 71.0 6.4 0.056 5.17 ID6 56 101.0 5.5 0.058 5.41 ID7 50 58.0 6.0 0.056 4.45 ID8 41 46.5 5.7 0.056 4.40 ID9 41 118.0 5.9 0.062 3.72 ID10 52 60.0 5.4 0.059 4.14 ID11 37 30.0 6.7 0.057 3.37 ID12 51 49.0 6.1 0.059 3.17 ID13 50 25.5 6.3 0.063 5.85 ID14 41 47.0 6.5 0.061 2.94 ID15 40 130.5 5.6 0.049 5.08 ID16 42 65.0 6.5 0.058 3.45 ID17 57 63.5 6.6 0.047 9.99 ID18 37 55.0 5.3 0.062 3.89 ID19 45 38.0 6.9 0.048 3.42 ID20 38 60.5 5.4 0.060 3.27 ID21 42 76.0 6.6 0.050 5.67 ID22 43 78.5 6.3 0.058 4.08 ID23 57 89.5 5.9 0.062 7.68 ID24 32 118.0 6.1 0.048 3.70 ID25 37 32.5 7.0 0.066 2.73 ID26 61 118.5 5.2 0.051 4.75 ID27 43 34.5 7.6 0.056 4.77 ID28 34 59.0 6.0 0.063 3.67 ID29 40 78.0 6.3 0.052 4.29 ID30 40 54.5 6.5 0.066 2.70 ID31 44 73.5 8.6 0.047 4.64 ID32 40 49.0 5.7 0.066 3.04 ID33 36 60.5 6.3 0.058 3.77 ID34 41 76.0 6.1 0.052 5.62 ID35 58 112.5 5.6 0.038 7.30 ID36 44 80.5 6.0 0.057 4.65 ID37 43 25.0 5.9 0.063 2.92 ID38 35 68.0 6.6 0.055 5.18 ID39 57 70.0 5.8 0.033 4.96 ID40 39 67.5 7.1 0.059 9.52 ID41 53 101.5 6.7 0.061 3.97 ID42 55 58.0 6.7 0.036 7.77

Participant Age 6 Months Later Measurements Scalp
Capacitance
(microsiemens) Scalp pH Hair Thickness (mm) IHMA (cm ² ) ID1 23 49.0 6.4 0.082 2.13 ID2 38 82.5 6.2 0.073 1.94 ID3 37 38.0 6.2 0.070 2.67 ID4 38 42.0 6.4 0.074 1.93 ID5 47 41.0 5.5 0.065 4.13 ID6 56 80.0 5.1 0.063 3.22 ID7 50 81.0 5.5 0.059 4.49 ID8 41 36.0 6.3 0.075 2.95 ID9 41 141.5 5.4 0.070 2.62 ID10 52 63.5 5.5 0.061 4.24 ID11 37 40.5 6.2 0.059 3.86 ID12 51 63.5 6.5 0.076 2.22 ID13 50 24.5 6.3 0.082 3.64 ID14 41 51.0 6.4 0.067 2.58 ID15 40 66.0 6.1 0.072 4.27 ID16 42 46.5 5.9 0.067 3.08 ID17 57 50.0 5.3 0.071 4.77 ID18 37 58.0 5.9 0.055 3.81 ID19 45 42.5 6.2 0.066 3.46 ID20 38 46.5 6.3 0.064 3.47 ID21 42 55.0 5.7 0.067 4.90 ID22 43 74.0 6.4 0.065 4.48 ID23 57 125.5 4.8 0.071 7.04 ID24 32 70.5 6.2 0.074 3.46 ID25 37 36.0 5.8 0.064 1.76 ID26 61 69.5 5.1 0.052 4.58 ID27 43 29.0 6.4 0.079 3.95 ID28 34 47.0 6.2 0.059 3.61 ID29 40 73.5 6.1 0.067 3.71 ID30 40 45.5 6.4 0.078 2.42 ID31 44 52.5 6.3 0.061 4.77 ID32 40 57.0 6.5 0.072 3.00 ID33 36 54.5 6.2 0.064 3.29 ID34 41 50.5 6.2 0.068 5.70 ID35 58 101.0 4.9 0.042 4.54 ID36 44 49.5 5.9 0.077 5.45 ID37 43 51.0 6.4 0.071 2.27 ID38 35 54.0 6.2 0.059 3.03 ID39 57 79.0 4.8 0.051 3.13 ID40 39 57.0 5.6 0.062 8.47 ID41 53 108.5 5.2 0.070 2.93 ID42 55 46.5 6.2 0.057 3.45

That is, the decision tree re-learning unit 17 may relearn the decision tree through the data shown in Tables 1 and 2 above. On the other hand, such re-learning may be performed when data is input by a consumer.

Again, FIG. 2 will be described.

The controller 11 controls the guide unit 14 to output a management index for delaying the progression to the predicted hair loss state (S50), and a product recommendation unit to recommend a hair-related product suitable for reaching a state according to the management index (19) can be controlled (S70).

The controller 11 may control the display unit 18 to display at least one of a management index and a recommended product.

11 is an exemplary diagram of a method in which the hair loss management apparatus according to an embodiment of the present disclosure displays a management index, and FIG. 12 is an example of a method in which the hair loss management apparatus according to an embodiment of the present disclosure displays a management index and a recommended product It is a drawing.

Referring to FIG. 11 , the display unit 18 may display a current hair loss state, a predicted hair loss state, and a management index. The predicted hair loss state is a hair loss state expected after the lapse of a predetermined time calculated according to the current hair loss state, and the management index may be management information required to delay progress to the predicted hair loss state.

According to an embodiment, the hair loss management apparatus 1 may store in advance a management index according to the current hair loss state and the predicted hair loss state in a memory (not shown). That is, the hair loss management device 1 is a management index when the current hair loss state is the first grade, the predicted hair loss state is the second grade, the current hair loss state is the first grade, and the predicted hair loss state is the third grade The management index at the time, the current hair loss state is the second grade, the predicted hair loss state is the third grade, and the management index and the like may be stored in a memory (not shown), respectively. The controller 11 may obtain a management index from a memory (not shown) according to the current hair loss state and predicted hair loss state, and control the display unit 18 to display the management index obtained from the memory (not shown). .

According to another embodiment, the hair loss management device 1 may store in advance a management index derivation formula (not shown) for deriving a management index according to the current hair loss state and the predicted hair loss state, and a management index derivation formula (not shown). City) can be used to control the display unit 18 to display the extracted management index.

The user may manage the hair thickness to 0.070 mm, the scalp moisture to 25 microsiemens, and the scalp pH to 6 with reference to the management index as shown in FIG. 11 .

Meanwhile, referring to FIG. 12 , the display unit 180 may further display a recommended product in the current hair loss state, the predicted hair loss state, and the management index. The recommended product may be a hair-related product that helps management as a management index. The recommended product may be a suitable product to reach a target value according to a management index.

The user can more easily manage the hair thickness to 0.070mm, the scalp moisture to 25 microsiemens, and the scalp pH to 6 through the use of the recommended product with reference to the management index and the recommended product as shown in FIG. 12 .

On the other hand, when displaying the predicted hair loss state as shown in FIG. 11 or 12 , the display unit 18 may further display a parting area or a predicted parting image according to the predicted hair loss state.

According to another embodiment of the present disclosure, the hair loss management apparatus may predict the hair loss state according to the input index.

13 is a control block diagram of a hair loss management apparatus according to another embodiment of the present disclosure.

The hair loss management apparatus 1 according to another embodiment of the present disclosure includes a controller 11, a measurement unit 12, a prediction unit 13, a guide unit 14, a counseling unit 15, a decision tree generation unit ( 16), it may include at least some or all of the decision tree re-learning unit 17 , the display unit 18 , and the product recommendation unit 19 . That is, the hair loss management apparatus 1 according to another embodiment of the present disclosure may further include a counseling unit 15 unlike the embodiment described above.

Since the components other than the counseling unit 15 are the same as those described above, overlapping descriptions will be omitted.

When receiving an input of at least one index of hair thickness, scalp moisture, and scalp pH, the counseling unit 15 may predict a hair loss state according to the input index. Alternatively, the counseling unit 15 may predict a future hair loss condition based on the current hair loss condition. In addition, the counseling unit 15 may predict a future hair loss state by reflecting the index input to the current hair loss state.

14 is a flowchart illustrating a first example of a method of operating an apparatus for managing hair loss according to another embodiment of the present disclosure.

The controller 11 may control the measurement unit 12 to diagnose the current user state ( S110 ).

This is the same as described in step S10 of FIG. 2 , and thus a redundant description will be omitted.

The controller 11 may receive an input of at least one index of hair thickness, scalp moisture, and scalp pH (S120).

For example, the hair loss management device 1 may further include an input module (not shown) such as a physical key button or a touch pad, and the controller 11 may control hair thickness and scalp moisture through an input module (not shown). and at least one indicator of scalp pH may be received.

The controller 11 may display the predicted hair loss state when managed by the input index (S130).

The controller 11 may calculate and display the predicted hair loss state when managed by the index input in step S120 through the counselor unit 15 . Specifically, the counselor unit 15 may calculate the predicted hair loss state when the current user state diagnosed through the measurement unit 12 is managed by the input index.

15 is an exemplary diagram of a method for displaying a predicted hair loss state by a hair loss management apparatus operating according to the flowchart shown in FIG. 14 .

The display unit 18 may display the current hair loss state, the input index, and the predicted hair loss state. According to the example of FIG. 15 , when the current hair loss state is I-2 and the input index is 0.05 hair thickness, the counselor 15 calculates the predicted hair loss status as II-4, and the calculated predicted hair loss status is displayed The display unit 18 can be controlled so as to be possible.

Through this, the user can feel the importance and necessity of management according to the input index. That is, there is an advantage that the user can easily check the predicted hair loss state when managed by each index while variously inputting the index.

On the other hand, the controller 11 may not receive an index, and in this case, the predicted hair loss state according to the current user state may be displayed.

16 is a flowchart illustrating a second example of a method of operating a hair loss management apparatus according to another embodiment of the present disclosure.

The controller 11 controls the measurement unit 12 to diagnose the current user state (S110), receives an input of at least one index among hair thickness, scalp moisture, and scalp pH (S120), and manages it with the input index If so, the predicted hair loss state may be displayed (S130).

This is the same as described in steps S110, S120, and S130 of FIG. 14, and thus a redundant description will be omitted.

The controller 11 displays the predicted hair loss state when managed with the input index, and then outputs a management index for slowing the progression to the predicted hair loss state (S140), and hair-related products suitable for reaching the state according to the management index can be recommended (S150).

That is, in FIG. 2, management indicators and hair-related products for delaying the progress to the predicted hair loss state according to the current user state are displayed. However, according to another embodiment of the present disclosure, as shown in FIG. 17 , the hair loss management device 1 is related to a management index and hair for delaying the progress to a predicted hair loss state according to the current user state and the input index. product can be displayed.

17 is an exemplary diagram of a method for displaying a predicted hair loss state by the hair loss management apparatus operating according to the flowchart shown in FIG. 16 .

The display unit 18 may display the current hair loss state and the management index for managing the predicted hair loss state and the predicted hair loss state according to the input index, and hair-related products together. According to the example of FIG. 17 , when the current hair loss state is I-2 and the input index is 0.05 hair thickness, the counselor 15 calculates the predicted hair loss status as II-4, and manages the calculated predicted hair loss status In order to be effective, the display unit 18 may be controlled to display required management indicators and hair-related products.

Through this, the user can not only feel the importance and necessity of management according to the input index, but also has the advantage of being more easily guided with the management guidelines for the management of the predicted hair loss state according to the input index.

According to an embodiment of the present disclosure, the above-described method may be implemented as a processor-readable code on a medium in which a program is recorded. Examples of the processor-readable medium include those implemented in the form of ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like.

In the hair loss management apparatus described above, the configuration and method of the above-described embodiments are not limitedly applicable, but all or part of each embodiment is selectively combined so that various modifications can be made. it might be

That is, the above description is merely illustrative of the technical idea of the present disclosure, and those skilled in the art to which the present disclosure pertains may make various modifications and variations without departing from the essential characteristics of the present disclosure. .

Accordingly, the embodiments disclosed in the present disclosure are for explanation rather than limiting the technical spirit of the present disclosure, and the scope of the technical spirit of the present disclosure is not limited by these embodiments.

The protection scope of the present disclosure should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present disclosure.

1: Hair loss management device
11: Controller
12: measurement unit
13: prediction unit
14: guide unit
16: Decision tree generator
17: Decision tree re-learning department
18: display unit
19: Product Recommendation Department

Claims (24)

a measurement unit for diagnosing the current user state;
a prediction unit for calculating a predicted hair loss state that is a hair loss state after a predetermined period has elapsed according to the current user status; and
Includes a guide unit for outputting a management index for slowing the progression to the predicted hair loss state,
the prediction unit
The predicted hair loss state is calculated through a decision tree in which the explanatory variable is at least one of age, hair thickness, scalp moisture, and scalp pH, and the dependent variable is at least one of part area and hair loss grade,
a decision tree generating unit for generating the decision tree by using at least one of the subject's age, hair thickness, scalp moisture, and scalp pH, and at least one of the subject's parting area and hair loss grade; and
Further comprising a decision tree re-learning unit for re-learning the decision tree generated by the decision tree generation unit
Hair loss management device.
delete delete According to claim 1,
The decision tree generating unit
generating the decision tree by comparing the prediction rates of the area of parts according to each of a plurality of combinations consisting of at least one of the age, hair thickness, scalp moisture, and scalp pH
Hair loss management device.
delete According to claim 1,
The decision tree re-learning unit
At the first time point, when the correlation between at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area follows the decision tree, the second time point elapsed from the first time point By comparing the predicted parting area output as the age, hair thickness, scalp moisture or scalp pH of the subject measured at the time point is input to the decision tree with the subject's parting area measured at the second time point, the doctor re-learning the decision tree
Hair loss management device.
According to claim 1,
Further comprising a display unit for displaying the area of the part according to the predicted hair loss state
Hair loss management device.
According to claim 1,
The management indicator is
Including at least one of hair thickness, scalp moisture, and scalp pH required to slow the progression to the predicted hair loss state
Hair loss management device.
9. The method of claim 8,
Further comprising a product recommendation unit for recommending a hair-related product suitable for reaching at least one of the required hair thickness, scalp moisture, and scalp pH
Hair loss management device.
According to claim 1,
The hair loss condition
Including the hair loss grade determined by the area of the part
Hair loss management device.
According to claim 1,
Further comprising a counseling unit that calculates a predicted hair loss state when the current user state is managed with an index including at least one of hair thickness, scalp moisture, and scalp pH input by the user
Hair loss management device.
12. The method of claim 11,
the guide part
Further outputting a management index to be managed in the predicted hair loss state calculated by the counseling unit
Hair loss management device.
In the method of the hair loss management device providing guidelines for hair loss management,
diagnosing the current user condition;
calculating a predicted hair loss state that is a hair loss state after a predetermined period has elapsed according to the current user state; and
Comprising the step of outputting a management index for slowing the progression to the predicted hair loss state,
The step of calculating the predicted hair loss state is
Calculating a predicted hair loss state through a decision tree in which the explanatory variable is at least one of age, hair thickness, scalp moisture, and scalp pH, and the dependent variable is at least one of part area and hair loss grade,
the method
generating the decision tree using at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area; and
Further comprising the step of re-learning the decision tree
A method of providing hair loss management instructions for a hair loss management device.
delete delete 14. The method of claim 13,
The step of generating the decision tree is
Comprising the step of generating the decision tree by comparing a plurality of combinations consisting of at least one of the age, hair thickness, scalp moisture, and scalp pH and the prediction rate of the area of parts according to each of the plurality of combinations
A method of providing hair loss management instructions for a hair loss management device.
delete 14. The method of claim 13,
The step of re-learning the decision tree is
At the first time point, when the correlation between at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the subject's parting area follows the decision tree, the second time point elapsed from the first time point Measuring at least one of the subject's age, hair thickness, scalp moisture, and scalp pH and the area of the part at the time point;
The predicted parting area output as at least one of the subject's age, hair thickness, scalp moisture, and scalp pH measured at the second time point is input to the decision tree is calculated as the subject's parting area measured at the second time point comparing with, and
Comprising the step of adjusting the correlation according to the decision tree according to the comparison result
A method of providing hair loss management instructions for a hair loss management device.
14. The method of claim 13,
Further comprising the step of displaying a parting area according to the predicted hair loss state
A method of providing hair loss management instructions for a hair loss management device.
14. The method of claim 13,
The step of outputting the management index is
Including the step of outputting at least one of the hair thickness, scalp moisture, and scalp pH required to slow the progression to the predicted hair loss state
A method of providing hair loss management instructions for a hair loss management device.
21. The method of claim 20,
The method further comprising the step of recommending a hair-related product suitable for reaching at least one of the required hair thickness, scalp moisture, and scalp pH
A method of providing hair loss management instructions for a hair loss management device.
14. The method of claim 13,
The step of diagnosing the current user state is
Comprising the step of measuring at least one of the user's age, hair thickness, scalp moisture, and scalp pH
A method of providing hair loss management instructions for a hair loss management device.
14. The method of claim 13,
receiving an index input of at least one of hair thickness, scalp moisture, and scalp pH; and
Further comprising the step of calculating a predicted hair loss state when the current user state is managed by the input index
A method of providing hair loss management instructions for a hair loss management device.
24. The method of claim 23,
Further comprising the step of outputting a management index to be managed in the predicted hair loss state when managed by the input index
A method of providing hair loss management instructions for a hair loss management device.

Images