Cerebral stroke posture control training method and system based on virtual reality technology

Abstract

The invention discloses a stroke posture control training method and system based on a virtual reality technology, and belongs to the technical field of stroke rehabilitation training. The problems that in traditional posture control training, the training effect is poor, the enthusiasm of a patient is not high, and personalized training is lacked are solved. The method comprises the following steps: S1, constructing a stroke posture training database; s2, generating a VR virtual evaluation scene based on a stroke posture training database according to the rehabilitation target of the patient and the severity of the illness state; s3, performing posture control training according to the generated VR virtual evaluation scene, and capturing and recording posture information of the patient in the training process; and S4, analyzing and processing the acquired posture information, and generating a posture control training plan according to an analysis result. The VR technology and the CPPC technology are combined, wide application of the CPPC technology can be promoted, the requirement of most patients for pursuing high-quality rehabilitation training is met, and the situation of medical resource shortage is relieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of stroke rehabilitation training, and in particular relates to a stroke posture control training method and system based on virtual reality technology. Background Art

[0002] Stroke is a common cerebrovascular disease with a high incidence and disability rate. Stroke patients often suffer from various limb motor dysfunctions, among which impaired postural control ability seriously affects the patient's quality of life. Postural control refers to the body's ability to maintain balance and coordination. Due to abnormal sensory input, central integration and motor output, stroke patients have reduced postural control ability and need long-term rehabilitation training to restore their postural control ability. However, traditional rehabilitation training methods are generally not interesting, resulting in poor patient compliance and poor rehabilitation training effects.

[0003] In recent years, VR technology has been widely used in the field of rehabilitation, mainly through visual and auditory feedback, to increase the fun of training, stimulate the enthusiasm of patients to train, and thus improve the training effect. Although VR technology shows great potential for stroke rehabilitation, its application in clinical practice is limited by high cost and technical difficulties.

[0004] At present, CPPC technology is mainly used in clinical practice to train stroke patients in posture control, and its clinical effect is remarkable, which has been highly recognized and evaluated by peers. However, the traditional rehabilitation model mainly relies on the clinical experience and time arrangement of therapists, which cannot meet the training needs of a large number of patients, nor can it provide an immersive training environment for patients, which is not conducive to patients' long-term persistence in exercise. Summary of the invention

[0005] In view of the problems existing in the prior art of traditional posture control training, such as poor training effect, low patient enthusiasm, and lack of personalized training, the present invention provides a stroke posture control training method and system based on virtual reality technology.

[0006] The technical solution adopted by the present invention is as follows:

[0007] A stroke posture control training method based on virtual reality technology comprises the following steps:

[0008] S1: Construct a stroke posture training database;

[0009] S2: Generate VR virtual assessment scenarios based on the stroke posture training database according to the patient's rehabilitation goals and the severity of the disease;

[0010] S3: Perform posture control training based on the generated VR virtual assessment scene, and capture and record the patient's posture information during the training process;

[0011] S4: Analyze and process the acquired posture information, and generate a posture control training plan based on the analysis results.

[0012] After adopting this technical solution, the present invention formulates a VR virtual assessment scene suitable for the patient according to the patient's own condition. The patient performs posture control training in the scene and collects posture information during the training process. After analyzing and processing the collected information, a VR virtual training scene that is more suitable for the patient is formulated to improve the training effect. By combining VR technology with CPPC technology, the present invention can promote the widespread application of CPPC technology, meet the needs of most patients for high-quality rehabilitation training, and alleviate the shortage of medical resources.

[0013] Preferably, S3 includes the following two posture control evaluations and two posture information collections:

[0014] For the first time: without any external intervention, posture control training was carried out based on the generated VR virtual assessment scene, and the patient's posture information was collected and recorded in real time during the training process;

[0015] The second time: posture control training is conducted again based on the generated VR virtual assessment scene. The patient's posture information is collected and recorded in real time during the training process, and the collected information is analyzed in real time. When the analysis results show that the deviation between the patient's posture and the standard posture exceeds the threshold, the voice prompt is activated to remind the patient to correct the posture through voice, and the corrected posture information is recorded.

[0016] After adopting this technical solution, since it is difficult for patients to correct themselves to the standard posture in the early stage of training, if patients are required to adjust to the standard posture through voice prompts all the time, it may undermine the patient's self-confidence and affect the smooth progress of the training. Therefore, the present invention conducts posture control training twice in the VR virtual assessment scene, one is completely without intervention, and the other is with voice prompts. The patient's posture can be corrected through voice prompts to obtain the corrected posture information. If the corrected posture differs too much from the standard posture, the threshold for starting the voice prompt is directly set according to the corrected posture information at the beginning of training. The patient's posture is compared with the corrected posture. If the deviation is within the set range, no voice prompt is given. If it exceeds the set range, a voice prompt is given. Requiring patients to meet the highest standards that they can achieve can greatly enhance their self-confidence and help the continuation of training.

[0017] Preferably, in S4, a posture control training plan is jointly formulated based on the posture information without intervention and the corrected posture information, and a deviation threshold for starting voice prompts during posture training is determined based on the difference between the corrected posture information or standard posture information and the posture information without intervention.

[0018] After adopting this technical solution, the patient's actual movement status can be more clearly understood based on the posture information without intervention, while the corrected posture information can understand the best state the patient can currently adjust to. The training plan formulated by combining the two is more objective.

[0019] Preferably, the deviation threshold in the posture control training program is adjusted according to the training progress.

[0020] After adopting this technical solution, as the training progresses, the patient's posture adjusted according to the voice prompts gradually approaches the standard posture. Therefore, it is necessary to continuously adjust the deviation threshold according to the training progress to improve the training requirements.

[0021] Preferably, when it is determined according to the corrected posture information that the deviation between the corrected posture and the standard posture is less than or equal to 5%, the standard posture is compared with the patient's posture to determine a threshold for starting the voice prompt;

[0022] When it is determined based on the corrected posture information that the deviation between the corrected posture and the standard posture is greater than 5%, the corrected posture is compared with the patient's posture to determine the threshold for starting the voice prompt.

[0023] After adopting this technical solution, when the deviation between the patient's corrected posture and the standard posture is less than or equal to 5%, it means that the patient's own condition is good and can be corrected to the standard posture through training in a shorter time. Therefore, the threshold can be set directly according to the standard posture.

[0024] Preferably, the stroke posture training database constructed by S1 consists of multiple VR standard training scene segments.

[0025] Preferably, S2 selects appropriate VR standard training scene segments from the posture training database according to the patient's rehabilitation goals and the severity of the disease, and sorts and splices the selected scene segments to combine the selected VR standard training scene segments into a continuous VR virtual training scene.

[0026] After adopting this technical solution, a training plan suitable for patients can be formulated by combining selected VR standard training scene fragments into continuous VR virtual training scenes, taking full account of individual differences of patients, improving the scientificity and effectiveness of training, and accelerating the rehabilitation process.

[0027] A system for implementing a stroke posture control training method based on virtual reality technology, comprising the following modules:

[0028] A virtual environment generation module is used to select VR standard training scene segments in the stroke posture training database and combine the selected scene segments into a continuous VR virtual training scene;

[0029] A posture capture module, used to collect the patient's posture information;

[0030] Feedback control module, used to provide real-time feedback on the patient's training status;

[0031] Individualized program customization module, used to develop posture control training plans based on the patient's basic conditions and collected posture information;

[0032] Human-computer interaction module, used to realize the interaction between patients and VR virtual training scenes;

[0033] The data analysis and processing module is used to receive the data collected by each module, analyze and process the data, and then feed it back to the corresponding module.

[0034] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0035] 1. The present invention develops a VR virtual assessment scene suitable for the patient according to the patient's own condition. The patient performs posture control training in the scene and collects posture information during the training process. After analyzing and processing the collected information, a VR virtual training scene that is more suitable for the patient is developed to improve the training effect.

[0036] 2. By combining VR technology with CPPC technology, the present invention can promote the widespread application of CPPC technology, meet the needs of most patients for high-quality rehabilitation training, and alleviate the shortage of medical resources.

[0037] 3. The present invention can greatly improve the patient's participation and enthusiasm through an interesting virtual reality environment and advanced human-computer interaction mode. Patients are more willing to take the initiative to train, which is conducive to the long-term implementation of the training plan.

[0038] 4. The present invention can comprehensively train patients' posture control ability in different situations through diversified training scenarios, and timely adjust the training posture according to real-time feedback information to improve the accuracy of training.

[0039] 5. The present invention formulates a posture control training plan according to the specific conditions of the patient, makes timely adjustments during the training process, fully considers the individual differences of the patients, improves the scientificity and effectiveness of the training, and accelerates the rehabilitation process.

[0040] 6. The present invention can accurately record and analyze the patient's training data, provide an objective evaluation basis for therapists and doctors, and facilitate tracking of the patient's rehabilitation progress and adjustment of the training plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the architecture diagram of the system in the present invention;

[0042] Figure 2 This is a distribution diagram of high-precision sensors in the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose and advantages of the embodiments of the present application clearer, the technical solution will be clearly and completely described below in combination with the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0044] A stroke posture control training method based on virtual reality technology. Figure 1 The stroke posture control training system based on virtual reality technology shown in the figure specifically includes the following steps:

[0045] S1: Collect the existing standard training action videos used for stroke posture control training, divide the standard training action videos according to single actions, and obtain several standard training action clips. These standard training action clips constitute the stroke posture training database. The stroke posture training database is divided based on the different treatment conditions, and several standard training action clips are stored in different groups according to the different treatment conditions, such as actions for treating walking difficulties, actions for treating body tilt, etc. The standard training action clips are classified and stored according to different conditions, which is convenient for the subsequent virtual environment generation module to capture the actions. On the basis of the division according to the treatment condition, it is also possible to further divide the training actions in each group according to the difficulty of the training actions, so that simple, general, and difficult VR virtual assessment or training scenes can be generated according to the patient's condition and training progress.

[0046] S2: During the evaluation, the operator inputs information such as the patient's rehabilitation goals and the severity of the disease into the system. The virtual environment generation module analyzes an evaluation plan suitable for the patient based on the above information. For example, for those with difficulty walking, core muscle strength training is mainly used to activate the patient's unconscious proprioception and medial motor system; for example, in the early stage of stroke, patients mainly undergo bed-chair transfer training, sitting balance training, and standing balance training. The evaluation plan includes the duration of the evaluation (the total duration of the clips that need to be captured) and the proportion of actions that need to be captured in each group in the posture training database (including the number and difficulty of each training action that needs to be captured). Based on the evaluation plan, the virtual environment generation module automatically captures several standard training action clips in each group of the posture training database according to the proportion and quantity, and converts them into VR standard training scene clips that can be displayed in the VR device (in other embodiments, they can also be converted into VR standard training scene clips before being stored in the posture training database). Then, according to the training principles of trunk first, then limbs, lower limbs first, then upper limbs, simple first, then complex, passive first, then active, the selected scene segments are sorted and spliced, so that the selected VR standard training scene segments are combined into a continuous VR virtual assessment scene;

[0047] S3: The patient puts on the VR device and Figure 2 The key parts of the body (such as joints, spine, etc.) shown are equipped with high-precision sensors for detecting the patient's movement state. The high-precision sensors are used to capture the patient's body posture information (such as center of gravity, support phase, swing phase, etc.) in real time, and the inertial measurement unit technology or optical tracking technology is used to accurately measure the movement angle, acceleration and other parameters, and the collected data is transmitted to the posture capture module for storage. In this embodiment, after the patient wears the VR device and the high-precision sensor, two posture control assessments and two posture information collections are performed, as follows:

[0048] First time: The patient undergoes posture control training based on the VR virtual assessment scene displayed in the VR device. The patient is not intervened at all during the training process, and the patient's posture information is collected and recorded in real time during the training process.

[0049] The second time: the patient performs posture control training again according to the VR virtual assessment scene displayed in the VR device, and analyzes the collected posture information in real time. When the analysis results show that the deviation between the patient's posture and the standard posture exceeds the threshold (the threshold is 5% in this embodiment), the feedback control module is started, and the feedback control module gives a voice prompt (in other embodiments, a red light prompt can also be given through a VR virtual character image, giving the patient a dual experience of vision and hearing to enhance the effect of interaction). At the same time, the patient can also be informed by voice how to correct it. For example, if the patient's center of gravity deviates from the movement trajectory, the patient is reminded by voice to adjust to the correct position. The patient adjusts the training posture according to the voice prompt, and obtains the adjusted posture information in real time through the posture capture module, and changes the content of the voice prompt according to the adjusted posture information to help the patient further adjust the training posture. When the data analysis and processing module determines that the posture information under the voice prompt deviates less than 1% from the posture information under the previous voice prompt, it means that the patient has reached the maximum adjustment degree, and the current posture information is recorded as the corrected posture information;

[0050] S4: The data analysis and processing module analyzes and processes the collected data, and transmits the analysis results to the individualized program customization module. The individualized program customization module jointly formulates a posture control training plan based on the posture information without intervention and the corrected posture information. When formulating a posture control training plan, the individualized program customization module will analyze in detail the training information captured in real time by the high-precision sensor during the evaluation process, and judge whether the formulated VR virtual evaluation scene is suitable for the actual training of the patient based on the training information. If the judgment result is suitable, the VR virtual evaluation scene can be directly used as the VR virtual training scene; if the judgment result is not suitable, such as according to the data analysis, some actions in the VR virtual training scene are difficult and obviously not suitable for the patient, and the patient completion rate is very low, these actions can be replaced and simpler training actions can be selected. As for the specific selection of simple actions, it can be judged according to the patient's completion rate of the action during the evaluation process; for example, if the patient is obviously physically exhausted during the evaluation process, the number of standard training action clips can be appropriately deleted to make the final VR virtual training scene more suitable for the patient. During the VR virtual training process, high-precision sensors are used to capture the patient's body posture information in real time, which is convenient for analyzing the patient's training effect. The training plan can be adjusted accordingly according to the training effect. At the same time, the deviation threshold for starting the voice prompt during the training process is determined according to the corrected posture information or the standard posture information. Specifically, when the deviation between the corrected posture and the standard posture is less than or equal to 5% according to the corrected posture information, the standard posture is compared with the patient's posture to determine the threshold for starting the voice prompt; when the deviation between the corrected posture and the standard posture is greater than 5% according to the corrected posture information, the corrected posture is compared with the patient's posture to determine the threshold for starting the voice prompt. The deviation threshold in the entire posture control training plan is adjusted according to the training process. For example, if the patient's muscle strength is weak in the early stage and the posture control ability is poor, the deviation threshold is determined according to the best level that the patient can achieve. As the training progresses, the patient's muscle strength gradually recovers, and the deviation threshold is continuously adjusted until the patient is required to use the standard posture.

[0051] During the posture control training, the patient can interact with the VR virtual training scene through the human-computer interaction module; for example, during the training process, the system conveys training instructions and feedback information to the patient through clear and easy-to-understand voice, and the patient can also interact through different gestures (such as clenching a fist to stop training, waving to switch training scenes, etc.). At the same time, the human-computer interaction module can also convert the training plan into several levels, calculate the proportion according to the duration and difficulty of the training action, and display the training progress in the virtual scene after the patient passes the level, and give the patient positive visual and auditory feedback, so as to improve the patient's training effect through gamification. Before each training session, the system uses animation to help patients review the problems in the last training, increase the interaction with patients, and improve the patient's enthusiasm for training. In addition, each time the patient completes a training session, he can light up an area. After completing all the training, a certificate can be automatically generated, and congratulations to the patient for completing all the training plans.

[0052] The above-mentioned embodiments only express the specific implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A stroke posture control training method based on virtual reality technology, characterized in that: The following steps are involved: S1: Construct a stroke posture training database; S2: Generate VR virtual assessment scenarios based on the stroke posture training database according to the patient's rehabilitation goals and the severity of the disease; S3: Perform posture control training based on the generated VR virtual assessment scene, and capture and record the patient's posture information during the training process; S4: Analyze and process the acquired posture information, and generate a posture control training plan based on the analysis results.

2. The method for stroke posture control training based on virtual reality technology according to claim 1, characterized in that: S3 includes the following two posture control assessments and two posture information collections: For the first time: without any external intervention, posture control training was carried out based on the generated VR virtual assessment scene, and the patient's posture information was collected and recorded in real time during the training process; The second time: posture control training is conducted again based on the generated VR virtual assessment scene. The patient's posture information is collected and recorded in real time during the training process, and the collected information is analyzed in real time. When the analysis results show that the deviation between the patient's posture and the standard posture exceeds the threshold, the voice prompt is activated to remind the patient to correct the posture through voice, and the corrected posture information is recorded.

3. The method for stroke posture control training based on virtual reality technology according to claim 2, characterized in that: In S4, a posture control training plan is jointly formulated based on the posture information without intervention and the corrected posture information. At the same time, a deviation threshold for starting the voice prompt during the posture control training process is determined based on the difference between the corrected posture information or the standard posture information and the posture information without intervention.

4. The method for stroke posture control training based on virtual reality technology according to claim 3, characterized in that: The deviation threshold in the posture control training program is adjusted according to the training progress.

5. The method for stroke posture control training based on virtual reality technology according to claim 3, characterized in that: When it is determined based on the corrected posture information that the deviation between the corrected posture and the standard posture is less than or equal to 5%, the standard posture is compared with the patient's posture to determine a threshold for starting the voice prompt; When it is determined based on the corrected posture information that the deviation between the corrected posture and the standard posture is greater than 5%, the corrected posture is compared with the patient's posture to determine the threshold for starting the voice prompt.

6. A stroke posture control training method based on virtual reality technology according to any one of claims 1 to 5, characterized in that: The stroke posture training database constructed by S1 consists of multiple VR standard training scene clips.

7. The method for stroke posture control training based on virtual reality technology according to claim 6, characterized in that: S2 selects appropriate VR standard training scene segments from the posture training database according to the patient's rehabilitation goals and the severity of the disease, and sorts and splices the selected scene segments to combine the selected VR standard training scene segments into a continuous VR virtual training scene.

8. A system for implementing the stroke posture control training method based on virtual reality technology according to any one of claims 1 to 7, characterized in that: Includes the following modules: A virtual environment generation module is used to select VR standard training scene segments in the stroke posture training database and combine the selected scene segments into a continuous VR virtual training scene; A posture capture module, used to collect the patient's posture information; Feedback control module, used to provide real-time feedback on the patient's training status; Individualized program customization module, used to develop posture control training plans based on the patient's basic conditions and collected posture information; Human-computer interaction module, used to realize the interaction between patients and VR virtual training scenes; The data analysis and processing module is used to receive the data collected by each module, analyze and process the data, and then feed it back to the corresponding module.

Images