KR20240046992A - Method for providing diagnosis and treatment of skin dermatitis based on machine-learning - Google Patents

Abstract

According to an embodiment of the present disclosure, a method for diagnosing and treating dermatitis using machine learning-based skin image analysis technology is provided. The method of the present disclosure includes taking a picture of the skin; Analyzing the skin photo to identify all types of dermatitis appearing in the skin photo based on a deep learning model that identifies dermatitis types; calculating the sum of the effective areas of the bounding boxes for the corresponding dermatitis areas for each identified dermatitis type; Calculating the ratio of the bounding box area for each type of dermatitis to the skin area to be analyzed; Determining the severity level of dermatitis for each type of dermatitis shown in the skin photo; and setting the intensity value of the treatment medium of the dermatitis treatment device according to the severity level of dermatitis. may include.

Description

Dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology {Method for providing diagnosis and treatment of skin dermatitis based on machine-learning}

This disclosure relates to a method of providing a mobile service for dermatitis diagnosis and treatment using machine learning-based skin image analysis technology. More specifically, it relates to a method of diagnosing and treating dermatitis by analyzing dermatitis photos based on machine learning.

In general, people with dermatitis visit a dermatologist, receive a diagnosis of dermatitis from a dermatologist, and depending on the diagnosis results, receive treatment with a treatment device such as a laser or UV light, and then receive treatment such as taking a treatment and applying it to the affected area.

In the case of dermatitis treatment by visiting a dermatologist, not only does it take a lot of time to receive treatment from a specialist, but also expensive laser or UV treatment devices are used every time you visit and receive treatment, resulting in high treatment costs. Because it relies solely on the knowledge and experience of a specialist, diagnostic errors due to dermatitis diagnosis and subsequent side effects of treatment may occur.

Due to various reasons such as the inconvenience of visiting a dermatologist in person, the long waiting time to receive treatment from a dermatologist, the burden of expensive treatment costs due to treatment methods using expensive treatment devices, and side effects due to diagnostic errors by specialists, timely treatment is required. Because people do not visit a dermatologist, they miss the treatment period for dermatitis, which causes dermatitis to worsen.

In order to solve the above problems, Korea Patent Publication No. 10-2019-0051256 compares the normal skin image with the skin image of the diagnostic target based on characteristic values such as skin color, skin wrinkles, and skin flatness. Although a method for assessing the condition has been disclosed, there are clear limitations in detecting the severity of skin inflammation as a disease and proposing a treatment method as it can identify the simple condition of the skin.

Korean Patent Publication No. 10-2022-0057793 specifies skin inflammation called atopic skin disease and provides a method for accurately and quickly determining the type and severity of atopic dermatitis based on a deep-learned atopic dermatitis learning module. In general, there are limitations in determining the state of dermatitis, which is a mixture of various types of dermatitis that can exist on the skin, and providing the optimal intensive management plan and appropriate treatment method.

Patent Document 1: Republic of Korea Patent Publication No. 10-2019-0051256 Patent Document 2: Republic of Korea Patent Publication No. 10-2022-0057793

According to an embodiment of the present disclosure, dermatitis diagnosis and treatment services can be provided using machine learning-based skin image analysis technology.

According to an embodiment of the present disclosure, the type and severity of dermatitis is calculated based on machine learning through an algorithm that considers skin conditions in which multiple types of dermatitis are mixed for skin photos, and according to the calculated severity of dermatitis. By automatically providing a suitable treatment method, the type and severity of dermatitis can be calculated relatively accurately through objective data.

According to another embodiment of the present disclosure, it is possible to diagnose and treat dermatitis in a dermatitis patient by analyzing the skin condition in which multiple types of dermatitis are mixed in the dermatitis photo, so that the dermatitis patient can be diagnosed and treated in a timely manner without having to visit a dermatology hospital. You can receive dermatitis diagnosis and treatment services.

According to one feature of the present disclosure, methods for diagnosing and treating dermatitis are provided. Dermatitis diagnosis and treatment methods include taking pictures of the skin; Analyzing the skin photo to identify all types of dermatitis appearing in the skin photo based on a deep learning model that identifies dermatitis types; calculating the sum of the effective areas of the bounding boxes for the corresponding dermatitis areas for each identified dermatitis type; Calculating the ratio of the bounding box area for each type of dermatitis to the skin area to be analyzed; Determining the severity level of dermatitis for each type of dermatitis shown in the skin photo; and setting the intensity value of the treatment medium of the dermatitis treatment device according to the severity level of dermatitis.

According to an embodiment of the present disclosure, types of dermatitis (comedogenic dermatitis, papular dermatitis, pustular dermatitis, nodular dermatitis, cystic dermatitis) mixed on the skin are determined based on machine learning through an algorithm that considers multiple situations for skin photos. Dermatitis, coagulative dermatitis, atopic dermatitis) can be identified, the severity of dermatitis can be calculated based on the identified type of dermatitis, and an appropriate treatment method can be provided according to the calculated severity of dermatitis.

According to an embodiment of the present disclosure, the type and severity of dermatitis can be relatively accurately calculated through objective data.

According to an embodiment of the present disclosure, it is possible to diagnose and treat dermatitis in dermatitis patients through a mobile app, thereby providing user convenience.

In addition, according to an embodiment of the present disclosure, people are reluctant to visit dermatologists due to reasons such as the burden of expensive treatment costs due to treatment methods using expensive treatment devices provided by dermatologists and side effects due to diagnostic errors by specialists. It can reduce the risk of missing the treatment period for dermatitis.

Figure 1 illustrates the configuration of a dermatitis diagnosis and treatment system using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.
Figure 2 shows a flowchart of a method of providing a mobile service for diagnosing and treating dermatitis using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.
Figure 3 is a configuration diagram of an application (mobile app) for a mobile service for dermatitis diagnosis and treatment using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.
Figure 4(a)(b) is a diagram illustrating an image of skin with mixed nodular, papular, and comedonal dermatitis, according to an embodiment of the present disclosure.
Figure 5 is a flowchart for generating and learning a dermatitis diagnosis and treatment model using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Below, if it is judged that there is a risk of unnecessarily obscuring the gist of the present disclosure, detailed descriptions of already known functions and configurations will be omitted. In addition, it should be noted that the content described below only relates to one embodiment of the present disclosure and is not limited thereto.

The terms used in this disclosure are only used to describe specific embodiments and are not intended to limit the disclosure. For example, a component expressed in the singular should be understood as a concept that includes plural components unless the context clearly indicates only the singular. It should be understood that the term “and/or” as used in this disclosure encompasses any and all possible combinations of one or more of the listed items. Terms such as 'include' or 'have' used in this disclosure are only intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in this disclosure, and the meaning of these terms is The use is not intended to exclude the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

In the embodiments of the present disclosure, 'module' or 'unit' refers to a functional part that performs at least one function or operation, and may be implemented as hardware or software, or as a combination of hardware and software. Additionally, a plurality of 'modules' or 'units' may be integrated into at least one software module and implemented by at least one processor, except for 'modules' or 'units' that need to be implemented with specific hardware. there is.

In addition, unless otherwise defined, all terms used in this disclosure, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which this disclosure pertains. It should be noted that terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the context meaning of the related technology, and should not be interpreted in an overly restrictive or expanded manner unless clearly defined otherwise in the present disclosure. .

Hereinafter, a dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology according to an embodiment of the present disclosure will be described in detail with reference to the drawings.

Figure 1 illustrates the configuration of a dermatitis diagnosis and treatment system using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.

The dermatitis diagnosis and treatment system 100 using machine learning-based skin image analysis technology may include a dermatitis treatment device 110, a user terminal 120, a server 130, and a communication network 140.

In one embodiment of the present disclosure, the dermatitis treatment device 110 generates a medium that has a skin sterilizing and treatment effect, such as plasma, laser, UV, etc., on the dermatitis area, so that it can be easily applied to the dermatitis area and is configured to be convenient to carry. It can be. In one embodiment, the dermatitis treatment device 110 is connected to the user terminal 120 through a short-distance wireless communication medium, receives the intensity value of the treatment medium transmitted from the application installed on the user terminal 120, and converts the intensity value to the corresponding intensity value. It may be configured to generate an intensity of the therapeutic medium.

In one embodiment of the present disclosure, the user terminal 120 may be configured to capture a skin image and analyze diagnosis and treatment information of the captured skin image. In one embodiment, the user terminal 120 may include an application (mobile app) installed on the user's terminal that is configured to enable camera photography of the skin area for which the user wants to be diagnosed. For example, the user terminal 120 is configured to enable camera photography by selecting the skin area for which diagnosis is desired. The application uses machine learning images to classify the type of dermatitis in order to determine the mild or severe state of dermatitis for the photographed skin area. It has a deep learning dermatitis type inference engine to enable object detection, determines the intensity of the appropriate treatment medium according to the severity of dermatitis based on the identified dermatitis type, and matches the intensity value of the determined treatment medium to the user terminal. It may be configured to be transmitted to a personal portable dermatitis treatment device connected through a wireless communication medium.

In one embodiment, the application installed on the user terminal 120 may be configured to provide the user with a dermatitis treatment and prescription method appropriate for the determined severity of dermatitis. In one embodiment of the present disclosure, the user terminal 120 may include a screen unit of a mobile app, and the screen unit allows the user to view dermatitis status diagnosis results for dermatitis photos taken with a camera by diagnosis date, and dermatitis It recommends a treatment appropriate for the dermatitis condition based on the diagnosis results and provides a prescription method. Additionally, if the dermatitis condition is urgently needed based on the established criteria as a result of the diagnosis, the system is structured so that the person can immediately contact the dermatologist to receive treatment. , it can be configured to minimize the risk of missing the treatment period for dermatitis by allowing the user to receive treatment in a timely manner according to the user's dermatitis condition.

In one embodiment of the present disclosure, the server 130 identifies the types of dermatitis mixed on the skin using a deep learning model in a method of providing a mobile service for dermatitis diagnosis and treatment using machine learning-based skin image analysis technology. According to the identified type of dermatitis, a model that determines the severity of dermatitis of the skin being analyzed can be created and trained.

In one embodiment of the present disclosure, the communication network 140 may be a short-range wireless communication medium. In one embodiment, short-range wireless communication media may include Bluetooth, NFC, Wi-Fi, etc. In one embodiment, the dermatitis treatment device 110 and the user terminal 120 may be configured to automatically connect through the communication network 140 when the user terminal 120 and the user's dermatitis treatment device 110 are mutually activated. there is.

Figure 2 shows a flowchart of a method of providing a mobile service for diagnosing and treating dermatitis using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.

First, the user terminal 120 may capture a skin image and transmit it to the server 130 (S202). In one embodiment, the user terminal 120 may capture a skin image using an application (mobile app) installed on the terminal and transmit it to the server 130 (S204).

Next, the server 130 can create and learn a deep learning dermatitis analysis engine. The server 130 can analyze the skin image received from the user terminal 120 using a deep learning dermatitis analysis engine. In one embodiment, the server 130 may generate diagnosis and prescription information for dermatitis (S206). In one embodiment, the server 130 may transmit diagnosis and prescription information for dermatitis to the user terminal 120 (S208).

Next, the user terminal 120 displays the diagnosis and prescription information about dermatitis received from the server 130 on the screen of the user terminal 120 so that the user can view the dermatitis information about the transmitted skin image. can do. In one embodiment, the user terminal 120 may generate dermatitis treatment or prescription recommendation information based on the diagnosis and prescription information for dermatitis received from the server 130 (S210).

Additionally, the user terminal 120 may transmit device setting information to the dermatitis treatment device 110 based on the dermatitis diagnosis and prescription information received from the server 130 (S212). For example, the user terminal 120 may set the treatment medium level of the dermatitis treatment device 110 based on diagnosis and prescription information for dermatitis.

Next, the dermatitis treatment device 110 may be driven at a set treatment medium level (S214). In one embodiment, dermatitis treatment device 110 may be driven with a set treatment medium level in response to receiving a user's start command.

Figure 3 is a configuration diagram of an application (mobile app) for a mobile service for dermatitis diagnosis and treatment using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.

In one embodiment of the present disclosure, the user terminal 120 includes a module described later, and as a result of the module described later, it is possible to provide the user with a final dermatitis condition severity diagnosis result and action information appropriate for the dermatitis condition diagnosis result. . In one embodiment, the user terminal 120 provides a treatment medium (plasma, laser, UV, etc.) can be configured to automatically set the intensity.

In one embodiment of the present disclosure, the user terminal 120 may include a camera module 310, a plasma generation energy control API 380, and a display unit 390.

In one embodiment of the present disclosure, the camera module 310 may be configured to capture skin images.

In one embodiment of the present disclosure, the real-time image processing 320 may be configured to process and analyze images captured or received by the camera module 310 in real time. In one embodiment, real-time image processing 320 may be configured to convert skin image pixel data captured by a camera module into data that can be used for machine learning analysis.

In one embodiment of the present disclosure, the data filtering module 330 may be configured to remove noisy data from data to be used for machine learning image analysis.

In one embodiment of the present disclosure, the CNN-based deep learning module 340 uses a convolutional neural network method verified as an image deep learning method to detect target objects (dermatitis types) from the input image. It may be configured to extract features and detect dermatitis types present in the image from the extracted features.

In one embodiment of the present disclosure, the knowledge base 350 may be composed of a database that structures and systematizes treatments and treatment methods appropriate for the severity of dermatitis by type of dermatitis, medically and scientifically verified from dermatitis experts and clinical result data. In one embodiment, the knowledge base 350 may provide input data to the skin condition diagnosis and action module 360, which will be described later.

In one embodiment of the present disclosure, the skin condition diagnosis and action module 360 may be configured to diagnose the severity of dermatitis in the user's skin area and to find and recommend treatments and prescription methods associated with the diagnosed dermatitis condition from the knowledge base 350. You can.

In one embodiment of the present disclosure, the module for setting the energy level generated by the treatment medium (plasma, laser, UV, etc.) (370) is used to determine dermatitis diagnosed by a personal portable dermatitis treatment device using treatment medium (plasma, laser, UV, etc.) technology. It may be configured to set the intensity value of the treatment medium suitable for treatment according to the severity of the condition.

In one embodiment of the present disclosure, the energy control API 380 generated by the treatment medium (plasma, laser, UV, etc.) is registered in the mobile app installed on the user terminal and is automatically connected wirelessly as soon as the mobile app is running. It may be configured to transmit the generated intensity value of the treatment medium set in the treatment device. In one embodiment, the generated energy control API 380 allows the user terminal 120 and the dermatitis treatment device to feed back the results processed by the treatment medium intensity setting value transmitted to the personal portable dermatitis treatment device to the user terminal 120. (110) It can be configured by determining a two-way wireless communication protocol between them.

In one embodiment of the present disclosure, the skin condition diagnosis and action information display module 390 can display dermatitis diagnosis results, recommended treatments and prescription methods, and changes in treatment results according to the dermatitis condition through the screen of the user terminal 120. It can be configured so that

Figure 4(a)(b) is a diagram illustrating an image of skin with mixed nodular, papular, and comedonal dermatitis, according to an embodiment of the present disclosure.

In one embodiment of the present disclosure, a method of identifying the type of dermatitis shown in a skin image involves learning each dermatitis object distributed on the skin as a data point as shown in the picture below, and recognizing each dermatitis object present in the skin picture. It can be configured to do so. In one embodiment, the number of learning data points for recognition of dermatitis-type entities can be increased by more than 10 times with a small amount of training image data.

In addition, the dermatitis objects recognized using the learned deep learning dermatitis recognition model are calculated by calculating the number of objects recognized for each dermatitis type, the bounding box area of each object, and the total effective bounding box area for each dermatitis type using a predetermined equation described later. It can be.

Figure 5 is a flowchart for generating and learning a dermatitis diagnosis and treatment model using machine learning-based skin image analysis technology according to an embodiment of the present disclosure.

As shown in Figure 5, the dermatitis objects recognized using the learned deep learning dermatitis recognition model include the number of objects recognized by dermatitis type, the bounding box area of each object, and the total effective bounding box area by dermatitis type as shown below [ It is characterized by being calculated through Equation 1].

In addition, the ratio of the total area of the bounding box for each type of dermatitis to the target skin area to be analyzed is calculated using Equation 2 below.

In addition, the calculation of the KPI value that determines the grade of dermatitis severity for each type of dermatitis shown in the skin photo is characterized by being calculated using [Equation 3] below.

It is characterized by determining the severity of dermatitis for the skin subject to dermatitis diagnosis based on the dermatitis status grade standard as [Equation 4] according to the KPI value calculated using [Equation 3] below.

[Equation 1]

: Area of the bounding box of the jth individual of identified dermatitis type i

N(i): number of individuals of dermatitis type i identified

m: total number of individuals identified with dermatitis

N(i) = , here

= upper leftmost pixel point coordinate of the jth object bounding box of dermatitis type i

= Bottom rightmost pixel point coordinate of the jth object bounding box of dermatitis type i

= , here

Total effective area of dermatitis type i ( ) =

: Calculated from a deep learning model for the jth object of recognized dermatitis type i

confidence score

[Equation 2]

Total analyzed skin area (S) =

Proportion of dermatitis type i in the analyzed skin ( ) =

K: Number of classes of dermatitis types to be recognized by the deep learning model

[Equation 3]

=

: Weighting of dermatitis type i , from normal to very severe on a scale of 1 to 10.

Weights assigned to each type of dermatitis up to and including dermatitis

[Equation 4]

If KPI < k1, skin condition is considered normal

If k1≤KPI<k2, mild skin inflammation

If k2 ≤KPI<k3, somewhat severe skin inflammation condition

If k3 ≤KPI<k4, treatment by a specialist is recommended for severe skin inflammation.

If KPI≥k4, it is a very severe skin inflammation condition and treatment by a specialist is absolutely necessary.

A method of providing dermatitis diagnosis and treatment services using machine learning-based skin image analysis technology according to a preferred embodiment of the present invention involves photographing skin areas with various skin inflammations through a mobile app that can capture skin with a camera using a user terminal. And, the color, shape, size, distribution, etc. of inflammation, which are the unique characteristics of each skin inflammation mixed in the photographed skin area, are learned using a deep learning model, and the learned deep learning model shows the dermatitis area in the photo. The type of dermatitis can be classified into not only atopic dermatitis, but also comedonal, papular, pustular, nodular, cystic, concretory, keloid, and folliculitis, and can determine the severity of dermatitis and provide appropriate treatment services.

Programs executed by the terminals and servers described in this disclosure may be implemented with hardware components, software components, and/or a combination of hardware components and software components. A program can be executed by any system that can execute computer-readable instructions.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software may be implemented as a computer program including instructions stored on computer-readable storage media. Computer-readable recording media include, for example, magnetic storage media (e.g., ROM (Read-Only Memory), RAM (Random-Access Memory), floppy disk, hard disk, etc.) and optical read media (e.g., CD-ROM). (CD-ROM), DVD (Digital Versatile Disc), etc. The computer-readable recording medium is distributed among computer systems connected to a network, so that computer-readable code can be stored and executed in a distributed manner. The media may be readable by a computer, stored in memory, and executed by a processor.

Computer-readable storage media may be provided in the form of non-transitory storage media. Here, 'non-transitory' only means that the storage medium does not contain signals and is tangible, and does not distinguish whether the data is stored semi-permanently or temporarily in the storage medium.

Additionally, programs according to embodiments of the present disclosure may be included and provided in a computer program product. Computer program products are commodities and can be traded between sellers and buyers. A computer program product may include a software program and a computer-readable storage medium on which the software program is stored. For example, a computer program product is a product in the form of a software program (e.g., a downloadable application) that is distributed electronically by the manufacturer of the device or through an electronic marketplace (e.g., Google Play Store, App Store). may include. For electronic distribution, at least a portion of the software program may be stored on a storage medium or created temporarily. In this case, the storage medium may be a storage medium of a manufacturer's server, an electronic market server, or a relay server that temporarily stores a software program.

A computer program product, in a system comprised of a server and a device, may include a storage medium of a server or a storage medium of a device. Alternatively, if there is a third device (e.g., a smartphone) connected to the server or device for communication, the computer program product may include a storage medium of the third device. Alternatively, the computer program product may include a software program itself that is transmitted from a server to the device or a third device, or from a third device to the device. In this case, one of the server, the device, and the third device may execute the computer program product to perform the method according to the disclosed embodiments. Alternatively, two or more of a server, a device, and a third device may execute the computer program product and perform the methods according to the disclosed embodiments in a distributed manner. For example, the server may execute a computer program product stored on the server and control a device connected to the server to perform the method according to the disclosed embodiments. As another example, a third device may execute a computer program product to control a device communication-connected to the third device to perform the method according to the disclosed embodiment. When the third device executes the computer program product, the third device may download the computer program product from the server and execute the downloaded computer program product. Alternatively, a third device may perform the method according to the disclosed embodiments by executing a computer program product provided in a pre-loaded state.

As described above, although the embodiments have been described with limited examples and drawings, various modifications and variations can be made by those skilled in the art from the above description. For example, the described techniques may be performed in a different order than the described method, and/or components, such as a described computer system or module, may be combined or combined in a form different from the described method, or other components or equivalents may be used. Appropriate results can be achieved even if replaced or replaced by .

110: Dermatitis treatment device
120: user terminal
130: server
140: communication network

Claims (7)

A dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology,
Taking a picture of the skin;
Analyzing the skin photo to identify all types of dermatitis appearing in the skin photo based on a deep learning model that identifies dermatitis types;
calculating the sum of the effective areas of the bounding boxes for the corresponding dermatitis areas for each identified dermatitis type;
Calculating the ratio of the bounding box area for each type of dermatitis to the skin area to be analyzed;
Determining the severity level of dermatitis for each type of dermatitis shown in the skin photo; and
Setting the intensity value of the treatment medium of the dermatitis treatment device according to the severity level of dermatitis;
Dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology, including. According to claim 1,
The step of setting the intensity value of the treatment medium of the dermatitis treatment device according to the severity level of dermatitis is
automatically controlling the intensity value of the treatment medium of the dermatitis treatment device to the intensity value of the treatment medium appropriate for the severity level of dermatitis; and
Dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology, including providing expert treatment recommendations and prescription methods according to the dermatitis severity grade. In paragraph 1
Dermatitis using machine learning-based skin image analysis technology, characterized in that dermatitis entities are calculated using the following equation: the number of entities recognized for each dermatitis type, the bounding box area of each entity, and the total effective bounding box area for each dermatitis type. Diagnosis and treatment methods.
: Area of the bounding box of the jth individual of identified dermatitis type i
N(i): number of individuals of dermatitis type i identified
m: total number of individuals identified with dermatitis
N(i) = , here
= upper leftmost pixel point coordinate of the jth object bounding box of dermatitis type i
= coordinate of the lower rightmost pixel point of the bounding box of the jth object of dermatitis type i
= , here
Total effective area of dermatitis type i ( ) =
: Calculated from a deep learning model for the jth object of recognized dermatitis type i
confidence score
In paragraph 1
Dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology, wherein the ratio of the total effective area of the bounding box for each dermatitis type to the target skin area to be analyzed is calculated using the equation below.
Total analyzed skin area (S) =
Proportion of dermatitis type i in the analyzed skin ( ) =
In paragraph 1
Dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology, wherein the calculation of the KPI value for determining the grade of dermatitis severity for each type of dermatitis shown in a skin photo is calculated using the equation below.
=
: Weight of dermatitis type i , weight given to each dermatitis type up to dermatitis on a scale of 1-10.
In paragraph 1
Dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology, characterized in that the severity of dermatitis for the skin subject to dermatitis diagnosis is determined based on the dermatitis grade criteria below according to the KPI value.


If KPI < k1, skin condition is considered normal
If k1≤KPI<k2, mild skin inflammation
If k2 ≤KPI<k3, somewhat severe skin inflammation condition
If k3 ≤KPI<k4, treatment by a specialist is recommended for severe skin inflammation.
If KPI≥k4, it is a very severe skin inflammation condition and treatment by a specialist is absolutely necessary.
In paragraph 1
According to the dermatitis severity grade based on the KPI value, the intensity of the treatment medium (plasma, UV, laser, etc.) appropriate for the severity grade of the dermatitis is determined, and the intensity value of the determined treatment medium is transmitted to the dermatitis treatment device connected to enable communication, thereby preventing the dermatitis. A dermatitis diagnosis and treatment method using machine learning-based skin image analysis technology, wherein the treatment device is automatically set to the intensity value of the delivered treatment medium.

Images