A rehabilitation training method and device based on virtual reality technology

Abstract

The application discloses a rehabilitation training method and device based on virtual reality technology, and relates to the field of virtual reality rehabilitation technology.The method comprises the following steps: after completing the activity range and corresponding orientation calibration, generating rehabilitation training tasks with different difficulty coefficients; the rehabilitation training task is continuous target tracking, specifically, in response to a user control request, controlling a ring-shaped cursor to move along the movement track of a target ball, so that the target ball is kept in the ring-shaped cursor; difficulty coefficient calibration is performed on the training load of the rehabilitation training task according to system parameters of the rehabilitation training task, wherein the system parameters comprise the complexity of the movement track of the target ball, the radius of the target ball, the radius of the ring-shaped cursor, the error range, the fixed effect factor and the relative effect factor; and the rehabilitation training task with corresponding system parameters is executed according to the required difficulty coefficient.The application can realize accurate regulation and control of the training load, and is suitable for rehabilitation training requirements of different recovery degrees.

Description

Technical Field

[0001] This invention relates to the field of virtual reality rehabilitation technology, and more specifically, to a rehabilitation training method and apparatus based on virtual reality technology. Background Technology

[0002] Motor dysfunction is a major cause of disability after stroke. Among the brain's motor function areas, the cortical regions that control upper limb movement account for the largest proportion, and 80% of stroke patients experience hand function impairment. Upper limb motor dysfunction can severely affect daily living abilities.

[0003] Currently, clinical rehabilitation training for patients mainly involves one-on-one, hands-on instruction. This method has several drawbacks: training efficiency and intensity are difficult to guarantee, and the training effect is influenced by the therapist's skill level; secondly, it's impossible to precisely control and record training parameters such as speed, trajectory, and intensity, making it difficult to optimize these parameters for the best treatment plan; thirdly, there's a lack of objective data to evaluate the relationship between training parameters and rehabilitation effects, hindering in-depth research into the patterns of neurorehabilitation; and fourthly, it cannot provide patients with real-time, intuitive feedback, making the training process monotonous and tedious, with patients often passively receiving treatment and lacking initiative in their participation.

[0004] Virtual reality (VR) technology offers advantages such as low cost and high operability in clinical neurorehabilitation, serving as an effective tool for cognitive research, assessment, and rehabilitation treatment. In VR-based rehabilitation training systems, training tasks are typically designed as games. Engaging and interactive activities, along with real-time feedback, help enhance patient engagement during training. Furthermore, precise control and recording of various parameters during training allows for dynamic adjustment of training strategies to achieve optimal rehabilitation outcomes.

[0005] Generally, adjusting the rehabilitation training load by changing the difficulty of the training tasks is crucial for improving the effectiveness of rehabilitation training. Excessively high or low task loads can cause patients to lose motivation. Studies have shown that rehabilitation training under appropriate loads can effectively improve the patient's rehabilitation outcomes. Therefore, achieving precise control of the rehabilitation training load is essential for the design of a rehabilitation training system.

[0006] Adjusting training load can usually be achieved by changing the training difficulty. Common methods include the following:

[0007] 1. Experimental paradigms based on cognitive function research. Typical paradigms include N-back and stroop tests. These paradigms are relatively mature in technology and are mainly used to study cognitive function. They typically employ rapid-response experimental designs and are not suitable for continuous, long-term motor training tasks.

[0008] 2. Experimental paradigms based on motor function research. These include the following types:

[0009] a) Operational tasks, such as piloting aircraft, driving cars, or shooting at flying objects. These tasks typically adjust their difficulty by changing their complexity (e.g., setting up multi-tasks), but lack an objective quantitative description of the training load.

[0010] b) Reaching task. Fitts' theorem can be used to quantitatively describe training load, but this type of exercise is intermittent and discontinuous.

[0011] c) Continuous target tracking. This is a common form of continuous motion training. Difficulty can be adjusted by changing the complexity of the tracking trajectory (e.g., increasing dimensionality, increasing trajectory perturbation), and increasing speed. However, there is no quantitative description of the task load, and the difficulty level is not linked to subjective perception.

[0012] Overall, existing virtual reality rehabilitation training systems have a vague definition of training load, the setting of difficulty coefficients is somewhat arbitrary, there is a lack of quantitative description of training load, and no connection is established between difficulty coefficients and subjective feelings. Summary of the Invention

[0013] To address the problems of existing virtual reality rehabilitation training systems, such as vague definitions of training load, arbitrary setting of difficulty coefficients, lack of quantitative description of training load, and failure to establish a link between difficulty coefficients and subjective feelings, this invention provides a virtual reality rehabilitation training method and device that can achieve quantitative description and precise control of rehabilitation training load.

[0014] To achieve the above-mentioned objectives of this invention, the technical solution adopted is as follows:

[0015] A rehabilitation training method based on virtual reality technology includes the following steps:

[0016] After completing the activity range and corresponding location marking, a series of rehabilitation training tasks with a certain degree of difficulty are generated;

[0017] The rehabilitation training task is continuous target tracking, specifically, in response to a user control request, controlling a circular cursor to move along the trajectory of the target ball so that the target ball remains within the circular cursor;

[0018] The difficulty coefficient of the rehabilitation training task is calibrated based on the system parameters of the rehabilitation training task. The system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor.

[0019] Perform rehabilitation training tasks with corresponding system parameters according to the required level of difficulty.

[0020] Specifically, the activity range and corresponding orientation calibration are used to adapt to the rehabilitation training needs of users with different recovery levels, including: obtaining the activity range boundary of the user's control cursor and initially calibrating the activity range; then obtaining the spatial coordinates of the cursor in each orientation of the activity range boundary, and completing the spatial coordinate calibration of each orientation in response to the orientation calibration request; finally obtaining the user's name or ID and completing the calibration work in response to the calibration request.

[0021] Furthermore, the formula for calculating the difficulty coefficient is as follows:

[0022]

[0023] Where DI is the difficulty coefficient, a is the fixed effect factor, b is the relevant effect factor, and P is the P-value. C Let r be the complexity of the trajectory of the target ball. cursor r is the radius of the circular cursor. target Let be the radius of the target ball, and ∈ be the error range.

[0024] Furthermore, a and b are obtained through regression analysis using preliminary experiments; P C It is obtained by evaluating entropy.

[0025] Furthermore, the r cursor Greater than or equal to r target .

[0026] Preferably, the continuous target tracking further includes: providing feedback to the user whether the target ball is located within the ring cursor by changing the color of the ring cursor.

[0027] Furthermore, the method for determining whether the target ball is located within the circular cursor is as follows:

[0028] Let d be the distance between the center point of the target ball and the center point of the circular cursor. When d ≤ r cursor -r target When +∈, the target ball is determined to be within the circular cursor.

[0029] The rehabilitation training method based on virtual reality technology further includes: displaying statistical information after completing rehabilitation training, including training results, training growth curve, tracking stability, and ranking of training results for all users; and storing various data generated by rehabilitation training.

[0030] A rehabilitation training device based on virtual reality technology, used to implement the rehabilitation training method based on virtual reality technology, includes:

[0031] Processor, display, controller;

[0032] The processor is electrically connected to the display and the controller, respectively;

[0033] After completing the activity range and corresponding orientation calibration, the processor generates a series of rehabilitation training tasks with a certain difficulty level. The rehabilitation training task is continuous target tracking. Specifically, in response to user control requests, it controls a circular cursor to move along the trajectory of a target ball, keeping the target ball within the circular cursor. The processor calibrates the difficulty level of the rehabilitation training task based on the system parameters of the rehabilitation training task. The system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor. Based on the required difficulty level, the processor executes the rehabilitation training task with the corresponding system parameters.

[0034] The display is used to show the target ball, the circular cursor, and statistical information;

[0035] The controller is used to respond to user control requests and control the circular cursor to move along the trajectory of the target ball, so that the target ball remains within the circular cursor.

[0036] Furthermore, the rehabilitation training device based on virtual reality technology also includes a memory and an optical locator;

[0037] The memory and the optical locator are electrically connected to the processor, respectively.

[0038] The memory stores various data generated for rehabilitation training; the optical locator is used to obtain the spatial coordinates corresponding to the controller and transmit them to the processor for processing.

[0039] The beneficial effects of this invention are as follows:

[0040] Compared with existing technologies, this invention provides a rehabilitation training method and device based on virtual reality technology. It adopts a training load difficulty coefficient calibration method based on continuous target tracking. This method can quantitatively describe the training load and establish a connection between the difficulty coefficient of the training load and subjective feelings, enabling precise control of the rehabilitation training load. In addition, it also proposes an activity range calibration method, which can adapt to the rehabilitation training needs of users with different recovery levels. Attached Figure Description

[0041] Figure 1 This is a flowchart of the method of the present invention.

[0042] Figure 2 A schematic diagram of the interface used to define the activity area.

[0043] Figure 3 This is a schematic diagram of the rehabilitation training interface.

[0044] Figure 4 This is a schematic diagram of the device of the present invention. Detailed Implementation

[0045] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0046] Example 1

[0047] The present invention describes a rehabilitation training method based on virtual reality technology, such as... Figure 1 As shown, the steps are as follows:

[0048] After completing the activity range and corresponding location marking, a series of rehabilitation training tasks with a certain degree of difficulty are generated;

[0049] The rehabilitation training task is continuous target tracking, specifically, in response to a user control request, controlling a circular cursor to move along the trajectory of the target ball so that the target ball remains within the circular cursor;

[0050] The difficulty coefficient of the rehabilitation training task is calibrated based on the system parameters of the rehabilitation training task. The system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor.

[0051] Perform rehabilitation training tasks with corresponding system parameters according to the required level of difficulty.

[0052] In one specific embodiment, the required level of difficulty is selected by the medical professional.

[0053] Preferably, the activity range and corresponding orientation calibration are used to adapt to the rehabilitation training needs of users with different recovery levels, including: obtaining the activity range boundary of the user's control cursor and initially calibrating the activity range; then obtaining the spatial coordinates of the cursor in each orientation of the activity range boundary, and completing the spatial coordinate calibration of each orientation in response to the orientation calibration request; finally obtaining the user's name or ID and completing the calibration work in response to the calibration request.

[0054] like Figure 2 The diagram shown is a schematic diagram for defining the range of motion.

[0055] Since different users have different abilities to move on the affected side, this method of defining the range of motion can ensure that each user can perform rehabilitation training to the maximum extent within their capabilities.

[0056] Specifically, the formula for calculating the difficulty level is as follows:

[0057]

[0058] Where DI is the difficulty coefficient, a is the fixed effect factor, b is the relevant effect factor, and P is the P-value. C Let r be the complexity of the trajectory of the target ball. cursor r is the radius of the circular cursor. target Let be the radius of the target ball, and ∈ be the error range.

[0059] Furthermore, a and b are obtained through regression analysis using preliminary experiments; P C It is obtained by evaluating entropy.

[0060] Furthermore, the r cursor Greater than or equal to r target .

[0061] In a specific embodiment, when the training scenario is a two-dimensional space, the coordinates of the target ball in the x and y axes can be represented as time-varying sequences x(t) and y(t), respectively. Furthermore, entropy can be used to evaluate the complexity of the motion trajectory, including approximate entropy, sample entropy, and fuzzy entropy. Here, approximate entropy is used for evaluation. The approximate entropy ApEn_x of x(t) and the approximate entropy ApEn_y of y(t) are calculated using the built-in approximateEntropy function in Matlab. Then, PC = ApEn_x + ApEn_y.

[0062] Approximate entropy is a method for describing the self-similarity of a time series in terms of patterns. It is used to describe the irregularity of complex systems. The greater the irregularity, the greater the complexity. Here, we only use approximate entropy as an example to illustrate the calculation of the complexity of a motion trajectory, but any other method that can quantitatively describe the complexity of a time series can be used.

[0063] In one specific embodiment, r target For a preset fixed value, by changing r cursor The size of the target ball thus changes the difficulty coefficient DI; the complexity P of the motion trajectories of different target balls generated using the same rule. C The difference between them is less than the first threshold, and its impact on the difficulty coefficient DI is negligible.

[0064] In a specific embodiment, the value range of ∈ is r. target 0.05-0.2 times.

[0065] like Figure 3 The image shown is a schematic diagram of the rehabilitation training interface.

[0066] In one specific embodiment, the training scenarios for the rehabilitation training include one-dimensional space, two-dimensional space, and three-dimensional space.

[0067] In one specific embodiment, the duration of a single rehabilitation training session ranges from 10 to 100 seconds.

[0068] Furthermore, the continuous target tracking also includes: providing feedback to the user whether the target ball is located within the circular cursor by changing the color of the circular cursor.

[0069] In this embodiment, when the target ball is located within the circular cursor, the circular cursor is displayed in green; otherwise, the cursor is displayed in red.

[0070] Furthermore, the method for determining whether the target ball is located within the circular cursor is as follows:

[0071] Let d be the distance between the center point of the target ball and the center point of the circular cursor. When d ≤ r cursor -r target When +∈, the target ball is determined to be within the circular cursor.

[0072] In one specific embodiment, 'a' is a fixed-effect factor reflecting the basic training load; 'b' is a relevant-effect factor reflecting the degree to which the training load is affected by regulatory factors, including P. C r cursor r target and ∈.

[0073] In one specific embodiment, the values ​​of a and b differ for different groups; the different groups are groups with different age ranges and different degrees of hemiplegia; all different groups should be covered when conducting the pre-test.

[0074] In a specific embodiment, a and b are obtained through regression analysis in a preliminary experiment, as follows:

[0075] First, set r target ∈ is a fixed value, and n different r are set. cursor , and r cursor The value of r should cover the preset maximum range, allowing the user to perform several rehabilitation training sessions, with each session's r value representing a specific value. target From the set n r cursor One is randomly selected from the options, ensuring that each level appears at least once, and the duration of each rehabilitation training session is fixed. After all rehabilitation training sessions are completed, the difficulty coefficient of each rehabilitation training session is quantitatively represented. Based on the results of the quantitative representation of the difficulty coefficient, regression analysis is performed using the least squares method to obtain a and b.

[0076] In one specific embodiment, the quantitative representation of the difficulty coefficient of each rehabilitation training session includes both behavioral performance quantitative representation and subjective scale quantitative representation, as detailed below:

[0077] (1) The quantitative representation of behavioral performance specifically involves calculating the root mean square error (RMSE) within each trial as a quantitative representation of the difficulty coefficient of the current trial. A larger RMSE indicates a greater difficulty coefficient. Taking a two-dimensional case as an example, assuming there are N sampling points in a trial, (x i ,y i ) and (x i ′,y i ′) are the center coordinates of the target ball and the circular cursor at the i-th sampling point, respectively, and the calculation formula is as follows:

[0078]

[0079] (2) The subjective scale represents the difficulty coefficient. Here, the NASA-TLX scale is used. The NASA-TLX scale is a workload assessment scale. After a trial, the user scores the training load under the current trial. Unlike the RMSE, this method can establish a connection between the user's subjective feelings and the radius of the circular cursor.

[0080] Preferably, a rehabilitation training method based on virtual reality technology further includes: displaying statistical information after completing rehabilitation training, the statistical information including training results, training growth curve, tracking stability and ranking of training results of all users; and storing various data generated by rehabilitation training.

[0081] In one specific embodiment, the statistical information is beneficial in motivating users' training enthusiasm.

[0082] In one specific embodiment, the difficulty level is adjusted based on the user's previous training performance. Specifically, this involves adjusting r based on the user's previous training performance data stored in the database. cursor The size of the load should be adjusted to avoid discouraging users, and this adjustment method can achieve precise adjustment of the training load.

[0083] In one specific embodiment, different users should select corresponding a and b from the stored data when performing rehabilitation training.

[0084] Example 2

[0085] The rehabilitation training method based on virtual reality technology described in this invention includes the following steps:

[0086] After completing the activity range and corresponding location marking, a series of rehabilitation training tasks with a certain degree of difficulty are generated;

[0087] The rehabilitation training task is continuous target tracking, specifically, in response to a user control request, controlling a circular cursor to move along the trajectory of the target ball so that the target ball remains within the circular cursor;

[0088] The difficulty coefficient of the rehabilitation training task is calibrated based on the system parameters of the rehabilitation training task. The system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor.

[0089] Perform rehabilitation training tasks with corresponding system parameters according to the required level of difficulty.

[0090] Preferably, a rehabilitation training method based on virtual reality technology further includes: displaying statistical information after completing rehabilitation training, the statistical information including training results, training growth curve, tracking stability and ranking of training results of all users; and storing various data generated by rehabilitation training.

[0091] The present invention provides a rehabilitation training device based on virtual reality technology, used to implement the rehabilitation training method based on virtual reality technology, such as... Figure 4 As shown, it includes:

[0092] Processor, display, controller;

[0093] The processor is electrically connected to the display and the controller, respectively;

[0094] After completing the activity range and corresponding orientation calibration, the processor generates a series of rehabilitation training tasks with a certain difficulty level. The rehabilitation training task is continuous target tracking. Specifically, in response to user control requests, it controls a circular cursor to move along the trajectory of a target ball, keeping the target ball within the circular cursor. The processor calibrates the difficulty level of the rehabilitation training task based on the system parameters of the rehabilitation training task. The system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor. Based on the required difficulty level, the processor executes the rehabilitation training task with the corresponding system parameters.

[0095] The display is used to show the target ball, the circular cursor, and statistical information;

[0096] The controller is used to respond to user control requests and control the circular cursor to move along the trajectory of the target ball, so that the target ball remains within the circular cursor.

[0097] A rehabilitation training method device based on virtual reality technology also includes a memory and an optical locator;

[0098] The memory and the optical locator are electrically connected to the processor, respectively.

[0099] The memory stores various data generated for rehabilitation training; the optical locator is used to obtain the spatial coordinates corresponding to the controller and transmit them to the processor for processing.

[0100] In one specific embodiment, the processor is a computer host; the display includes a head-mounted display and an external display, which can effectively avoid the discomfort caused by wearing the head-mounted display for a long time; different types of controllers can be selected according to the user's needs to carry out rehabilitation training for different limbs.

[0101] In this embodiment, the main program used for rehabilitation training is developed using Unity 3D; the controller, head-mounted display, and optical locator are HTC VIVE devices.

[0102] Specifically, the formula for calculating the difficulty level is as follows:

[0103]

[0104] Where DI is the difficulty coefficient, a is the fixed effect factor, b is the relevant effect factor, and P is the P-value.C Let r be the complexity of the trajectory of the target ball. cursor r is the radius of the circular cursor. target Let be the radius of the target ball, and ∈ be the error range.

[0105] In one specific embodiment, when the user selects different displays, the radius r of the target sphere changes accordingly. target Adjustments will be made, setting the radius r of the target sphere... target The angle should be no less than 1° from the user's perspective, as shown in the following formula:

[0106] r target ≥D×π / 180

[0107] Where D is the distance between the user and the monitor.

[0108] In one specific embodiment, initialization is required before defining the activity range, including loading relevant forms and loading hardware drivers.

[0109] In summary, this invention employs a training load difficulty coefficient calibration method based on continuous target tracking. This method can quantitatively describe the training load and establish a link between the difficulty coefficient of the training load and subjective feelings, enabling precise control of the rehabilitation training load. In addition, an activity range calibration method is proposed, which can adapt to the rehabilitation training needs of users with different recovery levels.

[0110] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A rehabilitation training method based on virtual reality technology, characterized in that, The steps include the following: After completing the activity range and corresponding orientation calibration, a series of rehabilitation training tasks with a certain degree of difficulty are generated. Before calibrating the activity range, initialization is required, including loading relevant forms and loading hardware drivers. The activity range calibration includes: obtaining the activity range boundary of the user control cursor and initially calibrating the activity range; then obtaining the spatial coordinates of the cursor in each orientation of the activity range boundary, and completing the spatial coordinate calibration in each orientation in response to the orientation calibration request; finally, obtaining the user's name or ID and completing the calibration work in response to the calibration request. The rehabilitation training task is continuous target tracking, specifically, in response to a user control request, controlling a circular cursor to move along the trajectory of the target ball so that the target ball remains within the circular cursor; The difficulty coefficient of the rehabilitation training task is calibrated based on the system parameters of the rehabilitation training task. These system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor. The calculation formula for the difficulty coefficient calibration is as follows: in, denoted as the difficulty coefficient, 'a' as the fixed effect factor, 'b' as the relevant effect factor, and 'PC' as the complexity of the target ball's trajectory. The radius of the circular cursor. Let the radius of the target ball be . This refers to the error range; Perform rehabilitation training tasks with corresponding system parameters according to the required level of difficulty.

2. The rehabilitation training method based on virtual reality technology according to claim 1, characterized in that: The a and b The results were obtained through regression analysis based on preliminary experiments; It is obtained by evaluating entropy.

3. The rehabilitation training method based on virtual reality technology according to claim 2, characterized in that: The Greater than or equal to .

4. The rehabilitation training method based on virtual reality technology according to claim 1, characterized in that: The continuous target tracking also includes: providing feedback to the user whether the target ball is located within the circular cursor by changing the color of the circular cursor.

5. A rehabilitation training method based on virtual reality technology according to claim 4, characterized in that: The method for determining whether the target ball is located within the circular cursor is as follows: set up d The distance between the center point of the target ball and the center point of the circular cursor, when If the target ball is within the circular cursor, then the target ball is determined to be within the circular cursor.

6. The rehabilitation training method based on virtual reality technology according to claim 1, characterized in that: Also includes: After completing rehabilitation training, statistical information will be displayed, including training results, training growth curve, tracking stability, and ranking of training results for all users. It will also store various data generated during rehabilitation training.

7. A rehabilitation training device based on virtual reality technology, used to implement the rehabilitation training method based on virtual reality technology according to any one of claims 1-6, characterized in that, include: Processor, display, controller; The processor is electrically connected to the display and the controller, respectively; After completing the activity range and corresponding orientation calibration, the processor generates a series of rehabilitation training tasks with a certain difficulty level. The rehabilitation training task is continuous target tracking. Specifically, in response to user control requests, it controls a circular cursor to move along the trajectory of a target ball, keeping the target ball within the circular cursor. The processor calibrates the difficulty level of the rehabilitation training task based on the system parameters of the rehabilitation training task. The system parameters include the complexity of the target ball's trajectory, the radius of the target ball, the radius of the circular cursor, the error range, the fixed effect factor, and the relative effect factor. Based on the required difficulty level, the processor executes the rehabilitation training task with the corresponding system parameters. The display is used to show the target ball, the circular cursor, and statistical information; The controller is used to respond to user control requests and control the circular cursor to move along the trajectory of the target ball, so that the target ball remains within the circular cursor.

8. A rehabilitation training device based on virtual reality technology according to claim 7, characterized in that: It also includes memory and optical locators; The memory and the optical locator are electrically connected to the processor, respectively. The memory stores various data generated for rehabilitation training; The optical positioner is used to obtain the spatial coordinates corresponding to the controller and transmit them to the processor for processing.

Images