KR102419567B1 - Method and apparatus for predicting diagnosis of skin diseases - Google Patents

Abstract

Disclosed are methods and apparatus for predicting diagnosis of skin diseases. A method for predicting diagnosis of a skin disease, the method comprising: receiving, from a user terminal, questionnaire data obtained based on a tree structure algorithm; determining diagnostic data for the skin disease using a neural network model to which the questionnaire data is input; and transmitting diagnostic data to the user terminal.

Description

Method and apparatus for predicting diagnosis of skin diseases

The following examples relate to diagnostic prediction technology for skin diseases.

History taking is a diagnostic method that aims to inquire from a doctor's point of view, listen to the patient's complaints, and inquire about the patient's family history. Questionnaire is an essential process in the medical treatment process and provides important clues in making a diagnosis. The doctor goes through the process of diagnosing the current disease by asking the patient about the onset situation and the progress or change of symptoms, especially the main pain.

In diagnosing skin diseases, an examination by looking at the lesion with the eyes is important, but an accurate and effective questionnaire is the basis of diagnosis and is a very important process in increasing the accuracy of diagnosis results. However, in the actual medical field in Korea, it is difficult for medical staff to spend a lot of time on questionnaires within a limited time, and even if they do, there are many questions that are omitted or missed in busy medical conditions.

Nowadays, there are many cases in which the patient actively explores his or her disease by using the Internet or the like to search for a disease similar to his or her skin disease. However, in reality, the amount of information on the Internet is so vast and there is a lot of information that is not medically proven, so there are problems that inaccurate self-diagnosis and try wrong folk remedies or treatment methods to aggravate symptoms.

In addition, a skin diagnosis technique using artificial intelligence technology has recently been developed, but it mainly diagnoses based on photos, and there is no function to collect detailed and accurate information about the questionnaire, so there is a limit to the accuracy of diagnosis.

Korean Patent Publication No. 10-2019-0140881 (published on December 20, 2019)
Korean Patent Publication No. 10-2019-0104660 (published on: 2019.09.11)

A method for predicting diagnosis of a skin disease according to an embodiment includes: receiving, from a user terminal, questionnaire data obtained based on a tree structure algorithm; determining diagnostic data for the skin disease using a neural network model to which the questionnaire data is input; and transmitting the diagnostic data to the user terminal.

The questionnaire data may be collected through a chatbot program executed in the user terminal.

The user terminal provides predetermined current query data to the user through a chatbot program, and when receiving a user input corresponding to the current query data, determines next query data based on the user input, and the user can provide the following query data to

The user terminal determines, based on a user input corresponding to the current query data, the next query data including a more detailed query than the current query data for the skin disease, and provides the next query data to the user. can provide

The questionnaire data may include user selection information on at least one of image data including a symptom of at least one of a primary and secondary outbreak of a skin disease and image data including information on a site of occurrence of a symptom.

The questionnaire data may include a user input for at least one of the user's gender, age, underlying disease, family history, living environment, occupation, hobbies, medications taken, cosmetics used, and symptoms of skin diseases.

The symptom of the skin disease may include at least one of a location where the symptom occurs and an aspect of the symptom.

The diagnosis data may include at least one of a predicted diagnosis name for the user's skin disease, a list of image data including symptoms of the skin disease, and information on the skin disease.

The neural network model may be trained using at least one of questionnaire data for each skin disease and image data of a lesion.

For the questionnaire data, data cleaning may be performed to recognize text, and image data may be subjected to image segmentation to detect features.

A method for predicting a diagnosis of a skin disease performed in a user terminal according to an embodiment includes: executing a program for collecting questionnaire data; receiving, through the program, a user input with respect to the query data provided to the user and a user selection of at least one of image data including symptoms and image data including information on the site of occurrence of symptoms; transmitting the questionnaire data based on the received user input and user selection to a diagnostic prediction device; and providing the diagnostic data to a user when the diagnostic data is received from the diagnostic prediction apparatus.

The step of receiving the input may include: providing predetermined current query data to the user through a chatbot program; and when a user input corresponding to the current query data is received, determining next query data based on the user input and providing the next query data to the user.

The next query data may include a more detailed query than the current query data on the skin disease.

The step of receiving the input includes providing the user with at least one of image data including symptoms of at least one of primary and secondary outbreaks of skin diseases and image data including information on the site of occurrence of symptoms through the chatbot program. step; and receiving a user selection for the provided image data.

The diagnosis prediction apparatus may determine the diagnosis data for the skin disease by using a neural network model to which the questionnaire data is input.

A method for predicting a diagnosis of a skin disease according to another embodiment includes executing a program for collecting questionnaire data; receiving, through the program, a user input with respect to the query data provided to the user and a user selection of at least one of image data including symptoms and image data including information on the site of occurrence of symptoms; determining diagnostic data for the skin disease by using a neural network model to which questionnaire data based on the received user input and user selection are input; and providing the determined diagnostic data to the user.

A diagnostic prediction apparatus for performing a method for predicting diagnosis of a skin disease according to an embodiment includes a memory and a processor, wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor When executed, the processor receives, from the user terminal, the questionnaire data obtained based on a tree structure algorithm, and uses a neural network model to input the questionnaire data to the diagnosis prediction device for the diagnosis data on the skin disease. and control the diagnosis prediction apparatus to transmit the diagnosis data to the user terminal.

A user terminal for performing the method for predicting diagnosis of a skin disease according to an embodiment includes a memory and a processor, the memory stores instructions executable by the processor, and the instructions are executed by the processor. When the processor executes a program for the user terminal to collect the questionnaire data, and through the program, the image data including the user input and symptoms for the query data provided to the user and the symptom occurrence site information When a user selection for at least one of the image data is received, the questionnaire data based on the received user input and the user selection is transmitted to the diagnostic prediction apparatus, and when the diagnostic data is received from the diagnostic prediction apparatus, the user The user terminal may be controlled to provide the diagnostic data to the user.

The processor determines the next query data based on the user input when the user terminal provides predetermined current query data to the user through a chatbot program, and receives a user input corresponding to the current query data, and , may control the user terminal to provide the next query data to the user.

The processor sends, through the chatbot program, at least one of image data including a symptom of at least one of a primary and secondary outbreak of a skin disease and image data including information on a site of occurrence of a symptom, to the user by the user terminal. and may control the user terminal to receive a user selection for the provided image data.

A diagnostic prediction apparatus for performing a method for predicting a diagnosis of a skin disease according to another embodiment, comprising a memory and a processor, wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor When executed by the processor, the diagnostic prediction device executes a program for collecting questionnaire data, and through the program, image data including symptoms and user input for query data provided to the user and symptoms Diagnosis data for the skin disease by using a neural network model that receives a user selection of at least one of image data including information on the occurrence site, and inputs the questionnaire data based on the received user input and user selection as input and control the diagnosis prediction apparatus to provide the determined diagnosis data to the user.

According to an embodiment, when a patient inputs medical questionnaire data on a mobile basis, an algorithm for diagnosing which skin disease the patient is symptomatic and providing information on the disease may be initiated.

According to an embodiment, the accuracy of the diagnostic prediction method may be improved by accumulating the questionnaire data input by the patient.

According to an embodiment, the questionnaire data may be collected more efficiently by using a chatbot program.

According to an embodiment, in collecting questionnaire data, it is possible to improve patient accessibility by using a mobile-based chatbot program.

According to an embodiment, questionnaire data optimized for the patient's symptoms may be collected by using the chatbot program function.

According to an embodiment, the type of skin disease may be determined based on questionnaire data input by the patient and image data including a region where the lesion occurs.

According to an embodiment, it is possible to provide an information collection method that reduces the time required for medical staff to identify a patient during preliminary treatment and does not omit information.

According to an embodiment, it is possible to provide a method that can be used as an auxiliary means of treatment by increasing the value of use in treatment of opinions.

According to an embodiment, it is possible to provide an auxiliary medical diagnosis method that allows a medical team other than a dermatologist to diagnose a skin disease.

According to an embodiment, the accuracy of diagnosis may be improved by using both the questionnaire data and the image data including the lesion site of the patient.

According to an embodiment, it is possible to provide an auxiliary medical diagnosis method that enables a non-medical person to predict a skin disease and conveniently and accurately obtain information related to a skin disease from the Internet.

1 is a diagram illustrating an outline of a diagnostic prediction system according to an embodiment.
2 is a flowchart illustrating a diagnosis prediction method performed by a diagnosis prediction apparatus according to an exemplary embodiment.
3 is a flowchart illustrating a diagnosis prediction method performed in a user terminal according to an exemplary embodiment.
4 is a flowchart illustrating a diagnosis prediction method performed by a diagnosis prediction apparatus according to another exemplary embodiment.
5 and 6 are diagrams illustrating an example of a screen on which a chatbot program is executed according to an embodiment.
7 is a diagram for describing a diagnosis prediction model according to an exemplary embodiment.
8 is a diagram illustrating a configuration of a diagnostic prediction apparatus according to an exemplary embodiment.
9 is a diagram illustrating a configuration of a user terminal according to an embodiment.
10 is a diagram illustrating a configuration of a diagnostic prediction apparatus according to another embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes may be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all modifications, equivalents and substitutes for the embodiments are included in the scope of the rights.

Terms used in the examples are used for the purpose of description only, and should not be construed as limiting. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present specification, terms such as “comprise” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It is to be understood that this does not preclude the possibility of the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

In addition, in the description with reference to the accompanying drawings, the same components are assigned the same reference numerals regardless of the reference numerals, and the overlapping description thereof will be omitted. In the description of the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted.

In addition, in describing the components of the embodiment, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only for distinguishing the elements from other elements, and the essence, order, or order of the elements are not limited by the terms. When it is described that a component is “connected”, “coupled” or “connected” to another component, the component may be directly connected or connected to the other component, but between each component another component It will be understood that may also be "connected", "coupled" or "connected".

Components included in one embodiment and components having a common function will be described using the same names in other embodiments. Unless otherwise stated, descriptions described in one embodiment may be applied to other embodiments as well, and detailed descriptions within the overlapping range will be omitted.

1 is a diagram illustrating an outline of a diagnostic prediction system according to an embodiment.

The skin disease diagnosis prediction system described herein may diagnose the most likely skin disease based on questionnaire data input by patients on mobile devices and provide information on the skin disease. The diagnosis prediction system is based on various questionnaire data input by the patient (underlying disease, drug intake, occupation, living environment, products in use, diet, lifestyle, etc.) and the location and pattern of skin disease symptoms. You can provide a diagnostic name. The diagnostic prediction system may collect questionnaire data through a chatbot program, thereby improving the accessibility of the user (or patient). The diagnostic prediction system may determine the next query based on the user's answer to the query provided to the user. The diagnosis prediction system collects questionnaire data based on a tree structure algorithm that gradually narrows the category range for predicting a diagnosis name by providing a detailed range of queries from a wide range of queries based on the user's answer to the provided query. can The diagnostic prediction system described herein may provide a diagnostic prediction method by focusing on allergic diseases among skin diseases. Therefore, the diagnosis prediction system, unlike the prior art of performing a diagnosis depending only on external symptoms for a skin disease, converts the questionnaire data including information on the diet and lifestyle of the user or patient to data including external symptoms. In combination, it is possible to perform a diagnosis for an allergic skin disease presented to the user.

The diagnosis prediction system compensates for the problem of the prior art of omitting questions when the medical staff asks, so that when the medical staff asks, the diagnosis can be performed without omitting information about the patient. In addition, the diagnosis prediction system can solve the problem of aggravating symptoms by self-diagnosing using the Internet or the like without a patient visiting a hospital, and trying wrong folk remedies or treatment methods. The diagnosis prediction system may provide an accurate diagnosis result when a patient wants to self-diagnose so that the patient can receive an appropriate treatment for a disease. The diagnosis prediction system may compensate for a limitation in accuracy that may occur because the conventional technology utilizing artificial intelligence technology mainly performs a diagnosis based on an image. The diagnosis prediction system may improve accuracy of diagnosis prediction by performing diagnosis based on not only image data but also questionnaire data.

Referring to FIG. 1 , a diagnostic prediction system may include a diagnostic prediction apparatus 110 , a network 120 , and a user terminal 130 . The diagnostic prediction apparatus 100 and the user terminal 130 may communicate with each other through the network 120 (eg, an Internet communication network, a wired/wireless local area network, or a wide area data communication network).

In an embodiment, the user terminal 130 may collect the questionnaire data 140 from the user (or patient). The user terminal 130 may transmit the collected questionnaire data 140 to the diagnosis prediction apparatus 110 through the network 120 . The diagnosis prediction apparatus 110 may determine the diagnosis data 150 based on the questionnaire data 140 . The diagnostic prediction apparatus 110 may transmit the determined diagnostic data 150 to the user terminal 130 through the network 120 . The user terminal 130 may provide the diagnostic data 150 to the user.

2 is a flowchart illustrating a diagnosis prediction method performed by a diagnosis prediction apparatus according to an exemplary embodiment.

Referring to FIG. 2 , in step 210 , the apparatus for predicting diagnosis of a skin disease may receive, from the user terminal, questionnaire data obtained based on a tree structure algorithm.

In an embodiment, the questionnaire data received by the diagnostic prediction apparatus may be collected through a chatbot program executed in the user terminal. The user terminal provides predetermined current query data to the user through the chatbot program, and when receiving a user input corresponding to the current query data, determines next query data based on the user input and provides the next query data to the user can do.

According to an embodiment, when the next query data is provided to the user, the provided next query data may be the current query data.

Based on a user input corresponding to the current query data, the user terminal may determine next query data including a more detailed query than the current query data on the skin disease, and provide the next query data to the user. The user terminal may determine the next query data including a more detailed query in determining the diagnosis name of the skin disease than the current query data, based on the user input for the current query data. For example, if the current query data is 'What are the symptoms of your skin?' and the user's answer is 'Rash', the next query data has a narrower range than the current query data, such as 'What is the pattern of rashes?' may include queries.

Next query data based on the current query data may be determined through artificial intelligence technology based on the current query data and a user input corresponding to the current query data. Therefore, the diagnostic prediction method described herein can present the most important and efficient questions for the diagnosis of skin diseases, and can gradually increase the level of questionnaires, thereby improving the accuracy of diagnosis.

The questionnaire data may include user selection information on at least one of image data including symptoms of at least one of the primary and secondary outbreaks of skin diseases and image data including information on the site of occurrence of symptoms. For example, the user terminal may provide the user with image data including the shape and pattern of the skin lesion and image data including information on the area where the skin lesion has occurred. The user may select and input image data corresponding to a symptom appearing on the user's skin from among the provided image data. The shape and pattern of skin lesions can be important clues in the diagnosis of skin diseases. In addition, the site of occurrence of skin lesions can also be important information for diagnosis along with the questionnaire data.

The questionnaire data may include user input for at least one of the user's gender, age, underlying disease, family history, living environment, occupation, hobby, medication, cosmetics used, and symptoms of skin disease. Here, the symptoms of the skin disease may include at least one of a location where the symptoms occur and an aspect of the symptoms. Since the diagnosis prediction apparatus performs a diagnosis on an allergic skin disease among skin diseases, it is possible to collect questionnaire data including information such as a patient's diet and lifestyle.

Skin diseases are known to occur in a specific age group, and skin diseases that can occur frequently depending on occupational groups or underlying diseases are known. Accordingly, the diagnostic prediction apparatus may use such data to increase the accuracy of diagnosis prediction for a skin disease. For example, skin diseases such as dermatitis and pruritus are common in kidney disease patients, and infectious diseases and vascular skin diseases are highly likely to occur in diabetic patients. Accordingly, the diagnosis prediction apparatus may increase the accuracy of diagnosing a skin disease by using information such as a patient's medical history or underlying disease.

In operation 220 , the diagnostic prediction apparatus may determine diagnostic data for a skin disease by using a neural network model to which the questionnaire data is input.

In an embodiment, the neural network model may be trained using at least one of questionnaire data for each skin disease and image data of a lesion. Here, the questionnaire data for each skin disease and the image data of the lesion may be constructed based on the patient data and open big data previously visited.

In the process of learning the neural network model, the questionnaire data is data cleaned so that text can be recognized, and image data is image segmentation based on a convolutional neural network (CNN). may be performed, and a K-Nearest Neighbor (KNN) feature may be detected.

The diagnosis data may include at least one of a diagnosis name predicted for the user's skin disease, a list of image data including symptoms for the skin disease, and information on the skin disease. For example, the diagnostic data includes a list of image data including a diagnosis name of a possible skin disease and a representative symptom of the skin disease, other symptoms and treatment methods that may be caused by the skin disease. May contain relevant medical information.

In operation 230, the diagnostic prediction apparatus may transmit diagnostic data to the user terminal.

3 is a flowchart illustrating a diagnosis prediction method performed in a user terminal according to an exemplary embodiment.

Referring to FIG. 3 , in step 310 , the user terminal may execute a program for collecting questionnaire data. Here, the program for collecting the questionnaire data may be, for example, a chatbot program. The user terminal may improve accessibility of the user by collecting questionnaire data from the user using the chatbot program.

In step 320, the user terminal may receive, through a program, a user input for the query data provided to the user and a user selection for at least one of image data including symptoms and image data including information on the site of occurrence of symptoms. have.

In an embodiment, the user terminal may provide predetermined current query data to the user through a chatbot program. When the user terminal receives a user input corresponding to the current query data, the next query data may be determined based on the user input and the next query data may be provided to the user. Here, the next query data may include a more detailed query than the current query data on the skin disease. The next query data may be the current query data when provided to the user.

The user terminal may provide the user with at least one of image data including symptoms of at least one of the primary and secondary outbreaks of skin diseases and image data including information on the site of occurrence of symptoms through the chatbot program. The user terminal provides image data to the user in the process of collecting the questionnaire data, so that the user can more easily and accurately input the questionnaire data. The user terminal may receive a user selection for the provided image data.

The questionnaire data collected by the user terminal may be shown in Table 1 below.

vaginal answer gender Male = 0, Female = 1 age subjective skin symptoms None=0, Tickle=1, Pain=2, Pain to touch=3, Burning=4 objective skin rash Macula=0, macule=1, papule=2, plaque=3, nodule=4, vesicle=5, blister=6, pustule=7, wheal=8 objective skin rash Scab=0, Peeling=1, Swelling & Ulcer=2, Cracking=3, Hardening=4, Atrophy=5, Lichenization=6, None=7 pattern of rash Circular=0, Linear=1, Irregular=2, Annular=3 Accompanying skin symptoms None=0, ooze=1, pus=2 accompanying systemic symptoms Fever=0, Cough/Runny nose/Sputum=1, Abdominal pain/Diarrhea/vomiting=2, Headache/Neuralgia/Neck stiffness/Paralysis=3, Lip edema and dyspnea=4, Arthralgia=5, None=6 tint of the lesion Erythema=0, brown spot=1, white spot=2, black spot=3, pigmentation=4, purpura=5 Distribution pattern of rash Both sides=0, one side=1 Whole body=0, Local=1 (If local, sub-items: face=1, scalp=2, belly=3, chest=4, back=5, belly=6, arm=6, hand=7, leg=8 , foot=9) skin disease history Immune skin disease (psoriasis, atopic dermatitis)-0, infectious skin disease=1, tumor skin disease=2, none=3 family history of skin disease Immune skin disease (psoriasis, atopic dermatitis)-0, infectious skin disease=1, tumor skin disease=2, none=3 Presence of underlying disease High blood pressure None=0, Yes=1 diabetes None=0, Yes=1 kidney disease None=0, Yes=1 hepatobiliary disease None=0, Yes=1 thyroid disease None=0, Yes=1 tumor None=0, Yes=1 drug use history Whether the medication was changed at the same time as the symptom onset None=0, Yes=1 herbal medicine intake None=0, Yes=1 drug allergy None=0, Yes=1 Whether you are taking health food None=0, Yes=1 Allergic disease factors new cosmetic products None=0, Yes=1 newly changed clothes None=0, Yes=1 Whether your occupation deals with chemicals or special substances None=0, Yes=1 pet None=0, Yes=1 dye or perm None=0, Yes=1 infectious disease factors overseas travel history None=0, Yes=1 outdoor activity None=0, Yes=1 insect bite None=0, Yes=1 Nursing home or nursing home None=0, Yes=1

The user terminal may receive a response corresponding to each query, and may collect the questionnaire data by receiving a user selection of image data for a symptom.

In operation 330, the user terminal may transmit the questionnaire data based on the received user input and user selection to the diagnosis prediction apparatus.

The diagnosis prediction apparatus may determine the diagnosis data for a skin disease by using a neural network model to which the questionnaire data is input. The neural network model may determine diagnostic data for a skin disease based on a combination of user responses for each question included in the questionnaire data. For example, the diagnostic prediction device may present diagnostic data of atopic dermatitis through algorithmic analysis when a patient with chronic erythematosus accompanied by itching at the folded arm enters the questionnaire data related to the donation image.

In operation 340, when the user terminal receives the diagnostic data from the diagnostic prediction apparatus, the user terminal may provide the diagnostic data to the user.

4 is a flowchart illustrating a diagnosis prediction method performed by a diagnosis prediction apparatus according to another exemplary embodiment.

As for the diagnostic prediction method according to another embodiment, both the diagnostic prediction method performed by the user terminal and the diagnostic prediction apparatus according to the embodiment may be performed by the diagnostic prediction apparatus.

Referring to FIG. 4 , in step 410 , the diagnostic prediction apparatus may execute a program for collecting questionnaire data. For example, the diagnostic prediction device may execute a chatbot program.

The diagnosis prediction apparatus may provide at least one of query data, image data including a symptom, and image data including information on an occurrence site of a symptom through the chatbot program. In response to the provision, in step 420 , the diagnostic prediction device provides information on at least one of image data including user input and symptoms provided to the user through a program and image data including information on the site of occurrence of symptoms. User selection may be input.

The diagnosis prediction apparatus may determine the diagnosis data for the skin disease by using a neural network model in which the user input received in step 430 and questionnaire data based on the user selection are input as inputs.

The diagnostic prediction apparatus may provide the determined diagnostic data to the user in operation 440 . For example, the diagnosis prediction device includes a list of image data including a possible diagnosis name of a skin disease and a representative symptom of the skin disease, other symptoms and treatment methods that may be caused by the skin disease, and Diagnostic data including related medical information may be provided.

5 and 6 are diagrams illustrating an example of a screen on which a chatbot program is executed according to an embodiment.

5 is a diagram illustrating an example of a screen in which the user terminal provides query data to the user through a chatbot program and receives user input and user selection information from the user when the user complains of skin pain in the trunk . Referring to FIG. 5 , the user terminal provides the user through the chatbot program with query data on whether pain appears only on one side or on both sides, image data including examples of lesions, and query data on the appearance of symptoms. can provide

For example, the user terminal may provide the user with query data, "Is the skin pain bilateral or the skin pain unilateral?" Based on the query data on bilaterality and unilaterality, the user may input a user input of “unilaterality” to the user terminal, and the user terminal may receive a user input of “unilaterality”. Based on the user input, the user terminal may provide image data including the symptoms of the user's skin disease along with the text "Choose a picture close to the symptom of the lesion from the following pictures" based on the user input. According to an embodiment, the user terminal may further provide a description for each image data. The user terminal may receive the user selection information “9 (blister)” for image data corresponding to the symptoms appearing on the user's skin. The user terminal may provide query data, "Is the symptom spread in a band shape forward and backward?" in response to the user selection information. The user terminal may receive a user input of “yes” in response to the provision of the query data.

The user terminal may transmit the questionnaire data based on the user input received through the chatbot program and the user selection to the diagnosis prediction device. The diagnosis prediction apparatus may predict that the user's skin disease is 'herpes zoster' based on the questionnaire data. The diagnostic prediction apparatus may determine and transmit diagnostic data corresponding to herpes zoster to the user terminal. The user terminal may provide diagnostic data to the user. Here, the diagnostic data provided to the user may include at least one of a diagnosis name of 'herpes zoster', a list of image data corresponding to 'herpes zoster', and information on 'herpes zoster'.

6 may be a diagram illustrating an example of a screen in which the user terminal provides query data to the user through a chatbot program and receives user input and user selection information from the user when a rash appears on the user's knee. Referring to Figure 6, the user terminal through the chatbot program, query data on whether the user has itchy symptoms, query data on whether pain appears only on one side or both sides, and image data including examples of lesions can provide

For example, the user terminal may provide the user with query data "Is there an itchy symptom at the site where the rash appears?" Based on the query data on the itch symptom, the user may input a user input of “no”, and the user terminal may receive the user input. The user terminal may provide query data, "Is the skin symptom bilateral or the skin symptom unilateral?" based on the user input. In response to the query data for bilaterality and unilaterality, the user terminal may receive a user input of “ambiguity”. Based on the user input, the user terminal may provide query data "Is there keratin as shown in the image below?" and image data including lesions in which symptoms of keratin appear. The user terminal may input a user input of “yes” based on the provided query data and image data, and the user terminal may receive the user input.

The user terminal may transmit the questionnaire data based on the user input received through the chatbot program and the user selection to the diagnosis prediction device. The diagnosis prediction apparatus may predict that the user's skin disease is 'psoriasis' based on the questionnaire data. The diagnostic prediction apparatus may determine and transmit diagnostic data corresponding to psoriasis to the user terminal. The user terminal may provide diagnostic data to the user. Here, the diagnostic data provided to the user may include at least one of a diagnosis name of 'psoriasis', a list of image data corresponding to 'psoriasis', and information on 'psoriasis'.

The screen on which the chatbot program is executed as shown in FIGS. 5 and 6 may be output from a user terminal or a diagnosis prediction device according to an embodiment. Also, the embodiments described as being performed by the user terminal in the detailed description of FIGS. 5 and 6 may also be performed by the diagnostic prediction apparatus according to the embodiment.

7 is a diagram for describing a diagnosis prediction model according to an exemplary embodiment.

Referring to FIG. 7 , in an embodiment, the diagnostic prediction model 720 may include a neural network. The diagnostic prediction model 720 may receive the questionnaire data 710 , pass the questionnaire data 710 through the input layer, the hidden layer, and the output layer, and output the diagnosis data 740 from the questionnaire data 710 .

The questionnaire data 710 may include information about the patient's underlying disease, living environment information, job information, drug use information, and cosmetic use information. Also, according to an embodiment, the questionnaire data 710 may include user selection information for at least one of image data including a symptom of a skin disease and image data of a region where a symptom of a skin disease appears. The questionnaire data 710 may include responses collected by the query of Table 1 for information.

The database 730 may store setting information learned by the diagnostic prediction model 720 to accurately predict the diagnostic data, and may store the questionnaire data input in the process of the diagnostic prediction model 720 predicting the diagnostic data. can

8 is a diagram illustrating a configuration of a diagnostic prediction apparatus according to an exemplary embodiment.

Referring to FIG. 8 , the diagnostic prediction apparatus 800 may correspond to the diagnostic prediction apparatus described herein. The diagnostic prediction apparatus 800 may include a communicator 810 , a processor 820 , and a memory 830 . Also, according to an embodiment, the diagnostic prediction apparatus 800 may include a database 840 .

The memory 830 is connected to the processor 820 and may store instructions executable by the processor 820 , data to be calculated by the processor 820 , or data processed by the processor 820 . Memory 830 may include non-transitory computer-readable media, such as high-speed random access memory and/or non-volatile computer-readable storage media (eg, one or more disk storage devices, flash memory devices, or other non-volatile solid state memory devices). may include

The communicator 810 provides an interface for communicating with an external device (eg, a user terminal). The database 840 may store data necessary for the diagnostic prediction apparatus 800 to perform a diagnostic prediction method. For example, the database 840 may store data on which the neural network model predicts a skin disease. In addition, the database 840 may store the questionnaire data of patients exhibiting symptoms of skin diseases. The processor 820 may control the diagnostic prediction apparatus 800 so that the diagnostic prediction apparatus 800 may perform one or more operations related to the operation of the diagnostic prediction apparatus described with reference to FIG. 1 .

For example, the processor 820 receives the questionnaire data obtained by the diagnosis prediction device 800 from the user terminal based on the tree structure algorithm, and uses a neural network model to input the questionnaire data for skin diseases. The diagnostic prediction apparatus 800 may be controlled to determine diagnostic data and transmit the diagnostic data to the user terminal.

9 is a diagram illustrating a configuration of a user terminal according to an embodiment.

Referring to FIG. 9 , a user terminal 900 may correspond to the user terminal described herein. The user terminal 900 may include a processor 910 , a memory 920 , and a communicator 930 . Also, according to an embodiment, the user terminal 900 may further include a user input interface 940 and a display 950 .

The memory 920 is connected to the processor 910 and may store instructions executable by the processor 910 , data to be calculated by the processor 910 , or data processed by the processor 910 . Memory 920 may include non-transitory computer-readable media, such as high-speed random access memory and/or non-volatile computer-readable storage media (eg, one or more disk storage devices, flash memory devices, or other non-volatile solid state memory devices). may include

The communicator 930 provides an interface for communicating with an external device (eg, a diagnostic prediction device). The display 950 may display a screen on which a chatbot program for collecting questionnaire data is executed. Also, the user input interface 940 may receive a user input and a user selection input by a user. For example, the user input interface 940 may be a touch screen, a keyboard, a mouse, or a voice recognition interface.

The processor 910 may control the user terminal 900 so that the user terminal 900 may perform one or more operations related to the operation of the user terminal described with reference to FIG. 2 .

For example, the processor 910 executes a program for the user terminal 900 to collect the questionnaire data, and through the program, the image data including the user input and symptoms for the query data provided to the user and symptoms The user terminal 900 may be controlled to receive a user selection for at least one of the image data including the occurrence site information. The processor 910 determines the next query data based on the user input when the user terminal 900 provides predetermined current query data to the user through the chatbot program, and receives a user input corresponding to the current query data, and , the user terminal 900 may be controlled to provide the following query data to the user. In addition, the processor 910 allows the user terminal 900 to use the chatbot program, at least one of image data including symptoms for at least one of primary and secondary skin diseases and image data including information on the site of occurrence of symptoms. may be provided to the user, and the user terminal 900 may be controlled to receive a user selection for the provided image data.

The processor 910 transmits the questionnaire data based on the user input and the user selection received by the user terminal 900 to the diagnosis prediction apparatus, and provides the diagnosis data to the user when the diagnosis data is received from the diagnosis prediction apparatus. The user terminal 900 may be controlled.

10 is a diagram illustrating a configuration of a diagnostic prediction apparatus according to another embodiment.

Referring to FIG. 10 , the diagnostic prediction apparatus 1000 may perform a diagnostic prediction method according to another exemplary embodiment. The diagnosis prediction apparatus 1000 according to another embodiment may perform all processes of collecting questionnaire data, determining diagnosis data based on the collected questionnaire data, and providing the determined diagnosis data to the user. That is, the diagnostic prediction apparatus 1000 according to another exemplary embodiment may perform both the diagnostic prediction apparatus according to the exemplary embodiment and the diagnostic prediction method performed by the user terminal.

The diagnostic prediction apparatus 1000 according to another embodiment may include a processor 1010 , a memory 1020 , and a storage 1030 . Also, according to an embodiment, the diagnostic prediction apparatus 1000 may further include a user input interface 1040 and a display 1050 .

The memory 1020 is connected to the processor 1010 and may store instructions executable by the processor 1010 , data to be calculated by the processor 1010 , or data processed by the processor 1010 . Memory 1020 may include non-transitory computer-readable media, such as high-speed random access memory and/or non-volatile computer-readable storage media (eg, one or more disk storage devices, flash memory devices, or other non-volatile solid state memory devices). may include

The storage 1030 may store data necessary for the diagnostic prediction apparatus 1000 to perform the diagnostic prediction method. For example, the storage 1030 may store setting information learned for the diagnostic prediction model accurately predicting the diagnostic data, and the questionnaire data collected in the diagnostic prediction method process performed by the diagnostic prediction apparatus 1000 so far. . The display 1050 may display a screen on which a chatbot program for collecting questionnaire data is executed. Also, the user input interface 1040 may receive a user input and a user selection input by a user. For example, the user input interface 1040 may be a touch screen, a keyboard, a mouse, or a voice recognition interface.

The processor 1010 may control the diagnostic prediction apparatus 1000 so that the diagnostic prediction apparatus 1000 may perform one or more operations related to the operation of the user terminal described with reference to FIG. 4 .

For example, the processor 1010 executes a program for the diagnostic prediction apparatus 1000 to collect the questionnaire data, and image data and symptoms including a user input and symptoms for query data provided to the user through the program Receives a user selection of at least one of image data including information on the occurrence site of The diagnosis prediction apparatus 1000 may be determined to provide the determined diagnosis data to the user.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

The software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

As described above, although the embodiments have been described with reference to the limited drawings, those skilled in the art may apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

110, 800: diagnostic prediction device 120: network
130, 900: user terminal 720: diagnostic prediction model
730, 840: database 810: communicator
820: processor 830: memory
910: processor 920: memory
930: Communicator 940: User input interface
950: display 1000: diagnostic prediction device
1010: processor 1020: memory
1030: storage 1040: user input interface
1050: display

Claims (22)

In the method of providing diagnostic data for a skin disease in which each step is performed by a diagnostic prediction device,
Receiving, from the user terminal, the questionnaire data obtained based on the tree structure algorithm;
determining diagnostic data for the skin disease using a neural network model to which the questionnaire data is input; and
transmitting the diagnostic data to the user terminal;
The user terminal is
When the user is provided with predetermined current query data related to the skin disease through the chatbot program, and a user input corresponding to the current query data is received, based on the user input corresponding to the current query data, the skin disease determining next query data including a more detailed query than the current query data for , and providing the next query data to the user,
The user terminal is
providing query data including image data including image data including the shape and pattern of the skin lesion and image data including information on the area where the skin lesion occurred, and receiving a user selection for the provided image data;
The user terminal provides a description of each image data provided through the chatbot program,
The step of determining the diagnostic data for the skin disease,
Determining the diagnostic data based on the questionnaire data including user selection information for image data including the symptoms of at least one of the primary and secondary outbreaks of the skin disease and the image data including information on the site of occurrence of the symptoms including,
The questionnaire data is
Includes user response information to questions about skin rash pattern, accompanying skin symptoms, accompanying systemic symptoms, distribution patterns of skin rashes, skin disease history, family history of skin diseases, medication history, allergic disease factors, and infectious disease factors do,
The diagnostic data is
The predicted diagnosis name for the user's skin disease, a list of image data including the symptoms for the predicted diagnosis name skin disease, other symptoms that may be caused by the predicted diagnosis name skin disease, and information on treatment methods containing,
Way. delete delete delete delete According to claim 1,
The questionnaire data is
Including a user input for at least one of the user's gender, age, underlying disease, occupation, hobbies, and cosmetics used,
Way. delete delete According to claim 1,
The neural network model is
It is learned using at least one of the questionnaire data for each skin disease and the image data of the lesion,
Way. 10. The method of claim 9,
The paperweight data, data cleaning is performed, text is recognized,
The image data, image segmentation is performed, features are detected,
Way. A method of providing diagnostic data for a skin disease performed in a user terminal, the method comprising:
executing a program for collecting questionnaire data;
receiving, through the program, a user input with respect to the query data provided to the user and a user selection of at least one of image data including symptoms and image data including information on the site of occurrence of symptoms;
transmitting the questionnaire data based on the received user input and user selection to a diagnostic prediction device; and
providing the diagnostic data to a user when receiving diagnostic data from the diagnostic prediction device;
The step of receiving the user selection input,
providing the user with predetermined current query data related to the skin disease through a chatbot program;
When a user input corresponding to the current query data is received, determining next query data including a more detailed query than the current query data on the skin disease based on the user input corresponding to the current query data ; and
providing the next query data to the user;
The step of providing the current query data to the user comprises:
Providing query data including image data including the shape and pattern of the skin lesion and image data including information on the area where the skin lesion occurred,
The step of receiving the user selection input,
Comprising the step of receiving a user selection for the provided image data,
The user terminal provides a description of each image data provided through the chatbot program,
The diagnostic prediction device comprises:
determining the diagnosis data based on the questionnaire data including user selection information for image data including the symptom of at least one of the primary and secondary outbreaks of the skin disease and the image data including information on the site of occurrence of the symptoms,
The questionnaire data is
Includes user response information to questions about skin rash pattern, accompanying skin symptoms, accompanying systemic symptoms, distribution patterns of skin rashes, skin disease history, family history of skin diseases, medication history, allergic disease factors, and infectious disease factors do,
The diagnostic data is
The predicted diagnosis name for the user's skin disease, a list of image data including the symptoms for the predicted diagnosis name skin disease, other symptoms that may be caused by the predicted diagnosis name skin disease, and information on treatment methods containing,
Way. delete delete delete 12. The method of claim 11,
The diagnostic prediction device comprises:
A method of determining diagnostic data for the skin disease by using a neural network model to which the questionnaire data is input. delete A computer-readable storage medium storing instructions for executing the method of any one of claims 1, 6, 9 to 11, and 15. A diagnostic prediction apparatus for performing a method for predicting a diagnosis of a skin disease, comprising:
memory and processor;
the memory stores instructions executable by the processor;
When the instructions are executed by the processor, the processor determines that the diagnostic prediction device is
Receive the questionnaire data obtained based on the tree structure algorithm from the user terminal,
Determining diagnostic data for the skin disease using a neural network model to which the questionnaire data is input,
controlling the diagnostic prediction device to transmit the diagnostic data to the user terminal;
The user terminal is
When the user is provided with predetermined current query data related to the skin disease through the chatbot program, and a user input corresponding to the current query data is received, based on the user input corresponding to the current query data, the skin disease determining next query data including a more detailed query than the current query data for , and providing the next query data to the user,
The user terminal is
providing query data including image data including image data including the shape and pattern of the skin lesion and image data including information on the area where the skin lesion occurred, and receiving a user selection for the provided image data;
The user terminal provides a description of each image data provided through the chatbot program,
The processor includes the diagnostic prediction device,
Control to determine the diagnostic data based on the questionnaire data including user selection information for image data including symptoms of at least one of the primary and secondary outbreaks of skin diseases and image data including information on the site of occurrence of symptoms do,
The questionnaire data is
Includes user response information to questions about skin rash pattern, accompanying skin symptoms, accompanying systemic symptoms, distribution patterns of skin rashes, skin disease history, family history of skin diseases, medication history, allergic disease factors, and infectious disease factors do,
The diagnostic data is
The predicted diagnosis name for the user's skin disease, a list of image data including the symptoms for the predicted diagnosis name skin disease, other symptoms that may be caused by the predicted diagnosis name skin disease, and information on treatment methods containing,
Diagnostic prediction device. A user terminal for performing a method for predicting a diagnosis of a skin disease, the user terminal comprising:
memory and processor;
the memory stores instructions executable by the processor;
When the instructions are executed by the processor, the processor causes the user terminal to
Run the program for collecting questionnaire data,
Through the program, a user input for the query data provided to the user and a user selection for at least one of image data including symptoms and image data including information on the site of occurrence of symptoms are inputted,
Transmitting the questionnaire data based on the received user input and user selection to a diagnostic prediction device,
when receiving diagnostic data from the diagnostic prediction device, control to provide the diagnostic data to a user;
The processor, the user terminal
When providing the user with predetermined current query data related to the skin disease through the chatbot program, and receiving a user input corresponding to the current query data, based on the user input corresponding to the current query data, the skin determining next query data including a more detailed query than the current query data about the disease, and controlling to provide the next query data to the user;
The processor, the user terminal
Provides query data including image data including image data including the shape and pattern of skin lesions and image data including information on the area where the skin lesion occurs, and controls to receive a user selection for the provided image data,
The user terminal provides a description of each image data provided through the chatbot program,
The diagnostic prediction device comprises:
determining the diagnosis data based on the questionnaire data including user selection information for image data including the symptom of at least one of the primary and secondary outbreaks of the skin disease and the image data including information on the site of occurrence of the symptoms,
The questionnaire data is
Includes user response information to questions about skin rash pattern, accompanying skin symptoms, accompanying systemic symptoms, distribution patterns of skin rashes, skin disease history, family history of skin diseases, medication history, allergic disease factors, and infectious disease factors do,
The diagnostic data is
The predicted diagnosis name for the user's skin disease, a list of image data including the symptoms for the predicted diagnosis name skin disease, other symptoms that may be caused by the predicted diagnosis name skin disease, and information on treatment methods containing,
user terminal. delete delete delete

Images