KR20230024233A - Method and apparatus for estimating cause of allegic contact dermatits using user questionnaire data and skin image data - Google Patents

Abstract

A method and apparatus for estimating the cause of contact dermatitis by using user medical interview data and skin image data are disclosed. The method for estimating the cause of contact dermatitis may comprise the steps of: receiving skin image data for estimating the cause of a skin disease from a user terminal; receiving user medical interview data related to the skin disease; acquiring result data including information on the cause of the skin disease from the skin image data and the user medical interview data by using a learned deep learning model; and providing the result data to the user terminal, wherein the result data includes cause items of the skin disease and a probability value for each of the cause items.

Description

Method and apparatus for estimating cause of contact dermatitis using user medical examination data and skin image data

Embodiments described below relate to a technology for diagnosing skin diseases based on deep learning technology.

Skin disorders are generally caused by environmental factors along with other causes. However, methods necessary for early detection of skin diseases have not yet been developed. In order to solve this problem, active research is being conducted all over the world.

In order to find the cause of the itching or swelling of the patient's skin, information on history and topographic approach is important. For example, if a red rash accompanied by itchiness occurs on the forehead, ears, etc. of a patient who dyed a week ago, the doctor will determine whether the hair dye is the cause of dermatitis with information about the patient's history of dyeing and the area where the rash occurred. It can be inferred that

However, in order to confirm the presumed causative substance, a patch test should be performed 2 days after the causative substance is applied to the patient's skin and then 4-7 days later to see if reactions such as skin rash appear. do. The patch test takes about a week to confirm the results, and has limitations such as that the patient must be restricted from taking the treatment during the test. In addition, since the intelligence test is a cumbersome process, in practice, doctors in hospitals often guess and diagnose the cause of dermatitis only by visual inspection based on the patient's interview and the doctor's experience.

When allergic symptoms occur on a patient's skin, it is important to immediately predict the causative substance and have the patient avoid the causative substance, so there is a need for a method that can supplement the limitations of conventional examination methods. and the need is great.

A method for estimating a cause of contact dermatitis using user medical examination data and skin image data according to an embodiment includes receiving skin image data for estimating a cause of a skin disease from a user terminal; Receiving user medical examination data related to the skin disease; obtaining result data including information on a cause of the skin disease from the skin image data and the user medical examination data using a learned deep learning model; The method may include providing the result data to the user terminal, and the result data may include cause items of the skin disease and a probability value of each cause item.

The receiving of the user medical examination data may include identifying a body part having the skin disease from the skin image data or determining a body part having the skin disease according to the user's selection; providing predefined medical examination items corresponding to the body part where the skin disease appears through the user terminal; and acquiring user medical examination data including response information of the user corresponding to each of the medical examination items.

A cause estimation method according to an exemplary embodiment may include extracting a skin lesion area from the skin image data; and generating skin lesion image data in which the extracted skin lesion regions are separated, wherein the obtaining of the result data comprises the step of generating the skin lesion image data based on the skin image data, the skin lesion image data, and the user medical examination data. It may include acquiring result data.

The obtaining of the result data may include determining a weight representing a degree of influence on the result data for each of the skin image data and the user medical examination data; and obtaining the result data based on a result of applying the weight to the skin image data and the user medical examination data.

The user medical examination data includes user estimation information on cause items of the skin disease, and the obtaining of the result data includes weighting the skin disease cause factors based on the user estimation information on the skin disease items. It may include applying to the probability values of the causal items of .

An apparatus for estimating the cause of contact dermatitis using user medical examination data and skin image data according to an embodiment includes a memory and a processor, the memory stores instructions executable by the processor, and the instructions When executed by the processor, the processor receives skin image data for estimating the cause of a skin disease from a user terminal, receives user medical examination data related to the skin disease, and uses a learned deep learning model to determine the skin image data. Obtaining result data including information on the cause of the skin disease from skin image data and the user medical examination data, providing the result data to the user terminal, and the result data includes cause items and Possibility values of each cause item may be included.

The processor identifies a body part where the skin disease appears from the skin image data or determines a body part where the skin disease appears according to the user's selection, and determines a predefined body part corresponding to the skin disease It is possible to provide medical examination items through the user terminal, and obtain user medical examination data including user response information corresponding to each of the medical examination items.

The processor determines a weight representing a degree of influence on the result data for each of the skin image data and the user medical examination data, and based on a result of applying the weight to the skin image data and the user medical examination data, Result data can be obtained.

According to one embodiment, it is possible to provide a method for determining whether it is allergic contact dermatitis and estimating the cause of contact dermatitis when an itchy or swelling rash occurs, and an application related thereto.

According to one embodiment, it is possible to prevent the patient's dermatitis from worsening and treat the patient's dermatitis at an early stage by allowing the patient with symptoms of contact dermatitis to immediately avoid the causative agent.

According to an embodiment, the type of skin disease may be classified based on the medical examination data and the image data of the skin lesion, and diagnosis information regarding severity may be provided.

According to one embodiment, it is possible to prevent contact dermatitis or the like from recurring in the patient by providing the patient with information about a product or situation that can be exposed to a causative agent of skin lesions, such as contact dermatitis.

According to an embodiment, the level of subjective diagnosis results of medical personnel may be qualitatively improved through artificial intelligence skin disease diagnosis technology provided through a mobile device.

According to an embodiment, it is possible to diagnose skin diseases at an early stage and contribute to improving the health of users through artificial intelligence-based non-invasive diagnosis technology that improves the quality of medical services.

1 is a diagram showing the overall configuration of a skin disease diagnosis system according to an embodiment.
2 is a diagram for explaining a deep learning model for estimating the cause of contact dermatitis according to an embodiment.
3 is a flowchart illustrating the operation of a method for estimating the cause of contact dermatitis according to an embodiment.
4 is a diagram for explaining adjusting a diagnosis result based on user medical examination data according to an exemplary embodiment.
5 is a block diagram showing the configuration of an apparatus for estimating the cause of contact dermatitis according to an embodiment.
6 is a block diagram illustrating a configuration of a user terminal according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can 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 changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Terms used in the examples are used only for the purpose of explanation and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

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

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing 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 will be omitted.

1 is a diagram showing the overall configuration of a skin disease diagnosis system according to an embodiment.

Referring to FIG. 1 , the system for diagnosing skin diseases according to an embodiment may estimate the cause or causative substance of contact dermatitis based on image data showing skin lesions and user questionnaire data, and provide the information to the user. In some embodiments, the skin disease diagnosis system is capable of diagnosing multiple skin conditions (eg, acute and chronic urticaria, seborrheic keratosis, psoriasis, acne, shingles, viral warts, contact dermatitis, atopic dermatitis). The skin disease diagnosis system may include a user terminal 100 , a skin disease diagnosis server 110 , a data backup server 120 and a diagnoser terminal 130 . The skin disease diagnosis server 110 may be implemented as a cloud server and may include a service server 112 , an analysis server 114 , and a storage 116 . The skin disease diagnosis server 110 and the data backup server 120 may be replaced with a 'skin disease diagnosis device' and a 'data backup device', respectively, and the service server 112 and analysis server 114 are respectively 'service management' It can be replaced by 'device' and 'analysis device'. In this specification, the skin disease diagnosis server 110 may correspond to a device for estimating the cause of contact dermatitis (eg, the device for estimating the cause 500 of FIG. 5 ). In the present specification, 'diagnosis' of a skin disease may mean that a computing device analyzes input data (eg, user questionnaire data, image data) to generate diagnostic data for a skin disease.

The user terminal 100 is a terminal of a user who wants to diagnose a skin disease, and may be, for example, a smart phone, a tablet computer, a PDA, a personal computer, a laptop computer, or the like, but is not limited thereto. The user terminal 100 may transmit user data including user information (eg, age, gender, medical history, etc.), user medical examination data, and image data related to a skin disease to the skin disease diagnosis server 110 . The image data related to the skin disease may be skin image data showing skin lesions of the user. In one embodiment, the user of the user terminal 100 may input user data through an application program and create user questionnaire data. In addition, the user may generate image data by capturing a skin lesion area through a function provided by a corresponding application program.

The service server 112 may receive user data, user medical examination data, and image data from the user terminal 100 and store the received user data, user medical examination data, and image data in the storage 116 . In addition, the service server 112 may transmit a request signal for requesting an analysis of a skin disease requested by a user to the analysis server 114 . Upon receiving the request signal, the analysis server 114 retrieves data (user data, user medical examination data, image data) to be used for analysis from the storage 116, and analyzes the user's skin disease using the searched data. . The analysis server 114 may derive an analysis result by analyzing the user's skin disease based on the analysis model of the deep learning model. The analysis server 114 may provide, for example, analysis results regarding the type of skin disease of the user, possible causes of the skin disease, and the like.

The analysis model is, for example, a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted boltzmann machine (RBM), DBN ( It may be one of a deep belief network, a bidirectional recurrent deep neural network (BRDNN), a deep Q-network, or a combination of two or more of them, and is not limited to the above examples. At least part of the analysis model may be implemented in software, hardware including a neural processor, or a combination of software and hardware.

In one embodiment, the analysis model may be a model learned by a learning device. The learning device may update parameters of the analysis model (eg, weight and bias of the neural network) based on the result value output from the analysis model. For example, the learning device calculates a loss based on a difference between a result value output by inputting training data to an analysis model and a target value corresponding to the learning data, and sets parameters of the analysis model to reduce the loss. can be adjusted Various loss functions may be used to calculate the loss, and adjustments of the parameters may be made by, for example, a back propagation algorithm. The learning device may repeatedly perform this process for each of a large number of training data, and through this, the parameters of the analysis model may be gradually adjusted in a desirable direction. The learning device may train the analysis model using various machine learning algorithms in addition to the learning method described herein.

The analysis result data generated by the analysis server 114 may be transferred to the storage 116 and stored therein. The analysis result data stored in the storage 116 is transmitted to the service server 112 , and the service server 112 may provide the analysis result data to the user terminal 100 . For example, information on the type of skin disease, a list of suspected causes of skin disease, countermeasures for each cause of contact dermatitis, and dermatology to be recommended to the user may be transmitted to the user terminal 100 . The skin disease diagnosis server 110 may recommend a suitable dermatology clinic around the user through a location-based service, and with the consent of the user, user data, user questionnaire data, skin image data, and diagnosis result data may be sent to a medical institution of the dermatology department. can transmit The application program of the user terminal 100 may provide the corresponding information to the user. The user may be provided with information on skin disease and information on countermeasures through an application running on the user terminal 100 without visiting a hospital or meeting a doctor.

The skin disease diagnosis server 110 may perform data backup by transferring data used for skin disease diagnosis (eg, user data, user medical examination data, image data) and diagnosis result data to the data backup server 120 . The data backup server 120 may continuously update the analysis model used by the analysis server 114 .

Depending on the embodiment, the skin disease diagnosis server 110 may provide data (eg, user data, user medical examination data, image data) received from the user terminal 100 to the diagnoser terminal 130, which is a medical staff terminal. Alternatively, corresponding data may be directly transmitted from the user terminal 100 to the diagnoser terminal 130 . In addition, diagnosis result data may be transmitted from the skin disease diagnosis server 110 to the diagnoser terminal 130 . The diagnoser (eg, medical staff) of the diagnoser terminal 130 diagnoses the user's skin disease based on the received data, and provides the diagnosis result to the user of the user terminal 100 through the skin disease server 110 or Alternatively, the skin disease diagnosis result diagnosed by the skin disease diagnosis server 110 may be verified.

The user can perform self-monitoring for the cause of allergic contact dermatitis using the application program of the user terminal 100 through the above skin disease diagnosis system. In addition, a diagnoser such as a medical practitioner can more accurately diagnose a skin disease by using the diagnostic information on allergic contact dermatitis provided by the skin disease diagnosis system as an aid. In addition, through the skin disease diagnosis system described above, the user can quickly and accurately diagnose skin diseases.

2 is a diagram for explaining a deep learning model for estimating the cause of contact dermatitis according to an embodiment.

Referring to FIG. 2 , the cause estimating device (eg, the cause estimating device 500 of FIG. 5 ) may estimate the cause of allergic contact dermatitis based on user medical examination data 220 and skin image data 210. .

Depending on the embodiment, the cause estimation device may extract a skin lesion area from the skin image data 210 to generate skin lesion image data 215 in which the skin lesion area is distinguished, and the cause estimation device may extract the skin image data ( 210), the cause of contact dermatitis may be estimated based on the skin lesion image data 215 and the user questionnaire data 220. The cause estimation device may extract a skin lesion area from the skin image data 210 using the learned deep learning model to generate skin lesion image data 215 in which the skin lesion area is distinguished. Brightness information of each region may be obtained through the image skin image data 210 , and topological pattern information of the skin may be obtained through the skin lesion image data 215 . The deep learning model 200 may estimate the cause of contact dermatitis based on the obtained brightness information and phase pattern information.

In one embodiment, the user medical examination data 220 may be obtained from information input by the user through an application or site installed on the user terminal. For example, the user questionnaire data 220 may include the user's age, primary skin eruption, rash patterns (eg, round, oval, irregular shape), size, patch, skin It may include history data on whether the lesions are bilateral or unilateral, the color of the rash, and a personal history of inflammatory dermatosis. For example, age is one of the important features in contact skin diseases. In general, atopic dermatitis and acne vulgaris are prevalent in adolescence, but seborrheic keratosis is prevalent in old age, so information about age helps determine the type and cause of contact dermatitis.

In one embodiment, the application of the user terminal may provide the user with questions for obtaining the user questionnaire data 220 . For example, the application is 'time of symptoms of skin disease', 'current symptoms of affected area', 'past symptom history', 'feeling of affected area', 'family history of skin disease', 'presence of other diseases', 'Medicine in use', 'Possible cause related to allergic skin disease', 'Possible cause related to infectious skin disease', 'Primary skin rash', 'Rash within skin', 'Rash pattern', 'Shape of skin lesion border', ' Receive selection/input information from the user regarding 'accompanying skin symptoms', 'companing systemic symptoms', 'tone of skin lesions', 'number of skin lesions', 'size of skin lesions' and 'distribution pattern of skin rash'; , It is possible to determine the user medical examination data 220 based on the received selection / input information.

The cause estimation device may obtain a diagnosis result for the cause of contact dermatitis using the learned deep learning model 200 . The trained deep learning model 200 may be based on, for example, a convolutional neural network (CNN), a deep neural network, a recurrent neural network (RNN), a random forest, or the like. However, it is not limited thereto. Skin image data 210 (or additional skin lesion image data 215) and user medical examination data 220 are input to the deep learning model 200, and diagnostic data for the cause of contact dermatitis from the deep learning model 200 230 may be output. The diagnosis data 230 includes information on the cause of skin disease, and specifically, may include cause items of the skin disease and probability values for each cause item. The diagnostic data 230 may provide a list of items (eg, hair dye, cosmetics, ointments, accessories, etc.) that may correspond to the cause of allergic contact dermatitis and probability information that each item corresponds to the actual cause of contact dermatitis. .

The deep learning model 200 may be a model learned based on a dataset of supervised learning data or unsupervised learning data built based on existing patient data and open big data. The deep learning model 200 may be trained based on learning data including image data showing skin lesions, user medical examination data, and diagnosis result data corresponding to the image data. In the learning process of the deep learning model 200, parameters (eg, connection weights) of the deep learning model 200 are used to receive image data and user medical examination data and output diagnosis result data corresponding to the image data and user medical examination data. can be adjusted

The deep learning model 200 extracts features from the input skin image data 210, skin lesion image data 215, and user medical examination data 220, and generates diagnostic data 230 based on the extracted features. can The cause estimation device can quickly and accurately estimate the cause of allergic contact dermatitis from the patient's image data and user questionnaire data (additionally user data) to replace the existing patch test using the deep learning model 200.

The deep learning model 200 is, for example, a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted boltzmann machine (RBM) ), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), deep Q-networks, or a combination of two or more of them, and is not limited to the above examples. At least part of the analysis model may be implemented in software, hardware including a neural processor, or a combination of software and hardware.

In one embodiment, the deep learning model 200 may be a model learned by a learning device. The learning device may update parameters (eg, weights and biases of a neural network) of an analysis model based on result values output from the deep learning model 200 . For example, the learning device calculates a loss based on the difference between a result value output by inputting training data to the deep learning model 200 and a target value corresponding to the training data, and deep learning to reduce the loss. Parameters of the learning model 200 may be adjusted. Various loss functions may be used to calculate the loss, and adjustments of the parameters may be made by, for example, a back propagation algorithm. The learning device may repeatedly perform this process for each of a large number of training data, and through this, parameters of the deep learning model 200 may be gradually adjusted in a desirable direction. The learning device may train the deep learning model 200 using various machine learning algorithms in addition to the learning method described herein.

3 is a flowchart illustrating the operation of a method for estimating the cause of contact dermatitis according to an embodiment. The cause estimation method may be performed by a cause estimation device described herein (eg, the cause estimation device 500 of FIG. 5 ).

Referring to FIG. 3 , in step 310, the cause estimating device may receive skin image data for estimating the cause of a skin disease and user questionnaire data related to the skin disease from a user terminal.

According to an embodiment, the cause estimation apparatus may perform image preprocessing on skin image data in which skin lesions appear. For example, the cause estimating apparatus may obtain preprocessed image data by removing noise from the image data and adjusting the size of the image to a predetermined reference size.

The user may select corresponding items or input related contents to the paperweight items output to the application program of the user terminal (eg, the user terminal 100 of FIG. 1), and through this, user paperweight data may be generated. There is, and the generated user medical examination data may be transmitted to the cause estimation device. In the case of user medical examination data, medical examination items may be selected according to a body part indicated in the skin image data or a target body part according to a user's selection, and then provided to the user. The cause estimation device may identify a body part with a skin disease from skin image data or determine a body part with a skin disease according to a user's selection. The cause estimating device may provide predefined medical examination items corresponding to the body part where the skin disease appears through the user terminal, and obtain user medical examination data including user response information corresponding to each of the medical examination items. there is.

The user questionnaire data may include, for example, at least one of the user's job at which the skin lesion appears, behavior before the skin lesion appears to the user, the body part where the skin lesion appears, and the user's characteristics related to the skin lesion. . The characteristics of the user associated with the skin lesion may include information about whether the user has a history of travel and, if there is a history of travel, information about a travel destination. In addition, the user's characteristics related to skin lesions may cause itchiness or rash on the user's skin, such as the user's hobbies, whether the user dyes hair, whether the user wears accessories, whether the user uses perfume, and whether the user has recently changed cosmetics. It may include user specifics related to the element in the

In one embodiment, a shaping algorithm of the user medical examination data may be performed in the course of acquiring the user medical examination data. For example, a process of removing outliers from the questionnaire results or a process of selecting and providing clinically meaningful questionnaire items may be performed.

In step 320, the cause estimation device may obtain result data including information about the cause of the skin disease from the skin image data and the user medical examination data using the learned deep learning model. The result data may include cause items of the skin disease and probability values of each cause item.

In another embodiment, the cause estimating device determines a weight representing the degree of influence on the result data for each of the skin image data and the user medical examination data, and the result is obtained based on a result of applying the weight to the skin image data and the user medical examination data. data can be obtained. The cause estimating device may apply weights to the skin image data and the user medical examination data to determine which data to place more weight on. For example, if it is determined that there is little meaningful information in the user medical examination data, the cause estimating device controls to generate result data mainly based on the skin image data by setting the weight of the skin image data higher than that of the user medical examination data. can These weights may be determined through a separate user questionnaire data analysis model.

In another embodiment, the cause estimation device may extract a skin lesion area from skin image data and generate skin lesion image data in which the extracted skin lesion area is divided. For example, the cause estimation device may generate skin lesion image data by extracting a skin region from skin image data and extracting a region where a skin lesion appears from the extracted skin region. In this case, the cause estimation device may obtain result data based on skin image data, skin lesion image data, and user medical examination data. Skin image data, skin lesion image data, and user medical examination data are input to the trained deep learning model, and the deep learning model may output result data for allergic contact dermatitis diagnosis based on the input data.

Depending on the embodiment, the user medical examination data may include user estimation information on causes of skin diseases. For example, the user medical examination data may include a cause item selected by the user as having a high probability of corresponding to the cause of contact dermatitis or a cause item selected as not likely to correspond to the cause of contact dermatitis for each of the exemplary skin disease cause items. information may be included. In this case, the cause estimating apparatus may adjust the diagnosis result data by applying weights based on user estimation information on the cause items of the skin disease to the probability values of the cause items of the skin disease. For example, the cause estimation device may exclude a cause item selected by the user as having a high probability of not corresponding to the cause of contact dermatitis from the cause item of the result data or apply a weight to reduce the probability value of the corresponding cause item. The cause estimating device may apply a weight to increase the probability value of the cause item selected by the user as being highly likely to correspond to the cause of contact dermatitis.

In step 330, the cause estimation device may provide the result data obtained in step 320 to the user terminal. The user terminal may receive result data from the cause estimation device and provide the result data to the user through an application program. Through the result data, the user may be provided with information on the cause items of allergic contact dermatitis, the probability values for each cause item, and a countermeasure plan for each causative substance. Accordingly, the user can perform immediate treatment or prevention of allergic contact dermatitis.

As described above, the cause estimation device increases the accuracy of skin disease diagnosis through a configuration of estimating the cause of skin disease using user medical examination data and skin image data, and enables immediate treatment by providing information on the cause of skin disease to the user. It provides a beneficial effect that As a result, the device for estimating the cause provides excellent accessibility to patients suffering from allergic contact dermatitis and enables quick treatment, thereby reducing medical expenses and freeing them from skin diseases, leading to a high quality of life.

4 is a diagram for explaining adjusting a diagnosis result based on user medical examination data according to an exemplary embodiment.

Referring to FIG. 4 , the cause estimation apparatus receives user medical examination data together with skin image data from a user terminal (410). The user medical examination data is the cause item selected by the user as likely to correspond to the cause of contact dermatitis or the cause item selected as not likely to correspond to the cause of contact dermatitis for each of the exemplary skin disease cause items. It may include user estimation information about the cause item indicating information about

The cause estimation device analyzes the received user medical examination data (420). The cause estimating device determines (430) the weight to be applied to each cause item based on the user estimation information included in the user medical examination data, and uses the determined weight to estimate the cause items of the user's contact dermatitis and the probability value of each cause item. Result data including may be applied (440). For example, the cause estimation device excludes a cause item selected by the user as having a high probability of not corresponding to the cause of contact dermatitis from the cause item of the result data, or applies a weight to reduce the probability value of the corresponding cause item to the result data, and the user For the cause item selected as likely to correspond to the cause of contact dermatitis, a weight to increase the probability of the corresponding cause item may be applied to the result data.

As the weight is applied to the result data, the cause item list or the probability value for each cause item may be modified, and the cause estimation device may provide the corrected result data to the user terminal (450).

5 is a block diagram showing the configuration of an apparatus for estimating the cause of contact dermatitis according to an embodiment.

Referring to FIG. 5, the cause estimation device 500 is a computing device for estimating the cause of allergic contact dermatitis based on skin image data and user questionnaire data provided by the user, and the skin disease diagnosis server 110 of FIG. can respond to The cause estimating device 500 may include a processor 510, a memory 520, and a communication device 530, and each component of the cause estimating device 500 may communicate with each other through a communication bus 540. there is.

The processor 510 controls the overall operation of the cause estimating device 500 . The processor 510 may include one or a plurality of processors, and the one or plurality of processors may include a general-purpose processor such as a central processing unit (CPU), an application processor (AP), a digital signal processor (DSP), a graphic processor (GPU), and the like. unit), a graphics-only processor such as a vision processing unit (VPU), or a neural processing unit (NPU).

Memory 520 stores information necessary for processor 510 to perform processing operations. For example, the memory 520 may store instructions executable by the processor 510 and a deep learning model, and may store related information while software or programs are executed in the cause estimation device 500 . Memory 520 may include volatile memory, such as RAM, DRAM, SRAM, and/or non-volatile memory known in the art, such as flash memory.

The processor 510 may control the cause estimation device 500 to perform one or more operations described above through FIGS. 1 to 4 by executing executable instructions stored in the memory 520 . For example, the processor 510 may receive skin image data for estimating the cause of a skin disease and user medical examination data related to a skin disease from a user terminal. In one embodiment, the processor 510 may identify the body part where the skin disease appears from the skin image data or determine the body part where the skin disease appears according to the user's selection. In this case, the processor 510 provides predefined medical examination items corresponding to the body part where the skin disease appears through the user terminal, and obtains user medical examination data including user response information corresponding to each of the medical examination items. can do.

The processor 510 may obtain result data including information on the cause of a skin disease from skin image data and user questionnaire data using the learned deep learning model. The result data may include cause items of the skin disease and probability values of each cause item.

In one embodiment, the processor 510 determines a weight representing the degree of influence on the result data for each of the skin image data and the user questionnaire data, and the result is based on a result of applying the weight to the skin image data and the user questionnaire data. data can be obtained.

In another embodiment, the user medical examination data may include user estimated information on cause items of the skin disease, and the processor 510 assigns a weight based on the user estimated information on the cause items of the skin disease. It can be applied to the probability number of items. For example, the user medical examination data may include a cause item selected by the user as having a high probability of corresponding to the cause of contact dermatitis or a cause item selected as not likely to correspond to the cause of contact dermatitis for each of the exemplary skin disease cause items. information may be included. The processor 510 may adjust diagnosis result data by applying weights based on user estimation information on cause items of skin diseases to probability values of cause items of skin diseases. For example, the processor 510 may exclude a cause item selected by the user as having a high probability of not corresponding to the cause of contact dermatitis from the cause item of the result data or apply a weight to reduce the probability value of the corresponding cause item. The processor 510 may apply a weight to increase the probability value of the cause item selected by the user as having a high probability of being the cause of contact dermatitis.

The processor 510 may provide result data of the contact dermatitis diagnosis obtained through the above process to the user terminal.

The communication device 530 may support establishment of a direct (eg, wired) communication channel or wireless communication channel between the cause estimation device 500 and an external device, and communication through the established communication channel. According to an embodiment, the communication device 530 is a wireless communication module (eg, a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (eg, a local area network (LAN)). ) communication module, or power line communication module). A wireless communication module may be used in a short-distance communication network (eg Bluetooth, wireless fidelity (WiFi) direct or infrared data association (IrDA)) or a long-distance communication network (eg legacy cellular network, 5G network, next-generation communication network, the Internet, or a computer network (eg Ex: LAN or WAN)) can communicate with external devices.

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

Referring to FIG. 6 , a user terminal 600 may correspond to the user terminal described herein (eg, the user terminal 100 of FIG. 1 ). The user terminal 600 may include a processor 610 , a storage device 620 and a communicator 630 . Elements of the user terminal 600 may communicate with each other through the communication bus 635 . Also, the user terminal 600 may further include a user input interface 640, a display 650, and a camera 660 according to embodiments.

The storage device 620 is connected to the processor 610 and may store instructions executable by the processor 610, data to be calculated by the processor 610, or data processed by the processor 610. For example, storage device 620 may include non-transitory computer readable media such as high-speed random access memory and/or non-volatile computer readable storage media (e.g., one or more disk storage devices, flash memory devices, or other non-volatile solid state media). state memory device), and an optical recording medium.

The communicator 630 provides an interface for communication with external devices (eg, the cause estimation device 500 and the skin disease server 110). For example, the communicator 630 may communicate with an external device through a wired or wireless network. The communicator 630 may transmit skin image data and user medical examination data to the cause estimation device, and may receive result data for skin disease diagnosis from the cause estimation device. According to an embodiment, the communicator 630 may be a wireless communication module (eg, a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (eg, a local area network (LAN) communication module). module, or a power line communication module). A wireless communication module may be used in a short-distance communication network (eg Bluetooth, wireless fidelity (WiFi) direct or infrared data association (IrDA)) or a long-distance communication network (eg legacy cellular network, 5G network, next-generation communication network, the Internet, or a computer network (eg Ex: LAN or WAN)) can communicate with external devices.

The display 650 may display a screen related to the user terminal 600 receiving cause analysis information of allergic contact dermatitis. For example, the display 650 may display a screen associated with executing an application program related to detecting the cause of allergic contact dermatitis and display result data received from the cause estimation device. The camera 660 may obtain skin image data showing skin lesions.

The user input interface 640 may receive an execution request for an application program input by a user and input data for a user questionnaire item. For example, the user input interface 640 may correspond to a touch screen, a keyboard, or a touch input device.

The processor 610 may control the user terminal 600 so that the user terminal 600 can perform one or more operations related to the operation of the user terminal 600 .

For example, the processor 610 may control the user terminal 600 to execute an application program, obtain user medical examination data through a user input, and control the user terminal 600 to acquire skin image data. . In addition, the processor 610 allows the user terminal 600 to transmit the skin image data and user medical examination data to the cause estimation device, and in response to the transmission, the user terminal to receive diagnosis result data for skin lesions from the cause estimation device. (600) can be controlled.

The user terminal 600 may perform one or more operations of the user terminal described above with reference to FIGS. 1 to 5 as it is, and duplicate descriptions will be omitted.

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 on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. - includes hardware devices specially configured to store and execute program instructions, such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. 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.

Software may include a computer program, code, instructions, or a combination of one or more of the foregoing, which configures a processing device to operate as desired or processes independently or collectively. You can command the device. Software and/or data may be any tangible machine, component, physical device, virtual equipment, or computer storage medium or It can be permanently or temporarily embodied in a device. Software may be distributed on networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer readable media.

As described above, although the embodiments have been described with limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques may be performed in an order different from the method described, and/or components of the described system, structure, device, circuit, etc. may be combined or combined in a different form than the method described, or other components may be used. Or even if it is replaced or substituted by equivalents, appropriate results can be achieved.

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

Claims (10)

In the method for estimating the cause of contact dermatitis using user medical examination data and skin image data,
Receiving skin image data for estimating the cause of a skin disease from a user terminal;
Receiving user medical examination data related to the skin disease;
obtaining result data including information on a cause of the skin disease from the skin image data and the user medical examination data using a learned deep learning model; and
Providing the resultant data to the user terminal;
The resulting data,
A method for estimating the cause of contact dermatitis, including the cause items of the skin disease and a probability value of each cause item. According to claim 1,
Receiving the user medical examination data,
identifying a body part where the skin disease appears from the skin image data or determining a body part where the skin disease appears according to the user's selection;
providing predefined medical examination items corresponding to the body part where the skin disease appears through the user terminal; and
Obtaining user medical examination data including user response information corresponding to each of the medical examination items
A method for estimating the cause of contact dermatitis comprising a. According to claim 1,
extracting a skin lesion area from the skin image data; and
Generating skin lesion image data in which the extracted skin lesion area is divided
Including more,
Obtaining the result data,
obtaining the result data based on the skin image data, the skin lesion image data, and the user medical examination data;
A method for estimating the cause of contact dermatitis comprising a. According to claim 1,
Obtaining the result data,
determining a weight representing a degree of influence on the result data for each of the skin image data and the user medical examination data; and
Obtaining the result data based on a result of applying the weight to the skin image data and the user medical examination data
A method for estimating the cause of contact dermatitis comprising a. According to claim 1,
The user questionnaire data,
Including user estimation information for cause items of the skin disease,
Obtaining the result data,
Applying weights based on user estimated information on the cause items of the skin disease to probability values of the cause items of the skin disease
A method for estimating the cause of contact dermatitis comprising a. In the apparatus for estimating the cause of contact dermatitis using user medical examination data and skin image data,
including memory and processor;
the memory stores instructions executable by the processor;
When the instructions are executed by the processor, the processor:
Receiving skin image data for estimating the cause of a skin disease from a user terminal;
Receiving user medical examination data related to the skin disease;
Obtaining result data including information on the cause of the skin disease from the skin image data and the user medical examination data by using the learned deep learning model;
Providing the resultant data to the user terminal;
The result data includes cause items of the skin disease and probability values of each cause item,
An apparatus for estimating the cause of contact dermatitis. According to claim 6,
the processor,
Identifying the body part where the skin disease appears from the skin image data or determining the body part where the skin disease appears by the user's selection;
Providing predefined medical examination items corresponding to the body part where the skin disease appears through the user terminal,
Obtaining user questionnaire data including user response information corresponding to each of the questionnaire items,
Apparatus for estimating cause of contact dermatitis According to claim 6,
the processor,
Determining a weight representing a degree of influence on the result data for each of the skin image data and the user medical examination data;
Obtaining the result data based on a result of applying the weight to the skin image data and the user medical examination data,
An apparatus for estimating the cause of contact dermatitis. According to claim 6,
The user questionnaire data,
Including user estimation information for cause items of the skin disease,
the processor,
Applying weights based on user estimated information on the cause items of the skin disease to the probability values of the cause items of the skin disease,
An apparatus for estimating the cause of contact dermatitis. According to claim 6,
the processor,
extracting a skin lesion area from the skin image data;
Generating skin lesion image data in which the extracted skin lesion area is divided;
Obtaining the result data based on the skin image data, the skin lesion image data, and the user medical examination data;
An apparatus for estimating the cause of contact dermatitis.

Images