Data processing method and system based on virtual reality

Abstract

The invention discloses a data processing method and system based on virtual reality, and the method comprises the steps: analyzing the influence of expression information and muscle signals of a patient on training actions, adjusting a virtual training model according to an analysis result, analyzing the training amount of the patient under the adjusted training actions, and adjusting a rehabilitation training scheme of the patient. Whether the training action is excessive or not is judged by analyzing the facial expression change of the patient during the training action, for example, the conditions of eyebrow wrinkling, mouth micro-tension, heart rate acceleration and the like possibly caused by pain of the patient, then the expression change of the patient in the action process is analyzed, the limit action range of the virtual training model is determined, and the virtual training model is adjusted. The virtual training model meets the training requirements, the situation that the rehabilitation training effect is poor due to overtraining of the patient is avoided, then the better training effect is achieved, and the method has the advantages of being high in accuracy and good in rehabilitation training effect.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality, and specifically provides a data processing method and system based on virtual reality. Background Art

[0002] With the rapid development of society, people pay more and more attention to physical health and are more willing to carry out rehabilitation training to quickly recover their bodies.

[0003] Currently, in the field of rehabilitation training, traditional rehabilitation training modes usually rely on doctors' orders, rehabilitation training videos, etc. Under this kind of rehabilitation training mode, patients can only follow the doctor's orders or imitate the rehabilitation training videos to carry out rehabilitation training. However, patients cannot monitor their own rehabilitation training according to their own rehabilitation training process, resulting in patients being unable to accurately grasp their own rehabilitation training situation and unable to accurately control the training volume, leading to too much or too little rehabilitation training volume and poor rehabilitation training effects. Moreover, during the training process, patients may change the training actions or force application methods due to reasons such as pain caused by large training action amplitudes, resulting in the injured parts not being effectively trained, and the prior art being unable to accurately evaluate the training effects of patients. Therefore, it is necessary to design a data processing method and system based on virtual reality that can improve accuracy and rehabilitation training effects. Summary of the Invention

[0004] The purpose of the present invention is to provide a data processing method and system based on virtual reality to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: A data processing method based on virtual reality, including:

[0006] Step S1: Establish a rehabilitation database, collect sensor data in real time, store the collected sensors in the rehabilitation database, and input the patient's comprehensive information and doctor's order information into the rehabilitation database;

[0007] Step S2: Collect the action images and muscle signals of the patient's rehabilitation training, construct a virtual action model based on the action images, determine whether the training actions of the virtual action model are standard, analyze the influence of the training actions and muscle signals on the rehabilitation training score, and obtain the patient's rehabilitation training score;

[0008] Step S3: Collect the patient's expression information, analyze the influence of the patient's expression information and muscle signals on the training actions, and adjust the virtual training model according to the analysis results;

[0009] Step S4: Analyze the training volume of the patient under the adjusted training actions and adjust the patient's rehabilitation training plan.

[0010] According to the above technical solution, the steps of collecting the action images of the patient's rehabilitation training, constructing a virtual action model based on the action images, judging whether the training actions of the virtual action model are standard, analyzing the influence of the training actions and muscle signals on the rehabilitation training score, and obtaining the patient's rehabilitation training score are as follows:

[0011] Step A1: Construct a virtual training model according to the patient's rehabilitation training plan;

[0012] Construct a patient virtual action model based on the action images of the patient's rehabilitation training;

[0013] Map the virtual action model into the virtual training model for overlapping comparison to obtain whether the actions in the virtual action model are standard;

[0014] Mark the non-overlapping body feature nodes in the virtual action model as the first nodes, and mark the body feature nodes corresponding to the first nodes in the virtual training model as the second nodes;

[0015] Step A2: Match the muscle signals with the first nodes to obtain the number of successfully matched muscle signals, identify the distance between the first nodes and the corresponding second nodes, and obtain the patient's rehabilitation training score according to the number and the distance.

[0016] According to the above technical solution, the steps of comparing the similarity of muscle signals and obtaining the patient's rehabilitation training score according to the difference between the action model and the training model and the similarity of muscle signals are as follows:

[0017] Step A21: Identify the patient's muscle signals, match the muscle signals with the first nodes, identify the number of successfully matched muscle signals, and retrieve the corresponding influence coefficient α on the rehabilitation training score according to the number;

[0018] Step A22: Identify the intensity of the successfully matched muscle signals, calculate the difference from the set threshold, and retrieve the corresponding influence coefficient β on the rehabilitation training score in the database according to the difference;

[0019] Step A23: Identify the coordinates of the first nodes and the second nodes, calculate the distance between the first nodes and the second nodes through the distance formula, and retrieve the set influence coefficient δ on the rehabilitation training score in the database according to the average distance;

[0020] Step A24: Calculate the patient's rehabilitation training score P1 = α × β × δ × P0 through the formula, where P0 represents the basic score of the set rehabilitation training, judge whether the rehabilitation training score meets the set conditions, and mark the virtual action models that do not meet the set conditions.

[0021] According to the above technical solution, collecting the facial expression information and muscle signals of the patient, analyzing the influence of the facial expression information and muscle signals of the patient on the training action, and adjusting the virtual training model according to the analysis result include the following steps:

[0022] Step B1: Obtain the facial expression information of the patient, identify the facial expression of the patient, analyze the change of the facial expression of the patient to obtain the pain value of the patient, and adjust the movement amplitude of the rehabilitation training action of the patient according to the pain value;

[0023] Step B2: Obtain the muscle signal of the patient, identify the position where the muscle signal is emitted, match the position with the target muscle position of the rehabilitation training, judge whether the muscle force of the patient is accurate according to the matching result, and adjust the muscle force of the patient.

[0024] According to the above technical solution, obtaining the facial expression information of the patient, identifying the facial expression of the patient, analyzing the change of the facial expression of the patient to obtain the pain value of the patient, and correcting the movement amplitude of the rehabilitation training action of the patient according to the pain value include the following steps:

[0025] Step B11: Identify the patient's eyebrows center, construct a coordinate system with the eyebrows center as the origin, identify the eyebrow edge nodes A and B on the right side of the eyebrows center, where A represents the eyebrow edge node far from the eyebrows center, B represents the eyebrow edge node close to the eyebrows center, identify the coordinates of the eyebrow edge nodes A and B, obtain the set coordinates C and D of the patient's eyebrows center edge nodes, connect AB and CD, and calculate the slopes K of the AB connection line and the CD connection line through the slope formula AB and K CD , calculate the difference K of the slopes AB -K CD , according to the difference K AB -K CD Retrieve the database, and retrieve the corresponding influence coefficient ε on the patient's pain value;

[0026] Step B12: Identify the coordinates of the patient's upper and lower lip nodes, calculate the distance L1 between the patient's upper and lower lip nodes through the distance formula, obtain the set distance L2 between the patient's upper and lower lip nodes, calculate the difference between L1 - L2, judge whether the difference is greater than the set threshold, and retrieve the influence coefficient θ set in the database;

[0027] Step B13: Obtain the real-time heart rate of the patient, calculate the difference between the real-time heart rate and the set heart rate, and retrieve the influence weight λ set in the database on the influence coefficient θ according to the difference;

[0028] Step B14: Calculate the pain value Q1 of the patient through the formula Q1 = (ε+(θ×λ))Q0, where Q0 represents the basic pain value of the patient;

[0029] Step B15: Extract frames from the virtual action model to obtain intercepted pictures, identify the pain values of the patient in each intercepted picture, select the intercepted pictures with pain values less than the set threshold, identify the action features of the virtual action model in the intercepted pictures, and use the action features as the limit range of the virtual training model to obtain an adjusted first virtual training model.

[0030] According to the above technical solution, the steps for obtaining the muscle signals of the patient, identifying the position where the muscle signals are emitted, matching the position with the target muscle position of the rehabilitation training, judging whether the patient's muscle force is accurate according to the matching result, and adjusting the patient's muscle force include the following steps:

[0031] Step B21: Identify the position of the patient's muscle signal, match the position with the patient's body feature nodes, identify the labels of the body feature nodes. When the label of the body feature node is the first node, identify the pain value of the patient. If the pain value is less than the threshold, calculate the difference between the pain value and the threshold, and retrieve the muscle signal intensity adjustment coefficient η in the database according to the difference.

[0032] Step B22: Identify the intensity H0 of the muscle signal, calculate the adjusted muscle signal intensity H1 = η × H0, retrieve the corresponding training action in the database according to the muscle signal intensity H1, and use the training action as the limit range of the virtual training model to obtain an adjusted second virtual training model.

[0033] Step B23: Compare the first virtual training model with the second virtual training model, and select the virtual training model with the smallest limit range as the target virtual training model for the patient.

[0034] According to the above technical solution, the steps for analyzing the training volume of the user under the corrected training action and adjusting the patient's rehabilitation training plan include the following steps:

[0035] Step S41: Obtain the patient's rehabilitation training plan, identify the number of training actions M0 of the patient's rehabilitation training, identify the limit range of the target virtual training model, identify the joint rotation angle θ0 of the patient under the limit range, identify the joint rotation angle θ1 of the patient in the virtual training model constructed according to the rehabilitation training plan, and retrieve the set training volume attenuation coefficient μ in the database according to the joint rotation angle difference θ0 - θ1.

[0036] Step S42: Identify the muscle signal intensity of the patient under the limit range, and retrieve the set training volume influence coefficient in the database according to the muscle signal intensity

[0037] Step S43: Calculate the actual training volume of the patient through a formula Among them, Y represents the training volume of the patient for each training action. Determine whether the training volume meets the standard, and calculate the number of training actions that need to be compensated through a formula Among them, W0 represents the training volume set in the patient's rehabilitation training plan. Adjust the patient's rehabilitation training plan according to the compensated training action volume. During the patient's training process, monitor the change in the limit range of the patient's training actions in real time, and update the patient's rehabilitation training plan in real time according to the change in the limit range

[0038] According to the above technical solution, the virtual reality-based data processing system includes: a data acquisition module and a model construction module

[0039] The data acquisition module is used to establish a rehabilitation database, collect sensor data in real time, store the collected sensors in the rehabilitation database, and input the patient's comprehensive information and doctor's order information into the rehabilitation database

[0040] The model construction module is used to collect the action images of the patient's rehabilitation training, determine whether the patient's rehabilitation training actions are standard according to the action images, analyze the influence of the patient's rehabilitation training actions and muscle signals on the rehabilitation training score, and obtain the patient's rehabilitation training score

[0041] According to the above technical solution, the system further includes a training evaluation module

[0042] The training evaluation module is used to collect the patient's expression information and muscle signals, analyze the influence of the patient's expression information and muscle signals on the rehabilitation training actions, and correct the patient's rehabilitation training actions according to the analysis results

[0043] According to the above technical solution, the system further includes a training guidance module

[0044] The training guidance module is used to analyze the training volume of the user under the corrected training actions and adjust the patient's training volume

[0045] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In the present invention, by overlapping and comparing the virtual action model and the virtual training model, it is possible to quickly determine whether the training actions of the patient meet the standards, analyze the muscle signals of the patient during the training actions, judge the connection relationship between the muscle signals and the body feature nodes of the patient, determine whether the muscles exercised by the patient using the training actions are the target muscles, and at the same time judge the intensity of the muscle signals, so as to accurately judge the training score of the patient during the training process, avoid the situation that the patient's training actions are not standard or the exercised muscles are inaccurate, greatly improve the rehabilitation training effect. By analyzing the facial expression changes of the patient during the training actions, such as the patient may frown, slightly open the mouth, and have an accelerated heart rate due to pain, judge whether the training actions are excessive, and then analyze the expression changes of the patient during the action process, determine the limit action amplitude of the virtual training model, adjust the virtual training model, so that the virtual training model meets the training requirements, avoid the situation that the patient has poor rehabilitation training effect due to overtraining, and thus achieve a better training effect. By analyzing the muscle signals of the patient during the training actions, identify the muscle exertion of the patient, judge whether the exertion of the patient is correct, and then provide exertion guidance to the patient. At the same time, analyze the intensity of the patient's muscle signals, judge whether the muscle signal intensity corresponding to the training actions of the virtual training model fits the patient's current rehabilitation situation, and be able to adjust the movement amplitude of the virtual training model according to the muscle signal intensity, be able to provide a more accurate training plan, make the virtual training model more in line with the patient's rehabilitation situation, and thus achieve a more excellent rehabilitation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the accompanying drawings:

[0047] Figure 1 is a flowchart of the method steps of the present invention.

[0048] Figure 2 is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0050] Please refer to Figure 1 , the present invention provides a technical solution: A virtual reality-based data processing method, including:

[0051] Step S1: Establish a rehabilitation database, collect sensor data in real time, store the collected sensors in the rehabilitation database, and input the comprehensive patient information and doctor's advice information into the rehabilitation database;

[0052] Step S2: Collect the action images and muscle signals of the patient's rehabilitation training, construct a virtual action model according to the action images, determine whether the training actions in the virtual action model are standard, analyze the influence of the training actions and muscle signals on the rehabilitation training score, and obtain the patient's rehabilitation training score;

[0053] Step S3: Collect the patient's facial expression information, analyze the influence of the patient's facial expression information and muscle signals on the training actions, and adjust the virtual training model according to the analysis results;

[0054] Step S4: Analyze the training volume of the patient under the adjusted training actions, and adjust the patient's rehabilitation training plan.

[0055] In the present invention, by analyzing the changes in the patient's facial expressions when performing training actions, for example, the patient may frown, slightly open the mouth, and have an accelerated heart rate due to pain, etc., to determine whether the training actions are excessive, and then analyze the changes in the patient's facial expressions during the action process, determine the limit action amplitude of the virtual training model, adjust the virtual training model, so that the virtual training model meets the training requirements, avoid the situation that the patient has poor rehabilitation training effects due to overtraining, and thus achieve better training effects. By analyzing the changes in the patient's facial expressions when performing training actions, for example, the patient may frown, slightly open the mouth, and have an accelerated heart rate due to pain, etc., to determine whether the training actions are excessive, and then analyze the changes in the patient's facial expressions during the action process, determine the limit action amplitude of the virtual training model, adjust the virtual training model, so that the virtual training model meets the training requirements, avoid the situation that the patient has poor rehabilitation training effects due to overtraining, and thus achieve better training effects.

[0056] In some preferred embodiments, step S2 further includes the following steps:

[0057] Step A1: Construct a virtual training model according to the patient's rehabilitation training plan;

[0058] Construct a patient virtual action model according to the action images of the patient's rehabilitation training;

[0059] Map the virtual action model into the virtual training model for overlapping comparison to obtain whether the actions in the virtual action model are standard;

[0060] Mark the non-overlapping body feature nodes in the virtual action model as the first nodes, and mark the body feature nodes corresponding to the first nodes in the virtual training model as the second nodes;

[0061] Step A2: Match the muscle signals with the first nodes to obtain the number of successfully matched muscle signals, identify the distance between the first nodes and the corresponding second nodes, and obtain the rehabilitation training score of the patient according to the number and the distance.

[0062] In some preferred embodiments, step A2 further includes the following steps:

[0063] Step A21: Identify the muscle signals of the patient, match the muscle signals with the first nodes, identify the number of successfully matched muscle signals, and retrieve the corresponding influence coefficient α on the rehabilitation training score according to the number;

[0064] Step A22: Identify the intensity of the successfully matched muscle signals, calculate the difference from the set threshold, and retrieve the corresponding influence coefficient β on the rehabilitation training score in the database according to the difference;

[0065] Step A23: Identify the coordinates of the first nodes and the second nodes, calculate the distance between the first nodes and the second nodes through the distance formula, and retrieve the set influence coefficient δ on the rehabilitation training score in the database according to the average distance;

[0066] Step A24: Calculate the rehabilitation training score P1 of the patient through the formula P1 = α × β × δ × P0, where P0 represents the basic score of the set rehabilitation training, determine whether the rehabilitation training score meets the set conditions, and mark the virtual action models that do not meet the set conditions.

[0067] Specifically, through the overlapping comparison of the virtual action model and the virtual training model, it is possible to quickly judge whether the training actions of the patient meet the standards, analyze the muscle signals during the patient's training actions, judge the connection relationship between the muscle signals and the patient's body feature nodes, determine whether the muscles exercised by the patient using the training actions are the target muscles, and at the same time judge the intensity of the muscle signals, so as to accurately judge the training score of the patient during the training process, avoid the patient having non-standard training actions or inaccurate exercised muscles, and greatly improve the rehabilitation training effect.

[0068] In some preferred embodiments, step S3 further includes the following steps:

[0069] Step B1: Obtain the expression information of the patient, identify the facial expression of the patient, analyze the change of the facial expression of the patient to obtain the pain value, and adjust the movement amplitude of the rehabilitation training action of the patient according to the pain value;

[0070] Step B2: Obtain the muscle signals of the patient, identify the positions where the muscle signals are emitted, match the positions with the target muscle positions of the rehabilitation training, judge whether the patient's muscle exertion is accurate according to the matching result, and adjust the patient's muscle exertion.

[0071] In some preferred embodiments, step B1 further includes the following steps:

[0072] Step B11: Identify the patient's eyebrow center, construct a coordinate system with the eyebrow center as the origin, identify the eyebrow edge nodes A and B on the right side of the eyebrow center, where A represents the eyebrow edge node far from the eyebrow center, and B represents the eyebrow edge node close to the eyebrow center, identify the coordinates of the eyebrow edge nodes A and B, obtain the set coordinates C and D of the patient's eyebrow center edge nodes, connect AB and CD, and calculate the slopes K of the AB connection line and the CD connection line through the slope formula AB and K CD , calculate the difference K of the slopes AB -K CD , according to the difference K AB -K CD Retrieve the database and retrieve the corresponding influence coefficient ε on the patient's pain value;

[0073] Step B12: Identify the coordinates of the patient's upper and lower lip nodes, calculate the distance L1 between the patient's upper and lower lip nodes through the distance formula, obtain the set distance L2 between the patient's upper and lower lip nodes, calculate the difference between L1 - L2, judge whether the difference is greater than the set threshold, and retrieve the set influence coefficient θ on the patient's pain value in the database;

[0074] Step B13:: Obtain the patient's real-time heart rate, calculate the difference between the real-time heart rate and the set heart rate, and retrieve the influence weight λ set in the database on the influence coefficient θ according to the difference;

[0075] Step B14: Calculate the patient's pain value Q1 = (ε+(θ×λ))Q0 through the formula, where Q0 represents the patient's basic pain value;

[0076] Step B15: Extract frames from the virtual action model to obtain intercepted pictures, identify the pain values of the patient in each intercepted picture, select the intercepted pictures with pain values less than the set threshold, identify the action features of the virtual action model in the intercepted pictures, and use the action features as the limit range of the virtual training model to obtain an adjusted first virtual training model.

[0077] Specifically, by analyzing the facial expression changes of the patient during the training actions, such as the patient may frown, slightly open the mouth, and have an accelerated heart rate due to pain, it is determined whether the training actions are excessive. Furthermore, by analyzing the expression changes of the patient during the action process, the limit movement amplitude of the virtual training model is determined, and the virtual training model is adjusted to make the virtual training model meet the training requirements, avoiding the situation where the patient has poor rehabilitation training effects due to overtraining, and thus achieving a better training effect.

[0078] In some preferred embodiments, step B2 further includes the following steps:

[0079] Step B21: Identify the position of the patient's muscle signal, match the patient's body feature nodes according to the position, identify the label of the body feature node. When the label of the body feature node is the first node, identify the pain value of the patient. If the pain value is less than the threshold, calculate the difference between the pain value and the threshold, and retrieve the muscle signal intensity adjustment coefficient η in the database according to the difference.

[0080] Step B22: Identify the intensity H0 of the muscle signal, calculate the adjusted muscle signal intensity H1 = η × H0, retrieve the corresponding training action in the database according to the muscle signal intensity H1, and use the training action as the limit range of the virtual training model to obtain the adjusted second virtual training model.

[0081] Step B23: Compare the first virtual training model with the second virtual training model, and select the virtual training model with the smallest limit range as the target virtual training model for the patient.

[0082] Specifically, by analyzing the muscle signals of the patient during the training actions, the muscle exertion of the patient is identified, and it is determined whether the patient's exertion is correct. Furthermore, the patient is given exertion guidance. At the same time, by analyzing the intensity of the patient's muscle signals, it is determined whether the muscle signal intensity corresponding to the training action of the virtual training model fits the patient's current rehabilitation situation. The movement amplitude of the virtual training model can be adjusted according to the muscle signal intensity, and a more accurate training plan can be provided, making the virtual training model more in line with the patient's rehabilitation situation, and thus achieving a better rehabilitation effect.

[0083] In some preferred embodiments, step S4 further includes the following steps:

[0084] Step S41: Obtain the patient's rehabilitation training plan, identify the number M0 of training actions in the patient's rehabilitation training, identify the limit range of the target virtual training model, identify the joint rotation angle θ0 of the patient under the limit range, identify the joint rotation angle θ1 of the patient in the virtual training model constructed according to the rehabilitation training plan, and retrieve the set training volume attenuation coefficient μ in the database according to the joint rotation angle difference θ0 - θ1;

[0085] Step S42: Identify the muscle signal intensity of the patient under the limit range, and retrieve the set training volume influence coefficient in the database according to the muscle signal intensity

[0086] Step S43: Calculate the actual training volume of the patient through a formula Among them, Y represents the training volume for each training action of the patient. Determine whether the training volume meets the standard, and calculate the number of training actions that need to be compensated through a formula Among them, W0 represents the training volume set in the patient's rehabilitation training plan. Adjust the patient's rehabilitation training plan according to the compensated training action volume, and monitor the change of the limit range of the patient's training actions in real time during the patient's training process, and update the patient's rehabilitation training plan in real time according to the change of the limit range.

[0087] With the same inventive concept as the above embodiment, the present application also provides a virtual reality-based data processing system, including: a data acquisition module, a model construction module, a training evaluation module, and a training guidance module;

[0088] The data acquisition module is used to establish a rehabilitation database, collect sensor data in real time, store the collected sensors in the rehabilitation database, and input the patient's comprehensive information and doctor's advice information into the rehabilitation database.

[0089] The model construction module is used to collect the action images of the patient's rehabilitation training, judge whether the patient's rehabilitation training actions are standard according to the action images, analyze the influence of the patient's rehabilitation training actions and muscle signals on the rehabilitation training score, and obtain the patient's rehabilitation training score.

[0090] The training evaluation module is used to collect the patient's expression information and muscle signals, analyze the influence of the patient's expression information and muscle signals on the rehabilitation training actions, and correct the patient's rehabilitation training actions according to the analysis results.

[0091] The training guidance module is used to analyze the training volume of the user under the corrected training actions and adjust the patient's training volume.

[0092] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0093] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A data processing method based on virtual reality, characterized in that: The method includes: Collecting the action images and muscle signals of the patient's rehabilitation training, constructing a virtual action model based on the action images, determining whether the training actions of the virtual action model are standard, analyzing the influence of the training actions and muscle signals on the rehabilitation training score, and obtaining the patient's rehabilitation training score; Collecting the patient's expression information, analyzing the influence of the patient's expression information and muscle signals on the training actions, and adjusting the virtual training model according to the analysis results; Analyzing the training volume of the patient under the adjusted training actions and adjusting the patient's rehabilitation training plan.

2. The data processing method based on virtual reality according to claim 1, wherein: The steps of collecting the action images of the patient's rehabilitation training, constructing a virtual action model based on the action images, determining whether the training actions of the virtual action model are standard, analyzing the influence of the training actions and muscle signals on the rehabilitation training score, and obtaining the patient's rehabilitation training score include the following steps: Constructing a virtual training model according to the patient's rehabilitation training plan; Constructing a patient virtual action model based on the action images of the patient's rehabilitation training; Mapping the virtual action model into the virtual training model for overlapping comparison to obtain whether the actions in the virtual action model are standard; Marking the non-overlapping body feature nodes in the virtual action model as the first nodes, and marking the corresponding body feature nodes in the virtual training model as the second nodes; Matching the muscle signals with the first nodes to obtain the number of successfully matched muscle signals, identifying the distance between the first nodes and the corresponding second nodes, and obtaining the patient's rehabilitation training score according to the number and the distance.

3. A data processing method based on virtual reality according to claim 2, characterized in that: The steps of comparing the similarity of the muscle signals and obtaining the patient's rehabilitation training score according to the difference between the action model and the training model and the similarity of the muscle signals include the following steps: Identifying the patient's muscle signals, matching the muscle signals with the first nodes, identifying the number of successfully matched muscle signals, and retrieving the corresponding influence coefficient α on the rehabilitation training score according to the number; Identifying the intensity of the successfully matched muscle signals, calculating the difference from the set threshold, and retrieving the corresponding influence coefficient β on the rehabilitation training score in the database according to the difference; Identifying the coordinates of the first nodes and the second nodes, calculating the distance between the first nodes and the second nodes through the distance formula, and retrieving the set influence coefficient δ on the rehabilitation training score in the database according to the average distance; Calculating the patient's rehabilitation training score P1 = α × β × δ × P0 by formula, where P0 represents the basic score of the set rehabilitation training, determining whether the rehabilitation training score meets the set conditions, and marking the virtual action models that do not meet the set conditions.

4. A data processing method based on virtual reality according to claim 1, characterized in that: The steps of collecting the patient's expression information and muscle signals, analyzing the influence of the patient's expression information and muscle signals on the training actions, and adjusting the virtual training model according to the analysis results include the following steps: Obtaining the patient's expression information, identifying the patient's facial expression, analyzing the change of the patient's facial expression to obtain the pain value of the patient, and adjusting the movement amplitude of the patient's rehabilitation training actions according to the pain value; Obtain the muscle signals of the patient, identify the position where the muscle signals are emitted, match the position with the target muscle position of the rehabilitation training, judge whether the patient's muscle force is accurate according to the matching result, and adjust the patient's muscle force.

5. A data processing method based on virtual reality according to claim 4, characterized in that: The steps of obtaining the expression information of the patient, identifying the facial expression of the patient, analyzing the change of the facial expression of the patient to obtain the pain value of the patient, and correcting the movement amplitude of the rehabilitation training action of the patient according to the pain value include the following steps: Identify the patient's glabella, construct a coordinate system with the glabella as the origin, identify the eyebrow edge nodes A and B on the right side of the glabella, where A represents the eyebrow edge node far from the glabella and B represents the eyebrow edge node close to the glabella, identify the coordinates of the eyebrow edge nodes A and B, obtain the set coordinates C and D of the patient's glabella edge nodes, connect AB and CD, and calculate the slopes K of the AB connection and the CD connection through the slope formula AB and K CD , calculate the difference K of the slopes AB -K CD , according to the difference K AB -K CD Retrieve the database and retrieve the corresponding influence coefficient ε on the patient's pain value; Identify the coordinates of the upper and lower lip nodes of the patient, calculate the distance L1 between the upper and lower lip nodes of the patient through the distance formula, obtain the set distance L2 between the upper and lower lip nodes of the patient, calculate the difference between L1 and L2, judge whether the difference is greater than the set threshold, and retrieve the influence coefficient θ of the set pain value of the patient in the database; Obtain the real-time heart rate of the patient, calculate the difference between the real-time heart rate and the set heart rate, and retrieve the influence weight λ of the set influence coefficient θ in the database according to the difference; Calculate the pain value Q1 of the patient through the formula Q1=(ε+(θ×λ))Q0, where Q0 represents the basic pain value of the patient; Extract frames from the virtual action model to obtain intercepted pictures, identify the pain values of the patient in each intercepted picture, select the intercepted pictures with pain values less than the set threshold, identify the action features of the virtual action model in the intercepted pictures, and use the action features as the limit range of the virtual training model to obtain the adjusted first virtual training model.

6. A data processing method based on virtual reality according to claim 4, characterized in that: The steps of obtaining the muscle signals of the patient, identifying the position where the muscle signals are emitted, matching the position with the target muscle position of the rehabilitation training, judge whether the patient's muscle force is accurate according to the matching result, and adjust the patient's muscle force include the following steps: Identify the position of the patient's muscle signal, match the patient's body feature nodes according to the position, identify the mark of the body feature nodes, when the mark of the body feature nodes is the first node, identify the pain value of the patient, if the pain value is less than the threshold, calculate the difference between the pain value and the threshold, and retrieve the muscle signal intensity adjustment coefficient η in the database according to the difference; Identify the intensity H0 of the muscle signal, calculate the adjusted muscle signal intensity H1 = η×H0, retrieve the corresponding training action in the database according to the muscle signal intensity H1, and use the training action as the limit range of the virtual training model to obtain the adjusted second virtual training model; Compare the first virtual training model with the second virtual training model, and select the virtual training model with the smallest limit range as the target virtual training model of the patient.

7. A data processing method based on virtual reality according to claim 1, characterized in that: The steps of analyzing the training volume of the user under the corrected training action and adjusting the rehabilitation training plan of the patient include the following steps: Obtain the patient's rehabilitation training plan, identify the number of training actions M0 in the patient's rehabilitation training, identify the limit range of the target virtual training model, identify the joint rotation angle θ0 of the patient under the limit range, identify the joint rotation angle θ1 of the patient in the virtual training model constructed according to the rehabilitation training plan, and retrieve the set training volume attenuation coefficient μ in the database according to the joint rotation angle difference θ0 - θ1; Identify the muscle signal intensity of the patient within the limit range, and retrieve the training volume influence coefficient set in the database according to the muscle signal intensity Calculate the actual training volume of the patient through a formula Among them, Y represents the training volume of the patient for each training action. Determine whether the training volume meets the standard, and calculate the number of training actions that need to be compensated through a formula Among them, W0 represents the training volume set in the patient's rehabilitation training plan. Adjust the patient's rehabilitation training plan according to the compensated training action volume. During the patient's training process, monitor the change in the limit range of the patient's training actions in real time, and update the patient's rehabilitation training plan in real time according to the limit range change 8. A data processing system based on virtual reality, characterized in that: The system includes: a data acquisition module and a model construction module; The data acquisition module is used to establish a rehabilitation database, collect sensor data in real time, store the collected sensors in the rehabilitation database, and enter the patient's comprehensive information and doctor's order information into the rehabilitation database; The model construction module is used to collect the action images of the patient's rehabilitation training, judge whether the patient's rehabilitation training actions are standard according to the action images, analyze the influence of the patient's rehabilitation training actions and muscle signals on the rehabilitation training score, and obtain the patient's rehabilitation training score.

9. A data processing system based on virtual reality according to claim 8, characterized in that: The system further includes a training evaluation module The training evaluation module is used to collect the patient's facial expression information and muscle signals, analyze the influence of the patient's facial expression information and muscle signals on the rehabilitation training actions, and correct the patient's rehabilitation training actions according to the analysis results.

10. A data processing system based on virtual reality according to claim 9, wherein: The system further includes a training guidance module; The training guidance module is used to analyze the training volume of the user under the corrected training actions and adjust the patient's training volume.

Images