AI-based Customized Healthcare Platform System for Providing Personalized Information for Musculoskeletal Pain Relief and Rehabilitation Assistance

Abstract

The AI-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance according to the present invention is characterized by comprising: a user information collection module that receives symptoms and bio-information related to a user's musculoskeletal system; a prediction module that generates prediction information predicting the user's need for rehabilitation by analyzing the symptoms and bio-information based on an AI algorithm; and an information provision module that provides customized healthcare information, including a rehabilitation exercise program, to the user based on the prediction information. According to the AI-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance of the present invention, the system provides the effect of accurately determining the individual's musculoskeletal condition and rehabilitation needs by comprehensively analyzing the user's symptoms and biometric information using AI, and effectively providing customized healthcare information that reflects this.

Description

Technology Field

[0001] The present invention relates to an artificial intelligence-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance. More specifically, it relates to a healthcare platform system capable of effectively providing customized rehabilitation exercise programs and healthcare information most suitable for individual users by comprehensively analyzing a user's musculoskeletal symptoms and bio-information through an artificial intelligence algorithm. Background Technology

[0002] In general, musculoskeletal disorders are becoming increasingly frequent due to changes in lifestyle and the increase in the elderly population. Chronic and repetitive pain and functional decline caused by these musculoskeletal disorders not only significantly reduce an individual's quality of life but also increase social and economic burdens.

[0003] Accordingly, healthcare apps or systems are recently being proposed to evaluate the condition of the musculoskeletal system and the need for rehabilitation by providing basic diagnostic and rehabilitation guidelines based on the user's various symptom information and simple physical information.

[0004] However, these conventional healthcare apps or systems rely primarily on fragmentary and uniform evaluation methods for the user's symptoms or condition, and are limited to providing healthcare information such as rehabilitation exercise programs in a uniform manner. They have limitations in that it is difficult to provide personalized healthcare information optimized for the individual by comprehensively analyzing the user's biometric and symptom information or by reflecting the user's condition changes and progress in real time.

[0005] Furthermore, conventional technology had a problem in that it was difficult to provide practically effective pain relief and rehabilitation assistance functions to users because there was not enough sophisticated system in place to accurately determine the need for rehabilitation based on changes in the user's individual condition and to provide customized healthcare information in real time based on this.

[0006] Therefore, there is a need to develop a novel and advanced healthcare platform system capable of receiving input regarding a user's musculoskeletal symptoms and biometric information, comprehensively analyzing this based on artificial intelligence algorithms, and providing customized healthcare information optimized according to the user's rehabilitation needs. Prior art literature

[0007] Korean Registered Patent No. 10-2438725 The problem to be solved

[0008] The present invention was devised to overcome the problems of the above technology, and its main purpose is to provide a system that receives user's musculoskeletal symptoms and bio-information, comprehensively analyzes them using an artificial intelligence algorithm to accurately predict the need for rehabilitation, and provides personalized healthcare information optimized for the individual user based on this.

[0009] Another objective of the present invention is to provide a system that converts user symptom information into graph-shaped data based on human anatomical structures, and uses this to perform a linked analysis of symptoms and biological information to clearly identify areas deemed to require rehabilitation and determine priorities.

[0010] Another objective of the present invention is to support effective rehabilitation of myofascial damage by accurately predicting the need for rehabilitation regarding the user's myofascial damage condition and, accordingly, providing information that recommends the use of customized rehabilitation exercise tools, such as myofascial microcurrent therapy patches, to the user. means of solving the problem

[0011] To achieve the above objective, the AI-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance according to the present invention is characterized by comprising: a user information collection module that receives symptoms and bio-information related to a user's musculoskeletal system; a prediction module that generates prediction information predicting the user's need for rehabilitation by analyzing the symptoms and bio-information based on an AI algorithm; and an information provision module that provides customized healthcare information, including a rehabilitation exercise program, to the user based on the prediction information.

[0012] In addition, the prediction module is characterized by comprising: a graph data generation unit that converts the symptoms into graph-shaped data based on human anatomical structures based on a Graph Neural Network (GNN) algorithm; an importance calculation unit that calculates the correlation between adjacent musculoskeletal areas connected to the pain occurrence site in the graph data and selects areas requiring rehabilitation by auxiliaryly referencing the bio-information; and a priority determination unit that generates prediction information determining the areas requiring priority rehabilitation exercises by classifying the need for rehabilitation of the selected areas. Effects of the invention

[0013] According to the artificial intelligence-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance according to the present invention,

[0014] 1) It has the advantage of being able to comprehensively analyze users' symptoms and biometric information using artificial intelligence to accurately determine individual musculoskeletal conditions and rehabilitation needs, and effectively provide customized healthcare information reflecting this.

[0015] 2) By recommending the use of effective and practical rehabilitation exercise tools, such as myofascial microcurrent therapy patches, through accurate prediction and evaluation of the user's myofascial damage, the rehabilitation and pain relief effects of myofascial damage can be maximized, and

[0016] 3) By coating the skin contact surface with a biocompatible-highly conductive composite mixture, the efficiency and stability of microcurrent delivery are significantly increased, and skin irritation is minimized, thereby maintaining excellent comfort and stable rehabilitation effects even with prolonged patch use. Brief explanation of the drawing

[0017] FIG. 1 is a block diagram illustrating the configuration of the system of the present invention. FIG. 2 is a conceptual diagram illustrating the state of receiving user symptoms and biometric information in a user information collection module. Figure 3 is a conceptual diagram illustrating the state of providing rehabilitation exercise videos to a user in an information provision module. Figure 4 is a conceptual diagram illustrating the state in which a rehabilitation exercise tool is recommended to a user by an information provision module. Figure 5 is a conceptual diagram illustrating graph-shaped data based on human anatomical structures based on GNN. Figure 6 is a conceptual diagram illustrating the structure of a myofascial microcurrent therapy patch. FIG. 7 is a conceptual diagram illustrating a state in which a user wears a myofascial microcurrent therapy patch including an IoT sensor. Specific details for implementing the invention

[0018] Preferred embodiments of the present invention will be described in detail below with reference to the attached drawings. The attached drawings are not drawn to scale, and the same reference numerals in each drawing refer to the same components.

[0019] The system of the present invention can be implemented centering on a main server and consists of a complex structure of hardware, software, and platform to realize an AI-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance.

[0020] In terms of hardware, the main server of the present invention includes a processor of a central processing unit (CPU), a graphics processing unit (GPU), or a processor dedicated to artificial intelligence computation (NPU, TPU, etc.) to rapidly and accurately process various input data including the user's musculoskeletal-related symptoms and biometric information, and is equipped with memory (RAM) for high-speed computation processing and a high-speed storage means (SSD, etc.) for data storage. In addition, it includes wired and wireless network interfaces and various input / output interfaces for stably receiving and transmitting biometric information and sensor data obtained from the user. Furthermore, it is equipped with a structure capable of wirelessly linking data in real time with separate smart healthcare equipment, such as an IoT-based myofascial microcurrent therapy patch (200) linked with the platform of the present invention.

[0021] In terms of software, the main server of the present invention can be implemented in a structure in which various software modules are installed, stored, and operated based on an operating system (OS). Specifically, it consists of various software components such as a user information collection module (110) that collects and manages user's musculoskeletal symptoms and bio-information input data, a prediction module (120) that converts input data into a graph form according to anatomical structure based on an artificial intelligence algorithm, particularly a Graph Neural Network (GNN) algorithm, determines correlations between adjacent parts and rehabilitation priorities to generate prediction information regarding the need for rehabilitation, and an information provision module (130) that processes this into customized healthcare information for the user.

[0022] These software modules are not simply implemented as logical software, but may also be implemented in specialized hardware forms such as FPGA (Field Programmable Gate Array) and ASIC (Application-Specific Integrated Circuit), and may be implemented in various forms such as processes, functions, procedures, subroutines, program code segments, drivers, firmware, circuits, databases, data structures, and variables.

[0023] The platform of the present invention is configured around a main server and organically interconnects with various user terminals, such as smartphones, tablet PCs, desktops, and laptops used by users, through wired and wireless networks. It can be configured with a platform structure that efficiently collects symptoms and biometric information from the user through a client application installed on the user terminal, and intuitively and effectively provides various personalized healthcare information, such as user-customized rehabilitation exercise programs, customized exercise videos, recommendations for rehabilitation exercise tools, and recommendations for myofascial microcurrent therapy patches (200).

[0024] Through this structural design, the present invention is implemented to effectively provide a customized management and rehabilitation program according to the user's musculoskeletal condition, thereby maximizing the efficiency of musculoskeletal pain relief and rehabilitation.

[0025] The configuration and operation of the system of the present invention will be described in detail below with reference to the attached drawings.

[0026] FIG. 1 is a block diagram illustrating the configuration of the system of the present invention.

[0027] As can be seen from FIG. 1, the system of the present invention is based on including a user information collection module (110), a prediction module (120), and an information provision module (130).

[0028] Figure 2 is a conceptual diagram illustrating the state of receiving user symptoms and biometric information in a user information collection module.

[0029] The user information collection module (110) of the present invention provides a function to receive symptoms and bio-information related to the user's musculoskeletal system.

[0030] The musculoskeletal system is one of the organ systems that constitute the human body; it refers to the structural system of the human body that includes muscles, bones, joints, tendons, ligaments, and cartilage, which maintain the body's shape and enable movement.

[0031] The musculoskeletal system performs the functions of supporting the body, providing stability, absorbing external shocks, and regulating body movements. It is also an essential structure that plays a role in protecting major internal organs.

[0032] Musculoskeletal symptoms refer to various physical discomforts and abnormal phenomena, such as discomfort, pain, limited movement, swelling, and stiffness, that occur in areas of the musculoskeletal system, including muscles, bones, joints, tendons, ligaments, and cartilage.

[0033] These symptoms can generally occur due to various causes, such as repetitive movements, poor posture, trauma or injury, degenerative changes, or other diseases.

[0034] These musculoskeletal symptoms can be distinguished into various types, such as pain, limited range of motion, joint swelling, muscle stiffness, numbness and paresthesia, muscle weakness, fatigue and fatigue pain, and joint sounds.

[0035] Pain can manifest acutely or chronically and includes joint pain (knees, shoulders, etc.), muscle pain, lower back pain, and neck pain. Limited range of motion is a condition in which joint movement is restricted, movement in specific directions is difficult, or accompanied by pain; this can occur due to shoulder-related conditions such as frozen shoulder and rotator cuff disease, or degenerative arthritis.

[0036] Joint swelling is a phenomenon in which a joint area swells due to inflammation or trauma within the joint, and includes rheumatoid arthritis and osteoarthritis.

[0037] Muscle stiffness is a condition in which movement is uncomfortable or restricted due to reduced muscle flexibility and primarily occurs when maintaining a fixed posture for a long time; it is closely related to myofascial pain syndrome and tension headaches of the neck and shoulders.

[0038] Numbness and paresthesia is a phenomenon in which the limbs become numb or lose sensation due to compression of nerves around muscles or joints, and includes carpal tunnel syndrome and spinal stenosis.

[0039] Muscle weakness is a condition in which specific muscles have difficulty exerting force or a persistent decline in muscle strength, and it is associated with conditions such as muscle atrophy or herniated discs.

[0040] Fatigue and fatigue-related pain refer to symptoms where muscles or joints become exhausted and painful due to repetitive and continuous movement, such as chronic fatigue syndrome and repetitive strain injury (RSI).

[0041] Joint crepitus is a condition in which a clicking or frictional sound occurs during joint movement; initially, there is no pain, but as the condition progresses, it may be accompanied by pain, and it commonly occurs in the knees and jaw joints.

[0042] Since these symptoms can manifest individually or in combination and cause discomfort and functional impairment in daily life, it is important to diagnose, manage, and treat them early.

[0043] Biometric information includes basic demographic information such as the user's age, gender, height, and weight, as well as physical activity-related information such as average daily exercise volume, exercise intensity, primary exercise type and frequency, and the degree of fatigue and pain during exercise.

[0044] In addition, it may include physiological information such as heart rate, heart rate variability, blood pressure, and body temperature, as well as musculoskeletal-related details such as joint range of motion, muscle strength and endurance, pressure distribution in muscles and joints, and pain sensitivity, and biomechanical information such as gait patterns, posture analysis, and muscle activity, and additional information such as sleep patterns, stress levels, and lifestyle habits.

[0045] The reason for collecting such diverse bio-information in this invention is to accurately identify and analyze the user's individual physical condition and musculoskeletal symptoms, thereby enabling the real-time evaluation and management of the effectiveness and suitability of individual rehabilitation exercise programs, and to provide a customized healthcare solution that prevents the risk of musculoskeletal injury and maximizes the efficiency of rehabilitation.

[0046] The user information collection module (110) can input and collect biometric information and musculoskeletal symptoms of the user through various methods.

[0047] For example, a method can be used in which visual data is input by the user directly inputting into an application provided via a mobile device such as a smartphone or tablet PC or a computer, or by using a mobile device's camera to capture the user's body movements, posture, muscle condition, etc., and physiological data such as the user's heart rate, blood pressure, body temperature, and muscle activity can be collected in real time using biosensors (wearable devices).

[0048] In addition, data can be collected by automatically measuring and inputting information on pressure, microcurrent delivery status and muscle condition through a dedicated medical device such as a myofascial microcurrent therapy patch (200) with an embedded IoT sensor (210), and additionally, various input methods can be combined to input natural language by having the user describe their symptoms and condition through voice recognition via a chatbot, or by automatically receiving and inputting the user's health checkup records or clinical data from a medical institution or other health care institution through a cloud-based database, thereby allowing for more accurate and convenient collection of the user's musculoskeletal condition and bio-information.

[0049] The prediction module (120) of the present invention provides a function to generate prediction information that predicts the user's need for rehabilitation by analyzing symptoms and biometric information obtained through the user based on an artificial intelligence algorithm.

[0050] In the present invention, predictive information refers to various forms of data indicating the necessity of rehabilitation or the need for a rehabilitation exercise program and the level thereof, based on the user's current condition, as a result of analyzing the user's musculoskeletal symptoms and bio-information through an artificial intelligence algorithm.

[0051] This predictive information can be provided in a quantitative (numerical) or qualitative (state) form.

[0052] Predictive information in a quantitative form can be provided by expressing the level of rehabilitation necessity according to the user's musculoskeletal condition as a score or graded numerical value ranging from 0 to 100 points. For example, the rehabilitation necessity of a specific joint or muscle area can be expressed in the form of a score such as '85 points' or quantified as 'Level 1 (low), Level 2 (medium), Level 3 (high)' so that the user can easily understand it.

[0053] On the other hand, qualitative predictive information can be provided by expressing the user's need for rehabilitation as specific conditions or recommended measures; for example, it can be clearly conveyed to the user by expressing it concretely and intuitively in forms such as 'immediate rehabilitation exercises recommended,' 'additional diagnosis and testing needed,' 'maintain current condition,' or 'precision rehabilitation program needed.'

[0054] Furthermore, the predictive information of the present invention may also be provided as visual information in a manner that intuitively conveys the parts of the user requiring rehabilitation by displaying or highlighting them in the form of graph-like images or diagrams generated based on human anatomical structures; in other words, the damaged parts of specific joints or muscles or the need for rehabilitation may be visually conveyed to the user through color changes or highlighting.

[0055] The predictive information of the present invention is not limited to only one form of such quantitative, qualitative, or visual information, but is provided in a complex manner by combining them, thereby enabling the user to more easily understand their musculoskeletal condition and rehabilitation needs, and can be generated in a form for effective delivery.

[0056] Artificial intelligence algorithms refer to computer-based data processing methods designed to learn specific patterns based on input data and thereby predict or analyze specific results; they are particularly utilized to efficiently analyze complex and vast amounts of data and derive meaningful predictive information.

[0057] First, an artificial intelligence algorithm suitable for the prediction module (120) of the present invention is a Convolutional Neural Network (CNN). The CNN is an algorithm specialized in analyzing visual data, such as images or videos, and can receive image data capturing the user's musculoskeletal condition, identify damaged areas of muscles or joints, analyze the degree of damage, predict the need for rehabilitation, and intuitively visualize the location and extent of the damaged area within the image and provide it to the user.

[0058] Secondly, there are Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM) algorithms that improve upon them. These algorithms have strengths in analyzing time-series data such as users' physiological data, exercise volume, and fatigue levels. In particular, they can learn the trends of changes in a user's condition over time and predict whether the musculoskeletal condition will worsen or improve and the need for rehabilitation based on this, thereby providing customized management strategies based on changes in the user's condition.

[0059] Finally, the most suitable algorithm for the present invention is the Graph Neural Network (GNN).

[0060] GNN is an algorithm that converts information on a user's musculoskeletal condition and injury into graph-based data according to human anatomical structures to analyze the correlations and influences between adjacent joints or muscles. It can accurately model and analyze the impact of damage or abnormalities in specific areas on surrounding musculoskeletal regions, thereby determining areas requiring priority rehabilitation and generating predictive information to provide more precise and effective customized healthcare solutions for musculoskeletal injury management.

[0061] Among these, the process of generating quantitative predictive information regarding the need for rehabilitation by analyzing the user's symptoms and biometric data through a CNN algorithm is explained as follows.

[0062] First, the user information collection module (110) receives symptoms such as shoulder pain intensity, limited movement, and swelling from the user via a smartphone app or web application and transmits the symptom information along with an image of the shoulder joint or muscle area to the prediction module (120). Then, the prediction module (120) analyzes the image using an image analysis model trained with a CNN algorithm to recognize visually apparent abnormal signs, such as changes in the shape of the area, changes in skin color, or swelling.

[0063] At this time, biometric information such as the user's heart rate, blood pressure, and muscle activity can be collected through wearable sensors or IoT-based biometric signal measuring devices, and through this, the CNN model can determine the user's current musculoskeletal condition and the severity of the injury by comprehensively considering the input symptom information, image information, and biometric information.

[0064] The CNN model evaluates the severity of symptoms, such as pain or movement limitations, by comparing the input data with previously learned past cases. It then performs a similarity analysis based on the combination of symptoms and biometric information to determine how similar the user's injury status is to cases requiring rehabilitation, and based on this, generates predictive information that expresses the user's need for rehabilitation as a quantitative score.

[0065] For example, symptom information such as "pain intensity of 8 points (out of 10)" and "high degree of limitation of movement" entered by the user, and "severe swelling of the shoulder area" analyzed in the image, showed very high similarity to cases with high severity as a result of CNN analysis, and biosignal data was also analyzed as exceeding the abnormal range, so the user's need for rehabilitation is generated in the form of quantitative numerical prediction information such as "rehabilitation need score of 95" and provided to the user, thereby allowing the user to clearly and objectively understand their condition.

[0066] Figure 3 is a conceptual diagram illustrating the state of providing rehabilitation exercise videos to a user in an information provision module.

[0067] The information provision module (130) performs the function of providing customized healthcare information, including a rehabilitation exercise program, to the user based on predictive information.

[0068] Personalized healthcare information refers to information that includes various health management details, such as rehabilitation exercise programs, provided to each user based on predictive data analyzed by artificial intelligence algorithms based on the user's musculoskeletal symptoms and biometric data.

[0069] Furthermore, a rehabilitation exercise program refers to a customized exercise management program that systematically organizes and provides the types, intensity, duration, and frequency of exercises optimized for the individual to improve the user's musculoskeletal condition and alleviate pain, based on predictive information analyzing the degree of damage, pain, and rehabilitation needs of specific muscles or joints.

[0070] For example, programs designed to improve the range of motion of the shoulder joint and alleviate pain include shoulder stretching, progressive strength training using resistance bands, and rotator cuff stabilization exercises.

[0071] In addition, as a program to improve pain and function in the lower back and lumbar region, customized exercises such as bridge exercises performed while lying down, abdominal stabilization exercises, lumbar stabilization exercises, and core muscle strengthening exercises can be provided.

[0072] In addition, programs for relieving knee and lower extremity pain and improving function can be provided by subdividing exercises such as squats, stair climbing, leg extensions, hamstring strengthening, and knee stabilization using resistance bands, according to the individual's pain intensity and muscle strength.

[0073] Specifically, when the information providing module (130) receives a quantitative value (e.g., 90 points) from the prediction information received from the prediction module (120) indicating that the user's shoulder joint injury is severe and there is a high need for rehabilitation, the information providing module (130) selects a customized rehabilitation exercise program that requires a high level of rehabilitation from a pre-configured program database based on the value, and further incorporates the user's symptom information (pain, movement limitation, etc.) and bio-information (heart rate, muscle activity, etc.) to finally adjust and select the program most suitable for the user's condition.

[0074] Subsequently, it is possible to provide the selected program through the user's smartphone app or web-based platform, guide exercise methods and durations clearly and easily through videos, images, and text, and continuously provide an optimal customized rehabilitation exercise program by frequently checking status feedback after exercise to periodically evaluate and update the program's suitability.

[0075] Figure 4 is a conceptual diagram illustrating the state in which a rehabilitation exercise tool is recommended to a user by an information provision module.

[0076] Furthermore, personalized healthcare information can additionally include not only the aforementioned rehabilitation exercise programs but also recommendations for rehabilitation exercise tools, information on in-home rehabilitation services, hospital information, and health functional foods.

[0077] For example, if the user’s shoulder joint injury status is evaluated as ‘rehabilitation need 90 points’ in the prediction module (120), the information provision module (130) recommends customized rehabilitation tools such as elastic bands for shoulder muscle strengthening and stretching, massage balls, and shoulder-specific pulley exercise equipment as ‘rehabilitation exercise tool recommendation information’ as shown in FIG. 4, and provides the usage and effects of each tool along with visual materials to guide the user to use them easily.

[0078] In addition, in the case of ‘visiting rehabilitation exercise service information,’ when the need for rehabilitation is assessed as very high in the prediction information, the information provision module (130) can recommend visiting exercise service information of a professional rehabilitation exerciser or physical therapist in the area where the user is located based on the user’s condition and location information, and provide the profile of the visiting expert, service availability hours, and method of use so that the user can use the visiting service quickly and conveniently.

[0079] Next, regarding the provision of 'hospital information,' if the user's injury condition is assessed as severe and it is determined that precise diagnosis or additional treatment is required, the information provision module (130) can recommend a specialized hospital or clinic in the area where the user is located that specializes in musculoskeletal rehabilitation treatment and provide specific information such as the hospital's medical departments, doctor information, available appointment times, and contact information to encourage the user to visit the hospital in a timely manner.

[0080] When receiving prediction information indicating that the user's fatigue or muscle recovery speed is slow, and thus the effectiveness of rehabilitation exercises may be reduced, the information provision module (130) provides additional information, specifically recommending health functional foods containing amino acid supplements, magnesium supplements, and antioxidant components that are effective for muscle recovery and pain relief, and guides the user on the efficacy, dosage, and purchase methods of each product to help the user achieve faster musculoskeletal recovery and pain relief.

[0081] These additional details are individually customized to reflect the user's specific predictive information, personal condition, and needs, thereby enabling the user to benefit from more accurate and effective rehabilitation and health management.

[0082] In summary, the system of the present invention provides the characteristic of effectively providing customized rehabilitation exercise programs and healthcare information most suitable for individual users by comprehensively analyzing the user's musculoskeletal symptoms and bio-information through an artificial intelligence algorithm.

[0083] Figure 5 is a conceptual diagram illustrating graph-shaped data based on human anatomical structures based on GNN.

[0084] Among the various artificial intelligence algorithms exemplified above, the Graph Neural Network (GNN) is considered the most desirable in this invention because, due to the characteristics of the human musculoskeletal system, individual parts such as joints and muscles are organically connected to each other, and damage or abnormality in a specific part directly affects other adjacent parts, it can more accurately predict the need for rehabilitation by clearly reflecting the relationships between each part. The prediction module (120) that further elaborates on this is described as follows.

[0085] Specifically, the prediction module (120) based on a Graph Neural Network (GNN) algorithm can be subdivided into a graph data generation unit (121), an importance calculation unit (122), and a priority determination unit (123).

[0086] The graph data generation unit (121) converts symptom information received from the user into graph-shaped data consisting of nodes and edges based on the musculoskeletal anatomical structure of the human body, such as muscles and joints, as shown in FIG. 5.

[0087] Specifically, the graph data generation unit (121) generates graph-shaped data by setting nodes based on the muscle and joint structures of the human body based on symptom information for each body part, such as the shoulder, neck, waist, and knee, entered by the user, and setting the mutual relationships between adjacent or anatomically directly connected parts as edges.

[0088] At this time, quantitative or qualitative symptom information, such as pain intensity, degree of movement limitation, and presence of edema, is assigned as attributes to each node, and information such as anatomical distance or connection strength is set as weights to the edges so that symptom information can be represented more accurately by reflecting the structural characteristics of the human musculoskeletal system.

[0089] The importance calculation unit (122) calculates the correlation between adjacent musculoskeletal areas connected to the pain occurrence area in the graph data and provides a function to select the area requiring rehabilitation by auxiliaryly referring to the bio-information.

[0090] First, the importance calculation unit (122) calculates the correlation between a specific area where pain has occurred and a musculoskeletal area that is anatomically directly connected or adjacent to it using graph data generated by the graph data generation unit (121).

[0091] In this context, correlation calculation refers to the process of quantitatively calculating and quantifying the potential impact that the severity of symptoms in a specific area may have on other adjacent areas. For example, if severe pain and limited movement occur in the shoulder joint, the impact on adjacent areas directly connected to the shoulder, such as the neck or elbow, is calculated as a correlation value (a value between 0 and 1) using a graph neural network algorithm.

[0092] Subsequently, the importance calculation unit (122) performs a more accurate evaluation by auxiliaryly referencing the user's bio-information based on the value obtained through such correlation calculation. For example, in a situation where the correlation value between the shoulder joint and the elbow joint is calculated to be high, if the user's physiological data such as muscle activity, heart rate, or fatigue level deviates significantly from the normal range, the correlation value can be adjusted or weights assigned by taking this into account to evaluate the need for more accurate and realistic rehabilitation.

[0093] Finally, the importance calculation unit (122) selects the area with the highest need for rehabilitation by synthesizing the correlation values ​​calculated in this way and the auxiliary evaluation results based on bio-information. For example, if the user’s shoulder area has very severe symptoms of pain and limited movement with a score of '9 points (out of 10)', and the correlation values ​​for the elbow and neck areas adjacent to the shoulder are calculated to be very high with scores of '0.8' and '0.7', respectively, and additionally, the muscle activity and fatigue measured in the elbow area are found to be abnormally high, then the elbow area is given priority importance, and finally, not only the shoulder but also the elbow area is selected as an area requiring priority rehabilitation.

[0094] In this way, the importance calculation unit (122) can determine the areas requiring rehabilitation more accurately and effectively through structural analysis of the graph neural network algorithm and auxiliary reference of bio-information.

[0095] The priority determination unit (123) provides a function to generate predictive information that determines the area requiring priority rehabilitation exercise by classifying the need for rehabilitation of the selected area.

[0096] To explain the function of the priority determination unit (123) in detail, the need for rehabilitation of the selected area in the priority calculation unit (122) is graded according to certain criteria. This grading process involves comprehensively evaluating the symptom severity, correlation values, and bio-information of each area to divide them into intuitive stages such as 'high', 'medium', and 'low'.

[0097] For example, if the importance score of the shoulder joint is very high at '0.9' or higher, and the elbow joint and neck area have importance scores of '0.7' and '0.6', respectively, the shoulder is graded as 'High', the elbow as 'Medium', and the neck as 'Low', providing a way for users to more easily understand the rehabilitation needs of each area.

[0098] Next, the priority determination unit (123) determines the area requiring priority rehabilitation exercise based on the graded information and generates prediction information. For example, the shoulder area, which is rated as 'high' in the graded results, is determined as the area requiring the most priority rehabilitation management, and then the elbow and neck areas are assigned priorities in that order, thereby generating prediction information with clear priorities for the user in the form of "1st priority: shoulder (high need for rehabilitation)", "2nd priority: elbow (medium)", "3rd priority: neck (low)".

[0099] Through this predictive information, the user can intuitively and clearly recognize their musculoskeletal condition and the areas requiring the most urgent rehabilitation. In addition, the information provision module (130) can provide customized rehabilitation exercise programs and additional health management information more effectively and systematically based on this priority information, thereby maximizing the user's musculoskeletal rehabilitation and pain management efficiency.

[0100] The prediction module (120) described above can generate prediction information that predicts the need for rehabilitation for various parts of the user. In particular, since the 'fascia' has the characteristic that when damaged, pain spreads not only to the fascia area but also to surrounding muscles and joints connected to the fascia, causing symptoms to appear more extensively, it is necessary to accurately predict the need for rehabilitation for the user's fascia damage by analyzing the degree of pain spread, muscle tension state, degree of movement limitation, and muscle activity among the user's symptoms and bio-information more precisely through an artificial intelligence algorithm.

[0101] That is, the degree of pain diffusion, muscle tension state, and muscle activity in the user's symptoms and bio-information have attributes that reflect the widespread symptom propagation and high recurrence potential characteristic of fascial injury, and the prediction module (120) can generate prediction information that more accurately evaluates the need for rehabilitation for fascial injury by comprehensively analyzing these attributes.

[0102] Specifically, when the user information collection module (110) receives symptom information such as the intensity of pain and muscle stiffness and the degree of pain spread in the neck or shoulder area through a smartphone app, and additionally collects biometric information such as muscle activity (EMG signal), muscle tension, and skin conductivity in real time from a wearable sensor or an IoT-based biometric measurement device, the prediction module (120) comprehensively analyzes these input data using an artificial intelligence algorithm.

[0103] At this time, the prediction module (120) analyzes the user's input data by comparing it with previously learned fascia injury case data, focusing on physiological characteristics such as pain spreading widely to surrounding areas as well as the fascia area, abnormal increase in muscle activity during normal movement, and intensification of muscle tension, particularly as characteristic symptoms appearing due to fascia injury.

[0104] For example, if the pain intensity in the user's shoulder and neck areas is very high, such as '8 points (out of 10)' and '7 points' respectively, and the muscle activity and tension in those areas are also higher than the reference value, and in particular, a distinct pain diffusion pattern from the shoulder to the neck is detected, the prediction module (120) can analyze that these characteristics show a very high similarity to a case of fascial injury and ultimately generate prediction information in a quantitative form such as "Need for fascial injury rehabilitation: 92 points".

[0105] The predicted information on the need for rehabilitation for fascial damage generated in this way can be used as a basis for recommending to the user, in particular, the use of a fascial microcurrent therapy patch (200) among rehabilitation exercise tools in the information provision module (130).

[0106] The information provision module (130) provides a function of providing information recommending the use of a myofascial microcurrent therapy patch (200) for rehabilitation of myofascial damage to the user among the rehabilitation exercise tools described above.

[0107] Figure 6 is a conceptual diagram illustrating the structure of a myofascial microcurrent therapy patch.

[0108] The myofascial microcurrent therapy patch (200) is a rehabilitation therapy tool more specialized for relieving pain and tension caused by myofascial damage. By directly delivering fine electrical stimulation (microcurrent) to the myofascial layer, it increases blood flow to the damaged myofascial tissue, relieves muscle tension, and reduces inflammatory response. This can promote faster and more effective pain relief and tissue recovery than general exercise therapy during the rehabilitation and recovery process of myofascial damage.

[0109] Referring to FIG. 6, it can be seen that the fascia microcurrent treatment patch (200) is manufactured in a thin and flexible form so as to adhere to the user's skin and deliver a microcurrent to the fascia layer, and specifically, it may be composed of an electrical signal generation control unit and a skin contact surface.

[0110] The electrical signal generation control unit generates microcurrent signals of specific frequency, amplitude, and waveform (sine wave, square wave, triangular wave, etc.) using a pulse generator and control circuit based on power supplied from a built-in battery, and provides the function of amplifying and stabilizing these signals and then directly transmitting them to the user's skin and fascia layer through the skin contact electrode.

[0111] The skin contact surface is made of a flexible and soft material, such as silicone or medical-grade polyurethane, so as to effectively transmit microcurrents delivered from the electrical signal generation control unit to the fascia layer, and electrodes for transmitting electrical signals (microcurrents) from the electrical signal generation control unit are placed on the surface of the skin contact surface.

[0112] In particular, as will be described later, the skin contact surface can be coated with the biocompatible-highly conductive composite mixture of the present invention to maximize the efficiency of microcurrent transmission and minimize skin irritation.

[0113] Since this myofascial microcurrent therapy patch (200) is the same as a known myofascial microcurrent therapy patch, it can be sufficiently explained by the configuration and function described above, so a separate additional explanation is omitted.

[0114] The information provision module (130) has significance in that it provides information recommending a myofascial microcurrent therapy patch (200) as an effective rehabilitation method for such a condition when, as a result of comprehensively analyzing the user's symptom information (spread of pain, increased tension) and bio-information (abnormal increase in muscle activity, tension state, etc.) from the prediction module (120) above, the user's condition shows a higher similarity to myofascial damage and the severity of the damage is evaluated to be high.

[0115] For example, the information providing module (130) can provide information to the user recommending a myofascial microcurrent therapy patch (200) as a rehabilitation method for this condition when the intensity of pain in the user's shoulder and neck areas is high at 8 points (out of 10) and 7 points respectively, and the pain spreads widely to surrounding areas, and at the same time bio-information such as muscle activity and tension state is abnormally increased, so that the possibility of myofascial damage is very high.

[0116] Furthermore, the information recommending this myofascial microcurrent therapy patch (200) may include usage time and a standard value for the minimum microcurrent intensity required to obtain a therapeutic effect during the usage time, so that the user can use the myofascial microcurrent therapy patch (200) more effectively and stably.

[0117] That is, the recommendation information for using the myofascial microcurrent therapy patch provided by the information provision module (130) can not only accelerate the recovery speed of the user's myofascial damage, but also effectively manage the widespread pain diffusion and the possibility of repeated recurrence characteristic of myofascial damage, thereby playing a role in maintaining the user's musculoskeletal health more long-term and stably.

[0118] Figure 7 is a conceptual diagram illustrating a user wearing a myofascial microcurrent therapy patch including an IoT sensor.

[0119] As can be seen from FIG. 7, the above-described fascial microcurrent therapy patch (200) may additionally include an IoT sensor (210) to more accurately measure the current intensity of the microcurrent actually delivered at the skin contact surface in real time to more accurately assess and update the need for rehabilitation for the user's fascial damage area, as well as to more effectively monitor the quality of microcurrent delivery and the skin contact condition.

[0120] In other words, since the actual current intensity measured at the actual skin contact surface of a general myofascial microcurrent therapy patch (200) frequently differs significantly from the setting value entered by the user when the skin contact state of the patch is unstable or when the user's skin condition and attachment state change, it is technically and practically more appropriate to analyze and evaluate based on the measurement value measured at the actual skin contact surface rather than the user's setting value in order to more accurately evaluate and manage the therapeutic effect of microcurrent and the need for rehabilitation. Therefore, an IoT sensor (210) is additionally included to implement this.

[0121] The IoT sensor (210) provides the function of generating detection information regarding the state of transmission of microcurrent transmitted from the microcurrent generation and control unit to the skin contact surface described above, and transmitting it to the user information collection module (110) through a communication interface built into itself.

[0122] In this case, the sensing information refers to electrical characteristic information that quantitatively represents the current intensity (μA) of the microcurrent actually transmitted through the skin contact surface.

[0123] Accurate measurement information regarding the current intensity of such microcurrents can be utilized as an indicator to accurately evaluate the actual delivery quality and effectiveness of microcurrent therapy in real time by determining in real time whether the microcurrent is being delivered normally according to the user's fascial damage status. Furthermore, it can be used to accurately assess the actual delivery quality of microcurrents in real time under various real-world usage environments, such as the user's skin contact status, attachment status, and changes in skin resistance, regardless of the microcurrent settings configured by the user through the patch.

[0124] For example, even if the user inputs a setting value of 200μA for the microcurrent treatment patch (200), if the actual microcurrent intensity measured at the skin contact surface through the IoT sensor (210) is continuously measured as 150μA, the IoT sensor (210) can transmit the detection information, which is the measured current intensity, as well as the setting value if necessary, to the user information collection module (110).

[0125] Specifically, the IoT sensor (210) includes a precision current sensor (current sensing sensor) for accurately measuring the current intensity of a microcurrent transmitted from the skin contact surface, and in particular, a current sensing method using a high-precision shunt resistor can be used to accurately measure the current intensity of the microcurrent. That is, the IoT sensor (210) has a structure that measures the minute voltage drop occurring across the shunt resistor when the microcurrent is transmitted through the skin contact surface and, based on this, precisely detects the accurate current intensity of the microcurrent in real time.

[0126] Additionally, detection information measured by the IoT sensor (210) is converted into digital data through a miniature analog-to-digital converter (ADC) placed inside the IoT sensor, and the converted data is transmitted in real time to the user information collection module (110) through a built-in wireless communication interface (low-power wireless protocol such as Bluetooth Low Energy, Wi-Fi, or Zigbee).

[0127] In addition, the setting value set by the user in the myofascial microcurrent treatment patch (200) is stored in the electrical signal generation control unit and can be transmitted to the user information collection module (110) through the wireless communication interface of the IoT sensor (210).

[0128] The user information collection module (110) receives the detection information through the wireless communication interface of the IoT sensor (210) and transmits it to the prediction module (120).

[0129] At this time, in order to further evaluate the current intensity performance of the myofascial microcurrent treatment patch (200) in the prediction module (120), the setting value set by the user in the myofascial microcurrent treatment patch (200) can also be received through the wireless communication interface of the IoT sensor (210) in the electrical signal generation control unit and transmitted to the prediction module (120) so that the detection information and the setting value can be compared.

[0130] In response to this, the prediction module (120) can provide a function to generate prediction information that updates the need for rehabilitation for the fascia damage area in real time by analyzing the detection information transmitted from the user information collection module (110).

[0131] Specifically, the prediction module (120) comprehensively analyzes the actual microcurrent intensity (sensing information) input from the IoT sensor (210) together with the user's symptom information (intensity of pain, degree of spread, degree of movement restriction, etc.) and bio-information (muscle activity, tension, etc.).

[0132] For example, if the microcurrent intensity significantly decreases compared to the existing level or fluctuates rapidly for an extended period, this may mean that the rehabilitation exercise effect on the user's fascia injury site is not fully exerted, potentially delaying pain relief or failing to improve muscle tension.

[0133] Accordingly, the prediction module (120) can comprehensively analyze the actual deterioration of microcurrent delivery quality and the user's symptom information and bio-information to re-evaluate the need for rehabilitation for the user's fascia injury site to a higher level than before (e.g., increasing the score quantifying the need for rehabilitation from 70 points to 90 points), and immediately reflect this to generate real-time updated prediction information (120) that enables more precise customized rehabilitation management, including additional management measures such as the intensity or frequency of the rehabilitation exercise program being performed by the user, the usage time of the fascia microcurrent treatment patch (200), and the status check of use.

[0134] Additionally, if the user information collection module (110) additionally receives the microcurrent setting value entered by the user through the patch along with the detection information of the IoT sensor (210), the prediction module (120) can compare and analyze the setting value with the actual measured detection information to perform an additional evaluation of the microcurrent delivery quality, such as the usage status or attachment status of the myofascial microcurrent treatment patch (200).

[0135] For example, in a situation where the set value is 200μA but the actual measured detection information consistently appears to be 150μA, which is significantly lower than the set value, the prediction module (120) can determine that the attachment status of the fascia microcurrent treatment patch (200) is poor or that the microcurrent delivery quality has deteriorated due to changes in skin condition, and can generate and provide additional information to guide the user to additional management measures, such as checking the attachment status of the patch or reattaching it, along with updating the user's rehabilitation needs.

[0136] In this way, the prediction module (120) basically provides a function to analyze the microcurrent delivery status measurement (sensing information) and update the need for rehabilitation in real time, and when additional setting values ​​are input, it provides a characteristic that can generate and provide more precise and proactive additional management information through an evaluation of the microcurrent delivery quality and patch usage status.

[0137] As previously explained, the above-described fascial microcurrent treatment patch (200) often shows a difference between the microcurrent setting value set by the user and the measurement value (sensing information) actually measured at the skin contact surface. This is because the microcurrent set is not transmitted directly to the skin and fascial layer but is lost as the contact resistance between the skin contact surface and the skin changes depending on the user's skin condition or changes in the patch's attachment state to the skin.

[0138] In particular, if foreign substances, sweat, or sebum are present between the skin contact surface and the skin, it may cause contact resistance to increase further or become irregular, making it difficult to achieve the targeted improvement in microcurrent transmission efficiency and effective rehabilitation and pain relief for the damaged fascia area.

[0139] To solve this, the present invention proposes a technique in which the skin contact surface of a myofascial microcurrent treatment patch (200) is fabricated from thermoplastic polyurethane (TPU), and then a biocompatible-highly conductive composite mixture is mixed in a specific ratio and coated onto the surface to significantly increase the microcurrent transmission efficiency while minimizing skin irritation and contact resistance, thereby providing excellent electrical conductivity.

[0140] Thermoplastic polyurethane (TPU), which is the material of the skin contact surface of the fascia microcurrent therapy patch (200), is a type of thermoplastic elastomer having a block copolymer structure in which diisocyanate, high molecular weight polyol, low molecular weight chain extender, etc. are polymerized stepwise.

[0141] In particular, the present invention utilizes the excellent elasticity, high surface uniformity, and stable adhesive properties of TPU to more uniformly and stably coat the biocompatible-highly conductive composite mixture described below onto the skin contact surface, thereby maximizing the transmission efficiency of microcurrents and exhibiting characteristics that minimize skin irritation and maintain high comfort and rehabilitation effects even when used in close contact with the skin for a long time.

[0142] In addition, the excellent electrical insulation properties of TPU ensure that microcurrents can be accurately delivered to the desired fascial damage site without loss at the skin contact surface, thereby providing a basis for further improving the quality of microcurrent delivery and the rehabilitation effect of the fascial microcurrent treatment patch (200) of the present invention.

[0143] The surface of the skin contact surface, made of TPU in this manner, is coated with a biocompatible-highly conductive composite mixture to enhance microcurrent transmission efficiency and alleviate skin irritation.

[0144] Specifically, the biocompatible-highly conductive composite mixture may comprise, based on 100 parts by weight thereof, 15 to 25 parts by weight of 1-ethyl-3-methylimidazolium tetrafluoroborate (EMIM-BF4), 20 to 30 parts by weight of poly(3,4-ethylene dioxythiophene)-polystyrene sulfonate (PEDOT:PSS), 15 to 20 parts by weight of molybdenum disulfide (MoS2), 15 to 20 parts by weight of hexagonal boron nitride (h-BN), and 10 to 20 parts by weight of chitosan oligosaccharide.

[0145] 1-Ethyl-3-methylimidazolium tetrafluoroborate (EMIM-BF4) is an ionic liquid with the chemical formula C6H11BF4N2, which is in a liquid state at room temperature and is a compound with high ionic conductivity, low vapor pressure, and excellent thermal and electrochemical stability.

[0146] EMIM-BF4 is generally used as an electrolyte additive for lithium-ion batteries or as an electrolyte material to ensure high ion conductivity and electrochemical stability in the fields of high-performance supercapacitors and electrochemical sensors.

[0147] In the biocompatible-highly conductive composite mixture of the present invention, EMIM-BF4 provides the function of minimizing contact resistance between the skin contact surface and the skin, and maximizing ion mobility when delivering microcurrents to the skin and fascia layers, thereby increasing the stability and accuracy of microcurrent delivery.

[0148] In addition, it can be equipped with the function of minimizing skin irritation by maintaining a stable electrical environment even during prolonged skin contact, and effectively enhancing the rehabilitation effect of the fascia-damaged area through the uniform delivery of microcurrents.

[0149] In addition, this material exhibits a synergistic effect that increases the mobility and uniformity of ions and electrons by combining with other materials in the biocompatible-highly conductive composite mixture, such as PEDOT:PSS and MoS₂, and also demonstrates a synergistic function that further enhances microcurrent transmission efficiency by simultaneously improving wettability and skin adhesion in a skin contact environment when mixed with chitosan oligosaccharides.

[0150] If EMIM-BF4 is less than 15 parts by weight in the biocompatible-highly conductive composite mixture, there is a problem where the ion transferability decreases and the microcurrent transfer efficiency is reduced, and if it exceeds 25 parts by weight, the viscosity on the skin surface increases, which may reduce skin adhesion and the mechanical properties of the entire mixture, so it is preferable to set it to 15 to 25 parts by weight.

[0151] Poly(3,4-ethylenedioxythiophene)-polystyrene sulfonate (PEDOT:PSS) is a polythiophene-based conductive polymer with the chemical formula (C_8H_8O_3S)_n, and is a representative conductive polymer compound that exhibits high electrical conductivity and thermal and electrochemical stability through polymer chains with continuous π-bond structures.

[0152] Generally, PEDOT:PSS is used as a functional material to provide excellent flexibility and high electrical conductivity, such as in flexible displays, transparent electrodes, and electrode materials for organic solar cells.

[0153] In biocompatible-highly conductive composite mixtures, PEDOT:PSS provides excellent electron transfer capability and uniform current flow during microcurrent delivery, thereby enhancing the quality and stability of microcurrent delivery to the skin and fascia layers.

[0154] In addition, based on biocompatibility, it can significantly reduce skin irritation and maintain high adhesion even when attached to the skin for a long time, thereby providing a stable long-term delivery effect of microcurrents.

[0155] When mixed with MoS2, another material in the biocompatible-highly conductive composite mixture, this material forms a more stable and uniform conductive network, exhibiting a synergistic effect that improves the electrical conductivity and microcurrent transfer efficiency of the entire mixture.

[0156] In a biocompatible-highly conductive composite mixture, if the amount of PEDOT:PSS is less than 20 parts by weight, there is a problem where the electrical conductivity of the mixture is reduced and the stability of current transmission decreases, and if it exceeds 30 parts by weight, the mechanical flexibility and skin adhesion of the mixture decrease, which may cause discomfort when in contact with the skin for a long time, so it is advantageous to set it to 20 to 30 parts by weight.

[0157] Molybdenum disulfide (MoS2) is a two-dimensional layered material of the Transition Metal Dichalcogenides (TMDs) series, having the chemical formula MoS2, in which molybdenum atoms and sulfur atoms are bonded in a sandwich structure. It is a compound that possesses excellent mechanical strength, lubricity, and thermal stability, along with excellent electrical properties similar to graphene.

[0158] Molybdenum disulfide is used industrially to provide high wear resistance in fields such as electrochemical sensors for ultra-high-speed electronic devices.

[0159] In biocompatible-highly conductive composite mixtures, molybdenum disulfide provides excellent electrical conductivity and mechanical durability simultaneously, offering the ability to minimize changes in contact resistance that may occur due to friction or skin movement during microcurrent delivery at the skin contact surface.

[0160] In addition, this can also have the function of assisting in the stable and uniform transmission of microcurrents.

[0161] Furthermore, this material provides a synergistic effect that significantly increases durability and long-term stability in skin-attachment environments through excellent friction-reducing properties, particularly when combined with hexagonal boron nitride (h-BN), another material in the biocompatible-high conductivity composite mixture.

[0162] If the amount of molybdenum disulfide in the biocompatible-highly conductive composite mixture is less than 15 parts by weight, the mechanical stability and electrical uniformity of the mixture decrease, which leads to a problem of reduced microcurrent transmission stability, and if it exceeds 20 parts by weight, the flexibility and skin adhesion of the mixture decrease, which may cause discomfort when attached to the skin, so it is preferable to set it to 15 to 20 parts by weight.

[0163] Hexagonal boron nitride (h-BN) is an inorganic compound with the chemical formula BN and a two-dimensional layered structure in which nitrogen (N) and boron (B) are repeatedly bonded in a hexagonal structure, and it is a compound that has excellent thermal conductivity, chemical and thermal stability, high insulation properties, and a low coefficient of friction.

[0164] Hexagonal boron nitride is primarily used as a high-performance heat dissipation material and a lubricity-enhancing additive.

[0165] In biocompatible-highly conductive composite mixtures, hexagonal boron nitride provides high mechanical durability and low friction characteristics, minimizing changes in contact resistance due to friction or skin movement between the skin contact surface and the skin.

[0166] In addition, it can combine the function of maintaining skin adhesion and the stability of microcurrent delivery even during long-term use.

[0167] If the hexagonal boron nitride in the biocompatible-high conductivity composite mixture is less than 15 parts by weight, the durability and friction reduction characteristics of the mixture are insufficient, which leads to a problem of reduced mechanical stability of the mixture in a skin contact environment, and if it exceeds 20 parts by weight, it can reduce the overall flexibility and uniformity of electrical conductivity of the mixture, so it is effective to set it to 15 to 20 parts by weight.

[0168] Chitosan oligosaccharide is an oligosaccharide produced by enzymatically or chemically reducing the molecular weight of chitosan, which is obtained by deacetylating the natural polysaccharide chitin. It is represented by the chemical formula (C6H11NO4)n and is a low-molecular-weight chitosan derivative that simultaneously possesses excellent biocompatibility, antibacterial properties, moisturizing ability, and skin adhesion.

[0169] Chitosan oligosaccharides are generally used in the cosmetics industry as ingredients for skin moisturizing and skin irritation relief, and in the pharmaceutical and medical fields as antibacterial dressings.

[0170] In the biocompatible-high conductivity complex mixture, chitosan oligosaccharides provide excellent moisturizing and superior antibacterial properties in a skin contact environment, thereby minimizing skin irritation. At the same time, by preventing moisture loss from the skin, they offer the function of preventing skin dryness or sensitivity when the patch is applied for an extended period.

[0171] Furthermore, it enhances adhesion and tackiness upon skin contact to maintain a uniform distribution of the mixture on the skin surface and minimizes changes in contact resistance caused by skin movement, sweat, or sebum, thereby possessing the function of maintaining the stability of microcurrent transmission for an extended period. Moreover, along with its biocompatible properties, it can provide the ability to effectively alleviate inflammation and skin irritation that may occur through anti-inflammatory action.

[0172] This provides a synergistic effect that further enhances skin adhesion and wettability in a skin contact environment when combined with other materials of the biocompatible-highly conductive composite mixture, such as EMIM-BF₄ and PEDOT:PSS, thereby improving the microcurrent delivery environment to be more uniform and stable.

[0173] If the amount of chitosan oligosaccharide in the biocompatible-highly conductive composite mixture is less than 10 parts by weight, there is a problem that the skin moisturizing power, antibacterial power, and adhesion are insufficient, resulting in insufficient skin irritation relief function and skin contact environment maintenance function. If the amount exceeds 20 parts by weight, the electrical conductivity and flexibility of the mixture are relatively reduced, which may lower the quality of microcurrent transmission and skin adhesion of the mixture. Therefore, it is preferable to set the amount to 10 to 20 parts by weight.

[0174] In summary, the biocompatible-highly conductive composite mixture of the present invention provides high microcurrent transmission efficiency and transmission stability based on excellent biocompatibility and skin adhesion, and possesses characteristics that enable continuous effective rehabilitation and pain relief of the fascia injury site by minimizing skin irritation and changes in contact resistance even during long-term skin attachment.

[0175] The following is a more specific demonstration of the performance regarding microcurrent transmission efficiency, skin irritation, etc., according to the composition and function described above, through experiments on the biocompatible-highly conductive composite mixture of the present invention.

[0176] <Comparative Example 1>

[0177] The skin contact surface of the myofascial microcurrent therapy patch was fabricated using ordinary silicone rubber (Silicone Rubber, Sylgard® 184 Silicone Elastomer from Dow Corning) without a separate coating layer. Specifically, a silicone base and a curing agent were weighed at a weight ratio of 10:1, which is the manufacturer's recommended mixing ratio, and mixed homogeneously. After removing air bubbles from the mixed material by vacuum degassing, it was injected into a mold and cured for 1 hour under heat curing conditions at 80°C to form the skin contact surface of the myofascial microcurrent therapy patch of the comparative example.

[0178] <Comparative Example 2>

[0179] The skin contact surface of the myofascial microcurrent therapy patch was fabricated using thermoplastic polyurethane (BASF’s Elastollan® TPU 1185A). Specifically, the skin contact surface of the myofascial microcurrent therapy patch of the comparative example was formed by injection molding the TPU resin under conditions of a cylinder temperature of 200°C and a mold temperature of 60°C in an injection molding machine.

[0180] <Comparative Example 3>

[0181] Graphite (Sigma-Aldrich, average particle size about 5-10 μm) and carbon black (Vulcan XC72) were uniformly mixed in a weight ratio of 50:50 as a conductive coating material on the skin contact surface of the fascia microcurrent therapy patch prepared in Comparative Example 2.

[0182] Polyvinylidene fluoride (PVDF, Solvay's Solef® 5130) was added to this mixture as a binder at 5 wt% relative to the total mixture, and acetone (Sigma-Aldrich) was used as a solvent to adjust the solid content of the mixture to 10 wt%. This mixture was uniformly coated to a thickness of 150 μm on the skin contact surface using a doctor blade, then dried in an 80°C drying oven for 1 hour, and further dried completely in a 60°C vacuum dryer for 12 hours to form a conductive coating layer on the skin contact surface of the myofascial microcurrent therapy patch of the comparative example.

[0183] <Example>

[0184] 20 parts by weight of 1-ethyl-3-methylimidazolium tetrafluoroborate (Sigma-Aldrich), 25 parts by weight of poly(3,4-ethylenediooxythiophene)-polystyrene sulfonate (Sigma-Aldrich, high conductivity grade), 18 parts by weight of molybdenum disulfide (Sigma-Aldrich, average particle size approx. 12 μm), 17 parts by weight of hexagonal boron nitride (Sigma-Aldrich, average particle size approx. 5 μm or less), and 20 parts by weight of chitosan oligosaccharide (Sigma-Aldrich, average molecular weight approx. 5,000 or less) are precisely weighed and mixed to a total solid content of 100 parts by weight, and then polyvinyl alcohol (Polyvinyl Alcohol, PVA, Sigma-Aldrich, average molecular weight approx. 30,000-50,000) is added as a binder at 5% by weight relative to the total amount of the mixture. Deionized water was used as the solvent to adjust the total solid content of the mixture to 15% by weight, and the mixture was homogeneously mixed using an ultrasonic disperser for 30 minutes.

[0185] This mixture was uniformly coated to a thickness of 100 μm on the skin contact surface of the myofascial microcurrent therapy patch prepared in Comparative Example 2 using a spray coating method, then pre-dried in a hot air dryer at 80°C for 30 minutes, and then further dried in a vacuum dryer at 60°C for 24 hours to form a biocompatible-highly conductive composite mixture coating layer on the skin contact surface of the myofascial microcurrent therapy patch of the present example.

[0186] [Experiment 1: Microcurrent Transfer Efficiency Evaluation Experiment]

[0187] This experiment was conducted in accordance with IEC 60601-2-10 (standard for basic safety and essential performance evaluation of medical electrical stimulation devices).

[0188] After uniformly attaching the fascial microcurrent treatment patches of Comparative Examples 1, 2, 3 and the Examples to the surface of an artificial skin model (Skin-mimicking phantom, Sigma-Aldrich, electrical resistance approx. 500Ω, thickness 3mm) with the same force (approx. 2.4N / cm²), a microcurrent of 200μA was set at the microcurrent output of each patch using a precision DC power supply (Keithley 2450 SourceMeter) and applied for 1 minute.

[0189] Next, the microcurrent passing through the actual skin model was measured using a precision current meter (Keithley DMM7510 7½ Digit Multimeter) connected to the opposite side of the artificial skin model, and the average value was calculated after a total of 60 measurements at 1-second intervals to calculate the microcurrent transmission efficiency (%) of each comparative example and example.

[0190] The above measurements were repeated a total of 5 times for each comparative example and example, and the average of the results was calculated and presented as the final result.

[0191] The following Table 1 shows the experimental results.

[0192] Microcurrent setting value (μA) Average measured value (μA) Transmission efficiency (%) Comparative Example 1 200 138.7 69.35 Comparative Example 2 200 151.4 75.70 Comparative Example 3 200 164.3 82.15 Examples 200 188.6 94.30

[0193] From the above results, it was confirmed that the example coated with the biocompatible-highly conductive composite mixture of the present invention exhibited excellent microcurrent transmission characteristics with significantly higher microcurrent transmission efficiency compared to Comparative Examples 1 to 3.

[0194] [Experiment 2: Skin Irritation and Skin Adhesion Evaluation Experiment]

[0195] This experiment was conducted in accordance with ISO 10993-10 (Biological evaluation of medical devices - skin irritation and skin sensitization test) and ASTM D3330 / D3330M-04 (method for evaluating peel strength of adhesive tape).

[0196] The fascial microcurrent therapy patches of each comparative example and embodiment were fabricated with an area of ​​3 cm × 3 cm, and a 3D artificial skin model (EpiDerm™, MatTek Corporation) with characteristics similar to human skin was used to evaluate skin irritation. Each sample patch was attached to the surface of the artificial skin for 24 hours in an environment of 37°C and 60% relative humidity. After the attachment was completed, the occurrence of skin irritation, such as redness or inflammation on the surface of the artificial skin, was evaluated using digital image analysis (high-resolution image capture and the image analysis program ImageJ) to calculate a skin irritation score (0 to 5 points, with lower values ​​indicating better performance).

[0197] Skin adhesion evaluation was performed by cutting each sample into a 25 mm wide test specimen according to ASTM D3330, attaching it to a stainless steel plate (SUS304, polished surface) at 37°C, and then measuring the average peel strength (N / cm) when peeling the sample from the steel plate using a 180-degree peel tester (Instron 3345, peel speed 300 mm / min).

[0198] The average value was calculated by repeating measurements 5 times for each comparative example and example.

[0199] The following Table 2 shows the experimental results.

[0200] Skin irritation score (points) Skin adhesion (peeling strength, N / cm) Comparative Example 1 3.13 0.93 Comparative Example 2 2.57 1.24 Comparative Example 3 2.15 1.58 Examples 0.68 2.17

[0201] From the above results, it was evaluated that the embodiment applying the biocompatible-highly conductive composite mixture of the present invention exhibits very low skin irritation and significantly high skin adhesion, demonstrating excellent biocompatibility and adhesive stability even in long-term skin attachment environments.

[0202] As explained above, the configuration and operation of the AI-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance according to the present invention have been described in the above description and drawings; however, this is merely an example, and the concept of the present invention is not limited to the above description and drawings. It is understood that various changes and modifications are possible within the scope of the technical concept of the present invention. Explanation of the symbols

[0203] 110: User Information Collection Module 120: Prediction Module 121: Graph Data Generation Unit 122: Importance Calculation Unit 123: Priority determination unit 130: Information provision module 200: Fascial microcurrent therapy patch 210: IoT sensor

Claims

Claim 1 As an AI-based customized information provision healthcare platform system for musculoskeletal pain relief and rehabilitation assistance, the system comprises: a user information collection module that receives symptoms and biometric information related to the user's musculoskeletal system; and a prediction module that analyzes the symptoms and biometric information based on an AI algorithm to generate prediction information predicting the user's rehabilitation needs. and, based on the prediction information, an information provision module that provides customized healthcare information to the user, including rehabilitation exercise programs, recommendations for rehabilitation exercise tools, information on visiting rehabilitation exercise services, hospital information, and health functional foods; wherein the prediction module generates prediction information that predicts the need for rehabilitation for the user's fascia injury by analyzing the symptoms and bio-information based on a Graph Neural Network (GNN) algorithm, and comprises a graph data generation unit that converts the symptoms into data in the form of a graph based on human anatomical structure, an importance calculation unit that calculates the correlation between adjacent musculoskeletal areas connected to the pain occurrence site in the graph data and selects the area requiring rehabilitation by auxiliaryly referencing the bio-information, and a priority determination unit that generates prediction information that determines the area requiring rehabilitation exercise by classifying the need for rehabilitation of the selected area, and wherein the information provision module provides information recommending the use of a fascia microcurrent therapy patch for rehabilitation for the fascia injury to the user among the rehabilitation exercise tools, wherein the skin contact surface of the fascia microcurrent therapy patch is made of thermoplastic polyurethane, It is made of TPU, and the surface of the skin contact surface is coated with a biocompatible-highly conductive composite mixture for improving microcurrent transmission efficiency and alleviating skin irritation, wherein the biocompatible-highly conductive composite mixture comprises, based on 100 parts by weight, 1-ethyl-3-methylimidazolium tetrafluoroborate,A healthcare platform system characterized by comprising 15 to 25 parts by weight of EMIM-BF4, 20 to 30 parts by weight of poly(3,4-ethylene dioxythiophene)-polystyrene sulfonate (PEDOT:PSS), 15 to 20 parts by weight of molybdenum disulfide (MoS2), 15 to 20 parts by weight of hexagonal boron nitride (h-BN), and 10 to 20 parts by weight of chitosan oligosaccharide. Claim 2 A healthcare platform system according to claim 1, wherein the fascia microcurrent therapy patch comprises an IoT sensor that generates sensing information indicating the current intensity of the microcurrent at the skin contact surface, the user information collection module receives the sensing information from the IoT sensor, and the prediction module analyzes the sensing information to generate prediction information that updates the need for rehabilitation for the fascia damage site in real time. Claim 3 delete Claim 4 delete Claim 5 delete Claim 6 delete

Images