Virtual reality physical rehabilitation system

By combining virtual domain systems, multi-sensory stimulation, and embodied assessment systems, multi-domain collaborative linkage of virtual reality embodied rehabilitation systems is achieved, solving the problem of lack of embodiedness in existing virtual reality rehabilitation systems and improving the effectiveness and personalization of rehabilitation training.

CN121731624AActive Publication Date: 2026-03-27BEIHANG UNIV

Patent Information

Application Number
CN202610238370.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-02-28
Publication Date
2026-03-27
Estimated Expiration
2046-02-28

AI Technical Summary

Technical Problem

Existing virtual reality rehabilitation systems lack precise shaping and control of embodiedness, failing to achieve deep integration between the virtual avatar and the real body. This results in limited improvement in rehabilitation training effectiveness and a lack of a comprehensive embodiedness assessment mechanism, making it difficult to achieve personalized and dynamic rehabilitation training.

Method used

By generating a virtual avatar that matches the patient's physical characteristics through a virtual domain system, and combining it with a multi-sensory stimulation and embodied assessment system, the difficulty of rehabilitation tasks and sensory stimulation can be adjusted in real time. By utilizing the auxiliary movement equipment of the device domain system, a closed-loop control system can be formed to achieve multi-domain collaborative linkage.

Benefits of technology

It improves the effectiveness and personalization of rehabilitation training, activates patients' motor neural pathways, promotes neural plasticity and functional recovery, and solves the problem of mismatch between training programs and patients' conditions in traditional rehabilitation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121731624A_ABST
    Figure CN121731624A_ABST
Patent Text Reader

Abstract

The invention provides a virtual reality body rehabilitation system. The virtual reality body rehabilitation system comprises a virtual domain system, a body feeling stimulation system, a body feeling evaluation system, an equipment domain system and a central control module. The virtual domain system specifically induces the sense of ownership, the sense of autonomy and the sense of position by creating a virtual environment of a virtual avatar, real-time action mapping and adaptation rehabilitation tasks; the body feeling stimulation system takes visual stimulation as a core and is matched with linkage of various kinds of sensory stimulation, so that the authenticity of body experience is enhanced; the body feeling evaluation system evaluates real-time body feeling and provides reliable data support for regulation and control. On the basis of the physical cognition theory, the virtual reality technology, the multi-mode sensing technology, the neuroscience and the rehabilitation medicine principle are fused, multi-channel feeling collaborative stimulation such as vision, body feeling and touch feeling is achieved by constructing the immersive virtual environment, a patient is induced to generate physical experience, abnormal body representation is remodeled, nerve plasticity and function recovery are promoted, and the rehabilitation effect of the patient is improved. The patient is helped to return to normal life and society.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of virtual reality rehabilitation equipment, and particularly relates to a virtual reality embodiment rehabilitation system. BACKGROUND

[0002] Embodiment refers to the dependence of thinking, emotion and consciousness on body structure, physiological state and interaction with the world, and the concept mainly includes body schema and body image. The concept of embodiment originates from the theory of embodied cognition, and the core is to subvert and break through the traditional concept. The brain is not considered as the only thinking tool, and the body is not considered as the only execution tool. Instead, the understanding of the world is considered to be shaped through the interaction of the body and the environment. Body schema is a body parameter useful for action planning and control through the sensorimotor matrix, such as planning the body movement path when trying to complete some action task or trying to avoid some dangerous harm. The Bayesian model believes that the body schema is not a fixed and static map, but a probabilistic prediction system that is dynamically updated based on a long-term body model and current multi-sensory input information. The brain will allocate weights according to the reliability of different sensory information according to the Bayesian rule. This mechanism is the core theoretical basis of the virtual reality embodiment and rehabilitation system. Body image is the image of our own body formed in our mind, which is a dynamic and changeable cognition, and in some cases is not completely separated from body schema. The virtual reality system can allow patients to view their own body from different perspectives and perform movements, which provides a powerful and clear body image to guide the movement. This visual input can be an effective tool for the consciousness to correct the body schema.

[0003] Based on the above background information, embodiment rehabilitation is the concept of embodied cognition applied to the field of rehabilitation. The core goal is not simply to train movements or muscle strength, but to repair the brain's cognitive bias of the body through body perception, movement execution and environmental interaction, and to rebuild the embodiment relationship between body-brain-environment. Traditional rehabilitation therapy focuses on mechanical repetitive training of limb function such as muscle strength or joint range of motion.

[0004] With the application of virtual reality (VR) technology in the field of rehabilitation, virtual reality rehabilitation systems gradually become an important supplement to traditional rehabilitation methods. They can provide patients with immersive training experiences by constructing virtual training environments, and to some extent, improve the interest of rehabilitation training. However, existing virtual reality rehabilitation systems mostly focus on the construction of visual immersion, and only guide patients to complete actions through virtual pictures, lacking precise shaping and regulation of embodiment. Patients are difficult to establish a sense of ownership, autonomy and location of the virtual avatar in the training process, and cannot achieve deep integration of the virtual avatar and the real body, so that rehabilitation training still stays at the level of "passive following", and it is difficult to effectively activate the motor nerve pathway of patients, and the rehabilitation effect is limited.

[0005] Meanwhile, existing virtual reality rehabilitation systems lack a perfect embodiment evaluation mechanism, and cannot obtain the embodiment state of patients in rehabilitation training in real time and accurately. They cannot dynamically adjust the rehabilitation task difficulty, virtual environment parameters and sensory stimulation mode based on the embodiment feedback of patients, and cannot optimize the assistance strategy of the auxiliary motion device according to the embodiment changes, resulting in that the training scheme does not match the real-time rehabilitation state of patients, and it is easy to cause the problem of too high training difficulty to cause the frustration of patients, or too low training difficulty to fail to achieve effective rehabilitation training intensity.

[0006] In addition, the existing rehabilitation system lacks cooperation and linkage among the functional modules. The scene construction in the virtual domain, the application of sensory stimulation, the regulation of the auxiliary motion device and the embodiment evaluation results are independent of each other, and it is difficult to form a closed-loop control system of "evaluation-feedback-adjustment-training". It is difficult to achieve personalized and dynamic rehabilitation training based on the embodiment of patients, and it is also difficult to strengthen the motor intention of patients through multi-sensory fusion embodiment stimulation, and then affect the efficiency of motor function reconstruction. SUMMARY

[0007] In view of this, the embodiment of the present application provides a virtual reality embodiment rehabilitation system to eliminate or improve one or more defects in the prior art.

[0008] A virtual reality embodiment rehabilitation system, comprising: a virtual domain system, configured to generate or call a virtual avatar according to a rehabilitation task, wherein parameters of the virtual avatar match characteristics of a real body of a patient, so as to enable the patient to have a sense of ownership of the virtual avatar and / or the real body; the virtual avatar is also configured to be able to map the body movement of the patient in real time, so as to enable the patient to have a sense of autonomy of the virtual avatar and / or the real body; and the virtual domain system is further configured to generate a virtual environment matching the rehabilitation task, and the virtual avatar is placed in the virtual environment, so as to enable the patient to have a sense of location of the virtual avatar and / or the real body; A body ownership stimulation system, by applying at least visual stimulation containing a virtual avatar to a patient, and at least one of tactile, auditory, olfactory and vestibular stimulation, so that the patient generates body ownership during rehabilitation, which includes at least one of the ownership, autonomy and location; A body ownership evaluation system, by obtaining real-time body ownership through inquiring a patient during performing a rehabilitation action, or by collecting physiological indicators of the patient, which includes at least one of heart rate, electromyogram, electroencephalogram, skin electricity, eye movement, micro-expression and stress, or by collecting kinematic data of the patient; A device domain system, including a plurality of assistive motion devices, for assisting the patient to implement the motion intention, or assisting the body ownership stimulation system to apply the sensory stimulation, or assisting the body ownership evaluation system to evaluate, to form an intermediary of the patient's embodied interaction with the environment; A central control module, for adjusting at least the difficulty of the rehabilitation task based on the real-time body ownership, or adjusting the type or intensity of the sensory stimulation applied by the body ownership stimulation system, or adjusting the controllable parameters of the assistive motion devices of the device domain system, or adjusting the mapping action of the virtual avatar.

[0009] The virtual reality embodied rehabilitation system in the application is based on the body ownership cognitive theory, and combines virtual reality technology, multi-modal sensing technology, neuroscience and rehabilitation medicine principles, to realize multi-channel sensory coordination stimulation such as vision, proprioception and touch, induce the patient to generate body experience, reshape abnormal body representation, promote neural plasticity and functional recovery, and help the patient return to normal life and society.

[0010] The application takes the construction, stimulation, evaluation and dynamic regulation of body ownership as the core logic, integrates five modules of virtual domain, body ownership stimulation, body ownership evaluation, device domain and central control, and forms a functional coordination and closed-loop efficient rehabilitation system. The virtual domain system matches the virtual avatar with the real body characteristics of the patient, realizes real-time action mapping and adapts to the virtual environment of the rehabilitation task, induces the ownership, autonomy and location, and lays the foundation for embodied rehabilitation; the body ownership stimulation system takes visual stimulation as the core, and cooperates with multi-type sensory stimulation linkage to strengthen the reality of body experience; the body ownership evaluation system realizes objective and accurate evaluation of real-time body ownership through multiple paths such as inquiry, physiological indicator collection and kinematic data collection, and provides reliable data support for regulation.

[0011] The auxiliary motion device of the device domain system builds a somatization interaction medium between the patient and the environment, effectively assists the landing of the motion intention, the application of the sensory stimulation and the promotion of the evaluation process, and makes up for the virtual-real interaction gap; the central control module flexibly adjusts the rehabilitation task, the sensory stimulation, the auxiliary device and the virtual avatar related parameters based on the real-time somatization evaluation result, realizes the complete closed loop of "evaluation-feedback-regulation".

[0012] Additional advantages, objects, and features of the application will be set forth in part in the description which follows, and will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the application. The objects and other advantages of the application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0013] It will be understood by those within the art that the objects and advantages of the application can be met by other specific arrangements, and that particular arrangements of the application described herein are shown by way of example and not limitation. For example, although the application has been described in terms of particular embodiments and illustrative figures, those of ordinary skill in the art will be able to determine numerous other alternative arrangements and equivalent schemes in accordance with the principles of the application, without departing from the scope of the application as set forth in the claims. BRIEF DESCRIPTION OF DRAWINGS

[0014] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description, explain the principles of the application. The drawings are intended for purposes of illustration only and the dimensions or other specific features within the drawings are not intended to be to scale. In order to simplify the drawings and aid in their understanding, corresponding parts are sometimes designated with common reference numerals and description thereof can be presented with reference to one or more of the drawings.

[0015] Figure 1 Structure schematic diagram of a virtual reality somatization rehabilitation system in an embodiment of the application.

[0016] Figure 2 Structure schematic diagram of a virtual reality somatization rehabilitation system in an embodiment of the application.

[0017] Figure 3 Structure schematic diagram of a virtual reality somatization rehabilitation system in an embodiment of the application.

[0018] REFERENCE NUMERALS: 10, rehabilitation cabin; 11, visual module; 12, auditory module; 13, olfactory module; 14, temperature adjustment module; 15, humidity adjustment module; 16, wind speed adjustment module; 17, light adjustment module; 18, gas partial pressure adjustment module; 21, motion posture monitoring module; 22, heart rate monitoring module; 23, electromyography monitoring module; 24, electroencephalography monitoring module; 25, skin electricity monitoring module; 26, eye movement monitoring module; 27, micro-expression recognition module; 28, pressure and gravity center monitoring module; 31, multi-degree-of-freedom balance platform; 32, vestibular electric stimulation device; 33, exoskeleton robot; 34, walking aid; 35, body weight support device; 36, treadmill; 41, virtual environment database; 42, virtual avatar generation module; 43, action mapping unit. DETAILED DESCRIPTION

[0019] In order to make the objects, technical solutions and advantages of the present application clearer, further detailed description will be given to the present application in combination with embodiments and drawings. Herein, the illustrative embodiments of the present application and the description thereof are used to explain the present application, but not as a limitation to the present application.

[0020] Herein, it is also necessary to point out that, in order to avoid the present application being obscured by unnecessary details, only the structures and / or processing steps closely related to the solutions according to the present application are shown in the drawings, and other details not closely related to the present application are omitted.

[0021] It should be emphasized that the term "comprise / comprising" is used herein to specify the presence of a feature, element, step or component, but not to exclude the presence or addition of one or more other features, elements, steps or components.

[0022] Herein, it is also necessary to point out that, if not specially stated, the term "connection" herein can not only mean direct connection, but also indirect connection with an intermediate.

[0023] In the following, embodiments of the present application will be described with reference to the drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0024] The virtual reality embodied rehabilitation system can help patients recover through the process of dynamic remodeling of body schema and body image with multi-sensory stimulation signals of various environmental factors, take the brain-body-environment as a whole, and create a safe and controllable immersive environment, which can also minimize the negative emotions caused by pain or failure. For example: when treating hemispatial neglect, a continuous visual bias can be introduced in the virtual reality scene to make the patient's action target deviate to the neglected side.

[0025] The present application is based on embodied cognition theory, and integrates virtual reality technology and multi-modal sensing technology to construct an embodied rehabilitation system capable of multi-domain collaborative linkage, precise perception and dynamic adjustment, which can induce patients to have embodied experience, remodel abnormal body representation, promote neural plasticity and functional recovery, and improve the effectiveness and individualization level of rehabilitation training.

[0026] Compared with the traditional rehabilitation technology and the existing virtual reality rehabilitation system, and clearly distinguished from the embodied intelligence concept, the embodiment of the application is precisely regulated through module cooperation and embodied feeling, realizes multi-dimensional technical breakthrough, and significantly improves the scientificity, effectiveness and compliance of rehabilitation training. The core of the application focuses on strengthening the embodied feeling of the patient himself, constructs the sense of possession, autonomy and position of the patient to his own body and virtual carrier through virtual avatar matching, multi-sensory stimulation and other means, activates neural remodeling and strengthens motor control as the goal, and finally improves the rehabilitation effect.

[0027] As shown in Figures 1-3 The application provides a virtual reality embodied rehabilitation system (hereinafter referred to as the system), which is based on the embodied cognition theory, combines virtual reality technology, multi-modal sensing technology, neuroscience and rehabilitation medicine principles, realizes multi-channel sensory coordination stimulation such as vision, proprioception and touch through the construction of an immersive virtual environment, induces patients to have physical experience, remodels abnormal body representation, promotes neural plasticity and functional recovery, and helps patients return to normal life and society. The system can be based on a rehabilitation cabin 10 and various hardware devices arranged in the rehabilitation cabin 10. The rehabilitation cabin 10 can adopt a fully closed, fully light-shielded and fully soundproof design to avoid external environmental interference. The rehabilitation cabin 10 can adopt any achievable structure, including but not limited to cabin body structure, secret room structure, etc.

[0028] In this embodiment, the system can include five core parts: a virtual domain system, an embodied feeling stimulation system, an embodied feeling evaluation system, a device domain system and a central control module, which closely cooperate to jointly build a complete rehabilitation training system.

[0029] The virtual domain system serves as a window for the patient to interact with the virtual rehabilitation world. After receiving a rehabilitation task instruction, it calls or generates a virtual environment corresponding to the task type and the patient's individual needs from a virtual environment database in real time, and generates or calls a virtual avatar according to the patient's physical characteristics to make the patient have a sense of possession of the virtual avatar and / or the real body. When the patient performs an action during the rehabilitation task, the virtual avatar is also configured to map the patient's limb movement in real time to make the patient have a sense of autonomy of the virtual avatar and / or the real body; the virtual avatar is also used to generate a virtual environment matching the rehabilitation task, and the virtual avatar is placed in the virtual environment to make the patient have a sense of position of the virtual avatar and / or the real body, laying a virtual foundation for embodied rehabilitation. The embodied feeling of the patient to the virtual avatar / real body can rebuild the damaged body schema and body image.

[0030] Understandably, the creation of virtual avatars is based on the patient's real physical characteristics and can be specifically optimized in terms of limb presentation, precisely inducing the patient's sense of ownership over the virtual avatar. Specifically, the creation of the virtual avatar can be based on some features of the patient's real body, achieving precise matching of core characteristics. This can involve collecting and replicating basic physical characteristics such as the patient's age, weight, and gender, ensuring that the virtual avatar completely corresponds to the patient's real body in core dimensions such as appearance, posture, and proportions. This establishes a basic perception that the "virtual avatar fits the individual," laying the morphological foundation for the sense of ownership.

[0031] Meanwhile, based on rehabilitation needs, some parameters of the virtual avatar can be designed using a non-matching approach to adapt to the rehabilitation scenarios of specific patients. For example, for patients wearing prostheses, to help them better carry out rehabilitation exercises and establish a normal understanding of limb function, after creating a virtual avatar based on the patient's real age, weight, gender, and other characteristics, the virtual domain system no longer replicates the real form of the prosthesis. Instead, it generates a healthy limb corresponding to the prosthesis. This preserves the connection between the virtual avatar and the patient to induce a sense of ownership and strengthen the perception that "the virtual limb is an extension of one's own limb." At the same time, the presentation of the healthy limb provides the patient with a target reference for functional recovery, helping them overcome the cognitive limitations caused by physical defects and improving the pertinence and effectiveness of rehabilitation training.

[0032] In this embodiment, the embodied stimulation system applies visual stimulation, which includes at least a virtual avatar, and at least one of touch, hearing, smell, and vestibular stimulation to the patient, so that the patient experiences embodied sensation during the rehabilitation process from two or more sensory stimuli. The embodied sensation includes at least one of the sense of possession, autonomy, and location.

[0033] Embodied stimulation systems play a crucial role in transforming environmental factors in virtual scenarios into real physical stimuli, providing the body with constantly updated sensory input. This helps the patient's brain continuously build new body schema models during the Bayesian integration process. Immersive virtual environments and multi-sensory stimulation reduce the monotony of traditional rehabilitation training. Dynamically adjusted training difficulty avoids patient frustration, while the immediate motor feedback and sense of accomplishment brought by embodied stimulation can alleviate patient anxiety and resistance, reduce rehabilitation interruptions caused by psychological factors, and ensure long-term training adherence.

[0034] Embodied stimulation systems can also provide multi-sensory input to the patient's brain. For example, various physical stimulation devices can simulate different environmental conditions. A wind simulator can generate airflow of varying intensity and direction to simulate outdoor wind; a vibration platform can generate vibrations of different frequencies and amplitudes to simulate the bumpy feeling of walking on uneven surfaces. Furthermore, embodied stimulation systems can apply targeted physical stimulation to patients based on their rehabilitation training needs, providing an interactive environment for embodied rehabilitation and improving its efficiency. For instance, during muscle strength training, resistance devices can apply adjustable resistance to the patient's limb movements to help strengthen muscles; in balance training, by changing the stability of the support surface, an unstable training environment can be created to improve balance. These realistic physical stimuli, combined with the virtual environment, allow patients to have a more realistic and lifelike experience during virtual training, effectively enhancing training outcomes.

[0035] In this embodiment, the embodied assessment system is a key component in achieving personalized rehabilitation training. This system provides real-time updated embodied feedback information for both the virtual domain and the embodied stimulation domain. During the execution of rehabilitation movements, the embodied assessment system obtains real-time embodied sensations by questioning the patient; alternatively, by collecting the patient's physiological indicators, including at least one of heart rate, electromyography, electroencephalography, conductancestokinesis, eye movements, micro-expressions, and stress; or by collecting the patient's kinematic data. The embodied assessment system can accurately acquire the patient's real-time embodied state during the rehabilitation process, providing data support for subsequent adjustments.

[0036] Optionally, the patient's kinematic data can be collected using a motion posture monitoring module, such as inertial sensors or visual capture technology, to collect the patient's kinematic data in real time and accurately, including information such as limb position, angle, speed, and acceleration, enabling precise tracking of every detail of the patient's movements during training. Simultaneously, physiological signal acquisition devices, such as heart rate sensors, electromyography (EMG) sensors, and electroencephalography (EEG) sensors, continuously collect the patient's real-time physiological signals, such as heart rate, muscle electrical activity, and brain electrical activity. These physiological signals can intuitively reflect the patient's physical state during rehabilitation, providing a concrete sense of ownership, autonomy, and position, or directly achieving a final real-time sense of embodiedness.

[0037] The system integrates and analyzes the collected kinematic data and physiological signals, adjusts the signal output of the virtual domain system and the embodiment sensation stimulation system in real time according to the preset algorithm and model, and thus forms an embodiment rehabilitation closed loop. For example, when the system detects that the patient's real-time embodiment sensation is low, the difficulty of the training task in the virtual environment is automatically reduced, or the movement speed and amplitude of the virtual avatar are adjusted, and the difficulty of the environment simulation in the embodiment sensation stimulation system is reduced at the same time, so as to ensure that the patient trains in a safe and comfortable state and improve the embodiment sensation and rehabilitation effect of the patient.

[0038] The embodiment sensation evaluation system builds a Bayesian-driven embodiment rehabilitation closed loop between the virtual domain, the embodiment sensation stimulation system and the patient through real-time updated feedback information. By accurately tracking each motion detail, the system can evaluate the patient's movement intention and repair state in real time, and ensure the matching of virtual feedback and physiological state of the patient.

[0039] In this embodiment, the device domain system is equipped with several auxiliary motion devices for building the mapping between the patient and the virtual avatar, making the patient perceive the consistency of his own action and virtual action and promoting the remodeling of body representation, and evaluating the integrity of action function and body space representation by quantifying the spatial features and time rules of body movement, which plays an important auxiliary role in rehabilitation training. These auxiliary motion devices can include mechanical exoskeletons, intelligent rehabilitation robots, wearable assistive devices, etc. During the training process, the central control module calls the matching auxiliary motion device according to the patient's movement intention and physical condition. For example, for patients with weak limb strength who are difficult to complete some actions independently, the mechanical exoskeleton can provide necessary power support to help the patient complete the movement action that cannot be completed originally and help the patient gradually recover the limb function; the intelligent rehabilitation robot can adjust the assistive force and movement trajectory in real time according to the patient's movement situation to provide personalized assistive training for the patient. In addition, the auxiliary motion device can also cooperate with the functions of the virtual domain and the embodiment sensation evaluation system, for example, through linkage with the virtual domain system, the action of the auxiliary motion device can be synchronized with the action of the virtual avatar, so that the patient can obtain physical assistance and also feel natural and smooth movement experience visually; in cooperation with the embodiment sensation evaluation system, the working mode and parameters of the auxiliary device are automatically adjusted according to the changes of the patient's physiological signals, further improving the safety and effectiveness of rehabilitation training.

[0040] In this embodiment, as shown in Figure 3 The central control module serves as the integration center of embodiment interaction and undertakes complex and critical control and coordination tasks. The central control module is used at least for adjusting the difficulty of the rehabilitation task, or adjusting the type or intensity of sensory stimulation applied by the embodiment sensation stimulation system, or adjusting the controllable parameters of the auxiliary motion device of the device domain system, or adjusting the mapping action of the virtual avatar based on the real-time embodiment sensation.

[0041] Upon receiving the rehabilitation task instruction, the central control module quickly starts the collaborative workflow of each system module. First, the control virtual domain system generates or selects a virtual environment and a virtual avatar that match the rehabilitation task, creating a suitable training scene for the patient; at the same time, the control embodiment assessment system starts the corresponding physiological signal acquisition device and motion posture monitoring module according to the preset physiological monitoring index acquisition process, and real-time obtains the body state information of the patient; then, according to the body condition, real-time embodiment and rehabilitation needs of the patient, the control device domain system calls the appropriate assisted motion device.

[0042] During the training process, the central control module constructs and updates the extensible body schema model based on the kinematic data collected by the motion posture monitoring module, accurately drives the action of the virtual avatar, and ensures that the virtual avatar can truly reflect the movement intention of the patient, including the internal representation matrix of the position, relationship and movement possibility of each part of the patient's body. At the same time, accurate control instructions can be generated in combination with the real-time embodiment obtained by the embodiment assessment system. These control instructions can be used on the one hand to control the embodiment stimulation system to apply corresponding physical stimulation, so that the scene change in the virtual environment is truly fed back in reality to better reconstruct the body image model; on the other hand, to drive the assisted motion device to perform embodiment collaborative rehabilitation action, provide personalized assistance for the patient, and assist the patient in constructing the damaged body schema model to better achieve motion control. In addition, the central control module is also responsible for realizing the decoding work between the "virtual domain-embodiment stimulation system-embodiment assessment system-device domain", ensuring that the information interaction and collaborative work between each system module can be accurately and efficiently carried out; at the same time, by fusing the model, the movement trend and rehabilitation needs of the patient are predicted, embodiment decision and control are made, and the automation, intelligence and personalization of the rehabilitation training process are realized.

[0043] In the present application, through the close cooperation and collaborative work of the above-mentioned system modules, the virtual reality embodiment rehabilitation system can effectively overcome the shortcomings of the existing rehabilitation training methods and fully exert the advantages of virtual reality embodiment remodeling. During the training process, the patient repairs the cognitive bias of the brain to the body through the interaction between the virtual environment and the body, and reconstructs the body image from the perspective of neural plasticity; not only can actively control the action of the virtual avatar and actively participate in the rehabilitation training, but also can perform rehabilitation training from the neural control level rather than only focusing on the rehabilitation of the limb part; when the body is tired or it is difficult to complete some actions, the patient can receive passive training assistance from the embodiment stimulation system and the device domain system, ensuring the smooth progress of the training and the rehabilitation effect.

[0044] The system can create movement scenes and training conditions that are difficult to achieve in reality for the patient by means of a virtual reality environment, help the patient complete movement intentions that cannot be completed originally, and effectively improve the effect and efficiency of rehabilitation training. Meanwhile, the system obtains real-time embodiment feeling data of the patient by collecting and analyzing physiological signals and kinematic data of the patient in real time, realizes adjustment of a personalized rehabilitation training scheme, helps the patient to rebuild the brain's cognition of the body, and ensures the safety and comfort of the training process, thereby providing the patient with a more high-quality and efficient rehabilitation training experience, and having a broad application prospect and significant social benefits.

[0045] The present application breaks through the limitation of existing VR rehabilitation which only focuses on single visual immersion, builds the patient's sense of possession, autonomy and location through a virtual domain system, and realizes precise shaping and strengthening of embodiment feeling by matching multi-sensory composite stimulation, thereby promoting rehabilitation training from "passive movement" to "embodiment interaction of brain-body-virtual environment", effectively activating brain neural plasticity, promoting damaged motor cortex reorganization and neural pathway reconstruction, and providing core support for motor function recovery.

[0046] Based on the core view of embodiment cognition that "cognition originates from the interaction between body and environment", the present application precisely matches the virtual avatar with the patient's body characteristics and maps the real-time limb movement, so that the patient perceives the virtual avatar as "self-body extension"; in combination with the real physical stimulation (such as tactile sensation, force feedback, etc.) provided by the embodiment feeling stimulation domain, a complete embodiment interaction link of "virtual environment-real body-physical feedback" is constructed, so that the patient's training in the virtual scene is converted into real body perception and cognitive experience. This embodiment experience not only effectively stimulates the patient's active rehabilitation willingness, but also accurately connects the cognitive needs of the patient's motor function recovery, solving the problem of insufficient rehabilitation targeting caused by the lack of embodiment perception linkage in traditional VR rehabilitation systems which only focus on visual immersion.

[0047] The present application follows the cognitive formation rule of embodiment cognition "perception-movement-feedback", and the central control module can realize information decoding and collaborative scheduling of the four domains, taking the patient's body movement as the core, dynamically evaluating the patient's real-time embodiment feeling through the kinematic data (such as body movement trajectory, joint activity, etc.) and real-time physiological signals (such as electromyography, electroencephalogram, etc. reflecting the embodiment cognition state) collected by the embodiment feeling evaluation domain; and then adjusting the scene parameters (such as difficulty, perspective, etc.) of the virtual domain, driving the embodiment feeling stimulation domain to apply matching physical stimulation, and controlling the device domain to assist movement, thereby forming a closed-loop rehabilitation process of "embodiment perception VR environment-movement execution-embodiment feeling feedback-parameter optimization". The closed loop always takes the patient's embodiment feeling as the core adjustment basis, ensures that the rehabilitation scheme is accurately adapted to the patient's body cognitive ability and movement control level, and greatly improves the precision and effectiveness of rehabilitation compared with traditional open rehabilitation training.

[0048] This invention, based on the differences in physical characteristics (such as limb length, muscle strength, etc.) and embodied cognitive abilities of different patients, can accurately match virtual avatar parameters (ensuring consistency in body mapping) and flexibly call upon different virtual environments (adapting to different cognitive levels), assistive movement devices (compensating for different degrees of motor function deficits), and physiological monitoring indicators (targeted assessment of the degree of embodiment) according to the needs of rehabilitation tasks. For example, for patients with severe motor dysfunction, assistive movement devices in the device domain can enhance the implementation of their motor intentions and help patients establish basic embodied interactive perception; for patients with mild functional impairment, complex scenarios can enhance the depth of embodied cognition, thereby adapting to patients with different disease types such as stroke, spinal cord injury, and Parkinson's disease, as well as patients at different stages of rehabilitation (mild, moderate, and severe), solving the problem that existing rehabilitation systems are unable to adapt to diverse embodied ability needs.

[0049] In some embodiments, the embodied assessment system is the core data acquisition and quantification unit of the virtual reality embodied rehabilitation system. Its core configuration is an embodied integration module, which uses standardized formulas to transform the patient's sense of ownership, autonomy, and location into quantifiable real-time embodied indicators, providing precise data support for the dynamic control of the rehabilitation process. The embodied integration module obtains the real-time embodied feeling using the following formula: in, It represents the real-time sense of embodiment, with a value between 0 and 1; the value is positively correlated with the strength of the sense of embodiment, that is, the higher the value, the stronger and more stable the patient's sense of embodiment of the virtual avatar and their own body.

[0050] Indicates a sense of ownership. Indicates the sense of autonomy factor, This represents the positional perception factor, with a value between 0 and 1. , , These represent the dynamic weighting coefficients, with values ​​between 0 and 1. Adjustments can be made based on different rehabilitation tasks and training statuses. These adjustments can be dynamically made according to different rehabilitation task types (such as limb coordination training, strength recovery training, balance training, etc.) and the patient's real-time training status (initial adaptation stage, advanced strengthening stage, consolidation and recovery stage, etc.) to ensure that the assessment results align with the rehabilitation priorities in different scenarios. For example, in the initial stage, adjustments can be made... The weighting is set relatively high, initially allowing patients to establish a sense of ownership over their own body or virtual avatar. Later, it can be... The weighting is relatively high, and rehabilitation training tasks are carried out by utilizing the sense of autonomy derived from embodied experience.

[0051] As a specific method, if A value greater than 0.8 indicates a high embodied state; at this stage, the patient has established a stable sense of ownership, autonomy, and place, with sufficient embodied experience to meet the needs of the current rehabilitation task. If If the value is less than 0.4, the patient is in a low state of embodiment; at this time, the patient's sense of embodiment is weak, and they cannot effectively establish a connection between the virtual avatar and their own body. It is necessary to adjust the rehabilitation parameters in a timely manner to strengthen the sense of embodiment. If the patient is in a moderate state of embodiment, the sense of embodiment is in a dynamic fluctuation range. It is necessary to monitor the training process in real time and fine-tune the parameters as needed to maintain or improve the sense of embodiment.

[0052] In the above embodiments, the formula calculation of the embodiedness integration module needs to rely on multi-dimensional basic data collection. For example, the sense of ownership (O), sense of autonomy (A), and sense of location (L) can be obtained through three types of data: subjective, physiological, and behavioral data. After processing and calculation, the real-time embodiedness E(t) can be obtained.

[0053] Firstly, the data was obtained based on a subjective questionnaire. The Embodiment Questionnaire (Botvinick & Cohen, 1998) was used to directly ask patients seven core questions. Combined with standardized scoring rules, subjective scoring data for the sense of possession (O), sense of autonomy (A), and sense of location (L) were extracted as one of the basic sources of these three types of sensory factors, directly reflecting the patient's subjective perception of their association with the virtual avatar.

[0054] Secondly, based on physiological indicators. By collecting multiple physiological indicators, three types of sensory factors are indirectly quantified. For example, (1) skin conductance response: when the virtual avatar is threatened (such as being chopped by an axe in the virtual scene), if the patient has a strong sense of ownership, the real body will trigger sympathetic nerve excitation, and the skin conductance will spike instantly. By monitoring the change amplitude of skin conductance signal, the intensity of ownership is quantified and converted into ownership factor (O). (2) EEG signal: after the patient has embodied the avatar, the event-related desynchronization (ERD) phenomenon will occur during the motor imagination. The degree of ERD is positively correlated with the intensity of embodiedness. By collecting EEG signals and analyzing ERD indicators, the sense of autonomy and the intensity of ownership are comprehensively mapped to help optimize the values ​​of the two types of factors. (3) EMG / movement posture monitoring: collect the patient's EMG signals and limb movement posture data, and compare them with the movement trajectory of the virtual avatar. If the overlap between the two is high and the time difference is ≤50ms, it indicates that the patient has a strong sense of autonomy. Based on this, the sense of autonomy factor (A) is quantified, and the stability of the sense of position is verified. (4) Eye movement signal: The patient's eye movement data is collected by eye movement tracking device. The correlation between eye movement focus, fixation duration and trajectory and virtual avatar and virtual environment is analyzed. If the eye movement behavior is consistent with the movement and position change of the virtual avatar, the position sense factor (L) can be quantified to reflect the patient's perception of their own position in the virtual environment.

[0055] Third, behavioral data acquisition. Behavioral data is collected through proprioceptive drift test to quantify the positional factor (L): patients are asked to point out the actual position of their own limbs (such as hands). If the pointed position is biased towards the position of the virtual limb (rather than the real limb position), the greater the deviation, the more the brain's judgment of the limb position is rewritten by the virtual vision, and the stronger the positional sense. This is then converted into the corresponding positional factor (L) value.

[0056] The above method can transform the previously difficult-to-capture sense of ownership, autonomy, and position into quantifiable E(t) indicators, replacing the traditional assessment model that relies on the patient's subjective description and the experience judgment of medical staff, thus improving the objectivity, accuracy, and repeatability of embodied assessment. Through the dynamic adjustment of α, β, and γ, the assessment focus can be optimized for different rehabilitation tasks and different patient training stages. For example, in balancing training, the weight of positional factors can be emphasized, while in limb control training, the weight of autonomy factors can be emphasized, making the assessment results more aligned with rehabilitation needs. The clear range of E(t) values ​​and state divisions provide the central control module with clear regulatory trigger conditions, making the adjustment of parameters such as the difficulty of rehabilitation tasks and the intensity of sensory stimulation more targeted, and helping to form a closed-loop rehabilitation system of "assessment-regulation-training".

[0057] The central control module, acting as the core of system coordination and scheduling, uses the real-time embodied sensation E(t) output by the embodied sensation assessment system as the basis for regulation. It coordinates with the virtual domain system, embodied stimulation system, and device domain system to achieve dynamic adaptation, ensuring that the embodied sensation remains within a reasonable range suitable for rehabilitation training. Based on the value of the real-time embodied sensation E(t), the central control module can enhance the patient's embodied sensation through multi-dimensional real-time regulation, for example: (1) When E(t) < 0.4 (low embodied state), the patient’s embodied feeling of the virtual hand is weak. The virtual domain system’s autonomous sense evoked module can be linked to optimize the action mapping logic, adjust the electromyographic feedback parameters of the virtual hand, and transform the patient’s weak electromyographic signals into large-amplitude movements of the virtual hand to enhance the sense of autonomy. At the same time, the tactile feedback intensity of the visual-tactile coupling module can be enhanced, and the device domain-assisted motion device can apply clearer tactile stimulation to improve the sensory linkage effect.

[0058] (2) For scenarios where the sense of ownership is insufficient, the embodied stimulation system can be linked to construct an electrocardiogram-visual coupling mechanism. The display parameters of the virtual hand can be adjusted through the central control module so that the color of the virtual hand skin flickers slightly with the patient's real heart rate, simulating the physiological characteristics of real skin, bringing the virtual limbs closer to the real body, and assisting the sense of ownership induction module to enhance the effect.

[0059] (3) When 0.4≤E(t)≤0.8 (medium embodied state), the difficulty of rehabilitation tasks can be appropriately increased (such as increasing the accuracy requirements of virtual hand grasping objects), while the perspective parameters of the position enhancement module are finely adjusted to optimize the visual presentation effect of the virtual scene and maintain a steady improvement in embodiedness.

[0060] (4) When E(t) > 0.8 (high embodied state), the current sensory stimulation intensity and equipment auxiliary parameters can be maintained, and the training difficulty can be gradually increased to maximize rehabilitation efficacy.

[0061] In some embodiments, the virtual domain system includes: an ownership-inducing module, an autonomy-inducing module, and a location-enhancing module, etc.

[0062] The possession-inducing module generates a suitable healthy virtual limb on the virtual body based on the patient's atrophied limb, thereby accurately inducing the patient's sense of ownership over the virtual body. Specifically, it first collects the basic characteristic parameters of the patient's atrophied limb (including limb outline, skeletal structure, joint range of motion, etc.). Based on these parameters, a healthy virtual limb with a shape and proportion matching the patient's healthy limb is generated on the virtual body, ensuring that the virtual limb not only conforms to the patient's overall body characteristics but also compensates for the atrophy defects of the affected limb. By presenting a healthy virtual limb, the patient's sense of physical rejection caused by the abnormal function of the affected limb is reduced, guiding the brain to establish an association between the virtual limb and its own body, thereby strengthening the cognition that "the virtual limb is an extension of one's own limb," effectively inducing a sense of ownership, and laying the foundation for embodied interaction in subsequent rehabilitation training.

[0063] The autonomy-inducing module amplifies the patient's small-amplitude limb movements, ensuring that the limb movements mapped by the virtual avatar are displayed within a normal range. This strengthens the patient's perception of control over the virtual avatar's movements and enhances their sense of autonomy. Specifically, it collects real-time data on the patient's limb movements during rehabilitation training. For small-amplitude limb movements that the patient can only perform due to functional impairment, the module uses an algorithm to proportionally amplify them, ensuring that the limb movements mapped by the virtual avatar present a normal physiological range. For example, if the patient's affected limb can only complete 10° of joint movement, the module amplifies the movement to 30° (within the normal range of limb movement) and simultaneously maps it to the virtual avatar. This allows the patient to intuitively perceive that their movement intentions are accurately executed and achieve the desired effect, reinforcing the perception that "the virtual avatar's movements are controlled by the patient." This addresses the problem of patients being unable to generate a strong sense of autonomy due to limited range of motion and stimulates their willingness to move actively.

[0064] The positional awareness enhancement module allows patients to view virtual scenes or avatars from either a first-person or third-person perspective, optimizing their positional perception of the virtual environment and avatar, and enhancing their sense of location. Specifically, it provides two switchable perspectives: first-person and third-person, which patients can choose according to their own adaptability and rehabilitation needs. When selecting the first-person perspective, patients perceive the virtual environment from the avatar's point of view, strengthening their sense of immersion in the virtual scene and accurately capturing the positional relationship between the avatar and the virtual environment. When selecting the third-person perspective, patients can directly observe the overall movements of the avatar and its spatial position within the virtual scene, aiding in correcting limb postures and clarifying their relative position to the virtual environment. This flexible perspective switching adapts to the positional awareness needs of patients at different stages of rehabilitation, ensuring that patients establish a clear and stable positional awareness in the virtual environment and enhancing the strength of their sense of location.

[0065] In some embodiments, such as Figure 1 and Figure 2 As shown, the virtual domain system includes a virtual environment database 41, a virtual avatar generation module 42, and an action mapping unit 43; each component has a clear division of labor and works closely together to create a highly adaptive and immersive virtual rehabilitation training environment for patients.

[0066] In this embodiment, the virtual environment database 41 stores contextualized task scenarios corresponding to different rehabilitation stages, which can match the patient's embodied cognitive level at different rehabilitation stages, avoiding a disconnect between scenario difficulty and physical ability. The virtual avatar generation module generates personalized avatars based on the patient's core body parameters such as gender, height, and limb length, allowing the patient to quickly establish embodied identity of "avatar as self." The motion mapping unit can achieve low-latency motion synchronization of ≤50ms through inverse kinematics algorithms, and even amplify and map movements that the patient cannot complete, ensuring accurate matching between physical movement intentions and virtual actions, and constructing a deep embodied interaction link of "real body - personalized avatar - contextualized scenario." This precisely adapted embodied experience allows the patient's training in the virtual scenario to fully conform to their own physical characteristics and rehabilitation needs, transforming virtual training into real physical perception and cognitive reshaping. This not only significantly improves the willingness to actively rehabilitate, but also solves the core pain points of traditional VR rehabilitation, such as weak embodied identity and mismatch between scenarios and physical abilities.

[0067] In this embodiment, the core function of the virtual avatar generation module 42 is to generate a personalized virtual avatar based on the patient's individual body parameters. For example, by collecting key parameters such as the patient's gender, height, limb length, and joint mobility, and using 3D modeling and digital image processing technology, a virtual avatar highly similar to the patient's physical characteristics can be constructed. For instance, for a male patient who is 175cm tall and has limited left leg joint mobility, the module will precisely create a virtual avatar according to his body proportions and simulate the state of limited left leg joint mobility. In this way, the virtual avatar seen by the patient in the virtual environment is like their own "digital clone," making it easier to feel a sense of immersion and identification during training, thus enhancing their enthusiasm and initiative. At the same time, a highly accurate virtual avatar can more accurately reflect the patient's physical condition, making rehabilitation training more scientific and reasonable. This personalized design helps patients quickly establish an embodied identity of "virtual avatar as an extension of oneself," breaking the sense of alienation of being a "bystander" in traditional rehabilitation training, and allowing patients to actively regard the training tasks in the virtual scene as their own body's real action needs. At the same time, a persona that fits the individual can reduce patients' psychological resistance to rehabilitation training, stimulate their willingness to participate actively, lay a psychological and cognitive foundation for the subsequent embodied rehabilitation loop of "perception-movement-feedback", ensure that patients always participate in the rehabilitation process in an "subjective" manner, and improve the focus and effectiveness of training.

[0068] In this embodiment, the motion mapping unit 43 is a key component for enabling real-time interaction between the patient and the virtual avatar, primarily relying on inverse kinematics algorithms to perform motion conversion. Once the device domain system collects the patient's kinematic data, the motion mapping unit immediately processes this data, converting it into synchronized movements of the virtual avatar using the inverse kinematics algorithm, with mapping latency controlled within 50ms. For example, if the patient raises their arm in reality, the motion mapping unit 43 quickly calculates the motion angle and displacement of the corresponding joint in the virtual avatar, enabling it to synchronously complete the arm-raising movement in the virtual environment. This low-latency motion synchronization ensures real-time matching between the patient's movement intentions and the virtual actions, allowing the patient to feel that "their own movements can directly control the virtual environment," strengthening the embodied association between body movement and environmental feedback, and aiding in the recoding and reshaping of motor functions by the brain.

[0069] More importantly, for movements that patients cannot perform due to physical limitations, the motion mapping unit 43 can amplify and map these movements in the virtual environment. For example, a patient unable to perform a squat due to insufficient leg muscle strength can have their limited leg movements amplified based on their existing limb motor abilities, allowing the virtual avatar to perform a near-squat in the virtual environment. Simultaneously, this unit can control the motion mapping delay to no more than 50ms, achieving near real-time synchronization between the patient's movements and the virtual avatar's movements. This significantly enhances the patient's immersion and interactive experience in the virtual environment, making them feel truly present in a virtual scene, thus more effectively promoting rehabilitation training. It also enables embodied decision-making and control based on fusion model predictions. This approach helps patients establish basic motor intention-motor feedback cognition, gradually awakening motor function and filling the gap in traditional rehabilitation where severely ill patients "difficult to participate in effective interaction," ensuring that patients at different rehabilitation stages can achieve effective embodied rehabilitation training.

[0070] The embodied stimulation system in this embodiment of the invention aims to stimulate patients to experience embodied sensations and can comprehensively simulate a virtual environment with physical properties, providing patients with realistic rehabilitation training scenarios. Specifically, the embodied stimulation system includes a visual module 11, and at least one of the following modules: an auditory module 12, a tactile module 13, an olfactory module 14, a temperature regulation module 15, a humidity regulation module 16, a wind speed regulation module 17, and a gas partial pressure regulation module 18. These modules are used to create multi-sensory channel stimulation (visual, tactile, auditory, olfactory, and vestibular) and simulation of environmental factors such as temperature, humidity, wind speed, light, and gas in a virtual environment with physical properties, and to reshape body representations through multi-sensory channel stimulation. Through the coordinated work of these modules, the system can accurately simulate different environmental factors and sensory stimuli according to rehabilitation needs, allowing patients to obtain near-realistic experiences in virtual training. This induces the reshaping of the patient's body schema with environmental stimuli as the core, while providing rich modal data for multi-sensory signal integration and achieving dynamic updates of body representations based on a Bayesian prediction model.

[0071] In some embodiments, the visual module 11 is used to assist the embodied rehabilitation system in constructing a spatial framework for body representation. It may include a spherical display screen, an arc display screen, or a giant screen display screen set in the rehabilitation cabin 10. From the perspective of immersion and scene realism, the surround display form can maximize the coverage of the patient's field of vision, reduce external environmental interference, and allow the patient to be fully visually integrated into the virtual scene.

[0072] The core of the sense of ownership is for patients to establish a sense of belonging that "the virtual avatar is an extension of their own body." The visual module reinforces this cognition through two layers: matching body representation and binding visual immersion. On the one hand, the module accurately presents a virtual avatar that matches the patient's real body characteristics through visual types such as ontological vision and mirror vision. This allows patients to visually perceive that the avatar's limb shape and posture are highly consistent with their own, constructing a visual cognition that "the virtual avatar and their own body share the same origin." On the other hand, the module relies on spherical or giant screen displays to form a surround visual space, enveloping the patient in the virtual environment, eliminating the sense of visual boundaries, and allowing the patient's visual attention to be fully focused on the connection between the virtual avatar and the virtual scene. This weakens the interference of the real environment, further deepening the binding relationship between the virtual avatar and their own body, and strengthening the sense of ownership.

[0073] The core of autonomy lies in the patient's recognition that "the actions of the virtual avatar are driven by their own movement intentions." The visual module reinforces this recognition through real-time visual feedback and motion-visual matching. Based on the patient's current position and angle, the module synchronously displays the virtual avatar's movement trajectory and the corresponding changes in the virtual scene. Furthermore, it accurately captures and presents the entire process of the patient's limb movements mapping to the virtual avatar through motion-visual analysis, allowing the patient to clearly see the complete visual chain of "self-generated movement intention → limbs performing the action → virtual avatar synchronously executing the same action." This real-time visual correspondence between movement intention and virtual action allows the patient to intuitively confirm that they are the sole master of the virtual avatar's movements, forming a perceptual closed loop of "self-movement → visual feedback → confirmation of autonomous control," significantly enhancing the sense of autonomy.

[0074] The core of spatial awareness is the patient's accurate perception of the "spatial position of the virtual avatar in the virtual rehabilitation environment and the relative position of themselves with the virtual environment." The visual module, as the core carrier of spatial awareness, deeply enhances spatial awareness through three layers: framework construction, precise presentation, and multi-dimensional assistance. First, the module directly assists the embodied rehabilitation system in constructing a spatial framework of bodily representation, laying the foundation for the patient to perceive the spatial relationship between themselves and the virtual avatar. Second, based on the spatial layout of the virtual rehabilitation environment and combined with the patient's real-time position and angle, the module dynamically adapts and presents corresponding virtual reality scenes and perspectives, allowing the patient to visually perceive the coordinates of the virtual avatar in the virtual space, the relative distance to virtual objects, and the spatial changes in the movement trajectory, forming the core perception link of "visual perception → spatial position judgment." Third, the surround design of the spherical / giant-screen display provides the patient with a 360° virtual spatial visual experience without blind spots, compensating for the limitations of local vision. Combined with proprioceptive vision's visual perception of the position of their own limbs, this makes the patient's perception of the virtual avatar's position in the virtual environment clearer and more three-dimensional. Simultaneously, it complements other sensory stimuli, further solidifying the foundation of spatial awareness.

[0075] Optionally, the visual module 11 can also display the corresponding virtual reality scene and perspective based on the simulated scene required for virtual reality rehabilitation and the current position and angle of the patient; for example, when the patient is doing walking training in the virtual scene, the visual module 11 will adjust the display screen synchronously according to the patient's steps and turns, so that the patient feels as if he is really moving in the virtual environment.

[0076] Optionally, the visual module 11 may also include one or more of proprioception, mirror vision, and motion vision; proprioception helps patients perceive the position and posture of their own body in space; mirror vision provides patients with body motion feedback from different perspectives through the principle of reflection; motion vision focuses on displaying the details and effects of the patient's movements. The combination of multiple visual forms enhances the patient's cognition and control of their own body.

[0077] In some embodiments, the tactile module is used to construct representations of body structure and surface state, and may include one or more of the following: vibrating pads, pneumatic / hydraulic tactile devices, microneedle array texture tactile simulation devices, or other wearable tactile devices disposed within the embodied rehabilitation system. These devices may be integrated into the wearable system, such as gloves or clothing. The tactile module constructs representations of body structure and surface state through various tactile devices, and, combined with tactile stimulation adapted to virtual scenes and multiple types of tactile feedback, enhances the patient's sense of embodiedness from three dimensions: ownership, autonomy, and location, providing crucial tactile support for embodied rehabilitation.

[0078] Optionally, the tactile module is also based on the simulated scene required for virtual reality rehabilitation and the surface features of the object that the patient is currently in contact with or simulates touch; for example, when the patient touches a virtual rough stone in a virtual scene, the microneedle array texture tactile simulation device will simulate a rough touch; if touching a soft fabric, the pneumatic / hydraulic tactile device will provide a gentle squeezing sensation.

[0079] Optionally, the tactile module further includes one or more stimulation types among passive tactile, active tactile, and pain sensation; passive tactile provides feedback when the patient unconsciously touches a virtual object; active tactile plays a role when the patient actively explores the virtual environment; and appropriate pain simulation (within a safe range) can help the patient develop a correct awareness of body protection and avoid injury caused by misoperation in real life.

[0080] The tactile module establishes a sense of belonging between the virtual avatar and the real body through tactile feedback. Relying on vibrating pads, tactile wearable devices, and other means, the module simulates tactile perception that matches the patient's real body structure. Combined with passive tactile stimulation, the module allows the patient to receive tactile feedback consistent with the real body when touching virtual objects. At the same time, it simulates the surface texture and structure of the patient's own limbs, reinforcing the understanding that "the tactile perception of the virtual avatar is the same as the perception of one's own body," deepening the bond between the virtual avatar and the real body, and enhancing the sense of ownership.

[0081] The tactile module uses active tactile stimulation to allow patients to receive corresponding tactile feedback when they control their virtual avatar to touch virtual objects, forming a closed loop of "self-motor intention → virtual action → tactile feedback". This allows patients to intuitively perceive that their own actions in controlling the virtual avatar can produce real tactile responses, clarify that they are the master of the virtual actions, further strengthen their sense of autonomy, and stimulate their willingness to actively train.

[0082] The tactile module can adapt the intensity and type of tactile stimulation to different locations based on the surface features of objects in the virtual scene. Similar to the spatial adaptation logic of the temperature regulation module, it allows patients to perceive the position of the virtual avatar in the virtual environment through tactile differences. For example, touching objects in different areas of the virtual scene can produce different tactile sensations, forming a link of "tactile difference → spatial position judgment". This makes up for the limitations of single vision and allows patients to clearly perceive the relative position of the virtual avatar and the virtual environment, thus strengthening their sense of position.

[0083] In some embodiments, the auditory module 12 is used to indirectly adjust body position perception in a virtual reality environment through sound, and may include a speaker array disposed at different locations within the embodied rehabilitation system; optionally, the auditory module 12 includes sounds associated with its own movements and ambient sounds surrounding the location of the virtual avatar in the environment. When the patient walks in the virtual scene, the speakers play footsteps that match the rhythm of the steps, allowing the patient to perceive their own movements auditorily; simultaneously, depending on the settings of the virtual scene, such as playing birdsong and wind sounds in a forest scene, and vehicle sounds and human voices in a street scene, a realistic environmental atmosphere is created. These sounds not only enhance the immersion of the virtual environment, but also help the patient determine their position and direction in the virtual space through changes in the location and intensity of the sound, thus improving spatial perception.

[0084] The auditory module 12 constructs a sense of belonging between the virtual avatar and the real body through sound association. The module synchronously plays sounds associated with the patient's own movements, such as the virtual avatar raising its hand or taking a step, producing sounds that match the real movements. This allows the patient to perceive that the virtual avatar's movements are consistent with their own, reinforcing the perception that "the virtual avatar's movements are synchronized with their own." Combined with other sensory stimuli, this deepens the bond between the virtual avatar and the real body, enhancing the sense of ownership. The auditory module 12 also strengthens the causal relationship between motor intention and virtual actions through sound feedback. When the patient generates a motor intention and controls their limb movements, the module plays corresponding sounds associated with their own movements in real time, forming a closed loop of "motor intention → limb movement → virtual movement → auditory feedback." This allows the patient to intuitively confirm that their control over the virtual avatar's movements produces corresponding sound responses, clarifying that they are the sole controller of the virtual actions and further strengthening their sense of autonomy. The auditory module 12, relying on the spatial layout of the speaker array, plays the surrounding environmental sounds of the virtual avatar's location. The speakers in different locations output sound intensity and direction differently, allowing patients to perceive the virtual avatar's position in the virtual environment through sound differences. This forms a link of "sound direction / intensity difference → spatial position judgment," which compensates for the limitations of single vision, clearly perceives the relative position of the virtual avatar and the virtual environment, and enhances the sense of position.

[0085] In some embodiments, the olfactory module 13 is used to connect to an edge system to regulate bodily state representation, including an odor release mechanism, a fragrance memory, and an odor concentration detection unit. The olfactory module 13 includes both intrinsically related odors and ambient odors; for example, it releases lavender scent in relaxation training scenarios to help patients soothe their emotions; and releases floral scents in simulated outdoor scenarios to enhance the realism of the scene. Intrinsically related odors can refer to physiological or pathological odors, such as body odor or the smell of excrement. The combination of intrinsically related odors and ambient odors allows patients to better integrate into the virtual environment at the olfactory level, thereby regulating their bodily state and emotions, and increasing their motivation for rehabilitation training.

[0086] Optionally, multiple odor-releasing mechanisms are provided, distributed at different locations within the embodied rehabilitation system; the perfume storage device is used to store liquid perfume or fragrance capsules, and is connected to the odor-releasing mechanism to provide it with odor; the perfume storage device corresponds one-to-one with the odor-releasing mechanism, or one perfume storage device provides odor to two or more odor-releasing mechanisms; different perfume storage devices are used to store the same or different odors. At least one odor concentration detection unit is provided to monitor the odor concentration within the embodied rehabilitation system, ensuring that the odor intensity is moderate and avoiding excessively strong or weak odors that could affect the rehabilitation effect.

[0087] The olfactory module transmits environmental odors and its own associated odors through multi-location odor release mechanisms, constructing an olfactory environment that fits the virtual scene. Smell directly connects to the limbic system; environmental odors can induce patients to regulate their emotions and physiological states (e.g., a refreshing environmental scent induces relaxation), thereby correcting the cognitive association between "environment-emotion-physiological state" in the body state schema. In multi-sensory integration, olfactory signals, as a supplement to the emotional and state dimensions, collaborate with visual scene atmosphere and auditory signal characteristics to enrich the dimensions of multi-sensory integration and enhance the immersiveness and consistency of embodied interaction.

[0088] In some embodiments, the vestibular module is used to simulate a virtual gravity field in a virtual reality scene and enable the patient to perform spatial orientation representation of the body in the virtual reality scene. It includes at least a multi-degree-of-freedom balance platform 31 and a vestibular electrical stimulation device 32. The vestibular module includes at least one type of stimulation: head movement stimulation and gravity change stimulation. In this embodiment, the multi-degree-of-freedom balance platform 31 can simulate tilting and swaying in different directions and degrees, coordinating with movement changes in the virtual scene to give the patient a realistic sense of body displacement. For example, in a scenario simulating climbing a virtual mountain, the balance platform tilts accordingly with the "climbing height" and "terrain changes," allowing the patient to feel changes in gravity and body imbalance, thereby improving balance and spatial orientation. The vestibular electrical stimulation device 32 can stimulate the vestibular organs with a weak current, further enhancing the patient's perception of head movement and gravity changes, and assisting in rehabilitation training. Different types of stimulation, such as head movement stimulation and gravity change stimulation, train the patient's vestibular function from multiple angles, improving their balance and coordination.

[0089] The vestibular system is the core sensory foundation for spatial orientation and balance perception. The environmental gravity changes and head movement-related stimuli it provides can directly induce patients to correct impaired spatial orientation cognition, reshaping the association between the body and the gravitational field and spatial motion. Through continuous vestibular stimulation feedback, it helps patients re-establish the association between "head posture - gravity direction - body balance," restoring their spatial orientation ability. At the multi-sensory integration level, vestibular signals are precisely synchronized with visual-spatial signals (such as the horizon and spatial references in virtual scenes) and limb movement signals (such as trunk posture and limb swing data), providing a core anchor point for balance-related multi-sensory integration. This ensures the consistency between the patient's balance perception and visual and motor perception in the virtual environment, avoiding sensory conflicts. Meanwhile, vestibular signals are used as key modal data inputs to the Bayesian prediction model. Based on the correlation between vestibular stimulation, visual space, and limb movement in historical interaction data, the model predicts the matching between the current vestibular feedback and the body's balance state. If the prediction result is consistent with the actual vestibular input, visual and motor signals, the existing body spatial orientation representation is strengthened, and the body schema reshaping effect is consolidated. If there is a prediction error (such as a mismatch between vestibular stimulation and visual spatial perception), the correlation parameters of "vestibular threshold - head posture - balance state" in the body representation are dynamically adjusted according to the source of the error, driving the real-time dynamic update of body balance and spatial orientation representation, ensuring that the body schema always adapts to the gravity field and movement scene requirements of the virtual environment.

[0090] In some embodiments, the temperature regulation module 14 is used to define the hot and cold surface states in the virtual reality rehabilitation environment, thereby indirectly affecting the internal bodily manifestation state. This includes cooling and heating equipment installed within the rehabilitation system to regulate the ambient temperature inside the chamber from 0°C to 45°C. For example, when simulating a cold snowy scene, the system lowers the temperature, making the patient feel cold and prompting corresponding physiological responses such as vasoconstriction and increased muscle tension. Conversely, when simulating a hot desert scene, the system raises the temperature, making the patient experience heat and triggering physiological changes such as sweating and increased breathing. This temperature simulation not only enhances the realism of the virtual environment but also assists rehabilitation training by regulating the body's physiological responses, such as helping patients improve their adaptability to temperature changes and improving blood circulation.

[0091] The temperature regulation module 14 can enhance the patient's embodied perception from three dimensions: sense of ownership, sense of agency, and sense of location by using the tactile feedback of the physical environment temperature and the temperature association of the virtual scene. This achieves the synergy between the virtual environment temperature and the real body sensation, making the patient's embodied experience more in line with real physiological perception.

[0092] This module regulates the temperature inside the cabin through cooling and heating equipment, allowing the patient's real body to perceive the same hot and cold stimuli as the virtual rehabilitation environment (e.g., when the virtual avatar touches an ice surface, the cabin temperature simultaneously drops to a low temperature range; when the virtual avatar touches a warm surface, the cabin temperature simultaneously rises to a suitable warm temperature range). This allows the patient to develop a sensory association that "the temperature perceived by the virtual avatar is the same as the temperature perceived by their real body," strengthening the bond between the virtual avatar and the real body, thereby enhancing the patient's sense of ownership over the virtual avatar and allowing the patient to more deeply identify that the virtual avatar is an extension of their own body.

[0093] When a patient controls their virtual avatar to make movements that involve contact with hot or cold surfaces, the temperature regulation module simultaneously applies corresponding hot or cold temperature stimuli to the real body. This closed loop of "making virtual movements → receiving instant temperature feedback" allows the patient to clearly perceive that their own movement intentions not only drive the virtual avatar's movements but also bring physiological feedback consistent with real body movements. This makes it clear that the virtual avatar's movements are completely under their control, thereby strengthening their sense of autonomy over the virtual avatar and enhancing their perception and participation in active movement.

[0094] In some embodiments, the humidity regulation module 15 indirectly adjusts the body surface condition by influencing the water evaporation rate. It includes at least one humidifier installed within the embodied rehabilitation system and a ventilation device connecting the inside and outside of the chamber, used to regulate the ambient humidity within the chamber in the range of 30%-90%. For example, in simulating a humid rainforest scenario, increasing humidity makes the patient feel moist skin, as if they were in a real rainforest environment; in simulating a dry desert scenario, decreasing humidity makes the patient experience dry skin. Different humidity environments affect the patient's skin sensations, respiration, etc., thereby helping the patient adapt to different humidity conditions during rehabilitation training. It also has a positive effect on the rehabilitation process by regulating the body's water metabolism.

[0095] Ambient humidity provides patients with continuous feedback on bodily surface comfort by influencing the rate of skin moisture evaporation. This can induce patients to reshape their association schema between bodily surface and ambient humidity, correcting impaired humidity perception. In multi-sensory integration, humidity signals work in conjunction with visual scene, temperature, and tactile signals (such as stickiness and dryness) to improve the multimodal consistency of environmental perception, helping patients better adapt to the perceptual needs of different environments.

[0096] In some embodiments, the wind speed adjustment module 16 provides a dynamic reference for the patient's spatial position, including at least one air blowing device located in front of or to the side of the patient within the embodied rehabilitation system, and capable of adjusting the airflow based on walking speed feedback and ambient wind speed sent by the central control module. For example, when the patient runs quickly in a virtual scene, the air blowing device will produce a larger airflow to simulate oncoming wind; when the patient walks slowly, the airflow will decrease accordingly. This wind speed simulation not only enhances the realism of movement in the virtual environment, but also assists the patient in perceiving their own movement speed and direction through changes in wind resistance and direction, further improving the body's motor control and spatial awareness.

[0097] The synchronous feedback between changes in the direction and magnitude of environmental airflow and the patient's movement state can induce the patient to reconstruct the relationship between body movement and environmental airflow, clarify the interactive cognition of "self-movement-airflow change," and correct impaired dynamic environmental adaptation capabilities. In multi-sensory integration, wind speed signals, as dynamic environmental feedback, work in conjunction with visual motion scenes, limb movement signals, and tactile airflow perception signals to form a dynamic multi-sensory link of "movement-vision-touch-airflow," enhancing the integration consistency of dynamic interaction.

[0098] Optionally, such as Figure 1 As shown, the temperature control module 14, humidity control module 15 and fan speed control module 16 can be integrated into a unit similar to a central air conditioner or wall-mounted air conditioner to reduce the space occupied by the equipment.

[0099] In some embodiments, the light regulation module 17 is used to adjust the circadian rhythm and emotional state of the body's state characteristics. This includes an array of lights positioned on top of the embodied rehabilitation system to adjust the light intensity and color. For example, when simulating a daytime rehabilitation scenario, the system can increase the light intensity to simulate bright sunlight, while adjusting the light color to a white spectrum close to natural light, giving the patient a daytime feeling and helping to improve their vitality and concentration. When simulating a nighttime rest scenario, the system reduces the light intensity and adjusts the light color to a warm yellow, creating a cozy and gentle atmosphere to help the patient relax and soothe their emotions. Furthermore, for patients with circadian rhythm disorders, this module can gradually adjust the intensity, color, and duration of light according to the rehabilitation plan, helping patients rebuild a normal circadian rhythm, improve sleep quality, and enhance overall bodily function. By precisely controlling light conditions, the light regulation module can provide physiological and psychological support to patients during rehabilitation training, promoting the rehabilitation process.

[0100] From the perspective of the auxiliary regulation of physiological and psychological states by light, this module can specifically improve the patient's training state through flexible switching of light intensity and color. For example, for patients with cognitive fatigue, using medium-intensity blue light (studies have shown that blue light can enhance alertness) in conjunction with task guidance in virtual scenarios can help them maintain attention; for anxious or tense patients, switching to low-intensity warm light (such as orange-yellow) creates a soothing atmosphere, helps reduce sympathetic nerve activity, and improves training tolerance; and in balance function training, the gradual change of light intensity (such as weakening from the center to the periphery) can construct "visual anchor points," guiding patients to naturally adjust their center of gravity to align with the center of light, strengthening the balance coordination reflex of "visual-vestibular-somatic".

[0101] Changes in ambient light can induce patients to correct impaired circadian rhythm cognition, reshape the temporal relationship between the body and the light environment, and simultaneously optimize the body's emotional and physiological state patterns. In multisensory integration, light signals, visual scene temporality, auditory environmental atmosphere, and physical emotional state work together to provide a benchmark for temporal multisensory integration and improve the temporal consistency of environmental perception.

[0102] In some embodiments, the gas partial pressure regulating module 18 is used to regulate the concentration of various gases required for the virtual reality environment or to directly act on the rehabilitation treatment itself, including the pressure monitoring and regulating device, gas supply device, and gas concentration monitoring device within the embodied rehabilitation system.

[0103] The gas pressure regulation module 18 monitors the total pressure inside the chamber. When the total pressure exceeds or falls below the safe pressure range, it controls the safety valve to open, connecting the inside and outside of the chamber. The gas supply device includes at least one gas source cylinder storing oxygen, hydrogen, carbon dioxide, or a mixture thereof. The gas concentration monitoring device includes at least one of an oxygen concentration sensor, a hydrogen concentration sensor, and a carbon dioxide concentration sensor to monitor the real-time concentration of the target gas. For example, the gas source cylinder of the gas supply device stores oxygen, hydrogen, carbon dioxide, or a mixture thereof, providing different gas ratios according to rehabilitation needs. For instance, in some rehabilitation treatments, it may be necessary to increase the oxygen concentration to improve the patient's respiratory function; during simulated high-altitude environment training, the gas composition is adjusted to simulate a low-oxygen environment. The gas concentration monitoring device, through oxygen concentration sensors, hydrogen concentration sensors, and carbon dioxide concentration sensors, monitors the target gas concentration in real time, ensuring a stable gas environment that meets rehabilitation requirements, providing a safe and effective rehabilitation training environment for patients.

[0104] From the perspective of targeted rehabilitation intervention, the gas regulation capabilities of the gas partial pressure regulation module 18 can directly serve specific rehabilitation goals. For example, for patients with chronic obstructive pulmonary disease, by setting a "hypoxia partial pressure gradient" (e.g., gradually reducing from normal oxygen concentration to 80%), combined with breathing training scenarios, their tolerance to hypoxic environments and respiratory muscle strength can be improved. For patients who need to improve cerebral blood flow, by appropriately increasing the oxygen partial pressure (e.g., maintaining it at 23%-25%), sufficient oxygen supply to the brain can be provided to assist in the repair of nerve function. As a potential antioxidant intervention gas, the precise control of its concentration (e.g., maintaining it within a safe range of 1%-3%) can be combined with virtual relaxation scenarios to assist in the rehabilitation of oxidative stress-related diseases. This design, which binds the gas environment with pathophysiological needs, makes the module not only an environmental simulation tool but also an important component of personalized rehabilitation programs.

[0105] Changes in environmental gas concentrations allow patients to perceive their body's adaptive responses to different gas environments, inducing them to reshape their association patterns with specific gas environments and correct impaired environmental adaptation cognition. In multi-sensory integration, gas concentration signals, visual environment (such as high altitude or conventional environment), and bodily physiological signals (such as respiratory status) work together to form a multi-sensory "environment-physiology-vision" link, ensuring the safety and consistency of interactions in special environments.

[0106] In the above embodiments, the system relies on body movement adaptation data from the virtual domain, combined with environmental stimulus data from various modules of the embodied sensory stimulation domain, to achieve deep integration of multimodal signals and dynamic updates of body representations through a Bayesian prediction model. Environmental stimulus signals (visual space, auditory orientation, tactile touch, vestibular balance, etc.) from each module are collaboratively input into the model along with body movement signals. The model predicts the matching and rationality of the current multisensory signals based on historical interaction data: if the prediction result matches the actual input, the existing body representation is strengthened, consolidating the body schema reshaping effect; if there is a prediction error, the correlation parameters of the corresponding dimensions (spatial orientation, balance, dynamic movement, etc.) in the body representation are dynamically adjusted according to the source of the error, driving real-time updates of the body representation. This collaborative mechanism ensures that the body schema always adapts to environmental interaction and rehabilitation needs, comprehensively reshaping damaged body cognition and providing core guarantees for achieving embodied rehabilitation goals.

[0107] The embodied assessment system in this invention utilizes multiple monitoring modules to comprehensively collect patients' physiological and motor data, providing a basis for achieving precise mapping between patients and virtual avatars and personalized rehabilitation training. The embodied assessment system undertakes the crucial task of real-time collection of patients' physiological and motor data. Its core objective is to construct a precise mapping relationship between patients and virtual avatars, allowing patients to clearly perceive the consistency between their own movements and those of the virtual avatar during rehabilitation training, thereby promoting the reshaping of body representation. The system consists of physiological signal acquisition devices and a movement posture monitoring module, which can accurately capture patients' movement intentions and internal sensations, providing core data support for Bayesian integration. This strengthens the sense of ownership and autonomy in both virtual and real worlds, driving real-time dynamic updates of the body model.

[0108] Furthermore, the embodied assessment system in this invention collects kinematic data (such as joint range of motion and limb position) and physiological signals (such as electromyography and electrodermal activity) in real time, which are not only used to drive the virtual avatar but can also be applied to the real body, providing the brain with a perceptual anchor point of "body existence." For example, the tactile module of the embodied stimulation system applies slight vibration stimulation to the patient's affected limb, while the electromyography module of the embodied assessment domain collects weak activation signals from the muscles of the affected limb. Both are fed back to the brain simultaneously: the tactile signal allows the brain to "perceive the existence of the affected limb," and the electromyography signal allows the brain to "confirm that the affected limb can respond to control." The dual signals work together to correct the brain's cognitive biases regarding the affected limb.

[0109] Optionally, the physiological signal acquisition device includes at least one module 28 of the following: heart rate monitoring module 22, electromyography monitoring module 23, electroencephalography monitoring module 24, electrodermal conductance monitoring module 25, eye movement monitoring module 26, micro-expression recognition module 27, and pressure and center of gravity monitoring module 28.

[0110] In some embodiments, the motion posture monitoring module 21 includes an inertial sensor array or motion capture device worn on the patient's body to acquire the patient's motion speed, acceleration, and position. It assesses the integrity of motor function and body spatial representation by quantifying the spatial characteristics and temporal patterns of body movement. As a specific approach, the inertial sensor array may consist of an accelerometer, a gyroscope, and a magnetometer. The accelerometer measures the acceleration of an object in three-dimensional space, the gyroscope detects the rotational motion of the object, and the magnetometer determines the orientation of the object. These three components work together to accurately capture every detail of the patient's limb movements. Another specific approach utilizes motion capture devices, such as optical motion capture systems. These systems use multiple cameras positioned around the training space to track markers on the patient's body, thereby acquiring high-precision body motion data.

[0111] The data acquired by these devices allows the motion posture monitoring module to quantify the spatial characteristics and temporal patterns of body movements. For example, when analyzing a patient's walking movements, it can calculate the changes in stride length, stride frequency, and joint range of motion over time. Based on this quantitative data, the system can accurately assess the patient's motor function, determine whether the patient's limb movements are coordinated, and identify any abnormal patterns. Simultaneously, it also helps assess the integrity of the body's spatial representation, understanding the patient's perception of their own body's position and posture in space.

[0112] In some embodiments, the motion posture monitoring module 21 is further configured to send the monitored motion speed, acceleration, position, joint range of motion, and postural symmetry to the central control module, enabling the patient to drive the virtual avatar in virtual reality using their own body. Based on this data, the central control module drives the virtual avatar in virtual reality to react in a manner highly consistent with the patient's body movements. For example, if the patient raises their arm in reality, the motion posture monitoring module quickly transmits the arm movement data to the central control module, which then controls the virtual avatar to synchronously complete the arm-raising action, achieving the goal of the patient driving the virtual avatar with their own body and enhancing the patient's immersion and interactivity in virtual training.

[0113] The motion posture monitoring module 21 sends kinematic data to the central control module in real time, driving the virtual avatar to precisely synchronize with the patient's body movements. This allows the patient to intuitively perceive the consistency between their own movements and the virtual movements, providing a motion data benchmark for establishing a sense of ownership (the virtual avatar is an extension of oneself). Simultaneously, the kinematic data, as the core input for Bayesian integration, is matched and analyzed with feedback data from the virtual environment: if the motion data matches the virtual avatar's movement data well, it strengthens the patient's sense of autonomy over the virtual avatar (they can control the virtual movements); if there is a discrepancy, it triggers real-time updates to the motion control parameters in the body model, correcting the completeness of the body's spatial representation and optimizing the accuracy of subsequent movement intention expression.

[0114] In some embodiments, the heart rate monitoring module 22 is used to reflect the patient's cardiovascular load and attention state, and may include an electronic heart rate monitor or a facial image acquisition device. An electronic heart rate monitor uses photoelectric sensors or electrode pads to detect electrical signals generated during heartbeats or changes in blood vessel volume, thereby accurately measuring heart rate; a facial image acquisition device, based on the principle of photoplethysmography (PPG), indirectly obtains heart rate information by analyzing changes in blood volume in blood vessels under the facial skin.

[0115] Optionally, the heart rate monitoring module's measurement range is set to 30-200 bpm, covering the normal resting and strenuous exercise heart rate ranges, with an error controlled within ≤5 bpm to ensure accurate heart rate measurement. In addition to real-time heart rate monitoring, the module also calculates heart rate variability (HRV), which reflects the regulatory function of the cardiac autonomic nervous system and serves as an important indicator for assessing a patient's mental stress and concentration. For example, when a patient is highly focused or under significant physical strain during rehabilitation training, their heart rate and HRV will change accordingly. This data is fed back to the central control module in real time, providing feedback on the patient's cognitive load and embodied status. The central control module can then adjust the training intensity and pace accordingly to prevent excessive patient fatigue and ensure training safety.

[0116] The heart rate monitoring module 22 assesses the intensity of embodiment by monitoring the patient's real-time heart rate changes. When the patient has a strong sense of ownership and autonomy over the virtual avatar, the heart rate will show stable fluctuations that are adapted to the training state. If the embodiment is weak (such as a low embodiment state), the heart rate is prone to abnormal fluctuations (too fast or too slow), which helps to judge the overall state of embodiment.

[0117] In some embodiments, the electromyography monitoring module 23 is used to reflect the patient's muscle activation and movement intention, and to enhance the sense of embodied control over the virtual avatar. It may include multiple electromyography electrodes worn on the patient's body to obtain surface electromyography signals of the target muscle area for rehabilitation, monitor muscle activation and the recovery of damaged muscles, identify movement intentions and evaluate muscle coordination patterns.

[0118] Optionally, the electromyography (EMG) monitoring module 23 has a sampling frequency ≥1000Hz, enabling it to quickly and accurately capture instantaneous changes in muscle electrical activity; a resolution ≥12bit ensures high precision of the acquired EMG signals, allowing it to distinguish subtle signal differences. Through analysis of surface EMG signals, the system can monitor muscle activation in real time, determine whether muscles contract and relax as expected, track the recovery of damaged muscles, and evaluate the effectiveness of rehabilitation training in improving muscle function. For example, during hand rehabilitation training, the EMG monitoring module can determine whether the patient has correctly performed the grasping action based on the EMG signals of the hand muscles and feed this information back to the central control module. The central control module then adjusts the virtual avatar's hand movements accordingly, providing the patient with more realistic training feedback and enhancing their sense of control over the virtual avatar.

[0119] After the electromyography (EMG) monitoring module 23 inputs EMG data into the central control module, it can enhance the sense of embodied control over the virtual avatar, allowing muscle exertion intentions to be directly converted into the movement intensity of the virtual avatar, thus improving the sense of autonomy. On the other hand, as a key modality of Bayesian integration, EMG data is matched with kinematic data and virtual movement feedback data: by comparing muscle activation intentions with the execution effect of virtual movements, the correlation accuracy of "muscle exertion-limb movement" in the body model is judged, and the model parameters are dynamically updated to optimize the conversion efficiency of movement intentions into movements. At the same time, the recovery process of damaged muscles is evaluated, providing a basis for adjusting the embodied rehabilitation program.

[0120] The electromyography (EMG) monitoring module 23 can be used to assess autonomy and position sense. By collecting the patient's EMG signals and comparing them with the EMG simulation signals of the virtual avatar's mapped movements, if the two have a high degree of overlap and a small time difference, it indicates that the patient has a strong sense of autonomy. At the same time, based on the intensity of the EMG signals, the accuracy of the patient's limb movements can be judged, which helps to assess the stability of the position sense.

[0121] In some embodiments, the EEG monitoring module 24 decodes brain neural activity, assesses central regulatory functions, drives neuro-machine interaction and promotes neural plasticity, and may include at least eight EEG electrodes worn on the patient's head with an impedance ≤5kΩ, and decodes motor intentions; the lower impedance can reduce attenuation and interference during signal transmission, ensuring that the acquired EEG signals are true and reliable.

[0122] When the brain engages in activities such as thinking and motor control, neurons generate electrical activity, which is transmitted through the scalp to form electroencephalogram (EEG) signals. The EEG signals collected by the EEG monitoring module contain various rhythmic components, such as alpha waves, beta waves, and gamma waves. Different rhythmic components are associated with different functional states of the brain. By analyzing the EEG signals, the system can understand the brain's neural activity patterns and assess the central nervous system's ability to regulate body movement. In rehabilitation training, the system decodes the patient's motor intentions based on EEG signals to achieve neuro-machine interaction. For example, when a patient intends to move their arm, the EEG monitoring module captures the corresponding neural signal changes and converts them into control commands, which are then sent to the central control module. The central control module then drives a virtual avatar to complete the arm movement. This neuro-machine interaction helps stimulate the brain's neuroplasticity and promotes the recovery of damaged nerve function.

[0123] The EEG monitoring module 24 can analyze the degree of event-related desynchronization (ERD) by collecting EEG signals. The higher the degree of ERD, the stronger the patient's sense of being. At the same time, it can distinguish the differences in the strength of possession, autonomy and position through EEG signal characteristics, providing core EEG data support for multi-dimensional embodiedness assessment.

[0124] In some embodiments, the skin conductance monitoring module 25 reflects the level of sympathetic arousal, assesses emotional state and training engagement, and enables the central control module to adjust the difficulty of the training task scenario based on the stress response level. This may include multiple skin conductance electrodes on the patient's hands or feet to obtain skin conductivity and verify the degree of embodiment and task participation. When a person's emotions change, sympathetic nerve excitation increases the activity of sweat glands in the skin, leading to increased sweat on the skin surface. Sweat contains electrolytes that can alter skin conductivity. Therefore, by measuring changes in skin conductivity, the patient's emotional state, such as tension, excitement, or anxiety, can be indirectly understood.

[0125] After skin conductance data is input into the Bayesian ensemble model, it is analyzed in conjunction with heart rate and kinematic data to determine the patient's stress response level. This drives the central control module to adjust the difficulty of training tasks, reducing task difficulty under high stress conditions to improve the comfort of the embodied experience, and maintaining task difficulty under moderate stress conditions to ensure training effectiveness. Simultaneously, based on the emotional sensations reflected in the skin conductance data, the correlation parameters between "environmental stimulus and emotional response" in the body model are dynamically updated to optimize the body's perceptual adaptation to the virtual environment.

[0126] The skin conductance monitoring module 25 can be used to assess the sense of ownership. When the virtual avatar is threatened by the virtual system, if the patient has a strong sense of ownership, it will trigger sympathetic nerve excitation, causing the skin conductance to spike instantly. By monitoring the amplitude of the skin conductance signal change, the intensity of the sense of ownership can be quantified.

[0127] In some embodiments, the eye-tracking monitoring module 26 tracks visual attention distribution, assesses spatial bias in body perception, and guides attention reshaping. It monitors eye movements and feeds them back to the central control module to adjust the viewing angle of the image displayed by the visual module. It may include an eye tracker, which can analyze the patient's spatial neglect through gaze heatmaps, combine training tasks with spatial exploration, and help reshape the body's position representation. The eye tracker uses technologies such as infrared cameras to track the patient's eye movement trajectory in real time, including information such as eyeball rotation, gaze point dwell position and time.

[0128] During rehabilitation training, a patient's visual attention distribution reflects their level of focus and interest in different elements in the virtual environment, and is also related to their spatial perception and cognition. For example, if a patient frequently stares at the ground during balance training, it may indicate a lack of confidence in their balance. The eye-tracking module feeds back the monitored eye-tracking data to the central control module, which adjusts the viewing angle of the image displayed by the visual module based on this data. For instance, when it detects that the patient's attention is focused on a certain area of ​​the virtual scene, the image is adjusted to make that area more prominent, guiding the patient's attention; or the viewing angle is changed to encourage the patient to focus on key information that helps improve balance, such as balance cues in the virtual environment, thereby helping the patient reshape correct visual attention patterns and spatial perception.

[0129] The eye-tracking module 26 can be used to assess position and autonomy. By tracking the eye focus and gaze trajectory, if the eye movement behavior matches the virtual avatar's actions and changes in the virtual environment's position, it indicates that the patient has a clear sense of position. At the same time, based on the eye movement response speed, it helps to determine the coordination between the movement intention under the control of autonomy and the virtual action.

[0130] In some embodiments, the micro-expression recognition module 27, which assesses subjective emotional experience and improves rehabilitation compliance, may include a facial image acquisition device for performing micro-expression recognition and feeding it back to the central control module. This device identifies micro-expressions representing a sense of body ownership and confirms that this ownership has been established. It also identifies moments of cognitive conflict in an unconscious state and uses this to predict task difficulty suitability, thereby adjusting the task difficulty. Micro-expressions are extremely short-lived facial expressions (typically between 0.04 and 0.25 seconds) that are difficult to control voluntarily, yet they can truly reflect an individual's inner emotional state. The facial image acquisition device captures subtle muscle movement changes in the patient's face using a high-resolution camera and employs computer vision and machine learning algorithms to recognize and analyze micro-expressions.

[0131] The micro-expression recognition module 27 can indirectly assess the embodied state by recognizing the patient's facial micro-expressions (such as focus, resistance, and pleasure). If the micro-expression shows focus and pleasure, it indicates a good embodied state; if resistance and irritability are present, it corresponds to a low embodied state, thus supplementing the embodied state assessment with subjective physiological feedback data.

[0132] In some embodiments, the pressure and center of gravity monitoring module 28 assesses balance function and body weight distribution, enhances the body's contact perception with the ground, and promotes the reshaping of balance representation. This includes weight sensors disposed inside or below the multi-degree-of-freedom balance platform 31, quantifying stability during embodied interaction. When the patient performs training movements such as standing and walking, the weight sensors monitor the pressure distribution applied by the patient to the balance platform in real time, and analyze the patient's center of gravity movement trajectory by calculating the position and changes of the pressure center.

[0133] After the pressure and center of gravity monitoring module 28 inputs data into the central control module, it can optimize the gravity feedback parameters in the virtual environment, allowing the patient's balance intentions to be directly translated into the balance posture of the virtual avatar, thus enhancing their sense of autonomy. Simultaneously, the pressure and center of gravity data, as Bayesian integrated balance perception dimension data, are matched with vestibular and kinematic data to determine the accuracy of the "center of gravity adjustment-limb balance" correlation in the body model, driving real-time updates to model parameters, reshaping the patient's balance representation, and improving the recovery effect of balance function in embodied rehabilitation.

[0134] This module helps patients enhance their sense of contact with the ground. For example, when a patient's center of gravity shifts, the system can use vibration feedback and voice prompts to help the patient perceive their imbalance, prompting them to adjust their posture and restore balance. Simultaneously, based on assessment data of balance function and weight distribution, the system can develop targeted training programs to gradually improve the patient's balance ability and promote the reshaping of balance representation. For patients with balance disorders, repeated training helps them re-establish correct balance sensation and body control patterns, improving their mobility and safety in real-life situations.

[0135] The pressure and center of gravity monitoring module 28 can be used to assist in the assessment of position and autonomy. By monitoring the patient's body pressure distribution and center of gravity changes, it can determine the patient's perception of the positional relationship between themselves and the virtual avatar. At the same time, based on the speed of center of gravity adjustment, it can assess the control ability of limb movements under the dominance of autonomy, and help improve the embodied sense assessment system.

[0136] In this embodiment of the invention, various assistive movement devices in the device domain system provide support for patient rehabilitation training through different mechanisms of action. The assistive movement devices include at least one of the following: an exoskeleton robot 33, a walking aid 34, a weight reduction device 35, a multi-degree-of-freedom balance platform 31, and a vestibular electrical stimulation device. These devices are used to assist patients in strengthening the implementation of their movement intentions or to assist the functions of the virtual domain and embodied assessment system, forming an intermediary between the patient and the environment to better achieve embodied interaction. These devices provide support for the patient's rehabilitation training from different dimensions. Through collaboration with the virtual environment and adaptation to the patient's physiological state, a comprehensive and personalized rehabilitation training system is constructed to help patients reshape their movement patterns and physical representations, thereby improving the effectiveness of rehabilitation training.

[0137] In some embodiments, the exoskeleton robot 33 is worn on the patient's body, closely conforming to the patient's body, to provide assistance or damping to the patient's rehabilitation area, helping the patient complete active or passive motor training and reshape movement patterns. In active motor training scenarios, when the patient expresses a movement intention, the robot understands the patient's movement intention and provides certain movement guidance. As the patient's motor ability improves, the guidance intensity is gradually reduced to promote body schema mastery of body control. For example, when attempting to raise an arm or take a step, the exoskeleton robot's built-in sensors capture weak movement signals and muscle electrical activity signals of the patient's limbs in real time. Based on these signals, the robot's power system provides corresponding assistance to help the patient complete the movement more easily. For example, for patients whose upper limb strength is weakened due to stroke, the exoskeleton robot 33 can assist them in completing arm extension and grasping movements, helping the patient gradually regain the voluntary movement ability of the limbs and strengthen the memory of correct movement patterns.

[0138] In passive movement training, the exoskeleton robot 33 acts as a "guide." Doctors or rehabilitation therapists can pre-program the movement, and the robot guides the patient's limbs according to the program, moving them at specific trajectories, speeds, and forces. During this process, the robot can also apply damping according to training needs, simulating resistance in a real environment to help patients strengthen their muscles and improve joint mobility. For example, during lower limb rehabilitation training, by setting appropriate damping, patients can feel resistance when lifting their legs, effectively exercising leg muscles, gradually reshaping normal movement patterns, and laying the foundation for patients to return to daily life.

[0139] The exoskeleton robot 33 assists patients in completing limb movements by conforming to the mechanical structure of their limbs, enhancing their sense of autonomy and ownership. At the same time, through movement constraints and guidance, it helps patients perceive the trajectory of limb movements and improve their sense of position. In terms of feedback adjustment, based on real-time embodiedness data (such as E(t) values), if the patient's embodiedness is weak (E(t) < 0.4), the assistance intensity is reduced and the movement guidance speed is slowed down to avoid excessive assistance weakening the sense of autonomy. If the embodiedness is sufficient (E(t) > 0.8), the assistance intensity is reduced to strengthen the patient's own movement intention and help restore autonomous movement ability.

[0140] In some embodiments, the walking aid 34 provides walking support to the patient based on their motor abilities, helping patients with balance disorders or lower limb muscle weakness to stand and walk safely. It works in conjunction with a virtual domain to provide similar walking posture feedback, helping the patient perceive their environment and gradually adapt to the user's gait rhythm. Optionally, the walking aid 34 is equipped with pressure sensors and posture detection devices, enabling it to sense the pressure distribution and changes in body posture applied by the patient in real time. When the patient stands or walks, the walking aid 34 automatically adjusts its support force to ensure the patient remains stable and avoids the risk of falls. For example, for patients with decreased muscle strength due to leg injuries, the walking aid 34 can share some of their body weight, allowing them to safely practice standing and walking.

[0141] Simultaneously, the walking aid 34 works closely with the virtual domain system. When the patient walks with the assistance of the walking aid 34, the motion sensors built into the walking aid 34 transmit data such as the patient's walking speed, stride length, and cadence to the virtual domain system. Based on this data, the virtual domain system generates virtual avatar motion feedback in the virtual environment that matches the patient's actual walking posture. During training, the patient can not only feel the support and protection of the walking aid in reality, but also intuitively see their own walking posture in the virtual environment. Through visual feedback, they can further adjust and correct their movements, improving the accuracy and effectiveness of rehabilitation training.

[0142] The walking aid 34 provides support to help patients maintain balance, helps them perceive the position of their center of gravity, strengthens their sense of position, and assists them in completing rehabilitation movements such as walking, thus enhancing their sense of autonomy. In terms of feedback adjustment, based on real-time embodied feedback, if the patient's sense of position is insufficient, the support height and center of gravity distribution can be adjusted to help the patient establish spatial position cognition; if the embodied sense is stable, the support force can be gradually reduced to guide the patient to control their center of gravity independently and improve the stability of the embodied sense.

[0143] In some embodiments, the weight-reducing device 35 is used to provide a set suspension force to the patient based on their mobility, reshaping the standing and movement posture of patients with lower limb muscle weakness and providing dynamic gravity environment simulation adjustment. Optionally, the weight-reducing device 35 may consist of a suspension system, a sling, and a control system. Doctors or rehabilitation therapists can precisely set the magnitude of the suspension force in the control system according to the patient's physical condition and rehabilitation stage. For example, for patients who have just begun standing training, a larger suspension force can be set so that the patient's lower limbs only need to bear less weight, helping the patient gradually adapt to the standing posture and enhance leg muscle strength and balance.

[0144] As the patient's recovery progresses, the suspension force is gradually reduced, guiding the patient to rely more on their own strength to complete standing and movement movements. During the training process, the weight-reduction device 35 is linked with the virtual domain system. The patient's movement data under weight-reduction conditions is synchronously fed back into the virtual environment. The virtual avatar simulates the patient's movement posture. By observing the virtual avatar's movements and combining them with their own feelings, the patient can adjust their body posture, thereby reshaping their standing and movement posture and preparing them for the restoration of normal walking function.

[0145] The weight-reduction device 35 can reduce the load on the patient's limbs through precise weight reduction, helping the patient to easily complete rehabilitation movements and enhance their sense of autonomy. At the same time, it assists the patient in perceiving the range of limb movements and improves their sense of position in conjunction with the virtual domain system. In terms of feedback adjustment, the weight reduction intensity is dynamically adjusted based on real-time embodied assessment results. When the embodied sense is weak, the weight reduction intensity is increased to reduce the difficulty of the movements and strengthen the embodied sense. As the embodied sense improves, the weight reduction intensity is gradually reduced to gradually strengthen the patient's own exercise load and adapt to the rehabilitation progress.

[0146] In some embodiments, the treadmill 36 is used to provide patients with walking training adapted to their walking speed and motor abilities, creating an environmental angle identical to that of a virtual reality environment and monitoring walking speed, providing a controllable space and facilitating a transition from rehabilitation to real-life scenarios. Optionally, the treadmill 36 is equipped with a high-precision speed adjustment system and an incline adjustment system, which can create an environmental angle identical to that of a virtual reality environment according to the patient's rehabilitation needs, such as simulating different terrains like flat ground, uphill, and downhill. Simultaneously, the sensors built into the treadmill 36 monitor the patient's walking speed, cadence, stride length, and other data in real time, and feed this data back to the central control module.

[0147] The central control module adjusts scene elements in the virtual environment based on data, ensuring a high degree of synchronization between the patient's movement on the treadmill 36 and the virtual experience. For example, when the patient increases their walking speed on the treadmill 36, the scene in the virtual environment will correspondingly produce a visual effect of rapid movement, accompanied by sound effects such as wind sounds to enhance immersion. This walking training within a controlled space not only helps patients improve their walking ability and endurance but also enhances their training motivation and enjoyment through interaction with the virtual environment, allowing them to achieve their rehabilitation goals in a training environment that more closely resembles real-life scenarios.

[0148] In some embodiments, the multi-degree-of-freedom balance platform 31 is used to drive the patient to perform pitching, rolling, or yaw movements at a set angular velocity and acceleration, or to perform vertical or horizontal planar movements at a set linear velocity and acceleration, in order to simulate different vestibular sensory information and transform the patient's abstract balance movements into embodied goals in a scenario. The vestibular electrical stimulation device is worn on the patient's vestibular organs and can apply corresponding vestibular sensory stimulation to reshape body representation. For example, in a simulated rocking scenario while riding a boat, the multi-degree-of-freedom balance platform 31, through rolling and pitching movements, allows the patient to experience a feeling similar to a boat rising and falling in waves, stimulating the patient's vestibular organs and prompting the patient to adjust their body posture to maintain balance.

[0149] During training, patients need to constantly perceive and adapt to the movement changes of the balance platform, thereby improving their balance perception and body posture control. Simultaneously, the multi-degree-of-freedom balance platform 31 is integrated with a virtual domain system. The virtual environment synchronously presents corresponding scene changes based on the balance platform's movement. For example, in simulated airplane takeoff training, the balance platform undergoes ascent, descent, and tilting movements, while the virtual environment displays the airplane accelerating and taking off on the runway. This allows patients to more effectively perform balance rehabilitation training under dual stimulation of vision and vestibular sense, reshaping correct body balance representations.

[0150] The multi-degree-of-freedom balance platform 31 can help patients perceive changes in their body's center of gravity and enhance their sense of position and balance by simulating posture adjustments at different angles and amplitudes. At the same time, it can enhance patients' spatial position cognition in a virtual environment by working with virtual domain scenarios. In terms of feedback adjustment, based on real-time embodied data, if the patient's sense of position is insufficient or their balance ability is weak, the amplitude and speed of the balance platform's movement can be reduced to help the patient establish a stable sense of position. If the embodied sense is good, the complexity of the movement can be increased to enhance the patient's autonomous balance control ability and further improve their sense of position.

[0151] In some embodiments, the vestibular electrical stimulation device 32 is worn on the vestibular organ of a patient and can apply corresponding vestibular stimulation to regulate balance perception and body posture control, helping sensory integration and reshaping body representation. The vestibular organ is an important organ for the human body to perceive changes in head position and movement. When the vestibular organ is impaired, it can lead to symptoms such as balance disorders and dizziness. The vestibular electrical stimulation device stimulates receptors in the vestibular organ with a weak current, activating relevant neural pathways and affecting the transmission and processing of nerve signals.

[0152] This electrical stimulation can help patients re-establish correct balance perception and improve their postural control. For example, for patients with vestibular dysfunction due to inner ear disease, a vestibular electrical stimulation device can promote the recovery and compensation of vestibular function through regular electrical stimulation, allowing patients to gradually adapt to and correct their abnormal balance during training. Simultaneously, this device works in conjunction with other rehabilitation equipment and systems, adjusting electrical stimulation parameters such as current intensity and frequency based on the patient's rehabilitation progress and physical response to achieve personalized rehabilitation intervention. This further assists in the reshaping of the patient's physical manifestations and improves their self-care ability and motor function.

[0153] The vestibular electrical stimulation device 32 can assist patients in establishing spatial orientation cognition and strengthening their sense of position through vestibular stimulation. At the same time, it can link with the visual stimulation of the virtual domain system to enhance the realism of embodied sensation. In terms of feedback regulation, based on real-time embodied sensation feedback, the intensity and frequency of vestibular electrical stimulation are dynamically adjusted. When embodied sensation is weak, the stimulation intensity is reduced to avoid overstimulation interfering with perception. When embodied sensation is stable, the stimulation intensity is appropriately increased to further strengthen the sense of position and help patients establish a stable spatial perception system.

[0154] In some embodiments, the central control module undertakes the task of decoding and converting information between the virtual domain, the embodied stimulation system, the embodied assessment system, and the device domain, constructing a closed-loop control system that links the four domains. The central control module receives motion posture data and physiological signals from the embodied assessment system, as well as operating status parameters from the device domain. After parsing and processing, it generates control commands for the virtual domain and the embodied stimulation system, enabling real-time interaction and precise mapping between the patient's body and the virtual reality environment and virtual avatar. This cross-domain collaborative capability ensures the consistency of actions and the synchronization of information among the various system modules, creating a highly realistic and immersive rehabilitation experience for the patient.

[0155] To create the illusion of physical possession for the patient, the central control module achieves this by precisely regulating the coordinated operation of the virtual domain and the embodied sensory stimulation system. The virtual domain system first generates a virtual avatar hand model that highly matches the patient's hand features (such as size, shape, and skin color). During interaction, the central control module strictly controls the visual-tactile synchronization parameters between the virtual avatar and the patient's real body, keeping the delay error within ≤50ms, far below the threshold of human perception delay.

[0156] When the system simulates a virtual body receiving tactile stimulation (such as a virtual object touching a virtual avatar's hand), the central control module immediately sends instructions to the tactile feedback module of the embodied stimulation system. This instructs the tactile device (such as a wearable tactile glove or patch) to apply tactile stimulation (including touch intensity, frequency, and range) to the corresponding location on the patient's real hand, consistent with the pattern observed in the virtual scene. This precise synchronization of visual information and tactile feedback allows the patient's brain to achieve cognitive integration, perceiving the virtual hand as part of their own body, thus creating a strong illusion of bodily possession.

[0157] In cognitive rehabilitation training, the central control module utilizes the illusion of body ownership to reshape the patient's body cognition by gradually adjusting virtual body parameters. Specifically, based on the patient's established sense of ownership of the virtual hand, the system slowly changes the size (e.g., gradually restoring it to normal proportions from a 10% reduction), shape (e.g., correcting abnormal finger proportions), or position parameters (e.g., adjusting the relative spatial position of the virtual hand and body) of the virtual hand according to a preset rehabilitation plan.

[0158] Because patients perceive the virtual hand as part of themselves, this gradual parameter adjustment is perceived by the brain as a change in their own body, subtly correcting the patient's original body image biases. For example, for patients with unilateral limb neglect due to stroke, the system can gradually bring the virtual avatar's affected hand into visual focus and simultaneously provide tactile feedback, helping patients to re-recognize the existence of the affected limb. For patients with body shape recognition impairment, adjusting the virtual avatar's proportional parameters can help them establish correct body spatial perception, ultimately improving body cognition or motor function impairments and enhancing rehabilitation outcomes.

[0159] The virtual reality embodied rehabilitation system according to embodiments of the present invention can achieve at least the following beneficial effects: (1) This invention uses three sub-modules of the virtual domain system to specifically induce a sense of ownership, strengthen a sense of autonomy, and optimize a sense of location. It combines the synergistic stimulation of multiple sensory modules such as temperature, vision, touch, and hearing to break the limitations of a single sensory experience. At the same time, it is suitable for special patients such as those with atrophied limbs and those wearing prostheses. It adopts a virtual avatar design with core feature matching and functional optimization to make the embodied feeling construction more in line with the patient's physical characteristics and rehabilitation needs, laying a solid foundation for rehabilitation training.

[0160] (2) This invention relies on the embodied assessment system to achieve multi-dimensional assessment through subjective questionnaires, physiological signal collection, behavioral data and other paths. Among them, the physiological signal collection modules can accurately capture objective data such as heart rate, electromyography, and electroencephalography. Combined with quantitative formulas, the embodied feeling that is difficult to capture is transformed into a quantifiable E(t) index, clarifying the criteria for judging the embodied state, providing accurate and reliable data support for subsequent dynamic regulation, and solving the pain point of traditional assessment relying on subjective judgment.

[0161] (3) Based on the real-time embodied assessment results, the central control module of the present invention links various systems to dynamically adjust the difficulty of rehabilitation tasks, sensory stimulation parameters, auxiliary movement equipment parameters and virtual avatar mapping actions, forming a complete closed loop of "embodied application-assessment-feedback-adjustment". It can be flexibly adapted according to different rehabilitation tasks, patient training stages and embodied states (high / medium / low), taking into account the individual differences of different diseases and different patients.

[0162] (4) The virtual domain system of the present invention constructs a virtual foundation, the sensory module enhances the embodied experience, the evaluation system provides data support, the auxiliary sports equipment builds a virtual-real interaction medium, and the central control module coordinates and schedules, breaking the limitation of independent operation of each module, realizing the deep integration of virtual scene, real body feeling, sports assistance, evaluation and control, and improving the embodied interactive experience.

[0163] (5) This invention reduces the boredom of traditional rehabilitation training by using immersive virtual scenes and multi-sensory synergistic stimulation; it helps patients complete movements and amplifies the effects of exercise by using assistive movement equipment, and combines the immediate sense of accomplishment brought by the improvement of embodiedness to alleviate patients' resistance and anxiety, reduce rehabilitation interruptions caused by psychological factors or movement difficulty, and ensure long-term training compliance.

[0164] The virtual reality embodied rehabilitation system of the present invention can be adapted to patients with various motor dysfunctions such as stroke, spinal cord injury, and prosthesis wearing. It is compatible with various assistive movement devices and multiple types of rehabilitation tasks. The virtual avatar design, sensory stimulation mode, and assessment parameters can all be flexibly adjusted. At the same time, it takes into account the needs of different rehabilitation stages. It has strong clinical applicability and has broad promotion value.

[0165] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0166] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0167] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0168] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A virtual reality embodied rehabilitation system, characterized in that, include: A virtual domain system is used to generate or invoke virtual avatars based on rehabilitation tasks. These virtual avatars are created based on characteristics of the patient's real body to give the patient a sense of ownership over the virtual avatar and / or their real body. The virtual avatars are also configured to map the patient's limb movements in real time to give the patient a sense of autonomy over the virtual avatar and / or their real body. The system is also used to generate a virtual environment that matches the rehabilitation task, within which the virtual avatars are placed to give the patient a sense of location over the virtual avatar and / or their real body. An embodied stimulation system, by applying visual stimulation that includes at least a virtual avatar, and at least one of touch, hearing, smell and vestibular stimulation, enables the patient to experience embodiedness during the rehabilitation process through stimulation of two or more senses, said embodiedness including at least one of said sense of possession, autonomy and position; An embodied assessment system obtains real-time embodied feelings by asking patients questions during the execution of rehabilitation movements; or by collecting patients' physiological indicators, including at least one of heart rate, electromyography, electroencephalography, conductance of skin, eye movement, micro-expression, and stress; or by collecting patients' kinematic data. The device domain system includes several assistive motion devices used to assist patients in reinforcing their movement intentions, or to assist the sensory stimulation applied by the somatosensory stimulation system, or to assist the assessment process of the somatosensory assessment system, so as to form a somatosensory interaction medium between the patient and the environment. The central control module is used at least to adjust the difficulty of the rehabilitation task based on the real-time embodied sensation, or to adjust the type or intensity of sensory stimulation applied by the embodied stimulation system, or to adjust the controllable parameters of the assistive movement device of the device domain system, or to adjust the mapped actions of the virtual avatar.

2. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The embodiedness assessment system includes an embodiedness integration module, which obtains the real-time embodiedness using the following formula: in, Indicates real-time embody feeling, with a value between 0 and 1; Indicates a sense of ownership. Indicates the sense of autonomy factor, This represents the positional perception factor, with values ​​ranging from 0 to 1. , , These represent dynamic weighting coefficients, with values ​​ranging from 0 to 1, which are adjusted according to different rehabilitation tasks and different training states. like A value greater than 0.8 indicates a high state of self-containment; if If the value is less than 0.4, the patient is in a low embodied state.

3. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The virtual domain system includes: The device has a sensory induction module, which is used to generate corresponding virtual limbs on the virtual body based on the patient's own limbs. The autonomous sense-evoking module is used to amplify the patient's small-amplitude limb movements, so that the limb movements mapped by the virtual avatar are displayed at a normal amplitude. The positional enhancement module allows patients to choose between first-person or third-person perspectives to view the virtual scene or virtual avatar.

4. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The virtual domain system includes a virtual environment database, a virtual avatar generation module, and an action mapping unit; The virtual environment database stores contextualized task scenarios corresponding to different stages of rehabilitation. The virtual avatar generation module can generate personalized virtual avatars based on the patient's gender, height, limb length, and joint range of motion parameters. The motion mapping unit uses an inverse kinematics algorithm to convert the kinematic data collected by the embodied assessment system into synchronized movements of the virtual avatar, or to amplify and map movements that the patient cannot perform in the virtual environment, with a motion mapping delay of no more than 50ms.

5. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The embodied stimulation system includes a visual module, an auditory module, and at least one of a tactile module, an olfactory module, a vestibular module, a temperature regulation module, a humidity regulation module, a wind speed regulation module, a light regulation module, and a gas partial pressure regulation module. It is used to form multi-sensory channel stimulation in vision, touch, hearing, smell, and vestibular sense in a virtual environment with physical properties, and to simulate environmental factors such as temperature, humidity, wind speed, light, and gas. It also reshapes bodily representations through multi-sensory channel stimulation. The visual module assists the embodied rehabilitation system in constructing a spatial framework for body representation, including a spherical display screen or a giant screen set within the rehabilitation system to surround the patient; the visual module also displays the corresponding virtual reality scene and perspective based on the simulated scene required for virtual reality rehabilitation and the patient's current position and angle; the visual module also includes one or more of proprioception, mirror vision, and motion vision. The tactile module is used to construct body structure representation and surface state representation, including one or more of the following: vibrating pads, pneumatic / hydraulic tactile devices, microneedle array texture tactile simulation devices, or other wearable tactile devices disposed within the embodied rehabilitation system; the tactile module is also based on the simulated scene required for virtual reality rehabilitation and the surface features of objects that the patient is currently in contact with or simulated touch; the tactile module also includes one or more of the following stimulation types: passive touch, active touch, and pain. The auditory module is used to indirectly adjust body position perception in the virtual reality environment through sound, including a speaker array set in different positions within the embodied rehabilitation system; the auditory module includes sounds associated with one's own movements and ambient sounds around the location of the virtual avatar in the environment; The olfactory module, used to connect to the limbic system to regulate bodily state representation, includes odor release mechanisms, fragrance storage units, and odor concentration detection units. Multiple odor release mechanisms are distributed at different locations within the embodied rehabilitation system. The olfactory module includes both personal odors and environmental odors. The fragrance storage unit stores liquid perfume or scent capsules and is connected to each odor release mechanism to provide odors. Each fragrance storage unit corresponds one-to-one with an odor release mechanism, or one fragrance storage unit can provide odors to two or more odor release mechanisms. Different fragrance storage units can store the same or different odors. At least one odor concentration detection unit is provided to monitor the odor concentration within the embodied rehabilitation system. The vestibular module is used to simulate the virtual gravity field in a virtual reality scene and enable the patient to perform orientation representation of body space in the virtual reality scene, including at least a multi-degree-of-freedom balance table and a vestibular electrical stimulation device. The vestibular module includes at least one type of stimulation, namely head movement stimulation and gravity change stimulation. The temperature regulation module is used to define the hot and cold surface states in the virtual reality rehabilitation environment, thereby indirectly affecting the internal body manifestation state. It includes a cooling and heating device installed in the rehabilitation system to regulate the ambient temperature inside the cabin from 0°C to 45°C. The humidity regulation module indirectly adjusts the body surface condition by affecting the water evaporation rate. It includes at least one humidifier installed in the embodied rehabilitation system and a ventilation device connecting the inside and outside of the cabin, for regulating the ambient humidity in the cabin within a range of 30%-90%. The wind speed adjustment module provides a dynamic reference for the patient's body position, including at least one air blowing device located in front of or to the side of the patient within the embodied rehabilitation system, and can adjust the air volume according to the walking speed feedback and ambient wind speed sent by the central control module. The light regulation module is used to adjust the circadian rhythm and emotional state of the body state, including a lighting array set on the top of the embodied rehabilitation system for adjusting the light intensity and light color; The gas partial pressure regulation module is used to adjust the concentration of various gases required for the virtual reality environment or to directly affect the rehabilitation treatment itself. It includes a pressure monitoring and regulation device, a gas supply device, and a gas concentration monitoring device within the embodied rehabilitation system. The pressure monitoring and regulation device monitors the total pressure inside the chamber and controls the safety valve to open when the total pressure exceeds or falls below the safe pressure range, thus connecting the inside and outside of the chamber. The gas supply device includes at least one gas source cylinder containing oxygen, hydrogen, carbon dioxide, or a mixture thereof. The gas concentration monitoring device includes at least one of an oxygen concentration sensor, a hydrogen concentration sensor, and a carbon dioxide concentration sensor, used to monitor the real-time concentration of the target gas.

6. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The embodied assessment system collects kinematic data and physiological signals in real time to construct a mapping between the patient and the virtual avatar, enabling the patient to perceive the consistency between their own movements and virtual movements and promoting the reshaping of body representation. The embodied assessment system includes a motion posture monitoring module and a physiological signal acquisition device, which includes at least one of the following modules: heart rate monitoring module, electromyography monitoring module, electroencephalography monitoring module, electrodermal conductance monitoring module, eye movement monitoring module, micro-expression recognition module, and pressure and center of gravity monitoring module. The motion posture monitoring module includes an inertial sensor group or motion capture device worn on the patient's body to obtain the patient's motion speed, acceleration and position, and to evaluate the integrity of motor function and body spatial representation by quantifying the spatial characteristics and temporal patterns of body movement. The motion posture monitoring module is also used to send the monitored motion speed, acceleration and position, joint range of motion and posture symmetry to the central control module, so that the patient can use his own body to drive the virtual avatar in virtual reality. The heart rate monitoring module is used to reflect the patient's cardiovascular load and attention status. It includes an electronic heart rate monitor or a facial image acquisition device with a measurement range of 30-200 bpm and an error of ≤5 bpm. It monitors the patient's heart rate and heart rate variability and feeds it back to the central control module, thereby providing feedback on the patient's cognitive load and embodiedness level. The electromyography (EMG) monitoring module is used to reflect the patient's muscle activation and movement intention, and enhance the sense of embodied control over the virtual avatar. It includes multiple EMG electrodes worn on the patient's body, with a sampling frequency of ≥1000Hz and a resolution of ≥12bit, to obtain surface EMG of the target muscle area for rehabilitation, monitor muscle activation and recovery of damaged muscles, identify movement intention and evaluate muscle coordination patterns. The EEG monitoring module decodes brain neural activity, assesses central regulatory functions, drives neuro-machine interaction and promotes neural plasticity, including at least eight EEG electrodes worn on the patient's head with an impedance ≤5kΩ, and decodes motor intentions. The skin conductance monitoring module reflects the level of sympathetic nerve arousal, assesses emotional state and training engagement, and enables the central control module to adjust the difficulty of the training task scenario based on the stress response level, including multiple skin conductance electrodes on the patient's hands or feet to obtain skin conductivity and verify the degree of embodiment and task participation. The eye-tracking monitoring module tracks visual attention distribution, assesses spatial bias in body perception, and guides attention reshaping. It monitors eye movements and feeds them back to the central control module to adjust the viewing angle of the image displayed by the visual module, including the eye tracker. By analyzing the patient's spatial neglect through fixation heatmaps, the training task is combined with spatial exploration to help reshape body position representation. The micro-expression recognition module assesses subjective emotional experience and improves rehabilitation compliance. It includes a facial image acquisition device for micro-expression recognition and feedback to the central control module. It identifies micro-expressions representing body ownership and confirms that body ownership has been established. It identifies moments of cognitive conflict in an unconscious state and uses this to predict task difficulty fit, thereby adjusting the task difficulty. The pressure and center of gravity monitoring module assesses balance function and body weight distribution, enhances the body's contact perception with the ground, and promotes the reshaping of balance representation. This includes weight sensors installed inside or below the multi-degree-of-freedom balance platform to quantify stability during embodied interaction.

7. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The assistive movement device includes at least one of the following: a multi-exoskeleton robot, a walking aid, a weight reduction device, a treadmill, a multi-degree-of-freedom balance platform, and a vestibular electrical stimulation device, used to assist patients in strengthening the implementation of movement intentions or to assist in the functions of the virtual domain and the embodied assessment domain. The exoskeleton robot is worn on the patient's body to provide assistance or damping to the patient's rehabilitation area, help the patient complete active or passive movement training, reshape movement patterns, understand the patient's movement intentions and provide certain movement guidance. As the patient's movement ability improves, the guidance intensity is gradually reduced to promote the body schema to master body control. The walking aid is used to provide walking support to patients based on their motor ability, helping patients with balance dysfunction or lower limb muscle weakness to stand and walk safely. It works in conjunction with the virtual domain to provide the same walking posture feedback, helping patients perceive the environment and gradually adapt to the user's gait rhythm. The weight reduction device is used to provide a set suspension force to the patient based on the patient's mobility, to reshape the standing and movement posture of patients with insufficient lower limb muscle strength, and to provide dynamic gravity environment simulation adjustment; The treadmill is used to provide patients with walking training that adapts to their walking speed and motor ability, creates the same environmental angles as the virtual reality environment and monitors walking speed, provides a controllable space, and provides a transition from rehabilitation to real-life scenarios. The multi-degree-of-freedom balance platform is used to drive the patient to perform pitching, rolling, or yaw movements with a set angular velocity and a set angular acceleration, or to perform vertical or horizontal planar movements with a set linear velocity and a set linear acceleration, in order to simulate different vestibular sensory information and transform the patient's abstract balance movements into embodied goals in the scene; the vestibular electrical stimulation device is worn on the patient's vestibular organ and can apply corresponding vestibular sensory stimulation to it, reshaping the body representation; The vestibular electrical stimulation device is worn on the vestibular organ of the patient and can apply corresponding vestibular stimulation to regulate balance perception and body posture control, help sensory integration and reshape body representation.

8. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The central control module is used to decode the virtual domain, embodied stimulation domain, embodied assessment domain, and device domain, and to realize the interaction and mapping between the patient's body and virtual reality. Through visual-tactile synchronous stimulation, it enables the patient to have the illusion of body ownership, which is applied to cognitive rehabilitation. The virtual domain system generates a virtual avatar that matches the characteristics of the patient's hand. By adjusting the visual-tactile synchronization parameters between the virtual avatar and the patient's real body, the delay error is ≤50ms. When the patient observes that the virtual body is touched, the tactile feedback module of the embodied stimulation system synchronously applies the same pattern of tactile stimulation to the corresponding part of the patient's real body, making the patient have the illusion that the virtual body belongs to them. In cognitive rehabilitation training, by gradually changing the size, shape, or position parameters of the virtual body, it helps patients reshape hand cognition and improve body image deviation in patients with body cognition or motor dysfunction.

9. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The central control module is used to realize embodied decision-making based on the prediction of the fusion model. It constructs and updates an scalable body schema model through dynamically fed-back embodied assessment domain signal data, including the intrinsic representation matrix of the position, relationship and movement probability of various parts of the patient's body. The central control module is also used to implement a multimodal signal fusion control method, including: spatiotemporal registration of EEG signals, EMG signals and kinematic data collected by the embodied assessment system and the device domain system; generating a comprehensive motor state assessment result through a weighted fusion algorithm; identifying the patient's movement intention and generating an internal dynamic model based on the body; analyzing the accuracy and completion of the movement intention and actual movement and generating a comprehensive motor state assessment result; when the accuracy of the movement task is less than 80% or the completion of the movement is less than 60%, the central control module sends a signal and manipulates the virtual domain system to display the correct task completion method, and manipulates the device domain system to guide the user's body to discover and consolidate the correct movement pattern rather than passively execute it; The central control module can also dynamically adjust the difficulty parameters based on the patient's real-time performance, including the size of the virtual target, resistance, intensity of the virtual gravity field, and task time limit.

10. The virtual reality embodied rehabilitation system according to claim 1, characterized in that, The central control module is used to realize the embodied control of the fusion model prediction, including an EEG signal parsing unit and a control command generation unit; The EEG signal analysis unit can extract features from the EEG signals collected by the embodied assessment system, and identify EEG features such as event-related desynchronization potentials corresponding to motor intentions; the control command generation unit can convert the identified motor intentions into motor parameters for the assistive movement device, including movement direction, speed, torque, and action parameters for the virtual avatar, including joint angles and range of motion, with a command generation delay of ≤200ms; the virtual domain system allows the patient's virtual avatar to automatically and smoothly demonstrate a correct task completion action, and the device domain system provides biomechanically minimized assistive force to guide the patient's real limbs to follow the trajectory of the virtual avatar; gradually reducing the virtual weight of the device domain, promoting the transformation of the device domain from external assistance to expansion outside the body; The EEG signal analysis unit can also analyze EEG features related to virtual hand illusion, whole-body illusion, sense of body ownership, and body agency, providing a basis for evaluating their effectiveness.

Citation Information

Patent Citations

  • VR rehabilitation system based on mirror neuron

    CN107433021A

  • Rehabilitation system and method based on interactive virtual reality and electrical nerve stimulation

    CN107626040A

  • Immersive upper limb rehabilitation system

    CN113101612A

  • Hand rehabilitation exoskeleton control system with multi-feeling synergistic stimulation

    CN117426948A

  • Method, device and system for evaluating virtual avatar in virtual reality rehabilitation scene

    CN118919017A

Cited By

  • Rehabilitation simulation method and system based on virtual other and heuristic state machine

    CN122314231A