Medical health management system and method based on virtual reality

Abstract

The invention provides a medical health management system and method based on virtual reality, and the method comprises the steps: predicting the training adaptability of a target patient to a current medical rehabilitation training item through the difficulty level of the current medical rehabilitation training item and the posture deviation in a medical rehabilitation training process, and obtaining a fitness prediction value; determining behavior association degrees between the target patient and other patients in the medical virtual reality interaction process, and further performing collaborative mapping on the adaptability of the current medical rehabilitation training item according to all the behavior association degrees to obtain collaborative characteristics of the target patient and other patients in the current medical rehabilitation training item; and carrying out cooperative constraint on the fitness predicted value through the cooperative features, and when the constrained fitness predicted value is lower than a preset fitness threshold value, adjusting the guidance speed of rehabilitation guidance of the current medical rehabilitation training item in the virtual reality equipment. By adopting the scheme of the invention, personalized medical health management based on the patient group effect can be realized.

Description

Technical Field

[0001] This application relates to the field of virtual reality technology, and more specifically, to a virtual reality-based medical and health management system and method. Background Technology

[0002] Virtual Reality (VR) technology has seen rapid development in healthcare management in recent years, becoming an important tool for enhancing rehabilitation training, surgical simulation, psychotherapy, and telemedicine. VR generates highly realistic immersive virtual environments through computers, allowing patients to undergo medical training or treatment in a safe and controllable environment, reducing the risks of real-world procedures and improving treatment adherence. In the field of medical rehabilitation, VR can provide personalized training programs to help patients recover limb function, undergo neurological rehabilitation, and perform motor training. It can also optimize training effects by monitoring patients' posture, movements, and training progress through real-time data.

[0003] In traditional medical rehabilitation training, the adjustment of training difficulty usually relies on individual patient performance data, such as training completion rate, error rate, or rehabilitation progress. However, this approach has certain limitations, primarily the lack of collaborative analysis of patient group data. This results in overly simplistic difficulty settings that fail to adequately meet the individualized needs of different patients. Existing methods for adjusting training difficulty typically employ fixed evaluation criteria, such as adjusting training levels based on patient completion time or postural deviations. However, this ignores individual patient differences and the impact of group rehabilitation models on individual adaptability. Furthermore, the lack of collaborative optimization based on group data leaves difficulty adjustments without a reference basis, leading some patients to encounter excessively high or low challenges during training, thus affecting rehabilitation outcomes. Introducing group data for intelligent optimization, such as dynamically adjusting training difficulty by analyzing the adaptation of similar patients, can improve the rationality of training, making training programs more precise and scientific. This helps improve patient rehabilitation efficiency and adaptability. Therefore, how to achieve personalized medical and health management based on patient group effects has become a challenge for the industry. Summary of the Invention

[0004] This application provides a virtual reality-based medical and health management system and method that can realize personalized medical and health management based on patient group effects.

[0005] In a first aspect, this application provides a virtual reality-based medical and health management method, comprising the following steps: The simulated and actual postures of the target patient during medical rehabilitation training using virtual reality equipment are monitored. The motion joints in the simulated and actual postures are differentially perceived, and then the posture deviations of the target patient during medical rehabilitation training are extracted. The training adaptability of the target patient to the current medical rehabilitation training program is predicted by the difficulty level of the current medical rehabilitation training program and the postural deviation, and the fitness prediction value is obtained. Determine the behavioral correlation between the target patient and other patients during medical virtual reality interaction, and then perform a collaborative mapping of the adaptability of the current medical rehabilitation training program based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program. The fitness prediction value is constrained by the cooperative feature to obtain the constrained fitness prediction value. When the constrained fitness prediction value is lower than the preset fitness threshold, the guidance speed of the current medical rehabilitation training project in the virtual reality device is adjusted.

[0006] Preferably, the differential sensing of the motion joints in the simulated posture and the actual posture, and the extraction of posture deviations in the target patient during medical rehabilitation training, specifically includes: The simulated attitude and the actual attitude are aligned in coordinates to obtain the coordinate-aligned simulated attitude and actual attitude. Extract the simulated coordinates of all motion joints from the simulated pose after coordinate alignment; Extract the actual coordinates of all motion joints from the actual pose after coordinate alignment; Differential matching is performed on all simulated coordinates and actual coordinates to obtain the attitude difference degree of all motion joints; Postural deviations in target patients during medical rehabilitation training are determined by the postural differences at all motion joints.

[0007] Preferably, the predicted fitness level of the target patient to adapt to the current medical rehabilitation training program is obtained by using the difficulty level of the current program and the postural deviation, specifically including: The difficulty level of the current medical rehabilitation training program is determined based on the historical completion status of the current medical rehabilitation training program. Initialize an adaptive evaluation model for assessing patients' adaptation to medical rehabilitation training programs; The difficulty level of the current medical rehabilitation training program and the posture deviation are used as input features of the adaptive evaluation model; The adaptive evaluation model is used to predict the fitness level of target patients in adapting to the current medical rehabilitation training program.

[0008] Preferably, determining the behavioral correlation between the target patient and other patients during medical virtual reality interaction specifically includes: Acquire training logs of the target patient and other patients during the medical rehabilitation training process, and extract behavioral characteristics of the target patient and other patients during virtual reality interaction from the training logs; The behavioral correlation between the target patient and each other during medical virtual reality interaction is determined by the Euclidean distance of the behavioral characteristics between the target patient and each other.

[0009] Preferably, the adaptability of the current medical rehabilitation training program is collaboratively mapped based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program, specifically including: Based on all behavioral correlations, a target patient group with similar rehabilitation training behaviors to the target patient was identified; Extract the fitness value of each patient in the target patient group for the current medical rehabilitation training program; Based on the fitness value of each patient in the current medical rehabilitation training program and the behavioral correlation between the target patient and each patient, a personalized adaptation mapping label value between the target patient and each patient in the current medical rehabilitation training program is determined; The collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program are determined by all the mapped label values.

[0010] Preferably, the fitness prediction value is obtained by applying collaborative constraints to the collaborative features, specifically including: The collaborative constraint coefficients for the target patient's adaptation to the current medical rehabilitation training program are determined using the aforementioned collaborative features. The fitness prediction value after constraints is determined by the collaborative constraint coefficient and the fitness prediction value.

[0011] Preferably, the virtual reality device includes a VR headset, motion tracking gloves, force feedback devices, and position tracking devices.

[0012] Secondly, this application provides a virtual reality-based medical and health management system, comprising: The monitoring module is used to monitor the simulated posture and actual posture of the target patient during medical rehabilitation training using virtual reality equipment, perform differential perception on the motion joints in the simulated posture and the actual posture, and then extract the posture deviation of the target patient during medical rehabilitation training. The processing module is used to predict the training adaptability of the target patient to the current medical rehabilitation training program based on the difficulty level of the current medical rehabilitation training program and the posture deviation, and obtain the fitness prediction value. The processing module is also used to determine the behavioral correlation between the target patient and other patients during the medical virtual reality interaction process, and then perform a collaborative mapping on the adaptability of the current medical rehabilitation training program based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program. The execution module is used to perform collaborative constraints on the fitness prediction value through the collaborative features to obtain a constrained fitness prediction value. When the constrained fitness prediction value is lower than a preset fitness threshold, the guidance speed of the rehabilitation guidance of the current medical rehabilitation training project in the virtual reality device is adjusted.

[0013] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described virtual reality-based medical and health management method.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described virtual reality-based medical and health management method.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In this embodiment, the simulated and actual postures of the target patient during medical rehabilitation training using virtual reality equipment are monitored. Differential perception is performed on the motion joints in the simulated and actual postures to extract posture deviations during the medical rehabilitation training. The training adaptability of the target patient to the current medical rehabilitation training program is predicted based on the difficulty level of the current program and the posture deviations, resulting in a predicted fitness value. The behavioral correlation between the target patient and other patients during the medical virtual reality interaction is determined, and the adaptability of the current medical rehabilitation training program is collaboratively mapped based on all behavioral correlations to obtain collaborative features between the target patient and other patients in the current medical rehabilitation training program. The predicted fitness value is collaboratively constrained using these collaborative features to obtain a constrained predicted fitness value. When the constrained predicted fitness value is lower than a preset fitness threshold, the guidance speed of the rehabilitation guidance in the current medical rehabilitation training program on the virtual reality equipment is adjusted.

[0016] Therefore, this application uses the collaborative characteristics between the target patient and other patients to collaboratively constrain the predicted fitness value of the target patient. When the constrained predicted fitness value is lower than a preset fitness threshold, the guidance speed of the rehabilitation guidance in the current medical rehabilitation training program in the virtual reality device is adjusted. First, the training adaptability of the target patient to the current medical rehabilitation training program is predicted by the difficulty level of the current medical rehabilitation training program and the postural deviations of the target patient during the medical rehabilitation training process, resulting in a predicted fitness value. By combining the challenge of the training program and the patient's own motor ability, the scientific nature of the difficulty assessment is ensured, and a quantitative indicator is provided for personalized difficulty adjustment. Compared with traditional methods, this scheme not only considers individual motor ability but also dynamically tracks rehabilitation progress, making the difficulty setting more targeted. Second, the adaptability of the current medical rehabilitation training program is collaboratively mapped based on all behavioral correlations, obtaining the adaptation of the target patient and other patients in the current medical rehabilitation training program. The collaborative features of the project, by analyzing the training behaviors of similar patients, construct an adaptive mapping based on group collaboration. This ensures that adjusting the training difficulty for the target patient is no longer an isolated decision, but rather an optimization that incorporates the rehabilitation trajectories of similar patients. This process effectively avoids assessment bias caused by limited individual data, making the difficulty adjustment more stable and reasonable. Finally, collaborative features are used to constrain the fitness prediction value, ensuring that the training difficulty adjustment has higher accuracy and dynamic adaptability. When the constrained fitness prediction value is lower than the preset fitness threshold, the system will automatically adjust the difficulty of the health guidance actions in the training program, making the training more in line with the patient's actual recovery ability. Compared with traditional methods that rely on manual adjustment or fixed rule adjustments, this solution can dynamically and intelligently optimize the training difficulty, avoiding overly challenging or overly simple training, thereby improving rehabilitation efficiency and ensuring continuous optimization of training effects. In summary, the proposed solution can realize personalized medical and health management based on the patient group effect. Attached Figure Description

[0017] Figure 1 This is an exemplary flowchart of a virtual reality-based medical and health management method according to some embodiments of this application; Figure 2 This is a logical framework diagram of a virtual reality-based medical and health management system according to some embodiments of this application; Figure 3 This is a flowchart illustrating the determination of behavioral correlation degree according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a virtual reality-based medical and health management system according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a virtual reality-based medical and health management method, according to some embodiments of this application. Detailed Implementation

[0018] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] refer to Figure 1 The figure is an exemplary flowchart of a virtual reality-based medical and health management method 100 according to some embodiments of this application. The virtual reality-based medical and health management method 100 mainly includes the following steps: In step 101, the simulated posture and actual posture of the target patient during medical rehabilitation training using virtual reality equipment are monitored, and the motion joints in the simulated posture and the actual posture are differentially perceived to extract the posture deviation of the target patient during medical rehabilitation training.

[0020] It should be noted that the reference Figure 2 As shown, this figure is a logical framework diagram of a virtual reality-based medical and health management system according to some embodiments of this application. The computing system is responsible for processing motion signals from sensors and generating control commands. The virtual reality control program receives the control commands from the computing system and controls the generation and changes of the virtual reality scene. The virtual reality scene refers to providing a virtual treatment environment for the patient according to the instructions of the control program. Sensors can capture the patient's motion signals and feed them back to the computing system. The virtual human is used to interact with the patient and provide motion guidance commands. The patient receives treatment in the virtual environment, and their motion and motion signals are captured by sensors and fed back to the system.

[0021] It should be noted that Virtual Reality (VR) devices are devices that create and simulate three-dimensional virtual environments using computer technology. Patients can interact with this virtual environment by wearing or using VR devices. In medical rehabilitation training, VR devices play the following roles: providing a safe training environment. Medical rehabilitation training often involves certain physical activities, such as movement and coordination training. VR devices can provide patients with a risk-free training environment, avoiding problems such as falls and injuries caused by training, especially in the early stages of rehabilitation; increasing patient participation and enthusiasm. Traditional rehabilitation training can be monotonous and boring, which may lead to a lack of motivation among patients. However, through VR, patients can engage in gamified rehabilitation training in an immersive virtual environment, enhancing the fun and challenge of training, and improving participation and continuity; providing real-time feedback and evaluation. VR devices can accurately record every movement, posture, and reaction of the patient during training. Doctors can monitor rehabilitation progress in real time based on this data and generate feedback reports to help the medical team adjust the training plan and accurately assess the patient's rehabilitation status.

[0022] It should also be noted that the virtual reality device in this application includes a VR headset, motion tracking gloves, force feedback devices, and position tracking devices. The VR headset is used to provide an immersive visual experience, the motion tracking gloves are used to track hand movements or other body parts movements, the force feedback devices are used to simulate tactile feedback, allowing users to perceive the texture and weight of objects in the virtual environment, and the position tracking devices are used to track the user's spatial position to ensure the realism of movements in the virtual environment.

[0023] Additionally, it should be noted that the simulated posture in this application refers to the simulated movement posture of the patient in a virtual environment. The simulated posture can provide the patient with an ideal movement posture reference and is a standardized movement model in rehabilitation training; the actual posture refers to the patient's real movement posture in the real world.

[0024] In practice, monitoring the simulated and actual postures of the target patient during medical rehabilitation training using virtual reality equipment can be achieved in the following ways: the simulated patient's movement posture in the virtual environment can be obtained from the simulation unit of the virtual reality equipment, and the simulated patient's movement posture can be used as the simulated posture; the patient's real movement posture in the real world can be captured in real time using an inertial measurement unit or motion capture equipment, and the real movement posture can be used as the actual posture.

[0025] In some embodiments, differential sensing of the motion joints in the simulated posture and the actual posture, and then extraction of posture deviations in the target patient during medical rehabilitation training, can be achieved through the following steps: The simulated attitude and the actual attitude are aligned in coordinates to obtain the coordinate-aligned simulated attitude and actual attitude. Extract the simulated coordinates of all motion joints from the simulated pose after coordinate alignment; Extract the actual coordinates of all motion joints from the actual pose after coordinate alignment; Differential matching is performed on all simulated coordinates and actual coordinates to obtain the attitude difference degree of all motion joints; Postural deviations in target patients during medical rehabilitation training are determined by the postural differences at all motion joints.

[0026] It should be noted that the motion joints in this application refer to important joint positions with high degrees of freedom and function during human movement. They are usually key support points for the human body during movement and can reflect the state and posture of the human body during movement. Each motion joint corresponds to one or more joints or parts and has relatively obvious changes during movement, which can serve as a marker of the human body's movement state. Examples of motion joints include the head, shoulders, elbows, knees, and ankles.

[0027] It should also be noted that the posture difference degree in this application is an indicator that measures the degree of difference between the simulated coordinates and the actual coordinates at the motion joint; the posture deviation in this application is an indicator that measures the degree of difference between the actual posture and the simulated posture of the patient during medical rehabilitation training.

[0028] In specific implementation, aligning the simulated posture and the actual posture with coordinates can be achieved in the following way: Homogeneous coordinate transformation, a technique used in the prior art, can be employed to align the simulated posture and the actual posture to the same coordinate system. The simulated posture in this coordinate system is then used as the aligned simulated posture, and the actual posture in this coordinate system is used as the aligned actual posture. Coordinate alignment allows for comparison between the simulated posture and the actual posture in the same space. Extracting the simulated coordinates of all motion joints from the aligned simulated posture can be achieved in the following way: A preset annotation model can be used to annotate the joints in the aligned simulated posture, and the coordinates of the annotated joints are then used as the simulated coordinates of the corresponding motion joints. Extracting the actual coordinates of all motion joints in the actual posture after alignment can be achieved in the following way: a preset annotation model can be used to annotate the joints in the actual posture after coordinate alignment, and then the coordinates of the annotated joints can be used as the actual coordinates of the corresponding motion joints; performing differential matching on all simulated coordinates and actual coordinates to obtain the posture difference of all motion joints can be achieved in the following way: the Euclidean distance between the simulated coordinates and actual coordinates at each motion joint can be used as the posture difference of each motion joint; determining the posture deviation of the target patient during medical rehabilitation training through the posture difference of all motion joints can be achieved in the following way: the sum of the posture difference of all motion joints can be used as the posture deviation of the target patient during medical rehabilitation training.

[0029] It should be noted that the annotation model in this application is based on a preset joint position template or human skeleton model. Using computer vision and machine learning techniques, it identifies and annotates the positions of key motion joints in the coordinate-aligned simulated and actual postures. This annotation model is usually trained to learn the positional relationships and motion patterns of different joints in three-dimensional space. It can automatically detect and annotate the position of each joint. In this way, the system can accurately map the joints in the coordinate-aligned simulated and actual postures to the human body model, thereby extracting the simulated and actual coordinates of each motion joint for subsequent analysis and calculation.

[0030] In step 102, the training adaptability of the target patient to the current medical rehabilitation training program is predicted by the difficulty level of the current medical rehabilitation training program and the posture deviation, and an adaptation prediction value is obtained.

[0031] In some embodiments, predicting the training adaptability of a target patient to the current medical rehabilitation training program based on the difficulty level of the program and the postural deviation, and obtaining the predicted fitness value, can be achieved through the following steps: The difficulty level of the current medical rehabilitation training program is determined based on the historical completion status of the current medical rehabilitation training program. Initialize an adaptive evaluation model for assessing patients' adaptation to medical rehabilitation training programs; The difficulty level of the current medical rehabilitation training program and the posture deviation are used as input features of the adaptive evaluation model; The adaptive evaluation model is used to predict the fitness level of target patients in adapting to the current medical rehabilitation training program.

[0032] It should be noted that the adaptation evaluation model in this application is a model used to assess the degree of adaptation of patients in the current training program; the adaptation prediction value in this application is a prediction value that measures the adaptability of patients in the current medical rehabilitation training program.

[0033] In specific implementation, determining the difficulty level of the current medical rehabilitation training program based on its historical completion status can be achieved as follows: Training data from different patients during the completion of the current program is obtained. This training data includes the patient's training duration, completion rate, and error rate. The weighted sum of the average training duration, average completion rate, and average error rate is then used as the difficulty level index for the current medical rehabilitation training program. This index reflects the difficulty level of the current medical rehabilitation training program. It should be noted that in this application, the weights for training duration, completion rate, and error rate are 0.4, 0.3, and 0.3 respectively. In other embodiments, the weights for training duration, completion rate, and error rate can be set to other values ​​depending on actual needs; no specific limitations are made here. Initializing an adaptive evaluation model for assessing patient adaptation to the medical rehabilitation training program can be achieved as follows: a gated recurrent loop unit can be selected. The Generic Random Unit (GRU) neural network serves as the core architecture of the adaptive evaluation model. Due to its ability to effectively handle the dynamic characteristics of temporal data and posture deviations, its network structure comprises three GRU layers (with 64, 32, and 16 hidden units respectively) and a fully connected output layer. Initial weights are loaded onto the adaptive evaluation model using pre-trained data (such as RehabNet), and hyperparameters such as Dropout regularization, Huber loss function, and Nadam optimizer are configured to establish the basic framework and parameter starting point for subsequent personalized fine-tuning. Using the difficulty level of the current medical rehabilitation training program and the posture deviations as input features of the adaptive evaluation model can be achieved as follows: the posture deviations of the target patient during medical rehabilitation training can be used as the primary feature, while the difficulty level (i.e., the difficulty level index) is one-hot encoded. The two types of features are then concatenated into an input vector after Z-score standardization. The adaptive evaluation model can predict the target patient's fitness level in adapting to the current medical rehabilitation training program using the predicted value of the model's output.

[0034] It should be noted that the training process of the adaptive evaluation model in this application is as follows: Model architecture: A gated recurrent unit (GRU) neural network can be selected as the core architecture of the adaptive evaluation model. The input layer integrates one-hot encoded difficulty level and pose deviation features, and the output layer is mapped to the fitness prediction value via a sigmoid function. The functional relationship is: fitness prediction value = w1 * difficulty level + w2 * pose deviation, where w1 and w2 are weight coefficients, which can be adjusted successively through experiments. In addition, the LeakyReLU activation function can be used to prevent gradient vanishing. Feature engineering: The FastDTW algorithm can be used to quantify the dynamic difference between the action and the standard template, combined with Robust... Scaling eliminates noise interference; the training strategy involves pre-training GRU weights on the RehabNet public dataset and then fine-tuning the target patient data using the Nadam optimizer. The Huber loss function and temporal consistency constraints (L1 smoothing of adjacent frame predictions) are used to balance outliers and dynamic continuity; in the validation and optimization phase, hyperparameters such as the dropout rate (0.3) and learning rate (1e-3) are determined based on patient-wise cross-validation and Bayesian search.

[0035] In step 103, the behavioral correlation between the target patient and other patients during the medical virtual reality interaction process is determined, and then the adaptability of the current medical rehabilitation training program is collaboratively mapped based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program.

[0036] In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the process of determining behavioral correlation in some embodiments of this application. In this embodiment, determining the behavioral correlation between a target patient and other patients during medical virtual reality interaction can be achieved through the following steps: In step 1031, the training logs of the target patient and other patients during the medical rehabilitation training process are obtained, and the behavioral characteristics of the target patient and other patients during the virtual reality interaction process are extracted from the training logs. In step 1032, the behavioral correlation between the target patient and each other patients during the medical virtual reality interaction is determined based on the Euclidean distance of the behavioral characteristics between the target patient and each other patients.

[0037] It should be noted that the training log in this application refers to a detailed record of all data, behaviors, and events related to the training process recorded when the patient undergoes medical rehabilitation training; the behavioral characteristics in this application refer to the performance characteristics of the patient's interactive behavior during virtual reality interaction; and the behavioral correlation in this application is an indicator that measures the similarity of the interactive behavioral characteristics between the target patient and other patients in the virtual reality environment.

[0038] In specific implementation, obtaining training logs of the target patient and other patients during the medical rehabilitation training process, and extracting behavioral characteristics of the target patient and other patients during virtual reality interaction from the training logs can be achieved in the following way: The training logs of the target patient and other patients during the medical rehabilitation training process can be obtained from the background database. These training logs contain training data of the target patient and other patients during the medical rehabilitation training process. Behavioral characteristics of the target patient and other patients' virtual interaction behavior during rehabilitation training using virtual reality devices can then be extracted from the training logs. These behavioral characteristics include the patient's interaction frequency, interaction duration, task completion rate, and error rate. These characteristics are usually collected through sensors or motion capture devices and then processed by noise reduction and normalization. Determining the behavioral correlation between the target patient and each other patient during the medical virtual reality interaction process based on the Euclidean distance of the behavioral characteristics between the target patient and each other patient can be achieved in the following way: The Euclidean distance of the behavioral characteristics between the target patient and each other patient can be used as the behavioral correlation of the target patient's virtual reality interaction behavior with each other patient during the medical rehabilitation training process.

[0039] In some embodiments, the adaptiveness of the current medical rehabilitation training program is collaboratively mapped based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program. This can be achieved through the following steps: Based on all behavioral correlations, a target patient group with similar rehabilitation training behaviors to the target patient was identified; Extract the fitness value of each patient in the target patient group for the current medical rehabilitation training program; Based on the fitness value of each patient in the current medical rehabilitation training program and the behavioral correlation between the target patient and each patient, a personalized adaptation mapping label value between the target patient and each patient in the current medical rehabilitation training program is determined; The collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program are determined by all the mapped label values.

[0040] It should be noted that the target patient group in this application refers to a patient group with similar rehabilitation training behaviors to the target patient; the mapping label value in this application is a mapping reference index used to measure the adaptability of the target patient and other patients in the medical rehabilitation training program; and the synergistic feature in this application is used to characterize the adaptive association feature information formed between the target patient and other patients in the process of medical rehabilitation training.

[0041] In specific implementation, the selection of a target patient group with similar rehabilitation training behaviors to the target patient based on all behavioral correlations can be achieved in the following way: First, a correlation threshold can be set, and then patients with behavioral correlations greater than the correlation threshold can be considered as similar patients with similar rehabilitation training behaviors to the target patient. Further, the set of all similar patients can be considered as the target patient group with similar rehabilitation training behaviors to the target patient. Extracting the fitness value of each patient in the target patient group for the current medical rehabilitation training program can be achieved in the following way: For each patient in the target patient group, the postural deviation of the patient during the completion of the current medical rehabilitation training program is obtained, and this postural deviation and the difficulty level of the current medical rehabilitation training program are input into the aforementioned fitness evaluation model. The output value of the fitness evaluation model is then used as the patient's fitness value for the current medical rehabilitation training program, thus obtaining the fitness value of each patient in the target patient group for the current medical rehabilitation training program. Based on the fitness value of each patient in the current medical rehabilitation training program... The mapping label value for personalized adaptation of the target patient to the current medical rehabilitation training program can be determined by the following method: First, the behavioral correlation between the target patient and each patient is added together to obtain the sum of all behavioral correlations, and this sum is used as the correlation sum. Then, for each patient, the ratio between the behavioral correlation between the patient and the target patient and the sum of the correlations is used as the mapping weight. The product of the mapping weight and the patient's fitness value in the current medical rehabilitation training program is used as the mapping label value for personalized adaptation of the target patient to the current medical rehabilitation training program. Thus, the mapping label value for personalized adaptation of the target patient to the current medical rehabilitation training program is obtained. The collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program can be determined by the following method: The sum of all mapping label values ​​can be used as the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program.

[0042] It should be noted that the collaborative features in this application are mainly used to optimize the fitness prediction results of the target patient in order to improve the personalized effect of medical rehabilitation training. By analyzing the behavioral characteristics and adaptive performance of the target patient and other patients in the virtual reality interaction process, the group fitness pattern is extracted and mapped to the training fitness calculation of the target patient. This can not only make up for the possible bias caused by single patient data, but also learn from the successful experience of similar patients, making the fitness assessment more accurate. Finally, the collaborative features provide data support for the dynamic adjustment of the training program, ensuring that the training content not only meets individual needs, but also promotes patients to complete rehabilitation training more efficiently.

[0043] In step 104, the fitness prediction value is constrained by the cooperative feature to obtain the constrained fitness prediction value. When the constrained fitness prediction value is lower than the preset fitness threshold, the guidance speed of the current medical rehabilitation training project in the virtual reality device is adjusted.

[0044] In some embodiments, the fitness prediction value is obtained by applying collaborative constraints to the collaborative features, which can be achieved through the following steps: The collaborative constraint coefficients for the target patient's adaptation to the current medical rehabilitation training program are determined using the aforementioned collaborative features. The fitness prediction value after constraints is determined by the collaborative constraint coefficient and the fitness prediction value.

[0045] It should be noted that the synergistic constraint coefficient in this application is a coefficient index that measures the group's adaptive constraint; the synergistic constraint in this application refers to the synergistic adjustment of the target patient's fitness prediction value based on the group's adaptiveness.

[0046] In specific implementation, the synergistic constraint coefficient for the target patient and other patients to adapt to the current medical rehabilitation training program can be determined by the following method: the natural exponential function value of the inverse of the synergistic feature can be used as the synergistic constraint coefficient for the target patient and other patients to adapt to the current medical rehabilitation training program; the fitness prediction value after constraint can be determined by the synergistic constraint coefficient and the fitness prediction value, which can be used as the fitness prediction value after constraint.

[0047] It should be noted that, in this application, adjusting the guidance speed of the current medical rehabilitation training program in the virtual reality device when the constrained fitness prediction value is lower than the preset fitness threshold means reducing the guidance speed of the current medical rehabilitation training program in the virtual reality device when the constrained fitness prediction value is lower than the preset fitness threshold. It should be noted that when the constrained fitness prediction value is lower than the preset fitness threshold, it indicates that the current training difficulty for the target patient exceeds their adaptability, affecting the rehabilitation effect. Therefore, it is necessary to adjust the guidance speed of the health guidance actions in the virtual reality device. This allows for dynamic optimization of the training content based on the patient's actual adaptation. Through this personalized adjustment method, medical rehabilitation training can more accurately match the patient's recovery progress, improving the effectiveness and safety of the training. The fitness threshold can be set independently based on historical fitness values, and its specific setting is not limited.

[0048] It should also be noted that this application's solution monitors the simulated posture and actual posture of the target patient during virtual reality rehabilitation training, extracts posture deviations, and predicts the patient's fitness level in conjunction with the training difficulty level, thereby accurately assessing the patient's rehabilitation progress. Simultaneously, this application's solution introduces behavioral correlations between users, utilizes collaborative mapping to obtain group fitness characteristics, and applies collaborative constraints to the fitness prediction value to ensure more accurate and reliable fitness assessment. When the patient's fitness level falls below a threshold, the system can dynamically adjust the training difficulty (i.e., guidance speed) to make the rehabilitation training more suitable for the patient's current state. Through these mechanisms, this application's solution can not only track the patient's rehabilitation status in real time but also optimize the training plan by combining group data, ultimately achieving personalized medical and health management, thereby improving rehabilitation effectiveness and the scientific nature of training.

[0049] On the other hand, in some embodiments, this application provides a virtual reality-based medical and health management system, referencing... Figure 4 The figure is a schematic diagram of the structure of a virtual reality-based medical and health management system 400 according to some embodiments of this application. The virtual reality-based medical and health management system 400 includes: a monitoring module 401, a processing module 402, and an execution module 403, which are described below: Monitoring module 401, in this application, is mainly used to monitor the simulated posture and actual posture of the target patient during medical rehabilitation training using virtual reality equipment, to perform differential perception of the motion joints in the simulated posture and the actual posture, and then to extract the posture deviation of the target patient during medical rehabilitation training. Processing module 402, in this application, is used to predict the training adaptability of the target patient to the current medical rehabilitation training program based on the difficulty level of the current medical rehabilitation training program and the posture deviation, and obtain the fitness prediction value. In this application, the processing module 402 is also used to determine the degree of behavioral correlation between the target patient and other patients during the medical virtual reality interaction process, and then perform a collaborative mapping of the adaptability of the current medical rehabilitation training program based on all the behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program. The execution module 403 in this application is mainly used to perform collaborative constraints on the fitness prediction value through the collaborative features to obtain the constrained fitness prediction value. When the constrained fitness prediction value is lower than the preset fitness threshold, the guidance speed of the current medical rehabilitation training project in the virtual reality device is adjusted.

[0050] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described virtual reality-based medical and health management method.

[0051] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device implementing a virtual reality-based medical and health management method according to some embodiments of this application. The virtual reality-based medical and health management method in the above embodiments can... Figure 5 The computer device 500 shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0052] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0053] The communication bus 502 can be used to transmit information between the aforementioned components.

[0054] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0055] The memory 503 stores program code for executing the solution of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. In the above embodiments, the virtual reality-based medical and health management method can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0056] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0057] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0058] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0059] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described virtual reality-based medical and health management method.

[0060] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0061] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A medical and health management method based on virtual reality, characterized in that, Includes the following steps: The simulated and actual postures of the target patient during medical rehabilitation training using virtual reality equipment are monitored. The motion joints in the simulated and actual postures are differentially perceived, and then the posture deviations of the target patient during medical rehabilitation training are extracted. The training adaptability of the target patient to the current medical rehabilitation training program is predicted by the difficulty level of the current medical rehabilitation training program and the postural deviation, and the fitness prediction value is obtained. Determine the behavioral correlation between the target patient and other patients during medical virtual reality interaction, and then perform a collaborative mapping of the adaptability of the current medical rehabilitation training program based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program. The fitness prediction value is constrained by the cooperative feature to obtain the constrained fitness prediction value. When the constrained fitness prediction value is lower than the preset fitness threshold, the guidance speed of the current medical rehabilitation training project in the virtual reality device is adjusted.

2. The method as described in claim 1, characterized in that, Differential sensing of the motion joints in the simulated posture and the actual posture, and then extraction of posture deviations in the target patient during medical rehabilitation training, specifically include: The simulated attitude and the actual attitude are aligned in coordinates to obtain the coordinate-aligned simulated attitude and actual attitude. Extract the simulated coordinates of all motion joints from the simulated pose after coordinate alignment; Extract the actual coordinates of all motion joints from the actual pose after coordinate alignment; Differential matching is performed on all simulated coordinates and actual coordinates to obtain the attitude difference degree of all motion joints; Postural deviations in target patients during medical rehabilitation training are determined by the postural differences at all motion joints.

3. The method as described in claim 1, characterized in that, The training adaptability of the target patient to the current medical rehabilitation training program is predicted by using the difficulty level of the current medical rehabilitation training program and the aforementioned postural deviation. The predicted adaptability value specifically includes: The difficulty level of the current medical rehabilitation training program is determined based on the historical completion status of the current medical rehabilitation training program. Initialize an adaptive evaluation model for assessing patients' adaptation to medical rehabilitation training programs; The difficulty level of the current medical rehabilitation training program and the posture deviation are used as input features of the adaptive evaluation model; The adaptive evaluation model is used to predict the fitness level of target patients in adapting to the current medical rehabilitation training program.

4. The method as described in claim 1, characterized in that, Determining the behavioral correlation between the target patient and other patients during medical virtual reality interaction specifically includes: Acquire training logs of the target patient and other patients during the medical rehabilitation training process, and extract behavioral characteristics of the target patient and other patients during virtual reality interaction from the training logs; The behavioral correlation between the target patient and each other during medical virtual reality interaction is determined by the Euclidean distance of the behavioral characteristics between the target patient and each other.

5. The method as described in claim 1, characterized in that, Based on all behavioral correlations, the adaptability of the current medical rehabilitation training program is collaboratively mapped to obtain the collaborative characteristics between the target patient and other patients in the current medical rehabilitation training program, specifically including: Based on all behavioral correlations, a target patient group with similar rehabilitation training behaviors to the target patient was identified; Extract the fitness value of each patient in the target patient group for the current medical rehabilitation training program; Based on the fitness value of each patient in the current medical rehabilitation training program and the behavioral correlation between the target patient and each patient, a personalized adaptation mapping label value between the target patient and each patient in the current medical rehabilitation training program is determined; The collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program are determined by all the mapped label values.

6. The method as described in claim 1, characterized in that, The fitness prediction value is obtained by applying collaborative constraints to the collaborative features, specifically including: The collaborative constraint coefficients for the target patient's adaptation to the current medical rehabilitation training program are determined using the aforementioned collaborative features. The fitness prediction value after constraints is determined by the collaborative constraint coefficient and the fitness prediction value.

7. The method as described in claim 1, characterized in that, The virtual reality device includes a VR headset, motion tracking gloves, force feedback devices, and position tracking devices.

8. A virtual reality-based medical and health management system, characterized in that, include: The monitoring module is used to monitor the simulated posture and actual posture of the target patient during medical rehabilitation training using virtual reality equipment, perform differential perception on the motion joints in the simulated posture and the actual posture, and then extract the posture deviation of the target patient during medical rehabilitation training. The processing module is used to predict the training adaptability of the target patient to the current medical rehabilitation training program based on the difficulty level of the current medical rehabilitation training program and the posture deviation, and obtain the fitness prediction value. The processing module is also used to determine the behavioral correlation between the target patient and other patients during the medical virtual reality interaction process, and then perform a collaborative mapping on the adaptability of the current medical rehabilitation training program based on all behavioral correlations to obtain the collaborative characteristics of the target patient and other patients in the current medical rehabilitation training program. The execution module is used to perform collaborative constraints on the fitness prediction value through the collaborative features to obtain a constrained fitness prediction value. When the constrained fitness prediction value is lower than a preset fitness threshold, the guidance speed of the rehabilitation guidance of the current medical rehabilitation training project in the virtual reality device is adjusted.

9. A computer device comprising a memory and a processor, the memory storing code, characterized in that, The processor is configured to acquire the code and execute the virtual reality-based medical and health management method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the virtual reality-based medical and health management method as described in any one of claims 1 to 7.

Images