Method and device for controlling MR glasses for dizziness rehabilitation and MR glasses for dizziness rehabilitation

Abstract

The invention is suitable for the technical field of MR glasses, and particularly relates to an MR glasses control method and device for dizziness rehabilitation and MR glasses for dizziness rehabilitation, and the method comprises the following steps: obtaining current rehabilitation training data of a patient; based on the current rehabilitation training data of the patient, determining whether the patient reaches a rehabilitation guidance condition; wherein the rehabilitation guidance condition is used for representing that training guidance is needed when the patient uses the MR glasses for dizziness rehabilitation to carry out dizziness rehabilitation training; if the patient reaches the rehabilitation guidance condition, generating a rehabilitation guidance scheme according to the current rehabilitation training data of the patient; and outputting the rehabilitation guidance scheme to guide the patient to perform dizziness rehabilitation training. According to the method, a personalized and real-time rehabilitation guidance scheme can be provided, and the training efficiency of the patient can be improved.

Description

Technical Field

[0001] The present application belongs to the technical field of MR glasses, and in particular, relates to a control method and device for MR glasses for vertigo rehabilitation, and MR glasses for vertigo rehabilitation. Background Art

[0002] MR glasses for vertigo rehabilitation are a type of medical device based on mixed reality (MR) technology, specifically used to assist in the treatment and rehabilitation of patients with vertigo-related diseases. MR glasses for vertigo rehabilitation use virtual scenes and interactive technology to help patients perform functional training of the vestibular system, improve balance, and thus relieve vertigo symptoms.

[0003] In the prior art, when using MR glasses for vertigo rehabilitation for vertigo rehabilitation training, patients may not be able to get timely training feedback or personalized guidance, especially when the complexity of the rehabilitation task is high, it is difficult for patients to judge whether the action is correct or the goal is achieved. If there is a lack of proper guidance during the rehabilitation training process, patients may repeatedly perform ineffective training, resulting in low rehabilitation efficiency. At the same time, training guidance without proper feedback and dynamic adjustment can easily confuse patients or make them lose confidence.

[0004] In summary, when using MR glasses for vertigo rehabilitation training, there is a problem of low patient training efficiency due to lack of targeted guidance. Summary of the invention

[0005] The embodiments of the present application provide a control method and device of MR glasses for vertigo rehabilitation, and MR glasses for vertigo rehabilitation, which can solve the problem in the related art that when using MR glasses for vertigo rehabilitation for vertigo rehabilitation training, there is a lack of targeted guidance, resulting in low patient training efficiency.

[0006] In a first aspect, an embodiment of the present application provides a method for controlling MR glasses for vertigo rehabilitation, comprising: Acquire the patient's current rehabilitation training data; wherein the patient's current rehabilitation training data includes current motion data and current training task information, the current training task information includes task type, standard motion data, motion threshold, and preset duration, and the current motion data is acquired through a data acquisition module of MR glasses for vertigo rehabilitation; Based on the current rehabilitation training data of the patient, determining whether the patient meets the rehabilitation guidance condition; wherein the rehabilitation guidance condition is used to indicate that the patient needs training guidance when performing vertigo rehabilitation training using the MR glasses for vertigo rehabilitation; If the patient meets the rehabilitation guidance conditions, generating a rehabilitation guidance plan according to the patient's current rehabilitation training data; The rehabilitation guidance program is output to guide the patient to perform vertigo rehabilitation training.

[0007] The above technical solutions in the embodiments of the present application have at least the following technical effects: The MR glasses control method for dizziness rehabilitation provided in the present application first obtains the patient's current rehabilitation training data, which is conducive to knowing the patient's current motion data (obtained through the data acquisition module of the MR glasses for dizziness rehabilitation) and the current training task information, and then determines whether the patient meets the rehabilitation guidance conditions based on the patient's current rehabilitation training data, which is conducive to knowing whether the patient needs training guidance when using the MR glasses for dizziness rehabilitation for dizziness rehabilitation training. If the patient meets the rehabilitation guidance conditions, a rehabilitation guidance plan is generated according to the patient's current rehabilitation training data, and finally the rehabilitation guidance plan is output to guide the patient to perform dizziness rehabilitation training. By analyzing the patient's current motion data and training task information, the method can determine in real time whether the patient has met the conditions for rehabilitation guidance, and can tailor the most suitable rehabilitation plan for each patient, thereby improving the pertinence and personalization of rehabilitation training. If the patient fails to meet the training standard or motion threshold, the method can provide the patient with a guidance plan in a timely manner to help the patient adjust his movements and avoid inefficient or erroneous training, thereby improving the training efficiency. Through precise data monitoring and analysis, the patient's rehabilitation training plan is no longer a one-size-fits-all approach. Instead, a personalized guidance plan is generated based on the patient's real-time performance, which can help patients avoid overtraining or undertraining during the rehabilitation process, thereby promoting the rehabilitation process more scientifically.

[0008] In a second aspect, an embodiment of the present application provides a MR glasses control device for vertigo rehabilitation, comprising: An acquisition unit, used for acquiring the patient's current rehabilitation training data; wherein the patient's current rehabilitation training data includes current motion data and current training task information, the current training task information includes task type, standard motion data, motion threshold, and preset duration, and the current motion data is acquired through a data acquisition module of MR glasses for vertigo rehabilitation; A judgment unit, used for determining whether the patient meets the rehabilitation guidance condition based on the current rehabilitation training data of the patient; wherein the rehabilitation guidance condition is used to indicate that the patient needs training guidance when performing vertigo rehabilitation training using the MR glasses for vertigo rehabilitation; a program generating unit, configured to generate a rehabilitation guidance program according to the current rehabilitation training data of the patient if the patient meets the rehabilitation guidance conditions; The output unit is used to output the rehabilitation guidance program to guide the patient to perform vertigo rehabilitation training.

[0009] In a third aspect, an embodiment of the present application provides MR glasses for vertigo rehabilitation, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any one of the embodiments of the first aspect when executing the computer program.

[0010] It can be understood that the beneficial effects of the second to third aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 It is a flowchart of a method for controlling MR glasses for vertigo rehabilitation provided in one embodiment of the present application; Figure 2 It is a schematic diagram of the implementation flow of the real-time control method in the MR glasses control method for vertigo rehabilitation provided in the embodiment of the present application; Figure 3 It is a schematic diagram of the implementation flow of the control method based on the patient's voice in the MR glasses control method for vertigo rehabilitation provided in the embodiment of the present application; Figure 4 is a schematic diagram of the structure of a MR glasses control device for vertigo rehabilitation provided in an embodiment of the present application; Figure 5 It is a schematic diagram of the structure of MR glasses for vertigo rehabilitation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0013] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0014] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0015] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0016] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0017] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0019] In the related art, when using MR glasses for vertigo rehabilitation for vertigo rehabilitation training, patients may not be able to get timely training feedback or personalized guidance, especially when the complexity of the rehabilitation task is high, it is difficult for patients to judge whether the action is correct or the goal is achieved. If there is a lack of proper guidance during the rehabilitation training process, patients may repeatedly perform ineffective training, resulting in low rehabilitation efficiency. At the same time, training guidance without proper feedback and dynamic adjustment can easily confuse patients or make them lose confidence.

[0020] To solve the above problems, the embodiment of the present application provides a control method and device of MR glasses for dizziness rehabilitation and MR glasses for dizziness rehabilitation. In the method, the current rehabilitation training data of the patient is first obtained, which is conducive to knowing the current action data of the patient (obtained by the data acquisition module of the MR glasses for dizziness rehabilitation) and the current training task information. Then, based on the current rehabilitation training data of the patient, it is determined whether the patient meets the rehabilitation guidance conditions, which is conducive to knowing whether the patient needs training guidance when using the MR glasses for dizziness rehabilitation for dizziness rehabilitation training. If the patient meets the rehabilitation guidance conditions, a rehabilitation guidance plan is generated according to the current rehabilitation training data of the patient, and finally the rehabilitation guidance plan is output to guide the patient to perform dizziness rehabilitation training. By analyzing the current action data and training task information of the patient, the method can judge in real time whether the patient has met the conditions for rehabilitation guidance, and can tailor the most suitable rehabilitation plan for each patient, thereby improving the pertinence and personalization of rehabilitation training. If the patient fails to meet the training standard or action threshold, the method can provide the patient with a guidance plan in time to help the patient adjust the action, avoid inefficient or erroneous training, and thus improve the training efficiency. Through precise data monitoring and analysis, the patient's rehabilitation training plan is no longer a one-size-fits-all approach. Instead, a personalized guidance plan is generated based on the patient's real-time performance, which can help patients avoid overtraining or undertraining during the rehabilitation process, thereby promoting the rehabilitation process more scientifically.

[0021] The MR glasses control method for vertigo rehabilitation provided in the embodiment of the present application can be applied to MR glasses for vertigo rehabilitation. In this case, the MR glasses for vertigo rehabilitation are the executor of the MR glasses control method for vertigo rehabilitation provided in the embodiment of the present application. The embodiment of the present application does not impose any restrictions on the specific type of MR glasses for vertigo rehabilitation.

[0022] Exemplarily, MR glasses for vertigo rehabilitation may include a display module, a voice playback module, a data acquisition module and a control module. The display module is a module capable of displaying rehabilitation guidance content, and may include an optical waveguide display, an OLED / LED screen display, a head-mounted display (HMD), etc.; the voice playback module is a module capable of playing rehabilitation guidance content, and may include an integrated speech synthesis module (Text-to-Speech, TTS), natural language processing (NLP) speech recognition, two-way voice communication, etc.; the data acquisition module is a module capable of collecting patient motion data, and may include an inertial measurement unit (IMU), an eye tracker, a temperature sensor, etc.; the control module is a module capable of performing data processing and controlling the display module, the voice playback module and the data acquisition module, and may include a central processing unit (CPU / SoC), a motion capture and analysis module, a wireless communication module (such as Bluetooth, Wi-Fi), a battery management and power control module, an environmental perception module, a biofeedback module, etc.

[0023] In order to better understand the MR glasses control method for vertigo rehabilitation provided in the embodiment of the present application, the specific implementation process of the MR glasses control method for vertigo rehabilitation provided in the embodiment of the present application is exemplarily introduced below.

[0024] Figure 1 A schematic flow chart of a method for controlling MR glasses for rehabilitation of vertigo provided in an embodiment of the present application is shown. The method for controlling MR glasses for rehabilitation of vertigo includes: S100, obtaining the patient's current rehabilitation training data. The patient's current rehabilitation training data includes current motion data and current training task information, the current training task information includes task type, standard motion data, motion threshold, and preset duration, and the current motion data is obtained through a data acquisition module of MR glasses for vertigo rehabilitation.

[0025] It can be understood that the patient's current motion data may include head motion data, eye motion data, acceleration and gyroscope data, and physiological signals. Head motion data may include the rotation angle, tilt angle, displacement, etc. of the head, which can be used as a basis for evaluating the patient's balance control and dizziness reaction; eye motion data, through eye motion sensors, records the patient's eye movement trajectory, eye rotation angle, etc., and eye movement control is closely related to balance; acceleration and gyroscope data can detect the patient's overall motion state and determine whether the patient has an unstable posture or abnormal movements; physiological signals, such as the patient's heart rate, blood oxygen saturation, etc., are used to evaluate the patient's physiological reactions and help determine whether the patient is overly fatigued or has other health problems.

[0026] The data acquisition module may include: an inertial measurement unit (IMU) with integrated accelerometers and gyroscopes to monitor the movement of the patient's head and eyes; an eye tracker to track the dynamic changes of the eyes and assess whether the patient's eye movements are normal; other sensors, such as temperature sensors, heart rate sensors, etc., to further supplement the data during the training process.

[0027] For example, the patient's motion and physiological data can be collected in real time through built-in sensors (IMU, eye tracker, heart rate sensor, etc.).

[0028] Exemplarily, the current training task information is pre-stored in the MR glasses for vertigo rehabilitation, and the current training task information includes task type, standard action data, action threshold, and preset duration. Task type is used to define the nature of the current task, such as balance training, rotation training, eye movement training, etc. Different types of tasks will have different requirements for the patient's actions; standard action data, in order to ensure the standardization and effectiveness of rehabilitation training, standard action data can be pre-set as a reference for patients to complete tasks. Standard action data may include the frequency, angle, duration, speed, position, etc. of the action; action threshold, in order to monitor the quality of the patient's action, a threshold can be set. When the patient's action deviates from the standard action data to a certain extent, a warning can be issued or the training content can be adjusted. For example, when the angle deviation of the action exceeds a certain range, the patient can be prompted to make adjustments; preset duration, each training task can set a predetermined duration to limit the training time to avoid excessive fatigue of the patient. The training duration can be dynamically adjusted according to the patient's physical condition or rehabilitation progress.

[0029] Through this step, the patient's current motion data can be collected in real time, which helps to determine the patient's rehabilitation status and provide a basis for whether the patient needs subsequent rehabilitation training guidance.

[0030] S200, based on the patient's current rehabilitation training data, determining whether the patient meets the rehabilitation guidance condition, wherein the rehabilitation guidance condition is used to indicate that the patient needs training guidance when performing vertigo rehabilitation training using the vertigo rehabilitation MR glasses.

[0031] It can be understood that the rehabilitation guidance conditions can be a set of preset standards, which are intended to determine whether one or more rehabilitation guidance conditions are met through the analysis of the patient's current training data, which indicates that the patient may encounter difficulties in training and need additional guidance. Rehabilitation guidance conditions may include: movement accuracy, whether the patient's current movement matches the standard movement data (such as angle, frequency, speed, etc.). If the patient's movement deviates too much from the standard movement, guidance may be needed to adjust the movement; movement intensity, whether the patient's movement reaches the preset movement threshold, such as the rotation or balance task in the training requires a certain movement intensity. If the movement intensity is not enough, it may mean that the training effect is not ideal; movement duration, whether the patient can maintain enough time to complete the training task. If the patient fails to maintain the prescribed movement duration, or the training duration is not enough to meet the rehabilitation requirements, guidance can be given; movement stability, whether the patient's balance is stable during training. If the patient frequently has unstable balance performance during rehabilitation training, auxiliary guidance may be required; physiological state monitoring, such as whether physiological parameters such as heart rate and blood oxygen are within the normal range. If the patient's physiological state deviates from the normal range (such as excessive fatigue, too fast heart rate, etc.), it may be necessary to reduce the training intensity or provide restorative guidance.

[0032] For example, the patient's current motion data can be compared with the standard motion data to check whether the patient is within the specified range of motion angles, whether the motion frequency and duration required for the task are completed, whether the motion is completed within the predetermined time, and whether unstable balance or abnormal posture occurs. For example, if the training task requires the patient to rotate his head 10 times, if the patient's frequency is less than 10 times, the patient will be prompted to increase the motion frequency.

[0033] The patient's movements can be judged to be qualified according to the preset movement threshold. For example, a movement amplitude threshold is set. When the patient's movement amplitude is less than the movement amplitude threshold, it can be judged that the patient has failed to complete the task requirements, and the patient is prompted to strengthen the movement.

[0034] The patient's physiological state can be assessed through physiological monitoring (such as heart rate, blood oxygen, etc.). If the patient's physiological indicators exceed the safe range (such as heart rate is too fast or too slow, blood oxygen saturation is too low), it can be judged that the patient needs to interrupt training, recover or adjust the training intensity. For example, if the heart rate exceeds 120 beats / minute or the blood oxygen saturation is less than 90%, the patient will be prompted to slow down or suspend training.

[0035] If the patient frequently loses balance, becomes dizzy or has other unstable behaviors during training, the sensor can be used to detect the balance state and determine whether the rehabilitation guidance conditions are met. For example, if the patient is frequently out of balance during training, the sensor may prompt the patient to adjust the training posture or pause the training to restore stability.

[0036] Through this step, the current training data is compared and analyzed with the preset standard action data, action threshold, etc., to determine whether the patient meets the rehabilitation guidance conditions. When the patient's performance does not meet the training standards, rehabilitation guidance can be automatically triggered to provide personalized training adjustments and suggestions, which is conducive to the scientificity and effectiveness of rehabilitation training and avoids errors or inappropriateness in training that lead to low rehabilitation efficiency.

[0037] In one possible implementation, see Figure 2 S200, based on the patient's current rehabilitation training data, determining whether the patient meets the rehabilitation guidance conditions, including: S210, extracting action data corresponding to the task type from the current action data according to the task type in the current training task information.

[0038] It can be understood that the task types may include balance training, rotation training, eye movement training, gaze tracking training, and vestibular training. The goal of balance training is to improve the patient's balance ability and reduce symptoms such as dizziness. The task of balance training is to train the patient to perform stable control of the head, body, or eyes. The motion data corresponding to balance training may include stability data of the head, eyes, or body, such as angle changes, position movement, etc.

[0039] The goal of rotation training is to improve the patient's head or body rotation control ability, and the task of rotation training is to train the patient to perform head, body or eye rotation movements. The action data corresponding to the rotation training may include rotation angle, rotation speed, rotation direction and duration.

[0040] The goal of eye movement training is to improve eye coordination and control the stability of eye movement. The task of eye movement training is to train patients to perform eye movements. The action data corresponding to eye movement training may include the eye rotation angle, rotation speed, and duration of eye movement.

[0041] The goal of eye tracking training is to improve the patient's eye tracking ability. The task of eye tracking training is to train the patient to track a moving target with his eyes. The action data corresponding to the eye tracking training may include the eye's tracking path, tracking speed, and tracking accuracy.

[0042] The goal of vestibular training is to improve the patient's vestibular adaptability and reduce the occurrence of vertigo. The task of vestibular training is to train the patient's vestibular system's response and control. The corresponding motion data of vestibular training can include the acceleration, rotation angle, and stability of the head posture.

[0043] For example, each task type has corresponding motion data, and the relevant motion data can be captured and extracted in real time through the data acquisition module. Balance training: The angle data of the head, eyes or body can be extracted from the current motion data, and the angle data of the head, eyes or body can be obtained through devices such as accelerometers, gyroscopes, and eye trackers. For example, an IMU sensor is used to monitor the angle change of the head, or the rotation data of the eyeball is extracted through an eye tracking device.

[0044] Rotation training: The rotation angle, rotation speed, rotation direction and other data can be extracted from the current motion data, and the rotation angle, rotation speed, rotation direction and other data can be used to monitor the rotation process in real time through a gyroscope or motion capture device. For example, the rotation angle and rotation speed of the head can be monitored, and the amplitude and duration of each rotation can be recorded.

[0045] Eye movement training: The eye's rotation angle (such as horizontal and vertical angles), rotation speed and other data can be extracted from the current motion data, and the eye's rotation angle, rotation speed and other data can be obtained through an eye tracker or a head-mounted eye tracking device.

[0046] Eye tracking training: The path data, tracking speed and tracking accuracy of the eye-tracking object can be extracted from the current motion data, and the path data, tracking speed and tracking accuracy of the eye-tracking object can be obtained through an eye tracking system (eye tracker or head-mounted eye tracking device).

[0047] Vestibular training: Data such as acceleration, rotation angle and position of the head can be extracted from the current motion data, and data such as acceleration, rotation angle and position of the head can be obtained through the head accelerometer and gyroscope data.

[0048] The data extracted through this step are used for subsequent analysis, comparison and rehabilitation guidance.

[0049] S220, calculating the deviation between the action data corresponding to the task type and the standard action data in the current training task information to obtain a deviation value.

[0050] It can be understood that the deviation value reflects the difference between the patient's current movement and the standard requirements. The larger the deviation value, the less the patient's movement meets the standard, and the training task may need to be adjusted or further rehabilitation guidance may be provided. The size of the deviation value is the basis for judging whether the conditions for rehabilitation guidance are met.

[0051] The action data corresponding to each task type mentioned in step S210 has corresponding standard action data. For example, the action data corresponding to rotation training may include rotation angle, rotation speed, rotation direction and duration, and the standard action data in the rotation training task information may include standard rotation angle, standard rotation speed, standard rotation direction and standard duration.

[0052] For example, for rotation training, balance training, or eye movement training, the angle deviation between the angle data of the patient's current action and the angle data of the standard action can be calculated, that is, , Indicates the angle deviation value, Indicates the angle data of the current action, The angle data representing the standard action. For example, if the patient's head rotation angle is 25° and the standard action requirement is 30°, then the deviation value is .

[0053] For training tasks involving time (such as the duration of maintaining a certain action), the duration deviation can be calculated, that is, , Indicates the duration deviation, Indicates the duration of the current action. Indicates the standard time. For example, if the standard time is 60 seconds and the patient completed 50 seconds, then the deviation value is .

[0054] For training tasks involving speed (such as rotation speed in rotation training and tracking speed in eye tracking training), the speed deviation between the patient's actual movement speed and the standard movement speed can be calculated, that is, , Indicates the speed deviation, Indicates the current action speed. Indicates the standard motion speed. For example, if the standard rotation speed is 60° / second and the patient's rotation speed is 55° / second, then the deviation value is .

[0055] In balance training, the position deviation between the patient's actual movement and the standard movement can be calculated. The position deviation refers to the difference between the relative position of the body or head and the standard position, that is, , Indicates position deviation, Indicates the current action position. Indicates the position of the standard action. For example, the standard action position is 0° (representing the center of balance), but the patient deviates by 5°, then the deviation value is .

[0056] The deviation value obtained by comparing the patient's actual movement data with the standard movement data reflects the patient's performance in rehabilitation training. The deviation value can help determine whether the patient needs rehabilitation guidance and provide a basis for subsequent rehabilitation plans.

[0057] S230, comparing the deviation value with the action threshold in the current training task information, determining whether the deviation value exceeds the action threshold, and obtaining a first determination result.

[0058] It can be understood that the action threshold refers to the maximum allowable deviation between the patient's action data and the standard action data, which is used to determine whether the patient's action meets the standard. The action threshold can be set according to the requirements of the training task to ensure the training effect and safety.

[0059] Exemplarily, the calculated deviation value is compared with the action threshold in the training task information. If the deviation value is greater than the action threshold, the first judgment result is that the deviation exceeds the standard, indicating that the patient's action does not meet the standard requirements, and rehabilitation guidance can be provided. If the deviation value is within the action threshold, the first judgment result is that the deviation is normal, indicating that the patient's action meets the requirements and no rehabilitation guidance is needed temporarily.

[0060] S240: When the first judgment result indicates that the deviation value exceeds the action threshold, it is determined that the patient meets the rehabilitation guidance conditions.

[0061] Exemplarily, when the deviation value exceeds the set action threshold, it can be judged that the patient has met the rehabilitation guidance conditions and further rehabilitation guidance programs can be initiated, such as action correction, training adjustment, etc. For example, the standard action data requires the patient to maintain a rotation angle of 30° for 10 seconds, but the patient actually rotates 20° and maintains it for 12 seconds, and the action threshold is an angle threshold of 3° and a duration threshold of 3s. Then the angle deviation is 10° and the duration deviation is 2s. The angle deviation exceeds the angle threshold, and the duration deviation does not exceed the duration threshold. The patient's action does not meet the standard requirements and rehabilitation guidance can be provided.

[0062] By calculating the deviation from the standard action data and comparing the deviation value with the set action threshold, it is possible to effectively determine whether the patient meets the rehabilitation guidance conditions and provide a basis for whether the patient needs to be guided in rehabilitation training in the future.

[0063] S300: If the patient meets the rehabilitation guidance conditions, a rehabilitation guidance plan is generated according to the patient's current rehabilitation training data.

[0064] Exemplarily, the training task can be adjusted according to the patient's performance. The movement difficulty of the training task can be adjusted. If the patient fails to meet the standard movement requirements (such as movement angle, frequency, etc.), the difficulty of the task can be reduced or the task can be simplified. For example, if the patient's rotation movement angle is insufficient, it can be recommended to lower the standard angle. The training duration can be adjusted. If the patient cannot maintain the training duration (such as failing to reach the set 1 minute training time), the training duration can be shortened and gradually increased to avoid excessive fatigue. The task type can be changed. If the patient encounters difficulties in a specific training task, the task type can be temporarily changed. For example, if the patient has difficulty in balance training, it can be switched to simpler head movement training.

[0065] The patient's current movement can be corrected. Instant movement correction guidance can be provided through a display module (such as the display screen of MR glasses) or a voice playback module. For example, if the patient's head rotation angle deviates greatly, a voice prompt can be given to ask the patient to turn the head to 30 degrees, or a movement demonstration can be presented on the display screen. If the patient's movement intensity does not meet the requirements, suggestions for strengthening the movement can be provided, such as increasing the rotation speed or increasing the range of motion.

[0066] If the patient's balance is unstable or there are obvious symptoms of dizziness, the training content can be adjusted and more balance exercises can be added. For example, dizziness relief training can provide slow and gradual movements, avoid rapid head movements, and reduce the patient's dizziness; strengthen support training, such as suggesting that the patient support on a wall or chair for more stable balance training.

[0067] If the patient's physiological parameters are monitored to be abnormal (such as a fast heart rate, low blood oxygen, etc.), the rehabilitation guidance plan may include adjusting the training intensity. If the patient's physiological indicators are high, the training intensity can be reduced or the training can be suspended until the physiological parameters return to the normal range; rest or recovery. If the patient's physiological state is unstable, the patient can be prompted to rest, or perform restorative training, such as deep breathing training, static balance training, etc.

[0068] Through this step, the training tasks can be adjusted in real time according to the patient's current rehabilitation training data, corrective feedback can be provided, and appropriate guidance can be given according to the patient's physiological state and training progress. This helps to optimize the rehabilitation process, improve training efficiency, avoid errors or excessive fatigue in training, and ultimately achieve better rehabilitation results.

[0069] In one possible implementation, see Figure 2 In step S300, a rehabilitation guidance program is generated according to the patient's current rehabilitation training data, including: S310, determining the deviation direction of the action data corresponding to the task type according to the action data corresponding to the task type and the standard action data in the current training task information.

[0070] It can be understood that the deviation direction can represent the difference direction between the patient's actual action and the standard action. According to different task types and calculation methods of the deviation value, the deviation direction can have multiple manifestations, which can be specifically determined according to the task type and the characteristics of the action data.

[0071] Exemplarily, the rotation deviation direction is determined as follows: if the current rotation angle is greater than the standard angle (e.g., the patient rotates 40° and the standard requirement is 30°), the deviation direction may be excessive. If the current rotation angle is less than the standard angle (e.g., the patient rotates 20° and the standard requirement is 30°), the deviation direction may be insufficient.

[0072] Action duration deviation direction judgment: If the current duration is greater than the standard duration (e.g. the patient's action time is 70 seconds and the standard is 60 seconds), the deviation direction is too long. If the current duration is less than the standard duration (e.g. the patient's action time is 50 seconds and the standard is 60 seconds), the deviation direction is too short.

[0073] Speed ​​deviation direction determination: If the patient's rotation speed is greater than the standard speed (such as the patient's rotation speed is 65° / second, the standard is 60° / second), the deviation direction is too fast. If the patient's rotation speed is less than the standard speed (such as the patient's rotation speed is 55° / second, the standard is 60° / second), the deviation direction is too slow.

[0074] Determination of position deviation direction: If the patient's balance angle deviates from the standard position and the deviation is greater than the standard allowable range, the deviation direction is too far.

[0075] The deviation direction can be determined for the action data whose deviation value exceeds the corresponding action threshold, and there is no need to determine the deviation direction for the action data whose deviation value is within the corresponding action threshold. For example, the action data of rotation training involves the action duration and rotation angle. The patient rotates 40°, the standard is 30°, and the corresponding threshold is 5°, then the deviation value (10°) exceeds the threshold (5°); the patient's action time is 61 seconds, the standard is 60 seconds, and the corresponding threshold is 3 seconds, then the deviation value (1 second) does not exceed the threshold (3 seconds); then it is only necessary to determine the patient's rotation deviation direction, and the deviation direction is excessive, and there is no need to determine the deviation direction of the action duration.

[0076] S320: Acquire a first text template according to a deviation direction of the action data corresponding to the task type.

[0077] It can be understood that based on step S210 and step S220, it can be known that each task type corresponds to multiple action data, and a text template can be set for each action data. The text template can be preset as a general guidance statement, and the template content can include information such as how to correct deviations and how to improve training.

[0078] Exemplarily, the first text template for rotation angle (such as balance training, rotation training, etc.) may include a text template for excessive deviation direction and a text template for insufficient deviation direction. The text template for excessive deviation direction may be set to "Your rotation angle is {current angle}°, which exceeds the standard requirement of {standard angle}°. Please reduce the rotation amplitude to the standard angle."; the text template for insufficient deviation direction may be set to "Your rotation angle is {current angle}°, which is lower than the standard requirement of {standard angle}°. Please increase the rotation amplitude to the standard angle.".

[0079] The first text template of the action duration (the patient maintains a certain action posture for a certain period of time) can include a text template with a deviation direction of too long and a text template with a deviation direction of too short. The text template with a deviation direction of too long can be set to "Your training time is {current time} seconds, which exceeds the standard time {standard time} seconds. It is recommended to reduce the training time to the standard range."; the text template with a deviation direction of too short can be set to "Your training time is {current time} seconds, which is lower than the standard time {standard time} seconds. Please continue to maintain it until the standard is reached.".

[0080] The first text template of speed may include a text template for a deviation direction of too fast and a text template for a deviation direction of too slow. The text template for a deviation direction of too fast may be set to "Your training speed is {current speed}° / second, which exceeds the standard requirement of {standard speed}° / second. It is recommended to slow down the speed to the standard range."; the text template for a deviation direction of too slow may be set to "Your training speed is {current speed}° / second, which is lower than the standard requirement of {standard speed}° / second. Please increase the speed to the standard range." The first text template of a position (such as balance training) may include a text template for a deviation direction of too far, and the text template for a deviation direction of too far may be set to "Your balance angle is {current angle}°, which deviates from the standard balance position. Please adjust your posture and try to keep it within the standard balance range.".

[0081] A first text template corresponding to the deviation direction may be selected from a collection of pre-stored text templates. For example, based on the example in step S310, it can be known that the rotation deviation direction of the rotation training is excessive, and the duration deviation value does not exceed the corresponding threshold, then a text template with an excessive rotation deviation direction is selected from the collection of pre-stored text templates.

[0082] S330, generating rehabilitation guidance text data based on the action data corresponding to the task type, the standard action data in the current training task information and the first text template.

[0083] Exemplarily, the action data corresponding to the task type and the standard action data in the current training task information are filled into the selected first text template to generate rehabilitation guidance text data. For example, the rotation deviation direction of the rotation training is excessive, and the selected first text template is "Your rotation angle is {current angle}°, which exceeds the standard requirement of {standard angle}°. Please reduce the rotation amplitude to the standard angle." The patient rotated 40°, and the standard is 30°. These data are filled into the text template to obtain the rehabilitation guidance text data "Your rotation angle is 40°, which exceeds the standard requirement of 30°. Please reduce the rotation amplitude to the standard angle.".

[0084] S340, performing voice conversion processing on the rehabilitation guidance text data to obtain rehabilitation guidance voice data.

[0085] For example, a suitable text-to-speech engine can be selected, such as Google TTS, Amazon Polly, Baidu speech synthesis, etc. After selecting the text-to-speech engine, parameters such as speech speed and pitch can be adjusted to make the speech more suitable for the rehabilitation training environment, and the rehabilitation guidance text data is input into the engine, and the engine converts the text into an audio file (rehabilitation guidance voice data), which can be an MP3, WAV or other audio format.

[0086] S350, marking the standard action data in the current training task information to obtain the marked standard action data.

[0087] It can be understood that marking standard action data means making the patient more clearly aware of the requirements of the standard action through visualization or other forms of marking.

[0088] For example, the standard motion data can be marked by means of graphic markings, which are virtual markings, graphics or logos to display the standard motion data. Assuming that the standard rotation angle in rotation training is 30°, the MR glasses can display a virtual arc marking on the screen to indicate the standard rotation angle, and use color changes or other visual effects (for example, green represents compliance with the standard, and red represents deviation). In balance training, the standard position is 0°, and a center point or vertical line can be displayed in the MR glasses to help patients judge their own balance angle.

[0089] Numerical marking can be used to mark standard action data. Numerical marking is to guide patients by displaying the numerical requirements of standard actions (such as angle, duration, speed, etc.), which can be done through a combination of text, numbers or graphics. For example, MR glasses can display standard duration and countdown to help patients track time progress.

[0090] In complex training tasks, graphics and numerical markings can be combined to provide dual feedback, allowing patients to see the numerical requirements of standard movements and graphical visual guidance at the same time.

[0091] By displaying graphic marks and numerical marks in MR glasses, patients can be helped to perform training tasks more accurately, especially in rehabilitation training that requires precise control of angle, duration or speed. Graphic marks provide intuitive visual references, while numerical marks provide clear standard requirements. The combination of the two can enhance patients' understanding and execution of standard movements and improve the effectiveness of rehabilitation training.

[0092] S360: Determine the rehabilitation guidance text data, the rehabilitation guidance voice data, and the marked standard movement data as a rehabilitation guidance plan.

[0093] For example, text data, voice data and marked data can be integrated into a complete rehabilitation guidance program and provided to the patient for the next step of training.

[0094] Through the above steps, specific rehabilitation guidance text can be generated and converted into voice for real-time prompts. At the same time, by marking standard movement data, patients can understand the correct training requirements, which can improve the training effect of patients and make rehabilitation training more personalized and precise.

[0095] S400, outputs a rehabilitation guidance program to guide the patient to perform vertigo rehabilitation training.

[0096] For example, a rehabilitation guidance program can be output through the voice playback module to provide patients with specific movement correction suggestions, training adjustments, and physiological state recovery prompts.

[0097] Rehabilitation guidance programs can be output through a display module (such as the screen of MR glasses) to provide patients with dynamic movement demonstrations, real-time prompts, or comparisons between standard movements and current movements to help patients better understand training tasks.

[0098] A training progress report can be generated to provide feedback on the patient's current training status, goals achieved, and areas for improvement. The report can be provided to the patient or rehabilitation specialist via MR glasses, mobile devices, or telemedicine platforms.

[0099] Rehabilitation training is a process of continuous optimization. After each training session, data can be recorded based on the patient's performance, and the data can be used to optimize the rehabilitation program. For example, if a patient often fails to meet the standard requirements in a certain task, the training difficulty can be increased or the training method can be changed. The patient's physiological state changes can also be tracked to adjust the training intensity and task difficulty, so that rehabilitation training can challenge the patient while avoiding excessive load.

[0100] Through this step, patients can be provided with personalized and timely guidance, which can improve the training effect and avoid delayed recovery due to improper training or wrong posture.

[0101] In a possible implementation, in step S400, outputting a rehabilitation guidance plan includes: S410, sending the rehabilitation guidance text data and the marked standard action data to the display module of the MR glasses for vertigo rehabilitation, so that the display module displays the rehabilitation guidance text data and the marked standard action data.

[0102] For example, the text data can be sent to the display module of the MR glasses through a suitable protocol (such as HTTP, WebSocket, etc.), and the display module receives the text data and renders the data on the display screen. The marked standard action data (such as virtual marking lines, angle values, etc.) are transmitted to the display module through wireless transmission technology (such as Wi-Fi, Bluetooth, etc.).

[0103] After receiving the data, the display module can render the rehabilitation guidance text and standard action graphic marks in real time through the built-in rendering engine of the MR glasses, and display the rehabilitation guidance text and standard action graphic marks in the patient's field of view. According to the text format, the rehabilitation guidance text is displayed in an appropriate font and layout; according to the standard action data, virtual graphic markings or marks are superimposed in the patient's field of view.

[0104] S420, sending the rehabilitation guidance voice data to the voice playing module of the MR glasses for vertigo rehabilitation, so that the voice playing module plays the voice content in the rehabilitation guidance voice data.

[0105] Exemplarily, voice data can be transmitted as a stream to the voice playback module of the MR glasses using a wireless transmission method (such as Wi-Fi, Bluetooth, etc.). After receiving the voice data, the voice playback module plays the voice content through the built-in speakers or headphones of the MR glasses.

[0106] Timestamps, event-driven, or polling mechanisms can be used to achieve consistency between display and voice content. For example, audio and display content can be synchronized based on timestamps so that they are updated at the same time.

[0107] Through the above steps, the rehabilitation guidance text data and the marked standard action data can be effectively sent to the display module of the MR glasses for vertigo rehabilitation, and the rehabilitation guidance voice data can be sent to the voice playback module, thereby achieving dual visual and auditory feedback, helping patients to perform rehabilitation training tasks more accurately, and significantly improving the training effect and patient participation.

[0108] In one possible implementation, see Figure 3 The MR glasses control method for vertigo rehabilitation also includes: S10, obtaining the patient's voice data.

[0109] For example, the voice data of the patient can be recorded in real time using a voice acquisition module (such as a microphone or a built-in audio sensor) of the MR glasses for vertigo rehabilitation. The voice data can include the patient's voice feedback or interactive content during the training process. The recorded voice data can be stored in the form of an audio file (such as WAV, MP3, FLAC, etc.).

[0110] S20, performing noise reduction processing on the patient's voice data to obtain first voice data.

[0111] Exemplarily, after receiving the voice data, a noise reduction algorithm can be used to perform noise reduction on the voice data. For example, spectral subtraction estimates the noise spectrum and subtracts the noise from the signal to retain the main voice signal; Wiener filtering removes noise by filtering based on statistical information; deep learning noise reduction uses deep neural networks (DNNs) or convolutional neural networks (CNNs) to enhance the voice signal. The voice data (first voice data) after noise reduction is clearer, the background noise is effectively suppressed, and the voice content is easier to process.

[0112] S30, performing voice content recognition on the noise-reduced voice data to obtain first text data.

[0113] Exemplarily, a speech recognition model such as Google Speech-to-Text, Microsoft AzureSpeech, or other APIs may be used to recognize the noise-reduced speech data and convert the speech data into text (first text data), or local speech recognition technology may be used to accomplish this.

[0114] S40, performing purpose prediction on the first text data to obtain a prediction result, and determining whether the patient meets the rehabilitation guidance conditions based on the prediction result.

[0115] Exemplarily, the purpose of the first text data can be predicted based on the rules, and the patient's intention can be judged based on keywords or fixed sentence patterns. For example, keywords such as completed, reached, and standard identify progress feedback; keywords such as next and what to do identify the need for guidance. Deep learning models (such as BERT, GPT and other language models) can be used to understand the patient's intention based on context. The prediction results may include progress feedback (indicating that the patient feeds back his or her own training progress), standard confirmation (indicating that the patient asks whether the standard has been met), guidance request (indicating that the patient requests the next step of guidance), and fatigue / rest request (indicating that the patient requests a rest).

[0116] If the patient's feedback indicates that the standard action has been completed (such as "I have achieved a 30° rotation"), the prediction result is progress feedback, which can verify whether the patient's action meets the standard. If the angle and duration of the action meet the set standards, no further rehabilitation guidance is required. If the patient asks whether the standard has been met (such as "Am I doing it right?"), the prediction result is standard confirmation, which can be compared with the standard based on the current training data. If the patient does not meet the standard, a rehabilitation guidance plan is generated to inform the patient that further training is required. If the patient expresses doubts about the next step of training (such as "What to do next?" or "Can I continue training?"), the prediction result is a guidance request, and appropriate guidance can be generated based on the current training task and progress to guide the patient to the next step of training. If the patient expresses a need for rest (such as "I'm too tired"), the prediction result is a fatigue / rest request, and the training can be paused or the training intensity can be reduced.

[0117] Through the above steps, the patient's purpose can be identified in a timely manner, the patient's training status can be evaluated, and personalized feedback can be provided to the patient, thereby effectively improving the efficiency of rehabilitation training.

[0118] Optionally, see Figure 3 S40, performing purpose prediction on the first text data to obtain a prediction result, and determining whether the patient meets the rehabilitation guidance conditions according to the prediction result, including: S41, extracting text features of the first text data to obtain a text feature vector.

[0119] Exemplarily, the first text data may be preprocessed, such as word segmentation, stop word removal, case conversion, etc. A suitable feature extraction method (such as bag-of-words model, TF-IDF, Word2Vec, GloVe, BERT, etc.) may be selected to convert the preprocessed first text data into a feature vector (text feature vector). The generated text feature vector is a set of numerical values ​​that represents the semantics and structure of the text and can be processed by a machine learning model.

[0120] S42, inputting the text feature vector into a purpose prediction model so that the purpose prediction model predicts the patient's purpose and obtains a prediction result. The purpose prediction model is a machine learning model.

[0121] Exemplarily, the purpose prediction model training process: the patient's voice data can be collected (collected through MR glasses) and converted into text data through voice recognition technology. Each piece of text data can be manually or semi-automatically labeled with the purpose (such as pause request, guidance request, fatigue / rest request, training task change request, standard confirmation, action adjustment request), and each piece of text data is cleaned and preprocessed. The processed text data is input into the feature extraction method to generate the corresponding feature vector. The feature vector represents the vocabulary and semantic information in the text and can be understood by the machine learning model. The text data is divided into a training set (training model parameters), a validation set (validating the performance of the model and adjusting hyperparameters), and a test set (evaluating the generalization ability of the model).

[0122] You can choose a suitable machine learning model to make predictions, such as support vector machine (SVM), random forest, gradient boosting tree (XGBoost, LightGBM). Use the training set to train the selected machine learning model. During the training process, the model will continuously adjust the parameters through optimization algorithms (such as gradient descent) so that the model can accurately predict the purpose of the training data. The model's hyperparameters (such as learning rate, tree depth, regularization parameters, etc.) are adjusted through methods such as cross-validation or grid search to improve the performance of the model. Evaluating the performance of the model on the validation set and using the test set to evaluate the generalization ability of the final trained model will help the model make accurate predictions on actual data.

[0123] The trained purpose prediction model is deployed into the MR glasses system. When a new text feature vector is input, the purpose prediction model can predict the patient's purpose in real time.

[0124] S43, matching the prediction result with the target label, and when the prediction result and the target label are successfully matched, it is determined that the patient meets the rehabilitation guidance conditions.

[0125] Exemplarily, the prediction result is compared with a pre-set target label. If there is a label corresponding to the prediction result in the target label, it can be confirmed that the patient has met the rehabilitation guidance conditions.

[0126] In one possible implementation, see Figure 3 The MR glasses control method for vertigo rehabilitation also includes: S101, determining a rehabilitation guidance template according to a purpose tag, wherein the rehabilitation guidance template includes a task adjustment template and a second text template.

[0127] It can be understood that the task adjustment template is used to adjust the content or intensity of the current training task, and can guide the patient to adjust the training, such as increasing or decreasing the intensity, speed, task content, etc. of the training. The task adjustment template is conducive to adapting the rehabilitation plan to the patient's current condition and avoiding adverse consequences caused by overtraining or too low intensity training. For example, a pause request adjusts the task to pause training and adds a prompt. A fatigue request reduces the load of the current training, extends the rest time, and adjusts to low-intensity training.

[0128] The second text template can provide text-based guidance to patients, help them understand how to adjust their training, and provide necessary guidance and suggestions. The second text template can include training suggestions, recovery tips, emotional support, etc., which can help patients get clear and detailed rehabilitation guidance. For example, the text template for a guidance request can be "It is recommended that you take a deep breath, relax your body, and maintain a relaxed posture."; the text template for a task change request can be "Based on your current physical condition, you have been changed to a certain training.".

[0129] Exemplarily, through the purpose label, the task adjustment template and the second text template that best suit the patient's current state can be selected from a set of predefined templates, and can be matched through a machine learning model (such as a decision tree, SVM, etc.) or a rule engine. For example, when the purpose label is a label for fatigue / rest request, a task adjustment template that increases rest time and reduces training intensity can be selected, and matched with a corresponding text template, such as "The current task has been paused, please rest for five minutes." As the patient's state changes (such as recovery, pain, mood swings, etc.), the target label can be dynamically updated and a suitable rehabilitation guidance template can be reselected.

[0130] S102: Generate a rehabilitation guidance plan based on the current training task information and the rehabilitation guidance template, wherein the rehabilitation guidance plan includes the second text data, the second voice data and the task adjustment information.

[0131] Exemplarily, task adjustment information can be generated according to the current training task information and the task adjustment template. Task adjustment information can be a specific operation, such as adjusting the training intensity, duration, rest time, etc. For example, in the current training task information, the task intensity is medium intensity, the task type is balance training, the preset duration is 10 minutes, the patient's purpose label is fatigue / rest request, and the task adjustment template is to increase the rest time and reduce the training intensity. Then the generated task adjustment information can be to reduce the current balance training intensity to low intensity and increase the rest time by 5 minutes.

[0132] The second text data may be generated based on the current training task information, task adjustment information, and the second text template. For example, the text template with the purpose tag of fatigue / rest request is "The current training intensity is too high, and the training has been paused for you to rest." The second text data generated based on the current training task information, task adjustment information, and text template may be "The current balance training intensity is too high, and the training has been paused for you to rest for 5 minutes, and then low-intensity training will be performed." The second text data may be converted into second voice data, and the conversion method is the same as step S340, which will not be described again.

[0133] The second text data, the second voice data and the task adjustment information are integrated to obtain a rehabilitation guidance program.

[0134] S103, generating a task adjustment instruction according to the task adjustment information, and performing an adjustment operation on the current training task according to the task adjustment instruction.

[0135] Exemplarily, the task adjustment information may be parsed, and a task adjustment instruction may be generated according to the parsed task adjustment information. For example, if the task adjustment information is to reduce the current balance training intensity to low intensity and increase the rest time by 5 minutes, the generated task adjustment instruction may include an instruction type (increase the rest time and adjust the training intensity) and instruction parameters (increase the rest time by 5 minutes and adjust the current balance training intensity from medium intensity to low intensity).

[0136] The task adjustment instruction can be passed to the training task control module, and the training task control module adjusts the current training task according to the task adjustment instruction. The adjustment process can be completed automatically through the control interface, sensor system, feedback system, etc. of the device. For example, the process of adjusting the training intensity: the intensity of visual and motion stimulation can be adjusted through the control module of the MR glasses, and the intensity of exercise in the virtual training environment can be reduced from medium intensity to low intensity, so that the patient can experience a lower training load; the process of increasing the rest time: stop the task content, or display the rest countdown interface to remind the patient to rest.

[0137] S104, after the adjustment operation of the current training task is completed, the second text data is sent to the display module and the second voice data is sent to the voice playback module.

[0138] Exemplarily, after the adjustment operation of the current training task is completed, the second text data is sent to the display module in accordance with the method in step S410; and the second voice data is sent to the voice playback module in accordance with the method in step S420.

[0139] Through the above steps, patients can receive dynamic and personalized guidance during the rehabilitation process, which is conducive to the accuracy and flexibility of the rehabilitation plan.

[0140] In a possible implementation, the MR glasses control method for vertigo rehabilitation further includes: S401, when the patient completes the current training task, obtain the patient's training time.

[0141] Exemplarily, the training duration of the patient can be obtained from the time tracking module, where the training duration is the total time the patient spends from the start of training to the end of the current training task. The time tracking module starts timing when the patient starts training and stops timing when the training ends.

[0142] S402, comparing the patient's training time with the preset time in the current training task information, determining whether the training time is greater than the preset time, and obtaining a second determination result.

[0143] Exemplarily, the training time of the patient is compared with the preset time to determine whether the actual training time of the patient is greater than the preset time. The second determination result is that the training time is greater than the preset time or the training time is within the preset time.

[0144] S403: When the second judgment result indicates that the training duration is greater than the preset duration, it is determined that the patient meets the rehabilitation guidance conditions.

[0145] Exemplarily, when the second judgment result is that the training time is longer than the preset time, it means that the patient has met the rehabilitation guidance conditions, and further rehabilitation guidance can be initiated to provide training task adjustment, rest suggestions or guidance on continuing training.

[0146] In another possible implementation, in step S400, generating a rehabilitation guidance program according to the patient's current rehabilitation training data includes: When the patient completes the current training task, the patient's training time is longer than the preset time of the current task, and the patient meets the rehabilitation guidance conditions, perform the following steps: S501, determining the patient's current task completion accuracy based on current motion data and standard motion data.

[0147] It can be understood that the current task completion accuracy can refer to the degree of completion of the patient in the current training task based on the comparison between the current action data and the standard action data. The current task completion accuracy can be quantified by calculating the difference between the patient's actual action and the standard action.

[0148] For example, in order to calculate the current task completion accuracy, a Euclidean distance method, an angle error method, or a dynamic time warping (DTW) method may be used to quantify the difference between the current action and the standard action.

[0149] Euclidean distance is a method for calculating the difference between two points (or two sets of data). For the action data at each time point, the Euclidean distance between the current action data and the standard action data can be calculated. The completion accuracy is evaluated by calculating the total difference of the entire action sequence.

[0150] If the training task involves rotation or angle adjustment, the angular difference between the current action and the standard action can be calculated. By calculating the error between the two angles, the accuracy of the task can be quantified.

[0151] DTW is an algorithm used to compare time series data. For some complex action sequences, the DTW algorithm can best match misaligned actions and calculate the difference between the current action and the standard action.

[0152] By comparing the difference between the current action data and the standard action data, the completion accuracy of the task can be calculated, and the calculation result of the completion accuracy is a value ranging from 0 to 1. High accuracy (close to 1) indicates that the patient performs well in the current task, the completed action is close to the standard, and the training goal is close to being achieved; medium accuracy means that the patient may need more training to improve the action, and may need to make fine adjustments or adjust the training difficulty based on the accuracy data; low accuracy (close to 0) indicates that the patient's current task completion is low, the action is greatly different from the standard, and higher intensity or different types of training are needed to improve.

[0153] S502, obtaining historical rehabilitation training data corresponding to the task type, wherein the historical rehabilitation training data includes the patient's historical motion data, historical training duration, and historical training task information.

[0154] Exemplarily, historical rehabilitation training data corresponding to the task type can be obtained from the database of MR glasses or the cloud platform. For example, if the current task is balance training, the balance training data that the patient has performed in the past can be obtained. When the patient uses MR glasses for vertigo rehabilitation training, the MR glasses can record relevant data of the patient's training process, such as training task information, patient motion data during training, training duration, etc. When the training is completed, the MR glasses can upload the recorded data to the cloud platform or save it locally.

[0155] If the training duration and data dimensions of different tasks are different, the data can be adjusted to the same scale through normalization to facilitate comparison. The historical action data, historical training duration, historical training task information and other data are aggregated to form a complete training file. The data can be arranged on a timeline to view the changes in the patient's performance in different time periods.

[0156] S503, determining the patient's historical task completion accuracy based on the historical action data and the standard action data in the historical training task information.

[0157] Exemplarily, historical task completion accuracy may refer to the degree of matching between historical motion data and standard motion data of a patient in a historical training task. The difference between historical motion data and standard motion data may be evaluated using the same method as the current task completion accuracy calculation method.

[0158] S504, calculating the average value of the historical training duration in the historical rehabilitation training data to obtain the historical duration mean.

[0159] For example, assuming that the patient has performed 5 balance training sessions in historical rehabilitation training, and the training durations of the five training sessions are 35 minutes, 40 minutes, 30 minutes, 20 minutes, and 25 minutes respectively, then the total historical training duration is 150 minutes, and the historical average duration is 30 minutes.

[0160] S505: Compare the historical duration average with the preset duration in the historical training task information to obtain a comparison result.

[0161] For example, the historical duration average can be compared with the preset duration in the historical training task information. If the historical duration average is less than or close to the preset duration, it indicates that the patient was able to complete longer tasks in past training and is suitable for the current task; if the historical duration average is greater than the preset duration, the training duration or the difficulty of the training task may need to be adjusted. The comparison result can be quantified. If the historical duration average is less than or close to the preset duration, the comparison result is 0; if the historical duration average is greater than the preset duration, the comparison result is 1.

[0162] S506, based on the current task completion accuracy, the historical task completion accuracy and the comparison result, determine the task difficulty of the current training task and obtain task difficulty information.

[0163] For example, the calculation of task difficulty can be performed by combining the current task completion accuracy, historical task completion accuracy and comparison results to obtain the comprehensive difficulty of the task by weighted average or scoring. Here, the weighted average method is used to calculate the task difficulty, and weights can be assigned to the current task completion accuracy, historical task completion accuracy and comparison results, respectively. , , The task difficulty calculation formula is: ,in, Indicates the difficulty of the task, Indicates the current task completion accuracy, Indicates the historical task completion accuracy, Indicates the comparison results.

[0164] Assume that the current task completion accuracy is 0.8, the historical task completion accuracy is 0.7, the historical average duration is 40 minutes, the historical preset duration is 30 minutes, and the comparison result is 1. 0.4, is 0.3, is 0.3, then the task difficulty is .

[0165] S507, generating a rehabilitation guidance plan based on the task difficulty information and the current training task information.

[0166] It can be understood that the task difficulty information provides an evaluation result of the current training task, which can help determine the adaptability of the training task. If the training task is of high difficulty, the training time can be increased, the training intensity can be reduced, and the rest time can be increased to prevent the patient from feeling overly tired or injured; if the training task is of medium difficulty, the existing task duration or intensity can be maintained, but some challenges can be added to enhance the patient's sense of participation and training effect; if the training task is of low difficulty, the training intensity or task complexity may need to be increased to promote faster progress of the patient.

[0167] Exemplarily, the intensity, duration, and training content of the current training task can be adjusted accordingly according to the difficulty level of the task (low, medium, high), and the difficulty level of the task can be divided according to the task difficulty information, for example, 0~0.4 is low difficulty, 0.4~0.6 is medium difficulty, and 0.6~1 is high difficulty.

[0168] The difficulty level of the task is low difficulty. You can increase the difficulty of the task and increase the intensity or duration of the training. For example, increase the duration of the training task, such as extending the current 20-minute task to 25 minutes; increase the complexity of the task, such as adding more posture changes in posture training.

[0169] If the difficulty level of the task is medium, you can continue to maintain the difficulty of the current task, or fine-tune the task duration and intensity. For example, you can continue the current task, but you can fine-tune the task duration, or add new action challenges (such as higher rotation frequency, balance angle, etc.).

[0170] If the difficulty level of the task is high, the duration or intensity of the task can be appropriately reduced, or the way the task is performed can be changed. For example, the task duration can be extended from 30 minutes to 40 minutes; if the patient's balance accuracy is low, the complexity of the task can be reduced, such as reducing the balance requirements, or increasing auxiliary support in the task.

[0171] According to the above method, task adjustment information can be generated and task adjustment instructions can be generated based on the task adjustment information, and the task adjustment instructions can be sent to the training task control module.

[0172] New text data and voice data can be generated based on the generated rehabilitation guidance plan to provide task guidance and feedback.

[0173] For example, the current task is difficult and the patient's task completion accuracy is low. The rehabilitation guidance plan generated by the MR glasses may include task adjustment information to reduce the task duration, adjust 30 minutes to 40 minutes, reduce training complexity, and reduce step frequency; the new text data is "Your training duration has been adjusted to 40 minutes, and the step frequency has been reduced. Please train according to the new requirements and keep your steps stable."; the new voice data is "Your training duration has been adjusted to 40 minutes, and the step frequency has been reduced. Please train according to the new requirements and keep your steps stable.".

[0174] By comprehensively analyzing the patient's current task completion accuracy, historical task completion accuracy, and historical average duration, it can intelligently assess the difficulty of the task and generate targeted rehabilitation guidance plans. This is conducive to rehabilitation training that not only meets the patient's needs, but also gradually improves the ability to complete tasks, thereby promoting better rehabilitation results.

[0175] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0176] Corresponding to the MR glasses control method for vertigo rehabilitation described in the above embodiment, the embodiment of the present application also provides a MR glasses control device for vertigo rehabilitation, and each unit of the device can implement each step of the MR glasses control method for vertigo rehabilitation. Figure 4 A structural block diagram of a MR glasses control device for vertigo rehabilitation provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0177] Reference Figure 4 , the device comprises: The acquisition unit is used to acquire the patient's current rehabilitation training data. The patient's current rehabilitation training data includes current motion data and current training task information. The current training task information includes task type, standard motion data, motion threshold, and preset duration. The current motion data is acquired through the data acquisition module of the MR glasses for vertigo rehabilitation.

[0178] The judgment unit is used to determine whether the patient meets the rehabilitation guidance condition based on the patient's current rehabilitation training data. The rehabilitation guidance condition is used to indicate that the patient needs training guidance when using the MR glasses for vertigo rehabilitation to perform vertigo rehabilitation training.

[0179] The program generation unit is used to generate a rehabilitation guidance program according to the patient's current rehabilitation training data if the patient meets the rehabilitation guidance conditions.

[0180] The output unit is used to output a rehabilitation guidance program to guide the patient to perform vertigo rehabilitation training.

[0181] It should be noted that the information interaction, execution process and other contents between the above-mentioned units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0182] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0183] The present application also provides a MR glasses for vertigo rehabilitation. Figure 5 The structural diagram of the MR glasses for vertigo rehabilitation provided by an embodiment of the present application. The MR glasses for vertigo rehabilitation include a display module, a voice playback module, a data acquisition module and a control module. Figure 5 As shown, the control module 6 of the MR glasses for vertigo rehabilitation of this embodiment includes: at least one processor 60 ( Figure 5 Only one is shown), at least one memory 61 ( Figure 5 Only one is shown in the figure) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the MR glasses for vertigo rehabilitation implement the steps of any of the above-mentioned embodiments of the MR glasses control method for vertigo rehabilitation, or implement the functions of each unit in the above-mentioned device embodiments.

[0184] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 62 in the control module 6 of the MR glasses for vertigo rehabilitation.

[0185] The control device 6 of the MR glasses for vertigo rehabilitation can be a central processing unit (CPU / SoC), a motion capture and analysis module, a wireless communication module (such as Bluetooth, Wi-Fi), a battery management and power control module, etc. The MR glasses for vertigo rehabilitation can include, but are not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Figure 5 It is merely an example of MR glasses for vertigo rehabilitation and does not constitute a limitation on MR glasses for vertigo rehabilitation. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, buses, etc.

[0186] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0187] In some embodiments, the memory 61 may be an internal storage unit of the control module 6 of the MR glasses for vertigo rehabilitation, such as a hard disk or memory of the MR glasses for vertigo rehabilitation. In other embodiments, the memory 61 may also be an external storage device of the MR glasses for vertigo rehabilitation, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the MR glasses for vertigo rehabilitation. Further, the memory 61 may also include both the internal storage unit of the MR glasses for vertigo rehabilitation and an external storage device. The memory 61 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0188] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0189] An embodiment of the present application provides a computer program product. When the computer program product is run on MR glasses for vertigo rehabilitation, the MR glasses for vertigo rehabilitation implement the steps in any of the above-mentioned method embodiments.

[0190] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to MR glasses for dizziness rehabilitation, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0191] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0192] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0193] In the embodiments provided in the present application, it should be understood that the disclosed MR glasses control method, device and MR glasses for vertigo rehabilitation can be implemented in other ways. For example, the MR glasses control device for vertigo rehabilitation and the MR glasses for vertigo rehabilitation embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0194] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0195] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for controlling MR glasses for vertigo rehabilitation, characterized in that: include: Acquire the patient's current rehabilitation training data; wherein the patient's current rehabilitation training data includes current motion data and current training task information, the current training task information includes task type, standard motion data, motion threshold, and preset duration, and the current motion data is acquired through a data acquisition module of MR glasses for vertigo rehabilitation; Based on the current rehabilitation training data of the patient, determining whether the patient meets the rehabilitation guidance condition; wherein the rehabilitation guidance condition is used to indicate that the patient needs training guidance when performing vertigo rehabilitation training using the MR glasses for vertigo rehabilitation; If the patient meets the rehabilitation guidance conditions, generating a rehabilitation guidance plan according to the patient's current rehabilitation training data; The rehabilitation guidance program is output to guide the patient to perform vertigo rehabilitation training.

2. The MR glasses control method for vertigo rehabilitation according to claim 1, characterized in that: The determining whether the patient meets the rehabilitation guidance conditions based on the current rehabilitation training data of the patient includes: According to the task type in the current training task information, extracting action data corresponding to the task type from the current action data; Calculating the deviation between the action data corresponding to the task type and the standard action data in the current training task information to obtain a deviation value; Comparing the deviation value with the action threshold in the current training task information, determining whether the deviation value exceeds the action threshold, and obtaining a first determination result; When the first judgment result indicates that the deviation value exceeds the action threshold, it is determined that the patient meets the rehabilitation guidance condition.

3. The MR glasses control method for vertigo rehabilitation according to claim 2, characterized in that: Generating a rehabilitation guidance program according to the current rehabilitation training data of the patient includes: Determining a deviation direction of the action data corresponding to the task type according to the action data corresponding to the task type and the standard action data in the current training task information; Acquire a first text template according to the deviation direction of the action data corresponding to the task type; Generate rehabilitation guidance text data based on the action data corresponding to the task type, the standard action data in the current training task information and the first text template; Performing voice conversion processing on the rehabilitation guidance text data to obtain rehabilitation guidance voice data; Marking the standard action data in the current training task information to obtain the marked standard action data; The rehabilitation guidance text data, the rehabilitation guidance voice data and the marked standard action data are determined as the rehabilitation guidance plan.

4. The MR glasses control method for vertigo rehabilitation according to claim 3, characterized in that: The outputting of the rehabilitation guidance program includes: Sending the rehabilitation guidance text data and the marked standard action data to a display module of MR glasses for vertigo rehabilitation, so that the display module displays the rehabilitation guidance text data and the marked standard action data; The rehabilitation guidance voice data is sent to a voice playing module of the MR glasses for vertigo rehabilitation, so that the voice playing module plays the voice content in the rehabilitation guidance voice data.

5. The MR glasses control method for vertigo rehabilitation according to claim 4, characterized in that: The method further comprises: Acquiring voice data of the patient; Performing noise reduction processing on the patient's voice data to obtain first voice data; Performing voice content recognition on the noise-reduced voice data to obtain first text data; The first text data is subjected to purpose prediction to obtain a prediction result, and based on the prediction result, it is determined whether the patient meets the rehabilitation guidance condition.

6. The MR glasses control method for vertigo rehabilitation according to claim 5, characterized in that: The performing purpose prediction on the first text data to obtain a prediction result, and determining whether the patient meets the rehabilitation guidance condition according to the prediction result, includes: Extracting text features of the first text data to obtain a text feature vector; Inputting the text feature vector into a purpose prediction model so that the purpose prediction model predicts the patient's purpose and obtains a prediction result; wherein the purpose prediction model is a machine learning model; The prediction result is matched with the target label, and when the prediction result is successfully matched with the target label, it is determined that the patient meets the rehabilitation guidance condition.

7. The MR glasses control method for vertigo rehabilitation according to claim 6, characterized in that: The method further comprises: Determining a rehabilitation guidance template according to the purpose tag; wherein the rehabilitation guidance template includes a task adjustment template and a second text template; Based on the current training task information and the rehabilitation guidance template, generating the rehabilitation guidance plan; wherein the rehabilitation guidance plan includes the second text data, the second voice data and the task adjustment information; Generate a task adjustment instruction according to the task adjustment information, and perform an adjustment operation on the current training task according to the task adjustment instruction; After the adjustment operation of the current training task is completed, the second text data is sent to the display module and the second voice data is sent to the voice playback module.

8. The MR glasses control method for vertigo rehabilitation according to claim 1, characterized in that: The method further comprises: When the patient completes the current training task, obtaining the training time of the patient; Compare the training duration of the patient with the preset duration in the current training task information, determine whether the training duration is greater than the preset duration, and obtain a second determination result; When the second judgment result indicates that the training duration is greater than the preset duration, determining that the patient meets the rehabilitation guidance condition; Generating a rehabilitation guidance program according to the current rehabilitation training data of the patient includes: Determining the patient's current task completion accuracy according to the current motion data and the standard motion data; Acquire historical rehabilitation training data corresponding to the task type; wherein the historical rehabilitation training data includes the patient's historical movement data, historical training duration, and historical training task information; Determining the patient's historical task completion accuracy based on the historical action data and the standard action data in the historical training task information; Calculating the average of the historical training durations in the historical rehabilitation training data to obtain a historical duration mean; Compare the historical duration average with the preset duration in the historical training task information to obtain a comparison result; Determine the task difficulty of the current training task based on the current task completion accuracy, the historical task completion accuracy and the comparison result, and obtain task difficulty information; The rehabilitation guidance plan is generated based on the task difficulty information and the current training task information.

9. A MR glasses control device for vertigo rehabilitation, characterized in that: include: An acquisition unit, used for acquiring the patient's current rehabilitation training data; wherein the patient's current rehabilitation training data includes current motion data and current training task information, the current training task information includes task type, standard motion data, motion threshold, and preset duration, and the current motion data is acquired through a data acquisition module of MR glasses for vertigo rehabilitation; A judgment unit, used for determining whether the patient meets the rehabilitation guidance condition based on the current rehabilitation training data of the patient; wherein the rehabilitation guidance condition is used to indicate that the patient needs training guidance when performing vertigo rehabilitation training using the MR glasses for vertigo rehabilitation; a program generating unit, configured to generate a rehabilitation guidance program according to the current rehabilitation training data of the patient if the patient meets the rehabilitation guidance conditions; The output unit is used to output the rehabilitation guidance program to guide the patient to perform vertigo rehabilitation training.

10. A MR glasses for vertigo rehabilitation, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images