Method and system for providing patient-customized feedback based on artificial intelligence

Abstract

The present invention relates to a method and system for providing patient-customized feedback based on artificial intelligence The method for providing patient-customized feedback based on artificial intelligence, according to the present invention, may comprise the steps of: collecting patient information related to indications of a patient; extracting, from the patient information, emotional information related to an emotional state of the patient; analyzing the emotional state related to the indications on the basis of the extracted emotional information and generating a prompt requesting generation of feedback content related to the emotional state; processing the generated prompt as an input to a pre-trained feedback generation model to obtain the feedback content related to the emotional state; and providing the feedback content related to the emotional state to a user terminal.

Description

AI-based method and system for providing patient-customized feedback

[0001] The present invention relates to a method and system for providing patient-customized feedback based on artificial intelligence.

[0002] With the rapid advancement of artificial intelligence (AI) technology in recent years, generative AI models capable of natural conversation with humans (e.g., ChatGPT) have emerged. In particular, unlike existing chatbots, Large Language Models (LLMs) provide conversational quality similar to humans based on their ability to understand various contexts and generate natural language, and are being rapidly adopted in various industrial fields such as medical, education, and healthcare.

[0003] With the advancement of such artificial intelligence technology, interest in AI-based patient-tailored treatment support technologies is rapidly expanding in the medical field. Specifically, the medical industry is seeing a growing need for technologies that utilize AI to provide non-face-to-face rehabilitation treatment, and to offer feedback that reinforces motivation for treatment or induces emotional stability by analyzing data on the patient's condition, emotions, and behaviors during the rehabilitation process.

[0004] In this regard, existing non-face-to-face feedback methods have limitations in that they fail to adequately reflect individual patients' situations or emotional states because they provide predefined, fixed feedback uniformly to various patients. Furthermore, the uniform feedback provided by existing non-face-to-face methods is perceived as noise by patients, which lowers their motivation to participate in treatment.

[0005] Specifically, under existing non-face-to-face feedback methods, patients are unable to engage with content irrelevant to their condition and are more likely to drop out early, which can significantly lower treatment continuation rates and overall treatment satisfaction.

[0006] Accordingly, there is a need for technology that can simultaneously enhance patient motivation for treatment and emotional support by analyzing changes in a patient's condition and emotions through artificial intelligence and generating and providing personalized feedback in response to the analysis results.

[0007] The present invention is intended to provide a method and system capable of providing patient-customized feedback based on artificial intelligence.

[0008] Specifically, the present invention aims to provide an artificial intelligence-based method and system for providing patient-customized feedback capable of analyzing a patient's emotional state based on emotional information extracted from patient information.

[0009] More specifically, the present invention aims to provide an artificial intelligence-based method and system for providing patient-customized feedback capable of generating patient-customized feedback content based on the emotional state related to the patient's indications.

[0010] Furthermore, the present invention aims to provide an AI-based patient-customized feedback provision method and system capable of providing feedback content to a patient by monitoring the patient's emotional state and determining the timing of feedback content provision according to emotional changes.

[0011] To solve the problem described above, the present invention proposes a method for providing patient-customized feedback corresponding to the emotional state by analyzing the emotional state related to the patient's indication based on artificial intelligence. The artificial intelligence-based method for providing patient-customized feedback according to the present invention may include the steps of: collecting patient information related to the patient's indication; extracting emotional information related to the patient's emotional state from the patient information; analyzing the emotional state related to the indication based on the extracted emotional information and generating a prompt requesting the generation of feedback content related to the emotional state; processing the generated prompt as input to a pre-trained feedback generation model to obtain the feedback content related to the emotional state; and providing the feedback content related to the emotional state to a user terminal.

[0012] Furthermore, the emotional information may include at least one of a first type of emotional information related to the patient's general emotions and a second type of emotional information related to the patient's indications.

[0013] Furthermore, in the step of generating the above prompt, the prompt is generated based on at least one of the above emotional information, the emotional state analysis result, and the above patient information, and the emotional state analysis result may include at least one of the first type emotional state analysis result corresponding to the first type emotional information and the second type emotional state analysis result corresponding to the second type emotional information.

[0014] Furthermore, the prompt may include at least one of a first type prompt generated based on at least one of the patient information, the first type emotion information, and the first type emotion state analysis result, and at least one of a second type prompt generated based on at least one of the patient information, the second type emotion information, and the second type emotion state analysis result.

[0015] Furthermore, in the step of providing the feedback content to the user terminal, the occurrence of a pre-set feedback event can be detected, and in response to the detection of the occurrence of the feedback event, the feedback content can be provided to the user terminal.

[0016] Furthermore, the method further includes a step of monitoring the emotional state associated with the above indication, and in the monitoring step, changes in the emotional class corresponding to the emotional state analyzed sequentially over time can be monitored.

[0017] Furthermore, in the step of providing the feedback content to the user terminal, the feedback event occurs in response to the detection of a change in the type of the emotion class, the time at which the feedback event occurs is determined as the time of providing feedback, and the feedback content can be provided to the user terminal at the time of providing feedback.

[0018] Furthermore, in the step of providing the feedback content to the user terminal, a user interface for providing at least one of a feedback program and a counseling program corresponding to the emotional state together with the feedback content may be provided to the user terminal.

[0019] Furthermore, the counseling program can provide an answer to a user query entered in relation to at least one of the patient's indications, emotional state, and feedback content, based on a previously trained large language model.

[0020] Furthermore, in the step of providing the feedback content to the user terminal, at least one of a first emotion state analysis report generated by analyzing the first type emotion information and a second emotion state analysis report generated by analyzing the second type emotion information may be provided.

[0021] Meanwhile, the artificial intelligence-based patient-customized feedback provision system according to the present invention comprises, in an electronic device, a memory for storing instructions and at least one processor electrically connected to said memory, and when said instructions are executed by said at least one processor, said at least one processor collects patient information related to the patient's indications, extracts emotional information related to the patient's emotional state from said patient information, analyzes said emotional state related to said indications based on the extracted emotional information, generates a prompt requesting the generation of feedback content related to said emotional state, processes said generated prompt as input to a pre-trained feedback generation model to obtain said feedback content related to said emotional state, and can provide said feedback content related to said emotional state to a user terminal.

[0022] Meanwhile, the program is executed by one or more processes in an electronic device and is stored on a computer-readable recording medium, and the program may include instructions for performing the steps of: collecting patient information related to a patient’s indication; extracting emotional information related to the patient’s emotional state from the patient information; analyzing the emotional state related to the indication based on the extracted emotional information and generating a prompt requesting the generation of feedback content related to the emotional state; processing the generated prompt as input to a pre-trained feedback generation model to obtain the feedback content related to the emotional state; and providing the feedback content related to the emotional state to a user terminal.

[0023] The artificial intelligence-based patient-customized feedback provision method and system according to the present invention can improve patient participation and immersion in treatment by analyzing patient information based on artificial intelligence and providing patient-customized feedback content that takes into account the individual patient's condition.

[0024] Furthermore, the AI-based patient-customized feedback provision method and system according to the present invention can analyze the patient's emotional state based on emotional information extracted from patient information and provide feedback content corresponding to the emotional state. Through this, an emotion-centered treatment environment can be realized that enhances psychological stability and willingness to participate in treatment by comprehensively considering the patient's psychological and emotional factors.

[0025] Furthermore, the AI-based patient-customized feedback provision method and system according to the present invention can continuously monitor changes in a patient's emotional state and control the timing of feedback provision based on these changes. Accordingly, by detecting changes in the patient's emotions in real time, immediate and empathetic feedback content can be provided at the necessary time, which can simultaneously improve the patient's emotional stability and treatment engagement.

[0026] Furthermore, the artificial intelligence-based patient-customized feedback provision method and system according to the present invention can realize an integrated treatment environment that simultaneously promotes psychological and physical recovery by providing feedback programs and counseling programs in conjunction with the patient's emotional state.

[0027] FIG. 1 is a conceptual diagram illustrating an artificial intelligence-based patient-customized feedback provision system according to the present invention.

[0028] FIG. 2 is a flowchart illustrating the overall method for providing patient-customized feedback based on artificial intelligence according to the present invention.

[0029] FIGS. 3a to 3d are conceptual diagrams for specifically explaining patient information and user feedback information according to the present invention.

[0030] FIGS. 4a and FIGS. 4b are conceptual diagrams illustrating the process of extracting different types of emotional information from patient information according to the present invention.

[0031] FIG. 5 is a conceptual diagram illustrating the process of generating a prompt according to the present invention.

[0032] FIGS. 6a to 6c are conceptual diagrams illustrating the process of providing feedback content to a user terminal according to the present invention.

[0033] FIGS. 7a and FIGS. 7b are conceptual diagrams for explaining a feedback program according to the present invention.

[0034] FIGS. 7c and FIGS. 7d ​​are conceptual diagrams for explaining a counseling program according to the present invention.

[0035] FIGS. 7e to 7g are conceptual diagrams for explaining an emotional state analysis report according to the present invention.

[0036] FIG. 8 is a block diagram illustrating a computing system in which the present invention can be implemented.

[0037] FIGS. 9 and FIGS. 10 are block diagrams illustrating an embodiment of a computing device according to the present invention.

[0038] Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Identical or similar components are assigned the same reference number regardless of the drawing symbols, and redundant descriptions thereof will be omitted. The suffixes "module" and "part" used for components in the following description are assigned or used interchangeably solely for the ease of drafting the specification and do not have distinct meanings or roles in themselves. Furthermore, in describing the embodiments disclosed in this specification, if it is determined that a detailed description of related prior art could obscure the essence of the embodiments disclosed in this specification, such detailed description will be omitted. Additionally, the attached drawings are intended only to facilitate understanding of the embodiments disclosed in this specification; the technical concept disclosed in this specification is not limited by the attached drawings, and it should be understood that they include all modifications, equivalents, and substitutions that fall within the spirit and technical scope of the present invention.

[0039] Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. These terms are used solely for the purpose of distinguishing one component from another.

[0040] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0041] A singular expression includes a plural expression unless the context clearly indicates otherwise.

[0042] In this application, terms such as “comprising” or “having” 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.

[0043] The present invention relates to a method and system for providing patient-customized feedback content based on artificial intelligence. Specifically, the present invention relates to a method and system for generating feedback content based on artificial intelligence according to the emotional state related to a patient's indications and providing the generated feedback content to a user terminal.

[0044] For example, the present invention utilizes a feedback generation model based on a large language model to generate feedback content according to the emotional state related to the patient's indications. Here, a large language model may refer to an artificial intelligence model capable of understanding and generating natural language by learning from a vast amount of data. Specifically, a large language model may refer to a generative artificial intelligence model capable of learning semantic relationships between texts included in an input prompt, understanding the context of the prompt, and generating natural and consistent sentences corresponding to it.

[0045] In addition, according to the present invention, the emotional state may refer to an emotional state based on at least one of emotions directly related to the patient's physical condition or treatment progress, such as pain, discomfort, or recovery stage in relation to the patient's indication, and general emotions arising from daily experiences, interpersonal relationships, environmental factors, etc., which are not directly related to the indication.

[0046] Meanwhile, the feedback content according to the present invention may refer to content comprising at least one of text, image, video, and audio for alleviating or improving emotional states such as pain, anxiety, and recovery process according to the indication of a patient.

[0047] In this context, the “indication” according to the present invention refers to a symptom or clinical situation requiring specific treatment or examination, which can be understood as the user’s disease or symptoms. For the convenience of explanation, the present invention focuses on “indications related to musculoskeletal diseases,” but is not necessarily limited thereto. As an example, the indications described in the present invention may include all indications arising from various diseases, including musculoskeletal diseases (e.g., cancer, diabetes, hypertension, etc.). The “patient” described in the present invention may refer to a user experiencing pain due to an indication, and in the present invention, the term “patient” may be used interchangeably with “user.”

[0048] In contrast, the present invention may provide feedback content related to exercises performed for health promotion in daily life, rather than rehabilitation exercises for therapeutic purposes related to the patient's indications. For example, in the exercise according to the present invention, there are no specific restrictions on the purpose of the exercise, and it may refer to exercises performed for various purposes, such as rehabilitation exercises, fitness exercises, ball sports, and dance exercises, for therapeutic, health promotion, or cosmetic purposes. In this case, the body part related to the indication may also be interpreted as the body part that the user targets for exercise.

[0049] In the present invention, patient information related to a user account logged into a user terminal is collected from a pre-established database, and the patient's emotional state can be analyzed from the patient information. Here, "patient information" may include various information related to the user, such as the patient's medical information, exercise history information, and user feedback information.

[0050] Furthermore, the feedback generation model previously trained in the present invention can generate patient-customized feedback content based on the patient's emotional state and provide it to a user terminal. For example, in the present invention, a patient (or user) can receive customized feedback content related to the patient's indications through an application pre-installed on the user terminal.

[0051] In the foregoing, the provision of patient-customized feedback based on artificial intelligence according to the present invention has been generally described, and this can be implemented by the feedback provision system described below. Below, with reference to FIG. 1, the artificial intelligence-based patient-customized feedback provision system according to the present invention will be described in detail. FIG. 1 is a conceptual diagram illustrating the artificial intelligence-based patient-customized feedback provision system according to the present invention.

[0052] As illustrated in FIG. 1, the artificial intelligence-based patient-customized feedback providing system according to the present invention (hereinafter referred to as the “feedback providing system,” 100) may include at least one of a communication unit (110), a storage unit (120), and a control unit (130). At this time, the feedback providing system (100) according to the present invention is not limited to the components described above and may further include components that perform the same or similar roles as the functions described in the present specification.

[0053] Meanwhile, the feedback providing system (100) according to the present invention may be implemented as an application or software. The feedback providing system (100) implemented as software in this manner may be downloaded via a program (e.g., Play Store) that allows the application to be downloaded on the user terminal (10), or implemented via an initial installation program on the user terminal (10). In this case, the communication unit (110), storage unit (120), and control unit (130) according to the present invention may be utilized as components of the user terminal (10). In the present invention, the user terminal (10) can be understood to mean an application installed on the user terminal (10). Such an application (or software) can be understood as a component of the feedback providing system (100) according to the present invention.

[0054] In the present invention, the user terminal (10) may also be named a 'mobile terminal' or 'electronic device', and the user terminal (10) described in this specification may include a mobile phone, a smartphone, a smart TV, a laptop computer, a digital broadcasting terminal, a PDA (personal digital assistants), a PMP (portable multimedia player), a navigation device, a slate PC, a tablet PC, an ultrabook, a wearable device (e.g., a smartwatch, a smart glass, a head-mounted display), etc.

[0055] More specifically, the user terminal (10) according to the present invention is not limited to an electronic device in which an application is activated, but may refer to an electronic device connected to an electronic device in which an application is activated. As an example, based on the fact that the user terminal (10) according to the present invention is a smartphone, the user terminal (10) may refer to a smart TV connected to said smartphone.

[0056] Meanwhile, the feedback providing system (100) may exist inside a server (hereinafter referred to as the server) built to perform a specific purpose (e.g., providing counseling services), or it may exist as a separate device from the server. When the feedback providing system (100) exists inside the server, the feedback providing system (100) according to the present invention may provide a user-customized counseling service through at least one component among a communication unit (110), a storage unit (120), and a control unit (130) located inside the server, or through a module that performs a function similar to each of the above components.

[0057] In this case, the application can provide a consultation service on the user terminal (10) where the application is installed through communication with the server. Furthermore, the feedback providing system (100) according to the present invention can provide artificial intelligence-based patient-customized feedback according to the present invention to the user terminal (10) by linking with a plurality of different external servers.

[0058] A user (U, or patient) of the present invention may receive patient-customized feedback content regarding the indications of the patient (U) through an application or webpage provided by the feedback providing system (100) according to the present invention. At this time, the user (or patient, U) of the present invention may possess a user account registered with the feedback providing system (100) according to the present invention. For convenience of explanation, the account of the patient user in this specification is referred to as a "user account (or patient account)." The "account" described above may be created through a page linked to the feedback providing system (100).

[0059] Alternatively, an ‘account’ may be created on at least one other server (e.g., a medical staff server) linked to the feedback providing system (100) according to the present invention. Accordingly, in this specification, without distinguishing the server where the account was issued, all accounts based on the feedback providing system (100) according to the present invention are referred to as “accounts already registered in the feedback providing system (100) according to the present invention.”

[0060] Meanwhile, the “medical staff (D)” described in the present invention refers to a person employed at a medical institution (e.g., a hospital) and may include, for example, at least one of a doctor, a nurse, or a physical therapist. For convenience of explanation, the present invention describes medical staff by citing doctors and physical therapists as examples. However, medical staff are not limited thereto, and any user employed at a medical institution to provide patient-customized feedback content may be considered medical staff according to the present invention.

[0061] A medical professional according to the present invention may possess a medical professional account already registered in the feedback provision system (100) according to the present invention. In this specification, a user terminal logged in with a medical professional account is referred to as a medical professional terminal. As an example, the feedback provision system (100) according to the present invention may receive medical information including prescription information prescribed by a medical professional (D) to a user (U) by linking with a medical professional server.

[0062] Meanwhile, according to the present invention, the communication unit (110) may be connected via a wireless or wired network to a user terminal (10), a medical staff terminal, an LLM server (140), a central server, a device, and at least one network, and configured to receive or transmit overall data and information necessary for the operation of the feedback provision system (100) according to the present invention. Specifically, the communication unit (110) may receive medical information related to the indications of the user (U) from the medical staff terminal in order to generate patient-customized feedback content.

[0063] Furthermore, the communication unit (110) may receive a user query from the user terminal (10). Here, “receiving a user query” may mean receiving an input signal (or selection signal) corresponding to the user query based on the user query being input by the user through the user terminal (10). According to the present invention, the user query may include at least one of a document, text, an image (or video), and voice. In this case, the feedback providing system (100) may further include a module that converts voice into text. For example, the feedback providing system (100) may include a voice recognition model capable of analyzing voice data corresponding to voice received through a microphone provided in the user terminal (10). For example, a speech recognition model can convert speech into text based on at least one of the following: a STT (Speech-to-Text) algorithm, a HMM (Hidden Markov Model), a HMM-GMM (Hidden Markov Model-Gaussian Mixture Model), a CTC (Connectionist Temporal Classification) based model, a Beam Search based model, a DNN (Deep Neural Network) based model, an RNN (Recurrent Neural Network) based model, a Seq2Seq (Sequence-to-Sequence) model, and a Transformer based model.

[0064] Here, the user terminal (10) may include at least one of a mobile phone, a smartphone, a notebook computer, a laptop computer, a slate PC, a tablet PC, an ultrabook, a desktop computer, a digital broadcasting terminal, a PDA (personal digital assistants), a PMP (portable multimedia player), a navigation device, and a wearable device (e.g., a smartwatch, a smart glass, a head-mounted display).

[0065] The communication unit (110) may include at least one communication module capable of wireless communication and wired communication between the feedback providing system (100) and the communication target. Additionally, the communication unit (110) may include a communication module that connects the feedback providing system (100) to at least one network.

[0066] Meanwhile, the communication unit (110) can support various communication methods depending on the communication standard of the communicating device. For example, the communication unit (110) may be configured to perform communication using at least one of the following technologies: WLAN (Wireless LAN), Wi-Fi (Wireless-Fidelity), Wi-Fi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), 5G (5th Generation Mobile Telecommunication), Bluetooth (Bluetooth™ Frequency Identification), Infrared Communication (Infrared Data Association; IrDA), UWB (Ultra-Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi Direct, and Wireless USB (Wireless Universal Serial Bus).

[0067] Next, the storage unit (120) may be configured to store various information related to the present invention. In the present invention, the storage unit (120) may be provided in the feedback providing system (100) itself, or alternatively, at least a part of the storage unit (120) may mean a database (Database: DB, 200).

[0068] The storage unit (120) may include one or more non-transient computer-readable storage media that can be read and / or accessed by at least one processor. One or more computer-readable storage media may include volatile and / or non-volatile storage components such as optical, magnetic, organic, or other memory or disk storage devices. In some examples, the storage unit (120) may be implemented using a single physical device (e.g., one optical, magnetic, organic, or other memory or disk storage device), whereas in other examples, the storage unit (120) may be implemented using multiple physical devices.

[0069] The storage unit (120) may include computer-readable instructions and additional data. The storage unit (120) may include a storage necessary to perform at least some of the methods and techniques described herein and / or at least some of the functions of the device and network.

[0070] Furthermore, at least a portion of the storage unit (120) may be a cloud storage or a cloud server. That is, the storage unit (120) is sufficient as long as it is a space where information necessary for the operation of the feedback providing system (100) according to the present invention is stored, and it can be understood that there are no restrictions on the physical space. Accordingly, the storage unit (120) and the database (200) may be used interchangeably without being separately distinguished below.

[0071] Meanwhile, the storage unit (120) may store patient information including at least one of medical information, exercise history information, question and answer information, and user feedback information. As an example, the medical information according to the present invention may include at least one of the patient's (or user's) age, the user's gender, the user's medical history, indication information, and treatment plan. Additionally, the exercise history information according to the present invention may include at least one of the date of exercise performance, the composition of exercise movements included in the exercise program performed, the results of motion analysis for each exercise movement, and the results of exercise history analysis. Furthermore, the exercise history information according to the present invention may include at least one of the name of each of a plurality of exercise items constituting an exercise program assigned to a user account, the number of exercises, the time of the exercise, information on the difficulty of the exercise, and an exercise video and exercise description corresponding to each of the plurality of exercise items.

[0072] Additionally, indication information may refer to medical condition information and clinical course data related to the target disease, symptoms, and rehabilitation purposes applicable to the patient. For example, the storage unit (120) may contain information directly related to indications in the patient's physical condition or treatment process, such as diagnosis name, pain location, symptom intensity, treatment stage, and treatment history.

[0073] Furthermore, the user feedback information according to the present invention may include at least one of the user's survey response information and memo information. Specifically, the storage unit (120) may store survey response information including user response data for at least one survey provided to the user terminal (10). At this time, the at least one survey includes at least one of a multiple-choice survey and a short-answer survey, and the survey response information may include at least one of a multiple-choice survey response information and a short-answer survey response information. Additionally, the storage unit (120) may store memo information including natural language text entered by the user (or patient) through the user terminal (10).

[0074] Meanwhile, the database (DB, 200) may be configured to store various information related to providing patient-customized feedback content. Specifically, the database (200) may store patient information for each of multiple users (or patients). For example, the database (200) may store at least one of medical information, exercise history information, and user feedback information corresponding to each of multiple users.

[0075] Furthermore, the database (200) may store exercise motion content (e.g., exercise name, number of exercises, timing of exercises, difficulty of exercises, etc.) corresponding to each of multiple body parts (e.g., shoulder, elbow, wrist & hand, hip & pelvis, knee, ankle & foot, neck, back, waist, abdomen) related to the indication. Furthermore, the database (200) may store at least one of an exercise video and an exercise description corresponding to each exercise motion content.

[0076] In the present invention, the storage unit (120) is described as existing separately from the database (200), but is not limited thereto, and the storage unit (120) may include the database (200).

[0077] Meanwhile, user authentication information may be stored in the storage unit (120). Here, “user authentication information” may refer to information used in a user authentication process performed to log in to a user account on a user terminal (10). As an example, user authentication information may be various, such as i) ID, ii) password, iii) password pattern, iv) user’s fingerprint authentication information, v) face authentication information, vi) voice authentication information, vii) iris authentication information, viii) vein authentication information, etc., set by the user.

[0078] Meanwhile, data and commands necessary for the operation of the feedback providing system (100) according to the present invention may be stored in the storage unit (120). Specifically, commands for the operation of the emotion analysis module (131) may be stored in the storage unit (120). The emotion analysis module (131) according to the present invention may refer to a module that extracts emotional information of a patient based on patient information and analyzes the emotional state of the patient corresponding to the emotional information. Here, emotional information may refer to information related to emotional expression (or emotional state) among the patient information.

[0079] There may be a wide variety of methods for extracting emotional information of a patient based on patient information according to the present invention and analyzing the patient's emotional state corresponding to the emotional information, and in this specification, any module capable of extracting emotional information of a patient based on patient information and analyzing the patient's emotional state corresponding to the emotional information is not limited to any specific type or method.

[0080] The feedback providing system (100) according to the present invention may further include at least one of a module and an algorithm that perform the same function as the sentiment analysis module (131). At this time, the sentiment analysis module (131) may include at least one artificial intelligence model among GPT, BioGPT, Gemini, BERT (Bidirectional Encoder Representations from Transformers), RoBERTa (Robustly optimized BERT approach), DeBERTa, KoBERT, KorBERT, DistilBERT, ClinicalBERT, MentalBERT, KoPsychBERT, ELECTRA, and ALBERT.

[0081] Additionally, the emotion analysis module (131) can perform operations to interpret contextual meaning from input patient information and infer an emotional state by using a Transformer neural network structure based on at least one algorithm among Self-Attention, Multi-Head Attention, and Position-Wise Feedforward Network.

[0082] Additionally, the sentiment analysis module (131) may include at least one classifier among LSTM (Long Short-Term Memory), GRU (Gated Recurrent Unit), or Transformer-based classifiers (e.g., BERT, Classification head), and may improve precision by ensembling with a CNN-based text classifier if necessary. Additionally, the sentiment analysis module (131) may calculate Emotion Intensity scores in parallel to account for subtle differences in emotional expression, or derive probability values ​​for each emotion using Softmax or Sigmoid-based activation functions.

[0083] The emotion analysis module (131) according to the present invention may be pre-trained or fine-tuned based on a large-scale emotion labeling dataset (KoNLPy, AIHub Sentimental Conversation Corpus, etc.) or actual clinical-based emotion response data. Furthermore, a domain-specific learning strategy may be applied to the emotion analysis module (131) to reflect differences in emotional expression according to the user's age and cultural expression characteristics.

[0084] Meanwhile, commands for the operation of a prompt generation unit (not shown) may be stored in the storage unit (120). The prompt generation unit according to the present invention may refer to a module capable of generating a prompt that requests the generation of feedback content related to a patient's emotional state based on at least one of patient information, emotional information, and an emotional state analysis result. At this time, the prompt generation unit may generate a prompt to be input into a pre-trained feedback generation model using at least one artificial intelligence model or algorithm. For example, the prompt generation unit may include at least one of a large language model based on T5 (Text-to-Text Transfer Transformer), BART (Bidirectional and Auto-Regressive Transformer), GPT (Generative Pre-trained Transformer), or LLaMA (Language Model for Many Applications), a rule-based template matching algorithm, a conditional prompting technique, a contextual embedding selection module, or a few-shot prompt generator.

[0085] At this time, the model or algorithm included in the prompt generation unit according to the present invention is not limited to the examples described above and may further include artificial intelligence models and algorithms having the same function. The prompt generation unit according to the present specification is not limited in type or method, provided it is a module capable of generating a prompt requesting the generation of feedback content related to a patient's emotional state based on at least one of patient information, emotional information, and an emotional state analysis result.

[0086] Furthermore, the storage unit (120) according to the present invention may store commands for the operation of a pre-trained feedback generation model (132). Here, the pre-trained feedback generation model (132) according to the present invention may refer to an artificial intelligence model that generates feedback content related to the patient's emotional state based on an input prompt. Specifically, the pre-trained feedback generation model (132) may generate feedback content related to the patient's emotional state by using a large language model. For example, the feedback generation model (132) may generate feedback content related to the patient's emotional state by using at least one large language model based on at least one of GPT (Generative Pre-trained Transformer), Gemini, T5 (Text-to-Text Transfer Transformer), Vision-Language Models (e.g., CLIP, Flamingo), Multimodal Generative Models (GPT-4 Multimodal, PaLM-E), BERT (Bidirectional Encoder Representations from Transformers), and LaMDA (Language Model for Dialogue Applications).

[0087] Meanwhile, information related to a feedback program corresponding to the patient's emotional state may be stored in the storage unit (120). Specifically, information related to a feedback program mapped to an emotion class corresponding to the patient's emotional state may be stored in the storage unit (120). The information related to the feedback program according to the present invention may include information regarding an intervention-type program for regulating the patient's emotional state or inducing psychological recovery. For example, the information related to the feedback program stored in the storage unit (120) may include various forms of content such as breathing techniques, relaxation training, meditation, stretching, cognitive enhancement activities, and emotional stability-inducing content.

[0088] Next, the control unit (130) may be configured to control the overall operation of the feedback providing system (100) related to the present invention. Specifically, the control unit (130) may include at least one of an emotion analysis module (131) and a pre-learned feedback generation model (132). The control unit (130) may process signals, data, information, etc. that are input or output through the components described above, or provide or process appropriate information and functions to the user.

[0089] The control unit (130) can control the output of a service page for providing feedback content through a display unit (or touchscreen) provided in the user terminal (10) and the medical staff terminal. Such a service page may be output on the user terminal (10) and the medical staff terminal through an application or web page installed on the user terminal (10) and the medical staff terminal. The service page is a page linked to the feedback providing system (100) according to the present invention and is configured to be controlled by the feedback providing system (100) according to the present invention.

[0090] Furthermore, when a service page is provided in the form of an application, the service page may be controlled by a CPU (Central processing unit) of a user terminal (10) on which the application is installed. In this case, the CPU of the user terminal (10) may provide customized feedback content to a user (or patient) based on information provided by the feedback providing system (100) according to the present invention.

[0091] Meanwhile, the control unit (130) can collect patient information corresponding to a user account logged into the user terminal (10) from the storage unit (120, or database (200)). Furthermore, the control unit (130) can extract emotional information from the patient information using an emotional analysis module (131). Specifically, the control unit (130) can extract different types of emotional information from the patient information using the emotional analysis module (131). For example, the emotional analysis module (131) can extract emotional information from the patient information that includes at least one of a first type of emotional information related to general emotions and a second type of emotional information related to emotions regarding the patient's indications.

[0092] General emotions according to the present invention may refer to emotions arising from daily experiences, interpersonal relationships, environmental factors, or personal emotional responses that are not directly related to the indications or treatment status of the patient (or user). For example, general emotions may include emotions related to non-medical emotions formed by daily situations or external stimuli, such as fatigue, lethargy, stress, mood deterioration, tension, and instability.

[0093] In contrast, emotions regarding indications may refer to emotions associated with a physical condition directly related to the patient's disease, symptoms, treatment, or rehabilitation process. For example, emotions regarding indications may include emotions directly associated with the indication, such as at least one of anxiety due to pain, frustration due to delayed recovery, fear of worsening symptoms, tension during the treatment process, and anticipation of recovery.

[0094] The control unit (130) can analyze the emotional state related to the indication based on the extracted emotional information and generate a prompt requesting the creation of feedback content related to the emotional state. Specifically, the emotional analysis module (131) can analyze the extracted emotional information and generate an emotional state analysis result regarding the patient's emotional state.

[0095] Furthermore, the emotion analysis module (131) can extract (or identify) emotion keywords for the emotion state from the results of the emotion state analysis and calculate an emotion class corresponding to the patient's emotion state based on the extracted emotion keywords. For example, if the emotion analysis module (131) identifies "worry" as an emotion keyword in the sentence "I am worried because the pain is severe," it can correspond the user's emotion state to the 'anxiety' class.

[0096] As an example, the emotion analysis module (131) can extract emotion keywords regarding the patient's emotional state from the emotional state analysis results using a pre-prepared keyword extraction algorithm. The pre-prepared “keyword extraction algorithm” according to the present invention may refer to an algorithm that extracts emotion keywords regarding the emotional state from the emotional state analysis results based on at least one of Named Entity Recognition (NER), KeyBERT, GPT, and Transformer. The keyword extraction algorithm according to the present invention is not limited to the algorithm described above, and the storage unit (120) may further include commands that enable an algorithm to operate the same function as the keyword extraction algorithm according to the present invention.

[0097] Specifically, the control unit (130) can use a prompt generation unit to generate a prompt to be input into a pre-trained feedback generation model based on at least one of emotional information, an emotional state analysis result, and patient information. At this time, the prompt may include different types of prompts. For example, the prompt may include at least one of a first type prompt corresponding to first type emotional information and a second type prompt corresponding to second type emotional information.

[0098] In the present invention, the control unit (130) is described as including a prompt generation unit, but it is not limited thereto, and the control unit (130) and the prompt generation unit may exist separately. In this case, the control unit (130) may generate a prompt requesting the generation of feedback content related to an emotional state in conjunction with the prompt generation unit.

[0099] Furthermore, the control unit (130) processes the generated prompt as input to a pre-trained feedback generation model (132) to obtain feedback content related to an emotional state from the feedback generation model (132). Specifically, the pre-trained feedback generation model (132) may refer to an artificial intelligence model trained to generate patient-customized feedback content in conjunction with a large language model.

[0100] For example, the control unit (130) can fine-tune (or tune, optimize) a large language model based on multiple patient information. Specifically, the control unit (130) can train a feedback generation model (132) to generate customized feedback content that adaptively responds to the patient's emotional state by fine-tuning the parameters of the large language model using emotional information, indication information, and user feedback data included in the patient information as training data.

[0101] More specifically, the control unit (130) can adjust the decoder or adapter layer of a pre-trained large language model using a learning set consisting of pairs of a patient's emotion class (e.g., anxiety, helplessness, pain-fear, etc.) and corresponding feedback content (e.g., encouraging message, calming sentence, etc.).

[0102] At this time, the large language model according to the present invention may be included in an existing LLM server (140). For example, the LLM server (140) may include at least one large language model based on at least one of GPT (Generative Pre-trained Transformer), T5 (Text-to-Text Transfer Transformer), Vision-Language Models (e.g., CLIP, Flamingo), Multimodal Generative Models (GPT-4 Multimodal, PaLM-E), BERT (Bidirectional Encoder Representations from Transformers), and LaMDA (Language Model for Dialogue Applications).

[0103] In the present invention, the LLM server (140) is described as existing separately from the feedback providing system (100), but is not limited thereto, and the feedback providing system (100) may be configured to include the LLM server (140). That is, the feedback providing system (100) and the LLM server (140) according to the present invention may exist separately, or the LLM server (140) may be included in the feedback providing system (100). For convenience of explanation, the LLM server (140) and the large language model are used interchangeably below, and the use of the large language model by the control unit (130) can be understood as using at least one large language model included in the LLM server (140).

[0104] Furthermore, the control unit (130) can provide feedback content obtained from a previously learned feedback generation model (132) to the user terminal (10). Specifically, the control unit (130) can provide feedback content related to the patient's emotional state through a service page displayed on the user terminal (10).

[0105] Specifically, the control unit (130) can determine the timing of the feedback provision when providing feedback content. For example, the control unit (130) can monitor the emotional state of a patient related to an indication. Furthermore, the control unit (130) can determine the timing of the feedback provision when a feedback event occurs in response to detecting a change in the type of the emotion class, and the timing of the feedback event occurs as the timing of the feedback provision. Furthermore, the control unit (130) can provide the feedback content to a user terminal at the time of the feedback provision.

[0106] Meanwhile, the control unit (130) may provide at least one of a feedback program and a counseling program corresponding to an emotional state along with feedback content. Specifically, the control unit (130) may output (or provide) a user interface for providing at least one of a feedback program and a counseling program corresponding to an emotional state on a service page of the user terminal (10).

[0107] Here, the feedback program may include an intervention-type program designed to regulate the patient's emotional state or induce psychological recovery. Additionally, the counseling program may include a counseling service that generates and provides answers to the patient's user queries based on a large language model.

[0108] Furthermore, the control unit (130) can provide an emotional state analysis report generated by analyzing emotional information to the user terminal (10) along with feedback content. Specifically, the control unit (130) can generate an emotional state analysis report corresponding to each of different types of emotional information. Furthermore, the control unit (130) can generate a comprehensive emotional state analysis report based on different types of emotional state analysis reports and provide it to the user terminal (10).

[0109] Meanwhile, the feedback providing system (100) may include one or more processors, and such processors may include one or more general-purpose processors and / or one or more special-purpose processors (e.g., digital signal processors, tensor processing units (TPUs), graphics processing units (GPUs), neural network processing units (NPUs), application integrated circuits, application semiconductors (ASICs), etc.). One or more processors may be configured to execute instructions, computer-readable instructions, and / or other instructions described herein that are stored (or included) in the storage unit (120). The feedback providing system (100) may perform data processing described below in cooperation with memory and at least one processor. The processor may be electrically connected to memory and may perform a series of operations and data processing using data and information stored in memory. Here, “memory” may be a component of the storage unit (120), and “processor” may be used interchangeably with the control unit (130).

[0110] In the foregoing, the feedback providing system (100) of the present invention has been described, and it can be implemented based on the artificial intelligence-based patient-customized feedback providing method described below.

[0111] Hereinafter, with reference to FIG. 2 together with FIG. 3a to 3c, FIG. 4a, FIG. 4b, FIG. 5, FIG. 6a to 6c, and FIG. 7a to 7d, the method for providing artificial intelligence-based patient-customized feedback according to the present invention will be described in more detail. FIG. 2 is a flowchart for generally explaining the method for providing artificial intelligence-based patient-customized feedback according to the present invention. FIG. 3a to 3d are conceptual diagrams for specifically explaining patient information and user feedback information according to the present invention, and FIG. 4a and 4b are conceptual diagrams for explaining the process of extracting different types of emotion information from patient information according to the present invention. FIG. 5 is a conceptual diagram for explaining the process of generating a prompt according to the present invention, and FIG. 6a to 6c are conceptual diagrams for explaining the process of providing feedback content to a user terminal according to the present invention. FIGS. 7a and 7b are conceptual diagrams for explaining a feedback program according to the present invention, FIGS. 7c and 7d are conceptual diagrams for explaining a counseling program according to the present invention, and FIGS. 7e to 7g are conceptual diagrams for explaining an emotional state analysis report according to the present invention.

[0112] In the present invention, a process of collecting patient information related to the patient's indications may be carried out (S210, see FIG. 2).

[0113] As illustrated in FIG. 3a, the control unit (130) can verify the user (or patient) account logged into the user terminal (10). Furthermore, the control unit (130) can collect patient information (300) corresponding to the user account from at least one of the storage unit (120), the database (200), and the external server (e.g., the medical staff server). Here, “patient information (300)” may include at least one of medical information (310) related to the patient, exercise history information (320), and user feedback information (330). The method for collecting patient information (300) corresponding to the user account according to the present invention may be very diverse, and the present specification is not limited to the type and method as long as it is a module capable of collecting patient information (300) corresponding to the user account.

[0114] For example, the control unit (130) can verify user identification information corresponding to a user account in order to collect patient information (300). At this time, the user identification information may include various identification information such as a unique identifier (ID) of the user account, a session token, and a login token, and the control unit (130) can identify (or verify) the user account logged into the user terminal using the user identification information.

[0115] The control unit (130) can retrieve patient information (300) corresponding to the identified user account information from at least one of the storage unit (120), the database (200), and the linked external server based on the identification of the user account. Furthermore, the control unit (130) can collect the retrieved patient information (300) as patient information (300) corresponding to the user account.

[0116] As an example, the control unit (130) may collect medical information (310) including at least one of the user's (or patient's) age, the user's gender, the user's medical history, prescription information prescribed from the medical staff terminal (20), treatment plan, and the patient's indication information. Specifically, the control unit (130) may collect indication information including at least one of the medical condition information and clinical progress data related to the treatment target disease, symptoms, and rehabilitation purpose applicable to the patient. For example, the control unit (130) may collect indication information directly related to the patient's physical condition or treatment process, such as the patient's diagnosis, pain location, symptom intensity, treatment stage, and treatment history, from at least one of the database (200) and the linked external server.

[0117] Meanwhile, the control unit (130) can collect survey response information from at least one of the database (200) and the associated external server. Specifically, the control unit (130) can collect survey response information for at least one survey provided to the user terminal (10).

[0118] For example, the control unit (130) may provide at least one survey regarding the patient's indications on a service page output to the user terminal (10). Here, the at least one survey may include a multiple-choice survey consisting of at least one multiple-choice question item regarding the patient's indications and a plurality of selection items corresponding to the multiple-choice question item.

[0119] As illustrated in FIG. 3b, the control unit (130) may provide a survey page (or service page, 1000) containing a plurality of surveys on a user terminal (10). In this case, the service page (1000) containing a plurality of surveys may be output through a touch screen (or display) of the user terminal (10). Here, the survey (or question, or problem, or item, or test) provided on the page may include a survey related to the patient's indication.

[0120] Specifically, the control unit (130) may provide a plurality of multiple-choice survey questions (341, 342) related to indications on a service page (1000) output to a user terminal (10). At this time, the multiple-choice survey questions (341, 342) provided to the user terminal (10) in the present invention are not limited to the examples described and may be very diverse.

[0121] As an example, the control unit (130) may provide at least one of the first multiple-choice survey question (341) and the second multiple-choice survey question (342) to the service page (1000) and receive a response to at least one multiple-choice survey question (341, 342) from the user terminal (10). Specifically, the multiple-choice survey may be composed of a multiple-choice question item related to the patient's indication and a plurality of selection items corresponding to each of the plurality of different responses to the multiple-choice question item, forming pairs. At this time, the format of at least one multiple-choice question item constituting the multiple-choice survey according to the present invention may be very diverse.

[0122] Specifically, the first multiple-choice survey item (341) may be composed of a question related to the indication (e.g., “How is your current emotional state?”) and a plurality of choice items corresponding to each of the multiple different responses to the said question (e.g., “Good,” “Bad”), and each of the multiple different responses may have a different score related to the emotional state matched to it. For example, among the multiple different responses (e.g., “Good,” “Bad”) composed of the first multiple-choice survey item (311), the first response (e.g., “Good”) may have a first score related to cognitive behavioral therapy (e.g., “4 points”) matched to it, and the fifth response (e.g., “Bad”) may have a fifth score related to cognitive behavioral therapy (e.g., “0 points”) matched to it.

[0123] As another example, the second multiple-choice question item (342) may be composed of a pair of question items related to the patient's indication (e.g., “Please select the item that best represents the overall condition?”) and multiple selection items corresponding to each of the multiple different responses to the question item (e.g., “I feel refreshed but my concentration is slightly lacking,” “I feel refreshed after exercise and have a great sense of accomplishment,” etc.). In this case, the multiple different responses may include selection items expressed in different natural languages ​​related to the indication.

[0124] Accordingly, as a response to at least one question item constituting a multiple-choice survey, the user may select a specific option that is considered to be the most suitable (or appropriate or accurate) for the user's condition among multiple option items corresponding to each of multiple different responses.

[0125] Meanwhile, as illustrated in FIG. 3c, the control unit (130) may provide at least one open-ended survey in relation to cognitive behavioral therapy on a service page (1000) output to a user terminal (10). Specifically, the at least one survey provided on the service page may further include an open-ended survey capable of receiving natural language input for at least one open-ended question item from the user terminal in relation to the indications of the patient.

[0126] For example, the control unit (130) may provide at least one open-ended survey (351, 352, 353) corresponding to different attributes to the user terminal (10). For example, the control unit (130) may provide a first open-ended survey (351) corresponding to a first attribute (e.g., pain) to the user terminal (10). At this time, the first open-ended survey (351) may provide an open-ended question item related to the first attribute (e.g., “Please freely describe in your own words the pattern or characteristics of the pain you are currently feeling”) and an input area for inputting a natural language response to the open-ended question item. Here, the input area may refer to an interface capable of receiving a natural language response from at least one of text and voice.

[0127] At this time, the user (or patient) can input a user response to a subjective survey question that is entered into the input area through the user terminal (10). Specifically, the user can input response text to the subjective survey question into the input area.

[0128] The control unit (130) according to the present invention may further include a module that converts voice input into an input area into text. For example, the control unit (130) may input voice data received from a user terminal (10) into a voice conversion module. Specifically, the voice conversion module may include at least one voice recognition model based on a Speech-to-Text (STT) algorithm that converts voice data into text. As an example, the voice conversion module may include at least one voice recognition model among a Hidden Markov Model (HMM), a Hidden Markov Model-Gaussian Mixture Model (HMM-GMM), a Connectionist Temporal Classification (CTC) based model, a Beam Search based model, a Deep Neural Network (DNN) based model, a Recurrent Neural Network (RNN) based model, a Sequence-to-Sequence (Seq2Seq) model, and a Transformer based model.

[0129] At this time, the control unit (130) can convert voice (or voice data) into text based on a STT (Speech-to-Text) algorithm and process the converted text as a user response to a subjective survey question.

[0130] According to the present invention, at least one questionnaire is not limited to the examples of questionnaires (or questions) described above, and may include at least one questionnaire among multiple-choice questionnaires and open-ended questionnaires, and the at least one questionnaire may include various questionnaires (or questions) related to the indications of the patient.

[0131] In the present invention, the control unit (130) can collect survey response information for at least one survey provided to the user terminal (10). Specifically, the control unit (130) can receive objective survey response information corresponding to objective surveys included in at least one survey.

[0132] The control unit (130) may receive survey response information (or multiple-choice survey response information) for a multiple-choice survey that includes a response matched to an item selected by user input among a plurality of selection items from a user terminal (10). At this time, the multiple-choice survey response information may include natural language response information corresponding to a specific selection item selected by user input among a plurality of selection items.

[0133] For example, referring again to FIG. 3b, the control unit (130) can receive a natural language response (e.g., "I am emotionally sensitive or irritable due to high physical fatigue") corresponding to a selection item selected by user input among multiple selection items for a multiple-choice survey question (e.g., "Please select the item that best represents the overall condition") included in a specific multiple-choice survey question (342) related to the user's indication (e.g., "Please select the item that best represents the overall condition") as multiple-choice survey response information.

[0134] Furthermore, the control unit (130) may receive subjective survey response information for a subjective survey included in at least one survey. At this time, the subjective survey may include natural language input for a subjective question item and natural language response information corresponding to the natural language input. As an example, referring again to FIG. 3c, the control unit (130) may receive a natural language input for a subjective question item (351) as subjective survey response information.

[0135] That is, the control unit (130) may receive survey response information including at least one of objective survey response information and subjective survey response information from a user terminal (10) and store it in a database (or storage unit (120)). Furthermore, the control unit (130) may collect survey response information stored in the database (200) in order to collect patient information related to the patient's indications.

[0136] As illustrated in FIG. 3d, the control unit (130) can collect memo information (360) entered by the user (or patient) through the user terminal (10). Specifically, the control unit (130) can store text entered by the user through a memo input area provided on the service page (1000) of the user terminal (10) as memo information in the database (or storage unit (120)). For example, the control unit (130) can store text (361, 362) entered periodically according to a preset execution date in relation to an exercise program performed by the user (or patient) as daily memo information (360) in the database (or storage unit (120)).

[0137] Alternatively, the control unit (130) can receive text entered by the user (or patient) in the memo input area at any point in time in real time, regardless of the exercise program execution, and store it in the database (or storage unit (120)) as memo information (360) along with a timestamp of the time of text input.

[0138] That is, the control unit (130) can receive not only periodic memo information associated with the user's exercise program execution, but also all memo information that is frequently entered during daily life. Furthermore, the control unit (130) can collect memo information (360) stored in the database (200) in order to collect patient information related to the patient's indications.

[0139] Next, in the present invention, a process of extracting emotional information related to the patient's emotional state from patient information may be performed (S220, see FIG. 2).

[0140] As illustrated in FIG. 4a, the control unit (130) can extract emotional information (400) based on patient information using an emotional analysis module (131). Here, emotional information (400) may refer to information related to emotional expression (or emotional state) among the patient information.

[0141] Specifically, the sentiment analysis module (131) can extract sentiment information (400) related to a patient's sentiment expression based on at least one of medical information (310) and user feedback information (330). For example, the sentiment analysis module (131) can extract sentiment information (400) related to a sentiment expression from at least one of survey response information (370) and memo information (380). Specifically, the sentiment analysis module (131) can identify at least one word, phrase, and sentence related to a sentiment expression from at least one of survey response information (370) and memo information (380). As an example, the sentiment analysis module (131) can identify at least one word, phrase, and sentence related to a sentiment expression through Natural Language Processing (NLP)-based morphological analysis, part-of-speech tagging, syntactic parsing, and attention weight analysis.

[0142] Specifically, the control unit (130) can extract emotion information corresponding to each day based on patient information entered sequentially by day using an emotion analysis module (131). For example, the control unit (130) can extract emotion information (401 to 403) for a plurality of performance days from at least one of survey response information (370) and memo information (380) entered for each performance day set in the exercise program performed by the user.

[0143] At this time, the emotional information may include at least one of a first type of emotional information related to the patient's general emotions and a second type of emotional information related to the patient's indications. Specifically, the control unit (130) can extract different types of emotional information from the patient information using an emotional analysis module (131). Specifically, the emotional analysis module (131) can extract emotional information from the patient information that includes at least one of a first type of emotional information related to general emotions and a second type of emotional information related to the patient's indications.

[0144] As previously explained, general emotion according to the present invention may refer to emotion arising from daily experiences, interpersonal relationships, environmental factors, or personal emotional responses that are not directly related to the indications or treatment status of the patient (or user). For example, general emotion may include emotions related to non-medical emotions formed by daily situations or external stimuli, such as fatigue, lethargy, stress, mood depression, tension, and instability.

[0145] In contrast, emotions regarding indications may refer to emotions associated with a physical condition directly related to the patient's disease, symptoms, treatment, or rehabilitation process. For example, emotions regarding indications may include emotions directly associated with the indication, such as at least one of anxiety due to pain, frustration due to delayed recovery, fear of worsening symptoms, tension during the treatment process, and anticipation of recovery.

[0146] As illustrated in FIG. 4b, the emotion analysis module (131) can classify the type of emotion information based on indication information (311) corresponding to the patient's indication. Specifically, the emotion analysis module (131) can classify the type of input emotion information by analyzing at least one of the disease name, symptom area, and prescription information included in the patient's indication information (311) in conjunction with the emotion expression.

[0147] For example, the sentiment analysis module (131) can classify a sentence (e.g., “I cannot sleep well because the pain in my lower back after surgery gets worse every night”, 412) included in at least one of the survey response information (370) and memo information (380) as a second type of sentiment information (420) related to the patient’s sentiment regarding the indication, if the sentence includes the name of the pain area (e.g., “lower back”) within the indication information (311).

[0148] In contrast, the emotion analysis module (131) may classify a sentence (e.g., “I get tired easily these days and don’t feel motivated,” 411) included in at least one of the survey response information (370) and memo information (380) as a first type of emotion information (410) related to general emotion if it is not directly associated with indication information (311), such as daily fatigue, stress, or mood deterioration.

[0149] Next, in the present invention, based on the extracted emotional information, a process may be carried out to analyze the emotional state related to the indication and generate a prompt requesting the creation of feedback content related to the emotional state (S230, see FIG. 2).

[0150] The control unit (130) can extract emotional information related to the patient's emotional state from patient information and analyze the emotional state based on the extracted emotional information. Specifically, the control unit (130) can analyze the emotional information using an emotional analysis module (131) and generate an emotional state analysis result corresponding to the emotional information. In the following description, the operation of the emotional analysis module (131) may also be understood as being performed by the control unit (130).

[0151] The method for analyzing emotional information and generating an emotional state analysis result corresponding to the emotional information according to the present invention can be very diverse, and the present specification is not limited to any type or method as long as it is a module capable of analyzing emotional information and generating an emotional state analysis result corresponding to the emotional information.

[0152] For example, the emotion analysis module (131) can be linked with a large language model (140) to generate an emotion state analysis result regarding the patient's emotional state from emotion information. Specifically, the emotion analysis module (131) can use a prompt generation unit to generate a prompt that analyzes the patient's emotional state based on emotion information. Furthermore, the emotion analysis module (131) can process the prompt as input to the large language model (140) to obtain an emotion state analysis result from the large language model (140).

[0153] Referring again to FIG. 4b, the emotion analysis module (131) can generate an emotion state analysis result (430) corresponding to each of different types of emotion information (400). Specifically, the emotion analysis module (131) can analyze the extracted emotion information to generate an emotion state analysis result for the patient's emotion state. For example, the emotion analysis module (131) can generate at least one of a first type emotion state analysis result corresponding to a first type emotion information (410) related to general emotion and a second type emotion state analysis result corresponding to a second type emotion information (420) related to the patient's indications.

[0154] Furthermore, the emotion analysis module (131) can extract (or identify) emotion keywords for the emotion state from the emotion state analysis results and produce an emotion class corresponding to the patient's emotion state based on the extracted emotion keywords. Here, the emotion class is a predefined set of emotion types that express the patient's emotion state and may refer to a standardized emotion expression unit for classifying the direction (positive, neutral, negative), intensity, and major emotion categories (e.g., anxiety, depression, lethargy, stability, confidence, etc.) of the emotion derived from the emotion state analysis results.

[0155] For example, if the emotion analysis module (131) identifies “worry” as an emotion keyword in the sentence “I am worried because the pain is severe,” it can map the user’s emotional state to the ‘anxiety’ class.

[0156] For example, the emotion analysis module (131) can extract emotion keywords regarding the patient's emotional state from the emotional state analysis results using a pre-prepared keyword extraction algorithm. Furthermore, the emotion analysis module (131) can generate an emotion keyword vector by embedding the extracted emotion keywords and calculate a similarity with a pre-stored emotion class embedding vector.

[0157] Specifically, the control unit (130) can calculate the similarity between an emotion class vector and an emotion keyword vector that are stored in a database. For example, the control unit (130) can calculate the similarity based on at least one of cosine similarity and Euclidean distance between the emotion class vector and the emotion keyword vector.

[0158] Furthermore, the control unit (130) may specify a specific emotion class corresponding to at least one emotion class vector whose calculated similarity satisfies a preset similarity condition as an emotion class corresponding to an emotion state. Here, the preset similarity condition may refer to a condition satisfied based on the similarity between a preset emotion class vector and an emotion keyword vector being greater than or equal to a threshold value. At this time, the threshold value is not limited to a specific value and can be set to various values ​​and can be changed by the system.

[0159] As illustrated in FIG. 5, the control unit (130) can generate a prompt (500a) requesting the generation of feedback content related to an emotional state. Specifically, the control unit (130) can generate a prompt (500a) requesting the generation of feedback content related to an emotional state based on at least one of emotional information (400), emotional state analysis result (430), and patient information (300) using a prompt generation unit (133).

[0160] More specifically, the control unit (130) can generate a first type prompt (510) requesting the generation of feedback content related to the patient's general emotions. Additionally, the control unit (130) can generate a second type prompt (520) requesting the generation of feedback content related to the patient's indications.

[0161] For example, the prompt generation unit (133) can generate a first type prompt (510) based on at least one of patient information, first type emotion information, and first type emotion state analysis result. Additionally, the prompt generation unit (133) can generate a second type prompt (520) related to patient information, second type emotion information, and second type emotion state analysis result.

[0162] That is, the prompt generation unit (133) can generate a prompt (500a) including at least one of a first type prompt (510) and a second type prompt (520).

[0163] Next, in the present invention, the generated prompt may be processed as input to a pre-trained feedback generation model to obtain feedback content related to an emotional state (S240, see FIG. 2).

[0164] As illustrated in FIG. 6, the control unit (130) can process the generated prompt (500) as input to a pre-trained feedback generation model (132). The pre-trained feedback generation model (132) according to the present invention may refer to an artificial intelligence model trained to generate patient-customized feedback content in conjunction with a large language model.

[0165] For example, the control unit (130) can fine-tune (or tune, optimize) a large language model based on multiple patient information. Specifically, the control unit (130) can train a feedback generation model (132) to generate customized feedback content that adaptively responds to the patient's emotional state by fine-tuning the parameters of the large language model using emotional information, indication information, and user feedback data included in the patient information as training data.

[0166] More specifically, the pre-trained feedback generation model (132) according to the present invention may refer to a generative model that fine-tunes a pre-trained large language model by utilizing emotional information, indication information, and user feedback data included in patient information as training data. There may be a wide variety of methods for fine-tuning the pre-trained large language model according to the present invention, and the present specification is not limited to the type and method as long as it is a module capable of fine-tuning the pre-trained large language model.

[0167] For example, a previously trained feedback generation model (132) can adjust the decoder or adapter layer of a previously trained large language model using a training set consisting of pairs of emotion classes (e.g., anxiety, helplessness, pain-fear, etc.) corresponding to the patient's emotional state and corresponding feedback content (e.g., encouraging messages, calming sentences, etc.).

[0168] Additionally, the control unit (130) can retrain some parameters of the large language model using the training set. At this time, the control unit (130) can perform overfitting prevention procedures such as validation set separation, early stopping, and weight decay, perform iterative training (epoch), and adjust hyperparameters such as the learning rate and batch size.

[0169] Alternatively, the control unit (130) may use at least one pre-trained large language model included in the LLM (140) as a pre-trained feedback generation model (132). In this case, the pre-trained feedback generation model (132) may be included in the LLM server (140), and the LLM server (140) may transmit a prompt (500) generated by the control unit (130) to the LLM server (140), so that the LLM server (140) may process the prompt (500) received from the control unit (130) as an input to the pre-trained feedback generation model (132).

[0170] That is, the pre-learned feedback generation model (132) according to the present invention is not limited to an artificial intelligence model that operates independently in the feedback providing system (100), but can be operated in a distributed processing form by linking with an external server or a cloud-based LLM server (140).

[0171] In addition, the feedback generation model (132) according to the present invention is not limited to a model structure or learning method and may include any form of generative model capable of generating feedback content related to a patient's emotional state based on emotional information.

[0172] The control unit (130) can obtain feedback content corresponding to the input prompt (500) from the previously learned feedback generation model (132). Specifically, the control unit (130) can obtain feedback content of a type corresponding to the type of the input prompt from the previously learned feedback generation model (132). For example, the previously learned feedback generation model (132) generates first type feedback content (610) corresponding to the first type prompt, and the control unit (130) can obtain the first type feedback content (610) from the previously learned feedback generation model (132). At this time, the first type feedback content (610) may refer to feedback content related to the patient's general emotions.

[0173] As another example, a previously learned feedback generation model (132) generates second-type feedback content (620) corresponding to a second-type prompt, and the control unit (130) can obtain the second-type feedback content (620) from the previously learned feedback generation model (132). In this case, the second-type feedback content (620) may refer to feedback content related to the patient's feelings regarding the indication.

[0174] That is, the control unit (130) can obtain feedback content (600a) including at least one of a first type feedback content (610) and a second type feedback content (620) from a previously learned feedback generation model (132).

[0175] Next, in the present invention, a process of providing feedback content related to an emotional state to a user terminal may be carried out (S250, see FIG. 2).

[0176] As previously explained, the control unit (130) can obtain feedback content to be provided to the user terminal (10) from a previously learned feedback generation model (132). Furthermore, the control unit (130) can provide feedback content to the patient through a service page provided to the user terminal (10).

[0177] The feedback content according to the present invention may refer to content comprising at least one of text, image, video, and audio for alleviating or improving emotional states such as pain, anxiety, and recovery process in relation to a patient's indication.

[0178] At this time, the feedback content can be provided to the user terminal (10) in a wide variety of ways, and the present invention does not limit the method of providing the content or the specific output medium. For example, the control unit (130) can provide the feedback content to the user terminal (10) in at least one of text notifications, pop-up messages, voice guidance, and video playback.

[0179] Meanwhile, the control unit (130) can determine the timing for providing feedback content to the user terminal (10). Specifically, the control unit (130) can detect the occurrence of a pre-set feedback event. Furthermore, the control unit (130) can provide feedback content to the user terminal in response to detecting the occurrence of a pre-set feedback event. The feedback event according to the present invention can be set in a wide variety of ways. For example, the control unit (130) can pre-set the time of the first login of a user account to the user terminal (10) on a specific date as a feedback event.

[0180] As an example, the control unit (130) may set the first login of the user account on each of the execution dates of the exercise program performed by the user as a feedback event. Specifically, the control unit (130) may detect the time when the user account first logs into the user terminal (10) on the execution date and detect that a pre-set feedback event has occurred based on the fact that the user account has logged in.

[0181] As another example, the control unit (130) can detect a change in the patient's emotional state and determine the timing for providing feedback content to the user terminal (10). Specifically, the control unit (130) can detect that a preset feedback event has occurred in response to detecting a change in the type of the emotion class corresponding to the patient's emotional state.

[0182] To this end, the control unit (130) can monitor an emotional state related to the indication. Specifically, the control unit (130) can monitor a change in the type of emotional class corresponding to the patient's emotional state.

[0183] As illustrated in FIG. 6b, the control unit (130) can monitor changes in emotion classes corresponding to the emotion states analyzed sequentially over time.

[0184] As previously explained, an emotion class is a predefined set of emotion types representing a patient's emotional state, and can refer to a standardized unit of emotional expression used to classify the direction (positive, neutral, negative), intensity, and major emotional categories (e.g., anxiety, depression, lethargy, stability, confidence, etc.) of emotions derived from the emotional state analysis results. In this case, the type of the emotion class serves as a superordinate category for classifying the emotion class, and can refer to a criterion indicating the overall tendency of the emotion class derived from the emotional state analysis results.

[0185] The method of distinguishing the types of emotion classes according to the present invention can be very diverse and is not limited to a specific method. For example, the types of emotion classes according to the present invention may include at least one of a first type emotion class (e.g., negative) and a second type emotion class (e.g., positive).

[0186] In this case, specific types of emotion classes may be pre-classified and mapped to multiple emotion classes. For example, an emotion class may be pre-mapped to either a Type 1 emotion class (e.g., negative) or a Type 2 emotion class (e.g., positive). As an example, emotion classes such as 'anxiety', 'lethargy', and 'pain-fear' may be mapped to the Type 1 emotion class (negative), while emotion classes such as 'confidence', 'sense of stability', and 'increased motivation' may be mapped to the Type 2 emotion class (positive).

[0187] The control unit (130) can monitor changes in the type of emotion class corresponding to the patient's emotional state based on the results of the emotional state analysis generated sequentially at different points in time according to the flow of time. For example, the control unit (130) can identify the emotion class corresponding to the first emotional state analysis result (440) regarding the patient's emotional state analyzed at the first point in time (e.g., day 1 of the exercise program). Furthermore, the control unit (130) can identify the emotion class corresponding to the second emotional state analysis result (450) regarding the patient's emotional state analyzed at the second point in time (e.g., day 2 of the exercise program).

[0188] The control unit (130) monitors changes in emotion classes corresponding to the emotion states analyzed sequentially over time, and in response to detecting a change in the type of emotion class, determines that a feedback event has occurred. Furthermore, the control unit (130) determines the time when the feedback event occurs as the time for providing feedback, and can provide feedback content to the user terminal (10) at the time for providing feedback.

[0189] For example, the control unit (130) may determine that a change in the type of emotion class has been detected if the first emotion state analysis result (440) regarding the patient's emotion state analyzed at the first time point (e.g., day 1 of the exercise program) corresponds to a first type emotion class (e.g., negative, 630) and the second emotion state analysis result (450) regarding the patient's emotion state analyzed at the second time point (e.g., day 2 of the exercise program) corresponds to a second type emotion class (e.g., positive, 640).

[0190] Furthermore, the control unit (130) determines that a change in the type of the emotion class has been detected, determines the time when the feedback event occurs as the time for providing feedback, and can provide feedback content to the user terminal (10) at the time for providing feedback.

[0191] At this time, the control unit (130) can generate feedback content based on the emotional state monitoring results using a previously learned feedback generation model. As illustrated in FIG. 6c, the control unit (130) can generate an emotional state monitoring result (650) related to a change in the type of emotional class corresponding to the emotional state analysis result.

[0192] Furthermore, the control unit (130) can use the prompt generation unit (133) to generate a prompt (500b) requesting the generation of feedback content (600b) related to the patient's emotional state change based on the emotional state monitoring result (650). The control unit (130) can process the generated prompt (500b) as input to a pre-trained feedback generation model (132).

[0193] Since the operation process of the prompt generation unit (133) and the previously learned feedback generation model (132) according to the present invention has been explained in detail above, it will be omitted to avoid duplication of explanation.

[0194] Furthermore, the control unit (130) can obtain feedback content (600b) related to changes in the patient's emotional state from a previously learned feedback generation model (132). Furthermore, the control unit (130) can provide the feedback content (600b) related to changes in the patient's emotional state to the user terminal (10).

[0195] Meanwhile, the control unit (130) may provide a user interface to the user terminal (10) for providing at least one of a feedback program and a counseling program corresponding to the emotional state, along with feedback content. The feedback program according to the present invention may include an intervention-type program for regulating the patient's emotional state or inducing psychological recovery. For example, the feedback program may include content for regulating the patient's emotional state or inducing psychological recovery. As an example, the feedback program may include at least one of breathing control, meditation guide, stretching guide, music for psychological stability, or simple self-training content for emotional recovery.

[0196] At this time, the feedback program according to the present invention may be stored in advance in a storage unit (120, database (200)) corresponding to an emotion class corresponding to the result of the emotional state analysis. The control unit (130) may extract the feedback program mapped to the emotion class corresponding to the result of the emotional state analysis of the patient's emotional state from the storage unit (120, database (200)) and provide it to the user terminal (10).

[0197] As illustrated in FIG. 7a, the control unit (130) may provide at least one of feedback content (600) and a feedback program through a service page (1000) provided to the user terminal (10). Specifically, the control unit (130) may provide a user interface (710) for providing a feedback program to the user terminal (10).

[0198] Furthermore, the control unit (130) can detect user input that is input into a user interface (710) for providing a feedback program. Specifically, it can detect user input (711) for a specific icon requesting the provision of said feedback program among at least one icon provided in the user interface (710) for providing a feedback program.

[0199] At this time, user input (711) can be performed in various ways, and the present invention does not limit the method of user input. For example, user input according to the present invention can be performed in at least one of tap, double tap, long press, click, swipe, drag, pinch in, pinch out, and rotate.

[0200] As illustrated in FIG. 7b, the control unit (130) can provide a feedback program (720) to a service page (1000) provided to a user terminal (10) in response to detecting user input (710) for a specific icon requesting the provision of a feedback program.

[0201] Meanwhile, the control unit (130) may provide a user interface to the user terminal (10) for providing a counseling program along with feedback content. At this time, the counseling program according to the present invention may include a counseling service that generates and provides answers to a patient's user query based on a large language model.

[0202] As illustrated in FIG. 7c, the control unit (130) may provide at least one of feedback content (600) and a counseling program through a service page (1000) provided to the user terminal (10). Specifically, the control unit (130) may provide a user interface (720) for providing a counseling program to the user terminal (10). The counseling program according to the present invention may provide an answer to a user query entered in relation to at least one of the patient's indications, emotional state, and feedback content, based on a pre-trained large language model (140).

[0203] For example, the control unit (130) can detect user input entered into a user interface (720) for providing a consultation program. Specifically, it can detect user input (721) for a specific icon requesting the provision of the consultation program among at least one icon provided in the user interface (720) for providing a consultation program. At this time, the user input (721) can be performed in various ways, and the present invention does not limit the method of user input. For example, the user input according to the present invention can be performed in at least one of a tap, double tap, long press, click, swipe, drag, pinch in, pinch out, and rotate.

[0204] As illustrated in FIG. 7d, the control unit (130) may provide a consultation page (730) to the user terminal (10) in response to detecting user input (721) for a specific icon requesting the provision of a consultation program. The consultation page (730) according to the present invention may include an input area in which a patient can input a user query. The control unit (130) may receive a user query (731) entered through the input area provided on the consultation page (730).

[0205] The control unit (130) can generate an answer (732) to the user query (731) in conjunction with the large language model (140). Specifically, the control unit (130) can process the user query (731) as input to the large language model (140) and obtain an answer (732) to the user query (731) from the large language model (140).

[0206] Furthermore, the control unit (130) can output an answer (732) to a user query (731) on a consultation page (730) provided to the user terminal (10). Through this, the control unit (130) can provide a counseling program to regulate the patient's emotional state or induce psychological recovery in relation to the patient's emotional state.

[0207] In the present invention, at least one of a feedback program and a counseling program is provided to a user terminal (10) along with feedback content, but is not limited thereto, and at least one of the feedback program and the counseling program may be provided at a time different from the time of providing the feedback content. For example, the control unit (130) may output a separate user interface (UI) for receiving at least one of the feedback program and the counseling program on a service page (1000) provided to the user terminal (10). At this time, the control unit (130) may control the user so that the user can directly request at least one of the feedback program and the counseling program at a time desired through the user interface.

[0208] That is, in the present invention, the timing and method of providing at least one of the feedback program and the counseling program are not fixed, and it can be understood that they may be provided in parallel with the feedback content or provided independently at a separate time depending on the system settings or the user's request.

[0209] Meanwhile, the control unit (130) can provide an emotional state analysis report related to the patient's emotional state along with feedback content to the user terminal (10). Specifically, the control unit (130) can generate different types of emotional state analysis reports based on emotional state analysis results corresponding to each type of emotional information.

[0210] For example, as illustrated in FIG. 7e, the control unit (130) can generate a first type emotional state analysis report (740) by analyzing first type emotional information. Specifically, the control unit (130) can generate a first type emotional state analysis report (740) based on at least one of the first type emotional state analysis results related to patient information and the patient's general emotions.

[0211] More specifically, the control unit (130) can generate a first type emotional state analysis report (740) based on at least one of the first type emotional state analysis results related to patient information and the patient's general emotions in conjunction with a large language model (140).

[0212] For example, the control unit (130) may use a prompt generation unit to generate a prompt requesting the generation of an emotional state analysis report according to a pre-set report format, based on at least one of the first type emotional state analysis results related to patient information and the patient's general emotions. At this time, the pre-set report format may refer to a data configuration format or output format for structuring and expressing the emotional state analysis results, and the configuration or expression method may be very diverse and is not limited to a specific format. As an example, the pre-set report format may refer to a report format configured to generate an emotional state analysis report based on a configuration that summarizes the emotional state, explains the trend of emotional change, and suggests an interpretation of the emotional state and related actions.

[0213] As an example, the control unit (130) may generate a prompt requesting the generation of a first-type emotional state analysis report (740) corresponding to a plurality of first-type emotional state analysis results accumulated over a specific period. Specifically, the control unit (130) may generate a prompt requesting the generation of a first-type emotional state analysis report (740) corresponding to a week of performance based on a plurality of first-type emotional analysis results analyzed during a week of performance set in an exercise program performed by a patient.

[0214] Furthermore, the control unit (130) processes the generated prompt as input to the large language model (140) and can obtain a first type emotional state analysis report (740) for at least one of the first type emotional state analysis results related to patient information and the patient's general emotions from the large language model (140).

[0215] The control unit (130) may provide a first type emotional state analysis report (740) related to the patient's general emotions to a service page (1000) provided to the user terminal (10). Specifically, the control unit (130) may provide a first type emotional state analysis report (740) comprising at least one of emotional state summary information (741) related to the patient's general emotions, emotional change trend information (742) related to the patient's general emotions, and interpretation information (743) related to the patient's general emotions.

[0216] As another example, as illustrated in FIG. 7f, the control unit (130) can generate a second type emotional state analysis report (750) by analyzing second type emotional information. The control unit (130) can generate a second type emotional state analysis report (750) based on at least one of the second type emotional state analysis results related to emotions regarding patient information and the patient's indications. Specifically, the control unit (130) can generate a second type emotional state analysis report (750) based on at least one of the second type emotional state analysis results related to emotions regarding patient information and the patient's indications in conjunction with a large language model (140).

[0217] For example, the control unit (130) can use the prompt generation unit to generate a prompt requesting the generation of an emotional state analysis report according to a preset report format, based on at least one of the second type emotional state analysis results related to patient information and the patient's indications.

[0218] As an example, the control unit (130) may generate a prompt requesting the generation of a second-type emotional state analysis report (750) corresponding to a plurality of second-type emotional state analysis results accumulated over a specific period. Specifically, the control unit (130) may generate a prompt requesting the generation of a second-type emotional state analysis report (750) corresponding to a week of performance based on a plurality of second-type emotional analysis results analyzed during a week of performance set in an exercise program performed by a patient.

[0219] Furthermore, the control unit (130) processes the generated prompt as input to the large language model (140) and can obtain a second type emotional state analysis report (750) for at least one of the second type emotional state analysis results related to patient information and the patient's indications from the large language model (140).

[0220] The control unit (130) may provide a second type emotional state analysis report (750) related to the patient's indications to a service page (1000) provided to the user terminal (10). Specifically, the control unit (130) may provide a second type emotional state analysis report (750) comprising at least one of summary information on the patient's indications (741), information on the trend of emotional changes related to the patient's indications (742), and interpretation information related to the patient's indications (743).

[0221] Furthermore, the control unit (130) can generate a comprehensive emotional state analysis report related to the patient's emotional state based on the first type emotional state analysis result and the second type emotional state analysis result. As illustrated in FIG. 7g, the control unit (130) can comprehensively analyze the patient's emotional changes by comparing the first type emotional state analysis result and the second type emotional state analysis result. For example, the control unit (130) can use a prompt generation unit to generate a prompt requesting to compare and analyze the results of different types of emotional state analysis results.

[0222] Specifically, the prompt generation unit can generate a prompt requesting analysis by comparing the Type 1 emotional state analysis results and the Type 2 emotional state analysis results according to a pre-configured report format. In this case, the pre-configured report format may refer to a data structure or output format for integrally expressing the results of different types of emotional state analysis, and the method of configuration may be very diverse and is not limited to a specific format.

[0223] For example, a pre-configured report format may mean a format that generates a comprehensive emotional state analysis report including at least one of a report analyzing the flow of changes in Type 1 emotions and Type 2 emotions, and a comprehensive evaluation report on Type 1 emotions and Type 2 emotions.

[0224] As an example, the control unit (130) may generate a prompt requesting the generation of a comprehensive emotional state analysis report corresponding to a plurality of emotional state analysis results cumulatively analyzed over a specific period. Specifically, the control unit (130) may generate a prompt requesting the generation of a comprehensive emotional state analysis report corresponding to a week of performance based on at least one of a comprehensive evaluation report for a plurality of first-type emotions and a comprehensive evaluation report for a plurality of second-type emotions analyzed during a week of performance set in an exercise program performed by a patient.

[0225] The control unit (130) processes the generated prompt as input to the large language model (140) and can obtain a comprehensive emotional state analysis report (760) regarding the first type emotional state analysis result and the second type emotional state analysis result from the large language model (140). Furthermore, the control unit (130) can output (or provide) the comprehensive emotional state analysis report (760) to the service page (1000) provided to the user terminal (10).

[0226] In the present invention, an emotional state analysis report is provided to a user terminal (10) along with feedback content, but this is not limited thereto, and only the emotional state analysis report may be provided at a time different from the time the feedback content is provided. For example, the control unit (130) may output a separate user interface (UI) for receiving the emotional state analysis report on a service page (1000) provided to the user terminal (10). At this time, the control unit (130) may control the user so that the user can directly select or request the emotional state analysis report at a time of their choice through the user interface.

[0227] In other words, it can be understood that in the present invention, the timing and method of providing the emotional state analysis report are not fixed, and may be provided in parallel with feedback content or independently at a separate time depending on the system settings or user requests.

[0228] As described above, the AI-based patient-customized feedback provision method and system according to the present invention can improve patient participation and immersion in treatment by analyzing patient information based on AI and providing patient-customized feedback content that takes into account the individual patient's condition.

[0229] Furthermore, the AI-based patient-customized feedback provision method and system according to the present invention can analyze the patient's emotional state based on emotional information extracted from patient information and provide feedback content corresponding to the emotional state. Through this, an emotion-centered treatment environment can be realized that enhances psychological stability and willingness to participate in treatment by comprehensively considering the patient's psychological and emotional factors.

[0230] Furthermore, the AI-based patient-customized feedback provision method and system according to the present invention can continuously monitor changes in a patient's emotional state and control the timing of feedback provision based on these changes. Accordingly, by detecting changes in the patient's emotions in real time, immediate and empathetic feedback content can be provided at the necessary time, which can simultaneously improve the patient's emotional stability and treatment engagement.

[0231] Furthermore, the artificial intelligence-based patient-customized feedback provision method and system according to the present invention can realize an integrated treatment environment that simultaneously promotes psychological and physical recovery by providing feedback programs and counseling programs in conjunction with the patient's emotional state.

[0232] The artificial intelligence-based patient-customized feedback provision system (100) according to the present invention can be implemented through a computing device described below and can perform data processing related to the artificial intelligence-based patient-customized feedback provision method described above.

[0233] Meanwhile, FIG. 8 illustrates an example of a block diagram of a computing system in which the present invention can be implemented.

[0234] Referring to FIG. 8, a computing system (10000) that performs an artificial intelligence-based patient-customized feedback provision method according to one embodiment of the present invention may include at least one computing device. At this time, the at least one computing device may be a single processor or a multi-processor computing device.

[0235] The components of at least one computing device of the present invention may include various hardware components such as one or more processors, memory, other hardware, and a system bus (not shown) that connects various system components so that they can transmit and receive data to and from each other (e.g., telecommutatively connected, physically connected, electrically connected), and the components of at least one computing device are not limited thereto and may be very diverse.

[0236] Meanwhile, at least one computing device included in a computing system (10000) that performs an artificial intelligence-based method for providing patient-customized feedback may be connected to communicate via a network (1070). For example, at least one computing device included in the computing system (10000) may be clustered or may be part of a local area network (LAN). Additionally, at least one computing device may be part of a wide area network (WAN) or connected to at least one of a client-server network and a peer-to-peer network within the cloud.

[0237] Meanwhile, when at least one computing device is used in at least one of a network environment and a cloud computing environment, the at least one computing device may be connected to at least one of a public and private network through a network interface or adapter. In one embodiment, other communication connection devices, such as a modem, may be used to establish communication through the network. The modem may be at least one of an internal modem and an external modem, and may be connected to a system bus through a network interface or a specific mechanism, etc. A wireless network component consisting of an interface and an antenna may be coupled to the network through a device such as an access point, a peer computer, etc. In the present invention, the method of connecting at least one computing device to communicate through the network (1070) is not limited, and it may be connected to communicate in a manner different from the described example.

[0238] Furthermore, other computer-type devices and / or systems not shown in FIG. 8 may also interact technically with at least one computing device or other system through one or more connections to the network (1070) via a network interface. Here, the network interface may include network interface equipment such as a physical network interface controller (NIC) or a virtual network interface (VIF).

[0239] The network (1070) of the present invention may include various forms such as the Internet, WLAN (Wireless LAN), Wi-Fi (Wireless-Fidelity), Wi-Fi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), 5G (5th Generation Mobile Telecommunication), Bluetooth (Bluetooth™ Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra-Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi Direct, Wireless USB (Wireless Universal Serial Bus), etc., and in the present invention, data transmission may be performed based on standard communication protocols such as TCP / IP, HTTP, SSL, etc.

[0240] A computing system (10000) that performs an artificial intelligence-based patient-customized feedback provision method according to the present invention may include at least one of a user computing device (1010, or user computing system), a training computing system (1050, or training computing device), and a server computing system (1030, or server computing device).

[0241] A user computing device (1010) according to the present invention may be understood as a computing device comprising at least one processor (1011) and a memory (1012) for performing an artificial intelligence-based patient-customized feedback provision method. For example, the user computing device (1010) may include at least one computing device among a smartphone, a smart TV, a laptop computer, a desktop computer, a digital broadcasting terminal, a PDA (personal digital assistants), a PMP (portable multimedia player), a navigation device, a slate PC, a tablet PC, an ultrabook, a wearable device (e.g., a smartwatch, a smart glass, and a head-mounted display).

[0242] At least one processor (1011) constituting the user computing device (1010) may include one or more general-purpose processors and / or one or more special-purpose processors. For example, at least one processor (1011) constituting the user computing device (1010) may be composed of at least one or a plurality of electrically connected processors among a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a neural network processing unit (NPU), an arithmetic logic unit (ALU), a floating-point arithmetic unit (FPU), an application integrated circuit, an application semiconductor (ASIC), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, and / or other electrical units for performing functions.

[0243] Furthermore, at least one processor (1011) may be configured to execute computer-readable instructions contained in memory (1012) and / or other instructions described herein.

[0244] The memory (1012) constituting the user computing system (1010) according to the present invention may include volatile memory, non-volatile memory, fixed media, removable media, magnetic media, optical media, semiconductor media and / or other types of physically durable storage media.

[0245] For example, the memory (1012) may include one or more non-transient / transient computer-readable storage media such as RAM, ROM, HDD (Hard Disk Drive), SSD (Solid State Disk), SSD (Silicon Disk Drive), EEPROM, EPROM, flash memory device, magnetic disk, and combinations thereof, and may include web storage of a server that performs the storage function of memory on the internet. This memory (1012) may store data and instructions necessary for the at least one processor (1011) to provide patient-customized feedback using artificial intelligence.

[0246] A user computing device (1010) may include one or more user input components (1021) that detect user input. For example, the user input component (1021) may also be referred to as a user interface module. The user input component (1021) may include a touch screen, a computer mouse, a keyboard, a keypad, a touchpad, a trackball, a joystick, a voice recognition module, or other similar devices. However, the present invention does not limit the type of user input component (1021). In this case, the user input component (1021) in the present invention does not necessarily mean a hardware means, but can be understood as a channel for receiving input from a user. Meanwhile, the user of the present invention may refer to an automated agent, script, playback software, etc., that operates on behalf of one or more people.

[0247] A user can interact with a computing system (10000) including at least one computing device through input text, touch, voice, movement, computer vision, gestures and / or other forms of input / output using a user input component (1021). For example, the user input component (1021) may include one or more of a command line interface (CLI), a graphical user interface (GUI), a natural user interface (NUI), a voice command interface and / or other user interface (UI) representations.

[0248] Between the user input component (1021) and the user computing device (1010), one or more application programming interface (API) calls may be made based on user input received from the user interface and / or network.

[0249] Here, the expression "based on" may be interpreted to include cases where it is based on the use of a specific configuration, modified from, derived from, influenced by, dependent on, or otherwise derived from a specific configuration. In some embodiments, an API call may be configured for a specific API, which may be interpreted or converted into an API call configured for another API. Here, an API may refer to a defined interface or connection between computers or computer programs.

[0250] In one embodiment, the user computing device (1010) may store at least one machine learning model (1020). For example, the user computing device (1010) may be various machine learning models, such as a plurality of neural networks (e.g., deep neural networks), or other types of machine learning models including non-linear models and / or linear models, which perform an artificial intelligence-based patient-customized feedback provision method based on user queries and patient information, and may be composed of a combination thereof.

[0251] According to an embodiment of the present invention, a user computing device (1010) may perform an artificial intelligence-based patient-customized feedback provision method using a local or / and external machine learning model (1020). Alternatively, the user computing device (1010) may perform an artificial intelligence-based patient-customized feedback provision method using a machine learning model (1040) provided by a server.

[0252] Additionally, according to another embodiment of the present invention, a server computing system (1030) communicating with a user computing device (1010) can provide patient-customized feedback to the user computing device (1010) on an application or / and the web in accordance with a request from a user received through the user computing device (1010).

[0253] In addition, according to another embodiment of the present invention, by linking at least a part of a user computing device (1010) and a server computing system (1030) with each other to perform an artificial intelligence-based patient-customized feedback provision method, patient-customized feedback can be provided to the user.

[0254] Additionally, according to various embodiments of the present invention, a user computing device (1010) and / or a server computing system (1030) can learn machine learning models (1020, 1040) performed in an artificial intelligence-based patient-customized feedback provision method through interaction with a training computing system (1050) that is communicatedly connected via a network (1070). In this case, the training computing system (1050) may be a computing system separate from the server computing system (1030). Alternatively, in some embodiments, the training computing system (1050) may be part of the server computing system (1030) or part of the user computing device (1010).

[0255] Meanwhile, the server computing system (1030) may include at least one processor (1031) and memory (1032). Here, the processor (1031) may be composed of at least one or a plurality of electrically connected processors among a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), a neural network processing unit (NPU), an application integrated circuit, an application semiconductor (ASIC), an arithmetic logic unit (ALU), a floating-point arithmetic unit (FPU), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, and / or other electrical units for performing functions. For example, at least one processor (1031) may include a circuit and a transistor configured to execute instructions from memory (1032).

[0256] The memory (1032) constituting the server computing system (1030) according to the present invention may include volatile memory, non-volatile memory, fixed media, removable media, magnetic media, optical media, semiconductor media, and / or other types of physically durable storage media. For example, the memory (1032) may include one or more non-transient / transient computer-readable storage media such as RAM, ROM, HDD (Hard Disk Drive), SSD (Solid State Disk), SSD (Silicon Disk Drive), EEPROM, EPROM, flash memory device, magnetic disk, etc., and combinations thereof, and may include web storage of a server that performs the storage function of memory over the internet. Additionally, the server computing system (1030) may further include a data storage (data store). For example, the data storage may be composed of at least one of a relational database, a NoSQL database, a data warehouse, and a local file system.

[0257] In the memory (1032) constituting the server computing system (1030) according to the present invention, data and instructions necessary for the at least one processor (1031) to perform the operation of an application for providing artificial intelligence-based patient-customized feedback may be stored.

[0258] In one embodiment, the server computing system (1030) may be composed of a single device or a plurality of computing devices, and these may be configured to operate according to a sequential or parallel computing architecture. Additionally, a distributed processing system may be configured with a plurality of networked devices.

[0259] Meanwhile, the training computing system (1050) may include at least one processor (1051) and memory (1052). The model trainer (1060) is a logical component that executes the training of at least one machine learning model (1020, 1040) and may be implemented in the form of hardware, firmware, or software. For example, the model trainer (1060) may be executed by the processor (1051) after loading training data (1061) stored in a storage device into memory (1052). For example, the model trainer (1060) may be configured to execute one or more operations (e.g., model training, model reconstruction, model validation, model testing) on ​​at least one machine learning model.

[0260] The machine learning model of the present invention may include at least one of a statistical model, an algorithm, a neural network (NN), a convolutional neural network (CNN), a generative neural network (GNN), a Word2Vec model, a Bag of Words model, a TF-IDF (document frequency-inverse document frequency) model, a GPT (Generative Pre-trained Transformer) model (or other autoregressive models), a PPO (Proximal Policy Optimization) model, a nearest neighbor model (e.g., a k-nearest neighbor model), a linear regression model, a K-means clustering model, a Q-learning model, a TD (Temporal Difference) model, a Deep Adversarial Network model, and all other types of models further described herein.

[0261] Specifically, the model trainer (1060) may execute operations to train a machine learning model, and said operations may include at least one of adding, removing, and modifying model parameters. At this time, the training of the machine learning model may be at least one of supervised learning, semi-supervised learning, and unsupervised learning. In one embodiment, the training of the machine learning model may include the step of repeatedly inputting training data (1061) based on epochs and repeatedly performing the machine learning model training process configured in this way. Here, an epoch may refer to a unit in which the entire set of training data (1061) undergoes forward and backpropagation processing once. In some implementations, different levels of training methods (e.g., supervised learning, semi-supervised learning, unsupervised learning) may be used for different epochs.

[0262] The training data (1061) of the present invention may include input data and / or data previously output from at least one machine learning model (e.g., recursive learning feedback). The parameters of at least one machine learning model may include at least one of a seed value, a model node, a model layer, an algorithm, a function, connections between different machine learning models, connections between parameters, machine learning model constraints, and other digital components that influence the output of the machine learning model. In this case, the model connections between different machine learning models may include or represent relationships between model parameters and / or models, which may be dependent or interdependent, hierarchical, and / or static or dynamic. The combinations and configurations of model parameters described herein may be too complex to be maintained or used by human cognitive abilities.

[0263] In the present invention, the machine learning parameters described according to the embodiments are not limited, and a single machine learning model may further include a plurality of model parameters.

[0264] Meanwhile, FIG. 9 illustrates an example of a block diagram of a computing device (1100) that may be included in a user computing device (1010), a server computing system (1030), and a training computing system (1050), as an embodiment of a computing system (10000) in which the present invention can be implemented.

[0265] As illustrated in FIG. 9, the computing device (1100) may include at least one application (e.g., Application 1 to Application N), and each of the at least one application may include a machine learning library and a model execution environment for performing an artificial intelligence-based patient-customized feedback provision method. The at least one application included in the computing device (1100) may communicate with the sensor, context manager, device state manager, or additional component(s) within the computing device (1100) via an Application Programming Interface (API). In one embodiment, the at least one application may interface with device components, such as receiving sensor data or state data or transmitting prediction results to an output device via a public or private API.

[0266] Meanwhile, FIG. 10 illustrates an example of a block diagram in another aspect of a computing device (1200), which is one of the components of a computing system (10000) that performs an artificial intelligence-based patient-customized feedback provision method according to an embodiment of the present invention.

[0267] A computing device (1200) according to the present invention may include at least one application (e.g., Application 1 to Application N), and at least one application may communicate with a central intelligence layer (1210). Each application may interact with a shared model within the central intelligence layer (1210) through an API (e.g., a common API).

[0268] The central intelligence layer (1210) includes one or more machine learning models and may share them among multiple applications or provide them independently to each. In one embodiment, the central intelligence layer (1210) may be integrated as part of an operating system or implemented as a separate logical layer.

[0269] Additionally, the central intelligence layer (1210) can communicate with the central device data layer (1220). The central device data layer (1220) can provide patient information and user queries stored within the computing device (1200) as input data necessary for providing AI-based patient-customized feedback. Each device component (e.g., sensor, state manager, etc.) can communicate with the central device data layer (1220) through a private API, etc.

[0270] The technology described in this specification may be composed of a single or multiple computing devices, and a machine learning model that performs an artificial intelligence-based patient-customized feedback provision method may be executed sequentially or in parallel on one component or multiple distributed components. Data storage, machine learning models, and applications may be distributed and operated locally or over a network, and these configurations can be flexibly applied to various system architectures.

[0271] Meanwhile, computer-readable media include all types of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable media include HDD (Hard Disk Drive), SSD (Solid State Disk), SSD (Silicon Disk Drive), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

[0272] Furthermore, the computer-readable medium may be a server or cloud storage that includes a storage and is accessible to an electronic device via communication. In this case, the computer may download the program according to the present invention from the server or cloud storage via wired or wireless communication.

[0273] Furthermore, in the present invention, the computer described above is an electronic device equipped with a processor, namely a CPU (Central Processing Unit), and no special limitations are placed on its type.

[0274] Meanwhile, the above detailed description should not be interpreted restrictively in all respects but should be considered exemplary. The scope of the invention shall be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the invention are included within the scope of the invention.

Claims

1. A step of collecting patient information related to the patient's indications; A step of extracting emotional information related to the emotional state of the patient from the patient information; A step of analyzing the emotional state related to the indication based on the extracted emotional information, and generating a prompt requesting the generation of feedback content related to the emotional state; A step of processing the generated prompt as input to a pre-trained feedback generation model to obtain the feedback content related to the emotional state; and An AI-based method for providing patient-customized feedback, characterized by including the step of providing feedback content related to the emotional state to a user terminal.

2. In Paragraph 1, The above emotional information is, Type 1 emotional information related to the general emotions of the above patient and An AI-based method for providing patient-customized feedback, characterized by including at least one of a second type of emotional information related to the patient's indications.

3. In Paragraph 2, In the step of generating the above prompt, Generate the prompt based on at least one of the above emotional information, the result of the emotional state analysis, and the above patient information, and The results of the above emotional state analysis are, The first type emotional state analysis result corresponding to the above first type emotional information and An artificial intelligence-based method for providing patient-customized feedback, characterized by including at least one of the second type emotional state analysis results corresponding to the second type emotional information.

4. In Paragraph 3, The above prompt is, A first type prompt generated based on at least one of the above patient information, the above first type emotion information, and the above first type emotion state analysis result, and An artificial intelligence-based method for providing patient-customized feedback, characterized by including at least one of a second type prompt generated based on at least one of the above patient information, the above second type emotion information, and the above second type emotion state analysis result.

5. In Paragraph 1, In the step of providing the above feedback content to the user terminal, An AI-based patient-customized feedback provision method characterized by detecting the occurrence of a preset feedback event and, in response to detecting the occurrence of the feedback event, providing the feedback content to the user terminal.

6. In Paragraph 5, It further includes a step of monitoring the emotional state associated with the above indication, and In the above monitoring step, An AI-based method for providing patient-customized feedback characterized by monitoring changes in emotion classes corresponding to the emotional states analyzed sequentially over time.

7. In Paragraph 6, In the step of providing the above feedback content to the user terminal, In response to detecting a change in the type of the above emotion class, the above feedback event occurs, and An AI-based patient-customized feedback provision method characterized by determining the time at which the above feedback event occurs as the time of feedback provision, and providing the feedback content to the user terminal at the time of feedback provision.

8. In Paragraph 1, In the step of providing the above feedback content to the user terminal, An artificial intelligence-based patient-customized feedback provision method characterized by providing a user interface to the user terminal for providing at least one of a feedback program and a counseling program corresponding to the emotional state along with the feedback content.

9. In Paragraph 8, The above counseling program is, An AI-based method for providing patient-customized feedback, characterized by providing an answer to a user query entered in relation to at least one of the patient's indications, emotional state, and feedback content, based on a previously trained large language model.

10. In Paragraph 2, In the step of providing the above feedback content to the user terminal, A method for providing AI-based patient-customized feedback characterized by providing at least one of a first emotional state analysis report generated by analyzing the first type of emotional information and a second emotional state analysis report generated by analyzing the second type of emotional information.

11. In an electronic device, Memory for storing instructions; and It includes at least one processor electrically connected to the memory, and When the above instructions are executed by the at least one processor, the at least one processor, Collect patient information related to the patient's indications, and Emotional information related to the emotional state of the patient is extracted from the patient information above, and Based on the extracted emotional information, analyze the emotional state related to the indication and generate a prompt requesting the creation of feedback content related to the emotional state, and The generated prompt is processed as input to a pre-trained feedback generation model to obtain the feedback content related to the emotional state, and An AI-based patient-customized feedback provision system characterized by providing the above-mentioned feedback content related to the above-mentioned emotional state to a user terminal.

12. A program that is executed by one or more processes in an electronic device and stored on a computer-readable recording medium, The above program is, A step of collecting patient information related to the patient's indications; A step of extracting emotional information related to the emotional state of the patient from the patient information; A step of analyzing the emotional state related to the indication based on the extracted emotional information, and generating a prompt requesting the generation of feedback content related to the emotional state; A step of processing the generated prompt as input to a pre-trained feedback generation model to obtain the feedback content related to the emotional state; and A program stored on a computer-readable recording medium characterized by including instructions that perform the step of providing the feedback content related to the above emotional state to a user terminal.

Images