VR chronic neck pain rehabilitation training system based on CPPC technology

Abstract

The application discloses a VR chronic neck pain rehabilitation training system based on a CPPC technology, relates to the technical field of intelligent medical treatment, and solves the problem that the existing rehabilitation system combined with VR cannot effectively help in improving the neck posture muscle control ability of a patient; the application comprises a data acquisition component used for collecting action posture data of the patient in an evaluation period and a training period; a data processing module used for analyzing and processing the action posture data to obtain analysis results; a control center used for pushing the analysis results to a medical treatment end, and then receiving guidance opinions and adjustment opinions from the medical treatment end; a plan making module used for generating a training plan according to the guidance opinions and adjusting the training plan according to the adjustment opinions; a VR device used for building a virtual scene according to the training plan and guiding the patient to execute the training plan in the virtual scene; and a database used for storing data related to rehabilitation training; and the application fundamentally solves the pain caused by the decline of the neck muscle posture control ability of the patient.

Description

Technical Field

[0001] This invention relates to the field of smart medical technology, specifically to a VR rehabilitation training system for chronic neck pain based on CPC technology. Background Technology

[0002] Chronic neck pain is a prevalent global health problem, with a reported prevalence of 10% to 24%. Studies indicate that the primary cause of chronic neck pain is decreased neck muscle strength and postural control, leading to cervical segmental instability. This results in additional stress on the neck structures, restricting neck movement and causing pain. According to the latest clinical research, enhancing neck muscle postural control is considered an effective way to alleviate pain in patients with chronic neck pain.

[0003] Currently, existing VR devices or systems for rehabilitation training of patients with neck pain mainly focus on improving patients' neck mobility and increasing neck muscle mass. Some patients experience short-term symptom relief, but the long-term effects are not good. This is because the patients' ability to control their neck posture muscles has not been significantly improved.

[0004] For example, the Chinese patent "Cervical Spine Rehabilitation Training System" (patent application number: 202010211003.5, publication number: CN111359158A). This patent monitors the rehabilitation patient's posture in real time during training and provides reminders to improve the standardization of exercise training; it generates exercise programs based on the rehabilitation patient's cervical spine mobility to ensure targeted training and improve training safety; at the same time, it combines games to guide the rehabilitation patient, divert their attention, reduce pain during training, increase the fun of training, and improve the rehabilitation patient's compliance.

[0005] However, the main purpose of this program is to design training programs that focus on the range of motion of the cervical spine. That is, in the process of treating neck pain, it only focuses on the patient's strength and range of motion, without considering the patient's posture control and neuromuscular regulation.

[0006] For example, the Chinese patent "A Rehabilitation Training Assistance Method and System Based on Brain-Computer Interface and Virtual Reality" (Patent Application No.: 202411414791.2, Publication No.: CN118919020A). This patent achieves a comprehensive and integrated training experience through the effective linkage between virtual devices and rehabilitation equipment. This linkage not only enhances the interactivity of training but also strengthens the user's sense of participation and enthusiasm.

[0007] However, the main purpose of this program is to enhance the fun of training and help patients actively participate in training, rather than designing a training program that is more conducive to patient recovery based on consideration of patients' postural control and neuromuscular regulation.

[0008] It is evident that existing technologies primarily focus on setting training programs based on superficial factors such as the patient's strength and range of motion, and encourage active patient participation by enhancing the fun of training. However, the assessment of the patient's physical condition is still not comprehensive enough, resulting in training failing to effectively help improve the patient's neck posture muscle control. Summary of the Invention

[0009] To address the problems existing in the prior art, this invention provides a VR chronic neck pain rehabilitation training system based on CPC technology, which solves the problem that existing VR-integrated rehabilitation systems cannot effectively help improve patients' neck posture muscle control.

[0010] A VR (virtual reality) rehabilitation training system for chronic neck pain based on CPC (central pathway and posture control) technology includes a data acquisition component, a data processing module, a plan creation module, VR equipment, a control center, a database, and a medical care terminal.

[0011] The data processing module is used to analyze and process the action posture data to obtain analysis results;

[0012] The control center is used to push the analysis results to the medical staff, and then receive guidance and adjustment opinions from the medical staff.

[0013] The planning module is used to generate training plans based on guidance and to adjust training plans based on adjustment suggestions.

[0014] The VR device is used to build virtual scenes according to the training plan and guide patients to perform the training plan in the virtual scenes;

[0015] The database is used to store data related to rehabilitation training.

[0016] Furthermore, the VR device includes a scene building module and a training guidance module. The scene building module is used to generate a specified training scene according to the training plan, and the training guidance module is used to generate prompting text or images in the training scene according to the training plan, and provide corresponding feedback after the patient completes the operation.

[0017] Furthermore, the data acquisition component includes an image acquisition unit and a sensor unit. The image acquisition unit is used to acquire images of the patient's movements according to instructions issued by the control center, and the sensor unit is used to acquire real-time movement data of the patient's head and neck.

[0018] Furthermore, the control center sends evaluation commands to the VR device, and the VR device executes the evaluation commands and broadcasts voice prompts. The evaluation commands include static evaluation commands and dynamic evaluation commands. The voice prompts corresponding to the static evaluation commands include "Evaluation begins," "Sit up straight," "Remain still for n seconds," and "Evaluation ends." The voice prompts corresponding to the dynamic evaluation commands include "Evaluation begins," "Please perform the specified action," "Remain still for n minutes," and "Evaluation ends." When the VR device broadcasts "Remain still for n seconds," the action posture data collected by the data acquisition component is used as static evaluation data. When the VR device broadcasts "Remain still for n minutes," the action posture data collected by the data acquisition component is used as dynamic evaluation data.

[0019] Furthermore, the specified actions include sitting upright, turning the head to the left, turning the head to the right, raising the head, and tucking the chin.

[0020] Furthermore, the data processing module processes the static assessment data to obtain static assessment results, and processes the dynamic assessment data to obtain dynamic assessment results. Then, the static and dynamic assessment results are input into the analysis model to obtain the patient's comprehensive neck assessment results. Finally, the static assessment results, dynamic assessment results, and comprehensive neck assessment results are sent to the medical staff.

[0021] Furthermore, the data processing module analyzes the static assessment data, including: the static assessment data is the patient's posture data when they feel they are sitting upright; the control center retrieves normal posture data for comparison; and the patient's head tilt angle is analyzed based on three-dimensional coordinates to analyze the patient's static head and neck posture perception and control ability, which serves as the static assessment result.

[0022] Furthermore, the data processing module analyzes the dynamic assessment data by dividing it into multiple data segments based on different specified actions. Each sub-data segment is then further divided into multiple time series. The data corresponding to each time series in a single sub-data segment are compared and analyzed to obtain the changes of a single specified action of the patient over time. The changes of multiple specified actions over time are then integrated and analyzed to assess whether the patient has neck movement disorders or postural control dysfunction, and to obtain the dynamic assessment results.

[0023] Furthermore, the planning module deploys a large language model, which is used to understand and analyze guidance opinions, combine with the database to clarify the training direction and generate a complete sequence of action instructions. The complete sequence of action instructions is then converted into several VR device game instructions and voice prompt groups as a training plan. The large language model is also used to understand and analyze adjustment opinions, combine with the database to clarify the action instructions that need to be adjusted, adjust the complete sequence of action instructions, and then convert the adjusted complete sequence of action instructions into a new training plan.

[0024] Furthermore, the training plan includes:

[0025] Posture control training: After wearing VR equipment, the patient controls the position of his head and aligns the crosshair with the crosshair. The system will simultaneously assess the patient's swing amplitude.

[0026] Neck muscle activation: Contraction of the muscles at the back of the neck retracts the head to select the crosshair. If the head is extended forward, the crosshair cannot be selected. Once the crosshair is selected, training begins.

[0027] APA training: Based on the principles of APA, the patient observes and controls head movements in a timely manner without being prompted by the APA.

[0028] CPA training: Based on the CPA principle, patients can anticipate the direction of movement and perform actions with prior cues to improve their CPA control.

[0029] The beneficial effects of this invention include:

[0030] 1. Based on the principle of posture control and combined with virtual reality technology, the existing VR rehabilitation training has been improved. It has been transformed from simple strength and speed training to neck neuromuscular posture control training, which can fundamentally solve the pain caused by the decline in the patient's neck muscle posture control ability.

[0031] 2. It can provide an immersive training environment for patients with neck pain, which can enhance the fun of training and make it easier for patients to get involved.

[0032] 3. Through various interactive devices, patients can interact with the virtual environment, truly perceive their progress, and improve their motivation for training.

[0033] 4. By providing real-time feedback information and through multi-sensory stimulation such as vision and hearing, timely feedback can be provided to adjust the coordination and control of the patient's neck muscles during the training process, thereby improving the accuracy and effectiveness of the training.

[0034] 5. Data analysis and processing enable real-time monitoring of patients' recovery progress, which helps doctors make scientific decisions and improves recovery efficiency.

[0035] 6. Develop individualized training plans based on the patient's specific condition to make rehabilitation training more targeted, improve training effectiveness, and accelerate the rehabilitation process. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the structure of a VR chronic neck pain rehabilitation training system based on CPC technology, which is an embodiment of this application.

[0037] Figure 2This is a schematic diagram illustrating the use of a VR chronic neck pain rehabilitation training system based on CPC technology, as described in an embodiment of this application.

[0038] Figure 3 This is the interface for running APA training on the VR device involved in the embodiments of this application.

[0039] Figure 4 This is the interface for running CPA training on the VR device involved in the embodiments of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0041] A VR rehabilitation training system for chronic neck pain based on CPC technology, such as Figure 1 As shown, it includes data acquisition components, data processing modules, planning modules, VR equipment, control center, database, and medical staff terminals;

[0042] The data acquisition component is used to collect patients' movement and posture data during the assessment and training periods.

[0043] The data processing module is used to analyze and process the action posture data to obtain analysis results;

[0044] The control center is used to push the analysis results to the medical staff, and then receive guidance and adjustment opinions from the medical staff.

[0045] The planning module is used to generate training plans based on guidance and to adjust training plans based on adjustment suggestions.

[0046] The VR device is used to build virtual scenes according to the training plan and guide patients to perform the training plan in the virtual scenes;

[0047] The database is used to store data related to rehabilitation training.

[0048] The VR device includes a scene building module and a training guidance module. The scene building module is used to generate a specified training scene according to the training plan. The training guidance module is used to generate prompting text or images in the training scene according to the training plan and provide corresponding feedback after the patient completes the operation.

[0049] In another embodiment, the data acquisition component includes an image acquisition unit and a sensor unit. The image acquisition unit is used to acquire images of the patient's movements according to instructions issued by the control center, and the sensor unit is used to acquire real-time movement data of the patient's head and neck.

[0050] In another embodiment, the control center sends an evaluation command to the VR device, and the VR device executes the evaluation command and broadcasts voice prompts; the evaluation command includes static evaluation command and dynamic evaluation command, and the voice prompts corresponding to the static evaluation command include "Evaluation begins", "Sit up straight", "Stay still for n seconds" and "Evaluation ends";

[0051] The voice prompts corresponding to the dynamic evaluation instructions include "Evaluation begins", "Please perform the specified action", "Remain still for n minutes", and "Evaluation ends". When the VR device announces "Remain still for n seconds", the action posture data collected by the data acquisition component is used as static evaluation data. When the VR device announces "Remain still for n minutes", the action posture data collected by the data acquisition component is used as dynamic evaluation data.

[0052] In another embodiment, the specified actions include sitting upright, turning the head to the left, turning the head to the right, raising the head, and tucking the chin.

[0053] In another embodiment, the data processing module processes the static assessment data to obtain static assessment results, and processes the dynamic assessment data to obtain dynamic assessment results. Then, the static assessment results and dynamic assessment results are input into the analysis model to obtain the patient's comprehensive neck assessment results. Finally, the static assessment results, dynamic assessment results, and comprehensive neck assessment results are sent to the medical staff.

[0054] In another embodiment, the data processing module analyzes the static assessment data as follows: the static assessment data is the patient's posture data when he feels that his body is sitting upright. The control center retrieves normal posture data for comparison. At the same time, the patient's head position tilt angle is analyzed based on three-dimensional coordinates to analyze the patient's static head and neck posture perception and control ability, which is used as the static assessment result.

[0055] In another embodiment, the data processing module analyzes the dynamic assessment data by dividing the dynamic assessment data into multiple data segments according to different specified actions, then dividing each individual sub-data segment into multiple time series, comparing and analyzing the data corresponding to each time series in each individual sub-data segment, obtaining the changes of the patient's single specified action over time, integrating and analyzing the changes of multiple specified actions over time, assessing whether the patient has neck movement disorders or postural control dysfunction, and obtaining dynamic assessment results.

[0056] The specific processing procedure is as follows:

[0057] Data preprocessing and input: Collect and input dynamic assessment datasets of patients, which contain posture data of the head or other body parts at multiple time points.

[0058] Data segments are divided by action: The data is divided into multiple sub-data segments according to the type of action (such as turning the head left, turning the head right, looking down, looking up, etc.).

[0059] Time series partitioning: Each sub-data segment is split into multiple time series according to time, and the changes of each action in the time dimension are analyzed.

[0060] Time series data comparison and analysis: Compare each time series data one by one to analyze the stability, smoothness and presence of anomalies in the movements.

[0061] Time-varying motion analysis: Evaluate the trend of each specified motion over time during execution to determine whether the patient can perform the motion stably.

[0062] Integrating the analysis results of multiple movements: The analysis results of multiple movements are combined to assess whether the patient has motor disorders or postural control dysfunction.

[0063] Generate dynamic assessment reports: Finally, based on the assessment results of the patient's various actions, a detailed dynamic assessment report is generated and provided to medical staff for further diagnosis and treatment.

[0064] By meticulously analyzing patients' dynamic assessment data through multiple steps, a comprehensive evaluation of their postural control and motor performance when performing various specified actions can be achieved. By dividing the dynamic data into multiple time series and conducting comparative analysis, the system ultimately detects neck movement disorders or postural control dysfunctions and outputs relevant dynamic assessment results. These results can also provide valuable references for healthcare professionals, helping to develop personalized treatment plans.

[0065] In another embodiment, the plan creation module deploys a large language model. This model is used to understand and analyze guidance, combine it with a database to clarify the training direction, and generate a complete sequence of motion instructions. This complete sequence of motion instructions is then converted into several VR device game commands and voice prompts as a training plan. The large language model is also used to understand and analyze adjustment suggestions, combine them with the database to clarify the motion instructions that need adjustment, and adjust the complete sequence of motion instructions accordingly. The adjusted sequence of motion instructions is then converted into a new training plan. Specifically, the guidance includes exercise difficulty, intensity, frequency, etc.

[0066] Specifically, the design process of the large language model is as follows:

[0067] (1) Preparation for fine-tuning the large language model:

[0068] In this embodiment, fine-tuning techniques are employed to enable the large language model to convert natural language guidance (such as exercise difficulty, intensity, and frequency) into executable training instruction codes. By using a specialized corpus containing patient rehabilitation training cases, medical expertise, and training plans, the pre-trained large language model is fine-tuned to understand medical terminology, action sequences, and motor control requirements within the guidance, and to automatically generate corresponding instruction codes based on these inputs. During fine-tuning, the model maps natural language input (such as "moderate intensity head rotation training") to specific training instructions (such as "turn your head 10 degrees to the left for 5 seconds, at a frequency of 30 times per minute").

[0069] (2) Fine-tuning the dataset and training process:

[0070] During fine-tuning, the dataset includes a large number of real-world training cases and their corresponding training instructions, covering various factors such as posture control, neck muscle activation, movement intensity, and difficulty. Each training case includes not only natural language descriptions of the training requirements but also precise training motion parameters (such as angle, duration, and frequency). Using this data, the large language model can generate code that can be directly used for training after understanding the patient's needs. The model training process employs supervised learning methods, using this labeled data for training, and optimizing the model by minimizing the error between the predicted and target instructions.

[0071] (3) Natural language to instruction code conversion mechanism:

[0072] The fine-tuned large language model can map natural language instructions into structured training command codes. Specifically, the model performs semantic parsing on the input instructions, identifies key action parameters (such as "turn head to the left," "10 degrees," "last 5 seconds," etc.), and converts them into command formats that the system can understand. These command formats can be:

[0073] Action instructions: For example, "Turn head left: angle = 10 degrees, duration = 5 seconds".

[0074] Voice prompts: For example, "Please turn your head to the left, hold the angle at 10 degrees, and hold for 5 seconds."

[0075] The generated instruction sequence is then converted into executable commands for the VR device through a predefined interface, including visual feedback, audio prompts, and training task settings. This ability to automatically generate instructions greatly improves the personalization and adaptability of training plans.

[0076] (4) Multi-level instruction generation and feedback mechanism:

[0077] The fine-tuned large language model supports multi-level training instruction generation. It can not only generate initial training instructions based on the patient's current state (such as exercise intensity and fatigue level), but also dynamically adjust the training plan through a real-time feedback mechanism. For example, if the system detects that the patient is having difficulty performing a certain action, the model will automatically adjust the exercise intensity, frequency, or other training parameters based on the patient's feedback and the performance of the action, to help the patient gradually improve their motor skills. These adjustments will be prompted to the patient in natural language (such as "reduce exercise intensity, slow down"), and new instruction codes will be generated by the model and translated into instructions for the VR device in real time.

[0078] (5) Model adjustment to match patient needs:

[0079] In practical applications, the large language model can be adjusted according to the individualized needs of patients. For example, a patient's age, medical history, motor ability, and rehabilitation progress will all affect the setting of the training plan. Through fine-tuning, the large language model can analyze the patient's training progress based on patient information and training records in the database, and adjust parameters such as exercise difficulty, intensity, and frequency in the training plan. In addition, the model can also make real-time adjustments based on patient feedback to ensure that training instructions continuously meet the patient's rehabilitation needs.

[0080] (6) Technical effects and advantages:

[0081] The fine-tuned large language model significantly improves the intelligence and automation of training plan generation, reducing the need for manual intervention. Simultaneously, the model can adapt to the personalized needs of different patients, flexibly adjusting training content and difficulty. Compared to traditional methods of manually writing training instructions, this technology not only improves the accuracy of training results but also accelerates the training plan development process, ensuring patients achieve optimal rehabilitation outcomes in the shortest possible time. Furthermore, due to the model's scalability, new training plans can be quickly generated and adjusted in the future based on different disease types and rehabilitation needs.

[0082] In another embodiment, the training plan includes:

[0083] (1) Posture control training: After wearing the VR device, the patient controls the position of their head, aligning the crosshair with the crosshair. The system will simultaneously assess the patient's swaying range. (e.g.) Figure 2 )

[0084] (2) Neck muscle activation: The head position is retracted by contracting the muscles at the back of the neck to select the control crosshair. If the head position is extended forward, the crosshair cannot be selected. After selecting the crosshair, the patient will play the task game through the game screen for 30 minutes.

[0085] (3) APA Training: The task-based game design is based on APA principles. Without giving the patient any prompts, game tasks are designed, such as dodging projectiles by controlling head position. (e.g.) Figure 3 )

[0086] (4) CPA Training: Task-based games are designed based on the CPA principle. With advance warning (an exclamation mark), patients anticipate the direction of the thrown object, improving their CPA control. (e.g.) Figure 4 )

[0087] Specifically, when faced with predictable or sudden internal and / or external postural disturbances, the central nervous system's regulation of the body's postural muscle responses mainly includes two mechanisms: anticipatory postural ladjustment (APA) and compensatory postural adjustment (CPA). APA refers to establishing stable kinetic chains to connect different body segments, preventing mechanical disturbances caused by local movements. CPA refers to the postural adjustments that occur in response to unpredictable postural disturbances.

[0088] The neural mechanism of APA (Activated Postural Awareness) likely stems from the transmission of perceptual and cognitive neural information, such as visual, auditory, proprioceptive, and motor intention, to the brain. After integration by the central nervous system, nerve impulses are generated, leading to movement prediction and pre-activation of relevant muscle activity, thus achieving optimized control of muscle activity. Studies have shown that APA of the neck postural muscles is significantly weakened before movement in patients with chronic neck pain, and the speed of muscle activation is significantly reduced.

[0089] The purpose of CPA (Postural Adjustment) is to adjust the position of the body's center of gravity after a disturbance occurs. Motor commands from the motor cortex act on the motor system, changing its state and inducing different sensory inputs to adjust posture. CPA requires correct sensory input and motor output.

[0090] In summary, when faced with unpredictable disturbances, the CPA is the main regulatory mechanism for postural control in the central nervous system; while when faced with predictable disturbances, the APA and CPA work together to maintain postural stability.

[0091] In another embodiment, the VR device further includes:

[0092] (1) Virtual Environment Generation Module: Used to generate a three-dimensional training space environment. For example, the patient aims at the target by adjusting their head position and fires accordingly. This requires the patient to continuously adjust their head position based on visual feedback during the process, enhancing the control ability of the neck posture muscles (e.g., ...). Figure 2 ).

[0093] (2) Posture capture module: used to capture the head position information of the patient's body in real time, including three-dimensional spatial motion trajectory, motion speed, motion acceleration, etc.

[0094] (3) Feedback control module: used to adjust the audiovisual feedback in the virtual environment and guide patients to perform correct posture control training.

[0095] (4) Training intensity adjustment module: Adjust the training difficulty according to the patient's training situation to prevent the patient from giving up training due to frustration.

[0096] When the completion rate of APA and CPA tasks falls below 60%, the training system automatically adjusts the training difficulty based on the patient's performance. If the patient fails to complete the predetermined movements within the allotted time, the system will slow down the game's pace, simplify the task complexity (e.g., reduce the number of throwables or lower the task difficulty), and increase voice or visual cues to help the patient improve task completion. This dynamic adjustment mechanism ensures that patients gradually master postural adjustment skills during rehabilitation without losing confidence due to overly difficult tasks.

[0097] If a patient fails to meet the set standards when performing APA or CPA tasks, the system can reduce the patient's exercise burden by lowering the training intensity, thus avoiding exacerbating pain or fatigue due to excessive load.

[0098] Reduce range of motion: For example, in the APA task, if the patient makes many mistakes or fails to complete the movement during training, the system can reduce the range of head rotation angles (e.g., from 30 degrees to 15 degrees) to reduce the intensity of the movement for the patient.

[0099] Reduce repetitions or duration: In CPA tasks, if the patient fails to make compensatory postural adjustments, the system can reduce the number of repetitions or the duration of each movement to ensure that the patient can complete the training with less burden.

[0100] Reduce movement speed: For patients with poor control, the system can reduce the speed of movement and the required precision of the action, thereby reducing the burden on the patient.

[0101] Reduce training frequency: For patients with slow progress, the system can appropriately reduce the training frequency (e.g., from 5 times a day to 3 times a day) to give patients more recovery time and avoid fatigue caused by overtraining.

[0102] Extended recovery intervals: If a patient performs poorly in a training cycle, the system can extend the rest time between training sessions (e.g., increase the rest time between each training session) so that the patient can recover and improve their performance in the next training session.

[0103] Gradually increase the frequency: When a patient makes good progress at a certain stage, the system will gradually increase the frequency of training based on their performance to improve the rehabilitation process.

[0104] (5) Data analysis and processing module: Analyzes the data during the patient training process, evaluates the training effect, and provides a basis for doctors to adjust the training plan.

[0105] (6) Human-computer interaction module: Enables patients to interact with the virtual environment, displaying training progress, training scores, and feedback information. Therapists can monitor the patient's real-time training screen and training status through a monitor and provide corresponding guidance and adjustments.

[0106] In another embodiment, the system usage process includes:

[0107] Step 1: The patient puts on the VR device and adjusts the position of the data acquisition components (camera, sensors);

[0108] Step 2: Determine if this is the patient's first time using the system; if yes, proceed to Step 3; otherwise, proceed to Step 4.

[0109] Step 3: Register an account, log in to the system to enter the evaluation process, and obtain the corresponding training plan after the evaluation is completed;

[0110] Step 4: Conduct training according to the training plan;

[0111] Step 5: If you choose to update the training plan, you will collect historical training data, evaluate it, and obtain the adjusted training plan.

[0112] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

Claims

1. A VR rehabilitation training system for chronic neck pain based on CPCC technology, characterized in that, Includes data acquisition components, data processing modules, planning modules, VR equipment, control center, database, and medical staff terminals; The data processing module is used to analyze and process the motion posture data to obtain analysis results. The motion posture data consists of static evaluation data and dynamic evaluation data. The data processing module processes the static evaluation data to obtain static evaluation results and processes the dynamic evaluation data to obtain dynamic evaluation results. Then, the static evaluation results and dynamic evaluation results are input into the analysis model to obtain the patient's comprehensive neck evaluation results. Finally, the static evaluation results, dynamic evaluation results, and comprehensive neck evaluation results are sent to the medical staff terminal. The control center is used to push the analysis results to the medical staff, and then receive guidance and adjustment opinions from the medical staff. The planning module is used to generate training plans based on guidance and to adjust training plans based on adjustment suggestions. The planning module is equipped with a large language model, which is used to understand and analyze the guidance, combine it with the database to clarify the training direction, and generate a complete sequence of action instructions. The complete sequence of action instructions is then converted into several VR device game instructions and voice prompt groups as a training plan. The large language model is used to understand and analyze the adjustment opinions, combine them with the database to identify the action instructions that need to be adjusted, adjust the complete action instruction sequence, and then transform the adjusted complete action instruction sequence into a new training plan. The database is used to store data related to rehabilitation training; The VR device is used to build a virtual scene according to the training plan and guide the patient to perform the training plan in the virtual scene. The training plan includes: Posture control training: After wearing VR equipment, the patient controls the position of his head and aligns the crosshair with the crosshair. The system will simultaneously assess the patient's swing amplitude. Neck muscle activation: Contraction of the muscles at the back of the neck retracts the head to select the crosshair. If the head is extended forward, the crosshair cannot be selected. Once the crosshair is selected, training begins. APA training: Based on the principles of APA, the patient observes and controls head movements in a timely manner without being prompted by the APA. CPA training: Based on the CPA principle, patients can anticipate the direction of the movement and perform the movement with prior prompts to improve their CPA ability. The CPCC technology refers to central conduction pathway and posture control technology, the APA refers to anticipatory posture adjustment, and the CPA refers to compensatory posture adjustment.

2. The VR chronic neck pain rehabilitation training system based on CPCC technology according to claim 1, characterized in that, The VR device includes a scene building module and a training guidance module. The scene building module is used to generate a specified training scene according to the training plan. The training guidance module is used to generate prompting text or images in the training scene according to the training plan and provide corresponding feedback after the patient completes the operation.

3. The VR chronic neck pain rehabilitation training system based on CPCC technology according to claim 1, characterized in that, The data acquisition component includes an image acquisition unit and a sensor unit. The image acquisition unit is used to acquire images of the patient's movements according to instructions issued by the control center, and the sensor unit is used to acquire real-time movement data of the patient's head and neck.

4. The VR chronic neck pain rehabilitation training system based on CPCC technology according to claim 1, characterized in that, The control center sends evaluation commands to the VR device, and the VR device executes the evaluation commands and broadcasts voice prompts. The evaluation commands include static evaluation commands and dynamic evaluation commands. The voice prompts corresponding to the static evaluation commands include "Evaluation begins," "Sit up straight," "Hold still for n seconds," and "Evaluation ends." The voice prompts corresponding to the dynamic evaluation commands include "Evaluation begins," "Please perform the specified action," "Hold still for n minutes," and "Evaluation ends." When the VR device broadcasts "Hold still for n seconds," the action posture data collected by the data acquisition component is used as static evaluation data. When the VR device broadcasts "Hold still for n minutes," the action posture data collected by the data acquisition component is used as dynamic evaluation data.

5. A VR chronic neck pain rehabilitation training system based on CPCC technology according to claim 4, characterized in that, The specified actions include sitting upright, turning the head to the left, turning the head to the right, raising the head, and tucking the chin.

6. The VR chronic neck pain rehabilitation training system based on CPCC technology according to claim 1, characterized in that, The data processing module analyzes static assessment data, including: static assessment data is the patient's posture data when they feel they are sitting upright; the control center retrieves normal posture data for comparison; and the module analyzes the patient's head position tilt angle based on three-dimensional coordinates to analyze the patient's static head and neck posture perception and control ability, which serves as the static assessment result.

7. A VR chronic neck pain rehabilitation training system based on CPPC technology according to claim 1, characterized in that, The data processing module analyzes the dynamic assessment data by dividing it into multiple data segments based on different specified actions. Each sub-data segment is then further divided into multiple time series. The data corresponding to each time series in a single sub-data segment are compared and analyzed to obtain the changes of a single specified action of the patient over time. The changes of multiple specified actions over time are then integrated and analyzed to assess whether the patient has neck movement disorders or postural control dysfunction, and to obtain the dynamic assessment results.

Images