Taijiquan exercise prescription rehabilitation training system and method based on virtual reality technology

Abstract

The invention belongs to but is not limited to the technical field of traditional Chinese medicine exercise therapy, and discloses a shadowboxing exercise prescription rehabilitation training system and method based on a virtual reality technology, a virtual reality presentation device is adapted to an immersive three-dimensional scene constructed by Unity or an equivalent three-dimensional engine; the motion capture and feedback module is integrated with an inertial measurement unit, a space positioning device and an attitude sensing node; the central processing unit is configured with an instruction scheduling module, a data integration module and a rendering control module; the training prescription generation module is provided with a joint state input interface and a training parameter output interface; the real-time rendering module is used for driving the three-dimensional model, the action animation and the interactive interface; the multi-channel audio management module comprises a resource loading interface and a spatialization audio output interface; the man-machine interaction module comprises a control input channel, a virtual image mapping engine and a state monitoring unit. According to the method, the scientificity, individuation and compliance of treatment are remarkably improved, and the method has a wide clinical application prospect.

Description

Technical Field

[0001] The present invention belongs to but is not limited to the technical field of traditional Chinese medicine exercise therapy, and in particular relates to a Tai Chi exercise prescription rehabilitation training system and method based on virtual reality technology. Background Art

[0002] Knee osteoarthritis (KOA) is a common degenerative joint disease, with patients often experiencing symptoms such as knee pain, stiffness, and limited mobility. While traditional treatments such as medication and surgery are effective, they come with increasing side effects and financial burdens, leading to the emergence of non-drug therapies as a key treatment approach for KOA. Exercise therapy, as an effective treatment, has been recommended by multiple medical guidelines for the treatment of KOA. However, existing exercise therapies suffer from low standardization, insufficient personalization, and poor patient compliance. Existing technologies lack personalized exercise prescriptions for KOA patients, and traditional Tai Chi exercise prescriptions fail to integrate modern technology to provide real-time feedback and adjustments. This lack of patient engagement and precise guidance makes it difficult to maximize treatment outcomes.

[0003] Prior art 1 (CN108345678A) discloses a Tai Chi rehabilitation training system based on video demonstration, which provides movement demonstrations for KOA patients through pre-recorded Tai Chi teaching videos and is equipped with a simple posture recognition module based on a two-dimensional camera to roughly compare and prompt the patient's overall movements.

[0004] Existing technical problems: 1. Lack of personalization: The system only provides unified demonstration videos and cannot dynamically adjust training content based on clinical indicators such as the patient's knee function score and pain index; 2. Delayed and coarse feedback: 2D camera gesture recognition has limited accuracy, providing only a "correct / incorrect" prompt and lacking real-time quantitative feedback on 3D motion deviations. 3. Poor compliance: Lack of immersive interaction and incentive mechanisms leads to insufficient patient participation and motivation for training, making it difficult for patients to persist in the long term.

[0005] Prior art 2 (CN109876543A) discloses a virtual reality rehabilitation system based on IMU and depth camera, which uses an inertial measurement unit (IMU) and a depth camera to collect user motion data and presents user movements in a virtual environment through Avatar mapping, for use in neurological rehabilitation and orthopedic postoperative training.

[0006] Existing technical problems: 1. Not optimized for KOA's characteristics: The system's movement library primarily consists of general rehabilitation movements, lacking the breakdown and grading of Tai Chi-specific movements tailored to KOA's pathological characteristics; 2. No prescription-level management: The system lacks a prescription generation module linked to medical scores and pain indexes, and cannot automatically output prescription parameters for training intensity, frequency, and duration. 3. High system complexity and lack of usability: The collaborative deployment and calibration process of multiple devices is cumbersome, making it difficult for patients to independently complete device installation and debugging at home, hindering its promotion and application. Summary of the Invention

[0007] In response to the problems existing in the existing technology, the present invention provides a KOA Tai Chi exercise prescription rehabilitation training system and method based on virtual reality (VR) technology, which screens and classifies the Tai Chi exercise prescriptions most suitable for KOA patients, and combines the application of VR technology to provide personalized and precise exercise intervention plans.

[0008] The present invention is implemented as follows: A Tai Chi exercise prescription rehabilitation training system based on virtual reality technology includes: a virtual reality presentation device, a motion capture and feedback module, a central processing unit, a training prescription generation module, a real-time rendering module, a multi-channel audio management module, a human-computer interaction module, and an intelligent interaction and remote monitoring module. Each module is connected via a bus and is synchronously called and command-controlled by the central processing unit. The virtual reality presentation device is adapted to an immersive three-dimensional scene built with Unity or an equivalent three-dimensional engine; it is used to display demonstration action animations, the user's movements and postures in the virtual environment, and the audio environment; The motion capture and feedback module integrates an inertial measurement unit, a spatial positioning device, and a posture sensing node; it is used to capture the user's head, hand, and leg data in real time, and transmit the data collected by each node to the central processing unit through a wireless synchronous transmission interface for posture fusion; The central processing unit is equipped with an instruction scheduling module, a data integration module and a rendering control module; The training prescription generation module has a joint status input interface and a training parameter output interface; it is used to formulate personalized exercise training prescriptions according to the user's status, automatically adjust the exercise training prescriptions, and provide personalized training suggestions and real-time corrections; The real-time rendering module is used to drive the 3D model, action animation and interactive interface; it is used to provide 3D model rendering, animation playback, light and shadow effects and other services for demonstration actions and virtual image mapping, and build a realistic virtual training environment; The multi-channel audio management module includes a resource loading interface and a spatialized audio output interface; it is used to load, play and control audio resources, and to broadcast text or guide user operations through voice prompts; The human-computer interaction module includes a control input channel, a virtual image mapping engine and a status monitoring unit; The intelligent interaction and remote monitoring module, through the integration of biofeedback equipment, can monitor the patient's physiological condition in real time and provide real-time feedback on the VR training process; it has remote data synchronization and real-time monitoring functions; it has the ability of self-learning and evolution, and can continuously optimize the training plan based on the patient's training data and feedback.

[0009] Furthermore, the virtual reality presentation device includes a head-mounted display, has a visual image interface, an inertial positioning unit and an audio playback channel, and supports the XR API protocol or an equivalent extended reality standard.

[0010] Furthermore, the motion capture component in the motion capture and feedback module includes: IMU sensor node for head positioning; Hand controller for upper limb movement detection; Leg-strap inertial module for lower limb motion acquisition.

[0011] Furthermore, the motion capture and feedback module is also provided with a three-dimensional joint mechanical model and dynamic feedback of the virtual joint state. Through the high integration of the inertial measurement unit, the spatial positioning device and the posture perception node, the system can construct an accurate joint motion model in three-dimensional space, and monitor and feedback the biomechanical state of the joint in real time, such as joint force, range of motion, gait and other parameters; in addition, the system uses AI algorithms to conduct in-depth analysis of motion data to determine whether the movement meets the individual's treatment needs, and issues prompts in real time to ensure that the execution of each movement is based on scientific biomechanics, avoid excessive or insufficient exercise, and reduce rehabilitation risks.

[0012] Furthermore, the training prescription generation module includes: Functional scoring input unit, used to import knee joint scoring data; Pain indicator access interface, used to input external physiological state parameters; Parameter matching operation unit, used to select training action combinations based on the rule base; The exercise intensity configuration unit is used to set the frequency, duration and amplitude information of the output prescription.

[0013] Furthermore, the real-time rendering module includes: Action animation manager, used to call standard action sequences; 3D model binding engine for implementing skeletal animation mapping; A scene construction unit, used to load the training scene resource package and output it to the display device; Rendering task scheduling interface, connecting to the central processing unit for frame-level control.

[0014] Furthermore, the multi-channel audio management module specifically includes: Audio content resource library; Audio playback scheduling module; Action phase trigger control interface; Spatial sound synthesizer, configured with three-dimensional sound field parameters; The audio module and the action module are linked together through the action identification ID.

[0015] Furthermore, the multi-channel audio management module features personalized voice guidance and virtual interaction. Leveraging deep learning technology, the system automatically adjusts the content and volume of voice guidance based on the patient's emotional state, training progress, and physiological feedback. This provides customized voice prompts, encouragement, and adjustment suggestions during training, enhancing the patient's emotional connection and motivation for recovery. Furthermore, the system includes virtual interaction capabilities, allowing patients to interact with virtual avatars through voice commands, controlling training content or adjusting exercise plans, further enhancing the system's human-computer interaction capabilities.

[0016] Furthermore, the human-computer interaction module includes: A control input recognition unit, configured to receive handle button or gesture input signals; an avatar bone mapping unit for redirecting the captured data to a three-dimensional character model; State feedback renderer, used to present the current user posture and interactive operation response information; An interface module that establishes a bidirectional data link with the graphics rendering module and the audio module.

[0017] Furthermore, the intelligent interaction and remote monitoring module has a bio-signal feedback mechanism; by integrating biofeedback equipment (such as heart rate sensors, electromyography, skin conductivity, etc.), the system can monitor the patient's physiological condition in real time and provide real-time feedback on the VR training process; for example, during rehabilitation training, if the system detects that the patient's heart rate is too fast or the muscles are fatigued, it can intelligently adjust the training intensity or switch to a more appropriate training mode; in addition, the dynamic training content and bio-signal data in the virtual scene are displayed in real time through the interactive interface, helping patients to perceive the training effects and physiological changes more intuitively, and enhance their sense of participation and effectiveness in training.

[0018] Furthermore, the intelligent interaction and remote monitoring module features remote data synchronization and real-time monitoring. Patient training data (including movement quality, joint forces, movement speed, and frequency) is synchronized to a cloud database in real time, allowing medical personnel to provide intervention and guidance through remote monitoring from any location and at any time. Doctors can review patients' exercise data through the system's data analysis platform and adjust training prescriptions or issue personalized recommendations based on their specific needs. This innovation significantly enhances the effectiveness of remote rehabilitation, ensuring that patients receive professional medical guidance and oversight from home.

[0019] Furthermore, the intelligent interaction and remote monitoring module has the ability of self-learning and evolution. The system's built-in self-learning module can continuously optimize the training plan based on the patient's training data and feedback. The training process of each patient will be continuously adjusted and optimized through machine learning algorithms to improve rehabilitation efficiency and personalization. The system can identify the patient's pain points and progress during training, automatically adjust the intensity and mode of exercise, and even predict and prepare future rehabilitation training content in advance through historical data. In addition, the system has a certain evolutionary ability. Through the experience accumulated during the long-term training process, it can not only optimize the rehabilitation plan for individual patients, but also gradually adjust the system's overall treatment strategy based on group data feedback to adapt to the needs of different patients.

[0020] Another object of the present invention is to provide a Tai Chi exercise prescription rehabilitation training method based on virtual reality technology, comprising the following steps: S101, using a virtual reality presentation device to load multiple virtual environments built by the Unity engine; S102, collecting user's motion data through a motion capture component; S103, transmitting the collected motion data to a data conversion unit via a predetermined data transmission interface; S104, transmitting the converted data to a training prescription generation module, and generating prescription data according to a predetermined data format; S105: Output the prescription data to the VR display interface through the feedback module.

[0021] Furthermore, the motion data collection and processing method based on virtual reality technology includes the following steps: S201, using a motion capture component to collect multi-dimensional motion data of the head, hands, and legs; S202, transmitting the collected motion data to a data fusion unit via a predetermined interface; S203, integrating the format of the transferred data and converting it into a predetermined data structure; S204: Output the integrated data to a graphics rendering module for 3D data mapping processing.

[0022] Furthermore, the real-time action feedback method based on virtual reality technology includes the following steps: S301, using a motion capture component to collect user motion data in real time; S302, transmitting the collected data to the engine interface via a predetermined transmission interface; S303, formatting the transmitted data according to a predetermined data conversion rule; S304, mapping the processed data to the user avatar data structure and outputting it via a graphics rendering module; S305: Present the mapping data in a 3D space using a human-computer interaction module.

[0023] First, the present invention provides an innovative KOA Tai Chi exercise prescription training system based on virtual reality technology. Through personalized exercise prescriptions, immersive training using virtual reality technology, and real-time feedback, this system significantly improves patient exercise compliance and treatment outcomes, and has broad clinical application prospects. The present invention's technical solution, combining VR technology with traditional Chinese medicine exercises, innovatively provides KOA patients with personalized, precise exercise rehabilitation treatment plans. Through this innovative model, patients can undergo rehabilitation training in an immersive virtual environment, and combined with artificial intelligence (AI) technology, real-time monitoring, feedback, and optimization of treatment plans, greatly enhancing the scientific nature, personalization, and compliance of treatment. This technological breakthrough not only overcomes the limitations of traditional rehabilitation methods that rely on professional physicians and equipment, but also addresses geographical and time constraints during the patient's rehabilitation process through remote monitoring capabilities, significantly improving the efficient utilization of medical resources.

[0024] Clinical Validation: Preliminary clinical data indicates that patients using this system experienced significant improvements in knee pain, stiffness, and joint function within just four weeks. Patients in the treatment group also showed significant improvements in exercise compliance, and their self-management abilities were enhanced compared to traditional treatments.

[0025] Second, as auxiliary evidence for the inventiveness of the claims of the present invention, it is also reflected in the following important aspects: (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are as follows: With the growth of the aging population worldwide, the number of patients with knee joint diseases continues to increase. Traditional rehabilitation treatment methods cannot meet the growing personalized needs, and the digital rehabilitation solution of the present invention can fill this market gap. Through the combination of remote rehabilitation and digital platforms, the time and space limitations of traditional treatments are broken through, allowing patients to undergo personalized and precise rehabilitation training at home, reducing the treatment costs and time costs of patients, while improving the accessibility of rehabilitation treatment. This feature not only provides patients with a more convenient and efficient treatment method, but also greatly reduces the economic burden on society and families, and has significant social value.

[0026] In terms of commercialization, the present invention has broad market application prospects. Its core technology can be promoted and applied in hospitals, rehabilitation centers, nursing homes and other places. In the future, it can also be expanded to more fields such as sports rehabilitation and aging medicine to form a diversified market model. Through various methods such as hardware equipment sales, subscription platform services, remote treatment and data management, the commercialization path of the project is not only clear but also has high growth potential. In addition, combined with the digital transformation of traditional Chinese medicine culture, the project also has the potential for international promotion, and is expected to promote the modern dissemination of Chinese medicine rehabilitation culture on a global scale. With the growth of market demand and the continuous maturity of technology, this project is expected to become a leader in the field of digital medical rehabilitation, create significant economic benefits, and drive the development of related industrial chains.

[0027] Through continuous optimization of technology and market expansion, the economic benefits it brings will gradually become apparent, driving the entire industry to transform in the direction of digitalization and intelligence, and becoming an important part of the future rehabilitation medical field.

[0028] (2) The technical solution of this invention fills the technical gap in the industry at home and abroad: The technical solution of the present invention fills the technological gap in the field of digital sports rehabilitation at home and abroad by innovatively integrating traditional Chinese medicine rehabilitation exercises with VR technology and AI algorithms. At present, although traditional Chinese medicine rehabilitation methods, such as Tai Chi and Ba Duan Jin, have accumulated valuable experience in many years of application, they rely on individual experience and manual guidance, lack personalized and standardized treatment methods, and patients have poor treatment compliance. Especially in the absence of professional guidance, it is difficult to maximize the rehabilitation effect. The present invention realizes the digitization, standardization and personalization of Chinese medicine exercises through VR immersive environment and motion capture technology, allowing patients to self-regulate and train according to their personal rehabilitation needs in virtual space, overcoming the time and space limitations of traditional treatments and providing patients with accurate exercise prescriptions.

[0029] Furthermore, the AI-driven biomechanical feedback system employed in this invention can monitor and analyze a patient's motion data in real time, automatically optimizing rehabilitation plans based on individual physiological characteristics and pathological progression. This technology not only enhances the personalization and precision of rehabilitation treatment, but also addresses shortcomings in existing exercise rehabilitation technologies, such as a lack of dynamic adjustment and compliance with exercise prescriptions. In particular, in remote rehabilitation scenarios, utilizing cloud-based data management and remote monitoring technology, patients can undergo efficient rehabilitation training at home without visiting a hospital, while maintaining real-time interaction with medical staff and adjusting treatment.

[0030] Compared to various rehabilitation products both domestically and internationally, this invention utilizes sensors combined with AI in motion capture and recognition technology, resulting in more accurate motion capture than other products that rely on computer vision or sensors alone. Its immersive rehabilitation training, compared to direct exercise, is more likely to increase patient interest and compliance. The innovative integration of traditional Chinese medicine exercises into KOA patient rehabilitation, coupled with real-time monitoring of rehabilitation data to ensure safety and prevent adverse events, is another key innovation, filling a gap in existing product offerings.

[0031] This invention breaks through the limitations of traditional remote rehabilitation treatment and fills the technological gap in intelligent and precise remote rehabilitation. Existing remote rehabilitation technologies mostly rely on video teaching or simple feedback mechanisms, making it difficult to achieve dynamic personalized adjustment and efficient treatment. In contrast, the present invention combines virtual reality with AI technology to create a comprehensive and intelligent remote rehabilitation platform, enabling precise interaction between patients and doctors, and enabling personalized rehabilitation training at any time and place, significantly improving patients' treatment compliance and rehabilitation effects.

[0032] (3) The technical solution of the present invention solves the technical problems that people have been eager to solve but have never been able to solve successfully: The present invention solves several key technical problems that have long been unsolved in the field of rehabilitation medicine, especially in the areas of personalized sports rehabilitation, precise real-time dynamic adjustment, digital transformation of traditional Chinese medicine exercises, and remote intelligent rehabilitation management.

[0033] First of all, traditional Chinese medicine rehabilitation methods, such as Tai Chi and Ba Duan Jin, have been proven to have significant effects on promoting physical health and regulating Qi and blood over thousands of years. However, these traditional exercises mostly rely on the patient's independent practice and lack systematic and personalized treatment and training guidance. In particular, the movements of traditional exercises of different schools are different, making it impossible for the rehabilitation process to accurately match the specific needs of each patient. Although traditional Chinese medicine rehabilitation methods have advantages in theory, they often lack quantitative standards and effective personalized adjustment mechanisms in practical applications. The present invention uses the combination of VR technology and AI to digitally transform traditional Chinese medicine exercises, realize the standardization, personalization and real-time feedback of movements, and enable traditional Chinese medicine rehabilitation methods to be accurately and flexibly adjusted in a virtual environment, thereby providing patients with scientific and accurate rehabilitation plans, overcoming the problem that traditional Chinese medicine exercises cannot be standardized.

[0034] Secondly, although exercise therapy is widely used in the rehabilitation of chronic diseases, its popularity is still limited by the lack of personalized treatment and poor treatment compliance. Traditional rehabilitation programs mostly rely on on-site guidance from doctors and cannot provide real-time adjustments according to changes in the patient's condition anytime and anywhere. Through the deep integration of VR and AI technologies, the present invention can not only realize the formulation and dynamic optimization of personalized exercise prescriptions, but also combine the treatment theory of traditional Chinese medicine exercises with modern technology to form a new, intelligent rehabilitation model. In a virtual reality environment, patients can perform rehabilitation training through traditional exercises such as Tai Chi and Ba Duan Jin, and use AI to adjust the accuracy and therapeutic effect of each movement in real time, solving the problem that traditional exercise therapy relies on manual guidance and cannot be flexibly adjusted according to the specific needs of patients.

[0035] Finally, telerehabilitation has always been a desired solution for the medical community and patients, especially in situations where resources are limited and regionally distributed. Existing telerehabilitation methods often rely on video guidance or simple feedback mechanisms, lacking precise personalized treatment plans and real-time data analysis, making it difficult to provide truly effective teletherapy. This invention addresses the issues of personalization, precision, and real-time dynamic adjustment of teletherapy. Through the digital application of traditional Chinese medicine exercises, patients can also undergo precise rehabilitation training that meets their individual needs in a home environment.

[0036] (4) The technical solution of the present invention overcomes technical prejudice: First of all, for a long time, traditional Chinese medicine exercises have been regarded as a relatively abstract and difficult to quantify treatment method. Many scholars and clinical experts believe that traditional Chinese medicine rehabilitation methods lack the scientific basis of modern medicine and are difficult to combine with modern medical technology. The ambiguity of traditional Chinese medicine theory and the difficulty in standardizing personalized treatment plans have led to certain technical biases in the modern medical system. Many technology developers separate modern technology and traditional Chinese medicine, believing that the two are difficult to integrate, resulting in limited application of traditional Chinese medicine exercises in the field of modern rehabilitation. The technical solution of the present invention successfully digitizes traditional Chinese medicine exercises by combining VR technology and AI, overcomes traditional prejudices, and enables traditional Chinese medicine pathological treatment methods to be accurately and standardized through modern technology, verifying the effectiveness and operability of traditional Chinese medicine exercises with the support of modern technology.

[0037] Secondly, although exercise rehabilitation is theoretically considered an effective treatment for patients with chronic diseases, many doctors and patients are skeptical about the controllability of its treatment effects and the feasibility of personalized treatment. Many traditional exercise therapies rely on unified treatment plans for most people and lack the ability to dynamically adjust according to the specific needs of each patient. Technical bias is reflected in the recognition of the limitations of exercise rehabilitation therapy. It is believed that once the exercise treatment plan is determined, it is difficult to optimize it according to the patient's real-time condition changes. The present invention solves this technical bias by integrating AI technology and real-time biomechanical data analysis. AI-driven personalized exercise prescriptions can automatically adjust the treatment intensity and movement sequence according to the patient's real-time data, breaking through the technical bottleneck that traditional treatment methods cannot flexibly respond to individual differences in patients, and improving the accuracy and effectiveness of exercise rehabilitation treatment.

[0038] In addition, remote rehabilitation therapy has always been considered difficult to achieve personalized treatment and real-time effective feedback. Traditional remote rehabilitation technologies mostly rely on video or simple data monitoring, which cannot provide in-depth interaction between patients and doctors, and cannot be accurately adjusted according to the actual situation of the patient. There is a widespread prejudice in the industry that remote treatment is difficult to achieve the same effect as on-site treatment, especially in terms of personalization and treatment quality. It is difficult to compare with traditional face-to-face treatment. The present invention successfully overcomes this technical prejudice. Through real-time data collection and analysis, it not only realizes the personalization of remote treatment, but also enables patients to obtain the same level of intelligent rehabilitation guidance as in the hospital in a home environment, completely breaking the technical prejudice that remote rehabilitation treatment is not effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a structural diagram of a Tai Chi exercise prescription rehabilitation training system based on virtual reality technology provided by an embodiment of the present invention.

[0040] Figure 2This is a flow chart of a Tai Chi exercise prescription rehabilitation training method based on virtual reality technology provided by an embodiment of the present invention.

[0041] Figure 3 This is a flow chart of the data collection and processing method provided by an embodiment of the present invention.

[0042] Figure 4 This is a flow chart of a real-time action feedback method provided by an embodiment of the present invention.

[0043] Figure 5 This is a diagram showing the effect of a user following a demonstration action during use provided by an embodiment of the present invention.

[0044] Figure 6 This is a playback effect diagram of the model avatar demonstration action provided by an embodiment of the present invention.

[0045] Figure 7 This is a schematic diagram of the initial interface of the system provided by an embodiment of the present invention.

[0046] Figure 8 2 is a schematic diagram of a device calibration interface provided by an embodiment of the present invention.

[0047] Figure 9 2 is a schematic diagram of an action selection interface provided by an embodiment of the present invention.

[0048] Figure 10 It is a system architecture diagram provided by an embodiment of the present invention.

[0049] Figure 11 This is a flowchart of clinical effectiveness verification provided by an embodiment of the present invention.

[0050] Figure 12 This is a diagram of a Tai Chi exercise prescription rehabilitation training device based on virtual reality technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] like Figure 1 As shown, an embodiment of the present invention provides a Tai Chi exercise prescription rehabilitation training system based on virtual reality technology, comprising: a virtual reality presentation device, a motion capture and feedback module, a central processing unit, a training prescription generation module, a real-time rendering module, a multi-channel audio management module, a human-computer interaction module, and an intelligent interaction and remote monitoring module. Each module is connected via a bus and is synchronously called and command-controlled by the central processing unit. The virtual reality presentation device is adapted to an immersive three-dimensional scene built with Unity or an equivalent three-dimensional engine; it is used to display demonstration action animations, the user's movements and postures in the virtual environment, and the audio environment; The motion capture and feedback module integrates an inertial measurement unit, a spatial positioning device, and a posture sensing node; it is used to capture the user's head, hand, and leg data in real time, and transmit the data collected by each node to the central processing unit through a wireless synchronous transmission interface for posture fusion; The central processing unit is equipped with an instruction scheduling module, a data integration module and a rendering control module; The training prescription generation module has a joint status input interface and a training parameter output interface; it is used to formulate personalized exercise training prescriptions according to the user's status, automatically adjust the exercise training prescriptions, and provide personalized training suggestions and real-time corrections; The real-time rendering module is used to drive the 3D model, action animation and interactive interface; it is used to provide 3D model rendering, animation playback, light and shadow effects and other services for demonstration actions and virtual image mapping, and build a realistic virtual training environment; The multi-channel audio management module includes a resource loading interface and a spatialized audio output interface; it is used to load, play and control audio resources, and to broadcast text or guide user operations through voice prompts; The human-computer interaction module includes a control input channel, a virtual image mapping engine and a status monitoring unit; The intelligent interaction and remote monitoring module, through the integration of biofeedback equipment, can monitor the patient's physiological condition in real time and provide real-time feedback on the VR training process; it has remote data synchronization and real-time monitoring functions; it has the ability of self-learning and evolution, and can continuously optimize the training plan based on the patient's training data and feedback.

[0053] After the system is activated, the motion capture and feedback module uses an inertial measurement unit (IMU) and optical or ultrasonic positioning nodes to capture the spatial posture of the user's key body parts (head, hands, and legs) in real time. Each sensor node collects raw sensor signals such as triaxial acceleration, angular velocity, and magnetic flux density. These signals are pre-processed using an internal Kalman filter algorithm to reduce noise interference and then bound to the corresponding joints of the user's virtual skeleton to form the initial posture recognition input.

[0054] Processed sensor data is transmitted to the central processing unit (CPU) in structured data frames (such as JSON or binary packets) via low-latency wireless communication protocols (such as Wi-Fi 6, Bluetooth Low Energy, or ZigBee). To ensure high-frequency, low-latency transmission, the system employs a timestamp synchronization mechanism and a buffered frame prediction algorithm to prevent packet out-of-order and frame loss, improving transmission stability and real-time performance. The data frames contain information such as the unique identifier of each acquisition node, timestamp, spatial coordinates, and attitude quaternion, forming a multi-channel asynchronous input source.

[0055] After receiving data from multiple nodes, the central processing unit enters the data integration module for posture fusion processing. This fusion process uses an extended Kalman filter (EKF) or complementary filter algorithm, combining spatial positioning data with IMU information to correct the Euler angles / quaternions of each node, thereby achieving multi-source posture fusion of the head, hands, and feet. The system further uses a forward kinematics (FK) algorithm combined with a human motion constraint model to reconstruct the user's dynamic skeletal structure, providing an accurate spatial input foundation for subsequent rendering and interaction.

[0056] After completing posture recognition, the training prescription generation module receives the processed time-series motion data and matches it with a built-in motion database and state recognition models (such as LSTM-RNN or Transformer architectures). The system automatically calls upon the training rule library based on user input, including range of motion (ROM), muscle strength assessment metrics, training history, and rehabilitation goals. Using multi-parameter decision logic (based on fuzzy logic or Bayesian networks), it generates a personalized training prescription structure, encompassing key parameters such as movement sequence, target amplitude, intensity threshold, frequency, and duration.

[0057] The prescription parameters and fused skeleton data are synchronously transmitted to the real-time rendering module, which internally calls a 3D model binding engine (such as Unity Animator + Humanoid Rig system) for motion mapping. The motion data is then processed on the GPU using skeleton-driven mapping (using the Skinned Mesh Renderer's bone weight transformation mechanism). This is combined with real-time lighting and shadow rendering technologies (such as PBR shaders or URP / Lit shaders) to produce highly realistic virtual character motion performance. This module also manages the unified rendering orchestration of system motion cues, feedback effects, and environmental response information.

[0058] After rendering is complete, the image is output to the virtual reality display terminal; the output timing and voice content of the audio guidance instructions are controlled by the multi-channel audio module. The module supports event-triggered binding (Event-based ID Mapping) with the action recognition ID to ensure that the audio and action are played synchronously. At the same time, the human-computer interaction module receives voice, button or gesture input from the user, and after passing through the instruction parsing module, it is fed back to the central processing unit in real time to realize the reflow of operation instructions such as switching, pausing, and resetting the training plan. The system can monitor the user's deviation from the prescription based on the set threshold and immediately generate correction prompts to complete the entire interactive closed loop.

[0059] Furthermore, the virtual reality presentation device includes a head-mounted display, has a visual image interface, an inertial positioning unit and an audio playback channel, and supports the XR API protocol or an equivalent extended reality standard.

[0060] Furthermore, the motion capture component in the motion capture and feedback module includes: IMU sensor node for head positioning; Hand controller for upper limb movement detection; Leg-strap inertial module for lower limb motion acquisition.

[0061] Furthermore, the motion capture and feedback module is also provided with a three-dimensional joint mechanical model and dynamic feedback of the virtual joint state. Through the high integration of the inertial measurement unit, the spatial positioning device and the posture perception node, the system can construct an accurate joint motion model in three-dimensional space, and monitor and feedback the biomechanical state of the joint in real time, such as joint force, range of motion, gait and other parameters; in addition, the system uses AI algorithms to conduct in-depth analysis of motion data to determine whether the movement meets the individual's treatment needs, and issues prompts in real time to ensure that the execution of each movement is based on scientific biomechanics, avoid excessive or insufficient exercise, and reduce rehabilitation risks.

[0062] Furthermore, the training prescription generation module includes: Functional scoring input unit, used to import knee joint scoring data; Pain indicator access interface, used to input external physiological state parameters; Parameter matching operation unit, used to select training action combinations based on the rule base; The exercise intensity configuration unit is used to set the frequency, duration and amplitude information of the output prescription.

[0063] Furthermore, the real-time rendering module includes: Action animation manager, used to call standard action sequences; 3D model binding engine for implementing skeletal animation mapping; A scene construction unit, used to load the training scene resource package and output it to the display device; Rendering task scheduling interface, connecting to the central processing unit for frame-level control.

[0064] Furthermore, the multi-channel audio management module specifically includes: Audio content resource library; Audio playback scheduling module; Action phase trigger control interface; Spatial sound synthesizer, configured with three-dimensional sound field parameters; The audio module and the action module are linked together through the action identification ID.

[0065] Furthermore, the multi-channel audio management module features personalized voice guidance and virtual interaction. Leveraging deep learning technology, the system automatically adjusts the content and volume of voice guidance based on the patient's emotional state, training progress, and physiological feedback. This provides customized voice prompts, encouragement, and adjustment suggestions during training, enhancing the patient's emotional connection and motivation for recovery. Furthermore, the system includes virtual interaction capabilities, allowing patients to interact with virtual avatars through voice commands, controlling training content or adjusting exercise plans, further enhancing the system's human-computer interaction capabilities.

[0066] Furthermore, the human-computer interaction module includes: A control input recognition unit, configured to receive handle button or gesture input signals; an avatar bone mapping unit for redirecting the captured data to a three-dimensional character model; State feedback renderer, used to present the current user posture and interactive operation response information; An interface module that establishes a bidirectional data link with the graphics rendering module and the audio module.

[0067] Furthermore, the intelligent interaction and remote monitoring module has a bio-signal feedback mechanism; by integrating biofeedback equipment (such as heart rate sensors, electromyography, skin conductivity, etc.), the system can monitor the patient's physiological condition in real time and provide real-time feedback on the VR training process; for example, during rehabilitation training, if the system detects that the patient's heart rate is too fast or the muscles are fatigued, it can intelligently adjust the training intensity or switch to a more appropriate training mode; in addition, the dynamic training content and bio-signal data in the virtual scene are displayed in real time through the interactive interface, helping patients to perceive the training effects and physiological changes more intuitively, and enhance their sense of participation and effectiveness in training.

[0068] Furthermore, the intelligent interaction and remote monitoring module features remote data synchronization and real-time monitoring. Patient training data (including movement quality, joint forces, movement speed, and frequency) is synchronized to a cloud database in real time, allowing medical personnel to provide intervention and guidance through remote monitoring from any location and at any time. Doctors can review patients' exercise data through the system's data analysis platform and adjust training prescriptions or issue personalized recommendations based on their specific needs. This innovation significantly enhances the effectiveness of remote rehabilitation, ensuring that patients receive professional medical guidance and oversight from home.

[0069] Furthermore, the intelligent interaction and remote monitoring module has the ability of self-learning and evolution. The system's built-in self-learning module can continuously optimize the training plan based on the patient's training data and feedback. The training process of each patient will be continuously adjusted and optimized through machine learning algorithms to improve rehabilitation efficiency and personalization. The system can identify the patient's pain points and progress during training, automatically adjust the intensity and mode of exercise, and even predict and prepare future rehabilitation training content in advance through historical data. In addition, the system has a certain evolutionary ability. Through the experience accumulated during the long-term training process, it can not only optimize the rehabilitation plan for individual patients, but also gradually adjust the system's overall treatment strategy based on group data feedback to adapt to the needs of different patients.

[0070] like Figure 2 As shown, the embodiment of the present invention provides a virtual reality technology-based Tai Chi exercise prescription rehabilitation training system and a virtual reality technology-based Tai Chi exercise prescription rehabilitation training method, including the following steps: S101, using a virtual reality presentation device to load multiple virtual environments built by the Unity engine; S102, collecting user's motion data through a motion capture component; S103, transmitting the collected motion data to a data conversion unit via a predetermined data transmission interface; S104, transmitting the converted data to a training prescription generation module, and generating prescription data according to a predetermined data format; S105: Output the prescription data to the VR display interface through the feedback module.

[0071] like Figure 3 As shown, an embodiment of the present invention provides a motion data acquisition and processing method based on virtual reality technology, comprising the following steps: S201, using a motion capture component to collect multi-dimensional motion data of the head, hands, and legs; S202, transmitting the collected motion data to a data fusion unit via a predetermined interface; S203, integrating the format of the transferred data and converting it into a predetermined data structure; S204: Output the integrated data to a graphics rendering module for 3D data mapping processing.

[0072] like Figure 4 As shown, an embodiment of the present invention provides a real-time action feedback method based on virtual reality technology, comprising the following steps: S301, using a motion capture component to collect user motion data in real time; S302, transmitting the collected data to the engine interface via a predetermined transmission interface; S303, formatting the transmitted data according to a predetermined data conversion rule; S304, mapping the processed data to the user avatar data structure and outputting it via a graphics rendering module; S305: Present the mapping data in a 3D space using a human-computer interaction module. Example

[0073] Functional interface of Tai Chi exercise prescription rehabilitation training system based on virtual reality technology System initial interface: After the system is started, the user will see the system initial interface as shown in the figure, which mainly includes the options of "Device Calibration", "Start Directly", "Select Action", and "Exit System". The "Device Calibration" option is used to calibrate the PICO4 VR all-in-one machine, handles, and somatosensory trackers to ensure that the device can accurately capture user movements and improve the accuracy of motion recognition. Click "Start Directly" and the system will start the sports rehabilitation training process according to the default settings. It is suitable for users who are familiar with system operation and do not need to select specific actions or perform equipment calibration. The "Select Action" function allows users to select different training actions according to their own needs, providing users with personalized choices. The "Exit System" is used to safely shut down the system after use.

[0074] Device Calibration: When the system initializes or the user re-wears the device, the system will force a calibration. The interface will display "Sensor Tracker 1.0" or "Sensor Tracker 2.0." The user will need to select the device type, click Next, and perform simple actions, such as standing or looking down at the tracker, to calibrate the device. Once calibration is complete, the system will return to the main menu. If any issues arise during calibration, click "Back" to restart the process or check the device connection.

[0075] Exercise Selection Screen: Click "Select Exercise" to enter the next screen, which presents a series of exercises, such as "Exercise 1" and "Exercise 2." Users can choose exercises based on their training plan, physical condition, and skill level. Once selected, click "Next" to begin training. To return to the previous screen and select again, click "Previous." Example

[0076] Operation process of Tai Chi exercise prescription rehabilitation training system based on virtual reality technology After selecting a movement and completing device calibration (if necessary), the system officially enters the operational phase. At this point, the demonstration module activates, displaying pre-set animations of the movements within the virtual environment. Simultaneously, the audio module plays corresponding voice prompts to help users understand and follow the movements. For example, when demonstrating a Tai Chi movement, the audio prompts, "Now begin the Lan Que Wei movement. First, turn your body to the left while slowly extending your arms..."

[0077] Users interact with the system through controllers and body movements. The motion capture module begins to work, capturing the user's head, hand, and leg movements in real time, including their position and posture. This motion data is quickly transmitted to the engine interface for processing. The avatar module then maps it to a 3D model of a virtual avatar, rendering the user's movements in real time within the virtual environment and enabling real-time interaction between the user and the virtual scene. During this process, the user can use the controller to control operations such as pausing, resuming, and replaying demonstration movements.

[0078] The present invention aims to explore the application potential of digital therapy in orthopedic diseases by constructing a digital rehabilitation all-in-one machine with precise evaluation, real-time feedback and personalized intervention functions, and loading traditional exercise rehabilitation prescription research combined with wearable sensors, motion capture systems and artificial intelligence algorithms to build integrated software including motion posture monitoring and motion quality assessment. The digital traditional exercise rehabilitation equipment realizes quantitative monitoring of rehabilitation training by establishing a dynamic evaluation model, and uses the VR user interaction interface to provide real-time motion correction guidance, and finally forms an intelligent training program that can adapt to different rehabilitation stages of KOA, realizing the optimization and intelligence of traditional treatment methods, and giving new era significance to traditional medicine. The system architecture is as follows Figure 10 shown.

[0079] System Modules: (1) Engine interface The virtual reality scene interaction system constructed in this invention uses the game engine interface as its core architectural component, responsible for the full-cycle management and control of virtual scene sessions. Standardized interface design enables organic collaboration between system components. This interface provides integration with the audio system, graphics rendering, file resource management, various development tools, and the Extended Reality Application Programming Interface (XRAPI), ensuring smooth operation and interaction of other modules within the engine. Among them, the multi-channel audio management module relies on the audio system provided by the engine interface to realize the loading, playback and control of audio resources such as environmental sound effects and background music, creating an immersive auditory experience for the virtual scene; the graphics rendering module uses the graphics rendering capabilities of the engine interface to provide three-dimensional model rendering, animation playback, light and shadow effects and other services for demonstration actions, virtual image mapping, etc., to build a realistic virtual training environment; the resource management module is connected to the file resource management system of the game engine, responsible for the loading, unloading and memory management of various resources (including models, audio, etc.), and provides an efficient resource access interface for the upper-level modules; XRAPI is connected to the mainstream extended reality device SDK to build a unified cross-platform XR service interface, responsible for the initialization, configuration and status management of various XR functions (including spatial positioning, tracking, rendering, etc.), to achieve an immersive interactive experience.

[0080] (2) Demonstration of movements The demonstration movements display preset animations through the engine's graphics rendering capabilities. The demonstration movements captured by motion capture technology are converted into a format that the engine can process, redirected to the character model and played when the system starts, so that the user can intuitively receive movement guidance when exercising. The demonstration movements are divided into multiple sections, each with a unique ID text. These ID texts communicate with the audio module to trigger corresponding audio prompts at the beginning of each movement. The human-computer interaction module can enhance the user's sense of participation and meet the individual needs of rehabilitation by controlling the playback progress of the demonstration movements. In addition, the data of the demonstration movement module is also monitored in real time by the motion capture module to ensure the accuracy and consistency of the movements.

[0081] (3) Motion capture and feedback module The motion capture module passes the captured data to the engine interface for real-time processing and feedback. The motion capture module supports multi-dimensional motion capture, including locating the user's position and movement through the head display, capturing hand movements through the image sensors in the controller or HMD, and capturing leg movements through sensor data strapped to the user's legs. The captured motion data is first passed to the engine interface for processing and then integrated into the system. The user avatar module maps the motion to the 3D model for real-time rendering. At the same time, the motion capture module works in conjunction with the user interaction module to ensure that every user's movement is promptly fed back to the interactive interface. In addition, the captured motion data is also monitored by the demonstration motion module to ensure the accuracy and synchronization of the demonstration motion, thereby improving the interactive experience of the entire system.

[0082] (4) Multi-channel audio management module The multi-channel audio management module relies on the engine-provided audio system to play ambient sounds and background music. While the system is running, soothing ambient sounds and background music are added, along with spatial audio effects. This not only increases the user's sense of immersion, but also helps regulate negative psychological states through music. The multi-channel audio management module works closely with the demonstration movements. At the beginning of each movement phase, a text description is broadcast through the multi-channel audio management module to help the user understand the current movement. This combination of text and audio not only enhances the user experience but also works in conjunction with the human-computer interaction module to guide user operations through voice prompts. Furthermore, the spatialization of audio effects relies on the engine-provided audio system to ensure accurate positioning and rendering of sound in 3D space.

[0083] (5) Real-time rendering module The real-time rendering module leverages the rendering capabilities of the engine interface to map the user's movements onto the 3D model, enabling real-time rendering and interaction. After collecting data from the user's head, hands, and legs, the real-time rendering module synthesizes this data into full-body motion data using a specific algorithm, applying multiple kinematic algorithms. This synthesized full-body motion data is mapped and redirected onto the 3D model within the engine, enabling real-time rendering and interaction within the virtual environment. The real-time rendering module is closely integrated with the motion capture module to ensure that every user movement is accurately reflected in the virtual avatar. Beyond displaying high-quality 3D models, the real-time rendering module also interacts with the audio module, triggering corresponding sound effects based on the user's movements. Furthermore, the real-time rendering module interacts with the human-computer interaction module, enabling instant feedback and interaction within the virtual environment, enhancing the overall sense of immersion and experience.

[0084] (6) Human-computer interaction The human-computer interaction module uses the engine to construct an interactive user interface in 3D space. Users can start and stop the entire system process using a controller. This module is tightly integrated with the motion capture module, capturing the user's full-body movements and feeding them into the graphical interface for real-time interaction. The human-computer interaction module also collaborates with the multi-channel audio management module to enhance the user experience through voice prompts and sound effects. Each control function in the interactive interface is implemented and managed using the API provided by the engine interface, ensuring smooth and responsive user operation. Furthermore, the human-computer interaction module collaborates with the real-time rendering module to reflect the user's movements and posture in the virtual environment in real time, enhancing immersion and interactivity.

[0085] Client development tools The system used in this paper uses Unity 2022.3.31f1, currently the most popular cross-platform game engine, offering powerful 3D graphics rendering capabilities and a wealth of development resources. Unity provides comprehensive virtual reality development support, including native integration with major VR headsets. Through Unity's XR Plugin Manager, developers can easily configure and manage interfaces for different VR platforms, achieving cross-platform compatibility.

[0086] Furthermore, Unity's input system supports a variety of VR controllers and interactive devices, ensuring a smooth and natural user experience within the virtual environment. Unity's built-in audio engine supports 3D spatial audio, accurately simulating the distribution and dynamics of sound within the virtual environment, enhancing immersion. Unity's UI Toolkit provides powerful and flexible user interface development tools, supporting the creation and management of both 2D and 3D interfaces. Using the Canvas system, developers can design responsive UI layouts that adapt to different resolutions and devices. The UI Toolkit is highly integrated with C#, facilitating the implementation of complex interaction logic and dynamic content updates. Unity's Mecanim animation system provides powerful animation control and state machine management, supporting complex character and object animations. Using Animator Layers and Animation Blending, developers can achieve smooth transitions and diverse animation expressions. Combined with OptiTrack's high-precision motion capture data, Mecanim can map user movements to virtual characters in real time, ensuring accurate and real-time animations. Furthermore, Unity supports Animation Events, making it easy to trigger specific logic operations during animations. Unity's asset management system supports efficient asset import, organization, and optimization. Unity also provides Unity Plastic SCM as a collaborative tool for project optimization, streamlining team collaboration.

[0087] The system in this invention uses the PICO 4 as a development device. The PICO 4 is a standalone VR headset with a 4320×2160 resolution, a 105° field of view, and is powered by the Snapdragon XR2 chip. PICO also provides developers with supporting development tools. The PICO Unity Integration SDK (hereinafter referred to as the PICO SDK) is a software development kit developed by PICO based on the Unity engine. The SDK encapsulates a range of functions covering rendering, input, tracking, mixed reality, platform services, and more. This SDK can be used to develop applications that run on the PICO VR standalone headset.

[0088] (1) User interaction The PICO SDK provides a comprehensive input management system that supports a variety of interactive devices, including controllers, gesture recognition, and eye tracking. Through the SDK, developers can easily implement natural and intuitive user interaction methods, such as gesture control, voice commands, and physical buttons. These interaction methods enhance user immersion and operational convenience within the virtual environment. Furthermore, the SDK supports multi-touch and complex gesture recognition, allowing users to complete complex tasks with simple movements, enhancing application usability and interactivity.

[0089] (2) Streaming test In the development of cross-platform VR systems, streaming testing is a very important function. It allows developers to quickly preview the effects without having to frequently deploy content to the target device. PICO provides a Live Preview Plugin that enables real-time transmission of content during the development process, allowing developers to view and test the latest modifications directly on the PICO VR device.

[0090] (3) Engine interface The PICO SDK is deeply integrated with the Unity engine, providing a rich API interface to streamline the development process. The SDK allows developers to easily access device hardware features such as position tracking, sensor data, and camera input. The engine interface supports real-time data transmission and processing, ensuring that dynamic elements in the virtual environment synchronize with changes in the real world. Furthermore, the SDK provides an optimized rendering pipeline and performance tuning tools to help developers achieve high-quality graphics while maintaining high frame rates and low latency.

[0091] (4) Performance optimization and resource management Performance optimization is crucial when developing applications for all-in-one VR headsets. The PICO SDK provides a series of performance analysis tools to help developers monitor and optimize application efficiency. Combined with Unity's resource management system, developers can efficiently manage and load resources, reducing memory usage and loading times. Furthermore, the SDK supports on-demand loading and resource compression technologies, ensuring smooth application execution on PICO 4 hardware, delivering a stable user experience.

[0092] (5) Cross-platform compatibility While this system is primarily developed for PICO 4, PICO's development tools also support a variety of other VR devices and platforms. Through a unified development interface, developers can seamlessly switch and adapt between different hardware, ensuring broad application compatibility. This cross-platform capability ensures future hardware upgrades and multi-device support.

[0093] (6) Integration with other modules The VR platform is tightly integrated with other system modules, such as audio, UI, animation management, and resource management, to create a high-quality user experience. Unity's audio system, combined with the PICO SDK's spatial audio support, enables realistic sound positioning. The UI Toolkit allows for the design of VR-adapted user interfaces, which integrate with PICO's input system for intuitive interaction. The Mecanim animation system, combined with PICO's motion tracking, enables natural animation of characters and objects. The resource management system ensures efficient sharing and access of various resources across modules, improving overall development efficiency and system performance.

[0094] Demonstration action capture and playback A high-precision motion capture scene was constructed using an OptiTrack PrimeX 13 motion capture camera to capture demonstration movements. The OptiTrack PrimeX 13 motion capture camera supports a native 240 frames per second (fps) rate and sub-millimeter accuracy, making it suitable for high-speed, precise tracking of medium-sized areas. It provides reliable data for motion recording, capturing even the finest details. After capture, the motion data is optimized and smoothed using Motive 3.0 and exported as the common skeletal animation FBX (.fbx) file format. This file is then imported into the Unity animation system and managed using Animator controllers for high-quality motion playback within the system.

[0095] Real-time capture and user interaction at runtime Using input from the head-mounted display, left and right controllers, and PICO motion trackers worn on each leg, the body tracking algorithm calculates the position and pose information of 24 basic bones throughout the body. The avatar utilizes the highly customizable Avatar solution provided by the PICO Avatar SDK. The PICO Avatar SDK is PICO's official avatar development kit, allowing the integration of official avatars into applications. It supports custom avatars, animation and expression control, real-time IK and rendering, and many other common features. By combining this body tracking algorithm with the skeletal retargeting capabilities of the PICO Unity Avatar SDK, the joint pose information from the body tracking algorithm can be directly mapped to user interactions on the avatar joints.

[0096] Development environment setup (1) Hardware Development equipment: The computer hardware configuration used in this invention is as follows: the processor is Intel Core i5-13600KF, the graphics card is NVIDIA GeForce RTX 4070, and the memory is 32GB DDR5.

[0097] VR equipment: Equipped with PICO 4 VR all-in-one machine, as well as supporting controllers and motion trackers for testing and debugging applications under development.

[0098] Motion capture equipment: Use the OptiTrack PrimeX 13 motion capture system, ensure it is properly connected to the development workstation, and perform basic calibration.

[0099] (2) Software Operating system: Windows 11 Standard Edition Engine: Unity 2022.3.31f1 XR development plugins: PICO Unity Integration SDK, PICO Avatar SDK Motion capture software: Motive 3.0 Other software: other necessary development tools, Visual Studio 2022, Blender, Unity PlasticSCM Module development steps (1) Demonstration action collection A world champion in Qigong, a National Games champion, and a national-level Qigong referee were invited to don motion capture suits and perform standardized demonstrations of Tai Chi based on the Delphi method. Twelve OptiTrack PrimeX 13 cameras were used to capture these demonstrations with high precision. Blackout curtains were used during recording, and all visible light sources were turned off to ensure the lighting and background conditions met the equipment requirements and reduce noise.

[0100] After recording the motion, optimize and smooth the recorded motion data in Motive software to correct any errors that occurred during the capture process. The processed motion data is exported as FBX animation format, imported into Unity's animation system, and managed and played back using the Animator controller.

[0101] (2) Action segmentation and voice prompts Divide the recorded continuous action into multiple sections, each corresponding to an independent action step and assigned a unique ID. For each action step, write a corresponding text description and record the corresponding voice prompt file. In Unity, write a script to synchronize the action playback with the voice prompt. Ensure that the corresponding voice prompt plays accurately at the beginning of each action.

[0102] (3) User action matching Use the PICO Avatar SDK to load avatars. At system startup, a specific avatar model is loaded based on its ID. This model features multiple levels of detail and can be dynamically adjusted when system overhead is high. The PICO Avatar SDK provides a standardized interface that receives motion data from motion capture devices in real time, converts and pre-processes it, and maps it to the avatar's skeletal system. Using animation controllers and IK (inverse kinematics) algorithms, natural transitions and accurate reproduction of movements are ensured.

[0103] like Figure 12 , integration and testing (1) Module integration Ensure that each module's interface is clearly defined and that data transmission formats are consistent to resolve inter-module compatibility issues. Integrate the demonstration action module, motion capture module, user interaction module, and audio module into the Unity project to ensure that all functional modules can work together. Use Unity's resource management system to efficiently share and access resources required by each module.

[0104] (2) System testing Perform independent functional testing on each module to ensure it works as expected. After system integration, conduct cross-module functional testing to verify the correctness of data transmission and interaction between modules. Use the performance analysis tools provided by the PICO SDK to monitor key performance indicators such as frame rate, latency, and resource usage, and optimize system performance.

[0105] (3) Repair and optimization We utilize version control systems and issue tracking tools to record and manage discovered bugs. Based on test feedback, we conduct multiple rounds of iterative development and optimization to fix bugs, optimize performance, and enhance the user experience. After each optimization, we conduct regression testing to ensure that resolved issues do not reappear in the new version and verify the overall stability of the system.

[0106] (4) Deployment and release Based on the characteristics of the PICO platform, we set the target platform to Android and the target device to PICO4. We then built the system into an APK file and deployed it to the device. We also developed a detailed user manual and operating guide to help users quickly get started with the system. We also established an ongoing maintenance and update mechanism to promptly fix any issues, introduce new features, and optimize performance to ensure the long-term stability of the system.

[0107] The research focuses on integrating digital technologies such as VR to create immersive rehabilitation environments, increasing the fun and compliance of training. Inertial sensors precisely capture patient motion data, providing an accurate basis for motion analysis. Artificial intelligence algorithms analyze sensor data to enable real-time motion monitoring and feedback. The developed equipment enables full-process monitoring of patient movement, real-time correction of movement standardization, and recommendations for advanced exercise progression. This provides technological support for the application of Traditional Chinese Medicine (TCM) exercise prescriptions in KOA treatment, enhancing their clinical effectiveness and promotional value.

[0108] 2. Relevant evidence of the technical effects obtained by the embodiments of the present invention.

[0109] KOA Tai Chi Exercise Prescription This invention provides a model for developing scientific and standardized Traditional Chinese Medicine exercise prescriptions through a Delphi-based Tai Chi treatment plan. Tai Chi movements, intensity, frequency, and duration are optimized to create personalized Traditional Chinese Medicine exercise prescriptions to enhance the effectiveness of Tai Chi in rehabilitation therapy. Potential risks during Tai Chi treatment are identified to optimize the plan. Expert consensus is reached through the Delphi method, providing a basis for standardizing Tai Chi treatment plans, enabling better application in clinical practice and improving the scientific and standardized nature of exercise therapy for KOA.

[0110] An expert working group was established, comprised of experts in Traditional Chinese Medicine (TCM) orthopedics, Tai Chi practitioners, methodologists, treatment protocol users, and graduate students in related fields. After a literature search and extensive research, the working group selected the 24-posture simplified Tai Chi (1956 edition, published by the General Administration of Sport of China), the most widely used form in KOA clinical research, as the initial list of items. Under the guidance of the expert steering committee, the first round of Delphi questionnaire development was conducted. The content included: 1. Introduction to the project background and research objectives; 2. Instructions and precautions for completing the questionnaire; 3. Basic information of the experts, including age, gender, highest degree, professional title, primary professional field, and years of experience; 4. Initial list of Tai Chi exercises for KOA treatment. This section combined the 9th and 11th single whip exercises. Using the 5-point Likert scale, each exercise was assigned a score for "recommendation" and "movement difficulty."

[0111] The significance and importance of the exercises for KOA prevention and treatment were assessed. "Recommendation" was categorized into five levels: strongly not recommended, not recommended, moderate or uncertain, recommended, and highly recommended, with scores assigned to them: 1, 2, 3, 4, and 5, respectively. "Difficulty of the exercises" was categorized into five levels: very difficult, difficult, moderate or uncertain, easy, and very easy, with scores assigned to them: 1, 2, 3, 4, and 5, respectively. ⑤ Expert familiarity with the questionnaire (Cs) was categorized into five categories: "unfamiliar," "not very familiar," "moderately familiar," "relatively familiar," and "very familiar," with scores assigned to them: 0.2, 0.4, 0.6, 0.8, and 1.0, respectively. ⑥ Expert judgment basis (Ca) was categorized into four categories: "practical experience," "theoretical analysis," "data acquaintance," and "personal intuition," with scores assigned to them: high, medium, and low, respectively. ⑦ Blank items were provided for experts to express their opinions. After completing the first round of questionnaire collection and data collection, a second round of questionnaires was designed. The second round of questionnaires covered the first round, with the initial Tai Chi treatment KOA exercise items refined based on expert feedback from the first round. Exclusions were explained and expert feedback solicited. Subsequent questionnaires followed the same structure as the second round until experts reached consensus on the items.

[0112] All questionnaire results were entered into Microsoft Excel 2021 by two researchers, and data consistency was checked. The data were statistically analyzed using SPSS 27.0. Descriptive statistical analysis was performed on the expert opinions, assessing the expert positivity coefficient, opinion concentration, coordination, and expert authority.

[0113] The expert engagement coefficient reflects the experts' interest in and enthusiasm for participating in the invention. It is expressed as the questionnaire recovery rate: Expert engagement coefficient = questionnaire recovery rate = (number of recovered questionnaires / number of distributed questionnaires) × 100%. A higher recovery rate indicates greater expert interest in the invention. It is generally considered that the valid questionnaire recovery rate should be no less than 60%. The degree of expert opinion convergence is reflected by the mean (x), the rank sum (S), the full score ratio (K), and the unimportance percentage (R). The mean (x) is the arithmetic mean of the scores for a particular item in the questionnaire, and the rank sum (S) is the sum of the scores for a particular item in the questionnaire: X = S / N = (X1 + X2 + … + XN) / N, where N is the number of experts who responded. A higher mean indicates a more important item and a higher degree of expert opinion convergence. The full score ratio (K) is the proportion of the highest score among all scores: K = n(X = 5) / N × 100%, where N is the number of experts who responded. A higher K indicates a more important item. The unimportance percentage (R) is the proportion of the lowest score (i.e., 1 point) in the total number of scores, that is, R=n(X=1) / N×100%, where N is the number of responding experts. The higher R is, the less necessary the item is.

[0114] The degree of consensus among experts' opinions was measured using the coefficient of variation (CV) and Kendall's coefficient of concordance (Kendall's W). The coefficient of variation (CV) reflects the degree of consensus among experts on a single item and is calculated as CV = (standard deviation / mean) × 100%. A smaller CV indicates higher consensus among experts on that item. The Kendall's W reflects the overall consensus among experts on all items and is calculated using a nonparametric test in SPSS software. Kendall's W measures consensus on a scale of 0 to 1, with a value of 0 indicating complete disagreement and 1 indicating complete consensus. Values ​​closer to 1 indicate greater consensus. A Kendall's W coefficient < 0.2 is generally considered to indicate poor consensus. Based on this, the Kendall's W coefficient was tested for significance, and consensus was considered statistically significant when P < 0.05.

[0115] The authority of expert opinions is expressed as the authority coefficient (Cr). This coefficient reflects the credibility and authority of an expert's judgment on an item. It is calculated as Cr = (Ca + Cs) / 2, where Ca represents the judgment basis coefficient and Cs represents the familiarity coefficient. Cr values ​​range from 0.1, with higher values ​​indicating greater credibility and authority. A value ≥ 0.7 is generally considered to indicate that the Delphi method results are reliable and acceptable for adoption.

[0116] All three rounds of Delphi questionnaires were distributed online using the online platform WJX. In the first round, 40 questionnaires were distributed, 38 of which were returned, with two experts not responding, for a recovery rate of 95%. In the second round, 38 questionnaires were distributed, 35 of which were returned, with three experts not responding, for a recovery rate of 92.10%. In the third round, 35 questionnaires were distributed, 33 of which were returned, with two experts not responding, for a recovery rate of 94.24%. Participating experts were from 11 provinces and municipalities: Tianjin, Shanghai, Xinjiang, Sichuan, Inner Mongolia, Zhejiang, Guizhou, Henan, Shandong, Jiangsu, and Yunnan.

[0117] The first round of Delphi method item recommendation score The authority of the experts in the first round of correspondence was Cr=0.87>0.7, indicating that the experts had a high level of authority and the results were highly reliable.

[0118] The second round of Delphi method item recommendation score The authority of the experts in the second round of correspondence was Cr=0.88>0.7, indicating that the experts had a high level of authority and the results were highly reliable.

[0119] The third round of Delphi method item recommendation score After two rounds of screening, the authority of the experts in the third round of correspondence was Cr=0.88>0.7, indicating that the experts were highly authoritative and the results were highly reliable.

[0120] The movements are arranged from easy to difficult according to their difficulty: starting position, closing position, flashing arms, reverse arm roll, cloud hands, wild horse's mane, single whip, hugging the knees and twisting the steps, left embracing the sparrow's tail, right embracing the sparrow's tail.

[0121] Combining the results of the Delphi questionnaire with the semi-structured interviews, the FITTVP principle was formulated according to the exercise prescription, and a traditional Chinese medicine exercise prescription was formed (taking the modified Tai Chi treatment program as an example). The plan is as follows: Scope of application: This program is suitable for the treatment of KOA patients in the remission and recovery stages. Patients should meet the following conditions: (1) No surgical indications (2) 18 ≤ ≤ 65 years old (1) Kellgren-Lawrence standard is rated as 0~Ⅱ Exercise frequency: no less than 3 days / week Exercise intensity: Subjective exertion scale 3-6 points, conversation test, can communicate normally while exercising (with or without accompaniment), can sing Exercise time: no less than 30 minutes / time (excluding warm-up and cool-down) Exercise method: Modified Tai Chi KOA treatment program (specific movements include starting and ending postures, flashing arms, reverse arm roll, cloud hands, wild horse mane, single whip, hugging the knees and twisting the steps, left grazing the peacock tail, right grazing the peacock tail) Total amount of exercise: no less than 150-300 minutes per week (according to the principle of gradual improvement) Exercise advancement: Patients should first adjust the frequency and time of exercise according to their own tolerance. After meeting the requirements, they should practice according to the stage movements. They should advance step by step until they feel slight sweat all over the body and muscle soreness but the pain can be relieved on the second day of exercise.

[0122] Phase 1: Starting position, reverse arm roll, cloud hand, flash arm, closing position Phase 2: Starting stance, Wild Horse Spreading its Mane, Reverse Arm Roll, Single Whip, Cloud Hands, Flashing Arms, Closing stance The third stage: starting posture, wild horse splitting its mane, hugging the knees and twisting the steps, reverse arm roll, left embracing the peacock tail, right embracing the peacock tail, single whip, cloud hands, flashing arms, closing posture On the basis of full mastery and fluency of movements, transition of Tai Chi postures is performed according to the progress of the movement.

[0123] Adaptation stage: high stance (knee bending angle of about 10°30°, lunge distance of about 11.5 times shoulder width) + first / second stage movements, coordinate with natural breathing, stretch the movements. Improvement stage: medium stance (knee bending angle of about 30°60°, lunge distance of about 1.52 times shoulder width) + third stage movements, breathe into the dantian and coordinate with the rhythm of the movements, the movements are smooth and coherent.

[0124] Maintaining the posture: Elevated + third stage movements, breathing deep into Dantian and coordinating with the rhythm of the movements, with smooth and coherent movements. Implementation Notes: (1) Conduct a comprehensive health assessment before use to identify the risk of cardiovascular and cerebrovascular events (2) Warm up and relax thoroughly before and after use to prevent sports injuries (3) When using for the first time, please do so under the guidance of a professional. Clinical effectiveness verification of KOA Tai Chi exercise prescription A single-center, prospective, observational comparative efficacy study (cohort study) was conducted. A total of 84 subjects diagnosed with KOA and not suitable for surgery who were admitted to the Department of Orthopedics at the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine between January and December 2024 were enrolled. Forty-two subjects were included in the observation group and 42 in the control group. The diagnostic criteria were based on the "Guidelines for the Diagnosis and Treatment of Osteoarthritis in China (2024 edition)" using the Kellgren-Lawrence X-ray grading system and the KOA clinical staging criteria in the "Guidelines for the Diagnosis and Treatment of KOA in Traditional Chinese Medicine (2020 edition)" developed by the Orthopedics Branch of the China Association for the Promotion of Traditional Chinese Medicine Research. Sample size was calculated using G*Power 3.1.9.7. This calculation was based on changes in knee pain related to the WOMAC index in previous literature studies. After 12 weeks of exercise intervention, the observation group scored 17.61±2.22 and the control group scored 20.58±3.95. With α=0.20 and 1β=0.80 specified, the software calculated that 37 cases needed to be included in each group, with no more than 10% of the samples dropping out. The sample size was expanded to 42 cases per group, for a total of 84 cases.

[0125] Baseline data analysis Group Observation Group control group Statistics P-value Gender n (%) 0.05 0.82 male 16(38.10) 17(40.48) female 26(61.90) 25(59.52) Age (y) 53.29±8.03 54.95±6.85 -1.02 0.31 Height (m) 1.64±0.05 1.65±0.05 -0.65 0.52 Weight(kg) 68.98±5.17 71.19±7.77 -1.54 0.13 BMI (kg / m²) 25.68±1.40 26.23±1.86 -1.53 0.13 K-L classification n(%) 0.48 0.79 Level 0 6(14.29) 4(9.52) Level I 13(30.95) 13(30.95) Level II 23(54.76) 25(59.52) This invention has been approved by the Ethics Committee of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine with the ethics review number TYLL2024[Z]002, and registered with the International Traditional Medicine Clinical Trial Registry with the registration number ITMCTR2024000201.

[0126] Inclusion criteria: (1) KOA subjects in the remission and recovery stages who meet the above-mentioned KOA diagnostic criteria and have no surgical indications; (2) 18 ≤ ≤ 65 years old; (3) Kellgren-Lawrence standard assessment level 0 to II; (4) Obtain informed consent and sign the informed consent form.

[0127] Exclusion criteria: (1) Those who have received other methods of KOA treatment within 2 weeks; (2) Those who have recently (in the past six months) regularly (at least three times a week, at least 30 minutes each time) performed traditional sports such as Tai Chi; (3) Those suffering from central nervous system diseases or limb movement disorders caused by other diseases; (4) Patients with serious diseases of the heart, brain, liver, kidney and hematopoietic system; (5) Subjects who are in the perioperative period, have been drinking alcohol for a long time, or are taking anticoagulants (such as warfarin), or have a suspected or confirmed history of drug abuse; (6) Cognitive impairment caused by mental illness or neurological disease; (7) Those who are unanimously determined by the physicians of the research team to be unsuitable for inclusion due to other reasons.

[0128] Shedding standard: (1) Subjects with extremely poor compliance, who perform required exercise less than or equal to once a week; (2) Subjects were lost to follow-up due to various reasons; (3) The subject is unwilling to continue the clinical trial and requests to withdraw from the trial to the attending physician.

[0129] If the subject meets any of the above conditions, he / she will be considered as dropout case.

[0130] Rejection criteria: (1) Those who do not meet the inclusion criteria and / or meet the exclusion criteria during the study; (2) Those who have never received treatment or have no records.

[0131] If the subject meets any of the above items, he / she will be excluded from the present invention.

[0132] Termination criteria: (1) If a serious adverse event occurs (threatening life or affecting normal work and life), the patient shall terminate the treatment if it is impossible to continue the treatment; (2) Symptoms completely disappear during the course of treatment and treatment is stopped.

[0133] Based on whether the patient consciously accepts KOA Tai Chi exercise prescription as the exposure factor, the specific exposure definition is: Observation Group (KOA Tai Chi Exercise Prescription Group): At initial enrollment, participants were recommended the Traditional Chinese Medicine (TCM) exercise prescription (modified Tai Chi treatment plan) developed in Study 2 based on their baseline exercise capacity and medical needs. Participants performed self-training exercises at home, following the doctor's recommended movements and referring to standard exercise videos. The videos demonstrated key exercises and precautions. Participants adjusted their movements at various follow-up time points. According to the exercise prescription, each session lasted at least 30 minutes (excluding warm-up and cool-down), and was repeated at least three times per week, with a cumulative weekly exercise volume of at least 150 minutes for four weeks.

[0134] Control group (traditional Tai Chi group): The subjects performed the Yang style 24-posture simplified Tai Chi training at home, with each training session lasting at least 30 minutes, repeated at least 3 times a week for 4 weeks.

[0135] All the subjects in the above groups received health education, emphasizing the importance of changing their lifestyle and work style, so that they could set correct treatment goals, reduce pain, and improve and maintain joint function. Based on their daily activities, they were advised to change their bad living and working habits, such as avoiding long periods of running, jumping, and squatting, and reducing or avoiding climbing stairs and mountains. They were also advised to control their weight, especially those with a BMI > 28 kg / m 2 For KOA subjects, it is recommended that they use mobility assistance support therapy in daily life. If necessary, appropriate mobility assistance devices such as canes, crutches, walkers, joint braces, etc. can be selected under the guidance of a doctor. Flat, thick, soft, and loose shoes can also be selected to assist walking.

[0136] Participants logged in daily for exercise, recording the number of exercise sessions, duration of each session, and their feelings about the exercise. Trained personnel from our institution regularly observed and evaluated the participants' conditions, treatment plans, and examination results before treatment and one and four weeks after treatment through completing medical records, regular outpatient follow-up visits, and telephone inquiries.

[0137] Knee symptoms were assessed using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the Visual Analog Scale (VAS) before treatment and at week 1 and week 4 after treatment. Exercise compliance was assessed using the Numeric Rating Scale (NRS). Gait was also assessed using a dynamic gait posture analysis system (RightGaitPostureMedical 3.0, Shenzhen Xingzheng Technology Co., Ltd.) before treatment and at week 1 and week 4 after treatment.

[0138] Record any adverse events that occur during the study, such as muscle or joint pain or injury, and other adverse medical events. Conduct targeted laboratory and imaging examinations for adverse events, if necessary.

[0139] Data Analysis: Data entry and processing were performed using Excel, and statistical analysis was performed using SPSS 27.0 software. Descriptive statistics were used to calculate the percentage of scores for each question, and the results were presented in pie charts and bar graphs. Enumeration data were presented as frequency (n) and percentage (%). Frequency differences of categorical variables between groups were compared using the chi-square test or Fisher's exact test. Continuous data were verified for normality using the Shapiro-Wilk test. Normally distributed data were presented as mean ± standard deviation (M ± SD), and inter-group comparisons were performed using the independent sample t-test. Non-normally distributed data were presented as median (first quartile, third quartile), i.e., M (Q1Q3), and inter-group comparisons were performed using the Mann-Whitney U test. Correlation analysis was performed using the Spearman rank test. If significant differences were found between groups, subsequent analyses were performed using analysis of covariance (ANCOVA) to adjust for confounding factors. Missing data were supplemented using intention-to-treat analysis. A 95% confidence interval (CI) was set, and a P < 0.05 was considered statistically significant. Clinical effectiveness verification process such as Figure 4 shown.

[0140] Evaluation results: Comparison of WOMAC total scores between the two groups (`x±s) Note: Intra-group comparison, compared with before treatment* P <0.05** P <0.001 Before treatment, the WOMAC total scores of the two groups were similar, indicating no significant difference in overall symptoms before treatment (P>0.05), indicating comparability between the two groups. One week after treatment (P<0.05) and four weeks after treatment (P<0.001), the control group's WOMAC total scores decreased slightly compared to pre-treatment. The observation group's WOMAC total scores decreased significantly both one and four weeks after treatment (P<0.001), indicating that the treatment regimen significantly improved overall symptoms in both groups, with the effect being more pronounced four weeks after treatment. The observation group showed a statistically significant improvement in WOMAC total scores compared to the control group one week after treatment (P>0.05). Four weeks after treatment, the observation group's total score decreased compared to the control group, but the difference was not significant (P>0.05).

[0141] Comparison of VAS scores between the two groups (mm, `x±s) Note: Intra-group comparison, compared with before treatment** P <0.001 VAS scores were compared between the two groups before treatment, one week after treatment, and four weeks after treatment. Pre-treatment VAS scores were similar between the two groups, indicating no significant difference in pain intensity before treatment (P>0.05), indicating comparability between the two groups. VAS scores in the observation group were significantly lower one and four weeks after treatment compared with pre-treatment scores (P<0.001), indicating that the treatment regimen in the observation group was significantly effective in alleviating pain. While VAS scores in the control group decreased one week after treatment, the difference was not significant compared with pre-treatment scores (P>0.05). Four weeks after treatment, there was no significant difference in VAS scores between the two groups, but scores in the observation group were still lower than those in the control group.

[0142] Comparison of NRS scores of exercise compliance between the two groups (`x±s) Note: Intra-group comparison, compared with before treatment* P <0.05** P <0.001 Before treatment, the NRS scores for exercise compliance were similar between the two groups (P>0.05), indicating no significant difference in expected exercise compliance between the two groups before treatment and comparability. The NRS scores for exercise compliance in the observation group were significantly lower than before treatment, both 1 and 4 weeks after treatment (P<0.001), with the decrease being more pronounced 4 weeks after treatment. This indicates that exercise compliance generally fell short of participants' expectations and decreased with prolonged treatment. Although compliance showed a downward trend in both groups, compliance in the observation group was higher than in the control group 1 and 4 weeks after treatment, with statistically significant differences between the two groups (P<0.05).

[0143] Comparison of gait analysis parameters between the two groups (`x±s) Note: Intra-group comparison, compared with before treatment* P <0.05** P <0.001 Before treatment, there were no significant differences in any gait parameters between the two groups (P>0.05), indicating comparability. Gait speed improved in both the observation and control groups. This improvement was not significant one week after treatment, but was significant four weeks later (P<0.001). Cadence increased in both the observation and control groups one week after treatment, but the increase was not statistically significant (P>0.05). Four weeks after treatment, cadence in the observation group was significantly higher compared to pre-treatment (P<0.001). Stride length increased in both the observation and control groups one week after treatment (P<0.05), and further increased four weeks after treatment (P<0.001). Single-leg support stability gradually improved in both groups. Compared to pre-treatment, stability in the control group increased significantly one week after treatment (P<0.05), and the difference between the two groups was significant four weeks after treatment (P<0.001). In terms of ground clearance, both groups showed a gradual increase at 1 and 4 weeks after treatment, with statistically significant differences compared to pre-treatment levels (P < 0.001). In terms of foot deviation angle, both groups showed a gradual decrease, but the difference was not significant (P > 0.05). At 1 and 4 weeks after treatment, no statistically significant differences were found between the two groups in changes in gait speed, cadence, stride length, single-leg support stability, ground clearance, and foot deviation angle (P > 0.05).

[0144] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0145] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A Tai Chi exercise prescription rehabilitation training system based on virtual reality technology, characterized in that: The settings are: Virtual reality presentation device, adapted for immersive 3D scenes built with Unity or an equivalent 3D engine; used to display demonstration action animations, the user's movements and postures in the virtual environment, and the audio environment; The motion capture and feedback module integrates an inertial measurement unit, a spatial positioning device, and a posture perception node; it is used to capture the user's head, hand, and leg data in real time, and transmit the data collected by each node to the central processing unit through a wireless synchronous transmission interface for posture fusion; The central processing unit is equipped with an instruction scheduling module, a data integration module, and a rendering control module; A training prescription generation module, which has a joint state input interface and a training parameter output interface; Used to formulate personalized exercise training prescriptions based on the user's status, automatically adjust exercise training prescriptions, provide personalized training suggestions and real-time corrections; A real-time rendering module is used to drive 3D models, action animations, and interactive interfaces. It provides 3D model rendering, animation playback, lighting effects, and other services for demonstration actions and virtual image mapping, thereby building a realistic virtual training environment. Multi-channel audio management module, including resource loading interface and spatialized audio output interface; used for loading, playing and controlling audio resources, as well as broadcasting text or guiding user operations through voice prompts; The human-computer interaction module includes a control input channel, a virtual image mapping engine, and a status monitoring unit.

2. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: Also provided are: The intelligent interaction and remote monitoring module, through the integration of biofeedback equipment, can monitor the patient's physiological condition in real time and provide real-time feedback on the VR training process; it has remote data synchronization and real-time monitoring functions; it has the ability of self-learning and evolution, and can continuously optimize the training plan based on the patient's training data and feedback.

3. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The virtual reality presentation device includes a head-mounted display, has a visual image interface, an inertial positioning unit and an audio playback channel, and supports the XR API protocol or an equivalent extended reality standard.

4. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The motion capture component in the motion capture and feedback module includes: IMU sensor node for head positioning; Hand controller for upper limb movement detection; Leg-strap inertial module for lower limb motion acquisition.

5. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The motion capture and feedback module is also equipped with a three-dimensional joint mechanical model and dynamic feedback of virtual joint status. Through the high integration of inertial measurement unit, spatial positioning device and posture perception node, the system can build an accurate joint motion model in three-dimensional space and monitor and feedback the biomechanical status of the joint in real time, such as joint force, range of motion, gait and other parameters; In addition, the system uses AI algorithms to conduct in-depth analysis of motion data to determine whether the exercise meets the individual's treatment needs, and issues prompts in real time to ensure that each movement is performed on the basis of scientific biomechanics, avoiding excessive or insufficient exercise and reducing rehabilitation risks.

6. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The training prescription generation module includes: Functional scoring input unit, used to import knee joint scoring data; Pain indicator access interface, used to input external physiological state parameters; Parameter matching operation unit, used to select training action combinations based on the rule base; The exercise intensity configuration unit is used to set the frequency, duration and amplitude information of the output prescription.

7. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The real-time rendering module includes: Action animation manager, used to call standard action sequences; 3D model binding engine for implementing skeletal animation mapping; A scene construction unit, used to load the training scene resource package and output it to the display device; Rendering task scheduling interface, connecting to the central processing unit for frame-level control.

8. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The multi-channel audio management module specifically includes: Audio content resource library; Audio playback scheduling module; Action phase trigger control interface; Spatial sound synthesizer, configured with three-dimensional sound field parameters; The audio module and the action module are linked together through the action identification ID.

9. The virtual reality technology-based Tai Chi exercise prescription rehabilitation training system according to claim 1, characterized in that: The intelligent interaction and remote monitoring module includes: Personalized voice guidance unit, based on deep learning algorithms, dynamically adjusts voice prompt content and volume according to the user's training status, physiological parameters and behavioral feedback; A virtual voice interaction control unit, used to recognize user voice commands and drive the virtual image to respond; The bio-signal feedback module includes a heart rate sensor, an electromyography acquisition unit, and a skin electrical sensor, which is used to collect the user's physiological status and control it in conjunction with the training content; Control input recognition unit and virtual skeleton mapping module, used to realize the mapping of user actions to three-dimensional virtual characters; State feedback renderer and interface module for real-time display of user gestures and interactive responses; Data synchronization and remote monitoring unit, used to upload training data to the cloud in real time and support remote analysis and prescription adjustment; The self-learning module uses training data for machine learning optimization to automatically adjust training content, intensity, and mode, supporting the dynamic evolution of individual training strategies and system-level treatment strategies.

10. A Tai Chi exercise prescription rehabilitation training method based on virtual reality technology according to any one of claims 1 to 9, characterized in that: The following steps are involved: S101, using a virtual reality presentation device to load multiple immersive virtual training environments constructed using a Unity engine or an equivalent three-dimensional graphics engine; S102, using a motion capture component to collect multi-dimensional motion data of the user's head, hands, and legs; S103, transmitting the collected motion data to a data fusion and conversion unit through a predetermined data interface; S104, structural integration and formatting of the received data, and conversion into the data structure required by the training prescription generation module; S105: Generate a personalized training prescription based on the user's motion data, and output it to the virtual reality display interface through a feedback module.

Images