Medical service solution for switching consultation mode based on artificial intelligence-based symptom assessment results

Abstract

The present disclosure relates to a method of operation of a server. The method of operation includes the steps of: obtaining patient symptom information based on user input received through a patient terminal; identifying the urgency of medical treatment based on data output by inputting the symptom information into an evaluation model for identifying the urgency of medical treatment; and selecting a consultation mode for the patient's treatment among a chatbot consultation mode, a remote consultation mode, and a face-to-face consultation mode based on the urgency.

Description

Technology Field

[0001] The present disclosure relates to a medical service provision solution that switches a consultation mode based on the results of an artificial intelligence-based symptom assessment, and more specifically, to a method of operation of a server that selects a consultation mode for a patient's treatment according to the urgency of treatment based on the patient's symptoms. Background Technology

[0002] While modern healthcare aims to provide convenient and efficient patient-centered services, problems arise where patients find it difficult to determine which medical services to use for their symptoms or where access to necessary medical services is cut off.

[0003] For example, there are many cases where it is difficult for patients to visit a hospital in person, or where it is hard to determine based on initial symptoms alone whether an in-person consultation is necessary or if remote consultation is sufficient. Furthermore, since a patient's health issues cannot be completely resolved with a single consultation, the importance of continuous monitoring after treatment and the ability to switch services as needed is increasing.

[0004] Existing medical service systems are primarily based on patients directly selecting and booking medical institutions, and chatbots or remote consultations often remain limited to providing information or basic medical care, while intelligent judgment and dynamic switching functions for service types based on changes in the patient's condition are rarely provided. Prior art literature

[0005] Registered Patent Publication No. 10-2314332 The problem to be solved

[0006] The present disclosure aims to provide a method of operation for a server that selects a consultation mode for patient treatment according to the urgency of treatment based on the patient's symptoms.

[0007] The purposes of the present disclosure are not limited to those mentioned above, and other purposes and advantages of the present disclosure not mentioned may be understood from the following description and will be more clearly understood from the embodiments of the present disclosure. Furthermore, it will be readily apparent that the purposes and advantages of the present disclosure can be realized by the means and combinations thereof set forth in the claims. means of solving the problem

[0008] A method of operation of a server according to one embodiment of the present disclosure includes the steps of: obtaining symptom information of the patient based on user input received through the patient's terminal; identifying the urgency of the medical treatment based on data output by inputting the symptom information into an evaluation model for identifying the urgency of the medical treatment; and selecting a consultation mode for the medical treatment of the patient among a chatbot consultation mode, a remote consultation mode, and a face-to-face consultation mode based on the urgency.

[0009] Here, the step of obtaining symptom information of the patient comprises providing at least one basic query for identifying symptoms to the patient's terminal, obtaining a basic response to the provided basic query from the patient's terminal, and inputting the basic response into an analysis model for identifying symptom types to identify the symptom information including at least one symptom type for the patient based on the output data.

[0010] At this time, the step of identifying the urgency of the medical treatment can identify the urgency of the medical treatment according to the patient's symptoms based on the output data obtained by acquiring the patient's biometric data and medical data including the patient's medical history and inputting the symptom information, the biometric data, and the medical data into the evaluation model.

[0011] Meanwhile, the step of selecting a consultation mode for the treatment of the patient may be to select the chatbot consultation mode if the urgency is less than a preset first standard urgency, select the remote consultation mode if the urgency is greater than or equal to the first standard urgency and less than a second standard urgency greater than the first standard urgency, and select the face-to-face consultation mode if the urgency is greater than or equal to the second standard urgency.

[0012] Meanwhile, the method of operation of the above server may include, when the chatbot consultation mode is selected, providing at least one deep query for diagnosis according to symptoms to the patient's terminal; obtaining a deep response to the deep query from the patient's terminal; inputting the obtained deep response into a diagnostic model trained based on the medical history of doctors to identify diagnostic information regarding the symptom information based on the output data; and providing the identified diagnostic information to the patient's terminal.

[0013] Meanwhile, the method of operation of the above server may include the steps of: obtaining consent information regarding whether to perform remote counseling according to the remote counseling mode from the patient's terminal when the remote counseling mode is selected; selecting a medical staff terminal to perform the remote counseling when it is identified that the patient's terminal has consented to the remote counseling according to the consent information; and providing code information to the patient's terminal and the selected medical staff terminal to connect to a video chat channel for performing the remote counseling with the selected medical staff terminal.

[0014] Meanwhile, the method of operation of the above server may include the step of providing a page to the patient's terminal to schedule a face-to-face consultation with at least one medical professional when the face-to-face consultation mode is selected. Effects of the invention

[0015] According to the present disclosure, by performing linkage and switching between medical services (chatbot, remote, face-to-face) through evaluation and diagnosis based on an artificial intelligence model, patients can conveniently and quickly access the medical service best suited to their condition anytime and anywhere.

[0016] In addition, according to the present disclosure, AI precisely evaluates a patient's initial symptoms and hyper-personalized data, and by providing diagnostic information that reflects the clinical knowledge of renowned doctors, it improves diagnostic accuracy, reduces the workload of medical staff, and enables the efficient distribution of medical resources.

[0017] In addition, according to the present disclosure, by monitoring the patient's condition in real time even after switching to a consultation mode, the patient's safety can be ensured by preventing disease exacerbation and strengthening the ability to respond to emergency situations. Brief explanation of the drawing

[0018] FIG. 1 is a flowchart for explaining the operation of a server according to one embodiment of the present disclosure. FIG. 2 is a diagram illustrating the operation of a server communicating with a medical record server and a biometric measurement device according to one embodiment of the present disclosure. FIG. 3 is a diagram illustrating the operation of a server communicating with a patient's terminal according to one embodiment of the present disclosure. FIG. 4 is a diagram illustrating the operation of a server communicating with a patient's terminal, a medical staff's terminal, and a video chat server according to one embodiment of the present disclosure. FIG. 5 is a block diagram illustrating the configuration of a server according to one embodiment of the present disclosure. Specific details for implementing the invention

[0019] Before specifically describing the present disclosure, the method of description in the specification and drawings is described.

[0020] First, the terms used in this specification and claims have been selected based on general terms considering their functions in the various embodiments of this disclosure. However, these terms may vary depending on the intent of those skilled in the art, legal or technical interpretations, and the emergence of new technologies. Additionally, some terms have been arbitrarily selected by the applicant. Such terms may be interpreted according to the meanings defined in this specification; in the absence of specific definitions, they may be interpreted based on the overall content of this specification and common technical knowledge in the relevant field.

[0021] In addition, the same reference numbers or symbols described in each drawing attached to this specification represent parts or components that perform substantially the same function. For convenience of explanation and understanding, the same reference numbers or symbols are used to describe different embodiments. That is, even if components having the same reference number are all depicted in multiple drawings, the multiple drawings do not imply a single embodiment.

[0022] Additionally, in this specification and claims, terms including ordinal numbers, such as "first," "second," etc., may be used to distinguish between components. These ordinal numbers are used to distinguish identical or similar components from one another, and the meaning of the terms should not be limited by the use of such ordinal numbers. For example, the order of use or arrangement of components combined with such ordinal numbers should not be restricted by the number. If necessary, each ordinal number may be used interchangeably.

[0023] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "consisting of" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0024] In the embodiments of the present disclosure, terms such as "module," "unit," "part," etc. are used to refer to a component that performs at least one function or operation, and such component may be implemented in hardware or software, or in a combination of hardware and software. Additionally, a plurality of "modules," "units," "parts," etc. may be integrated into at least one module or chip and implemented as at least one processor, except where each needs to be implemented in specific individual hardware.

[0025] Furthermore, in the embodiments of the present disclosure, when a part is described as being connected to another part, this includes not only a direct connection but also an indirect connection through another medium. Additionally, the meaning that a part includes a certain component implies that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0026] In the present disclosure, the expression “at least one of a, b, or c” may refer to “a”, “b”, “c”, “a and b”, “a and c”, “b and c”, “a, b, and c all”, or variations thereof.

[0027] The terms used in this disclosure have been selected to be as widely used and general as possible, taking into account their functions within this disclosure; however, these terms may vary depending on the intent of those skilled in the art, case law, the emergence of new technologies, etc. Additionally, in specific cases, terms have been selected at the applicant's discretion, and in such cases, their meanings will be described in detail in the relevant explanatory sections. Therefore, terms used in this disclosure should be defined not merely by their names, but based on their meanings and the overall content of this disclosure.

[0028] Singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as generally understood by those skilled in the art as described in this specification. Additionally, terms including ordinal numbers, such as "first" or "second," used in this specification may be used to describe various components, but said components should not be limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another.

[0029] When a part of a specification is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Furthermore, terms such as "part" or "module" as used in the specification refer to a unit that processes at least one function or operation, and this may be implemented in hardware or software, or as a combination of hardware and software.

[0030] Embodiments of the present disclosure are described below with reference to the attached drawings so that those skilled in the art can easily implement the invention. However, the present disclosure may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present disclosure in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0031] FIG. 1 is a flowchart for explaining the operation of a server according to one embodiment of the present disclosure.

[0032] The server (100) can select a consultation mode for the user based on the user's symptoms.

[0033] At this time, the server (100) can obtain symptom information of the user based on the consultation content obtained by performing consultation with the user through the chatbot, and can select a consultation mode for the user based on the symptom information.

[0034] Here, the consultation mode may include a chatbot consultation mode, a remote consultation mode, and a face-to-face consultation mode.

[0035] Here, the chatbot consultation mode is a consultation mode in which a patient obtains diagnostic information from a diagnostic model (20) through question-and-answer based on a diagnostic model to be described later, the remote consultation mode is a consultation mode in which a patient receives a diagnosis from a medical professional through video chat between the patient's terminal and the medical professional's terminal (500), and the face-to-face consultation mode is a consultation mode in which a patient visits a medical institution to receive a diagnosis.

[0036] To this end, the server (100) may provide at least one web page, UI (User Interface), code information (e.g., QR code, URL, etc.) for consultation according to the consultation mode to at least one terminal that has requested consultation (question and answer) for medical treatment.

[0037] In one embodiment, when a chatbot consultation mode is selected, the server (100) may provide at least one UI for inputting and providing questions and / or responses to the patient's terminal (200).

[0038] In addition, as an example, when a remote consultation mode is selected, the server (100) may provide code information to the patient's terminal (200) for accessing a video chat channel for remote consultation.

[0039] Additionally, in one embodiment, when a face-to-face consultation mode is selected, the server (100) can provide at least one web page for scheduling a medical appointment for a face-to-face consultation to the patient's terminal (200).

[0040] Referring to FIG. 1, the server (100) can obtain the patient's symptom information (S110).

[0041] In one embodiment, the server (100) provides at least one basic query to the patient's terminal (200) to identify symptoms, and obtains a basic response to the basic query from the patient's terminal (200) to identify the patient's symptom information.

[0042] Specifically, the server (100) can provide at least one basic query that is pre-set to the patient's terminal (200) to identify symptoms.

[0043] For example, the basic inquiry may include a sentence requesting an explanation of the patient's symptoms.

[0044] Afterward, the server (100) can obtain a basic response to a basic query from the patient's terminal (200). Here, the basic response may include a description of the symptoms, duration, etc.

[0045] Accordingly, the server (100) can identify the patient's symptom information based on the basic response.

[0046] In one embodiment, the server (100) can identify symptom information including at least one symptom type for a patient based on data output by inputting a response based on an analysis model for identifying symptom types.

[0047] Here, types of symptoms may include, but are not limited to, headache, abdominal pain, muscle pain, chest pain, back pain, delirium, amnesia, vomiting, dizziness, cold sweats, fatigue, phlegm, cough, bleeding, etc.

[0048] The analysis model may be a model trained based on Natural Language Processing (NLP) to identify at least one symptom type and the time of onset or duration of the symptom type from text in which symptoms are described in natural language.

[0049] To this end, the analysis model may correspond to a model trained based on training data consisting of multiple training texts in which symptoms are described in natural language and entity labels such as symptoms, onset time, and duration included in each training text.

[0050] Here, the analysis model may be a model trained based on various learning algorithms such as CRF (Conditional Random Field), BERT (Bidirectional Encoder Representations from Transformers), SVM (Support Vector Machine), RNN (Recurrent Neural Network), LSTM (Long Short-Term Memory), and GRU (Gated Recurrent Unit), but is not limited thereto.

[0051] The server (100) can identify the urgency of medical treatment for a patient based on the output data obtained by inputting symptom information into an evaluation model (10) for identifying the urgency of medical treatment (S120).

[0052] Here, an urgency can be identified that has a larger value as the need for rapid medical treatment increases. For example, the urgency may have a value between 0 and 100, but is not limited thereto.

[0053] For example, the evaluation model (10) may be a model trained to identify the urgency according to the symptoms.

[0054] Specifically, the evaluation model (10) can be trained to identify the urgency according to the symptoms, while reflecting the correlation between the symptoms and the history and the correlation between the symptoms and the biological items.

[0055] To this end, the evaluation model (10) may be a model trained based on training data in which the urgency is labeled for each training data set in which the symptom type, duration of the symptom type, medical history, and biometric data of each of the various patients are matched.

[0056] At this time, the evaluation model (10) may be a model trained based on a learning algorithm such as logistic regression, Random Forest, Gradient Boosting, SVM (Support Vector Machine), MLP (Multi-Layer Perceptron), Transformer, etc., but is not limited thereto.

[0057] Meanwhile, in one embodiment, the server (100) acquires medical data including the patient's biometric data and the patient's medical history, and inputs symptom information, biometric data, and medical data into the evaluation model (10) to identify the urgency of medical treatment according to the patient's symptoms based on the output data.

[0058] To this end, the server (100) can obtain the patient's medical data from the medical record server (300) and obtain the patient's biometric data from the biometric measurement device (400) that measures the patient's biometric signal.

[0059] The server (100) can select a consultation mode for patient treatment based on the urgency (S130).

[0060] In one embodiment, the server (100) can select a consultation mode for patient treatment among a chatbot consultation mode, a remote consultation mode, and a face-to-face consultation mode based on urgency.

[0061] Specifically, the server (100) can select a chatbot consultation mode if the urgency is less than a preset first standard urgency (e.g., 40), select a remote consultation mode if the urgency is greater than or equal to the first standard urgency and less than a second standard urgency (e.g., 70), and select a face-to-face consultation mode if the urgency is greater than or equal to the second standard urgency.

[0062] FIG. 2 is a diagram illustrating the operation of a server communicating with a medical record server and a biometric measurement device according to one embodiment of the present disclosure.

[0063] Referring to FIG. 2, the server (100) can communicate with the medical record server (300) and the biometric device (400).

[0064] The medical record server (300) may correspond to a server that manages the medical system, EMR (Electronic Medical Record) system, or ENR (Electronic Nursing Record) system of a medical institution.

[0065] For example, the medical record server (300) can electronically record and manage medical data including a patient's treatment schedule, medical history, diagnostic history, prescription, test results, treatment records, medical records, test data, patient registration number, etc., and may be an FTP server (File Transfer Protocol Server), a web server, a database server, or a cloud server, but is not limited thereto.

[0066] The biometric measuring device (400) is a device for measuring biosignals.

[0067] For example, the biometric device (400) may be medical equipment or wearable devices (e.g., smart watch, smart ring, blood glucose meter (e.g., CGM), fitness tracker, HMD (Head Mounted Device), smart glasses, etc.) for measuring biometric signals / items such as blood pressure, respiratory rate, brainwave, electrocardiogram, electromyogram, heart rate, activity level, blood glucose, skin temperature, blood oxygen saturation, and stress level (HRV), but is not limited thereto.

[0068] In one embodiment, the server (100) may obtain medical data of a patient from a medical record server (300) and obtain biometric data including at least one biometric item measured for the patient from a biometric measurement device (400).

[0069] Accordingly, the server (100) can input the patient's symptom information, medical data, and biometric data into the evaluation model (10) and identify the urgency of treatment for the patient based on the output data.

[0070] FIG. 3 is a diagram illustrating the operation of a server communicating with a patient's terminal according to one embodiment of the present disclosure.

[0071] Referring to FIG. 3, the server (100) can communicate with the patient's terminal (200).

[0072] Here, the patient's terminal (200) may correspond to a user's terminal that wishes to receive medical treatment according to symptoms, and may be implemented as a desktop PC, tablet PC, laptop PC, smartphone, wearable device, etc.

[0073] In one embodiment, when the chatbot consultation mode is selected, the server (100) can provide at least one deep query for diagnosis according to symptoms to the patient's terminal (200).

[0074] Specifically, the server (100) can identify at least one deep query based on symptom information and provide it to the patient's terminal (200).

[0075] For example, the server (100) can obtain deep queries based on output data by inputting symptom information into a diagnostic model (20) trained based on the medical history of doctors.

[0076] The diagnostic model (20) can perform natural language processing (NLP) that recognizes and interprets natural language based on a language model, and natural language generation (NLG) that generates, reinforces, and reconstructs words and sentences, and may be a model that is trained based on the medical history of doctors and learns the doctors' clinical knowledge and diagnostic patterns.

[0077] Here, the language model is a model trained to output the statistically most appropriate output value (e.g., a vector value corresponding to the answer to the input text) based on input values ​​(e.g., text such as natural language, sentences, words, etc.), and may be a large language model, such as, but not limited to, GPT, Phi, BERT, LaMDA, LLaMA, RoBERTa, etc.

[0078] Specifically, the diagnostic model (20) can convert input text data into vector values ​​in a latent space through text embeddings, and can identify the context of words and phrases with similar meanings as well as other relationships between words, such as parts of speech, based on the vector values ​​in the latent space, and can output result values ​​for natural language processing, natural language generation, etc. based on the relationships between these vector values, and may correspond to a model trained to output a response to text data by mimicking the judgment process of doctors based on medical history including medical record data for each medical treatment performed by each doctor (e.g., patient symptom data, types of tests performed according to symptoms (e.g., CT, MRI, blood test, etc.), test data, diagnosis certificate, diagnosis name, prescription, follow-up procedures according to diagnosis results (e.g., rehabilitation program, physical therapy program, appointment, etc.)).

[0079] Here, the diagnostic model (20) may generate and output diagnostic information based on data input to the diagnostic model (20), and if it is determined that additional information is needed to generate diagnostic information, it may generate and output a deep query to request the input of additional information. Here, the diagnostic information may include an expected disease name.

[0080] Accordingly, when the server (100) inputs the patient's symptom information into the diagnostic model (20) and obtains an advanced query based on the output data, it can provide the advanced query to the patient's terminal (200).

[0081] After an in-depth query is provided to the patient's terminal (200), the server (100) can obtain an in-depth response to the in-depth query from the patient's terminal (200).

[0082] In one embodiment, the server (100) can obtain audio data containing the patient's voice from the patient's terminal (200) after a deep query is provided to the patient's terminal (200).

[0083] In this case, the server (100) can perform speech recognition on the audio data and identify the acquired text as a deep response.

[0084] For example, the server (100) can input audio data into a speech recognition model to obtain text corresponding to the audio data.

[0085] Here, the speech recognition model may be a model that performs the process of converting input speech data into text. The speech recognition model may consist of, but is not limited to, a feature extraction module, an acoustic model (AM), a pronunciation dictionary, a language model (LM), and a decoder.

[0086] Speech recognition models can analyze signals to extract acoustic features (Feature Extraction). Specifically, in the feature extraction step, representative acoustic features such as Mel-Frequency Cepstral Coefficients (MFCC) or spectrograms can be extracted by analyzing the input speech signal.

[0087] The extracted features are converted into pronunciation units through an Acoustic Model (AM), combined with a Language Model (LM) to generate contextually correct sentences, and finally, through a decoding step, a final text output can be generated.

[0088] Acoustic models (AMs) perform the role of analyzing feature vectors and converting them into phonemes, and Hidden Markov Models (HMMs), deep learning-based RNNs, or Transformer models can be used.

[0089] An Acoustic Model (AM) can be a component that performs the role of converting input speech data into pronunciation units (phonemes). The acoustic model learns which phoneme corresponds to a specific sound in a given speech signal and can be implemented using Hidden Markov Models (HMM), Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), or Transformer-based models. By training on a large amount of text data corresponding to the speech data, the acoustic model optimizes the relationship between specific acoustic patterns and pronunciations, and can perform noise removal and correction for acoustic variations (speaker intonation, environmental noise, etc.).

[0090] A pronunciation dictionary can perform the operation of mapping the output of an acoustic model to a word-by-word basis by providing phoneme combinations of specific words.

[0091] Language models (LMs) can perform the task of reducing errors by predicting contextually appropriate word sequences. Language models (LMs) can convert phoneme and word combinations predicted by acoustic models (AMs) into contextually correct sentences. Specifically, N-gram methods capable of predicting the probability of word occurrence, deep learning-based LSTM (Long Short-Term Memory), and Transformer (GPT, BERT) models can be applied to language models.

[0092] The decoder can perform the operation of combining the outputs of the acoustic model and the language model to ultimately determine the most suitable text.

[0093] In addition, the server (100) may obtain a deep response in text format from the patient's terminal (200).

[0094] When an in-depth response to an in-depth query is obtained, the server (100) inputs the in-depth response into a diagnostic model (20) and can obtain diagnostic information regarding symptom information based on the output data.

[0095] In addition, the server (100) can provide the diagnostic information to the patient's terminal (200) when the diagnostic information is obtained based on the data output by inputting the patient's symptom information into the diagnostic model (20).

[0096] That is, the server (100) may not perform the action of providing an in-depth query if an in-depth query is not obtained from the diagnostic model (20), and while an in-depth consultation with a patient is in progress, the diagnostic model (20) may generate an output (e.g., in-depth query, diagnostic information, etc.) based on a new input (e.g., in-depth response) by reflecting information related to the patient entered into the diagnostic model (20) (e.g., symptom information of the patient, at least one in-depth query provided to the patient, at least one in-depth response obtained from the patient, etc.).

[0097] In addition, when multiple deep queries are obtained, the server (100) can identify the accuracy of recognition regarding the degree to which the patient recognizes symptoms based on the similarity between the multiple deep queries, and can identify whether to switch to a remote consultation mode or a face-to-face consultation mode according to the identified accuracy of recognition.

[0098] In one embodiment, the server (100) maintains the chatbot consultation mode if the recognition accuracy is greater than or equal to a preset value (e.g., 60%), and if the recognition accuracy is less than a preset value, it can identify whether to switch to a remote consultation mode or a face-to-face consultation mode based on the recognition accuracy and the urgency (identified based on the baseline response).

[0099] To this end, the server (100) can identify the patient's recognition accuracy according to the order in which each deep query with a similarity exceeding a threshold is provided to the patient's terminal (500) and the number of deep queries with a similarity exceeding a threshold.

[0100] For example, the server (100) can convert each of a plurality of deep queries into a vector value in a latent space through text embeddings and calculate the similarity between deep queries based on the vector value in the latent space.

[0101] Specifically, the server (100) may identify similarity based on the angle between vectors for different deep queries by utilizing cosine similarity. In this case, the server (100) can calculate a higher similarity as the angle between vectors becomes smaller. Additionally, the server (100) may identify similarity based on the distance between vectors for different deep queries. Here, the server (100) can calculate the distance between vectors based on the difference in vector values ​​by dimension, and can calculate a higher similarity as the distance between vectors becomes closer.

[0102] At this time, the server (100) sets one of the multiple deep queries as a reference query and identifies the similarity between the deep query provided to the patient's terminal (500) after the reference query and the reference query, and can identify the similarity with the reference query sequentially according to the order in which the deep queries are provided.

[0103] For example, the server (100) may set the deep query provided to the patient's terminal (500) first among a plurality of deep queries as the reference query, and then, if no deep query is identified whose similarity to the reference query exceeds a threshold, set the deep query provided to the patient's terminal (500) as the reference query in the order following the deep query set as the reference query.

[0104] Here, when a deep query is identified in which the similarity with a reference query exceeds a threshold, the server (100) identifies the deep query provided to the patient's terminal (500) next to the reference query as the first target query, and identifies the deep query provided to the patient's terminal (500) next to the deep query in which the similarity with the reference query exceeds a threshold as the second target query.

[0105] In this case, the server (100) can identify the similarity between the first target query and the second target query, and if the similarity between the first target query and the second target query is below a threshold (identifying that the patient has entered a different response to a similar query), the patient's recognition accuracy can be lowered.

[0106] If the similarity between the first target query and the second target query exceeds a threshold, the server (100) may identify each of the deepening queries provided to the patient's terminal (500) in the order following each of the first and second target queries as the third target query and the fourth target query, and may perform an operation to identify the similarity between the third target query and the fourth target query, and may repeat this process until the similarity between the identified target queries is below the threshold.

[0107] In this case, the server (100) can reduce the intensity of lowering the patient's recognition accuracy as the number of times the operation to identify similarity between target queries is repeated (since there are many cases where the patient has entered similar responses to similar queries) increases.

[0108] If no deep query with a similarity exceeding a threshold is identified, the server (100) can increase the patient's recognition accuracy, and increase the intensity of increasing the patient's recognition accuracy as the number of deep queries provided to the patient's terminal (500) increases.

[0109] Here, the server (100) can set an initial value of the patient's recognition accuracy and identify the recognition accuracy by adjusting it upward or downward starting from the initial value.

[0110] For example, the server (100) can set an initial value for the patient's recognition accuracy based on the patient's urgency, and can set the initial value lower as it determines that the patient's judgment ability is deteriorated as the patient's urgency increases.

[0111] Meanwhile, if the recognition accuracy is less than a preset value, the server (100) can correct the first standard urgency which serves as a criterion for selecting the consultation mode, and compare the corrected first standard urgency with the patient's urgency (identified based on the patient's symptom information) to select the consultation mode.

[0112] For example, the server (100) may lower the first standard urgency level, but the greater the difference between the patient's recognition accuracy and the preset value, the greater the intensity of lowering the first standard urgency level.

[0113] Accordingly, the server (100) can select a remote consultation mode if the patient's urgency is less than the corrected first standard urgency, and select a face-to-face consultation mode if the patient's urgency is greater than or equal to the corrected first standard urgency.

[0114] Here, the server (100) can provide a message to the patient's terminal (500) recommending remote or face-to-face consultation with medical staff for a more accurate diagnosis.

[0115] Through this, the server (100) can reduce the possibility of misdiagnosis and enhance patient safety by comprehensively considering the degree of self-awareness and urgency regarding the patient's symptoms and requesting a timely switch to an appropriate consultation mode (remote consultation or face-to-face consultation) rather than just a simple chatbot consultation.

[0116] FIG. 4 is a diagram illustrating the operation of a server communicating with a patient's terminal, a medical staff's terminal, and a video chat server according to one embodiment of the present disclosure.

[0117] Referring to FIG. 4, the server (100) can communicate with the patient's terminal (200), the medical staff's terminal (500), and the video chat server (600).

[0118] In one embodiment, the server (100) requests the opening of a video chat channel from the video chat server (600) and can receive and obtain code information (e.g., URL, QR code, etc.) for accessing the video chat channel opened in accordance with the opening request from the video chat server (600).

[0119] Here, the medical staff terminal (500) is a terminal of personnel (e.g., doctor) capable of performing medical acts, and may correspond to a user terminal such as a desktop PC, tablet PC, laptop PC, smartphone, or wearable device, or may be implemented as a kiosk installed in a specific location, public PC, etc.

[0120] Here, the video chat server (600) can open a channel (e.g., API (Application Programming Interface)) for video conferencing / chatting between multiple terminals and generate code information for accessing the opened video chat channel.

[0121] Accordingly, the video chat server (600) supports real-time data transmission and reception between terminals connected to the video chat channel, so that video conferencing / chatting can be performed between terminals connected to the video chat channel.

[0122] In addition, as an example, when a remote consultation mode is selected, the server (100) can obtain consent information from the patient's terminal (200) regarding whether to perform remote consultation according to the remote consultation mode.

[0123] For example, when a remote consultation mode is selected, the server (100) provides a message to the patient's terminal (200) to check whether the patient consents to the remote consultation, and can obtain consent information including whether consent is given from the patient's terminal (200).

[0124] If the patient's terminal (200) is identified as having consented to remote consultation based on the consent information, the server (100) can select the patient's terminal (200) and the medical staff's terminal (500) to perform remote consultation.

[0125] In one embodiment, the server (100) can identify at least one medical professional who performed medical treatment on the patient based on the patient's medical data.

[0126] At this time, the server (100) can select a medical professional among the identified medical professionals that matches the medical specialty (e.g., internal medicine, surgery, pediatrics, otolaryngology, dentistry, obstetrics and gynecology, etc.) according to the patient's symptom information.

[0127] To this end, the server (100) may include a classification model trained to identify medical specialties based on symptoms.

[0128] The classification model may be a model trained to identify medical specialties based on symptoms. For example, the classification model may be a model trained on training data in which medical specialties are labeled for at least one keyword (e.g., fever, body temperature, vomiting, depression, etc.) for each different symptom, and may be a model trained on learning algorithms such as Logistic Regression, Support Vector Machine (SVM), Random Forest, Gradient Boosting Machine, and Convolutional Neural Network (CNN), but is not limited thereto.

[0129] Accordingly, the server (100) can input the patient's symptom information into a classification model and, based on the output data, identify the medical specialty according to the patient's symptom information.

[0130] Through this, the server (100) can select a terminal (500) of a medical professional that matches the medical specialty according to the patient's symptom information among the terminals (500) of the medical professional who performed treatment on the patient.

[0131] If there is no medical staff terminal (500) among the medical staff terminals (500) that have performed medical treatment on the patient that matches the medical specialty according to the patient's symptom information, the server (100) may select a medical staff terminal (500) that matches the medical specialty according to the patient's symptom information among the medical staff terminals (500) capable of remote consultation.

[0132] Afterward, the server (100) can provide code information to the patient's terminal (200) and the selected medical staff's terminal (500) to access a video chat channel for performing remote consultation with the selected medical staff's terminal (500).

[0133] Meanwhile, if it is identified that the patient's terminal (200) has not consented to remote consultation based on the consent information, the server (100) may provide the patient's terminal (200) with a page to schedule an in-person consultation with at least one medical professional.

[0134] Additionally, when the face-to-face consultation mode is selected, the server (100) can provide a page to the patient's terminal (200) to schedule a face-to-face consultation with at least one medical professional.

[0135] For example, the server (100) may provide a page containing available appointment schedules for face-to-face consultations for each medical professional who matches the medical specialty according to the patient's symptom information. Here, the page may correspond to a web page.

[0136] As an additional example of implementation, the server (100) can monitor the patient's biometric data for a certain period (e.g., 12 hours, 1 day, 3 days, etc.) after the consultation mode for the patient is selected.

[0137] Specifically, the server (100) can identify the urgency of medical treatment for a patient based on the patient's biometric data received from the biometric measurement device at a set period (e.g., 1 minute, 10 minutes, 30 minutes, 1 hour, etc.).

[0138] At this time, if the time during which an emergency level greater than the second standard emergency level described above is continuously identified exceeds a threshold time (e.g., 3 hours, etc.), the server (100) may provide a warning notification to the patient's terminal (200) containing a message indicating that an immediate face-to-face medical examination is required.

[0139] FIG. 5 is a block diagram illustrating the configuration of a server according to one embodiment of the present disclosure.

[0140] Referring to FIG. 5, the server (100) may include memory (110), a communication interface (120), and a processor (130).

[0141] The memory (110) stores various programs or data temporarily or non-temporarily and transmits the stored information to the processor upon the call of the processor (130). Additionally, the memory (110) can store various information required for the operation, processing, or control operation of the processor (130) in an electronic format.

[0142] The memory (110) may include, for example, at least one of a main memory and an auxiliary memory. The main memory may be implemented using a semiconductor storage medium such as ROM and / or RAM.

[0143] ROM may include, for example, conventional ROM, EPROM, EEPROM and / or MASK-ROM. RAM may include, for example, DRAM and / or SRAM. The secondary storage device may be implemented using at least one storage medium capable of storing data permanently or semi-permanently, such as a flash memory device, SD (Secure Digital) card, solid state drive (SSD), hard disk drive (HDD), optical recording media such as a magnetic drum, compact disc (CD), DVD or laser disc, magnetic tape, magneto-optical disc and / or floppy disk.

[0144] In one embodiment, the memory (110) may include an evaluation model (10) and a diagnostic model (20).

[0145] In addition, as an example, the memory (110) may include an analysis model and a classification model.

[0146] Meanwhile, the training of each of the evaluation model (10), diagnosis model (20), analysis model and classification model may be performed on the server (100), may be performed on a separately provided AI server, or the server (100) may acquire and use a model that has been pre-trained by an external device.

[0147] The communication interface (120) may include a wireless communication interface, a wired communication interface, or an input interface. The wireless communication interface may communicate with various external devices using wireless communication technology or mobile communication technology. Such wireless communication technologies may include, for example, Bluetooth, Bluetooth Low Energy, CAN communication, Wi-Fi, Wi-Fi Direct, ultrawide band (UWB), Zigbee, infrared data association (IrDA), or near field communication (NFC), and mobile communication technologies may include 3GPP, Wi-Max, LTE (Long Term Evolution), 5G, etc.

[0148] A wireless communication interface can be implemented using an antenna, a communication chip, a substrate, etc., capable of transmitting electromagnetic waves to the outside or receiving electromagnetic waves transmitted from the outside.

[0149] A wired communication interface can communicate with various devices based on a wired communication network. Here, the wired communication network can be implemented using physical cables, such as, for example, pair cables, coaxial cables, fiber optic cables, or Ethernet cables.

[0150] Depending on the embodiment, either the wireless communication interface or the wired communication interface may be omitted. Accordingly, the electronic device (100) may include only a wireless communication interface or only a wired communication interface. In addition, the electronic device (100) may be equipped with an integrated communication interface (120) that supports both wireless connection via the wireless communication interface and wired connection via the wired communication interface.

[0151] The electronic device (100) is not limited to having one communication interface (120) that performs a communication connection in one manner, but may include a plurality of communication interfaces (120) that perform communication connections in multiple manners.

[0152] The processor (130) controls the overall operation of the electronic device. Specifically, the processor (130) is connected to the configuration of the electronic device including memory as described above, and can control the overall operation of the electronic device by executing at least one instruction stored in the memory as described above. In particular, the processor (130) can be implemented as a single processor as well as as a plurality of processors.

[0153] The processor (130) may be implemented in various ways. For example, one or more processors (130) may include one or more of a CPU (Central Processing Unit), GPU (Graphics Processing Unit), APU (Accelerated Processing Unit), MIC (Many Integrated Core), DSP (Digital Signal Processor), NPU (Neural Processing Unit), hardware accelerator, or machine learning accelerator. One or more processors (130) may control one or any combination of other components of an electronic device and may perform operations or data processing related to communication. One or more processors (130) may execute one or more programs or instructions stored in memory. For example, one or more processors (130) may perform a method according to one embodiment of the present disclosure by executing one or more instructions stored in memory.

[0154] In embodiments of the present disclosure, the processor (130) may mean a system-on-chip (SoC) in which one or more processors (130) and other electronic components are integrated, a single-core processor, a multi-core processor, or a core included in a single-core processor or a multi-core processor, wherein the core may be implemented as a CPU, GPU, APU, MIC, DSP, NPU, hardware accelerator or machine learning accelerator, etc., but the embodiments of the present disclosure are not limited thereto.

[0155] In the case where the method according to one embodiment of the present disclosure includes a plurality of operations, the plurality of operations may be performed by a single processor (130) or by a plurality of processors (130). For example, when a first operation, a second operation, and a third operation are performed by the method according to one embodiment, the first operation, the second operation, and the third operation may all be performed by a first processor, or the first operation and the second operation may be performed by a first processor (e.g., a general-purpose processor) and the third operation may be performed by a second processor (e.g., an artificial intelligence dedicated processor).

[0156] One or more processors (130) may be implemented as a single-core processor including one core, or as one or more multicore processors including multiple cores (e.g., homogeneous multicore or heterogeneous multicore). When one or more processors (130) are implemented as multicore processors, each of the multiple cores included in the multicore processor may include internal processor memory such as on-chip memory, and a common cache shared by multiple cores may be included in the multicore processor (130). Additionally, each of the multiple cores included in the multicore processor (130) (or some of the multiple cores) may independently read and execute program instructions for implementing a method according to one embodiment of the present disclosure, or all (or some) of the multiple cores may be linked together to read and execute program instructions for implementing a method according to one embodiment of the present disclosure.

[0157] When a method according to one embodiment of the present disclosure includes a plurality of operations, the plurality of operations may be performed by one of the plurality of cores included in a multi-core processor, or may be performed by a plurality of cores. For example, when a first operation, a second operation, and a third operation are performed by a method according to one embodiment, the first operation, the second operation, and the third operation may all be performed by a first core included in a multi-core processor, or the first operation and the second operation may be performed by a first core included in a multi-core processor and the third operation may be performed by a second core included in a multi-core processor.

[0158] In one embodiment, the processor (130) can control a chatbot module (131), a data collection module (132), a data security module (133), etc. Each of these modules may correspond to a functional unit module implemented in software and / or hardware.

[0159] The chatbot module (131) is a module for performing consultation with a patient.

[0160] In one embodiment, the chatbot module (131) can provide a basic query to the patient's terminal and obtain a basic response to the basic query from the patient's terminal.

[0161] Additionally, when the chatbot consultation mode is selected, the chatbot module (131) can input the patient's symptom information into the diagnosis model (20) and obtain at least one deep query based on the output data.

[0162] Accordingly, the chatbot module (131) can provide an in-depth query to the patient's terminal and obtain an in-depth response to the in-depth query from the patient's terminal, and the chatbot module (131) can input the in-depth response into the consultation model (20) and, based on the output data, obtain diagnostic information about the patient and provide it to the patient's terminal.

[0163] The data collection module (132) is a module for acquiring data.

[0164] In one embodiment, the data collection module (132) can acquire the patient's medical data from the medical record server (300) and the patient's biometric data from the biometric measurement device (400).

[0165] The data security module (133) can encrypt and store data.

[0166] In one embodiment, the data security module (133) can store data (e.g., symptom information, diagnostic information, etc.) in distributed storage based on IPFS (InterPlanetary File System), Ceph, HDFS (Hadoop Distributed File System), etc., and record the hash value of the data, storage location, access rights information, etc. on the blockchain.

[0167] In addition, as an example, the data security module (133) may encrypt data (e.g., symptom information, diagnostic information, etc.) by applying an encryption technique (e.g., Zero-Knowledge Proof, Homomorphic Encryption, etc.).

[0168] In addition, in one embodiment, the data security module (133) can identify whether consent has been given for the collection and use of data (e.g., symptom information, diagnostic information, etc.) based on user input received through the patient's terminal (200).

[0169] In this case, if the server (100) is identified as having consented to the collection and use of data, it can control the chatbot module (131) to provide a basic query to the patient's terminal (200).

[0170] Meanwhile, the various embodiments described above may be implemented by combining two or more embodiments, provided that they do not conflict or contradict each other.

[0171] Specifically, each of the multiple operations, steps, and configurations for implementing one embodiment may be embodied in another embodiment, or the operations and steps of another embodiment may be followed by the last operation or step of one embodiment, but are not limited thereto.

[0172] Meanwhile, computer instructions or computer programs for performing processing operations in the various embodiments of the present disclosure described above may be stored in a non-transitory computer-readable medium. When such computer instructions or computer programs stored in the non-transitory computer-readable medium are executed by a processor of a specific device, the specific device described above performs processing operations according to the various embodiments described above.

[0173] A non-transient computer-readable medium refers to a medium that stores data semi-permanently and can be read by a device, unlike media that store data for a short period of time such as registers, caches, and memory. Specific examples of non-transient computer-readable media include CDs, DVDs, hard disks, Blu-ray discs, USBs, memory cards, and ROMs.

[0174] According to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product (e.g., downloadable app) may be temporarily stored or temporarily created on a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0175] Although preferred embodiments of the present disclosure have been illustrated and described above, the present disclosure is not limited to the specific embodiments described above. It is understood that various modifications can be made by those skilled in the art without departing from the essence of the present disclosure as claimed in the claims, and such modifications should not be understood individually from the technical spirit or perspective of the present disclosure. Explanation of the symbols

[0176] 100: Server 110: Memory 120: Communication interface 130: Processor 131: Chatbot Module 132: Data Collection Module 133: Data Security Module 200: Patient's terminal 300: Medical record server 400: Biometric device 500: Medical staff terminal 600: Video chat server 10: Evaluation Model 20: Diagnostic Model

Claims

Claim 1 In a method of operation of a server, the method comprises: a step of obtaining symptom information of the patient based on user input received through the patient's terminal; and a step of identifying the urgency of the medical treatment based on data output by inputting the symptom information into an evaluation model for identifying the urgency of the medical treatment. The method comprises the step of selecting a consultation mode for the treatment of the patient among a chatbot consultation mode, a remote consultation mode, and a face-to-face consultation mode based on the above-mentioned urgency; and the step of acquiring the patient's symptom information comprises providing at least one basic query for identifying symptoms to the patient's terminal, acquiring a basic response to the provided basic query from the patient's terminal, inputting the basic response into an analysis model for identifying symptom types and identifying the symptom information including at least one symptom type for the patient based on the output data; and the method of operation of the server comprises the step of, when the chatbot consultation mode is selected, providing at least one deep query for diagnosis according to symptoms to the patient's terminal; the step of acquiring a deep response to the deep query from the patient's terminal; and the step of inputting the acquired deep response into a diagnostic model trained based on doctors' medical history and identifying at least one of at least one deep query regarding the symptom information and at least one diagnostic information regarding the symptom information based on the output data. The method of operation of the server includes the step of providing at least one of the identified deep query and the identified diagnostic information to the patient's terminal; wherein, when a plurality of deep queries are provided to the patient's terminal, the method includes the step of identifying the recognition accuracy regarding the degree to which the patient recognizes symptoms based on the similarity between the plurality of deep queries; and the step of identifying whether to switch to the remote consultation mode or the face-to-face consultation mode based on the recognition accuracy.A method of operation of a server comprising: a step of identifying recognition accuracy regarding the degree to which the patient recognizes symptoms, wherein one of the plurality of deep queries is set as a reference query based on the order in which the plurality of deep queries are provided to the patient's terminal; when a deep query among the plurality of deep queries is identified in which the similarity with the reference query exceeds a threshold, the deep query provided to the patient's terminal next to the reference query is identified as a first target query; the deep query provided to the patient's terminal next to the deep query in which the similarity with the reference query exceeds the threshold is identified as a second target query; and the recognition accuracy is identified based on the similarity between the first target query and the second target query. Claim 2 delete Claim 3 A method of operation of a server according to claim 1, wherein the step of identifying the urgency of the medical treatment comprises acquiring medical data including the patient’s biometric data and the patient’s medical history, and inputting the symptom information, the biometric data, and the medical data into the evaluation model to identify the urgency of the medical treatment according to the patient’s symptoms based on the output data. Claim 4 A method of operation of a server according to claim 1, wherein the step of selecting a consultation mode for the treatment of the patient comprises selecting the chatbot consultation mode if the urgency is less than a preset first standard urgency, selecting the remote consultation mode if the urgency is greater than or equal to the first standard urgency and less than a second standard urgency greater than the first standard urgency, and selecting the face-to-face consultation mode if the urgency is greater than or equal to the second standard urgency. Claim 5 delete Claim 6 In claim 1, the method of operation of the server comprises: a step of obtaining consent information from the patient's terminal regarding whether to perform remote counseling according to the remote counseling mode when the remote counseling mode is selected; a step of selecting a medical staff terminal to perform the remote counseling when the patient's terminal is identified as having consented to the remote counseling according to the consent information; and a step of providing code information to the patient's terminal and the selected medical staff terminal to connect to a video chat channel for performing the remote counseling with the selected medical staff terminal. Claim 7 A method of operation of a server according to claim 1, wherein the method of operation of the server includes the step of providing a page to the patient's terminal to schedule a face-to-face consultation with at least one medical staff member when the face-to-face consultation mode is selected.

Images