Virtual simulation training system and method for visual disorder rehabilitation after stroke

Abstract

The invention discloses a virtual simulation training system and method for post-stroke visual disorder rehabilitation. The virtual simulation training system comprises VR glasses, an image generation module, an eye movement detection module, a user interaction module, a feedback evaluation module and a power module. The VR glasses are used for being worn by a patient, so that visual impairment repair is carried out on the patient; the image generation module is used for generating different images so as to provide various different visual restoration schemes; the eye movement detection module is used for synchronizing sight and imaging and acquiring eyeball parameters; the user interaction module is used for enabling a patient to interact with the system and selecting to-be-used options; the feedback evaluation module is used for monitoring and analyzing the reaction time, the accuracy rate and the behavior pattern of the patient in real time; the power module is used for supplying power to other modules. According to the application, the VR technology in the VR equipment is utilized to locally process the image, so that the targeted stimulation on the diseased part is enhanced, and a better rehabilitation effect is provided for the user.

Description

Technical Field

[0001] This application belongs to the technical field of medical rehabilitation, and specifically relates to a virtual simulation training system and method for visual impairment rehabilitation after stroke. Background Art

[0002] Visual impairment is one of the common sequelae after stroke. Approximately 3 / 4 of the patients will present with this symptom, and about 28% - 52% of them show visual field defect (VFD), including hemianopia, quadrantanopia, visual constriction, and scotoma, etc. Among them, homonymous hemianopia is the most common, manifested as the loss of the same half visual field in both eyes. Visual impairment can be divided into three types: visual sensory disorder, visual motor disorder, and visual perception disorder, manifested as decreased visual acuity, visual field defect, strabismus or eye movement disorder, as well as visual neglect, visual hallucination, etc. These disorders seriously affect the patient's daily life and rehabilitation process. Traditional rehabilitation methods for visual field defect include alternative therapy using visual aids (such as prisms), compensatory therapy using intact residual abilities, and restoration therapy, which focus on regenerating the plasticity of nerve tissue by continuously presenting stimuli in the blind area. Currently, traditional visual rehabilitation methods mostly rely on physical training, with limited effects and difficulty in personalized training according to the specific needs of different patients. Therefore, there is an urgent need for a new type of training system to more effectively help patients restore visual function. Summary of the Invention

[0003] This application provides a virtual simulation training system and method for visual impairment rehabilitation after stroke to solve the above technical problems.

[0004] To solve the above technical problems, a technical solution adopted in this application is: A virtual simulation training system for visual impairment rehabilitation after stroke, comprising:

[0005] A VR glasses, used to carry each module, wherein the VR glasses are for the patient to wear, so as to repair the visual impairment of the patient;

[0006] An image generation module, located in the VR glasses, wherein the image generation module is used to generate different images to provide a variety of different visual repair solutions;

[0007] An eye movement detection module, located in the VR glasses, wherein the eye movement detection module is used to synchronize the line of sight with the imaging and obtain eye parameters;

[0008] A user interaction module, connected to the image generation module, wherein the user interaction module is used for the patient to interact with the system and select the options to be used;

[0009] A feedback evaluation module, connected to the image generation module, wherein the feedback evaluation module is used to monitor and analyze the patient's reaction time, accuracy rate, and behavior pattern in real time;

[0010] Power supply module, used to supply power to other modules.

[0011] Furthermore, the image generation module includes:

[0012] A display, located in the eye sockets of the VR glasses, where the display is used to display images;

[0013] An image generator, connected to the display, where the image generator is used to generate different display images on the display.

[0014] Furthermore, the eye movement detection module includes:

[0015] An optical basic module, connected to the feedback evaluation module, where the optical system is used to irradiate the patient's eyes from different angles through multiple light sources;

[0016] A camera, connected to the feedback evaluation module, where the camera is used to capture real-time images of the patient's eyes;

[0017] An image sensor, connected to the feedback evaluation module, where the image sensor is used to identify the patient's pupil and corneal reflection points.

[0018] Furthermore, the optical basic module includes:

[0019] A light source module, used to provide light sources;

[0020] A filter, located in front of the light source module, where the filter is used to control the color and intensity of the irradiated light of the light source;

[0021] A lens, used to reduce aberration and chromatic aberration, ensuring the clarity and accuracy of the image.

[0022] Furthermore, the user interaction module includes:

[0023] A touch screen, connected to the VR glasses through a data cable, where the touch screen is used to display a control interface to the patient and perform touch operations;

[0024] A voice component, connected to the touch screen, where the voice component is used to give voice prompts or receive voice commands from the patient.

[0025] Furthermore, the feedback evaluation module includes:

[0026] A main control chip, connected to the touch screen, camera, optical basic module, image sensor, and image generation module, where the main control chip is used to process image data and calculate eye parameters, and at the same time send control signals to the optical basic module and image generation module according to the instructions of the touch screen;

[0027] A monitoring and evaluation system is installed on the main control chip. Among them, the monitoring and evaluation system is used to record the reaction time of the patient in each training and evaluate each training of the patient.

[0028] A storage module is connected to the main control chip. Among them, the storage module is used to store the training data and evaluation data of each patient.

[0029] Another technical solution adopted in this application is: a virtual simulation training method for visual impairment rehabilitation after stroke, including the following steps:

[0030] Collect the eye movement parameters of the patient.

[0031] Based on the main menu interface and the eye movement parameters, select the training mode and training difficulty.

[0032] Based on different training modes and training difficulties, conduct rehabilitation training on the patient's eyes.

[0033] Based on the rehabilitation training, detect the reaction time and recognition accuracy of the patient, and analyze the patient's behavior pattern.

[0034] Based on the patient's behavior pattern and recognition accuracy, automatically adjust the next training difficulty and give voice suggestions.

[0035] Furthermore, the method for collecting the eye ball data of the patient includes:

[0036] Emit near-infrared light through a light source to irradiate the patient's eyes, and take images of the eyes.

[0037] Process the images of the eyes with an image processing algorithm to obtain the characteristic data of the patient's pupils and corneal reflection points.

[0038] Based on the characteristic data, perform P-CR vector calculation to obtain eye movement parameters.

[0039] Furthermore, the method for conducting rehabilitation training on the patient's eyes based on different training modes and training difficulties includes:

[0040] In response to the training mode being the hemianopia visual field diagnosis mode, set the brightness, frequency, and contrast parameters of the light source, conduct visual field evaluation, compare the visual field defects of the left and right eyes, and provide a basis for personalized visual field rehabilitation therapy.

[0041] In response to the training mode being the hemianopia rehabilitation therapy mode, use infrared or image training to eliminate the interference of slight eye movement, and at the same time perform flickering stimuli of different positions and brightness on the blind area.

[0042] In response to the training mode being the shape recognition image mode, generate different images for the patient to recognize and evaluate the patient's image recognition ability.

[0043] In response to the training mode being the color discrimination image mode, basic colors, color combinations, dynamic color changes, and real-life application training are sequentially performed to improve the patient's color recognition ability;

[0044] In response to the training mode being the motion tracking image mode, images containing different objects moving at various speeds and directions are generated, and the training difficulty is set, which is determined based on the object's moving speed, trajectory, and background interference, for improving the patient's ability to track moving objects;

[0045] In response to the training mode being the contrast adjustment image mode, images of daily objects with different contrasts are generated for improving the patient's contrast recognition ability.

[0046] The beneficial effects of this application are as follows: This application uses the VR technology in the VR device to perform local processing on the imaging, enhancing the targeted stimulation of the diseased area and providing better rehabilitation effects for users. By using the VR device, independent diagnosis and training programs are provided to achieve multi-functional training and provide more efficient training effects for users. By using the VR device, a real environment is constructed with high fidelity, providing a more comprehensive visual simulation for users. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a structural block diagram of an embodiment of the virtual simulation training system for post-stroke visual impairment rehabilitation of this application;

[0048] Figure 2 is Figure 1 a structural block diagram of an embodiment of the feedback evaluation module in

[0049] Figure 3 is Figure 1 a structural block diagram of an embodiment of the eye movement detection module in

[0050] Figure 4 is a schematic flowchart of an embodiment of the virtual simulation training system for post-stroke visual impairment rehabilitation of this application;

[0051] Figure 5 is Figure 1 a schematic flowchart of an embodiment of step S1 in DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] To make the objectives, technical solutions, and advantages of the present invention clearer, the following further elaborates on the present invention in detail with reference to specific embodiments.

[0053] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Therefore, the present invention is not limited by the specific embodiments disclosed in the following specification.

[0054] Refer to Figure 1 , Figure 1 which is a structural block diagram of an embodiment of the virtual simulation training system for post-stroke visual impairment rehabilitation of this application. The system includes a VR glasses 1, an image generation module 2, an eye movement detection module 3, a user interaction module 4, a feedback evaluation module 5, and a power supply module 6.

[0055] The VR glasses 1 are used to carry each module. Among them, the VR glasses 1 are for patients to wear, so as to repair visual impairment for patients; the image generation module 2 is located in the VR glasses 1 and is used to generate different images to provide a variety of different visual repair solutions. The eye movement detection module 3. Is located in the VR glasses 1, where the eye movement detection module 3 is used to synchronize the line of sight with the imaging and obtain eye parameters. The user interaction module 4 is connected to the image generation module 2, where the user interaction module 4 is used for patients to interact with the system and select options to be used. The feedback evaluation module 5 is connected to the image generation module 2, where the feedback evaluation module 5 is used to monitor and analyze the patient's reaction time, accuracy rate, and behavior pattern in real time. The power supply module 6 is used to supply power to other modules.

[0056] Refer to Figure 2 , the image generation module 2 includes a display 21 and an image generator 22. Among them, the display 21 is located in the eye socket of the VR glasses 1, where the display 21 is used to display images. The image generator 22 is connected to the display 21, where the image generator 22 is used to generate different display images on the display 21.

[0057] Refer to Figure 3 , the eye movement detection module 3 includes: an optical basic module 31, a camera 32, and an image sensor 33.

[0058] The optical basic module 31 is connected to the feedback evaluation module 5, where the optical system is used to irradiate the patient's eyes from different angles through a variety of light sources. The camera 32 is connected to the feedback evaluation module 5, where the camera 32 is used to take real-time images of the patient's eyes. The image sensor 33 is connected to the feedback evaluation module 5, where the image sensor 33 is used to identify the patient's pupil and corneal reflection point.

[0059] The optical basic module 31 includes: a light source module 311, a filter 312, and a lens 313. The light source module 311 is used to provide a light source. In this application, near-infrared light (NIR) is used as the light source. Near-infrared light causes less harm to the eyes. A near-infrared light-emitting diode (LED) with a wavelength between 780 - 950 nm is used, which has good reflection characteristics on eye tissues and provides clear corneal reflection signals. The light source layout can accurately obtain information on corneal reflection and the pupil, and the light source irradiates the eyes from multiple angles. Multiple low-power near-infrared light LEDs are distributed around or above and below the eye movement tracking device to ensure appropriate reflection signals can be generated at different eye positions and angles.

[0060] The filter 312 is located in front of the light source module 311. Among them, the filter 312 is used to control the color and intensity of the irradiated light of the light source. The lens 313 is used to reduce aberration and chromatic aberration and ensure the clarity and accuracy of the image. Among them, the lens 313 meets the requirements of photographing the eyes at close range, has a short focal length to obtain a larger field of view to completely photograph the eye area, has good optical performance, reduces aberration and chromatic aberration, and ensures the clarity and accuracy of the image.

[0061] The user interaction module 4 includes: a touch screen 41 and a voice component 42. The touch screen 41 is connected to the VR glasses 1 through a data cable. Among them, the touch screen 41 is used to display a control interface to the patient and perform touch operations. The patient directly selects images, adjusts the contrast, and confirms answers through the touch screen 41. The touch screen 41 should support multi-touch to facilitate complex operations by the patient.

[0062] The voice component 42 is connected to the touch screen 41. Among them, the voice component 42 is used to give voice prompts or receive voice commands from the patient. For some patients with limited motor ability, the voice component 42 integrates voice recognition technology, allowing the patient to select training modes, difficulty levels, and confirm answers through voice commands. For example, the patient can say "Select interference training" or "Confirm this object". During the training process, the system can provide encouragement to the patient through voice, such as "Well done!" or "Try again, you can do it!" These positive feedbacks help enhance the patient's confidence and motivate them to continue training.

[0063] In other embodiments, the touch screen 41 supports external input devices, such as switches, gesture recognition devices, etc., to help patients who cannot use the touch screen 41 to conduct training. This flexibility can ensure that different types of patients can participate in training smoothly.

[0064] Furthermore, the feedback and evaluation module 5 includes: a main control chip 51, a monitoring and evaluation system 52, and a storage module 53.

[0065] The main control chip 51 is connected to the touch screen 41, the camera 32, the optical basic module 31, the image sensor 33, and the image generator 22. Among them, the main control chip 51 is used to process image data and calculate eye parameters, and at the same time send control signals to the optical basic module and the image generation module 2 according to the instructions of the touch screen 41.

[0066] The monitoring and evaluation system 52 is loaded and connected to the main control chip 51. Among them, the monitoring and evaluation system 52 is used to record the reaction time of the patient in each training and evaluate each training of the patient.

[0067] The storage module 53 is connected to the main control chip 51. Among them, the storage module 53 is used to store the training data and evaluation data of each patient.

[0068] Refer to Figure 4 , Figure 4 is a schematic flowchart of an embodiment of the virtual simulation training system for post-stroke visual impairment rehabilitation of the present application. The method includes:

[0069] Step S1. Collect the eye movement parameters of the patient.

[0070] Specifically, refer to Figure 5 , step S1 includes:

[0071] Step S11. Emit near-infrared light through a light source to irradiate the patient's eyes and take images of the eyes.

[0072] Specifically, the light source emits near-infrared light to irradiate the eyes, and the camera 32 in the imaging system takes images of the eyes in real time. These images contain the pupil, corneal reflection, and some areas around the eyes, serving as the original data for subsequent processing.

[0073] Step S12. Process the images of the eyes with an image processing algorithm to obtain the characteristic data of the patient's pupil and corneal reflection points.

[0074] Specifically, through a series of image processing techniques, such as edge detection, threshold segmentation, morphological operations, etc., to identify the pupil and corneal reflection points. For example, using the edge detection algorithm can find the boundary of the pupil and determine the position of the pupil center; for the corneal reflection point, its position can be determined by finding the brightest point in the image or according to the specific shape and intensity distribution of the reflected light. In addition to the position information, other characteristic data of the pupil and corneal reflection points will also be extracted, such as the size change of the pupil (which may be related to factors such as light intensity, visual attention, etc.), the brightness and shape changes of the corneal reflection point, etc. These data can provide more information for subsequent eye movement analysis.

[0075] Step S13. Based on the characteristic data, perform P-CR vector calculation to obtain eye movement parameters.

[0076] Specifically, after determining the positions of the pupil center and the corneal reflection point, the vector between them (P-CR vector) is calculated. This vector has its specific coordinate representation in three-dimensional space. By comparing the P-CR vectors at different times, the change amount of the vector can be obtained; various eye movement parameters are calculated based on the change of the P-CR vector. For example, the component changes of the vector in the horizontal and vertical directions can be converted into the horizontal and vertical rotation angles of the eyeball. For the rotational movement of the eyeball, more complex geometric and mathematical models can be used to calculate in combination with the changes in the length and angle of the P-CR vector. These calculated eye movement parameters can be used to judge the fixation direction of the eyeball, the amplitude and speed of saccadic movements, etc.

[0077] Among them, the pupil-corneal reflection vector method (P-CR vector method) is an eye movement tracking technology based on optical principles. When light irradiates the eye, a reflection is formed on the corneal surface. The position of this corneal reflection point changes with the rotation of the eyeball. At the same time, the position of the pupil center also moves due to eye movement. By accurately measuring the relative position relationship between the pupil center and the corneal reflection point, that is, the P-CR vector, information about eye movement can be obtained.

[0078] In three-dimensional space, the movement of the eyeball includes horizontal direction (left and right rotation), vertical direction (up and down rotation), and rotational movement (rotation around the central axis of the eyeball). The P-CR vector method can comprehensively consider these movement dimensions. Because the position changes of the corneal reflection point and the pupil center in three-dimensional space can reflect different movement states of the eyeball. For example, when the eyeball rotates horizontally, the horizontal component of the P-CR vector changes; when the eyeball rotates vertically, its vertical component changes; and the rotational movement of the eyeball causes complex changes in the angle and length of the P-CR vector.

[0079] Step S2. Based on the main menu interface and eye movement parameters, select the training mode and training difficulty.

[0080] Specifically, the user interface includes: main menu interface, training mode selection, difficulty setting, help and support.

[0081] Main menu interface. After the patient starts the application, they first enter the main menu, and the interface shows four options: "Start Training", "View Progress", "Personalized Settings", and "Help". Each option uses large icons and clear text descriptions, making it convenient for the patient to understand at a glance;

[0082] Training mode selection. After clicking "Start Training", the patient can select different training modes. The modes include "Basic Training", "Gradually Decrease Contrast Training", "Interference Training", and "Practical Application Training". Under each mode, the system will briefly describe the training content and objectives to help the patient understand the purpose of each training.

[0083] Difficulty setting. After selecting a training mode, patients can further select the difficulty. Difficulty is divided into "easy", "medium" and "hard", and each difficulty corresponds to a different contrast range and object complexity. For example, the "easy" mode may show high-contrast images, while the "hard" mode shows low-contrast and complex background images. Patients can make selections using a slider or button.

[0084] Feedback and prompts: To improve patient participation and motivation, the user interaction module 4 should provide real-time feedback and prompts.

[0085] Patients can select "View Progress" in the main menu, and the system will display their training records, including training duration, success rate, progress, etc. Through charts and data visualization, patients can clearly see their rehabilitation progress. Personalized settings: In order to meet the needs of different patients, user interaction module 4 should provide personalized setting options. Visual settings: Patients can adjust the color contrast, font size, and background color of the interface according to their own visual preferences to ensure the best visual experience. Training plan customization: Patients can develop personalized training plans based on their own progress and needs, select training frequency and duration, and ensure the effectiveness and sustainability of the training.

[0086] User interaction module 4 should include a “Help” option to provide patients with detailed instructions and answers to frequently asked questions. This includes:

[0087] User Guide: Explain how to use the various functions of the system through pictures and texts to ensure that patients can carry out training smoothly;

[0088] Video tutorials: Provide short videos to demonstrate how to conduct different training modes and interactive methods, helping patients understand how to operate more intuitively;

[0089] Customer Support: Online or telephone support is provided so that patients can get timely help when they encounter any technical problems.

[0090] Step S3: Perform recovery training on the patient's eyes based on different training modes and training difficulties.

[0091] Specifically, step S3 includes:

[0092] Step S31. In response to the training mode being the hemianopsia visual field diagnosis mode, the brightness, frequency, and contrast parameters of the light source are set to perform visual field assessment and compare the visual field defects of the left and right eyes to provide a basis for personalized visual field restoration therapy.

[0093] Specifically, the above eye movement parameters are output to the image generation module 2 to synchronize the vector coordinate system of the image generation module 2, so as to ensure the relative stillness between the line of sight and the imaging coordinates. Enter the diagnosis stage. First, use green light points with the highest acceptable brightness to randomly appear at a fixed frequency. Without prompts, record the user feedback data through the interaction device to identify the visual field range. According to the visual field range data, gradually reduce the brightness of the visual targets point by point and level by level, and let the user operate the interactive device to determine the sensitivity threshold, so as to identify the visual field sensitivity. After system processing, repeat the test at the junction of the visual area and the blind area to reduce the deviation, and record the data. At the same time, record various data such as static visual field data, and form visual field range data after cleaning, transformation, standardization and grouped analysis. After diagnosis, conduct an end assessment, process the data, compare the visual field defects of the left and right eyes, provide a basis for personalized visual field restoration therapy, and record and upload the data before and after training for comparison.

[0094] Step S32. In response to the training mode being the hemianopia recovery therapy mode, use infrared or image training to eliminate the interference of eye movement, and at the same time perform flashing stimuli at different positions and brightness levels on the blind area.

[0095] Specifically, the above eye movement parameters are output to the image generation module 2 to synchronize the vector coordinate system of the image generation module 2, so as to ensure the relative stillness between the line of sight and the imaging coordinates. Use infrared or image training to eliminate the interference of eye movement, and at the same time select the visual field recovery treatment software to perform flashing stimuli at different positions and brightness levels on the blind area. In the initial stage, detect the line of sight fixation and tracking effect through simple flashing stimuli to determine the flashing stimulus parameters. In the training stage, according to the visual field disorder diagnosis results, perform grouped phased flashing stimuli on the visual defect boundary according to specific parameters and intersperse general stimuli to prevent fatigue. Perform flashing image stimuli on the blind spot with the same parameters. Subsequently, comprehensively perform image stimuli on the blind spot and the defect area of the visual field, reduce the parameters of the healthy side visual field to strengthen the stimulation of the blind area and the blind spot and control the wavelength, and at the same time record the relevant data and process and analyze it. Finally, comprehensively evaluate the rehabilitation effect from multiple dimensions through performance scoring, AI data analysis, subjective feedback collection and long-term tracking.

[0096] This mode uses multiple methods to evaluate the rehabilitation effect. The performance scoring is based on the user's performance in different training steps, and uses a quantitative evaluation method to judge the changes in the visual field range and sensitivity; the AI data analysis automatically records the training data by the AI system, and generates a visual progress report through algorithm analysis, providing a basis for adjusting the treatment plan; the subjective feedback is to ask the user about their understanding and acceptance of the training content after each training, helping to optimize the treatment plan. At the same time, through regular evaluation of the changes in the blind side visual field in the user's daily life for long-term tracking, judge the long-term effectiveness of the treatment method, and provide practical basis for improving the plan.

[0097] Step S33. In response to the training mode being the shape recognition image mode, generate different images for the patient to recognize and evaluate the patient's image recognition ability.

[0098] Specifically, step S33 includes:

[0099] Step S331. Output the above eye movement parameters to the image generation module 2 to record data such as eye movement speed, acceleration, and angle and output them to the personal database.

[0100] Select software to generate shape images and determine the training difficulty. In the initial evaluation stage, understand the patient's basic shape recognition ability through simple tests and formulate a personalized plan. In the training stage, conduct basic shape recognition in sequence, show simple shapes for the patient to click and give feedback; combined shape recognition, present complex combined images and limit the reaction time; dynamic shape recognition, show dynamically changing shapes to improve the patient's sensitivity; real-life application, use physical cards for the patient to find matching objects in life and describe their features. In the evaluation stage, test again and compare with the initial results, record the progress, and comprehensively evaluate the training effect through performance scoring, subjective feedback, and long-term tracking.

[0101] Step S34. In response to the training mode being the color discrimination image mode, conduct basic color, color combination, dynamic color change, and real-life application training in sequence to improve the patient's color recognition ability.

[0102] Specifically, step S34 includes:

[0103] Step S341. Select graphic design software such as Adobe Illustrator to generate a variety of color combination images, ensuring that basic colors, gradient colors, and complex patterns are covered. Determine the training difficulty level, including color saturation, contrast, and combination method.

[0104] Step S342. Output the above eye movement parameters to the image generation module 2 to record data such as eye movement speed, acceleration, and angle and output them to the personal database. In the initial evaluation, formulate a personalized plan through simple tests. In the training stage, conduct basic color, color combination, dynamic color change, and real-life application training in sequence, and the AI system provides feedback, adjusts the difficulty, and records the results. At the same time, the AI system automatically adjusts the subsequent training according to the training results. In terms of evaluation, comprehensively evaluate the training effect from four dimensions: performance scoring, AI data analysis, subjective feedback collection, and long-term tracking.

[0105] Step S35. In response to the training mode being the motion tracking image mode, generate images containing different objects moving at various speeds and directions, and set the training difficulty, which is determined according to the object movement speed, trajectory, and background interference, to improve the patient's ability to track moving objects.

[0106] Specifically, use image generation software to generate images containing different objects moving at various speeds and directions, and set the training difficulty, which is determined based on the object movement speed, trajectory, and background interference. Conduct an initial assessment, and through a simple motion tracking test, develop a personalized training plan for the patient. The training stage is carried out step by step: basic motion tracking training, presenting basic shapes moving slowly for the patient to track, and the system gives immediate feedback; complex motion tracking training, showing multiple moving objects and requiring the patient to track them separately, and the system adjusts the number, speed, and direction of the objects according to the patient's performance; environmental interference training, showing moving objects in a complex background, and the system records the patient's performance to adjust the training; dynamic response training, showing fast-moving objects and requiring the patient to react quickly, and the system adjusts the training difficulty based on the reaction speed. In addition, the system will analyze the patient's situation based on the training results and automatically adjust the subsequent training content and difficulty. During the assessment, comprehensively evaluate the training effect from multiple aspects such as performance scoring to quantify tracking-related indicators, data analysis to generate a progress report, collecting the patient's subjective feedback, and regular long-term tracking.

[0107] Step S36. In response to the training mode being the contrast adjustment image mode, generate images of daily objects with different contrasts to improve the patient's contrast recognition ability.

[0108] Specifically, use image generation software to create images containing daily objects with different contrasts and determine the training difficulty. Evaluate the patient's basic ability through a simple test, and then customize a personalized training plan. During the training, the system shows images with different contrasts to the patient, and some of the images are set with complex backgrounds to form interference. At the same time, guide the patient to find low-contrast objects in the real-life scenario. The system will record the training data, adjust the training according to the patient's performance, and give feedback. After the training, comprehensively evaluate the training effect from dimensions such as performance scoring, data analysis, subjective feedback, and long-term tracking, so as to enhance the patient's object recognition ability in a low-contrast environment, improve visual attention, and promote neural remodeling.

[0109] Step S4. Based on the rehabilitation training, detect the patient's reaction time and recognition accuracy, and analyze the patient's behavior pattern.

[0110] Specifically, the monitoring and evaluation system 52 will record the patient's reaction time in each training, including the time from the image display to the patient's reaction. This data will help analyze the patient's adaptability to different contrasts and image complexities. Using AI algorithms, the system can identify the changing trend of the reaction time, judge whether the patient is gradually adapting to the training content, or whether there is fatigue or frustration.

[0111] The monitoring and evaluation system 52 will automatically calculate the recognition accuracy rate of the patient in each training session, including the ratio of the number of correct recognitions to the total number of attempts. The change in the accuracy rate will become an important indicator for evaluating the patient's learning progress. By comparing the accuracy rates in different training stages, the rehabilitation therapist can better understand whether the patient's visual recognition ability is improving.

[0112] By recording the patient's behaviors during training (such as the number of incorrect selections, the number of abandonments, etc.), the monitoring and evaluation system 52 can analyze the patient's behavior patterns and identify the possible difficulties they may have with specific training content. AI technology can compare these behavior patterns with the performances of other patients to help the rehabilitation therapist develop a more targeted training plan for each patient.

[0113] Step S5. Automatically adjust the difficulty of the next training session based on the patient's behavior patterns and recognition accuracy rate, and provide voice suggestions.

[0114] Specifically, based on the patient's performance, the system will automatically generate personalized training suggestions, including the recommended training frequency, suitable training modes, and difficulty levels. The system will summarize the patient's performance through voice, including the accuracy rate, changes in reaction time, and overall progress. This summary can enhance the learning effect and motivate the patient to continue working hard in subsequent training.

[0115] Such personalized suggestions can ensure that the patient trains in the best state. Through regular evaluations and analyses, the rehabilitation therapist can observe the patient's long-term changes and timely adjust the training strategy to ensure the achievement of the rehabilitation goals.

[0116] The present invention has the following beneficial effects:

[0117] 1. Relying on the training and rehabilitation program, visual impairment repair training can be carried out alone or visual impairment repair training tasks can be carried out in specific task scenarios.

[0118] 2. In the hemianopia visual field diagnosis scheme and hemianopia recovery therapy scheme, by setting up an eye movement detection module to respond to the line-of-sight jitter during training, keeping the relative stillness between the line of sight and the imaging, and reducing the contamination of the jittery visual field to the data and the impact on training.

[0119] 3. Utilizing the VR technology in the VR device to perform local processing on the imaging, enhancing the targeted stimulation of the affected area, and providing better rehabilitation effects for users.

[0120] 4. Utilizing the VR device to provide independent diagnosis and training programs, realizing multi-functional training, and providing more efficient training effects for users.

[0121] 5. Utilizing the VR device to construct a real environment with high fidelity, providing a more comprehensive visual scene simulation for users.

[0122] The above are only embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall equally be included in the patent protection scope of the present application.

Claims

1. A virtual simulation training system for rehabilitation of post-stroke visual impairment, characterized in that: include: VR glasses, used to carry each module, wherein the VR glasses are worn by patients to repair visual impairment of patients; An image generation module, located in the VR glasses, wherein the image generation module is used to generate different images to provide a variety of different visual restoration solutions; An eye movement detection module is located in the VR glasses, wherein the eye movement detection module is used to synchronize sight and imaging and obtain eye parameters; A user interaction module connected to the image generation module, wherein the user interaction module is used for the patient to interact with the system and select options to be used; A feedback evaluation module connected to the image generation module, wherein the feedback evaluation module is used to monitor and analyze the patient's reaction time, accuracy, and behavior pattern in real time; Power module, used to supply power to other modules.

2. The system according to claim 1, characterized in that The image generation module comprises: A display, located in the eye socket of the VR glasses, wherein the display is used to display images; An image generator is connected to the display, wherein the image generator is used to generate different display images on the display.

3. The system according to claim 1, characterized in that The eye movement detection module comprises: an optical basic module connected to the feedback evaluation module, wherein the optical system is used to illuminate the patient's eyes from different angles through a plurality of the light sources; A camera connected to the feedback evaluation module, wherein the camera is used to capture images of the patient's eyes in real time; An image sensor is connected to the feedback evaluation module, wherein the image sensor is used to identify the pupil and corneal reflection point of the patient.

4. The system according to claim 3, characterized in that The optical basic module comprises: A light source module, used for providing light source; A filter, the filter is located in front of the light source module, wherein the filter is used to control the color and intensity of the irradiated light of the light source; Lens, used to reduce aberration and chromatic aberration, ensuring image clarity and accuracy.

5. The system according to claim 4, characterized in that The user interaction module comprises: A touch screen connected to the VR glasses via a data cable, wherein the touch screen is used to display a control interface to the patient and perform touch operations; A voice component is connected to the touch screen, wherein the voice component is used to provide voice prompts or receive voice instructions from patients.

6. The system according to claim 5, characterized in that The feedback evaluation module comprises: A main control chip connected to the touch screen, the camera, the optical basic module, the image sensor and the image generation module, wherein the main control chip is used to process image data and calculate eyeball parameters, and send control signals to the optical basic module and the image generation module according to instructions from the touch screen; A monitoring and evaluation system, which is loaded on the main control chip, wherein the monitoring and evaluation system is used to record the patient's reaction time in each training and evaluate the results of each training of the patient; A storage module is connected to the main control chip, wherein the storage module is used to store the training data and evaluation data of each patient.

7. A virtual simulation training method for rehabilitation of post-stroke visual impairment, characterized in that: The following steps are involved: Collect the patient's eye movement parameters; Based on the main menu interface and the eyeball data, select a training mode and a training difficulty; Based on different training modes and training difficulties, the patient's eyes are given recovery training; Based on the recovery training, detecting the patient's reaction time and recognition accuracy, and analyzing the patient's behavior pattern; Based on the patient's behavior pattern and the recognition accuracy, the difficulty of the next training session is automatically adjusted and voice suggestions are provided.

8. The method according to claim 7, characterized in that The method for collecting eyeball data of a patient comprises: irradiating the patient's eyes with near-infrared light from a light source and capturing an image of the eyes; Processing the eye image with an image processing algorithm to obtain feature data of the patient's pupil and corneal reflection point; Based on the feature data, P-CR vector calculation is performed to obtain the eye movement parameters.

9. The method according to claim 7, characterized in that: The method for performing recovery training on the patient's eyes based on different training modes and training difficulties includes: In response to the training mode being a hemianopsia visual field diagnosis mode, setting the brightness, frequency, and contrast parameters of the light source, performing visual field assessment, and comparing the visual field defects of the left and right eyes to provide a basis for personalized visual field restoration therapy; In response to the training mode being a hemianopsia recovery therapy mode, infrared or image training is used to eliminate visual micro-motion interference, and at the same time, flashing stimulation of different positions and brightness is performed on the blind area; In response to the training mode being a shape recognition image mode, generating different images for the patient to recognize, and evaluating the patient's image recognition ability; In response to the training mode being a color discrimination image mode, basic color, color combination, dynamic color change and real-life application training are sequentially performed to improve the patient's color recognition ability; In response to the training mode being a motion tracking image mode, images containing different objects moving at various speeds and directions are generated, and the training difficulty is set, wherein the difficulty is determined according to the object moving speed, trajectory, and background interference, so as to improve the patient's moving object tracking ability; In response to the training mode being a contrast-adjusted image mode, images of daily objects with different contrasts are generated to improve the patient's contrast recognition ability.

Images